Ship and Offshore Structures Congress PDF

PROCEEDINGS OF THE 19TH INTERNATIONAL SHIP AND OFFSHORE
STRUCTURES CONGRESS
Prelims_Vol-1.indd i
7/31/2015 10:57:24 AM
Proceedings of the 19th International

Ship and Offshore Structures
Congress
Editors
C. Guedes Soares & Y. Garbatov

Centre for Marine Technology and Ocean Engineering (CENTEC), Instituto
Superior Tcnico, Universidade de Lisboa, Lisbon, Portugal
VOLUME 1
Prelims_Vol-1.indd iii
7/31/2015 10:57:24 AM
CRC Press/Balkema is an imprint of the Taylor & Francis Group, an informa business
2015 Taylor & Francis Group, London, UK
Typeset by MPS Limited, Chennai, India
Printed and bound in Great Britain by CPI Group (UK) Ltd, Croydon, CR0 4YY
All rights reserved. No part of this publication or the information contained herein may be
reproduced, stored in a retrieval system, or transmitted in any form or by any means,
electronic, mechanical, by photocopying, recording or otherwise, without written prior
permission from the publishers.
Although all care is taken to ensure integrity and the quality of this publication and the
information herein, no responsibility is assumed by the publishers nor the author for any
damage to the property or persons as a result of operation or use of this publication
and/or the information contained herein.
Published by:
CRC Press/Balkema
P.O. Box 11320, 2301 EH Leiden, The Netherlands
e-mail: Pub.NL@taylorandfrancis.com
www.crcpress.com www.taylorandfrancis.com
ISBN set: 978-1-138-02895-1 (2 volumes hardback and CDROM)

ISBN Volume 1: 978-1-138-02896-8
ISBN Volume 2: 978-1-138-02897-5
ISBN: 978-1-315-64719-7 (eBook PDF)
Prelims_Vol-1.indd iv
7/31/2015 10:57:24 AM
Table of contents
Preface
xxv
VOLUME 1
Report of Committee I.1: Environment
1
2
Prelims_Vol-1.indd v
Introduction
Environmental data
2.1
Wind
2.1.1 Locally sensed wind measurements
2.1.2 Remotely sensed wind measurements
2.1.3 Numerical modelling to complement measured data
2.2
Waves
2.2.1 Locally sensed wave measurements
2.2.2 Remotely sensed wave measurements
2.2.4 Wave description from measured ship motions
2.3
Current
2.3.1 In-situ current measurements
2.3.2 Remotely sensed current measurements
2.4
Sea water level
2.4.1 Locally sensed sea water level measurements
2.4.2 Remotely sensed sea water level measurements
2.5
Ice and snow
2.5.1 Locally and remotely sensed ice and snow measurements
Environmental models
3.1
Wind
3.1.1 Analytical description of wind
3.1.2 Statistical and spectral description of wind
3.2
Waves
3.2.1 Analytical and numerical wave models
3.2.2 Experimental description of waves
3.2.3 Statistical description of waves
3.2.4 Spectral description of waves
3.3
Current
3.3.1 Analytical description of current
3.3.2 Statistical and spectral description of current
3.4
Sea water level
3.5
Ice and snow
Climate change
4.1
New IPPC scenarios and climate models
4.1.1 Temperature
4.1.2 Ice and snow
4.1.3 Sea water level
1
4
5
6
6
7
8
8
9
12
13
14
14
14
15
15
15
15
15
15
15
15
16
17
17
18
18
20
20
28
30
32
33
33
34
34
34
34
35
36
37
38
7/31/2015 10:57:24 AM
vi
Table of contents
4.1.4 Wind and waves

4.1.5 Ocean circulation
5
Special topics
5.1
Hurricane
5.2
Wave current interaction
5.2.1 Wave-current interaction model
5.2.2 Numerical and analytical method
5.2.3 Experiments and measurements
5.3
Wave and wind energy resource assessment
6
Design and operational environment
6.1
Design
6.1.1 Met-Ocean data
6.1.2 Design environment
6.1.3 Design for climate change and rogue waves
6.2
Operations
6.2.1 Planning and executing marine operations
6.2.2 Northern sea route, weather routing, warning criteria and current
6.2.3 Eco-efficiency ship operation
7
Conclusions
7.1
Advances
7.2
Recommendations
Acknowledgements
References
38
40
40
40
41
41
43
44
45
47
47
47
48
51
52
53
54
56
57
59
60
60
61
Report of Committee I.2: Loads
73
1
2
Introduction
Computation of wave-induced loads
2.1
Zero speed case
2.1.1 Body wave interactions
2.1.2 Body-wave-current interactions
2.1.3 Multibody interactions
2.2
Forward speed case
2.3
Hydroelasticity methods
2.4
Loads from abnormal waves
Ship structures specialist topics
3.1
Slamming and whipping
3.2
Sloshing
3.2.1 Analytical methods
3.2.2 Experimental investigations
3.2.3 Numerical simulation
3.2.4 Sloshing with internal suppressing structures
3.2.5 Sloshing and ship motions
3.3
Green water
3.4
Experimental and full scale measurements
3.5
Loads due to damage following collision/grounding
3.6
Weather routing and operational guidance
Offshore structures specialist topics
4.1
Vortex-induced vibrations (VIV) and vortex-induced motions (VIM)
4.1.1 VIV
4.1.2 VIM
4.2
Mooring systems
4.3
Lifting operations
Prelims_Vol-1.indd vi
75
75
75
75
79
79
80
83
85
87
87
91
91
92
93
94
95
96
99
101
102
104
104
104
106
108
111
7/31/2015 10:57:24 AM
Table of contents
vii
4.4
Wave-in-deck loads
4.5
Floating offshore wind turbines
5
Probabilistic modelling of loads on ships
5.1
Probabilistic methods
5.2
Equivalent design waves
5.3
Design load cases and ultimate strength
6
Fatigue loads for ships
7
Uncertainty analysis
7.1
Load uncertainties
7.2
Uncertainties in loading conditions
8
Conclusions
References
113
113
115
115
117
119
120
123
123
124
125
128
Report of Committee II.1: Quasi-static response
141
1
2
144
144
144
146
147
148
148
148
148
148
149
149
149
149
151
152
152
152
152
153
153
153
154
155
155
156
160
161
161
161
162
162
163
163
164
165
165
166
Introduction
Strength assessment approaches
2.1
Modelling of loads by quasi-static analysis
2.2
Response calculation
2.3
Reliability
Calculation procedures
3.1
Taxonomy of engineering assessment methods
3.1.1 Simplified analysis (rule-based design) / first principles
3.1.2 Direct calculations
3.1.3 Reliability analyses
3.1.4 Optimisation-based analyses
3.2
Design for production loads modelling
3.2.1 Rules versus rational based ship design
3.2.2 Direct simulations for global quasi-strength assessment
3.2.3 Loads extracted from experiments and testing
3.2.4 Loads from seakeeping codes
3.3
Structural modelling
3.3.1 Finite element modelling
3.3.2 Models for global and detailed analyses
3.3.3 Composite structures
3.4
Structural response assessment
3.4.1 Buckling and ultimate strength
3.4.2 Fatigue strength
3.4.3 Ship dynamics vibrations
3.5
Validation of calculation results
3.5.1 Model scale experiments and testing
3.5.2 Full scale hull stress monitoring
Uncertainties associated with reliability-based quasi-static response assessment
4.1
Uncertainties associated with loads
4.1.1 Still water and wave loads
4.1.2 Ice loads
4.1.3 Combination factors
4.2
Uncertainties in structural modelling
4.2.1 Corrosion
4.2.2 Structural characteristics
4.2.3 Reliability and risk-based structural assessment
4.2.4 Methods and criteria
4.2.5 Structural capacity
Prelims_Vol-1.indd vii
7/31/2015 10:57:24 AM
viii
Table of contents
4.3
Risk-based inspection, maintenance and repair

4.3.1 Inspection
4.3.2 Maintenance and repair
5
Ship structures
5.1
Developments in international rules and regulations
5.1.1 IMO goal-based standards
5.1.2 IACS common structural rules for bulk carriers and oil tankers
5.1.3 Development of structural design software systems
5.2
Special ship concepts
5.2.1 Service vessels for wind mills and offshore platforms
5.2.2 Container ships
5.2.3 LNG/LPG tankers
5.2.4 Other ship types
6
Offshore structures
6.1
Types of analysis for various floating offshore structures
6.2
Types of analysis for various fixed offshore structures
6.3
Uncertainty, risk and reliability in offshore structural analysis
7
Benchmark study
7.1
Methodology
7.2
Simplified methods
7.3
Quasi-static linear FE analysis
7.4
Nonlinear, transient dynamic FE analysis
7.5
Concluding remarks
8
Conclusions and recommendations
References
167
167
168
169
169
169
170
172
173
173
173
174
175
176
176
179
182
184
184
186
188
188
190
191
192
Report of Committee II.2: Dynamic response
209
1
2
211
211
211
211
219
220
220
220
221
222
222
224
224
227
227
228
229
229
229
229
230
232
234
234
Introduction
Ship structures
2.1
Environmental-induced vibrations
2.1.1 Wave-induced vibration
2.1.2 Ice-induced vibration
2.2
Machinery or propeller-induced vibrations
2.2.1 Propeller-induced vibration
2.2.2 Machinery-induced vibration
2.2.3 Numerical and analytical vibration studies of ship structures
2.3
Noise
2.3.1 Interior noise
2.3.2 Air radiated noise
2.3.3 Underwater radiated noise
2.4
Sloshing impact
2.4.1 Experimental approaches
2.4.2 Numerical modelling
2.4.3 CCS structural response
2.4.4 Current approaches for sloshing assessment
2.5
Air blast and underwater explosion
2.5.1 Air blast
2.5.2 Underwater explosion
2.6
Damping and countermeasures
2.7
Monitoring
2.7.1 Hull structural monitoring system
Prelims_Vol-1.indd viii
7/31/2015 10:57:24 AM
Table of contents
ix
2.7.2 New sensors technology and application

2.7.3 New full scale monitoring campaigns and related studies
2.8
Uncertainties
2.9
Standards and acceptance criteria
2.9.1 Habitability
2.9.2 Underwater noise
2.9.3 Others
3
Offshore structures
3.1
Vibration
3.1.1 Wind-induced vibration
3.1.2 Wave-induced vibration
3.1.3 Vortex-induced motion
3.1.4 Internal flow-induced vibration
3.1.5 Ice-induced vibration
3.2
Very large floating structures
3.3
Noise
3.3.1 Analysis of underwater noise by pile-driving
3.3.2 Measurement and mitigation of underwater noise
3.3.3 Equipment noise
3.4
Blast
3.5
3.6
Uncertainties
3.7
4 Conclusion
References
234
236
239
241
241
242
242
243
243
243
244
245
246
246
249
249
250
250
250
251
252
253
254
254
257
Report of Committee III.1: Ultimate strength
279
1
2
282
283
283
283
284
284
284
285
286
288
288
288
289
290
290
291
292
294
294
298
299
299
299
300
Introduction
Fundamentals
2.1
Design for ultimate strength
2.2
General characteristics of ultimate strength
Assessment procedure for ultimate strength
3.1
Empirical and analytical methods
3.1.1 Introduction
3.1.2 Hull structures
3.1.3 Residual strength of damage hull structures
3.1.4 Plates and stiffened plates
3.2
Numerical methods
3.2.1 Introduction
3.2.2 Nonlinear FE method
3.2.3 Idealized structural unit method
3.2.4 Conclusion
3.3
Experimental methods
3.4
Reliability assessment
3.5
Rules and regulations
3.5.1 Harmonized common structural rules
3.5.2 Updates to offshore rules and guides
Ultimate strength of various structures
4.1
Tubular members and joints
4.1.1 Tubular members
4.1.2 Tubular joints
Prelims_Vol-1.indd ix
7/31/2015 10:57:24 AM
Table of contents
4.2
Steel plate and stiffened plates

4.2.1 Introduction
4.2.2 Analytical formulations for ultimate strength of stiffened panels
4.2.3 Uniaxial compression
4.2.4 Multiple load effects
4.2.5 Panels with openings, cut-outs or rupture damage
4.2.6 Welding effects
4.2.7 In service degradation
4.2.8 Experimental testing
4.2.9 Optimization
4.2.10 Conclusions
4.3
Shells
4.4
Ship structures
4.4.1 Progressive collapse methods
4.4.2 Damaged structures
4.4.3 Corrosion
4.4.4 Complex ship structural components and complex loading
4.4.5 Reviews and applications
4.5
Offshore structures
4.6
Composite structures
4.6.1 Failure identification and material degradation models
4.6.2 Ultimate strength of composite stiffened panels and
box girders
4.6.3 Environmental effects
4.6.4 Compression after impact
4.7
Aluminum structures
4.7.1 Introduction
4.7.2 Weld-induced effects
4.7.3 Formulation development
4.7.4 Experimental investigation
4.7.5 Fiber-reinforced polymer strengthened
4.7.6 Sandwich panels
4.7.7 Hull girder
4.7.8 Summary and recommendation for future works
5
Benchmark study
5.1
Small box girder
5.1.1 Introduction
5.1.2 Model parameters
5.1.3 Baseline calculations
5.1.4 Comparison with solid element mesh
5.1.5 Comparison with Smith method
5.1.6 Effect of imperfection amplitude and shape
5.1.7 Effect of material model parameters
5.1.8 Effect of plating thickness
5.1.9 Summary/conclusions
5.2
Three hold model of hull girder
5.2.1 Calculation cases
5.2.2 Calculation results
5.3
Summary and recommendation for future works
6
Conclusion and recommendation
References
Prelims_Vol-1.indd x
301
301
302
302
303
304
304
305
305
306
306
306
308
309
310
310
310
312
312
314
315
316
317
317
318
318
318
320
320
321
321
321
322
322
322
322
323
324
327
328
329
331
331
332
332
332
335
338
339
340
7/31/2015 10:57:24 AM
Table of contents
xi
Report of Committee III.2: Fatigue and fracture
351
1
2
354
354
354
354
354
355
355
355
355
355
355
355
356
356
356
356
356
356
357
357
357
357
357
357
357
358
358
358
358
359
362
363
364
364
Introduction
Fatigue life-cycle design philosophies and methodologies
2.1
Fatigue and fracture in marine structures
2.2
Preliminary design
2.3
Detailed design
2.4
Fabrication
2.5
In-service maintenance
2.5.1 Inspection techniques
2.5.2 Inspection planning
2.6
Fatigue strength
2.6.1 S-N curves related to expected workmanship
2.6.2 Crack propagation parameters
2.7
Fracture strength
2.8
Fatigue loads
2.8.1 Wave loads
2.8.2 Loading unloading
2.8.3 Vibrations
2.9
Environmental effects
2.9.1 In air
2.9.2 Seawater
2.9.3 Other aggressive environments
2.9.4 Coating and coating life
2.10
Fatigue, fracture & failure criteria
2.10.1 Failure definition
2.10.2 Uncertainties
2.10.3 Safety factors
Factors influencing fatigue/fracture
3.1
Resistance
3.1.1 Thickness and size
3.1.2 Environment (corrosion)
3.1.3 Temperature
3.1.4 Residual stress & constraint, mean stress
3.2
Materials
3.2.1 Metallic alloys
3.2.2 Fatigue & fracture improvements through material changes,
surface treatment
3.3
Loading
3.3.1 Stochastic loading (load interaction effects (sequence))
3.3.2 Cycle counting spectral, time-domain, stress ranges, means stress effect
3.3.3 Complex stresses
3.3.4 Recent developments in multiaxial fatigue criteria
3.4
Structural integrity/life cycle management
3.4.1 Fabrication and repair
3.4.2 Inspection & monitoring of structure and coatings
3.4.3 Inspection and maintenance
3.5
Composites
Fatigue assessment methods
4.1
Overview
4.2
Fatigue damage models
4.2.1 Stress based concepts
Prelims_Vol-1.indd xi
364
365
365
365
366
369
373
373
374
376
377
378
379
381
381
7/31/2015 10:57:24 AM
xii
Table of contents
4.2.2 Strain concepts

4.2.3 Notch-intensity factor, -integral and -energy density concepts
4.2.4 Confidence and reliability
4.3
Fracture mechanics models
4.3.1 Crack growth rate model
4.3.2 Crack growth assessment
4.3.3 Fracture mechanics based fatigue evaluation of ship structures
4.4
Rules, standards & guidance
4.4.1 Ship rules
4.4.2 Design codes for offshore structures
4.4.3 IIW recommendation
4.4.4 ISO standards
4.5
Acceptance criteria
4.6
Measurement techniques
4.6.1 Crack growth and propagation
4.6.2 Fatigue
4.6.3 Material properties
4.6.4 Fracture toughness
5
Benchmarking study
5.1
Problem statement
5.2
Analytical methods
5.3
Numerical analysis using FEM
5.4
Results
5.5
Discussion & benchmarking study conclusions
6
Summary & conclusions
References
382
382
383
385
389
390
391
392
392
394
395
395
395
396
396
397
398
398
399
399
400
402
403
404
404
405
Report of Committee IV.1: Design principles and criteria
415
418
418
418
419
419
419
420
420
421
422
423
423
427
428
430
431
431
432
434
434
437
437
437
Introduction
1.1
General concept of sustainability oriented design
1.2
Goal oriented normative framework
1.3
Procedures for the impact analysis of regulations
Quantification of sustainability aspects
2.1
Economic aspects
2.2
Human aspects
2.3
GCAF and NCAF indicators for loss of life
2.3.1 Life Quality Index
2.3.2 DALY and QALY indicators
2.4
Environmental aspects
2.4.1 Cost of averting a tonne of oil spilt (CATS)
2.4.2 CO2 emissions costs
2.4.3 Other emissions costs
Depreciation rates in decision making
3.1
Pure time preferences
3.2
Precautionary approach vs standard economic theory
3.3
Integrated Assessment Models
3.4
Tails of the probability distributions
3.5
Role of the discounting rate
3.6
Conclusion (depreciation rates)
Examples related to sustainability oriented design
4.1
Probability based design
Prelims_Vol-1.indd xii
7/31/2015 10:57:24 AM
Table of contents
xiii
4.2
Lifecycle design
4.3
Lifecycle design considering future climate change
5
Regulatory framework for marine structures
5.1
Development of goal based standards at IMO
5.1.1 IACS harmonized common structural rules for bulk carriers and tankers
5.1.2 Goal based standards/safety level approach (GBS/SLA) at IMO
5.2
Regulatory actions implemented at IMO targeting environmental protection
5.2.1 Energy Efficiency Design Index (EEDI)
5.2.2 NOx SOx control
5.2.3 Emission control areas
5.2.4 MARPOL Annex V prevention of pollution by garbage from ships
5.2.5 IMO ship recycling (the Hong Kong convention)
5.2.6 Pre-normative investigations at imo in the field of noise
radiation into water
5.3
Other (non IMO) regulatory actions in the field of ships
5.3.1 Developments in the naval ship code
5.3.2 Inland vessels
5.3.3 EU directive on safety of offshore oil and gas operations
5.4
Comments on the recent developments in the normative framework
6
Studies focussing on environmental impact
6.1
Studies on green house gas emissions
6.2
Studies on countermeasures to limit emissions
6.2.1 Slow steaming
6.2.2 Scale effects and propulsive improvements
6.2.3 Discussions of the EEDI concept
6.2.4 Studies on control of NOx and SOx emissions
6.2.5 Emissions trading schemes
6.2.6 Alternative fuels
7
Conclusions
References
439
441
443
444
444
446
447
447
447
447
448
448
Report of Committee IV.2: Design methods
459
1
2
461
461
462
462
465
466
467
467
467
469
469
471
472
472
475
477
482
483
484
Introduction
Design methodology
2.1
Developments in procedural aspects of ship design methodology
2.2
Developments in Design-for-X and risk-based design
2.3
Developments in ship form-function mapping, tradespace searches
2.4
Handling uncertainty in future operating context
Design tools
3.1
Introduction
3.2
Development of design tools
3.3
Tools for lifecycle cost modeling and lifecycle assessment
3.4
Links between design tools and production and operational phases
3.5
Developments in integrated naval architecture packages
Optimization developments
4.1
Introduction to Design Support Systems (DESS)
4.2
Parallel processing and hardware developments
4.3
Developments in structural optimization algorithms (optimization solvers)
4.4
Surrogate modeling and variable fidelity approaches (surrogate solvers)
4.4.1 Surrogate modeling in design and optimization
4.4.2 Surrogate modeling in risk and safety analyses
Prelims_Vol-1.indd xiii
449
449
449
450
451
451
451
451
452
452
452
452
453
453
453
453
454
7/31/2015 10:57:25 AM
xiv
Table of contents
4.5
Optimization for production (design quality modulesPRODUCTION)
4.6
Optimization for lifecycle costing (design quality modulesLCC)
5
Classification society software review
5.1
Background, motivation, and aim
5.2
Tool analysis
5.2.1 Overall functionality
5.2.2 Evaluation criteria
5.3
Classification societies tools details
5.3.1 American Bureau of Shipping (ABS)www.eagle.org
5.3.2 Bureau Veritas (BV)www.bureauveritas.com
5.3.3 China Classification Society (CCS)www.ccs.org.cn
5.3.4 Croatian Register of Shipping (CRS)www.crs.hr
5.3.5 DNVGL
5.3.6 Korean Register of Shipping (KR)www.krs.co.kr
5.3.7 Nippon Kaiji Kyokai (ClassNK)www.classnk.com
5.3.8 Polish Register of Shipping (PRS)www.prs.pl
5.3.9 Registro Italiano Navale (RINA)www.rina.org
5.4
Conclusions and future challenges
6
Structural lifecycle management
6.1
Introduction
6.2
Tool development
6.3
Data interchange and standards
6.4
Integration with repair
6.5
Integration with structural health monitoring systems
6.6
Summary of the lifecycle structural management systems
7
Obstacles, challenges, and future developments
8
Conclusion
Acknowledgments
References
484
486
487
487
488
488
488
490
490
491
491
492
493
495
496
497
498
498
499
499
500
502
503
504
506
506
508
509
509
VOLUME 2
Report of Committee V.1: Accidental limit states
519
1
2
523
524
524
525
527
527
527
528
528
530
532
532
532
532
532
533
533
533
Introduction
Fundamentals of ALS design
2.1
Introduction
2.2
Codes and standards
2.3
Updates of codes and standards
2.4
Uncertainties in ALS in design
2.5
Practice for ships
Hazard identification
3.1
Introduction
3.2
Safety levels in ALS design
4.1
Introduction
4.2
Safety level of offshore structures in ALS
4.2.1 General
4.2.2 Discussion of new ISO standards for offshore structures
4.2.3 Characterization of hazards
4.2.4 Accidental design situations
4.2.5 ALS safety levels implied in structural codes
Prelims_Vol-1.indd xiv
7/31/2015 10:57:25 AM
Table of contents
4.3
Safety level of ship structures in ALS

4.3.1 General
4.3.2 GBS of ship structure design
4.3.3 Safety level in ULS in CSR
4.3.4 Safety level in ALS in CSR-H
Assessment of accidental loads
5.1
Introduction
5.2
Explosion load assessment
5.2.1 Deterministic approach
5.2.2 Probabilistic approach
5.2.3 Definition of explosion loads for design
5.3
Fire load assessment
5.3.2 Risk-based and probabilistic approach
5.4
Load assessment for collision accidents
5.4.2 Risk-based and probabilistic approach
5.5
Load assessment for dropped object accidents
5.5.2 Risk-based approach
Determination of action effects
6.1
Introduction
6.2
Review of numerical tools
6.3
Modelling geometries
6.4
Modelling loads
6.4.1 Ship collision
6.4.2 Dropped objects
6.4.3 Explosions
6.4.4 Fire
6.5
Material models
6.5.1 Plasticity model
6.5.2 Stress-strain curve
6.5.3 Failure criteria
6.6
Uncertainties of ALS models
6.7
6.8
Appendix A
6.8.1 True stress-strain curve for Ls-Dyna
Benchmark study. Resistance of topside structures Subjected to fire
7.1
Scope of work
7.2
Strategy of benchmark study
7.3
Input
7.3.1 Geometry of target structure
7.3.2 Material data
7.3.3 Boundary conditions
7.3.4 Loads
7.4
Results
7.4.1 Static analysis
7.4.2 Push-down analysis
7.4.3 Fire analysis
7.4.4 Design of PFP
7.4.5 Effects of boundary conditions
Prelims_Vol-1.indd xv
xv
535
535
535
536
536
538
538
538
539
539
542
542
542
543
544
545
545
546
546
547
547
547
549
550
552
552
553
553
554
554
557
557
557
560
560
560
560
561
561
562
562
562
563
564
564
566
566
567
568
570
571
7/31/2015 10:57:25 AM
xvi
8
9
Table of contents
7.4.6 Methods of controlling numerical instability for beam element model

7.4.7 Effects of local heat flux
7.5
Conclusion from the benchmark study
References
Annex 1. Material models for non-linear finite element analysis
9.1
Introduction
9.2
Guidelines and standards
9.3
Material model database
9.3.1 Steel
9.3.2 Aluminium
9.3.3 Foam, isolator, rubber
9.3.4 Ice
9.3.5 Air
9.3.6 Water
9.3.7 Explosives
9.3.8 Risers, umbilical or power cable
9.3.9 Composites
9.3.10 Concrete
9.3.11 Soil
9.4
References
571
573
575
576
579
579
580
580
580
583
584
584
585
586
586
587
587
588
588
589
Report of Committee V.2: Natural gas storage and transportation
591
1
2
3
593
593
595
595
595
595
596
596
598
Introduction
Background
Safety and design
3.1
Cargo containment
3.1.1 Non-self supporting tanksmembrane tanks
3.1.2 Independent tanks
3.1.3 New development of CCS
3.2
Structural integrity and rules
3.3
Sloshing
3.3.1 Global flow and sloshing-ship motion coupling, online
sloshing prediction
3.3.2 Long-term assessment
3.3.3 Experimental methods, benchmark
3.3.4 Sloshing model test benchmark
3.3.5 Sloshing physics, scaling ELPs, dominating physics and
relevant scaling laws
3.3.6 Numerical methods
3.4
Leakage
3.5
Fatigue
3.6
Collision, grounding, flooding
3.7
Sloshing control
3.8
Fire safety, temperature control of hull structures
LNG as fuel
4.1
Why LNG as fuel
4.2
LNG supply chain
Safety and design special applications
5.1
Floating LNG, FLNG, FSRU
5.2
Side by side or tandem mooring?
5.3
Arctic
Prelims_Vol-1.indd xvi
598
599
600
600
600
601
602
602
603
605
605
608
608
608
610
610
611
612
7/31/2015 10:57:25 AM
Table of contents
xvii
6 Conclusions
References
612
612
Report of Committee V.3: Materials and fabrication technology
619
1
2
622
622
Introduction
General trends
2.1
Developments in the maritime markets and their impact on the trends
in Fabrication and materials technologies
2.1.1 Korea
2.1.2 Japan
2.1.3 China
2.1.4 Europe
2.1.5 Brazil
2.2
Ongoing research programmes on fabrication and materials
2.2.1 Korea
2.2.2 Japan
2.2.3 China
2.2.4 Europe
2.2.5 Brazil
2.2.6 USA
Structural materials
3.1
Metallic materials
3.1.1 Aluminium alloys
3.1.2 Titanium
3.1.3 Metal foam
3.1.4 Application of metals in low temperatures
3.2
Non-metallic materials
3.2.1 Fire resistant materials
3.2.2 Bio-composites
3.2.3 Influence of sea water on non-metallic materials
3.2.4 Recycling and disposal
3.2.5 Application of non metallic materials at low temperatures
3.3
Hybrid materials
Joining and fabrication technology
4.1
Advances in joining technology
4.1.1 Welding automation and recent developments in joining technologies
4.1.2 Underwater welding
4.1.3 Frictions stir welding of steel
4.2
Innovations in fabrication technology
4.2.1 Plate bending with line heating
4.2.2 Post-treatment of welded joints and plate edges
4.2.3 Hybrid structures and joints
4.3
Influence of production quality on strength
4.3.1 Weld geometry and misalignments
4.3.2 Effect residual stress and distortions
4.3.3 Utilisation of high strength steel and thin plates
4.4
Dimension and quality control
Corrosion protection
5.1
Protection rules
5.2
Coating and paints
5.2.1 Epoxy-based coating systems
Prelims_Vol-1.indd xvii
622
624
624
624
624
625
625
625
626
626
626
627
628
629
629
629
630
630
631
633
634
634
636
636
637
637
637
637
637
638
638
640
640
640
641
642
642
643
643
644
644
644
645
645
7/31/2015 10:57:25 AM
xviii
Table of contents
5.2.2 Zinc-rich paints

5.2.3 Thermal spraying and deposition
5.2.4 Antifouling (AF) coatings
5.2.5 Self healing coatings
5.2.6 Intelligent coatings
5.2.7 Ice-breaker coatings
5.3
Cathodic protection
5.4
Corrosion resistant steels
5.5
Corrosion monitoring
5.6
Non destructive testing
5.6.1 Visual inspection of welds
5.6.2 Inspection for delayed (hydrogen induced) cracking
5.6.3 Methods of inspection
5.6.4 Under film corrosion detection
6 Manufacturing simulation
6.1
Discrete event simulation and production optimization
6.1.1 Layout planning
6.1.2 Production planning
6.1.3 Outfitting and customization
6.1.4 Logistic simulations
6.2
Virtual and augmented reality
7
Welding simulation
7.1
Computation welding mechanics
7.2
Arc welding simulation methodologies
7.2.1 Sequentially coupled thermos-mechanical models
7.2.2 Thermo-mechanical staggered coupled
7.3
Heat source models
7.4
Material models
7.5
Thermal- and mechanical boundary conditions
7.6
Mesh size
7.7
Computational time and cost
7.8
Weld residual stress measurements
7.9
Benchmark case
8
References
645
645
646
646
646
646
647
647
648
648
648
648
649
649
649
650
650
651
651
652
652
653
653
653
653
653
654
655
656
657
657
657
658
659
660
Report of Committee V.4: Offshore renewable energy
669
1
2
671
671
671
672
673
674
674
674
675
675
678
678
679
681
Introduction
Offshore renewable energy resources
2.1
Offshore wind energy resources
2.1.1 Resource assessment
2.2
Wave energy resources
2.3
Tidal and ocean current energy resources
2.3.1 Physical resource assessment
2.3.2 Numerical resource modelling
Offshore wind turbines
3.1
Recent industry and research development
3.2
Numerical modelling and analysis
3.2.1 Numerical tools state-of-the-art
3.2.2 Load and response analysis of bottom-fixed wind turbines
3.2.3 Load and response analysis of floating wind turbines
Prelims_Vol-1.indd xviii
7/31/2015 10:57:25 AM
Table of contents
3.3
xix
Physical testing
3.3.1 Laboratory testing
3.3.2 Field testing
3.4
Transportation, installation, operation and maintenance
3.4.1 Current industry and research development
3.4.2 Numerical simulations of marine operations
3.4.3 Guidelines on marine operations for offshore wind turbine
transportation, installation, operation and maintenance
3.5
Rules and standards
4
Wave energy converters
4.1
4.1.1 Load and motion response analysis
4.1.2 Mooring analysis
4.1.3 Power take-off analysis
4.2
Physical testing
4.2.1 Laboratory testing and validation of numerical tools
4.2.2 Field testing
4.3
Rules and standards
5
Tidal and ocean current turbines
5.1
Development, modelling and testing of tidal current energy converters
5.1.1 Device development
5.1.2 Numerical modelling and experimental testing
5.2
Environmental impact
5.2.1 Marine planning
5.3
Economic feasibility
6
Combined use of ocean space
7
Conclusions and recommendations for future work
References
692
692
693
695
695
698
699
700
701
701
702
703
703
703
703
704
704
704
705
707
709
Report of Committee V.5: Naval vessel design
723
1
2
726
726
726
727
728
729
729
730
730
730
730
731
731
731
732
732
733
733
733
733
733
Introduction
Naval class rule development/progress
2.1
Introduction
2.2
Military structural requirements
2.3
Military operational safety loads
2.4
Military performance loads
2.5
Concluding remarks
Military loads
3.1
Underwater weapon effects
3.1.1 Primary shock wave
3.1.2 Shock wave reflections and cavitation
3.1.3 Bubble dynamics and jetting
3.2
Above water weapons effects
3.2.1 External blast
3.2.2 Internal blast
3.2.3 Bullets and fragments
3.3
Maritime improvised explosive devices
3.4
Concluding remarks
Naval service life management
4.1
Introduction
Prelims_Vol-1.indd xix
687
687
689
689
690
691
7/31/2015 10:57:25 AM
xx
Table of contents
4.2
Ship service life in context

4.2.1 Australian LPA class
4.2.2 Australian Adelaide class FFG-07
4.2.3 ANZAC class
4.3
Determining the remaining life of a warship
4.4
Naval structural monitoring programs
4.5
Consequence of increasing displacement
4.6
Options for enhancing fatigue life of warships
Naval specific structure design
5.1
Structural uniqueness of naval ships
5.2
Naval integrated permanent structures
5.2.1 Flight decks (vertical)
5.2.2 Stern ramps (launch and recovery systems)
5.2.3 Blast resistant structures
5.3
Naval modular flexible structures
5.3.1 Mission bays
5.3.2 Weapon modules
5.3.3 Advanced enclosed masts/sensor (enclosed aperture stations)
5.4
Conclusion
Naval mast design
6.1
Introduction
6.2
Types of naval masts
6.3
Materials (composite vs. steel vs. aluminum)
6.4
Loads
6.4.1 Weight of equipment
6.4.2 Environmental loadings (includes wind and seaway loads)
6.4.3 Thermal
6.4.4 Shock and blast
6.4.5 Load combinations
6.5
Vibration and resonance
6.6
Structural analysis and design
6.7
Other considerations
6.8
Classification society rules for mast design
6.9
Conclusions
Progressive collapse analysis and residual strength assessment
7.1
Introduction
7.2
Progressive collapse method overview
7.3
Development of the progressive collapse method
7.4
Residual strength assessment by progressive collapse method
7.5
Use of FEA for progressive collapse assessment
7.6
Progressive collapse analysis within classification society rules
7.7
Discussion and conclusions
High speed naval craft
8.1
Naval applications
8.2
Defining a high speed craft
8.2.1 Principles
8.2.2 Hull form
8.2.3 Standards and regulations
8.3
Defining operational limitations
8.3.1 Operational profile
8.3.2 Operational envelope
Prelims_Vol-1.indd xx
734
734
734
735
735
737
738
738
739
739
739
739
740
741
742
742
743
743
744
744
744
745
746
747
747
747
747
747
747
748
748
749
749
749
750
750
750
751
751
752
752
753
754
754
755
755
756
756
757
757
757
7/31/2015 10:57:25 AM
Table of contents
8.4
xxi
Accelerations effects
8.4.1 Slamming
8.4.2 Human factors
8.4.3 Fatigue
8.5
Material technologies
8.5.1 Steel
8.5.2 Aluminium
8.5.3 Fibre reinforced plastics (FRP)
8.6
Unmanned naval high speed craft
8.7
Classification society rules
8.8
Conclusion
9
Benchmark studies
9.1
Whipping response of ship
9.1.1 Introduction
9.1.2 UNDEX bubble phenomena
9.1.3 Experimental investigations
10 Discussions and conclusions
References
758
758
758
759
759
759
760
760
760
760
761
761
761
761
762
763
764
766
Report of Committee V.6: Arctic technology
769
Introduction
1.1
Limitations
2 Present design methods
2.1
Ships
2.1.1 Rules
2.1.2 First principles
2.2
Offshore structures
2.2.1 Rules
2.3
Validation methods
3 Case 1: Ship transportation in arctic watersthe NSR
4 Case 2: Floating offshore structures in arctic waters
5 Future perspectives and challenges
5.1
Numerical simulations
5.2
Ice induced fatigue
6 Summary and recommendations
Acknowledgments
References
771
772
772
772
773
776
780
783
784
788
790
793
795
797
799
801
802
802
Report of Committee V.6: Arctic technology annex
807
1
2
3
Brief offshore structures code summaries

Full scale ice load measurement campaigns
References
809
813
816
Report of Committee V.7: Structural longevity
817
820
820
820
821
821
821
823
Introduction
1.1
Background & mandate
1.2
Relationship with other ISSC committees
Lifecycle assessment & management for structural longevity
2.1
Introduction
2.2
The need for lifecycle assessment and management
2.3
Conclusions
Prelims_Vol-1.indd xxi
7/31/2015 10:57:25 AM
xxii
Table of contents
Current practice
3.1
Introduction
3.2
The role of regulators and classification societies
3.3
Classification rules and guidance
3.4
Commercial shipping vessels
3.4.1 International trading vessels
3.4.2 High-speed craft (HSC)
3.4.3 Vessels operating in inland waterways
3.5
Offshore structures
3.5.1 Offshore drilling units
3.5.2 Floating production storage and offloading (FPSO) units
3.5.3 Fixed production platforms
3.6
Naval vessels
3.7
Conclusions
Prediction of longevity
4.1
Introduction
4.2
Prediction of longevity of merchant ships
4.2.1 Prediction of corrosion
4.2.2 Fatigue strength prediction
4.2.3 Buckling prediction
4.3
Prediction of longevity of fixed offshore structures
4.4
Conclusions
Prevention & repair of structural failures
5.1
Introduction
5.2
Prevention of failure design stage
5.2.1 Corrosion protection
5.2.2 Material selection
5.2.3 Structural design
5.3
Prevention of failure operation
5.3.1 Maintenance & inspection
5.3.2 Repair and rehabilitation
5.4
Inspection methods & techniques
6.1
Introduction
6.2
Inspection execution
6.3
Inspection techniques
6.4
Limitations
6.5
Sensing technologies
7.1
Introduction
7.2
Passive systems
7.2.1 Strain
7.2.2 Acoustic emission
7.2.3 Vibrations
7.2.4 Crack
7.2.5 Corrosion
7.2.6 Acceleration
7.2.7 Metocean information
7.3
Active system
7.3.1 Impedance-based methods
7.3.2 Lamb wave-propagation methods
Prelims_Vol-1.indd xxii
823
823
823
824
825
825
826
826
826
826
827
827
827
828
828
828
828
829
829
830
830
830
831
831
831
831
832
832
833
833
834
836
836
836
837
837
838
839
839
839
840
840
840
841
841
841
841
842
842
842
843
7/31/2015 10:57:25 AM
Table of contents
xxiii
7.4
Data acquisition and processing
7.5
Sensor network, wired and wireless
7.6
Maturity of structural hull monitoring systems
8
Methodologies for using inspection & sensed data
8.1
Introduction
8.2
Operational advice
8.2.1 Identifying loading to stay within safe operating envelope
8.2.2 Quantifying operational loading and changes
8.3
Lifecycle management advice
8.3.1 Condition based maintenance (CBM)
8.3.2 Reliability centered maintenance
8.3.3 Reliability based inspections
8.4
Design update based on lessons learned from analysis of failures
8.5
Discussion
8.6
Conclusions
9
Life time extension, comparison outside & within the maritime industry
9.1
Introduction
9.2
Lifetime extension of existing structures
9.3
Other industries
9.4
Differences in approaches for ships, offshore structures, and other
marine structures (ranging from navy to renewable energies)
9.5
Conclusions
10 Conclusions & recommendations
10.1
Conclusions
10.2
Recommendations
References
844
844
844
845
845
846
846
848
848
850
850
850
851
851
851
852
852
852
854
Report of Committee V.8: Risers and pipelines
865
1
2
867
867
867
867
868
868
868
869
870
871
871
871
873
878
878
879
879
879
880
880
881
882
882
883
Introduction
New design concepts
2.1
Latest design practice of flexible risers
2.1.1 Present application envelope
2.1.2 Deep water
2.1.3 Shallow water
2.1.4 Singing risers
2.1.5 Hybrid towers
2.2
Latest design practice of pipeline
Dynamic response investigation review
3.1
Riser
3.1.1 Wave load induced dynamic response
3.1.2 VIV
3.2
Free span VIV of pipeline
3.2.1 Assessment
3.2.2 Mitigation
Soil-pipeline interaction
4.1
Introduction
4.2
Soil behavior near pipelines
4.3
Pipeline as-laid embedment and riser touchdown
4.4
Lateral pipe-soil interaction
4.5
Axial pipe-soil interaction
4.6
Pipeline stability during trenching and backfilling
4.7
Pipeline stability during sediment transport and liquefaction
Prelims_Vol-1.indd xxiii
855
855
856
856
856
857
7/31/2015 10:57:25 AM
xxiv
Table of contents
Failure modes of risers and pipelines

5.1
Steel riser and pipelines
5.1.1 Buckling (buckle propagation), collapse and fatigue failure
5.1.2 Corrosion
5.1.3 Crack
5.1.4 Erosion
5.2
Flexible pipes
5.2.1 Failure modes
5.2.2 Design analysis
5.2.3 Monitoring
6
Installation
6.1
Risers
6.2
Pipelines
7
Inspection and repair
7.1
Risers
7.2
Pipelines
7.2.1 Maintenance
7.2.2 Inspection
7.2.3 Repair
8
Conclusions
References
884
884
884
885
886
886
886
886
886
887
888
888
888
889
889
891
891
891
892
893
895
Author index
903
Prelims_Vol-1.indd xxiv
7/31/2015 10:57:25 AM
Preface
The first volume contains the eight Technical Committee reports presented and discussed at the 19th
International Ship and Offshore Structures Congress (ISSC 2015) in Cascais, Portugal, 710 September
2015 and the second volume contains the reports of the eight Specialist Committees. The Official discussers reports, all floor discussions together with the replies by the committees will be published after the
Congress in electronic form.
The Standing Committee of the 19th International Ship and Offshore Structures Congress comprises:
Chairman:
Secretary:
Carlos Guedes Soares

Jrgen Amdahl
Yoo Sang Choo
Weicheng Cui
Segen Estefen
Stefano Ferraris
Wolfgang Fricke
Masahiko Fujikubo
Mirek Kaminski
Merv Norwood
Jeom Kee Paik
Jean-Yves Pradillon
Manolis Samuelides
Ajit Shenoi
Christina Wang
Yordan Garbatov
On behalf of the Standing Committee, we would like to thank DNV-GL, ClassNK (Nippon Kaiji Kyokai),
ABS (American Bureau of Shipping), CCS (China Classification Society), KR (Korean Register), and LR
(Lloyds Register) for sponsoring ISSC 2015.
Carlos Guedes Soares
Chairman
Yordan Garbatov
Secretary
Cascais, September, 2015
Prelims_Vol-1.indd xxv
7/31/2015 10:57:25 AM
19th INTERNATIONAL SHIP AND

OFFSHORE STRUCTURES CONGRESS
710 SEPTEMBER 2015
CASCAIS, PORTUGAL
VOLUME 1
COMMITTEE I.1
ENVIRONMENT
COMMITTEE MANDATE
Concern for descriptions of the ocean environment, especially with respect to wave, current and wind, in
deep and shallow waters, and ice, as a basis for the determination of environmental loads for structural
design. Attention shall be given to statistical description of these and other related phenomena relevant to
the safe design and operation of ships and offshore structures. The committee is encouraged to cooperate
with the corresponding ITTC committee.
COMMITTEE MEMBERS
Chairman:
E. Bitner-Gregersen, Norway
S. K. Bhattacharya, India
Z. Cherneva, Portugal
S. Dong, China
T. Fu, USA
G. Kapsenberg, The Netherlands
N. Ma, China
C. Maisondieu, France
R. Miyake, Japan
A. John Murphy, UK
I. Rychlik, Sweden
KEYWORDS
Environment, ocean, wind, wave, current, sea level, ice, deep water, shallow water, data source, modelling,
rogue waves, climate change, design condition, operational condition, uncertainty.
ISSC committee I.1: ENVIRONMENT
CONTENTS
1.
INTRODUCTION ......................................................................................................................... 4
2.
ENVIRONMENTAL DATA ........................................................................................................ 5

2.1 Wind .................................................................................................................................... 6
2.1.1 Locally sensed wind measurements ...................................................................... 6
2.1.2 Remotely sensed wind measurements .................................................................. 7
2.1.3 Numerical modelling to complement measured data ........................................... 8
2.2 Waves .................................................................................................................................. 8
2.2.1 Locally sensed wave measurements ..................................................................... 9
2.2.2 Remotely sensed wave measurements ................................................................ 12
2.2.3 Numerical modelling to complement measured data ......................................... 13
2.2.4 Wave description from measured ship motions ................................................. 14
2.3 Current............................................................................................................................... 14
2.3.1 In-situ current measurements .............................................................................. 14
2.3.2 Remotely sensed current measurements ............................................................. 15
2.4 Sea water level .................................................................................................................. 15
2.4.1 Locally sensed sea water level measurements .................................................... 15
2.4.2 Remotely sensed sea water level measurements ................................................ 15
2.5 Ice and snow...................................................................................................................... 15
2.5.1 Locally and remotely sensed ice and snow measurements ................................ 15
3.
ENVIRONMENTAL MODELS................................................................................................. 17
3.1 Wind .................................................................................................................................. 17
3.1.1 Analytical description of wind ............................................................................ 18
3.1.2 Statistical and spectral description of wind ........................................................ 18
3.2 Waves ................................................................................................................................ 20
3.2.1 Analytical and numerical wave models .............................................................. 20
3.2.2 Experimental description of waves ..................................................................... 28
3.2.3 Statistical description of waves ........................................................................... 30
3.2.4 Spectral description of waves .............................................................................. 32
3.3 Current............................................................................................................................... 33
3.3.1 Analytical description of current ......................................................................... 33
3.3.2 Statistical and spectral description of current ..................................................... 34
3.4 Sea water level .................................................................................................................. 34
3.5 Ice and snow...................................................................................................................... 34
4.
CLIMATE CHANGE.................................................................................................................. 34
4.1 New IPPC Scenarios and climate models ........................................................................ 35
4.1.1 Temperature ......................................................................................................... 36
4.1.2 Ice and snow ........................................................................................................ 37
4.1.3 Sea water level ..................................................................................................... 38
4.1.4 Wind and waves................................................................................................... 38
4.1.5 Ocean circulation ................................................................................................. 40
5.
SPECIAL TOPICS ...................................................................................................................... 40

5.1 Hurricane ........................................................................................................................... 40
5.2 Wave current interaction .................................................................................................. 41
5.2.1 Wave-current Interaction Model ......................................................................... 41
5.3
5.2.2 Numerical and Analytical Method ...................................................................... 43

5.2.3 Experiments and Measurements ......................................................................... 44
Wave and wind energy resource assessment ................................................................... 45
6.
DESIGN AND OPERATIONAL ENVIRONMENT ................................................................ 47

6.1 Design ............................................................................................................................... 47
6.1.1 Met-Ocean Data ................................................................................................... 47
6.1.2 Design Environment ............................................................................................ 48
6.1.3 Design for Climate Change and Rogue Waves .................................................. 51
6.2 Operations ......................................................................................................................... 52
6.2.1 Planning and executing marine operations ......................................................... 53
6.2.2 Northern Sea Route, Weather routing, Warning Criteria and Current .............. 54
6.2.3 Eco-Efficiency Ship Operation ........................................................................... 56
7.
CONCLUSIONS ......................................................................................................................... 57
7.1 Advances ........................................................................................................................... 59
7.2 Recommendations............................................................................................................. 60
ACKNOWLEDGEMENTS ............................................................................................................... 60
REFERENCES .................................................................................................................................... 61
1.
INTRODUCTION
This report is built upon the work of the previous Technical Committees in charge of Environment. The aim
is to review scientific and technological developments in the field since the last Committee, and to set them
in the context of the historical developments, in order to give a practicing engineer a balanced, accurate and
up to date picture about the natural environment as well as data and models which can be used to
approximate it in the most accurate way. The content of the present report also reflects the interests and
fields of competence of the Committee membership.
The mandate of the 2012 ISSC I.1 Committee has been adopted. It accords ice an equal status with
traditional interests such as wind, wave, current and sea water level, and recognizes the importance of
environmental data to the planning of marine operations and prediction of operability. Also in accordance
with the ISSC I.1 mandate, this Committee has reported on the resources available for design and the
operational environment. Additionally, the Committee has continued the initiated in 2010 cooperation with
the corresponding ITTC Committees.
The renewable energy installations are not mentioned explicitly in the Committee I.1 mandate as the
ISSC 2015 Committee V.4 Offshore Renewable Energy is addressing the topic. However, the increased use
of renewable energy sources has encouraged the Committee I.1 to put some attention to these issues,
focusing on metocean description only. The Committee I.1 would like to suggest extending, in
communication with the Committee V.4, the I.1 mandate in the future by giving renewable energy
installations an equal status with ships and offshore structures.
The Committee consisted of members from academia, research organizations, research laboratories and
classification societies. The Committee met four times: in Lisbon (1819 February 2013), Shanghai (910
December 201), San Francisco (8 June 2014) and in Hvik (1314 November 2014). Committee members
also met on an ad hoc basis at different scientific conferences and industrial workshops. The Committee I.1
contributed, together with the ISSC 2015 I.2 (Loads) Committee and the ITTC Ocean Engineering
Committee, to the organized by the ITTC Seakeeping Committee the 2nd ITTC-ISSC Joint Workshop on
uncertainty modelling which took place 30 August 2014 in Copenhagen.
The organisation of this report is an evolution of the outline used by the preceding Committee in their
report to the 18th ISSC Congress. Section 2 focuses on sources of environmental data for wind, waves,
current, sea water level and ice (including snow). Section 3 addresses modelling of environmental
phenomena while Section 4 climate change. Section 5 discusses some selected special topics. The design
and operating environment is presented in Section 6. The most significant findings of the report are
summarised in Section 7.
Three areas are considered as particularly important fields at the present time and have been selected for
special attention: hurricanes, wave-current interaction and resource assessment.
Rogue waves have been a topic of increasing interest over the past two decades. Two international
projects ShortCresT and EXTREME SEAS dedicated to these waves have been completed during the period
of the 2015 ISSC I.1 Committee. Following the two previous Committees this Committee felt that the rogue
waves could be adequately dealt with inside the normal wave sections: the wave data section (2.2) and wave
modelling section (3.2).
Major conferences held during the period of this Committee include the 31st33th International Ocean,
Offshore and Arctic Engineering (OMAE) conferences held in Rio de Jainerio (2012), Nantes (2013), San
Francisco (2014), and the 22nd24th International Offshore and Polar Engineering (ISOPE) conferences
held in Rhodes (2012), Alaska (2013) and Busan (2014). Also of great interest to the Committee were: the
13th International Workshop on Wave Hindcasting and Forecasting held in Banff (2013), the
MARSTRUCT (International Conference on Marine Structures) conference which took place in Espoo
(2013), the EUG (European Geosciences Union) conference in Vienna (2012, 2013, 2014), WISE
(Waves in Shallow Water Environment) in Washington (2013) and Reading (2014), COST (Predictive
Power of Marine Science in a Changing Climate) in Sopot (2014), the European Safety and Reliability
conference (ESREL) in Amsterdam (2013) and Wroclaw (2014), POAC (Port and Ocean Engineering under
Arctic Conditions) in Espoo (2013), IWMO (International Workshop on Modelling the Ocean) in
Yokohama (2012) and Bergen (2013), respectively, TRA (Transport Research Area) in Paris (2014),
MARTECH (International Conference on Marine Technology and Engineering) conference in Lisbon
(2014) and the U.S. Department of Energy Wave Energy Converter Extreme Conditions Modeling (ECM)
Workshop in Albuquerque, NM (2014). Works on resource assessment for Marine Renewable energy were
reported at EWTEC 2013 (European Wave and Tidal Energy Conference) in Aalborg (2013). Papers from
those sources have been reviewed and those of particular relevance are cited here. The articles published in
journals and conference proceedings not available to the Committee in the final forms by March 2015, are
not covered by the present review.
A number of Joint Industry Projects (JIPs) are also contributing to the world's knowledge base on the
metocean environment, with results released publicly in the form of academic papers. Several EU, JIP and
ESA (European Space Agency) projects have reported during the course of this Committee, including:
EXTREME SEAS, ShortCresT (both on extreme and rogue waves), HAWAII and LoWish (both on shallow
water), NavTronic (ship routing), SAFE OFFLOAD (LNG terminals), SHOPERA (energy efficient safe
ship operations) and DeepStar (metocean processes). A number of hindcast projects have also been in
operation or results from them have been reported during the period of this Committee, notably GROW2012
(global), GROW-Fine Northern Indian Ocean, GROW-Fine Mediterranean Sea, GROW-Fine Sea of
Okhotsk, the GROW Fine Caribbean (Caribbean Sea), the GROW Fine North Atlantic Basin, NAMOS
(NW Australia), SNEXT (North Sea), SEAFINE (SE Asia), BOMOSHU (Brazil, Atlantic waters), WASP
(WesT Africa Swell Project), and a Chinese national project in the South China Sea. The present status of
the ESA GlobWave project, making satellite derived data more widely available, is also reviewed; a
summary of services can be found on http://globwave.ifremer.fr /Products/Summary-of-Services.
Success of the global and basin-scale ocean models development with data assimilation under the
GODAE (Global Ocean Data Assimilation Experiment) program, initiated some years ago, opened a new
era of operational oceanography. This program ended in 2008 and has continued as GODAE OceanView
(https://www.godae-oceanview.org/). The 5 years of GODAE OceanView progress and priorities were
presented at the GODAE OceanView Symposium in Baltimore in 2013.
Climate change has also been a topic of continuing worldwide interest both regarding mitigation as well
as adaptation process. It has had impact on research activities within the shipping, offshore, emerging
renewable energy and coastal engineering industry sectors and the need to adaptation to climate changes is
getting increasing recognition within these sectors. The previous Committee reviewed this subject as a
special topic while the current Committee has addressed it in a separate Section 4 on climate change. During
the period of the 2015 ISSC I.1 Committee the 5th Assessment Report of the Intergovernmental Panel on
Climate Change (IPCC. 2013) has been issued and is reviewed herein. The present report makes an attempt
to provide the ISSC Congress with the most up-to-date information from leading scientists on the main
climate change issues of relevance to those working on the seas: temperature, sea ice extent, sea-level rise
and storm intensity and frequency. Particular attention is given to the Arctic environment and to tropical and
extra-tropical hurricanes and related wave climate.
Enhancing safety at sea through specification of uncertainties related to environmental description is
being increasingly recognized by the shipping, offshore, emerging renewable energy and coastal
engineering industries. Organization of the 2nd ITTC-ISSC Joint Workshop on uncertainty modelling is
confirming the importance of the topic. Generally, uncertainty related to wave description may be divided
into two groups: aleatory (inherent, intrinsic) uncertainty and epistemic (knowledge based) uncertainty
(Bitner-Gregersen et al., 2014a). Aleatory uncertainty represents a natural randomness of a quantity, also
known as intrinsic or inherent uncertainty, e.g. the variability in wave height over time. Aleatory uncertainty
cannot be reduced or eliminated. Epistemic uncertainty represents errors which can be reduced by collecting
more information about a considered quantity and improving the methods of measuring it. Recent scientific
and technological developments in the field of environment are presented in the report in the perspective of
these uncertainties.
The report is covering a wide ranging subject area and limited space as well as the boundaries presented
by the range of specialisms and competencies of the Committee members, this Committee report cannot be
exhaustive. However, the Committee believes that the reader will gain a fair and balanced view of the
subjects covered and we recommend this report for the consideration of the ISSC 2015 Congress.
2.
ENVIRONMENTAL DATA
Wind, ocean waves, current, sea water level, ice and snow conditions vary geographically and in time.
Physical, probabilistic and statistical models can approximate this variability. Environmental data
represent an important contribution to modelling of environmental phenomena. They can be collected by
in-situ instruments, remote sense techniques and/or generated by a model. Environmental data are
affected by measurement, statistical (sampling variability) and model uncertainties (Bitner-Gregersen
et al., 2014a), which are not fully quantified today. A question getting increasing attention in the last
years is: Are these measurements actually ground truth? This question was also raised up during the 13th
International Workshop on Wave Hindcasting and Forecasting and 4th Coastal Hazards Workshop taken
place in Banff, Canada (Jensen et al., 2013), but a final answer to it still does not exist.
The issue of data ownership remains a general problem (ISSC, 2009, ISSC, 2012, Bitner-Gregersen
et al., 2014a) even though some progress regarding access to the environmental data has been made since
2012. This makes work on comparison of different data sources and specification of uncertainties related to
them difficult, and consequently specification of the ground truth even more challenging. The data are often
of proprietary naturefor example, oil companies, ship owners, and agencies usually keep their data
confidential. In some cases, government agencies make data freely available in the public domain,
such as the NOAA, NIBCO data sources, but this is the exception rather than the rule. An example of
making data available without compromising their confidentiality is the SIMORC URL data base:
http://www.simorc.org/, administered by the University of Southampton (ISSC, 2012).
Also in 2010 TOTAL Oil & Gas Operator launched a project to give remote and public access to realtime wind, current and wave, or other metocean data monitored from many oil and gas platforms offshore
West and Central Africa (from Nigeria to Angola). Since 2013, with the support of the French
Meteorological Office Mto-France, the data from half dozen platforms offshore Nigeria, Congo and
Angola have become available on the World Meteorological Organizations (WMO) Global
Telecommunication System (GTS). (Quiniou-Ramus et al., 2013) present the type of metocean stations that
are part of this network (MODANET), the IT architecture that was selected to send the data out of the
TOTAL Companys network, the quality control undertaken by Mto-France before sending the data to the
GTS, and discuss future possible use of the data that are envisaged.
The stationarity and homogeneity assumption of measurements is obviously questionable and likely not
valid in some circumstances. It is getting increasing focus in academia and the marine and renewable energy
industries, e.g. (Ewans, 2014). The need to account for non-stationarity and non-homogeneity of
environment data is expected to continue, particularly because of changing climate but also due to needs of
engineering applications. Although several data uncertainties have been reported during the period of the
ISSC 2015 Committee I.1 a systematic investigation of them still is lacking.
2.1
Wind
Meteorological data of good quality are important not only for understanding of global and regional
climates but also for specification of design and operational criteria of ship, offshore and renewable energy
structures. Local measurements of the wind, traditionally at 10 m height above the sea surface, have been
the standard way to record wind characteristics for decades and remain important particularly for
verification of data from other sources. But as suitable measurement sites are scarce, and it is not possible to
enlarge this number significantly, the advent of remote measurement techniques and numerical simulations
has allowed for much more detailed descriptions of wind in the offshore data.
Apart from wind speed also wind direction, wind profile (describing variations of the mean wind speed
with height above the ground or above the sea water level), gust, wind spectrum and squalls represent
important characteristics of a wind field which can be determined from wind data, see (DNV, 2014).
2.1.1
Locally sensed wind measurements
Large and meso-scale wind fields have been studied for years leading to a wide variety of wind field data.
These measurements have been either focused on short term detailed observations with attention on specific
meteorological and oceanographic mechanisms or on longer term measurements of statistical behavior and
have been used to support weather forecasting. More recently with the increased activity in coastal regions a
number of efforts have established offshore observation capability providing a valuable source of
environmental data of all types. The U.S. Department of Commerces National Oceanographic Data Center
(NODC) is one of the national environmental data centers operated by the National Oceanic and
Atmospheric Administration (NOAA). They are part of the World Data Center System initiated in 1957 to
provide a mechanism for data exchange, and they operate under guidelines issued by the International
Council of Scientific Unions (ICSU). There are three World Data Centers for Oceanography:
World Data Center, Silver Spring, Maryland, United States

World Data Center, Moscow, Russia
World Data Center, Tianjin, Peoples Republic of China
In-situ wind measurements are collected by buoys, ships and platforms. Perhaps the most well-known
organization collecting wind data is the U.S. National Oceanic and Atmospheric Administration (NOAA)
National Data Buoy Center (NDBC) (www.ndbc.noaa.gov). As a part of the National Weather Service
(NWS), the center designs, develops, operates, and maintains a network of approximately 90 data collecting
buoys and 60 Coastal Marine Automated Network (C-MAN) stations. For each of these buoys and C-MAN
stations the NDBC provides hourly observations from a network of all stations measuring wind speed,
direction, and gust; barometric pressure; and air temperature. In addition, all buoy stations, and some
C-MAN stations, measure sea surface temperature and wave height and period. Conductivity and water
current are also measured at selected stations.
The wind is often assumed to be stationary over one-hour, and relationships between specific parameters,
such as the ratio of the maximum 1-minute mean wind speed in the hour, are required to be specified.
Recommended relationships are available in codes, such as (ISO, 2012, DNV, 2014).
Muyau et al. (2014) investigated short-term effects in wind data, sampled at 1 Hz off the west coast of
Borneo, including mainly monsoon conditions. Their data set consisted of 5952 1-hour records collected at
three offshore locations and one onshore location. Using the stringent run test, they found none of their 1 Hz
data records to be stationary over one hour, and even when the 1 Hz data were averaged over longer
intervals, no records of 3-second means were found to be stationary, and only 14% of the records of 1minute means were found to be stationary. The authors concluded that the run test was not reliable for the
10-minute mean wind speeds over the one hour, due to the small data set. The results for the original data
set and other averaging intervals, which were almost entirely non-stationary over one hour for their data
sets, indicate that the application of fixed short-term wind relationships is questionable.
The lack of stationarity in the wind measurements has long been recognised for highly transient wind
events, such as squalls. Bitner-Gregersen et al. (2014b) show the wind speeds measured during a major
North Atlantic Storm and the other during a squall event demonstrating clearly the transient nature of the
squall wind speeds.
Within the 15-year JIP DeepStar project all available hurricane wind data sets made in and around the
Gulf of Mexico since 1998 have been collected, quality control, and then analyzed (Cooper et al., 2013).
The data sets included offshore platform anemometer records, measurements from National Oceanic and
Atmospheric Administration (NOAA) buoys, Coastal-Marine Automated Network (C-MAN), Automated
Surface Observing System (ASOS), and National Ocean Service (NOS) stations, tower arrays of
anemometers deployed along the coast, coastal weather radars, and dropsonde observations made by
hurricane hunter aircraft. The aim of these investigations has been improving modelling of hurricanes based
on the historiocal data with reasonable statistical uncertainty (see also Section 6.1.2).
New needs for a detailed description of wind profiles and turbulence at regional and local scales, mostly
required by the developing wind offshore industry, appear to play a major role in the development of new
sensors today as well as the implementation of downscaled numerical models. The offshore wind industry
needs data on suitable locations for the installation and changes in the wind profile (beyond 200 m) as well
as spatial distribution of wind characteristics. These data are still very spare today.
2.1.2
Remotely sensed wind measurements
Remotely sensed surface wind data is available from the U.S. National Oceanic and Atmospheric
Administrations (NOAA) Center for Satellite Applications and Research (STAR). Both active (radar) and
passive (radiometer) microwave sensors are capable of retrieving ocean surface wind speed, and with active
microwave instruments being used to also retrieve the wind direction. The development and refinement of
instrumentation and algorithms for ocean surface wind retrieval is an ongoing process being conducted in
both the active and passive remote sensing areas. STARs Ocean Surface Winds Team (OSWT) web site
(http://manati.star.nesdis.noaa.gov/products.php) provides: wind vector fields and wind speed fields.
Additionally the STARs web site provides rain, sea ice, SST and water vapor data.
Information on specific storms as well as storm forecast data can be found at
http://www.nrlmry.navy.mil/tc_pages/tc_home.html. The web page was created to provide remote sensing
imagery and data sets derived from both geostationary and polar orbiter sensors. The limitation of the data
sets from the web site is that the data sets are updated automatically in near real-time generating the data
products, updating storm positions, adding new storms and deleting storms that have decayed and are no
longer active. The available data is global in nature and includes the standard visible/IR and water vapor
geostationary imagery in addition to passive and active microwave data.
Various remote sensing databases have been updated and made available at CERSAT
(http://cersat.ifremer.fr/) thanks to the new cloud computing facility Nephelae.
The complete ERS-1 & ERS-2 altimeter data archive from 1991 to 2003 has been reprocessed in the
context of ESA REAPER project. The ERS-1/2 REAPER Altimeter dataset is composed of the following
three product types which are freely accessible: GDR, the RA Geophysical Data Record product containing
radar range, orbital altitude, wind speed, wave height and water vapor from the ATSR/MWR as well as
geophysical corrections; SGDR, the RA Sensor Geophysical Data Record (SDGR) product containing all of
the parameters found in the REAPER GDR product (ERS_ALT_2_) with the addition of the echo waveform
and selected parameters from the Level 1b data; and the RA Meteo product containing only the 1 Hz
parameters for altimeter (surface range, satellite altitude, wind speed and significant wave height at nadir)
and ATSR/MWR data (brightness temperature at 23.8 GHz and 36.5 GHz, water vapor content, liquid water
content) used to correct altimeter measurements. It also contains the full geophysical corrections. Major
improvements with respect to the previous ESA RA products format (OPROcean Productand WAP
Waveform product) have been implemented (e.g. the 4 Envisat RA-2 retrackers, RA calibration
improvement, new reprocessed Precise Orbit Solution, ECMWF ERA-interim model, NICO09 ionospheric
correction until 1998, GIM ionospheric correction up to 2003, new SSB, etc.). The assessment of the
REAPER data quality versus the ERS OPR and WAP data shows a clear improvement in terms of accuracy
over the tandem periods between ERS-1, ERS-2 and Envisat missions (currently assessed periods).
However, the REAPER dataset presents some limitations (such as the use of poor MWR Wet tropospheric
correction, out of range PTR corrections, etc.) that are fully described in the Product Handbook.
Gridded daily wind vector and wind stress fields, estimated over global ocean from QuikSCAT
scatterometer (referred as DQSCAT) data, have been updated in 2013. Their spatial resolution is 0.25 in
longitude and in latitude. They are produced from the new QuikSCAT wind retrievals indicated as
QuikSCAT V3 (ftp://podaac.jpl.nasa.gov/OceanWinds/ quikscat/preview/ L2B12 /v3/). Wind retrievals are
provided over QuikSCAT swath at Wind Vector Cell (WVC) of 12.5km spatial resolution. The new
scatterometer product is assumed improving wind speed performance in rain and at high wind conditions.
The calculation of daily gridded wind fields from scatterometer wind observations is performed using same
objective method used for the estimation of daily ASCAT wind fields (DASCAT) (Bentamy et al., 2011).
The resulting wind field accuracy is investigated trough the comparisons with daily-averaged winds from
moored buoys. The overall statistics indicate that the daily scatterometer wind fields compare well to dailyaveraged buoy data. The rms differences do not exceed 2m/s and 20 for wind speed and direction,
respectively. Despite of difference in buoy and scatterometer sampling schemes used for the estimation of
daily winds, correlation values attest that satellite daily winds reproduce fairly well in-situ estimates.
Analyzing a 5-year dataset collected over two surface current and meteorological moorings, Plagge et al.
(2012) investigated the influence of surface currents on satellite scatterometer and altimeter ocean winds.
Comparing wind residuals between Ku-band Quick Scatterometer (QuikSCAT) and buoy measurements
they observed that scatterometer winds and buoy wind direction differences due to currents were negligible
for the range of surface velocities encountered and the length scales observed by QuikSCAT. As a
consequence; at length scales of 10 km and longer the scatterometer wind can be considered to be current
relative and not earth relative. Observed differences between earth-relative and current-relative winds of
order 10%20% of the wind velocity are not uncommon in the considered area and other ocean regions and
this study more fully validates that microwave remote sensing winds appear to respond to wind stress even
in the presence of larger-scale currents.
For further discussion of accuracy of satellite data see also (ISSC, 2009).
2.1.3
Numerical modelling to complement measured data
Numerically generated wind data are still commonly used in design and marine operations as well as
renewable energy applications. For some ocean areas they are the only data available. Although the number
of remote sense data is increasing (e.g. the GlobWave database) they not always exist for a location
considered in an engineering application. The numerical data refer usually as the 10-minute average wind
speed at the 10 m height above the ground or the still water level and include also wind direction. The wind
data can be converted to a different averaging period as well as to the different heights by appropriate
commonly used expressions; e.g. (DNV, 2014).
The recently updated or new developed metocean data bases such as: ERA-Interim (European
Reanalysis), http://www.ecmwf.int/en/research/climate-reanalysis, ERA-Clim (European Reanalysis of
Global Climate Observations) http://www.ecmwf.int/en/research/projects/era-clim, CFSR (Climate Forecast
System Reanalysis), https://climatedataguide.ucar.edu/climate-data /climate-forecast-system-reanalysis-cfsr,
NORA10 (Aarnes et al., 2012), HIPOCAS (www. mar.ist.utl.pt/hipocas/members_details.as), BMT-ARGOSS
(http://www.bmtargoss.com/met ocean-web-portals/wwwwaveclimatecom/) and Fugro-OCEANOR
(http://www.oceanor.no/ Services/Worldwaves/WW_database) include information about both wind and
waves. Further progress aiming at enhancing accuracy of these databases and/or extension of the time period
they covered has taken place since 2012. The improvement includes higher resolution, better quality-control
of assimilated data and/or improvement of validation procedures, see e.g. (Cardone et al., 2014).
2.2
Waves
Observations of waves in the open ocean still represent a challenge and they are limited. Most of wave
recordings take place in coastal areas. Therefore wave data from hindcast studies are the choice data sets for
development of design criteria of marine structures. However, measured wave data either locally or
remotely remain important for development, calibration, and validation of numerical wave models used for
generating hindcasts, particularly in coastal areas due to shallow-water aspects of wave dynamics. The
measured data are also important for providing description of individual wave characteristics in the open
ocean and coastal waters as well as validation of nonlinear short-term wave models. During the period of the
Committee new instrumental data sets including extreme waves have been collected/reported and further
improvements of hindcasts have taken place.
With the increased need and use of ocean wave measurements the question that remains to be addressed
is: Are these measurements actually ground truth? What are the uncertainties and limitations of the various
measurement systems? Apart from instrumental errors wave data are affected by statistical uncertainty due
to limited duration of wave records and model uncertainty associated with a method used for
derivation/generation of wave parameters. Additionally, whenever observations of the ocean wave
environment are made, the questions of stationarity and ergodicity need to be addressed.
Wave characteristics commonly used in applications include significant wave height, spectral (or zerocrossing) wave period, wave spectrum and wave directional spreading, see (DNV, 2014).
2.2.1
Locally sensed wave measurements
Wave buoys, a wave staff, radars, lasers, LASAR and a step gauge remain the most important sources of insitu measurements. Specific issues for the most common wave measurement systems include:
LidarFixed point measurements need to consider instrumentation accuracy in the range estimate
provided by the instrument. The absorption of water also needs to be accounted for. For free surfaces
with low void fraction this error is likely less than any uncertainty in the along range resolution. Typical
range resolutions are around +/2.5 cm and thus would likely only be of concern in very low sea states.
BuoyUncertainties in significant wave height Hs seem to be associated mostly with a question how
well the buoy tracks the free surface and whether non-linear effects are accurately measured.
Additionally issued related to specific installations may also be present. The buoy may cut the top
of the wave off, particularly if it is moored. One area that remains to be opened is what uncertainty
exists in estimation of wave direction. The accuracy is dependent on the number of degrees of
freedom and how the 2D spectrum is derived. Three DOF buoys have difficulty resolving directional
wave energy due to the poor directional resolution (again dependent on the processing method).
While they likely get the dominant wave direction, they will tend to smear energy in a given
frequency band if there are multiple systems that propagate at similar headings.
X-Band RadarThe dominant source of uncertainty for these systems comes from the calibration. If
there is an assumed linear relationship between Hs and the square of SNR (Signal-to-Noise Ratio),
then an issue is how is the spread in the fit applied to the measured Hs data? A universal uncertainty
cannot be applied for all radars since it is so dependent on the actual data used during calibration.
Nieto-Borge et al. (2008) shows a typical calibration curve. At a SNR of 1.5 the Hs values ranged
from 2.54 m. So while period and direction may be easily obtained from these systems, accurate Hs
measurements is still a topic for further research.
Significant wave height Hs and spectral (or zero-crossing) wave period Tp (Tz) represent important
parameters for design and operations of ships and offshore structures. They are used for validation of wave
models, wave climate studies and calculations of extremes for weather forecasting purposes. Whether
significant wave height Hs is determined using directly time series (H1/3), 4*standard deviation (std) of the
free surface or via calculation of the zeroth moment of the spectrum (M0) there are closed form expressions
for the statistical uncertainty that should be accounted for (Bitner-Gregersen and Hagen, 1990, BitnerGregersen and Magnusson, 2014). Also Hs computed via 4*std or by 4*sqrt(M0) wont necessarily be equal
due to windowing/overlapping of segments during calculation of the spectrum. Each system has a specific
frequency bandwidth it is able to measure, and no single system can measure the entire wave spectrum.
Further, as pointed out by (Bitner-Gregersen and Hagen, 1990) Hs computed via 4*std is only equal to the
one calculated via 4*sqrt(M0) if the sea surface is a narrow-banded Gaussian process.
In September 2013, a U.S. Office of Naval Research funded, nine-day experiment was conducted aboard
the research vessel R/V Melville, where the statistical and phase-resolved wavefield were measured using a
shipboard radar, airborne lidar, wave buoys, and a bow mounted lidar (Merrill et al., 2014). The measurement
area was just west of Southern California and around San Clemente Island and the Channel Islands (Figure
1). The local wave field was measured using the OceanWaves WaMoS II wave radar system coupled using
a post-processing routine developed by Scripps Institute of Oceanography (SIO) to obtain phase-resolved
results from the WaMoS radar intensity maps. Point measurements of the wave field were provided by SIO
miniature wave buoys and Datawell III Waverider buoys, both of which were modified to record buoy
motions at 1.0 Hz in addition to their normal statistical parameters. During three separate periods of the
cruise, an airborne lidar system provided five kilometer box sweeps of the wave field in the vicinity of the
ship. An experimental bow mounted wave measuring lidar system developed by SIO to record the wave
height in the vicinity of the ship (but outside of the ship generated wake) was also deployed. A small boat
with a wave measuring ultrasonic array system was also used when conditions allowed.
10
Figure 1. The R/V Melville operational area for the September 2013 field experiment.
Merrill et al. (2014) compare the wave measurements made from the various instrumentation systems and
notes that while comparison of the wave statistics shows that each system is correctly capturing the
generalized wavefield behavior, it also shows that significant work still needs to be done in regards to the
measurement of phase-resolved wave fields where much of the uncertainty due to registration in both time
and space is present.
Wave measuremets recorded by buoy and radar installed 6 km apart in a deep water region offshore
northeast Brazilian shelf were compared by Ribeiro et al. (2013). Differences in wave directional spectral
parameters calculated by both equipments were found. The Hs correlation was 91% and bias 0.06 m. The
radar Hs compared to the buoy Hs was overestimated in high sea states and underestimated in low ones. The
Tp correlation was 69% and bias 0.02 s. The peak direction was the worst correlated, 60% correlation was
found for E-ESE directional band. Time series matched each other showing the typical regional wave
climate for the area.
A new wave data processing system allowing deeper evaluation of the information that was stored inside
a wave buoy and was not transmitted in real-time was presented by Pereira et al. (2012). The main aims of
this study has been to increase the reliability of real-time data transmitted by heave-pitch-roll buoys and
verify the efficiency of a single board computer to execute the traditional wave processing including
automatic quality control.
Gemmrich and Garrett (2012) investigated the influence of inertial current on sea-states from offshore
buoy measurement pointing out that wave-current interaction is inducing wave height modulation. They
suggest that these interactions be taken into account in hindcast wave models.
An interesting element regarding wave measurement is the development of video processing techniques
which provide new and interesting insight in the assessment of wave characteristics and breaking.
Fedele et al. (2013) use stereo imaging techniques to identify the space-time evolution of the sea surface.
Creating data series of sea surface maps they analyze the characteristics of large waves. This study revealed
that the maximum wave surface height over an area during a given duration (spacetime extreme) is larger
than that expected at a given point in space (time extreme). If the area is large enough compared to the mean
wavelength, a spacetime extreme most likely coincides with the crest of a focusing wave group that passes
through the area.
Schwendeman et al. (2014) investigated energy dissipation in young wind sea by mean of in-situ
measurement. Their main conclusions are that there is a strong correlation between wave breaking
dissipation and the mean square slope of the waves, both of which increase along fetch. Video-derived
breaking rates and breaking crest distributions also increase with mean square slope. Conducting error
analysis they suggest that many bulk breaking parameter values from various recent field experiments are
likely biased by subtleties of video collection and processing.
A number of efforts have focused on improving the performance of wave radars. This work has
concentrated on both improved accuracy in measuring wave spectra and in measuring phase resolved wave
fields. Although techniques to extract wave parameters from radar measurements have been evolving over
11
the past several decades (Young et al., 1985, Nieto-Borge et al., 2008) along with our understanding of the
scatterers that contribute to sea clutter (Long, 2001), they still have limited accuracy and reliability even
under idealized conditions (Johnson et al., 2009). The primary reason is due to the large number of factors
that sea clutter depends on, both radar parameters as well as sea characteristics.
A number of these efforts have examined the use of coherent radar systems. These systems include both
high transmit power systems capable of long range and coherence to allow for Doppler processing and
systems that transmit non-coherently but are coherent on receive. Smith et al. (2013) describes the
development of a low cost, high power coherent on receive radar for making sea surface measurements.
Hacket et al. (2014) compare wave field measurements from incoherent and coherent measurements from a
dual-polarized pulse-Doppler X-band radar to examine the sensitivity of the extracted wave parameters to
the characteristics of the radar and the scatterers. These experiments were performed offshore of the Scripps
Institution of Oceanography pier in July 2010. Radar measurements in low wind speeds were performed
with dual-polarized high-resolution X-band pulse-Doppler radar at low grazing angles along with two
independent measurements of the surface waves using conventional sensors, a GPS-based buoy and an
ultrasonic array.
Comparison between radar cross section (RCS) and Doppler modulations show peak values occurring
nearly in-phase, in contrast with tilt modulation theory. Spectral comparisons between Doppler-based and
RCS-based spectra show that Doppler-based spectra demonstrate a greater sensitivity to swell-induced
modulations, while RCS-based spectra show greater sensitivity to small-scale modulations (or generally
have more noise at high frequency), and they equally capture energy at the wind wave peak. Doppler
estimates of peak period were consistent with the conventional sensors, while the RCS differed in
assignment of peak period to wind seas rather than swell in a couple cases. Higher-order period statistics of
both RCS and Doppler were consistent with the conventional sensors. Radar-based significant wave heights
are lower than buoy-based values, and contain nontrivial variability of ~33%. Comparisons between HH
and VV polarization data show VV data more accurately represents the wave field, particularly as the wind
speeds decreases.
It is interesting to note that new very high significant wave heights have been registered/reported since
2012, see (Cardone et al., 2014). The K-5 buoy in the eastern North Atlantic recorded a new high
measurement of HS = 19.0m on 4 February, 2013. The authors report in addition several cases of satellite
altimeter estimates of HS 20 m in the North Atlantic and the North Pacific.
Several rogue waves recorded in the ocean have also been reported in the period of this Committee.
During the Andrea storm, which crossed the central part of the North Sea on November 8th-9th, 2007 storm,
on November 9th, 2007 a rogue wave called Andrea was measured at the Ekofisk field (Magnusson and
Donelan, 2013). This wave is comparable in characteristics to the well-known New Year wave (called also
the Draupner wave) recorded by Statoil at the Draupner platform the 1st January 1995, see Table 1. Cmax in
Table 1 denotes the maximum crest height in the wave record, Hmax the maximum zero-downcrossing wave
height with the crest Cmax, CF is the maximum crest factor (crest criterion), HF is the maximum height factor
(height criterion). CF>1.3 (or >1.2 as suggested by Haver and Anderson (2000) and HF> 2 within a 20minute wave record represent simplified definitions of a rogue wave, see e.g. (Bitner-Gregersen and Toffoli,
2012a). If both criteria are fulfilled a rogue wave can be classified as a double rogue wave (Krogstad et al.,
2008). As seen in Table 1 both the New Year wave as well as the Andrea wave can be called the double
rogue wave. Note that both waves are recorded in the North Sea at the platforms located in the water depth
of ca. 75 m.
Table 1. Characteristics of the Andrea and New Year waves
derived from the ca. 20-minute wave records.
Wave parameters
Andrea wave
Draupner wave
9.2 m
11.9 m
Hs
13.2 s
14.4 s
Tp
15.0 m
18.5 m
Cmax
1.63
1.55
CF=Cmax/Hs
21.1 m
25.0 m
Hmax
2.3
2.1
HF=Hmax/Hs
Waseda et al. (2014) have reported extreme waves registered in 2009, 2012 and 2013 by point-positioning
GPS-based wave measurements conducted by deep ocean (over 5,000 m) surface buoys moored in the
North West Pacific Ocean. Two large rogue (freak) waves exceeding 13 m in height were observed in
October 2009 and three extreme waves around 20 m in height were observed in October 2012 and in
January 2013. These extreme events are associated with passages of a typhoon and a mid-latitude cyclone.
12
The two rogue waves recorded on 26 and 27 October 2009 had distinct directional characteristics, former
being narrow and latter being broad. On October 4, 2012, extreme waves of 22.8 m (H1/3 = 13.4 m), and 17.3
m wave height (H1/3 = 10.3 m) were registered during passage of typhoon while on January 14, 2013, an
extreme wave height 17.7 m (H1/3 = 10.0 m) was observed during passage of a bomb cyclone. These three
waves were not rogue waves with regard to maximum height factor criterion (HF > 2), the wave crest
criterion is not reported by Waseda et al. (2014). Up to 6% of observations collected in a few months in
2009, 2012 and 2013 represented rogue waves, but they were recorded in the intermediate sea states (the
average H1/3 = 2.75.6 m). The authors suggest that a direct measurement of extreme wave by GPS sensor
might become an attractive alternative for observing extreme waves offshore. The accuracy of such
measurement depends on how well the platform follows the wave motion.
Rogue waves have been also recorded by the wave buoy SBF3-1 in the sea area of mainland Jiangsu,
China (Wang et al., 2014) but the wave heights have not been high.
Wave records are often limited to 20-minutes and therefore parameters derived from these records are
affected by sampling variability, the statistical uncertainty due to limited number of observations. BitnerGregersen and Magnusson (2014) have provided estimates of sampling variability associated with
significant wave height and zero-crossing wave period based on measurements from the Ekofisk field in
central North Sea. The calculated sampling variability shows the same trend as the theoretical values due to
(Bitner-Gregersen and Hagen, 1990). The sampling variability is higher in Hs than in Tz and both increase
with increasing Hs and Tz. It is anticipate that sampling variability depent on the shape of a wave spectrum;
the JONSWAP spectrum gives higher variability than the Pierson-Moskowitz spectrum. The authors
demonstrated the impact of intrinsic and sampling variability on short-term and long-term description of
ocean waves as well as validation of wave spectral models. The intention of the study has been to put again
attention to intrinsic and sampling variability and to remind practitioners that sampling variability must be
taken into account for accurate use of wave measurements.
The limited duration of wind and waves time series has allowed adopting an assumption of stationarity on
which most of wind and waves models is based today. However, conditions such as e.g. wind-sea
developing as rapidly-moving tropical cyclones or hurricane passes will not be stationary (Ewans, 2014,
Bitner-Gregersen et al., 2014b). Ewans (2014) examined the stationarity (determined from the run test) of
12-months Directional Waverider data recorded at the US Corp of Army Engineers Field Research
Facility at Duck, North Carolina. He found the vast majority of the records were stationary up to 160 min.
Non-stationary records have been generally associated with changing wind-sea conditions occurring with
local wave growth.
2.2.2
Remotely sensed wave measurements
Investigations aiming at providing satellite wave products for users are continuously going on. The
GlobWave project (www.globwave.org), reported in the previous Committee I.1 Report, has been an
interesting initiative funded by the European Space Agency (ESA) to service the needs of satellite wave
product users.
Work was conducted over the years at CERSAT so as to provide relevant validated altimeter data sets.
As a major outcome of this work, altimeter significant wave height (SWH) measurements are presently
available almost continuously over a 20-year time period from the eight altimeter missions ERS-1&2,
TOPEX-Poseidon, GEOSAT Follow-On (GFO), Jason-1, Jason-2, ENVISAT and CryoSat. Each altimeter
data product has specific characteristics (format, flags), and in order to facilitate the access to SWH
altimeter measurements and the use of this long time series, data were extracted from the original products,
screened according to quality flag values, corrected and gathered into homogeneous daily data files
(Queffeulou, 2013).
SWH data from the CryoSat-2 IGDR data sets produced and provided by the NOAA Laboratory for
Satellite Altimetry (ftp://ibis.grdl.noaa.gov/pub/cs2igdr/), both low rate mode (LRM) and Pseudo LRM
were validated using comparisons with collocated altimeter measurements from Jason-1, Jason-2 and
ENVISAT RA-2. They were implemented in the data base. Additionally, preliminary results of the
validation work of the SARAL AltiKa, launched in February 2013, are provided by the author, showing a
very high accuracy of the AltiKa SWH (Significant Wave Height) measurement.
A well-known interest of remote sensing is the ability it offers to assess wave trains propagation across
oceanic basins. For instance Young et al. (2013) analyzed altimeter data from transects across the Southern
Ocean to determine the decay of oceanic swell. They observed that the decay rate is shown to be
proportional to wavenumber squared and swell amplitude cubed, confirming previous work by Ardhuin
(2009) and Babanin (2012). This decay relationship is consistent with turbulent interaction with the
13
background, either in the air or water and is in agreement with the limited previous studies. It presents a
source term suitable for use in wave prediction models.
However, coverage offered by satellite altimeters and space-borne Synthetic Aperture Radars is still
sparse and spectrally limited. In order to compensate for this, various authors investigate the possibility of
extracting relevant information from seismic noise data to complement the remote sensing
information(Ardhuin et al., 2012). Especially, Husson et al. (2013) analyze the signature of one swell event
in the seismic noise recorded all around the Pacific and show that it is a natural complement to the global
coverage provided by the Synthetic Aperture Radar wave mode data from ENVISAT. The great sensitivity
of seismometers to very long waves allowed revealing the presence of swell forerunners when arriving to
the coast, which by default are not detected by the SAR.
Analysis of the available SAR archives allows assessment of various specific features affecting wave
propagation. Using an archive of satellite ENVISAT ASAR images acquired for a period of about five
years, 20072011, over the White Sea and during periods when the water column was thermally stratified,
Koslov et al. (2014) identified and analyzed internal waves having a variety of spatial scales, propagation
directions and interpacket distances. Assumption that observed nonlinear internal waves group (IWs) are
generated at the consequent tidal cycles by the interaction of relatively strong barotropic tidal flow with the
frontal area located over the bank in the SW Gorlo Strait seems to be confirmed by the results of the
numerical model used to calculate the propagation of NIW packets generated in the SW Gorlo Strait which
agrees with the SAR observations and confirms the strong influence of M2 tidal cycles.
Using airborne and spaceborne interferometric synthetic aperture radars (InSARs) allowing surface
velocity measurements at very high spatial resolutions over a large area, Hwang et al. (2013) investigated
the breaking process over a coastal zone. Breaking can be detected by mean of various methods among
which the analysis of surface roughness decorelation. They show that the breaking fraction is strongly
correlated to the wind sea mean square slope, in agreement with previous observations showing that the
breaking length scale is considerably shorter than the dominant wavelength.
The 2013 Ocean Surface Topography Science Team (Willis and Bonnefond, 2013). Meeting was held in
Boulder, CO on October 8-11. The primary objectives of the OSTST Meeting were to provide updates on
the status of Jason-1 and OSTM/Jason-2, conduct splinter meetings on the various corrections and altimetry
data products, and discuss the science requirements for future altimetry missions.
2.2.3
Numerical wave models used for forecasting or building hindcast databases are under constant evolution
(see Section 3.2.3). Hindcast data (or corrected hindcast) are often used and they remain to be the main
source of metocean data for design and operational planning as well as for establishing joint environmental
description. Locations where high quality in-situ data are available are sparsely distributed, since buoy and
platform data are geographically limited, and though satellite observations offer global coverage, they suffer
from temporal sparsity and intermittency, making estimation of long term distributions and extreme analysis
difficult. The corrected hindcast may be unbiased on average but still can be corrupted by other types of
errors, which introduce a bias in the estimated return values of extreme sea states.
The limitation of the hindcast data has been for some time a lack of validation of numerical wave models
with instrumented data of significant wave height beyond 12 meters, but such data have started to exist and
used in the validation work recently. They confirm that 3rd Generation wave models are capable of
accurately hindcasting significant wave heights also in very extreme storms, see e.g. (ISSC, 2012, Cardone
et al., 2014).
As mentioned in Section 2.3.1 the recently updated or new developed metocean databases such as: ERAInterim, ERA-Clim, CFSR, NORA10, GROW12, HIPOCAS, BMT-ARGOSS and Fugro-OCEANOR
include information about both wind and waves and quality of these data bases is under continuous
improvement. ERA-Interim, ERA-Clim and CFSR databases have higher resolution and improved forcing
with better quality-control of assimilated data, see e.g. (Aarnes et al., 2012, Cardone et al., 2014). In the
extra-tropics these hindcasts can be expected to provide good estimates of the wave climate, especially for
the highest waves, whereas ship observations (collected since 1854) of the highest waves are notoriously
unreliable, and may be subject to some fair-weather bias (ship observations are discussed in Section 6.1
Design). The hindcast models are somewhat less reliable in the tropics, but for tropical storms the waves are
less extreme and do not define the design criteria for a sailing ship but may define design criteria for
offshore structures. Some recent result showing accuracy of hindcast models in extratropical storms is
presented e.g. by Ponce de Leon and Guedes Soares (2014) and in Ponce de Leon et al. (2014).
14
It is expected that hindcast data compare more satisfactory with measurements in conditions which are
lower than the design metocean conditions. Further, predictions giving by different metocean databases may
also be more consistent in these conditions but this is not sufficiently documented yet and new
investigations are called for.
Due to development of computers wave frequency-directional wave spectra have started to be archived
by met-offices opening new possibilities for environmental modelling as well as design and marine
operations.
2.2.4
Wave description from measured ship motions
Nielsen and Stredulinsky (2012) showed that it is possible to estimate the wave spectrum at the location of
an advancing ship by processing its wave-induced responses similar to the situation of a traditional wave
rider buoy. The study utilized a large set of full-scale motion measurements and the authors were able to
compute fairly accurate estimates of integrated sea state parameters when compared to corresponding
estimates from real wave rider buoys. The complete distribution of wave energy was also compared and
showed poorer agreement. The authors compared also their ship motion based estimates to observations
obtained from a commercial wave radar and showed that for the studied data set, the motion based estimate
provided, on average, slightly better sea state estimates than the wave radar system.
2.3
Current
Current data are important for studying ocean dynamics but also for the marine structure design and
operations. Information about current profile and velocities is of particular interest for structures that are
sensitive to currents, such as e.g. risers, riser towers, export lines, pipelines and umbilicals, especially in
connection with possible occurrences of vortex-induced-vibration (VIV) effects. Developments taken place
within renewable energy have brought the need for new current measurements as well as numerical current
data. Several studies showing how current energy can be utilized can be found e.g. in the OMAE 2012,
OMAE 2013 and OMAE 2014 Conference Proceedings. The most common categories of ocean currents
are: wind generated currents, tidal currents, circulational currents, loop and eddy currents, soliton currents,
longshore currents (DNV, 2014). Important characteristics of current field include mean current speed,
eddies, variations of current with water depth and current direction.
2.3.1
In-situ current measurements
Acoustic measurement techniques (both coherent and incoherent) for in-situ sensing of ocean current offer
an excellent space-time resolution of the velocity profile (Bitner-Gregersen et al., 2014a).
Ocean current observations can be found at a number of web-sites. NOAAs National Oceanographic
Data Center (NODC), http://www.nodc.noaa.gov/, provides current data from a number of sources as does
the Bundesamt fr Seeschifffahrt und Hydrographie (Federal Maritime and Hydrographic Agency),
http://www.bsh.de/en/index.jsp, of the German Federal Ministry of Transport, Building and Urban
Development. The Southern Oscillation Index/El Nino web site, http://www.pmel.noaa.gov/tao/elnino/ninohome.html provides access to a number of links to a number of data products including surface currents.
The utilisation of the Kuroshio Current power has initiated several measurements campaigns in Japan
and Taiwan. Kodaira et al. (2013) conducted an Acoustic Doppler Current Profiler (ADCP) measurement
around the island that revealed enhanced current speed of the Kuroshio Current under topographic
influences. Concurrent measurement by SAR revealed strong radar scatter where the current shear is strong.
Studies of the Loop Current carried out in the JIP DeepStar included the first measurements of the loop
inflow and turbulence and evaluation of existing numerical models, (Cooper et al., 2013). The Loop Current
is a strong permanent current that flows through the Yucatan Straits, loops northward, and then exits
through the Florida Straits where it is renamed as the Gulf Stream. In DeepStart the first time focus was
given on measuring the flow of the Loop through the Yucatan Straits providing fundamental information
that had never been gathered. The investigations showed that many of the models used were much worse
than simply assuming that the loop current remained unchanged (persistence). Further, it was documented
that the models were primarily limited by the accuracy of their initial conditions. These findings have been
utilized in other marine industry efforts to improve forecast models.
New measurements of current in the Brazilian waters have been reported during the term of the
Committee. Current data measured by an instrumented mooring line deployed at the Santos Basin, in a
water depth of 2200 m, show a mean velocity of 0.20 m/s with no preferential direction, (Andrioni et al.,
2012). Peak velocities 34 times higher than the average in a 3-year time series measured at the Santos
Basin, on Lula field have been identified associated with the passage of eddies dipoles. Current speeds up to
1.2 m/s at the first hundred meters of the water column have been generated.
15
Ceccopieri and Silveira (2012) used 2-year recorded current data series from an oceanographic mooring
array (F2200) at Lula Field to study the vertical variability of the ocean flows in this area. This variability
represents essentially the 1st baroclinic mode and a large part of it occupies the first 400600 m water depth,
with no predominant direction. It has been observed that the significant directional variability over the So
Paulo Plateau occurs far away from the mean current jets that flow parallel to the continental shelf
geometry. Further, significant seawater column seasonal stratification has been found.
2.3.2
Remotely sensed current measurements
Near-realtime global ocean surface currents derived from satellite altimeter and scatterometer data can be
found at NOAAs Ocean Surface Current AnalysesReal Time (OSCAR) web site
(http://www.oscar.noaa.gov/index.html). The data is validated against moored and floating buoy data, and
the method to derive surface currents with satellite altimeter and scatterometer data is the outcome of
several years NASA sponsored research.
2.3.3
Ocean model outputs have been used after the incident at the Deepwater Horizon platform in April 2010 in
the Gulf of Mexico to trace spilled oil in the Gulf Stream, and to trace debris and radioactive materials after
the earthquake and tsunami incidents on 11 March 2011 in north east Japan, see e.g. (Aoyama et al., 2012),
(Tsumune et al., 2012), providing promising results. In Massonnet et al. (2013) comparison of predictions of
five ocean models can be found.
2.4
Sea water level
Sea level variations have got special attention in the last decade due to the ongoing debate about climate
change. The sea level changes have been geographically non-uniform in the past and climate projections
show that they will be also in the future (see Section 4.1.4). They have little effects on ship design directly
but have impact on design and operations of offshore and coastal installations and may influence ship
operations (e.g. due to changes of harbour depth).
Sea level variations are collected by gauges, remote sensing techniques or generated by numerical
models.
2.4.1
Locally sensed sea water level measurements
Sea level observations by tide gauges are restricted to the coastal region and because of the natural
geographical inhomogeneity of the sea level rise; the global average sea level estimates become
erroneous. An obvious source of error of long-term sea level trends from in-situ measurements is the
change of the terrestrial reference frame which needs always to be checked.
2.4.2
Remotely sensed sea water level measurements
Satellite altimetry provides a means to measure directly the global sea surface topography and its
accuracy depends on the spatial scale. Although altimetry is not able to provide local short scale sea level
monitoring, it provides the long-term mean sea level change at global scale.
2.4.3
Numerical modelling can provide historical data sets which are essential for the analysis of long-term sea
level variations for marine and renewable energy applications.
Several studies have been carried out to project future sea water level changes using GCM (Global
Climate Model) or RCM (Regional Climate Model) models (see Section 4.1.4). The degree to which
GCM, or RCM, have sufficient resolution and/or internal physics to realistically capture the
meteorological forcing responsible for storm surges is regionally dependent.
2.5
Ice and snow
Trends in the Arctic and Antarctic regional climate are largely investigated as they are considered
markers of global climate change. Ice and snow melting conditions are analyzed mostly from remote
sensing and in-situ data. Sea ice evolution is also widely studied and large efforts have been made to
develop and validate coupled ice-ocean models. Changes of ice conditions are reported in Section 4.1.3.
2.5.1
Locally and remotely sensed ice and snow measurements
Barrand et al. (2013) used a data set combining in situ meteorological observations, spaceborne
scatterometer data (QuickScat), together with output from simulations of a regional climate model,
16
RACMO2, to analyze the melting conditions on the Antarctic Peninsula. Trends of the positive degree day
(PDD) sum were found largely positive at most stations. QuickScat data analysis showed that melt was
typically more variable over ice shelves than grounded ice but that no evident link was identified between
early melt season onset and ice shelf retreat or breakup events. However ice-shelf break-up or retreat events
were found to coincide with longer melt season durations. Finally QSCAT melt extent in the Antarctic
Peninsula was shown to be strongly correlated with the OctoberJanuary averaged SAM index, linking AP
melt trends to large-scale (global) modes of atmospheric variability.
Using CryoSat-2 data acquired between November 2010 and September 2013 and adapting a repeattrack method, McMillan et al. (2012) produced the first altimeter-derived estimates of volume and mass
change for the entire Antarctic ice sheet. Based on analysis of this new data set that provides nearcontinuous (96%) coverage of the entire continent, extending to within 215 km of the South Pole and
leading to a fivefold increase in the sampling of coastal regions, the authors estimate that, since 2010, the
average Antarctic ice sheet contribution to global sea level rise has been 0.45 0.14 mmyr-1, a value, which
is more than twice as large as the 20 year mean determined from an ensemble of geodetic techniques.
Comiso and D.K. (2014) provide a general assessment of the current state of the Arctic climate as
derived from historical satellite and in-situ data, as reflected in its major components, temperature, sea ice
cover and snow cover. They indicate that warming in the Arctic has been amplified, as expected from icealbedo feedback and other effects and as predicted by models, with a rate of warming increasing of about
0.6oC/decade in the Arctic during the period 1981 to 2012. They also list some requirements for future
observations, as for instance ice thickness and snow cover as well as for the development of numerical
models incorporating the physics of the system and able to simulate the observed variability and trends of
the various parameters in the system.
Low-resolution remotely sensed sea ice drift products are widely used for various purposes such as
validation of coupled ocean-sea ice models, model parameter estimation and data assimilation in the Arctic
Ocean. Sumata et al. (2014) made an intercomparison of four different Arctic Ocean low-resolution ice-drift
products derived from satellite observations, and also examined their differences to buoy drift data. The
products are inferred from different sensors with different time and space scales, and different motiontracking algorithms so that the data was processed to monthly mean vectors so as to standardize the
temporal representation. This comparison provides practical information to data users as well as uncertainty
estimates. Even though the transpolar drift speed differs among the products by 13% on average and
differences among the products are not spatially uniform, all the products are considered by the authors as
practical and useful for model validation and data assimilation, providing that the uncertainty and error
statistics are given in a proper way.
CERSAT provides a new database, Altiberg, which contains the small iceberg (<3km in length) detected
by altimeters using the high resolution waveforms. The level 2 products are the individual detected iceberg
files. The level 3 product contains Antarctic polar stereographic 100 km resolution grids of monthly volume
of ice, probability of presence of icebergs and mean iceberg area. The database is also available for other
several altimeter mission (ERS-1, ERS-2, Jason-1, Jason-2, CryoSat-2, Topex, Envisat, AltiKa). A merged
product combining all the available altimeters is also provided.
2.5.2
As important decreases in Arctic sea ice are predicted by some climate models it is critical to assess the
accuracy and reliability of high-latitude climate forecasts. This requires a better understanding of sea ice
dynamics and thermodynamics through the proper simulation of sea ice and its responses to atmospheric
forcing across a range of temporal and spatial scales. Johnson et al. (2012) assessed the ability of six
coupled Arctic Ocean Model Intercomparison Project (AOMIP) models to simulate sea ice thickness and to
identify trends. Sea ice thickness from six AOMIP models was compared with thickness across the Arctic
basin from various remote and in-situ measurement data sets. Authors indicate that on the one hand the
models overestimate thickness of ice thinner than 2 m, which is problematic for forecasting areas of open
water and perhaps the timing of the seasonal cycle, and on the other hand they underestimate the thickness
of measured ice thicker than 2 m possibly hindering long-term forecasts where the proper role of multi-year
ice is critical.
With about the same objective, Lindsay et al. (2013) evaluated seven different atmospheric reanalysis in
the Arctic regions and compared four data sets for use in simulating past sea ice conditions. Sea ice
thickness and total ice volume were computed with the PIOMAS coupled iceocean model. Obtained
results are broadly similar even though one atmospheric model provided a better agreement than the others
with submarine ice draft measurements.
17
Wave propagation through ice and in ice covered waters is rather intensively investigated. Campbell
et al. (2014) used a stereo imaging system to analyze propagation of waves interacting with three different
types of ice. Alteration of the frequency content of the wave spectra together with the damping of the energy
flux is found to be ice quality dependent. It is also noted that wave reflection is observed at the boundary of
each ice layer.
Meylan et al. (2014) analyzed wave propagation in the Antarctic marginal ice zone (MIZ) from in situ
measurements of ocean surface wave spectra. Analysis of the measurements provided by an array of five
wave sensors shows significant wave heights and peak periods do not vary appreciably in approximately the
first 80 km of the ice-covered ocean where dominant floe sizes are less than 6 m and ice concentrations are
relatively low. Beyond this region, where dominant floe sizes are greater than 10 m and concentrations are
higher, significant wave heights is found to decay whilst peak periods increase. Attenuation rates are
investigated and found to be insensitive to amplitudes for long-period waves but to increase with increasing
amplitude above some critical amplitude for short-period waves. They also decrease with increasing wave
period. Further, for long-period waves the decrease is shown to be proportional to the inverse of the period
squared. This relationship can be used to efficiently implement wave attenuation through the marginal ice
zone in ocean-scale wave models.
Unlike the rapid sea ice losses reported in the Arctic, satellite observations show an overall increase in
Antarctic sea ice concentration over recent decades. However, observations of decadal trends in Antarctic
ice thickness, and hence ice volume, do not currently exists. Holland et al. (2014) use a model of the
Southern Ocean and its sea ice, forced by atmospheric reanalysis to assess 19922010 trends in ice
thickness and volume. The model successfully reproduces observations of mean ice concentration,
thickness, and drift, and decadal trends in ice concentration and drift, imparting some confidence in the
hindcasted trends in ice thickness. The small overall increase in modeled ice volume is actually the residual
of much larger opposing regional trends. Even though the reasons these regional patterns of changes that
produces the overall increase in ice cover are not well known and proposed drivers are numerous, the model
results presented by the authors reproduce observations of mean ice concentration, drift, and thickness, as
well as trends in ice concentration and drift. Additionally, the simulated ice thickness trends also agree with
those of Masumoto et al. (2012), which can be regarded as a best estimate because of their use of data
assimilation.
3.
ENVIRONMENTAL MODELS
The environmental description often employs a mixture of mathematical, probabilistic, empirical and
statistical models. The following decoupling approach is commonly used. It is assumed that for a limited
period of time and in a particular geographical region metocean conditions vary in a stationary way called
sea state. Metocean conditions in a sea state can be described by means of mathematical models depending
on a number of characteristic parameters. Changes of sea state parameters, which vary much slower than sea
waves, wind, currents, sea water level and some ice characteristics, are modelled by means of probabilities.
The final description of environmental conditions is obtained by combining the models for sea states
evolution with the description of sea waves, wind, current, sea water level and ice in a sea state.
The increasing awareness of importance of environmental uncertainties for safety at sea is contributing
to increased number of benchmark studies where performance of different models is compared and
evaluated (e.g. the 2nd ITTC-ISSC Workshop). Both academia and industry is participating in these studies.
However, specification of model uncertainty is not an easy task. Model uncertainty is due to imperfections
and idealisations made in physical process formulations as well as in choices of probability distribution
types and fitting techniques applied for estimation of distribution parameters, see (Bitner-Gregersen et al.,
2014a). Several errors can contribute to model uncertainty each defined as the ratio between the true
quantity and the quantity as predicted by the model. A mean value not equal to1.0 expresses a bias. The
standard deviation expresses the variability of the predictions by the model. Experimental tests, field tests or
the average values of recognized models (or weighted models) are used often as a reference value (the true
value).
Most of the probabilistic and hybrid models are based today on the assumption of stationarity and
homogeneity which can be questioned in certain conditions (see Section 2) and may have significant impact
on model predictions. Attention to this model uncertainty has been given in the course of the Committee,
e.g. (Bitner-Gregersen et al., 2014b).
3.1
Wind
The classical modelling approach for engineering applications is to fit the probability distribution to a
known model and estimate statistical parameters such as the mean and variance. The common statistical
18
description techniques have some drawbacks as they lack the time variation properties and ignore crossdependencies between other meteorological data. No important changes to the statistical and spectral
description of wind have taken place in recent years. Wind models commonly applied can be found, e.g. in
Legacy DNV RP-C205 (DNV, 2014) and Legacy DNV-OS-J101 (DNV, 2011). The importance of
accounting for non-stationarity have been stressed during the period of the Committee. Also the
homogeneity assumption needs to be revisited in the future, particularly due to renewable energy
installations. CFD (Computational Fluid Dynamics) methodology is getting increasing attention in
modelling of wind conditions.
3.1.1
Analytical description of wind
As a result of social awareness of air emission due to use of fossil fuels, the utilization of the natural wind
power resources becomes an important option to avoid the dependence on fossil resources in industrial
activities. For example, the maritime industry, which is responsible for more than 90% of the world trade
transport, has already started to look for solutions to use wind power as auxiliary propulsion for ships. The
practical installation of the wind facilities often requires large amount of investment, while uncertainties for
the corresponding energy gains are large. Therefore a reliable models to describe the variability of wind
speeds in time and large geographical regions are needed to estimate the expected available wind power,
coefficient of the variation of the power and other statistics of interest, e.g. expected length of the wind
conditions favourable for the wind-energy harvesting.
Most models considers time series of wind data at fixed locations or in space keeping the time fixed.
Modelling spatial and temporal dependence of wind speed Uw, is a very complex problem. Often one
proposes to use the transformed Gaussian model, which assumes that there exists a deterministic function G,
possibly dependent on location, such that X = G(Uw) has Gaussain CDF. In Ailliot and Monbet (2012) a
switching Markov model has been used to describe the dependence structure of wind speed data. The
advantage of the model is that it uses a number of typical weather conditions to include meteorological
information into the description wind speed variability. The limitation is the complexity of the model which
limits its applicability to relatively small regions. In Rychlik and Mustedanagic (2013) a global nonstationary model for wind speeds variability has been proposed. In this work the transformed wind speed
X = G(Uw) is a non-stationary and non-homogeneous Gaussian field. The model is defined by means of its
mean values, variances and the covariances. The global covariances depends on local parameters which are
covariances of wind speeds gradient. The model was applied in Rychlik and Mao (2014) to describe
statistical properties of wind speeds encountered by vessels along commercial routes. Measured wind
speeds on-board Wallenius Lines AB ships were used to validate the model.
Installations of wind-harvesting equipments on-board of a vessel raise a question of structural safety. In
order to analyse the relevant loads both wind speed and wind direction is needed. Modelling wind velocity
in time and space is much harder problem than considering just wind speed. The reason is the complex
dependence between wind speed and the direction which cannot be described using the transformed
Gaussian models. In Ailliot et al. (2014) a spatio-temporal model for wind velocities has been proposed. It is
a generalization of the model proposed in Ailliot and Monbet (2012) and hence also limited to small
geographical regions.
The parameters in models presented in Ailliot and Monbet (2012), Rychlik and Mustedanagic (2013),
Rychlik and Mao (2014) and Ailliot et al. (2014) are all estimated using ERA-Interim data Dee et al. (2011)
produced by European Centre for Medium-Range Weather Forecasts.
Muyau et al. (2014) have investigated wind data sampled at 1 Hz from three offshore locations off the
west coast of Borneo and one onshore location (see Section 2.1.1) which included mainly monsoon
conditions. The data set analysed consisted of 5952 1-h records. The authors have found that none of their 1
Hz data records was stationary over one hour, and even when the 1 Hz data were averaged over longer
intervals, no records of 3-second means were found to be stationary, and only 14% of the records of 1-min
means were found to be stationary. The results indicate that application of fixed short-term wind
relationships is questionable.
3.1.2
Statistical and spectral description of wind
Wind, waves and current are the most important environmental loads that act on ships and offshore
structures. Statistical descriptions of the loads are often used in evaluation of reliability of structures details
as well as safety of marine activities. A standard model of the loads assumes that those are stationary for a
period of time lasting from 10 minutes to several hours. During such periods properties of loads (time
series) are described by means of their average values and power spectral densities (PSD). A Gaussian, or
transformed Gaussian, model is commonly used. This is convenient since the loads are then uniquely
defined by the mean, PSD and transformation, if the transformed Gaussian model is employed.
19
Several parametric families of PSD, i.e. functions of sea state parameters like; significant wave height
Hs; peak period Tp, some parameters that describing waves directional spreading, average wind speed Uw,
turbulence intensity, gust factor and other, exist. Means to simulate Gaussian time series are well
established. The simulated time series of artificial loads are then used to study responses of structures, ships
etc. During the period 20122014 there were a number of papers published presenting studies of responses
of offshore located wind mills, mooring lines etc. to the combine wind and wave loads. In several papers
one was choosing some typical values of sea states and wind parameters focusing mostly on derivation of
responses. Those papers will not be included in the present summary of the research on environmental
conditions.
The stationarity periods are very short compared to length of maritime operations or life of a structure
details. Consequently the parameters defining PSD forms time series that long-term variability needs to be
appropriately modelled. While marginal models for sea states or winds parameters are well established the
multivalued models for joint sea-states wind parameters variability are still under development. Two types
of applications are commonly met in the literature; reliability assessments or estimation of available energy
at offshore located wind mills and safety of structures or maritime operations. In the first type of problems
the general variability of environmental loads are of interest including both common and rare values of
parameters. In the second type of problems the focus is on extreme cases which can lead to failures and for
safe structures or operations are rarely encountered.
In Li et al. (2013) the authors have presented a model for the joint PDF (Probability Density Function)
of mean wind speed Uw, significant wave height Hs and peak period Tp. The long-term PDF of Uw was
assumed to be Weibull, the conditional PDF of Hs given Uw were also Weibull with parameters being a
polynomial functions of the mean wind speed. Finally, the conditional distribution of Tp given Uw and Hs
was lognormal with parameters dependent on the mean wind and significant wave heights. The joint PDF
were fitted to the 10 years of hindcast data in several locations in Atlantic Ocean along European coast. For
extreme responses analysis the authors proposed to use the 50-year contour surfaces of the fitted joint PDF.
Another approach was presented in Tao et al. (2013b). In that work the authors proposed to use copulas to
model the joint PDF of Uw and Hs. This is an approach to introduce dependence between non-Gaussian
variables when marginal PDFs are known. Basically one is first specifying marginal PDFs for Uw and Hs
and then the joint PDF for the parameters is modelled by means of an appropriately chosen function
(copula) of the marginal PDFs.
Finding an appropriate model for a joint PDF of extreme parameter values is a difficult problem. First of
all data are sparse and has to be gathered over long time periods. Secondly the data may not be stationary or
homogeneous. Modelling multivalued extremes is an active research field in statistics. One way to model
variability of extreme values, e.g. having 100 years return, when there are not sufficiently many
observations is to fit a joint PDF to the available data and extrapolate PDF to the regions were data are
missing. This approach was used in Li et al. (2013) and Tao et al. (2013b). The method relies on
assumptions that a chosen PDF fits correctly data both in central region, dominating the fit, and in tails for
which there are often only few observation available. This may not be the case and hence if distribution of
extreme values are of major interest, then one should look for the tailored models, valid solely in the tails.
One often taken approach is to select extreme episodes, for example severe storms, and propose a PDF
for the observed environmental parameters. Such an approach was taken in Towe et al. (2013) where one
was modeling variability of extreme Hs and Uw values. In the paper the authors assumes that the marginally
Hs and Uw have extreme value distribution (Frche). Then means to model and estimate dependence
measures for bivariate extremes were presented. The analysis was further generalized by introducing
covariates which allows to deal with the cases when data are not stationary (homogeneous). The method
was validated using the hindcast data at three locations in North Sea. It was shown that storm direction is
important covariate and that extreme wind speed and significant wave height are dependent.
Another modelling approach was presented in Naess and Karpa (2013). In that paper the authors
extended the ACER method (Naess and Karpa, 2013) to estimate the extreme quantiles to cover the
bivariate data. Wind speed and significant wave height data were considered. The hindcast at some location
close to Norway coast were used for validation. The method applies to time series and no extraction of
storms is needed. Finally the estimates of ACER empirical surface was extrapolated to very high quantiles
by means of a fit of a parametric function to the empirical ACER.
All the work discussed above used hindcast or buoys to study the distributions of wind speed or
significant wave height. The hindcast or reanalysis do not represent actual measurements of quantities but
extrapolations to the grid locations based on simulations from complex dynamical models. It is defined on
regular grids in time and space and hence convenient to use. Often the data has to be calibrated using in situ
measurements before these can be used. In Young et al. (2013) the satellite measurements of wind speed
and significant wave height were used to estimate the 100 years return wind speeds and significant wave
20
heights. The trends in these values were also identified and used in prediction of the future return values
over the globe.
The practical installation of the wind facilities often requires large amount of investment, while
uncertainties for the corresponding energy gains are large. Therefore a reliable models to describe the
variability of wind speeds in time and large geographical regions are needed to estimate the expected
available wind power, coefficient of the variation of the power and other statistics of interest, e.g. expected
length of the wind conditions favourable for the wind-energy harvesting.
Roughly the average available wind power during a stationarity period is proportional to average wind
speed raised to power three hence an accurate long-term distributions of wind speed at one or
simultaneously at several positions are of interest. There exists several complex numerical programs to
reconstruct wind fields using ocean-atmosphere models based on large-scale meteorological data, called
hindcast or reanalysis. In Salvacao et al. (2013) the authors presented a study in which results derived by
means of the mesoscale model WRF, a regional non-hydrostatic model and ERA-Interim data (Dee et al.,
2011), were calibrated using some buoy measurements. The method was validated and shown to be useful
to derive the long-term distributions of wind speeds and directions along the Portuguese coast. In the
calibration process a relation between the average wind speeds and height above the ground is needed. In
Babanin and McConochi (2013) wind measurements near surface of waves was studied and the accuracy of
the relation was investigated. The conclusion was that one needs to be cautious when extrapolating the wind
speeds measured by buoys to e.g. 10 meters. For example if the logarithmic law is used then U10 could be
overestimated. In Trombe et al. (2013) results of an experiment where weather radars were used for
monitoring the weather at the Horns-Rev offshore wind farm, in the North Sea were presented. Use of the
weather radar enabled the collection of meteorological observations at high spatio-temporal resolutions for
enhancing the understanding of meteorological phenomena that drive wind fluctuations. The measured
environment provided relevant inputs to prediction systems for anticipating changes in the wind fluctuation
dynamics, generating improved wind power forecasts.
In the field of energy production prediction models of wind vector characteristics are needed in order to
supervise and optimize the electricity generation and planning. Prediction by means of time series of wind
speeds have been extensively studied in the literature. For example a model for very short-term probabilistic
forecasts (essential for an optimal management of wind generation) was presented in Pinton (2012). The
author used mixtures of generalized Lognormal distributions combined with autoregressive and conditional
parametric autoregressive models to predicts wind speeds. Another innovative approach was presented in
Stefanakos et al. (2014) where fuzzy time series techniques were applied to forecast wind speeds. The
efficiency of the method has been compared with the forecasting using more traditional ARIMA models.
3.2
Waves
The wave spectral models remain being continuously developed. The GlobWave project initiated by ESA
in 2008 to improve the uptake of satellite-derived wind-wave and swell data is still on going. The present
status of the ESA GlobWave project making satellite derived data more widely available is on
http://globwave.ifremer.fr/Products/Summary-of-Services and the GLOBWAVE Data Handbook on the
project web site (http://www.globwave.org/).
The knowledge of extreme and rogue waves has advanced since the 2012. The predictions made by
theoretical and numerical models compare well with experimental results. Attention has been given during
the period of the Committee to including forcing sources such as wind and wave breaking in nonlinear wave
models and to studying wave-current interaction. Scale effects in model tests of extreme waves has been in
increasing focus.
Some progress has been made on long-term description of sea states, particularly on directionality of
several wave systems and spatial-temporal models of sea surface characterists. Also shallow water effects
are getting increasing attention due to the needs of renewable energy industry and adaptation to climate
change.
3.2.1
Analytical and numerical wave models
The WAM model and the WAVEWATCH-III model are the most generalized and tested wave prediction
models used for both hindcasting and forecasting purposes. Although both WAM and WAVEWATCH-III
are 3rd generation (3G) wave models, they now differ in a number of physical and numerical aspects and
may give different predictions. This is an indication that a single best solution has not yet been accepted.
The SWAN model remains to be used with shallow water wave climate. The quality of numerical wave and
surge hindcasts for offshore and coastal areas depends to a large extent on the quality and the accuracy of
the upper boundary conditions, i.e. in particular on the quality of the driving wind fields. Recent progress on
21
wave modelling has been presented at the 13th International Workshop on Wave Hindcasting and
Forecasting & 4th Coastal Hazard Symposium, taken place 27 Oct.1 Nov. 2013 in Banff, Canada
(http://www.waveworkshop.org/13thWaves/index.htm). Selected papers presented are published in the 2014
Special Issue of Ocean Dynamics.
Efforts to push exclusively non-stationary models such as WAM and WAVEWATCH-III closer to shore
reported in ISSC 2012 I.1 Report (ISSC, 2012) have continued during the period of this Committee, e.g.
(Tolman, 2013). This development is motivated to avoid learning, maintaining, and running multiple wave
models at a given operational centre. Continuous attempts are made also to establish a stronger interaction
between the wave and the circulation modelling community important for future development of the wave
and circulation models.
Wave models have largely improved over the recent years thanks to new developments in both
parameterization, introducing more consistent description of the physics based on observations, and
numerical choices. Roland and F. (2014) summarize the most important aspects of these improvements,
including the introduction of currents, coastal reflection, and bottom sediment types which are of prime
interest for modeling in coastal waters. Most of these improvements were implemented in a new version
(4.18) of the code WaveWatch III which was released in March 2014 by NOAA/NCEP allowing the use
of unstructured grids and introducing new parameterizations for wave dissipation together with new
parameterizations for bottom friction including movable bed roughness (Tolman and group, 2014). Details
on the parameterization and physics that were implemented can be found in earlier papers.
Ardhuin et al. (2012) investigated the description of wave-current interaction in models considering also
conditions with strong currents. They showed that including currents in the model resulted in error
reductions by up to 30% on the evolution of the significant wave height; even at locations where currents
are relatively weak but which are located down-wave of strong current gradients that cause large refraction
effects. Leckler et al. (2013) proposed an improved parameterization of energy dissipation by whitecapping
taking into account the physical relationship intrinsically linking spontaneous breaking dissipation with
dissipation induced by breaking waves.
In parallel to this effort for improvement of the models, operational services were developed and
hindcast databases were implemented (see Section 2.2.3).
The National Centers for Environmental Prediction (NCEP) (Chawla et al., 2012) implemented a new
global forecast operational system using the third public release of WAVEWATCH III (3.14). This new
multi-grid system was designed to provide the National Weather Service (NWS) and NCEP centers with
model guidance of suitable resolution for all areas where they have the responsibility of providing gridded
forecast products. New important features include two-way nesting between grids and carving out selected
areas of the computational domain. This allowed increasing spatial resolution and extending the global
domain closer to the North Pole, while at the same time optimizing the computational cost. A spectral
partitioning algorithm was also implemented to separate individual sea states from the overall spectrum,
thus providing additional products for multiple sea states. This modeling system has been validated against
data using a multiyear hindcast database as well as archived forecasts.
A thirty one year wave hindcast (19792009) was implemented by Chawla et al. (2012) using NCEPs
latest high resolution Climate Forecast System Reanalysis (CFSR) wind and ice database. This hindcast has
been generated using the third generation wind wave model WAVEWATCH III (3.14) with a mosaic of 16
two-way nested grids having a resolution ranging from 1/2 to 1/15. Validation made for bulk significant
wave height Hs and 10 m (above Mean Sea Level) wind speeds U10 comparing with both altimeter records
and NDBC buoys showed the general ability of the database to correctly representing the wave climate.
Agreement is excellent at most buoys between model and data out to the 99.9th percentile but of better
quality at the offshore buoys than at the coastal buoys because of unresolved coastal features (topographic/
bathymetric) as well as issues related to interpolating wind fields at the land-sea margins. It should be noted
though that the authors point out some concerns about the wave climate in the Southern Hemisphere due to
the over prediction of winds (early part of the database) as well as the lack of wave blocking due to icebergs
(in the model).
Another 20 year multi-scale global hindcast database of global wave parameters has been produced,
(Rascle 2013), using WaveWatch III with a new parameterization for wave, and forcing from a combination
of ECMWF analysis and CFSR reanalyzes, sea ice from CFSR and ECMWF and icebergs from CERSAT.
Validation of this database against altimeters and buoys pointed out different features, especially related to
wind forcing and in agreement with conclusions by Chawla et al. (2013). CFSR winds were found
anomalously high in the Southern Ocean for the years 19911993, compared to following years, resulting in
anomalous high biases for these years, including off the U.S. West coast. CFSR and NCEP analyses are
22
found to have systematically higher values than ECMWF analyses of the wind speed, especially for the
highest speed range.
Even if a simple calibration of the wind wave growth parameter corrected the average to high wave
heights, modeled wave heights are found to still be too low for the highest values (Hs > 12 m), likely due to
an underestimation of the winds in these conditions. The new parameterization implemented in the model
also allowed a more accurate estimation of parameters derived from spectral moments, including the surface
Stokes drift and mean square slopes that are relevant for wavecurrent interactions modeling and remote
sensing, and also spectra of seismic noise sources. However air-sea fluxes of momentum and energy are
found to be not very realistic and will require adjustments of the future generation and dissipation
parameterizations.
As pointed out in the previous papers, quality and accuracy of the wind forcing is important for wave
modeling, especially in complex coastal areas. Bricheno et al. (2013) investigated the influence of the
accuracy of representation of wind forcing and mean sea level pressure on waves and surges modeling.
Running the Weather Research and Forecasting (WRF) model at 12-, 4-, and 1.33-km resolution for a storm
event over the Irish Sea they used the outputs to force the coupled hydrodynamic and the Proudman
Oceanographic Laboratory Coastal Ocean Modeling System (POLCOMS)Wave Model (WAM) so as to
assess the effect of the forcing on storm surge and waves. They observed an improvement of the wind speed
estimation when moving from 12- to 4-km resolution but going to the 1.33 km resolution showed no further
significant improvement. Implementing the atmospheric model results at 12 and 4 km as input to the ocean
model, wave direction was seen to improve with increased ocean model resolution, and higher-resolution
forcing was found to generally increase the wave height over the Irish Sea by up to 40 cm in places.
Improved clustering of wave direction was observed when 4-km meteorological forcing was used. Large
differences were seen in the coastal zone because of the improved representation of the coastline and, in
turn, the atmospheric boundary layer. Hence, the combination of high-resolution atmospheric forcing and a
coupled wavesurge model gave the best result.
The effect of the grid resolution on modelling of fetch-limited wave growth has been studied using wave
model WAM by Touboul and Pelinovsky (2014). Three different methods to compile a grid for a wave
model in the case of an irregular shoreline were discussed. The effect of grid resolution on the growth rate
of the wave energy at short fetch was relatively large.
Fan et al. (2012) described a 29-yr (19812009) global ocean surface gravity wave simulation generated
by a coupled atmosphere-wave model using NOAA/GFDLs High-Resolution Atmosphere Model
(HiRAM) and the WAVEWATCH III. The comparisons with satellite altimeter measurements indicated
that the significant wave height (SWH) low bias in ERA-40 reanalysis had been improved in these model
simulations.
For several decades, the Discrete Interaction Approximation (DIA) for nonlinear resonant four-wave
interactions has been the engine of third-generation wind-wave models. Tolman (2013) presented a
Generalized Multiple DIA (GMD) which expands upon the DIA by (i) expanding the definition of the
representative quadruplet, (ii) formulating the DIA for arbitrary water depths, (iii) providing complimentary
deep and shallow water scaling terms and (iv) allowing for multiple representative quadruplets. The GMD is
rigorously derived to be an extension of the DIA, and is backward compatible with it. The free parameters
of the GMD have been optimized holistically, by optimizing full model behavior in the WAVEWATCH III
wave model (Tolman and Grumbine, 2013). The results showed that in deep water, GMD configurations
can be found which remove most of the errors of the DIA. The two-scale approximation (TSA) to the full
Boltzman integral representation of quadruplet wave-wave interactions has recently been presented as a new
method to estimate nonlinear transfer rates in wind waves, and has been tested for idealized spectral data, as
well as for observed field measurements (Willis and Bonnefond, 2013). TSA has been implemented in
WAVEWATCH III and shown to perform well for wave spectra from field measurements, even for cases
with directional energy shearing, compared to the DIA.
Gramstad and Babanin (2014) has made an attempt to include quasi-resonance interaction in the wave
model. The results are preliminary but promising and call for further investigations.
A semi-empirical determination of the spectral dependence of the energy dissipation due to surface wave
breaking has been studied by Romero et al. (2012) and then used to propose a model for the spectral
dependence of the breaking strength parameter b, defined in the O.M. Phillipss statistical formulation of
wave breaking dynamics. Numerical investigations based on full dynamic equations for wave breaking have
been studied by Chalikov and Babanin (2012) in a one-dimensional environment with a wave spectrum.
Besides, the role of breaking in an evolving wave field has been studied by Schwendeman et al. (2014).
23
Importance of high frequency tail in third generation wave models, SWAN and WAVEWATCH-III, has
been studied by Siadatmousavi et al. (2012) with different assumptions for high cut-off frequency used to
evaluate the interaction of low and high frequency components in wave spectral evolution. The results
showed that WAM Cycle 3 was more sensitive to cut-off frequency as well as to the exponent used in the
expression for the frequency tail, than other formulations in SWAN. Wang and Jiang (2012) proposed a new
spectral dissipation source term which comprises saturation based dissipation above two times of peak
frequency and improved whitecapping dissipation at lower frequency spectrum. The reciprocal wave age
was involved into the whitecapping model to adjust dissipation rate at different wind speed. The Phillips
higher frequency saturation parameter in the saturation-based dissipation was no longer taken as a constant,
but varies with wave age. In addition, limiters based on the CFL criteria have been proposed for the spectral
propagation velocities in SWAN (Dietrich et al., 2013). These limiters prevented the excessive directional
turning and frequency shifting of wave energy and improve accuracy by reducing local errors that would
otherwise spread throughout the computational domain. Rascle and Ardhuin (2013) presented a multi-scale
global hindcast of ocean waves which covered the years 19942012, based on improved source term
parameterizations for wind sea and swell dissipation (Ardhuin et al., 2010).
Waves propagating in shallow water dissipate energy in a thin, turbulent boundary layer near the bottom
experience friction. This friction can be estimated with a simple quadratic friction law scaled with an
empirical coefficient. Two values of this coefficient have been recommended by previous studies (for sandy
bottoms): a high value for waves in a storm and a low value for swell. However, the review of a large
number of more recent observations by Zijlema et al. (2012) gave a new wind drag parameterization with
lower values. Using this lower value also improved the estimates of wave growth in shallow water and of
low frequency wave decay in a tidal inlet, independent of the wind drag. Besides, modification of the
STWAVE bottom friction coefficients and boundary forcing conditions has been studied on the STWAVE
model (Christopher et al., 2013). The results indicated good agreement with the measured nearshore wave
data for an open water Manning n bottom friction coefficient equal to 0.03 s/m0.33, and good agreement
with the measured inshore wave data with Manning n bottom friction coefficients equal to values derived
from land classification data and applied in the ADCIRC model.
The effect on waves of the Wave-Current Interaction (WCI) process has been studied for wave
simulation. Nearshore propagation of cyclonic waves have been simulated (Panigrahi et al.) using state-ofart SWAN model coupled with hydrodynamic model POM (Princeton Ocean Model). Ardhuin et al. (2012)
did a study on performance of numerical models in conditions with strong currents. The results showed that
wave models can reduce the errors on significant wave heights by more than 30% in some macrotidal
environments, such as the coast of Brittany, in France. The structured grid circulation model ROMS has
been coupled with the unstructured grid Wind Wave Model II (Dutour-Sikiric, 2013). The chosen models
and coupling approach allowed the grids of both models to be chosen independently. In addition, the
unstructured-mesh SWAN spectral wave model and the ADCIRC shallow-water circulation model have
been integrated into a tightly-coupled SWAN+ADCIRC model (Dietrich et al., 2011). (Dietrich et al., 2012)
examined the SWAN+ADCIRC model applied to a high-resolution, 5M-vertex, finite-element SL16 mesh
of the Gulf of Mexico and Louisiana. Performance and validation of ADCIRC+SWAN model were also
studied by Prasad et al. (2013) and Choi et al. (2013). Their study signified the importance of coupled
parallel ADCIRC+SWAN model for operational needs during extreme events.
The University of Miami has presented a Fully Coupled Atmosphere-Wave-Ocean Modeling system
(UMCM). The UMCM includes three model components: atmospheric, surface wave, and ocean circulation
models. Chen et al. (2007) gave a brief introduction to UMCM and an overview for the coupled modeling
effort in the Coupled Boundary Layer Air-Sea Transfer (CBLAST)-Hurricane program. Chen et al. (2013)
described the results of a new directional windwave- coupling parameterization in a fully coupled model
developed based on the CBLAST-Hurricane observations and laboratory measurements. Currently, UMCM
can be configured with two different options in terms of component models: 1) coupled with the fifthgeneration Pennsylvania State University-National Center for Atmospheric Research Mesoscale Model
(MM5), a third generation wave model (WAVEWATCH III), and the three-dimensional Price-WellerPinkel (3DPWP) upper ocean model (UMCM-MWP); and 2) coupled with the Weather Research and
Forecasting Model (WRF), the University of Miami Wave Model, UMWM, (Donelan et al., 2012), and the
Hybrid Coordinate Ocean Model (HYCOM) (UMCM-WMH).
In addition, Li et al. (2012) did an investigation of the effects of wave state and sea spray on an idealized
typhoon using an air-sea coupled modeling system. The coupling between atmosphere and sea surface
waves considered the effects of wave state and sea sprays on airsea momentum flux, the atmospheric lowlevel dissipative heating, and the wave-state-affected sea spray heat flux. Smith et al. (2013) examined
24
tropical cyclone ocean-wave model interactions using an ESMF (Earth System Modeling Framework) based
tropical cyclone (TC) version of the Coupled Ocean/Atmosphere Mesoscale Prediction System (COAMPS).
The Coupled Ocean-Atmosphere-Wave-Sediment Transport (COAWST) modeling system was used to
investigate Semi-enclosed Gulf of Venice (Benetazzo et al., 2013). The results revealed that, when applied
to intense storms, the effect of coupling on waves results in variations of significant wave height up to
0.6 m, with some areas experiencing significant increase/decrease of wave spectral energy for
opposite/following currents respectively.
Pleskachevsky et al. (2012) investigated the impact of the gustiness on surface waves under storms in the
North Sea, focusing on the appearance of wave groups with extreme high amplitude and wavelength.
Optical and microwave satellite data were used to connect mesoscale atmospheric turbulences and extreme
waves measured near the German coast. Moving atmospheric open cells are observed to produce a local
increase in the wind field at the sea surface, moving as a consistent system with a propagation speed near to
swell wave-traveling speed which is showed to be the cause for the variability in height and length for wave
groups and the probability for individual high waves.
Output from the wave models provides sea state description in a form of the two-dimensional wave
spectrum but does not give any information about the instantaneous position of the sea surface in a given sea
state. To obtain the latter numerical wave models describing short-term variations of water surface elevation
need to be applied. A review of these models for deep, intermediate and shallow water has been presented
by the previous Committee I.1 (ISSC, 2012, Bitner-Gregersen et al., 2014a). Further, a recently issued
valuable book on wave modelling of Massel (2013) provides an overview of existing wave models. In the
present report we concentrate on modelling of extreme and rogue waves.
Extreme and rogue (called also freak or abnormal) waves have been studied extensively in the past few
decades and attention to these waves has remained also during the period of this Committee. Osborne,
(2010) and Osborne (2013) suggested to group nonlinear waves to:
Population IStokes waves, nonlinear and steep

Population IIUnstable packets, steep with a narrow band wave spectrum being able to be indetified by
the Benjamin-Feir Index, BFI, see (Onorato et al., 2006a).
Figure 2. The wave profile time series during the Andrea storm as measured by the laser within 20 min with 5 Hz
sampling frequency at the Ekofisk field 9 November 2007, after (Magnusson and Donelan, 2013).
Sea states responsible for formulations of these unstable packets and occurrence of abnormal waves are
characterized by large steepness and a narrow wave spectrum, both in frequency and direction. An
example of an abnormal wave, being much higher and steeper than the surrounding waves in the record is
shown in Figure 2.
A number of studies addressing rogue waves have been conducted theoretically, numerically,
experimentally and based on field data. The occurrence of rogue waves, their generation mechanism, and
detailed dynamic properties are now becoming clear. The recent state-of-the-art review on mechanisms
responsible for generating of these waves can be found in (Onorato et al., 2013), see also e.g. (BitnerGregersen et al., 2014c). They can be classified as follows:
linear Fourier superposition (frequency or angular linear focussing)

wavecurrent interactions
crossing seas
quasi-resonance nonlinear interactions (modulational instability)
shallow water effects
wind.
25
As demonstrated in the EC EXTREME SEAS project (EXTREME-SEAS, 2013) the numerical codes
used satisfactory for description of rogue waves today include: NLS (NonLinear Schrdinger) equation
[2D, 3D], Dysthe model [2D, 3D], HOSM (High-Order Spectral Method) [2D, 3D] and Conformal
Method [2D] (Figure 3). These codes have been applied and enhanced in EXTREME SEAS and validated
towards field and laboratory data. Numerical wave data generated by them compare well with the model
test data, e.g. (Toffoli et al., 2011a), (Slunyaev et al., 2012), (Oberhagemann J, 2012), (Bitner-Gregersen
and Toffoli, 2012a) and they are capable to reproduce the field data, e.g. (Bitner-Gregersen et al., 2014c).
It needs to be mentioned that forcing sources such as wind and wave breaking are not included in the
codes presented in Figure 3.
Figure 3. The flowchart of nonlinear wave models, after Slunyaev, (EXTREME-SEAS, 2013).
In the last decade most of the attention was given to the formation of rogue waves due to quasi-resonance
nonlinear interactions referred to as modulational instability. It has been shown that sea states responsible
for occurrence of modulational instability in deep water are characterized by high steepness and a narrow
wave spectrum, both in frequency and direction, and can be identified by the Benjamin Feir Index, BFI, (see
e.g. (Onorato et al., 2013)). Such sea states can be addressed as Rogue Sea States (M. Onorato, pers. comm.,
(EXTREME-SEAS, 2013)). The BFI is a measure of the relative importance of nonlinearity and dispersion.
It can be defined as BFI = (kpHs/2)/(/p), where kpHs/2 is the wave steepness (kp is the wavenumber at the
spectral peak) and /p is the frequency spectral bandwidth ( is the halfwidth at half-maximum of the
spectrum and p is the spectral peak frequency). It should be noted that the above definition of BFI is valid
for stationary conditions (A. Slunyaev, pers. comm., (EXTREME-SEAS, 2013)). Provided the wave field is
sufficiently steep, narrow banded and unidirectional, random waves are expected to become unstable when
BFI = O(1). This results in increased probability of occurrence of rogue waves associated with an
enhancement of non-Gaussian properties of the surface elevation, as shown by several authors, e.g. (Onorato
& Proment, 2012), (Xiao et al., 2013). Since 2012 several studies bringing further insight in the physics of
rogue waves have been carried out.
Stochastic numerical simulations (unidirectional, strongly nonlinear) of sea states characterized by the
JONSWAP spectrum have been conducted within the strongly nonlinear approach by Sergeeva and
Slunyaev (2013). The dominating horizontal asymmetry of rogue waves is obtained in severe situations: the
rare slope is usually higher than the front slope. The geometry of maximum wave shapes is analyzed by the
authors with the focus on issue how the extreme waves look like.
Slunyaev and Shrira (2013) have studied the nonlinear stage of modulational instability in detail by
means of fully nonlinear numerical simulations. The maximum attainable wave crest amplification (for
particular initial conditions, the cases leading to breaking are not concerned) is found to be about 4.2, wave
trough amplification slightly exceeds 3, and wave height amplification is slightly less than 3. The universal
maximum wave and wave group are formulated, and peculiarities of the wave shape are calculated and
analyzed.
26
Within the weakly nonlinear theory the nonlinear stage of the modulational instability has been described
by analytic breather solutions of the nonlinear Schrdinger equation (Chabchoub et al., 2012, Slunyaev et
al., 2013d). These solutions (single and many co-phased modes of the modulational instability) are tested in
the laboratory tank, and satisfactory agreement is obtained between the measurements and the theory. The
strongly nonlinear simulations of potential hydrodynamic equations exhibit very good agreement with the
measurements while the weakly nonlinear theory (nonlinear Schrdinger equation) is a reasonable first
approximation for nonlinear wave groups.
A few rogue wave events, which have been registered in-situ (the North Sea) by means of time-series of
surface displacement, have been reconstructed with help of numerical simulations of approximate
asymptotic equations by Slunyaev et al. (2014). The reconstruction procedure has been verified implicitly
by means of strongly nonlinear simulations. It is shown that the reconstruction may be reliable for the period
of about 10 min.
It is established that the directional spreading of wave energy weakens effects related to modulational
instability when one wave system is present, for references see e.g. (ISSC, 2012). Onorato et al., (2006b)
and Osborne (2010) have shown that the modulational instability and rogue waves can be triggered by a
peculiar form of directional sea state, where two identical, crossing, narrow-banded random wave systems
interact between each other. Such results have been confirmed through recent numerical simulations of the
Euler equations and experimental work carried out at the MARINTEK Laboratories (Toffoli et al., 2011b).
It was observed that extreme events are more likely to occur when 40o < < 60o confirming the findings of
Onorato et al. (2006b).
The study of Bitner-Gregersen and Toffoli (2014) based on hindcast data and numerical simulations
carried out by the HOSM method show that rogue-prone crossing sea states can occur in the ocean. Further,
the authors demonstrate that although directionality has an effect on occurrence of extreme waves in
crossing seas, extreme waves can occur not only for narrow-banded wave directional spreading but also
when it is broader. The most critical condition for occurrence of rogue waves in crossing seas is associated
with energy and frequency of two wave systems while the angle between the wave systems and directional
spreading will decide how large extreme waves will grow. The 40 degree angle and narrow-banded
directional spreading is generating the largest waves. It is interesting to note, that wind sea and swell wave
systems registered during the Louis Majesty accident Cavaleri et al. (2012) had approximately the same
wave energy and frequency but a typical wave directional spreading.
The accident to the Louis Majesty ship took place in the Mediterranean Sea on March 3, 2010 (Cavaleri
et al., 2012). The ship was hit by a large wave that destroyed some windows at deck number five and caused
two fatalities. Using the WAM wave model, driven by the COSMO-ME winds, a detailed hindcast of the
local wave conditions has been performed. The results have revealed the presence of two comparable wave
systems characterized by almost the same frequency (around 0.1Hz) and significant wave heights of
approximately 3.5 m. The total significant wave height, Hm0, at the time of the accident was estimated
around 5 m. These sea state conditions are discussed by the authors in the framework of a system of two
coupled Nonlinear Schrdinger (CNLS) equations, each of which describe the dynamics of a single spectral
peak. Even though, due to the lack of measurements, it is impossible to establish the nature of the wave that
caused the accident, it has been shown that the angle between the two wave systems during the accident is
close to the condition for which the maximum amplitude of the breather solution is observed (40o < < 60o).
Waseda et al. (2013) have revisited a well-studied marine accident case in Japan in 1980 (Onomichi
Maru incident) and hindcasted the sea states using both the DIA and SRIAM nonlinear source terms in the
wave spectral model. The findings indicate that the temporal evolution of the directional spreading and
frequency bandwidth agree reasonably well between the two schemes and therefore most commonly used
DIA method is qualitatively sufficient to predict freakish sea state. The authors show that in the case of
OnomichiMaru, a moving gale system caused the spectrum to grow in energy with limited down-shifting
at the accident site. The unimodal wave system grew under strong influence of local wind with a peculiar
energy transfer. This conclusion contradicts the marine inquiry report speculating that the two swell systems
crossed at the accident site.
When studying rogue waves the information given by the hindcast and the higher order solutions can be
utilized and their use is encouraged. The complementary nature of the two types of models has been pointed
out by the ISSC 2009 I.1 Official Discusser Prof. H. Tomita, ISSC (ISSC, 2009). As discussed by BitnerGregersen et al. (2014c) a spectral wave model (e.g. a WAM-like model) provides sea state description only
in a form of the two-dimensional wave spectrum withbout giving any information about the instantaneous
position of the sea surface in a given sea state. Note also that it accounts for wind forcing and resonant wave
interactions but not for quasi-resonant interactions, which are responsible for occurrence of modulational
instability and hence rogue waves (Onorato et al., 2013). Phase resolving wave models, on the other hand,
provide the water surface elevation from which statistical properties of individual waves can be extracted
and include quasi-resonant interactions. Further, these nonlinear wave models allow simulation of a wave
27
record for a required time duration and by repeating the 17-30 minute simulations sufficient number of
times significant reduction of sampling variability in estimated sea surface characteristics and their
probability of occurrence can be achieved.
Bitner-Gregersen et al. (2014c) have applied coupling of the wave spectral model WAM and the
numerical nonlinear wave model based on the Euler equations, solved with the HOSM method, proposed in
West et al. (1987), to investigate statistical properties of surface oscillations during the particularly severe
Andrea storm, which crossed the central part of the North Sea on November 8th9th, 2007. The analysis
shows that when the Andrea storm is passing the North Sea rogue waves can be expected in several
locations, not only at Ekofisk where the Andrea wave was recorded. Rogue waves are deducted during the
storm development; when storm builds up and when it decades, in the location considered by the study.
They are not observed when the storm reaches the largest Hs. The proposed approach for coupling the wave
spectral model with the nonlinear phase resolving model can be considered to be used for forecasting
purposes. Although these models are computationally intense, the great advance in enhancing computer
power has made the coupling between them feasible.
Spatial variation of nonlinear wave groups with different initial envelope shapes has been theoretically
studied by Zhang et al. (2014a). The results have confirmed that the simplest nonlinear theoretical model is
capable of describing the evolution of propagating wave packets in deep water. Further, three groups of
laboratory experiments run in the CEHIPAR wave basin have also been investigated and systematically
compared with the numerical simulations of the nonlinear Schrdinger equation. A small overestimation has
been detected, especially in the set of experiments characterized by higher initial wave steepness but
generally the numerical simulations show a high degree of agreement with the laboratory experiments
confirming again that the nonlinear Schrdinger equation catches the essential characteristics of the extreme
waves and provides an important physical insight into their generation.
Forcing sources such as wind and wave breaking have started to be included in the numerical codes.
Onorato and Proment (2012) considered the effect of the wind and the energy dissipation on the nonlinear
stages of the modulational instability by mapping the forced/damped nonlinear Schrdinger equation into
the standard NLS with constant coefficients. The results give some insights on the effects of wind and
dissipation on the formation of rogue waves. The authors show that the effect of wind/dissipation has an
impact on the modulational instability; in particular, an initially stable (unstable) wave packet could be
destabilized (stabilized) by the wind (dissipation). Xiao et al. (2013) proposed the energy dissipation model
in HOSM which is based in a low-pass filter in the wavenumber space. The energy dissipation process due
to breaking is simulated by filtering the frequencies higher than the frequencies given by the two filter
parameters established by comparison the HOSM simulations with measurements.
CFD (Computational Fluid Dynamics) methodology is getting increasing attention in modelling of water
waves, particularly extreme waves and wave breaking. The numerical simulation of multiphase and
multicomponent fluid flows is a challenging task in CFD. Iafrati et al. (2013) use direct numerical
simulation of the Navier-Stokes equations for a two-phase flow (water and air) to study the dynamics of the
modulational instability of free surface waves and its contribution to the interaction between the ocean and
atmosphere. The study shows that if the wave steepness of the initial wave exceeds a threshold value, wavebreaking events are observed and the formation of large-scale dipole structures in the air. In the case of
breaking due to modulational instability the dissipation of the energy is mostly concentrated in the air side.
Simulations carried out correspond to the propagation of waves without the presence of external wind; the
consequences of wind on the generation of vorticity during breaking event are under investigation.
Chella et al. (2013) investigated the wave breaking process over a submerged reef. They use a two-phase
numerical model, which solves the flow problem for air and water simultaneously. The Navier-Stokes
equations are solved on uniform Cartesian grids in two dimensions. A staggered grid is used for the
computation with the velocities defined at the cell edges and the pressure at the cell centres. The study
focused on simulating breaking waves over a submerged reef with differnt slope and on determining the
breaker depth and breaker height indices for various values of wave steepness and crest submergence.
Examination of the wave profile during wave breaking has shown that most of the energy is concentrated at
the crest of the breaking wave. The numerical results of the study are in reasonable agreement with the
experimental results and are consistent with the principles of wave hydrodynamics.
A new Lattice Boltzmann method (LBM) has been developed by Banari et al. (2013) in order to simulate
efficiently multiphase flows with high density ratios and to study complex air-sea interaction problems, such
as wind wave breaking and related sea-spray generation. The method is only currently implemented in a
two-dimensional (2D) framework but can be extended to 3D.
Toffoli et al. (2013a) present a laboratory experiment in a large directional wave MARINTEK basin
carried out to investigate the instability of a plane wave to oblique side band perturbations in finite water
28
depth. Experimental observations are supported by numerical simulations and confirm that a carrier wave
becomes modulationally unstable even for relative water depths kph < 1.36 (with k the wavenumber of the
plane wave and h the water depth), when it is perturbed by appropriate oblique disturbances. The authors
state that the results present evidence that the underlying mechanism is still a plausible explanation for the
generation of rogue waves in finite water depth. These findings are confirmed by Fernandez et al. (2014)
using direct numerical simulations of the Euler equations (HOSM) but the results, nonetheless, indicate that
modulational instability cannot sustain a substantial wave grow for kh < 0.8.
Using a Boussinesq model with improved linear dispersion, Gramstad et al. (2013) showed numerical
evidence that bottom non-uniformity can provoke significantly increased probability of rogue waves as a
wave field propagates into shallower water, in agreement with the earlier experimental results.
Kinematics of extreme waves has been compared versus extreme kinematics of all waves and with
predictions of 3-order and 5-order asymptotic nonlinear theory by Sergeeva and Slunyaev (2013). Though in
general the significant amount of data well correspond to each other, some difference between kinematics of
extreme waves and extreme kinematics is emphasized. Rogue waves are typically characterized by large values
of velocities, but high velocities do not necessarily correspond to rogue waves.
Toffoli et al. (2012) have carried out a detailed analysis of the surface wave kinematics of random
directional waves using HOSM. A number of sea states with different wave steepness, spectral bandwidth
and directional spreading were considered to address the combined effect of high order nonlinearity and
directional spreading on wave kinematics. Because the model is not capable to model breaking effects, input
sea states were defined so that an individual wave does not overcome the steepness for breaking onset.
Despite this limitation, the numerical simulations offer a complete overview of the velocity field with the
possibility to investigate the statistical properties of the orbital motion. The simulations revealed that the
growth of large amplitude waves as an effect of modulational instability leads to a significant departure of
the horizontal velocity from linear predictions which increased up to 60% of the initial (linear) condition for
unidirectional wave field. As short directional spreading increases, departure from the Gaussian distribution
gradually diminishes and eventually vanishes for sufficiently broad directional spreading.
The wave runup in narrow bays leads to larger runup heights than in the situation of a plane geometry.
This natural observation is confirmed and described theoretically by Rybkin et al. (2014). Touboul and
Pelinovsky (2014) have shown that the bottom pressure distribution is much improved (with respect to
linear description) using the GreenNaghdi model. Solitonic waves, travelling or fully reflected at a wall
have been considered as a particular case in the study. The fully nonlinear potential equations are used to
control the efficiency of the description. The GreenNaghdi model is found to predict well the bottom
pressure distribution, even when the quantitative representation of the runup height is not satisfactorily
described.
The interest in polar regions has brought attention to waveice interactions. Massel and Przyborska
(2013) proposed theoretical models of the surface waves generated by calving glaciers. Four case studies of
the ice blocks falling into water, representing selected modes of the glacier calving, and corresponding
water surface oscillations have been studied. The modes included: cylindrical ice block of small thickness
impacting on water, ice column sliding into water without impact, large ice block falling on water surface
with pressure impulse and ice column detaching from the glacier wall and falling on water surface. The
carried out calculations show that the glacier calving with pressure impulse creates higher oscillations than
the case of calving with ice column sliding into water. High waves are also generated when ice column
detaches from the glacier wall and impact on water surface. The developed models can be useful for
estimation of the wave amplitude as a function of distance from the glacier wall as well as a function of time
from the impact at a given location.
Wave-current interaction is reported under special topics in Section 5.2.
3.2.2
Experimental description of waves
Extensive laboratory tests of extreme and rogue waves have been carried out by two research projects, EC
EXTREME SEAS and ShorCresT JIP (continuation of the CresT JIP), and the EC-Hydralab IV program.
Some tests of the EC-Hydralab IV program were utilised by EXTREME SEAS.
In (EXTREME-SEAS, 2013) the model tests have been carried out in the tank of Technische Universitt
Berlin (TUB) and the Spanish basin Canal de Experiencias Hidrodinamicas de El Pardo (CEHIPAR). The
project utilized also model tests data from experiments carried out world-wide in which the project partners
had participated. The first phase of the tests consisted of generating the target deterministic wave sequences
with inserted extraordinarily large waves and irregular waves with rogue waves (unidirectional and with
varying directional spreading). Measurments, apart from water surface, recorded also kinematic
characteristics of these waves. In the second phase of the model tests ship behaviour in extreme and rogue
29
waves have been studied. Four ships have been investigated by the project: the LNG Tanker,
Product/Chemical Tanker, Cruise Ship and Large Container Ship. The ships were instrumented to measure
motions and wave induced loads. Both extraordinary rogue waves: the Draupner wave and the Andrea wave
have been reproduced in the model tanks, the Draupner wave in the TUB tank while the Andrea wave (see
Figure 2) in the CEHIPAR basin. At the Technical University of Berlin breather solutions of the Nonlinear
Schrdinger (NLS) equation have been successfully produced with help of the University of Torino and
used for the first time in sea-keeping tests (Clauss et al., 2012), opening up new perspectives in the
methodology of examining offshore structures and ships against rogue waves.
Model tests in the CresT and ShortCresT JIP projects have been carried out in the MARIN and Imperial
College basins and were addressing offshore structures. The ShortCresT investigations included long
crested and short crested waves (short crested waves at different scales) and frequency spectra of field and
basin waves. Buchner and Forristal (2012) observed in ShortCresT that short crested basin waves and field
waves show very similar behaviour. ShortCresT investigations of nonlinear waves are generally confirming
the findings of EXTREME SEAS. The final results of the ShortCresT project will be published in the
OMAE 2015 Conference Proceedings. A summary of the findings is given by Hennig et al. (2015).
Short intense long-living nonlinear wave groups (with steepness up to ka = 0.3 and a couple of
oscillations in the group) have been shown to exist in laboratory conditions (unidirectional propagation,
deep water flume), (Slunyaev et al., 2013b). They are the strongly nonlinear analogue of the envelope
solitons of the integrable nonlinear Schrodinger equation. The measured groups agree very well with the
strongly nonlinear simulations of the potential equations of hydrodynamics. It is found that the nonlinear
rise of the solitary group velocity is even larger than one of the uniform Stokes wave. These nonlinear
groups form extreme wave patterns.
The space-time evolution of high wave groups in crossing seas have been investigated in the
MARINTEK basin by Santoro et al. (2013). The experimental data have been supported by applying Quasideterminism theory of Boccotti given that a high crest takes place in a fixed point in the basin. It is observed
that the high waves group is given by the superposition of two wave groups, associated, one with the lowfrequency component of the frequency spectrum, and the other with the high-frequency one. Further, it is
shown how in crossing seas the change in direction of one system affects the evolution of the related group,
without any influence on the evolution of the other one. The effect of the change in the wave spectrum on
wave profiles have also been analysed.
Yan et al. (2013) performed a laboratory experiment on the instability of Stokes wave trains with large
steepness in finite water depths. Two class instabilities of Stokes wave, quartet interaction and quintet
interaction, were observed. It is found that the evolution of crescent wave pattern is affected by the
development of quintet interaction. The larger the relative water depth, the more quickly the growth of the
instability is observed. The impact of wave steepness for the occurrence of the competition is examined by
applying linear instability analysis of a Stokes wave. The results show that the observed Benjamin-Feir
instability is two dimensional.
The dependence of runup heights on the incident wave shape (sinusoidal wave or solitary wave,
particular polarity and wave asymmetry) has been studied. The theoretical formulas obtained in
Didenkulova et al. (2014) and Rybkin et al. (2014) have been confirmed in experiments in the large wave
tank of the Hannover University and in the wave facility of the Caen University (Ezersky et al., 2013a,
Ezersky et al., 2013b).
Results of model tests carried out in the MARINTEK model basin and used to establish statistics of
wave parameters are reported in Section 3.2.3. Several experiments addressing wave-current interaction
have taken place since 2012 and are reported in Section 5.2 (Special topics).
One of the main issue when performing experiments is reduction of sampling variability, the uncertainty
due to limited number of data. Sampling variability is an epistemic uncertainty and can be reduced opposite
to the intrinsic uncertainty of sea surface elevation which is always present. For stationary meteorological
conditions, due to randomness of sea surface, wave parameters derived from a wave record will depend on
which part of a wave record is used in an analysis as well as on the length of a wave record (BitnerGregersen and Hagen, 1990, Bitner-Gregersen and Magnusson, 2014). An error introduced by the limited
length of a wave record is an epistemic uncertainty and can be reduced by increasing duration of wave
measurements/numerical simulations. Ideally a wave record should be infinite to eliminate sampling
variability. Numerical simulations of water surface represent a good support to field and model tests as they
allow reducing sampling variability by increasing duration of simulations when wave input is kept constant
and intrinsic variability accounted for. This is more difficult in nature, where stationarity of sea states is an
issue, and in model tests due to the costs associated with repeated model test runs.
30
Model tests need to be repeated a sufficient number of times to reduce sampling variability otherwise
bias results maybe obtained, does not matter how accurate instruments used in an experiment are, see
(Bitner-Gregersen et al., 2014b, Bitner-Gregersen and Magnusson, 2014). Inaccuarcy introduced in test
results due to the limited number of data can be accounted for in load and response analysis in terms of
distribution functions or standard deviations. The bootstrapping technique can be used to establish these
uncertainties.
It needs also to be mentioned that to reflect intrinsic variability of linear and nolinear numerical
simulations of surface elevation as well as waves generated in model basins, wave models and laboratory
wave makers have to account for random ampliture and phase of a wave field. Further, modulational
instability occur typically after 10-30 wave lengths from a wavemaker thus a scale of model tests should to
be considered carefully to be able to generate abnormal waves in a basin.
3.2.3
Statistical description of waves
Wave height and wave crest are ones of the most important wave characteristics for engineering
applications. A recent review of wave crest distributions can be found in Bitner-Gregersen et al. (2014b).
Comparisons of individual wave height and crest distributions with the distributions of laboratory data for a
wide range of conditions, including highly nonlinear steep sea states and sea states with extensive and
intensive wave breaking have been conducted as part of the EC EXTREME SEAS and ShorTCresT JIP
projects. It has been demonstrated that linear and second-order models are not able to capture very nonlinear
waves like rogue waves. Some recent papers from these projects, developed since 2012, are reviewed
below. The review includes also investigations carried outside the projects.
In Petrova and Guedes Soares (2014) statistical properties of wave crest, height and trough are
investigated using experiments generated in the deep water MARINTEK basin with mixed crossing seas.
The findings are compared with the previous results from the same experiment. Waves have been generated
using bimodal spectra following the model of Guedes Soares (1984). The observed statistics and
probabilistic distributions exhibit, in general, increasing effects of third-order nonlinearity with the distance
from the wavemaker. However, this effect is less pronounced in the wave systems with two following wave
trains than in the crossing seas with identical initial spectral characteristics.
Deep-water waves with different initial steepness able to trigger modulational instability and measured in
the MARINTEK wave basin have been studied by Cherneva et al. (2013). The authors compare the statistics
of wave heights, crest and trough amplitudes observed in the basin with a variety of theoretical
approximations based on GramCharlier expansions. The results indicate that the theoretical
approximations describe the empirical distributions reasonably well, for the most part. The study also shows
that the zero-up- and zero-down crossing heights of the largest waves observed in the tests do not exceed
MicheStokes type upper limits.
Zhang et al. (2013a), using the NLS equations and experimental data from the CEHIPAR wave basin,
have confirmed the early findings, the occurrence of the modulation instability resulted from the quasiresonant four wave interaction in a unidirectional sea state in deep water, can be indicated by the coefficient
of kurtosis. The latter has shown correlation with the extreme wave height. The modified EdgeworthRayleigh distribution has been used to approximate the wave heights. The authors have related also some
statistical properties of the maximum wave heights in different sea states with the initial Benjamin-Feir
Index.
Having in mind engineering applications (Bitner-Gregersen and Toffoli, 2012b) proposed a 2-parameter
Weibull crest distribution derived based on experimental directional wave data (included rogue waves and
wave breaking) collected in the MARINTEK basin. The analysis has been supported by HOSM numerical
simulations. The related parameters have been parameterised as a function of a general version of the
Benjamin-Feir Index for directional sea states. On the whole, the proposed distribution captures better the
tail of the crest distribution than the (Forristall, 2000) crest distribution.
In the ShortCresT JIP, departures from the Forristall crest distribution were also found for steep sea
states by comparison with crest distributions from measured waves in the laboratory at Imperial College;
and a new parameterization that accounts for the additional nonlinear effects, including breaking has been
developed (Swan and Latheef, 2014).
Statistics of new field measurements of wave height and crest elevation measured in the North Sea
during a storm in December 2012 are presented by Gibson et al. (2014). The water surface elevation was
recorded by Saab WaveRadar REX instruments mounted on eight fixed-jacket platforms in addition to a
Datawell Directional Waverider buoy. 19 freak waves, following the definition of Haver and Anderson
(2000), occurred during this storm. An easterly sea state which peaked well in excess of the 100-year wave
height has been generated by the storm. The study shows that the significant steepness and spectral
bandwidth during the storm remain almost constant and the measured crest elevations and wave heights are
31
in good agreement with the Forristall and Boccotti distributions. However, for the wind-speed larger than 25
m/s the measured crest elevation lies above the second-order Forristall distribution.
Christou and Ewans (2014) took advantage of the development of a large in-situ measurement wave
database to jointly analyze sea-state parameters, environmental conditions and local characteristics so as to
provide an insight into the behaviour of rogue waves. The database, constructed using data sets of qualitycontrolled measurements from fixed sensors mounted on offshore platforms, mostly located in the North
Sea, contains 122 million individual waves, of which 3,649 are rogue waves. Analysis of sea-states
parameter showed that rogue-waves induced an increase of the kurtosis value. Examination of metocean
conditions for each sea-state suggested that there was not any particular combination of wind-sea, swell,
wind or current that is particularly conducive to the formation of rogue waves. Finally, a wavelet analysis
conducted on the rogue wave samples presented evidence to suggest that rogue waves are merely
extraordinary and rare occurrences of the normal population that are caused by dispersive focusing.
It is interesting to note that the analyses carried out by Christou and Ewans (2014) of measured data from
the Kvitebjrn platform did not find a link between low spreading and the occurrence of rogue waves in the
measured data. Opposite, Waseda et al. (2011) provided evidence of an increased likelihood of rogue waves
at the Kvitebjrn platform in the North Sea when directional spreading was relatively low (approximately
7.6 degrees narrower compared to observations without rogue waves present), and this corresponded to a
wave spreading of 30 degrees.
Wang et al. (2014) analysed preliminary one year observed wave data from Jiangsu (coastal water),
China. The horizontal symmetry, vertical symmetry, wave steepness, kurtosis and the BFI index have been
studied. Most of the rogue waves, which are not high, are present in 5 intermediate water depth what results
in the obscure correlation between BFI and rogue waves. The authors postulate that two different types of
rogue waves are present in the considered area.
The statistics of the elevation estimated by Waseda et al. (2014) from the buoy position record of the
JKEO (Jamstec Kuroshio Extension Observatory), NKEO (New Kuroshio Extension Observatory)
observation sites and the Hiratsuka observation tower has shown that the probability density function (PDF)
is nearly Gaussian, and the PDF of the extremum is well approximated by the analytical formula of
Cartwright and Longuet-Higgins (1956).
Babanin (2012) argues that different scenarios of wave evolution should be represented by different
wave height/crest probability distributions. Otherwise, the residual scatter is inevitable and will not be
improved even when the databases are enhanced and measurement accuracy is improved. Selection of such
scenarios needs to be based on the understooding of wave physics.
It is well recognized today that the effect of modulational instability is gradually suppressed when the
wave energy spreading increased and the second order wave theory is adequate to describe the statistical
behaviour of ocean waves up to a particular probability level (ISSC, 2012).
Stansberg (2012) applied a Hilbert envelope approach (energy envelope) to describe the wave groups
and showed that the sampling variability of the extreme energy envelope is much higher than for linear
phenomena, due to the quadratic nature. Note, that the Hilbert transform disregard wave nonlinearity,
being crucial when rogue waves are considered, see (ISSC, 2012).
The nonlinear and nonstationary properties of a special field wave record have been analysed by
Cherneva and Guedes Soares (2014) with the Wigner spectrum with the ChoiWilliams kernel. The wave
time series including the Andrea rogue wave recorded at the Ekofisk field has been used in the study. The
ability of the Wigner spectrum to reveal the wave energy distribution in frequency and time is demonstrated.
The results are compared with previous investigations for different sea states and also the state with
Draupners abnormal New Year wave.
Despite these recent achievements regarding dynamic properties of rogue waves a consensus on
probability of occurrence of rogue waves has not been achieved yet. Probability of occurrence of rogue
waves is related to mechanisms generating them. Since 2012 occurrence of rogue waves due to some of
these mechanisms have been studied.
To investigate the frequency of occurrence of seas states which may trigger the modulational instability
in deep water in the North Atlantic, Bitner-Gregersen and Toffoli (2012a) have used hindcast data from a
few North Atlantic locations generated by Oceanweather Inc. and European Centre for Medium-Range
Weather Forecast (ECMWF). The Oceanweather Inc. hindcast wind and wave covered the period 1988
1998 and were sampled every 3 h. Data have been post-processed by Shell using the program APL Waves
Hanson and Phillips (2001) for the partitioning of 3-D spectra (i.e. directional wave spectra) into separate
peaks. The ECMWF wind and wave data covered the period 20012009 and were archived at a sampling
frequency of 6 h. Results revealed rogue-wave-prone sea states can occur in the North Atlantic (the North
Atlantic wave climate is used for design of ships) more often than once in the 20/25-yr return period. Also
the highest sea state within the 10-yr time period analysed (Hs > 15 m) is characterised by kpHs/2 = 0.13, the
conditions which may trigger the modulational instability.
32
Also, as shown by Bitner-Gregersen and Toffoli (2014), rogue-prone crossing wave systems responsible
for the generation of abnormal waves can occur more often than once in the 20/25-yr return period in the
North Atlantic and the North and Norwegain Sea, primarily in low and intermediate sea states. Their
occurrence is location specific, depending strongly on local features of wave climate. This type of sea states
have not been found in locations where wind sea and swell components, or several swell components, are
well separated, characterized by significantly different spectral periods.
Nikolkina and Didenkulova (2012) divide the area of occurrence of rogue events into three zones: deep
water area, shallow-water regions and the coast. The shallow water zone is defined as the sea areas with
depths <50 m while deep waters are associated with water depths exceeding 50 m. The coastal rogue events
include unexpected and hazardous waves of extreme height or run-up that occurred at the shoreline. The list
of rogue and rogue-like events has been completed based on the literature review, existing databases and
information published in the media. The validity of the event (true or possible), its classification in terms of
its location (shallow, deep water or coast zone) and estimated Hs are given in the list developed by the
aurhors. The investigations have shown that observed rogue waves are widely spread around the World,
however, some regions seem to be more affected. The great number of rogue wave observations in shallow
waters (51% for all observed events and 38.5% for the selection of true events only) and at the coast (40%
for all observed events and 50% for the selection of true events only) can be explained by the dense
population in these areas and/or a heavy marine traffic in coastal waters. The annual variability of rogue
wave changes significantly from year to year.
On the basis of available information about eyewitness observations of rogue waves it is found later by
Didenkulova et al. (2013c) that the rogue waves occur with similar probability in situations when the
longitudinal modulational instability is efficient or inable (deep or shallow waters). The relative threshold
depth which splits the situations to shallow and deep water is about 20 m. Opposite, Wang et al. (2014)
found that in coastal waters of Jiangsu, China, the occurrence probability of rogue waves is similar to the
one predicted by the Rayleigh distribution.
Oh and Jeong (2013) have studied the maximum significant wave heights and periods observed at the
nine measurement stations at the Korean coast of the East Sea. The aim of the study has been identifying
meteorological conditions which in several recent years generated extremely high waves at the Korean coast
of the East Sea causing severe coastal disasters almost every winter season. The authors found that the
reason for appearance of the high waves were extra-tropical low pressure systems that had been rapidly
developing in the East Sea. The study has contributed to better understanding of mechanism forming such
strong low pressure systems and spatial evolution of them. To enhance predictability of these systems
further investigations are needed.
Aarnes et al. (2012) investigated the 100-yr return value estimates of significant wave height using the
recent Norwegian Reanalyzes 10 km (NORA10) hindcast database developed by the Norwegian
Meteorological Institute and covering the northeast Atlantic. They considered three different stationary
models commonly applied in extreme value statistics: the generalized extreme value (GEV) distribution, the
joint GEV distribution for the largest-order statistic (LOS), and the generalized Pareto (GP) distribution. In
general, the Hs 100-yr return value estimates differ by less than 5%, with local discrepancies peaking
around 20%. In these areas the annual maximum and the peaks-over-threshold methods were found to
outperform the largest-order statistics. However, no model has been found to be superior in all cases.
Extreme metocean conditions occur simultaneously in a storm process. Different techniques for
modelling simultaneous data have been investigated recently. Based on a bivariate equivalent maximum
entropy distribution, Dong et al. (2013b) estimated joint extreme parameters of the concomitant wave height
and wind speed at a site in the Bohai Sea. Tao et al. (2013b) hindcasted by a numerical model the annual
extreme wave height and corresponding wind speed at a platform in Bohai Bay for the period 1970 to 1993.
Bivariate normal copula and Frank copula are utilized to construct joint distribution of these two random
variables. Applicability of copula models to estimate joint probabilities needs still further investigations.
3.2.4
Spectral description of waves
Wave spectra represent a key input to the design process. Empirical formulations of waves spectra were
used extensively in engineering applications and this situation remains. Typical forms for the frequency
spectrum (see e.g. (Amurol et al., 2014)) are modified versions of the Pierson and Moskowitz (PM)
spectrum and the JONSWAP spectrum for unimodal conditions (one wave system), and the (Ochi and
Hubble, 1976) and (Torsethaugen, 1993, Torsethaugen, 1996) spectrum for bimodal sea states (two wave
systems present), see (DNV, 2014). The two wave system wave spectrum of Guedes Soares (1984) can also
be used.
33
The maritime industry has used traditionally the PM spectrum but recently also other empirical spectra
have started to be applied like the JONSWAP spectrum and a double peak spectrum particularly for
evaluation of ship operational criteria. The choice of formulation depends on the site and the conditions
being represented; for example, when the spectrum is being defined for extreme sea states for design, a
unimodal description will usually be most representative. It should be noted that the choice of spectral form
has implications for the spectral levels above and below the spectral peak. There is still uncertainty about the
form of the decay in the high frequency tailthe (modified) PM and JONSWAP formulae have an f-5 tail
decay, while the Torsethaugen (1993) spectrum had originally the f-4 tail which later was modified
(Torsethaugen, 1996) to become a function of Hs and is ranging from f-4 to f-5 (for details see e.g. (BitnerGregersen et al., 2014b)).
Waseda et al. (2014) analysing buoys deep water data in the North West Pacific Ocean in 2009, 2012,
and 2013 have shown that the tail of the averaged spectrum follows the frequency tail f-4, and the significant
wave height and period satisfies the Tobas 3/2 law. The observations compare well with a numerical wave
hindcast.
The universal power law for the spectrum of one-dimensional breaking Riemann waves in shallow water
is justified by Pelinovsky et al. (2013). The spectrum of spatial amplitudes at the breaking time has an
asymptotic decay of k4/3, with corresponding energy spectrum decaying as k8/3. This result remains valid
for arbitrary nonlinear wave speed and may be important for interpretation of measured spectra and
construction of theoretical model spectra of essentially nonlinear waves.
Shell Global Solutions, in association with Ifremer and Oceanweather, formed a Joint Industry Project
WASP (West Africa Swell Project), to analyze and compare the available data on swell and its spectral
description. Measurements were obtained from Shell, Ifremer, Chevron and Marathon. Hindcast data came
from Oceanweather and the NOAA WAVEWATCH model. Sites ranging from Nigeria to Namibia were
considered. The project was initiated in 2001 and has permitted improvement in the knowledge of swell
climatology in West Africa. The full description of the project is given by Olagnon et al. (2004). Some
project results were presented in the OMAE 2013 Conference (Forristall et al., 2013, Olagnon et al., 2013,
Prevosto et al., 2013).
Swell events show a large variety of configurations when they arrive at sites off West Africa after
generation and propagation of waves across the Atlantic Ocean (Prevosto et al., 2013). Within the West
Africa Swell Project (WASP JIP), these different configurations have been described and discussed and the
ability of numerical models to reproduce faithfully their properties has been assessed from comparisons with
in-situ measurements. The numerical wave models were not, at the time of the project, of sufficient quality
to accurately track all the swell events, but the models provided better information on superposition of
multiple swell systems than directional buoys. Since the end of the project the hindcast models have been
improved by the integration of new dissipation and nonlinear interaction source functions and also by the
addition of buoy and altimeter data assimilation. Therefore the comparison of the numerical model
prediction and buoy data is not conclusive. Analyses of some shallow water measurements have indicated
the presence of low-frequency components that could be due to the presence of infra-gravity waves but
further investigations are needed to use these results in engineering applications.
Modeling the dispersion of swell over long distances has indicated that the resulting power spectrum will
have a triangular or lognormal shape (Forristall et al., 2013). Sampling variability makes it difficult to
distinguish between those shapes or JONSWAP or Gaussian forms, but lognormal spectra generally
provided good fits. The models also indicate that the width of the spectrum in both frequency and direction
should be inversely related to the peak frequency.
3.3
Current
The ISSC 2009 I.1 Committee (ISSC, 2009) reported on success of the global and basin-scale ocean
models development with data assimilation under the GODAE (Global Ocean Data Assimilation
Experiment) which opened a new era of operational oceanography. GODAE ended in 2008 and continues
as GODAE Ocean View: https://www.godae-oceanview.org/ (see ISSC, 2012). The 5 years of GODAE
OceanView progress and future priorities were presented at the GODAE OceanView Symposium in
Baltimore, Maryland, 46 November 2013.
As pointed out by the previous I.1 Committee (ISSC, 2012) the GODAE product is global and it
provides boundary conditions for regional and high-resolution models. It does not aim at covering all
time-scales that are required for engineering.
3.3.1
Analytical description of current
Ocean model outputs have been used after the incident at the Deepwater Horizon platform in April 2010 in
the Gulf of Mexico to trace spilled oil in the Gulf Stream as well as to trace debris and radioactive materials
34
after the earthquake and tsunami incidents on 11 March 2011 in north east Japan, see e.g. (Aoyama et al.,
2012, Tsumune et al., 2012). The results are promising, showing that numerical ocean models can be
utilized in these type of analyses although they may give different predictions. In Masumoto et al. (2012)
comparison of predictions of five ocean models can be found when data around Japan are used.
Ocean current models such as HYCOM are used increasingly for engineering applications and have
recently been assessed using data at the location offshore West Africa by Jeans et al. (2013). Jeans and
Wade (2013) have shown that the HYCOM model does not represent dominant features of the local current
regime in the region offshore Namibia. This region is subjected to relatively strong diurnal sea breezes,
which can drive nearly resonant inertial responses in the ocean, which can, in turn, dominate the current
regime. The spatial resolution of the wind field used to drive the global HYCOM model is insufficient to
resolve this critical wind forcing, so the resulting model currents cannot represent the dominant features.
Utilization of current by energy devices has been presented at e.g. the OMAE 2013 and 2014
conferences.
3.3.2
Statistical and spectral description of current
(Jones et al.) has proposed a Monte Carlo simulation method for combining different independent current
components such as tide, through-flow, surge and high-frequency currents which are present in the
Singapore Straits. The suggested model is expanding the horizon of available measured and modelled data
and is facilitating the definition of design current speeds. The number of random picks from the nonexceedence probability distributions of the surge, through-flow and high-frequency components in a given
year for each component is defined by assuming its occurrence rate to be Poisson-distributed around a
known annual mean value. The method gives a reduction in design current when compared to values
derived by multiplying the exceedence probabilities of varying independent contributions directly. The
Kuroshio current is a major global current that flows near the east coasts of Taiwan and Japan. It is a
relatively strong current with typical speeds of 3 to 5 knots at the water surface. Agarwal and McNeill
(2013) present the derivation of extreme two-dimensional (i.e., directional) and planar profiles for Kuroshio
currents at a site in Nankai Trough, Japan; water depth is almost 2000 m. Totally about 6000 currents
profiles measured over six months in 2010 by JAMSTEC have been used in the study. The inverse firstorder reliability method (inverse FORM) and proper orthogonal decomposition (POD) techniqve have been
employed to preserve the directionality in extreme currents. A set of a limited number of extreme N-year
current profiles is derived using the proposed methodology. The introduced methodology for deriving
extreme directional current profiles from measured data seems to give promising results.
Ceccopieri and Silveira (2012) successfully applied statistical and dynamical orthogonal modes in order
to study the vertical variability pattern of the oceanic flows of the So Paulo Plateau of Brazil.
3.4
Sea water level
Several studies have been carried out to project future sea water level changes using GCM (Global Climate
Model) or RCM (Regional Climate Model) models. The findings are summarised in Section 4.1.4. The
degree to which GCM, or, RCM have sufficient resolution and/or internal physics to realistically capture the
meteorological forcing responsible for storm surges is regionally dependent. For example current GCMs are
unable to realistically represent tropical cyclones.
3.5
Ice and snow
Ice and snow models are discussed in Section 4.1.3 dedicated to climate change. Changes in the Arctic
Ocean and Antarctica weather have received widespread attention. The projected climate changes in these
regions are opening new opportunities for the marine industry. Potential seasonal shipping on the Northern
Sea Route, the Northwest Passage and maybe a Transpolar Route, will improve access to many offshore
resources in the Arctic region (see Section 6.2).
4.
CLIMATE CHANGE
During the course of the ISSC 2015 Committee I.1 the Fifth Assessment Report (AR5) summarising the
latest scientific findings regarding climate change has been issued by (IPCC, 2013). A significant
development has taken place since the issue of the Fourth Assessment Report (AR4) (IPCC, 2007),
particularly in the increased use of quantitative statistical measures simplifying synthesis and visualization
of model performance (see e.g. Sahany et al., 2012).
AR5 is confirming the conclusion of AR4, the observed climate changes are due to human activities.
Further, there are large regional variations in observed and projected climate driven changes in
environmental conditions. As in AR4, temperature, sea water level and ice has got more attention in the
AR5 report in comparison to wind and waves, but it is the first time the IPCC is providing an assessment of
35
wave models. The main uncertainties in the understanding of the climate system and the ability to project
changes in response to anthropogenic influences identified by IPPC (2013) include: uncertainties in
observation of changes in the climate system, uncertainties in drivers of climate change, uncertainties in
understanding the climate system and its recent changes, uncertainties in global and regional climate change
models.
It can be mentioned that in December 2014 a global climate conference has been held in Lima where the
participating nations agreed to collaborate to mitigate climate changes.
4.1
New IPPC Scenarios and climate models
The IPCC Fifth Assessment Report (IPCC, 2013) uses four scenarios to describe past and future driving
forces, called Representative Concentration Pathways (RCP). These can be described as four possible
trajectories for future greenhouse gas concentrations in the atmosphere. They include emissions of
greenhouse gases and land cover/use but not socio-economic, technology, demographic and other aspects as
the Fourth IPPC Assessment Report forcing covered. Each scenario leads to a certain change of average
global temperature over the next century (or more).
The four RCPs lead to radiative forcing levels of 8.5, 6.0, 4.5 and 2.6 W/m2, by 2100 and they are named
according to it, i.e. RCP 8.5, RCP 6, RCP 4.5 and RCP 2.6, which has been estimated based on the forcing
of greenhouse gases and other forcing agents. Each of the RCPs covers the 18502100 period and together
they cover the range of forcing found in literature; they are sufficiently separated to allow distinguishable
climate results (about 2 W/m2 was found to satisfy this) and the number of RCPs are manageable. They
include one mitigation scenario leading to a very low forcing level (RCP 2.6), two medium stabilization
scenarios (RCP 4.5/RCP 6) and one very high baseline emission scenario (RCP 8.5).
The main characteristics of four RCPs are summarized by van Vuuren and al. (2011a), Meinshausen
and al. (2011) and Moss and al. (2010), and they are described in more details by Riahi and al. (2011),
Masui and al. (2011), Thomson and al. (2011), Dutour-Sikiric (2013) and van Vuuren and al. (2011b). The
scenario RCP 4.5 is expected to achieve the political target of a maximum global mean temperature increase
of 2C while the scenario RCP8.5 is close to business as usual and believed to give a temperature rise of
4C or more.
Projections of changes in the climate system are made by (IPCC, 2013) using a hierarchy of climate
models ranging from simple climate models, to models of intermediate complexity, to comprehensive
climate models, and Earth System Models (ESMs). These models based on the set of new scenarios of
anthropogenic forcing have been used for the new climate model simulations carried out under the
framework of the Coupled Model Intercomparison Project Phase 5 (CMIP5) of the World Climate Research
Programme. A large number of comprehensive climate models and ESMs have participated in CMIP5, they
form the core of the climate system projections.
Most modes of interannual to interdecadal variability are now capture by the climate models. As in the
AR4 Report, their assessment presents a varied picture and the CMIP5 models only show a modest
improvement over CMIP3 (Coupled Model Intercomparison Project Phase 3), mostly due to fewer poorperforming models (IPCC, 2012, IPCC, 2013). New since the AR4 process-based model evaluation, is now
helping identify sources of specific biases, although the observational record is often too short or inaccurate
to offer full specification of model uncertainty.
Confidence in climate model projections in AR5 is based on physical understanding of the climate
system and its representation in climate models, and on a demonstration of how well models represent a
wide range of processes and climate characteristics on various spatial and temporal scales (Knutti et al.,
2010). A climate models credibility is increased if the model is able to simulate past variations in climate,
such as trends over the 20th century and palaeoclimatic changes.
Longer-term climate change projections are outside the range of historical observations making
assessment of models accuracy difficult. A frequently used approach to improve the climate model
projection is the re-calibration of model outputs to a given observed value, e.g. (Wang and Overand, 2012),
(Mahlstein and Knutti, 2012). Some studies explicitly formulate a statistical frameworks that relate future
observables to climate model output (reviewed in (Knutti et al., 2010) and (Stephenson et al., 2012). Such
frameworks not only provide weights for the mean response to forcing scenarios but also allow the
uncertainty in the predicted response to be quantified (Bracegirdle and Stephenson, 2012).
The following terms have been used by (IPCC, 2013) to indicate the assessed likelihood of climate
projections: virtually certain (99100% probability), very likely (90100% probability), likely (66100%
probability), about as likely as not (3366% probability), unlikely (033% probability), very unlikely (0
10% probability), exceptionally unlikely (01% probability). Additional terms extremely likely (95100%
probability), more likely than not (>50100% probability), and extremely unlikely (05% probability) have
been introduced.
36
4.1.1
Temperature
The observational record of 20th century changes in global surface temperature has been compared to that
simulated by each CMIP5 and EMIC model and the respective multi-model means. For the CMIP5 models,
interannual variability in most of the simulations is qualitatively similar to that observed although there are
several exceptions. The gradual warming evident in the observational record, particularly in the more recent
decades, is also evident in the simulations, with the multi-model mean tracking the observed value closely
over most of the century, and individual model results departing by less than about 0.5oC.
The observed global mean surface temperature (GMST) has shown a much smaller increasing linear
trend over the past 15 years than over the past 30 to 60 years. Depending on the observational data set, the
GMST trend over 19982012 is estimated to be around one-third to one-half of the trend over 19512012.
The reduction in observed GMST trend is most marked in Northern Hemisphere winter. However, the
decade of the 2000s has been the warmest in the instrumental record of GMST.
In the AR4 (IPCC, 2007) it was noted that the largest errors in SST (Sea Surface Temperature) in CMIP3
were found in mid and high latitudes. While this is still the case in CMIP5, there is marginal improvement
with fewer individual models exhibiting serious bias. The inter-model zonal mean SST error standard
deviation is significantly reduced at all latitudes north of 40oS, even though the multi-model mean is only
slightly improved.
The globally averaged combined land and ocean temperature data as calculated by a linear trend, show a
warming of 0.85 [0.65 to 1.06] C, over the period 18802012, when multiple independently produced
datasets exist, about 0.89 [0.69 to 1.08] C over the period 19012012, and about 0.72 [0.49 to 0.89] C
over the period 19512012 when based on three independently-produced data sets. Sea surface temperatures
have also increased. Intercomparisons of new SST data records obtained by different measurement methods,
including satellite data, have resulted in better understanding of errors and biases in the records. For average
annual NH temperatures, the period 19832012 was very likely the warmest 30-year period of the last 800
years (high confidence) and likely the warmest 30-year period of the last 1400 years (medium confidence).
This is supported by comparison of instrumental temperatures with multiple reconstructions from a variety
of proxy data and statistical methods, and is consistent with AR the results.
It is virtually certain that the upper ocean (above 700 m) has warmed from 1971 to 2010, and likely that
it has warmed from the 1870s to 1971. It is likely that the ocean warmed between 700-2000 m from 1957 to
2009, based on 5-year averages. It is also likely that the ocean warmed from 3000 m to the bottom from
1992 to 2005, while no significant trends in global average temperature were observed between 2000 and
3000 m depth from circa 1992 to 2005. Below 3000 m depth, the largest warming is observed in the
Southern Ocean.
These findings strengthens the conclusions from both AR4 (2007) and SREX, Risk of Extreme Events
and Disasters to Advance Climate Change Adaptation, (IPCC, 2012) that it is now very likely that
anthropogenic forcing has contributed to the observed changes in the frequency and intensity of daily
temperature extremes on the global scale since the mid-20th century. In terms of historical trends, CMIP3
and CMIP5 models generally capture observed trends in temperature extremes in the second half of the 20th
century (Sillmann et al., 2013) and there is high agreement that the global distribution of temperature
extremes are represented well by CMIP3 and CMIP5 models. Studies with fixed ice-sheet topography
indicate the increase of SST is greater than 2C but less than 4C (medium confidence) of global mean
surface temperature rise with respect to the pre-industrial level.
It is likely that the GMST anomaly for the period 20162035, relative to the reference period of 1986
2005 will be in the range 0.3C to 0.7C (medium confidence). It is very likely that globally averaged
surface and depth-averaged temperatures averaged for 20162035 will be warmer than those averaged over
19862005. According to (IPCC, 2013) global mean temperatures will continue to rise over the 21st century
under all of the RCPs. From around the mid-21st century, the rate of global warming begins to be more
strongly dependent on the scenario. GMSTs for 20812100, relative to 19862005 will likely be in the 5 to
95% range of the CMIP5 models; 0.3C to 1.7C (RCP2.6), 1.1 to 2.6C (RCP4.5), 1.4C to 3.1C
(RCP6.0), 2.6C to 4.8C (RCP8.5). The Arctic region is projected to warm most (very high confidence). It
is virtually certain that, in most places, there will be more hot and fewer cold temperature extremes as
global mean temperatures increase. Increases in the frequency, duration and magnitude of hot extremes
along with heat stress are expected; but occasional cold winter extremes will continue to occur.
Over the course of the 21st century, the global ocean will warm under all RCP scenarios. The strongest
ocean warming is projected for the surface in subtropical and tropical regions. At greater depth the warming
is projected to be most pronounced in the Southern Ocean. Best estimates of ocean warming in the top one
hundred metres are projected to be 0.6C (RCP2.6) to 2.0C (RCP8.5), and 0.3C (RCP2.6) to 0.6C
(RCP8.5) at a depth of about 1 km by the end of the 21st century. Due to the long time scales of this heat
37
transfer from the surface to depth, ocean warming will continue for centuries, even if GHG emissions are
decreased or concentrations kept constant.
4.1.2
Ice and snow
Realistic historical sea ice represent an important input to climate models (Wang and Overand, 2012),
(Massonnet et al., 2013), (Stroeve et al., 2012), (Massonnet et al., 2013), (Wang and Overand, 2012),
(Overland and Wang, 2013). (IPCC, 2013) reports that most of arge-scale sea ice processes, such as basic
thermodynamics and dynamics, are well understood and well represented in models. However, there are still
challenges in modelling important details of sea ice dynamics and deformation, particularly in small scales,
see e.g. (see e.g. (Hutchings et al., 2011). Currently, sea ice model development is focusing primarily on
more precise descriptions of physical processes including biological and chemical species.
Snow model development for sea ice has lagged behind terrestrial snow models. Lecomte et al. (2013)
introduced vertically varying snow temperature, density and conductivity to improve vertical heat conduction and melting in a 1D model intended for climate simulation. The model has, however, limitations,
many physical processes affecting the evolution of the snow pack, such as redistribution by wind, moisture
transport (including flooding and snow ice formation) and snow grain size evolution, still are not included in
most climate models.
Evaluation of sea ice performance requires accurate information on ice concentration, thickness,
velocity, salinity, snow cover and other factors. The most reliably measured characteristic of sea ice remains
sea ice extent (usually understood as the area covered by ice with a concentration above 15%). Caveats,
however, exist related to the uneven reliability of different sources of sea ice extent estimates (e.g., satellite
vs. pre-satellite observations), as well as to limitations of this characteristic as a metric of model
performance, see (Notz and Marotzke, 2012).
As in AR4, there is very high confidence that the Arctic sea ice extent (annual, multi-year and perennial)
decreased over the period 19792012. The average decrease was likely between 1.3 m and 2.3 m. The rate
of the annual decrease was very likely between 3.5 and 4.1% per decade (range of 0.45 to 0.51 million km2
per decade). The average decrease in decadal extent of annual Arctic sea ice has been most rapid in summer
and autumn (high confidence), but the extent has decreased in every season, and in every successive decade
since 1979 (high confidence). The extent of Arctic perennial and multi-year ice decreased between 1979 and
2012 (very high confidence). The rates are very likely 11.5 [9.4 to 13.6]% per decade (0.73 to 1.07 million
km2 per decade) for the sea ice extent at summer minimum (perennial ice) and very likely 13.5 [11 to 16] %
per decade for multi-year ice. There is medium confidence from reconstructions that the current (1980
2012) Arctic summer sea ice retreat was unprecedented and SSTs were anomalously high in the perspective
of at least the last 1,450 years.
It is likely that the annual period of surface melt on Arctic perennial sea ice lengthened by 5.7 [4.8 to 6.6]
days per decade over the period 19792012. There is high confidence that in the Arctic, where the sea ice
thickness has decreased, the sea ice drift speed has increased. There is very high confidence that, during the
last decade, the largest contributions to global glacier ice loss were from glaciers in Alaska, the Canadian
Arctic, the periphery of the Greenland ice sheet, the Southern Andes and the Asian mountains (80% of the
total ice loss). It is very likely that the annual Antarctic sea ice extent increased at a rate of between 1.2 and
1.8% per decade (0.13 to 0.20 million km2 per decade) between 1979 and 2012 (very high confidence).
It is very likely that the Arctic sea ice cover will continue shrinking and thinning year-round in the course
of the 21st century as GMST rises. At the same time, in the Antarctic, a decrease in sea ice extent and
volume is expected, but with low confidence. The CMIP5 multi-model projections give average reductions
in Arctic sea ice extent for 20812100 compared to 19862005 ranging from 8% for RCP2.6 to 34% for
RCP8.5 in February and from 43% for RCP2.6 to 94% for RCP8.5 in September (medium confidence). A
nearly ice-free Arctic Ocean (sea ice extent less than 106 km2 for at least five consecutive years) in
September before mid-century is likely under RCP8.5 (medium confidence). There is little evidence in
global climate models to a seasonally ice-free Arctic Ocean beyond which further sea ice loss is unstoppable
and irreversible. In the Antarctic the CMIP5 multi-model mean projects a decrease in sea ice extent that
ranges from 16% for RCP2.6 to 67% for RCP8.5 in February and from 8% for RCP2.6 to 30% for RCP8.5
in September for 20812100 compared to 19862005 (low confidence).
There is very high confidence that snow cover extent has decreased in the NH (Northern Hemisphere),
especially in spring. Satellite records indicate that over the period 19672012, snow cover extent very likely
decreased; the largest change, 53% [40 to 66%], occurred in June. No month had statistically significant
increases. In the Southern Hemisphere (SH), evidence is too limited to conclude whether changes have
occurred.
It is very likely that NH snow cover will reduce as global temperatures rise over the coming century.
Global model projections show that by the end of the 21st century a decrease of the NH spring snow
covered area will be 7 [3 to 10] % (RCP2.6) and 25 [18 to 32] % (RCP8.5); there is medium confidence in
38
these numbers. Further, by the end of the 21st century, projected near-surface permafrost area is projected to
decrease by between 37% (RCP2.6) to 81% (RCP8.5) (medium confidence).
Some recent references addressing climatic changes of ice include: (Box, 2013), (Box and Colgan,
2013), (Winkelmann et al., 2012), (Franco et al., 2013), (Fettweis et al., 2013), (Little et al., 2013.).
4.1.3
Sea water level
Both tide gauge and satellite altimetry data show that the GMSL (Global Mean Sea Level) has continued to
rise, relative to 19611990, (IPCC, 2013). Although the increase is fairly steady, observational records show
short periods of either no change or a slight decrease. The observed estimates of the GMSL increase lie
within the envelope of all the projections except perhaps in the very early 1990s. The sea level rise
uncertainty due to scenario-related uncertainty is smallest for the most recent assessments (AR4 and AR5)
and observed estimates lie well within this scenario-related uncertainty. It is virtually certain that over the
20th century sea level rose. The mean rate of sea level increase was 1.7 mm yr1 with a very likely range
between 1.5 to 1.9 between 1901 and 2010 and this rate increased to 3.2 with a likely range of 2.8 to 3.6 mm
yr1 between 1993 and 2010.
According to IPCC (2013) it is very likely that the rate of global mean sea level rise during the 21st
century will exceed the rate observed during 19712010 for all RCP scenarios due to increases in ocean
warming and loss of mass from glaciers and ice sheets. Model projections of sea level rise are larger than
reported in the AR4 (2007), primarily because of improved modeling of land-ice contributions. For the
period 20812100, compared to 19862005, global mean sea level rise is likely (medium confidence) to be
in the 5 to 95% range of projections from process-based models, which give 0.26 to 0.55 m for RCP2.6,
0.32 to 0.63 m for RCP4.5, 0.33 to 0.63 m for RCP6.0, and 0.45 to 0.82 m for RCP8.5. For RCP8.5, the rise
by 2100 is 0.52 to 0.98 m with a rate during 20812100 of 8 to 16 mm yr1.
Based on current knowledge, only the collapse of the Antarctic ice sheet, if initiated, could cause global
mean sea level to rise substantially above the likely range during the 21st century (medium confidence). It is
virtually certain that global mean sea level rise will continue beyond 2100, with sea level rise due to thermal
expansion to continue for many centuries but the amount is dependent on future emissions. The available
evidences today indicate that sustained global warming greater than a certain threshold above pre-industrial
would lead to the near-complete loss of the Greenland ice sheet over a millennium or more, causing a global
mean sea level rise of about 7 m. It is very likely that there will be a significant increase in the occurrence of
future sea level extremes in some regions by 2100, with a likely increase in the early 21st century. Sea level
rise of 1 to 3 m per degree of warming is projected if the warming is sustained for several millennia.
According to IPCC (2013) there is low confidence in region-specific projections of storminess and
associated storm surges.
Some recent references addressing sea level changes include: (Jevrejeva et al., 2012b), (Jevrejeva et al.,
2012a), (Gillet-Chaulet and al., 2012), (Gehrels et al., 2012), (Gillet-Chaulet and al., 2012), (Bamber and
Aspinall, 2013), (Dutton and Lambeck, 2012).
4.1.4
Wind and waves
The two last Assessment Reports (AR) form IPCC (2007) and IPCC (2013) are based on results from the
Coupled Model Intercomparison Project (CMIP) studying the output of climate models (CMs). The CMIP
was established by the Working Group on Coupled Modelling (WGCM) of the World Climate Research
Programme (WCRP) in 1995 and provides a community-based infrastructure in support of climate model
diagnosis, validation, and intercomparison.
CMIP has gone through five phases. The fifth phase, CMIP5, has provided a framework for assessments
in the Fifth IPCC Assessment Report (AR5) but so far only a limited number of wind and wave studies are
based on CMIP5 models (http://cmip-pcmdi.llnl.gov), and most post-AR4 wind and wave studies are based
on CMIP3 (http://cmip-cmdi.llnl.gov/cmip3_overview.html). A comprehensive summary of characteristics
and performance of climate models is given by IPCC (2013).
Storms can be classified into two groups: extra-tropical cyclones (regular storms) and tropical cyclones
(hurricanes and typhoons). IPCC (2013) reports that there is low confidence in long-term (centennial)
changes in tropical cyclone activity, after accounting for past changes in observing capabilities. Increases in
the frequency and intensity of the strongest storms in the North Atlantic are robust (very high confidence) but
the cause of this increase is debated (the relative importance of internal variability and anthropogenic and
natural forcings). The model projections indicate that it is likely that the global frequency of tropical cyclones
will either decrease or remain essentially unchanged, with a likely increase in both global mean tropical
cyclone maximum wind speed and rainfall rates, however, there is lower confidence in region-specific
projections of frequency and intensity. Because of improvements in model resolution and downscaling
techniques, it is more likely than not that the frequency of the most intense storms will increase substantially
39
in some basins under projected 21st century warming. The scientific community is in agreement that the
global number of extratropical cyclones is unlikely to decrease by more than a few per cent.
Recent investigations continue to support the findings of the AR4 and SREX, it is a likely poleward shift
of storm tracks since the 1950s (both in NH and SH). Storm track biases in the North Atlantic have been
improved slightly. There is low confidence in the magnitude of regional storm track changes, and the impact
of such changes on regional surface climate.
It is likely (medium confidence) based on reanalysis forced model hindcasts and ship observations that
mean significant wave height has increased since the 1950s over much of the North Atlantic north of 45N,
with typical winter season trends of up to 20 cm per decade. It is also likely (medium confidence) that
annual mean significant wave heights will increase in the Southern Ocean as a result of enhanced wind
speeds. Swells generated in the Southern Ocean are likely to affect heights, periods, and directions of waves
in adjacent basins. It is very likely (medium confidence) that wave heights and the duration of the wave
season will increase in the Arctic Ocean as a result of reduced sea-ice extent. In general, there is low
confidence in region-specific projections due to the low confidence in tropical and extratropical storm
projections, and to the challenge of downscaling future wind fields from coarse-resolution climate models.
The CMIP3 results of Hemer et al. (2012) and Hemer et al. (2013), Fan et al. (2013), Mori et al. (2012),
Semendo et al. (2013), support the earlier studies reviewed by Bitner-Gregersen et al. (2013a), i.e. negligible
changes in the projected mean SWH (Significant Wave Height) in all ocean basins except the Southern
Ocean and the South Pacific. Both Mori et al. (2013) and Hemer et al. (2013) found that the variance of
wave-climate projections associated with wave downscaling methodology dominated other sources of
variance within the projections, e.g. the climate scenario or climate model uncertainties. Further, Mori et al.
(2013), Hemer et al. (2013) and Fan et al. (2013) found some changes in wave periods and wave direction in
the 21st century. Wave periods were found to increase over the eastern Pacific and decrease in the North
Atlantic. Wave directions exhibit clockwise rotation in the tropics.
Wind and wave projections using CMIP5 models are limited. de Winter et al. (2013) compared the
Gumbel probability density functions (PDFs) for the annual maximum wind speed for 12 models for
RCP4.5 and RCP8.5. The chosen regions were the southern and the northern North Sea and the time period
20502100. The distributions show significant variability between the 12 models with respect the mean
values and standard deviations. It is particularly noteworthy that the 100 year wind speed for the northern
North Sea varied from around 24 m/s to 34 m/s, an extremely large deviation seen from an engineering
point of view.
Few similar studies appear to have been made on wave heights. Dobrynin et al. (2012) used one wave
model (WAM) forced by winds from one CM (EC-Earth) to study the evolution of the global wave climate
over 250 years (18502100) in terms of the mean projected SWH, using RCP4.5 and RPC8.5. On a global
scale only minor changes were detected in the mean projected wave height, 0.05 m from 1850 to 2010.
Between 2010 and 2100 the change was found to be an increase of 0.03 m for RCP8.5 and a decrease of
0.005 m for RCP4.5, respectively. The anomaly in Norwegian waters appears to have been close to 0%.
Uncertainties associated with wind and waves projections are presented in a flowchart in Figure 4.
Figure 4. Uncertainties associated with wind and waves projections.
40
4.1.5
Ocean circulation
The movement of fresh water between the atmosphere and the ocean can also influence oceanic salinity,
which is an important driver of the density and circulation of the ocean. It is very likely that changes in the
mean surface waters salinity have taken place since the 1950s (IPCC, 2013). The mean contrast between
high- and low-salinity regions increased by 0.13 [0.08 to o.17] from1950 to 2008. The Atlantic has become
saltier and the Pacific and Southern Ocean freshened. There is high confidence in this assessment.
According to IPCC (2013) it is likely that salinity will increase in the tropical and especially subtropical
Atlantic, and decrease in the western tropical Pacific over the next few decades. Overall, it is likely that
there will be some decline in the Atlantic Meridional Overturning Circulation by 2050 (medium
confidence). However, the rate and magnitude of weakening is very uncertain and decades when this
circulation increases are also to be expected. These changes are expected to have impact on the ocean
current patterns.
5.
SPECIAL TOPICS
5.1
Hurricane
Modelling of hurricanes has got increasing attention in the scientific community and media due the extreme
weather events reported in the last decade and projected climate changes. Reasons for occurrence of the
hurricanes is still a topic of discussion. It is interesting to note that the API (American Petroleum Institute)
hurricane metocean conditions in the Gulf of Mexico were revised some years ago to account for effects of
recent major hurricanes as well as changes in understanding regarding occurrence of hurricanes
Hurricane or typhoon (tropical cyclones) originate over tropical oceans in the latitudes between 5 to 20
from the equator where the surface water temperature is sufficiently warm. A strong low pressure pumps up
a large amount of air from just above the sea surface rises in spiral fashion and condenses to from clouds
and precipitation in the upper atmosphere. A large amount of latent heat energy is released into the
atmosphere generating strong winds Hurricane or typhoon frequently causes serious damage to marine
structure or navigating vessels. Wind wave modeling is essential method to predict wave height. High
accurate model plays an important role for disaster prevention.
Forecasting of wave heights is essential for planning and operation of maritime activities. Traditionally,
wave heights have been predicted using physics-based models, which rely primarily on the energy balance
equation. More recently, soft computing techniques such as Artificial Neural Network, Genetic
Programming (GP) have been used to generate forecasts with leads time from a few hours to several days.
Nitsure et al. (2012) improved the forecast of wave heights with lead times of 12 h and 24 h using GP. The
results were satisfactory, especially for the peak wave heights formed by the extreme events like hurricanes.
Besides, a spherical multiple-cell (SMC) grid was installed in a global wave model by Li et al. (2012) to
overcome the polar wave problems. A 2nd order upstream non-oscillatory advection scheme and a rotation
scheme for wave spectral refraction were used. The unstructured SMC grid allows time step to be relaxed
and land cells to be removed, saving over 1/3 of the total computation time in comparison with the original
latitude-longitude grid model. In addition, a quadtree-adaptive model was applied by Tsai et al. (2013) for
prediction of waves generated by tropical cyclones. The quadtree grid system can be adapted to the vicinity
of the hurricanes and/or some prescribed regions of interest which require higher resolutions, so that the
quadtree grid system can move with the hurricanes.
The University of Miami has presented a Fully Coupled Atmosphere-Wave-Ocean Modeling system
(UMCM). The UMCM includes three model components: atmospheric, surface wave, and ocean circulation
models. Chen et al. (2007) gave a brief introduction to UMCM and an overview for the coupled modeling
effort in the Coupled Boundary Layer Air-Sea Transfer (CBLAST)Hurricane program. Chen et al. (2013)
described the results of a new directional windwave-coupling parameterization in a fully coupled model
developed based on the CBLAST-Hurricane observations and laboratory measurements. Currently, UMCM
can be configured with two different options in terms of component models: 1) coupled with the fifthgeneration Pennsylvania State University-National Center for Atmospheric Research Mesoscale Model
(MM5), a third generation wave model (WAVEWATCH III), and the three-dimensional Price-WellerPinkel (3DPWP) upper ocean model (UMCM-MWP); and 2) coupled with the Weather Research and
Forecasting Model (WRF), the University of Miami Wave Model, UMWM, Donelan et al. (2012) and the
Hybrid Coordinate Ocean Model (HYCOM) (UMCM-WMH).
In addition, Liu et al. (2012) did an investigation of the effects of wave state and sea spray on an
idealized typhoon using an air-sea coupled modeling system. The coupling between atmosphere and sea
surface waves considered the effects of wave state and sea sprays on airsea momentum flux, the
atmospheric low-level dissipative heating, and the wave-state-affected sea spray heat flux. Smith et al.
(2013) examined tropical cyclone ocean-wave model interactions using an ESMF (Earth System Modeling
41
Framework) based tropical cyclone (TC) version of the Coupled Ocean/Atmosphere Mesoscale Prediction
System (COAMPS). The Coupled Ocean-Atmosphere-Wave-Sediment Transport (COAWST) modeling
system was used to investigate semi-enclosed Gulf of Venice Benetazzo et al. (2013). The results revealed
that, when applied to intense storms, the effect of coupling on waves results in variations of significant wave
height up to 0.6 m, with some areas experiencing significant increase/decrease of wave spectral energy for
opposite/following currents respectively.
The features of typhoon such as occurrence time, location, tracks and intensity have much uncertainty,
and corresponding probability research have a great development recently. Zhang and Xu (2005) proposed a
Modified Maximum Entropy Distribution (MMED) with four unknown parameters which can cover
different kinds fitting curves for extreme environmental elements. (Dong et al., 2013b) proposed maximum
likelihood method to estimate uncertain parameters, and compared it with method of moment and empirical
curve fitting method. Dong et al. (2013a) introduced Poisson-MMED to consider typhoon occurrence
numbers and process extreme values, and adopted it in the long-term extreme value calculations of storm
surges based on the data from Japan and North Sea. Ocean environmental elements such as wind, wave and
current induced by typhoon are generally statistically related. Dong (2013c) selected annual extreme
significant wave heights and corresponding peak periods based on storm process data firstly, then
constructed bivariate equivalent MMED to study the joint probability of these two elements. Tao et al.
(2013a) proposed several Poisson bivariate MMEDs by using four common-used copulas and Poisson
distribution (which fits the storm occurrence numbers), and established a Joint Tide-Wave Impact Grade to
judge the typhoon storm surge intensity passed Qingdao of China. For marine engineering designs, Dong
et al. (2012) presented interval estimations of return wave heights based on MMEDs while Tao et al.
(2013b) gave the joint design parameters of wave height and wind speed by using bivariate MMEDs, and by
taking account of typhoon occurrence frequency, the corresponding Poisson compound extreme value
distributions can also be used for designs. Figure 5 shows a flowchart of the hurricane or tyhpoon hindcast
model and statistical model used when diriving extremes.
Figure 5. Hurricane or tyhpoon hindcast model and statistical model.
5.2
Wave current interaction
Wave-current interaction is a common phenomenon in the ocean. It affects wave parameters, the shape of a
wave spectrum and has impact on the flow field of current. Consequently it influences the response and
safety of marine structures. The subject got increasing attention in the last decade due to the debate about
rogue waves and recently also climate change as well as the renewable energy needs.
In the present section the latest trend of investigations dedicated to wave-current interaction are reviewed
for the period 20092014, given main focus on the years 20122014.
5.2.1
Wave-current Interaction Model
Based on intensive former research, Smith (2006) summarized the energy, momentum, and mass-flux
changes between surface wave and underlying Eulerian mean flows. Besides classical wave radiation
stress, other various terms are identified with, for example, the integrated CL vortex force implemented.
They focused on interpreting these terms in term of physical mechanism and permitting reasonable
42
estimates of the associated dependency. Lane et al. (2007) compared the vortex-force representation of the
wave-averaged effects on currents with the radiation-stress representation in a region that is proper for
coastal and shelf waters.
Mellor (2011) presented surface wave equations appropriate to three-dimensional ocean model by
assuming that the depth dependence of wave motions is provided by linear theory. Expressions for vertically
dependent radiation stress and a definition of the Doppler velocity for a vertically dependent current field
are obtained. The momentum and wave energy equations included other quantities such as vertically
dependent surface pressure forcing and terms representing the production of turbulence energy by currents
and waves. This research provided sound base for three-dimensional ocean models that treat surface wave
and wind- and buoyancy-driven currents simultaneously. Later, Ardhuin et al. (2008a) criticized this paper
of inadequate approximations of the wave motion. They showed these equations are not consistent in the
simple case of shoaling waves without current. The modification needs a numerical evaluation of the waveforcing terms. Mellor (2011) revised their work, abandoning the a priori use of sigma coordinates as
characterization of waves derived for a flat bottom can be misunderstood in the sigma domain. Only the
depth-dependent stress radiation terms are added to the momentum equation to incorporate them into threedimensional circulation models. It is also believed that transport of the surface stress into the water column
is supported by pressure and turbulence instead of turbulence alone as described in former studies.
On the other hand, Ardhuin et al. (2008b) utilized the generalized Lagrangian mean theory to find exact
equations for three-dimensional wave-turbulence-mean flow interactions. To close their equations, they
specify the wave forcing terms under the hypotheses of small surface slope, weak horizontal gradients of
water depth and mean current, and weak curvature of the mean current profile, yielding analytical
expressions for the mean momentum and press forcing terms related to the wave spectrum. Furthermore,
glm2z-RANS equations with non-divergent mass transport in Cartesian coordinate are obtained by applying
a vertical change of coordinate. Their approximation provides an explicit extension of known waveaveraged equations to short scale variations of the wave field and vertically varying currents. The
underlying exact equations can provide a natural framework for extensions to finite wave amplitudes and
any realistic conditions. Meanwhile, Bennis and Ardhuin (2011) thought that work of Mellor (2011) is not
consistent with the known depth-integrated momentum balance in the presence of a sloping bottom.
Unrealistic surface elevations and currents can be produced by a numerical integration of the equations in
the absence of dissipation. It seemed to them that the inconsistency is caused by a different averaging for the
pressure gradient term and the advection terms of the same equation. Nevertheless, work of Mellor (2011)
can produce large errors for continental shelf applications such as the study of cross-shore transports outside
the surf zone. They recommended equations for the quasi-Eulerian velocity (McWilliams et al., 2004,
Ardhuin et al., 2008b) which is free from such problems. Mellor (2011) replied to this criticism, stating that
it is only partially correct but fallacious for the most part.
Hasanat Zaman and Baddour (2011) reported a study of three-dimensional interaction of a current-free
monochromatic surface wave field with a wave-free uniform current field under the assumption of
irrotational and inviscid flow. Particularly, the wave and current fields are not necessarily collinear with
each other. They also developed the expressions to describe the characteristics of the interacting flow by
mass, momentum and energy transport conservation. They used parameters such as the surface disturbance
amplitude and length, mean water depth, mean current-like parameter and direction of the combined flow to
illustrate the wave-current field after interaction.
Wave breaking influences wave-current interactions in real sea. Whitecapping affects the Reynolds
stresses. Restrepo et al. (2011) modified a model (McWilliams et al., 2004) for the conservative dynamics of
waves and currents to include the averaged effect of multiple, short-lived, and random wave-breaking events
on large spatio-temporal scales. They treated whitecapping by parameterizing stochastically as an additive
uncertainty to the fluid velocity. They coupled it to the Sokes drift as well as to the current velocity in the
form of nonlinear momentum terms in the vortex force and the Bernoulli head. They discussed
whitecappings effect on tracer dynamics, mass balance and boundary conditions.
Aiki and Greatbatch (2012) investigated the residual effect of surface gravity waves on mean flows in
the upper ocean by thickness-weighted mean (TWM) theory in a vertically Lagrangian and horizontally
Eulerian coordinate system. They derived depth-dependent equations for the conservation of volume,
momentum and energy, which allow for (i) finite amplitude fluid motions, (ii) the horizontal divergence of
currents, and (iii) a concise treatment of both kinematic and viscous boundary conditions at the sea surface.
Chen and Chen (2014) presented a three-dimensional Lagrangian solution up to the fifth-order found for
the Boundary Value Problem (BVP) of irrotational, progressive water waves propagating in the presence of
uniform current in water of constant depth. They embedded wave-current interaction in the Lagrangian
43
velocity potential assuming that pressure is not affected by current in the wave-current field. Further, they
also presented motion properties of particles such as the particle motion period, drift velocity, the
Lagrangian mean level and the 3D particle trajectory and streamline. The numerical results were in good
agreement with experiments (Chen et al., 2012). When currents vanish, the present solution deduces to the
solution of the progressive wave propagation (Chen et al., 2010).
5.2.2
Numerical and Analytical Method
Rusu and Guedes Soares (2011) investigated wave-current interaction using SWAN model, which is a stateof-art spectral model for the wave transformations. They explained the theoretical background of the wavecurrent interactions, including the transformation of the wave spectrum and breaking waves due to currents.
Generally, their experimental data and the simulations had good agreement. Moreover, Rusu et al. (2011)
studied the wave propagation and the consequences of the influence of currents on waves in an estuary.
They evaluated the effects of the wind and local currents on the incoming waves by performing SWAN
simulations with and without considering the tide level and tide induced currents. The model results were
also compared with measurements to validate the results of the wave prediction system developed. Bratland
et al. (2011) described the calculation of wave elevations in higher order unidirectional, irregular waves with
a uniform current in deep water.
Markus et al. (2013) using unsteady RANS equations simulated the flow field of a nonlinear wave in
combination with a non-uniform current. The methodology combines a non-linear wave model with a VOF
calculation to generate an unsteady sea state. A simulation strategy that focuses on capturing wave-current
interaction is introduced and is validated with respect to fluid particle kinematics.
Ardhuin et al. (2012) reviewed the performance of numerical models in conditions with strong current
with respect to currents effects on waves. Their analysis was supported by experiments and real sea
measurements. They showed that using different parameterizations with a dissipation rate proportional to
some measure of the wave steepness to the fourth power, the results deviate significantly, none being fully
satisfactory. So they called for experiments data with higher spatial resolution to better resolve the full
spatial evolution of the wave fields and to validate the parameterizations.
Moreira and Peregrine (2012) investigated numerically effects of nonlinearity on a linear water wave
train in deep water with underlying currents by using a boundary-integral method. Their research included a
slowly varying current and a rapidly varying current (see Figure 6). Especially, they paid attention to
wave breaking and blocking and qualitatively explained former experimental observation.
Teles et al. (2013) evaluated wave-current interactions by an advanced CFD solver based on RANS
equations. First, changes in the mean horizontal velocity and the horizontal velocity amplitude profiles and
the influence of various first and second order turbulence closure models were studied. The results of the
numerical simulations were compared to the experimental data from former studies. Secondly, they
conducted more detailed study of the shear stresses and the turbulence viscosity vertical profile changes.
Figure 6. Fully nonlinear results obtained for wave groups propagating over (a) still water, (b) a slowly varying
current and (c) a rapidly varying current, (Moreira and Peregrine, 2012).
44
Zou et al. (2013) formulated a new Boussinesq-type equations for wave-current interactions. They
included the mean water depth change caused by the presence of current and wavecurrent interaction.
Both strong and moderate current were considered. They revealed the effect of the current on the vertical
distributions of the wave velocity and the pressure. They performed numerical simulation to show the
effect of the mean water depth change, caused by the presence of current, on wave motions.
Zhang et al. (2014b) utilized RANS equations, with kt- turbulence model (kt denotes the turbulent
kinetic energy and is the dissipation rate) and VOF method to capture free surface, to study wavecurrent interactions. They referred to work by Lin and Liu (1999) and used a inertia wave generator in the
middle of numerical tank. The generation of current was realized by setting boundary conditions. So they
studied the wave in following and opposing current simultaneously (see Figure 7). They validated their
velocity profiles by comparing with experimental work by Umeyama (2011). Furthermore, they discussed
the effects of wave periods and current velocity on regular wave-current induced water surface profile and
velocity distribution. Finally, they studied the propagation of a solitary wave traveling with a
following/opposing current numerically.
Myrhaug and Holmedal (2014) provided a simple analytical tool to calculate the wave-induced current
beneath long-crested (2D) and short-crested (3D) random waves. As an example, they calculated the
significant values of the Stokes drift and transport in deep water and in finite water depth.
Figure 7. An illustrative sketch of computational domain and boundary conditions for modeling wavecurrent
interaction, Zhang et al. (2014).
Wave-current interaction could affect the wave focusing and freak wave in the ocean. Hu and Ma
(2011) investigated the influence of wave-current interaction on freak waves based on modified Nonlinear
Schrdinger Equations (NLS). Peng et al. (2013) developed a numerical wave tank by high order
spectrum method (HOSM) and simulated nonlinear wave-current interactions. Peng et al. (2013)
investigated similar issue by a RANS simulation taking viscous effect into consideration.
5.2.3
Experiments and Measurements
Umeyama (2009) investigated the turbulence intensities of the interaction between nonbreaking waves
with varying periods and uniform current in a two-dimensional water flume. The prediction of the Eulerian
mean velocity was discussed in following and opposing current. Furthermore, Umeyama (2009) conducted
intensive experiments under different wave parameters such as the ratio between wave height and
wavelength or that between particle velocity and depth-averaged velocity. He looked into the phaseaveraged turbulent intensities, wave-current Reynolds stress, and velocities for variable wave or hydraulic
characteristics such as wave height, period, direction, and water depth. This study also made up for the data
lacking an aspect of phase-averaged velocity and lateral turbulence in former study. Recently, Umeyama
(2011) employed Laser Doppler anemometer (LDA) and coupled Particle Image Velocimetry(PIV) and
particle tracking velocimetry (PTV) measurements to study kinetic aspects of surface waves propagating
with or without a current in a constant water depth. Physical properties of the velocity and trajectory of a
water particle during one wave cycle were investigated experimentally.
Ma et al. (2010) conducted a laboratory observation of the nonlinear evolution of waves propagating on
a spatially varying opposing current in a 2D wave-current flume. A frequency downshift was found in
opposing currents. The ultimate frequency downshift increases as increase initial steepness enhances. The
evolution of frequency modulation was observed using the instantaneous frequency extracted by the Morletwavelet transform. Wave blocking in strong current could increase the asymmetric modulation and
accelerate the effective frequency downshift.
Gemmrich and Garrett (2012) examined long records of surface wave heights from buoy observations in
the northeastern Pacific Ocean. Their findings showed the effect of near-inertial currents on surface waves.
45
The result had implications for wave forecasting and provided valuable information on the frequency,
strength, and intermittency of the associated near-inertial motions.
Toffoli et al. (2013b) performed experiments in two independent wave tanks and showed experimentally
that a stable wave propagating into a region characterized by an opposite current may become
modulationally unstable. The experimental results supported the recent conjecture based on a currentmodified nonlinear Schrdinger equation which establishes that rogue waves can be triggered by a
nonhomogeneous current characterized by a negative horizontal velocity gradient.
Robinson et al. (2013) provided a first insight into the dynamics of dense saline gravity currents moving
beneath regular progressive free-surface water waves. After the initial collapse, the gravity currents
propagated horizontally with two fronts, one propagating in the wave direction and the other against the
wave direction. The overall length of the gravity current, the position of the gravity current center, the shape
of the two leading profiles and an asymmetry in the shape of the upstream and downstream current heads
have been discussed in the paper.
Wijesekera et al. (2013) deployed several acoustic Doppler current profilers and vertical strings of
temperature, conductivity, and pressure sensors on and around the East Flower Garden Bank (EFGB) (180km southeast of Galveston, Texas) to examine surface wave effects on high-frequency flows over the bank
and to quantify spatial and temporal characteristic of these high-frequency flows.
Ma et al. (2013) carried out systematic experiments focusing on the evolution of wave trains with
initially sidebands on uniform currents in a wave-current flume.
The linear modal theory for waves on opposite jet current is given by Shrira and Slunyaev (2014a). It is
based on the approximate variable separation for transverse modes. The resulting one-dimensional
boundary-value problem allows one to perform theoretical analysis of waves on currents, and to employ a
series of analytic solutions. The Agulhas current represents of order 1001000 trapped wave modes. Shrira
and Slunyaev (2014b) has formulated also the weakly nonlinear equation for modes of waves trapped by an
opposite jet current. The mode evolution against the current is described by the nonlinear Schrdinger
equation with reduced nonlinear term. The weakly nonlinear solutions are verified by means of strongly
nonlinear simulations. The occurrence of localized breaking events from smooth initial conditions is shown.
The 3D intense solitary wave patters are shown to exist. The 3D problem corresponds to effectively
unidirectional dynamics, and thus the probability of rogue waves in the field of trapped waves is much
higher.
5.3
Wave and wind energy resource assessment
With the development of Marine Renewable Energy, a significant effort is made on resource assessment.
Objectives of resource characterization include not solely wind, wave or tidal current power assessment but
also an accurate description of the environment for the purpose of engineering applications such as design
and optimization of marine devices and marine operations. As a consequence, a refined characterization of
the space and time variability of the various relevant parameters is requested, especially in coastal areas, and
adapted analysis tools are to be developed.
Especially for the case of wave energy extraction there exists a strong demand for high resolution wave
hindcast databases allowing regional or site climatology assessment.
Boudiere et al. (2013) developed a sea-states hindcast database for the assessment of sea-states
climatologies that fulfills such requirements for MEC design and optimization. This database, covering the
Channel and Bay of Biscay over a 19 years period from 1994 to 2012 was built running an up-to-date
configuration of the WaveWatch III wave model on a refined unstructured grid with mesh size ranging
from about 200 m in coastal areas to about 10 km offshore. The wave model is forced by the wind field
from the CFSR reanalysis (Climate Forecast System Reanalysis) that was produced at NCEP (National
Centre for the Environmental Prediction) in 2010. Tidal currents and water levels were derived from atlases
of tidal harmonics obtained from the MARS 2D (Model for Applications at Regional Scale) hydrodynamic
model. Outputs include a large set of parameters relevant to marine energy applications at each point of the
computational grid together with directional spectra at over 4000 locations. Model data was validated
against in-situ (buoy), remote sensing measurement data as well as another hindcast database and proved to
be of good quality.
Combining the WAVEWATCH III and SWAN wave models, Garcia-Medina 2013 built a 7 year
hindcast at a 30 arc-second resolution along the coast of Oregon and southwest Washington, USA.The
hindcast accuracy was validated against buoy data and was used to analyze the alongshore variability of the
resource over the continental shelf. Beyond the classical general power decay with depth related to wave
refraction and shoaling, specific local features of the power distribution are identified. For instance due to
wave refraction, areas off the central and southwest Oregon coast are identified that show increased wave
46
power at 50 m of water in comparison with the 250 m value. It is also observed that areas with preferentially
narrower wave spectra in both frequency and direction are identified off southwest Oregon. However, it is
also pointed out that a near shore wave energy resource characterization for Oregon and Washington, in
water depth lesser than 50 m, may necessitate even higher-resolution assessments.
Influence of tide on wave propagation, hence on wave energy resource is worth being evaluated. Using a
non-linear coupled wave-tide model SWAN-ROMS (Hashemi and Neill, 2014) assessed the impact of tides
on the wave energy resource of the northwest European shelf seas. Comparing one month of wave power
computed using the SWAN model alone, that is without current, and the SWAN-ROMS coupled results
they show that the impact of tides is significant, and can exceed 10% in some regions of strong tidal currents
(e.g. headlands). Results also show that the effect of tidal currents on the wave resource is much greater than
the contribution of variations in tidal water depth, and that regions which experience lower wave energy
(and hence shorter wave periods) are more affected by tides than high wave energy regions.
Even though the Mediterranean Sea is usually considered as a basin with limited wave resource,
significant number of studies are conducted to provide high resolution assessment of the wave energy
resources in that area.
Using a parallel version of the WAM wave model Cycle 4.5.3 Liberti et al. (2013) built a ten year
hindcast data base of the entire Mediterranean basin on a 1/16 resolution grid to assess the wave energy
resource. Analysis of the model results allows the characterization of the regional distribution of the
resource. For instance, the western Sardinia coast and the Sicily Channel are found to be among the most
productive areas in the whole Mediterranean. Additionally, simulation results show the presence of
significant spatial variations of wave power availability even on relatively small spatial scales along these
two coastlines, confirming the needs for high resolution resource assessment. Seasonal variability is also
investigated and large amplitude seasonal variation range is observed in some areas.
Combining wave buoy data and a 22 year hindcast data set provided by ECMWF, Vicinanza et al. (2013)
assessed the power resource off the coast of Sardinia (Italy) in the Mediterranean Sea. The annual offshore
wave power was found to range between 8.91 kW/m and 10.29 kW/m, the bulk of which is provided by
north-westerly waves. They also studied the nearshore energetic patterns by means of a numerical coastal
propagation model (Mike21 NSW). The analyses highlight two hot spots in which, practically, the energy
dissipation appears to be compensated for by natural phenomena of energetic refraction and wave reflection
due to seabed interaction and where the wave power is respectively 9.95 and 10.91 kW/m.
Even though regional and local studies are a mandatory requirement for resource assessment, global
characterization of the available resource should also be considered as it presents some economical and
industrial interest. Using six years (20052011) of a hindcast database provided by NOAA and built using
WAVEWATCH III at a spatial resolution of 30 arc-minutes and a temporal resolution of 3 hours, Gunn
and Stock-Williams (2012) estimated from global parameters the wave power along the worlds oceanic
coastline. Introducing the Pelamis P2 power matrix they also estimated the extractable power for an array
of such wave energy converter (WEC). Beyond a quantitative description of the available power around
the world, this work allows to point out an interesting qualitative result as it reveals that the best areas for
wave power generation may not be exactly correlated with the best areas for raw resource.
Using a more general approach valid for any device to be installed at a specific site and focusing on
other parameters than the sole available power, namely the energy spread in frequency and direction, and
the seasonality, Portilla et al. (2013) show that other areas than the most energetic ones (located in the
storm-belts) can be similarly attractive simply because the lower energy level is compensated by a greater
potential efficiency due to wave characteristics more uniform in time and to the lack of severe conditions.
As for waves, tidal current resource assessment requires an accurate characterization of the available
power on rather high resolution grids but covering large areas over which tidal turbines are likely to be
deployed in arrays. It is also of prime importance to characterize the variability of the flow and therefore to
provide parameters, metrics and methodologies that are suitable for engineering applications. Even though
numerical modeling is the most widely used approach, there are still strong requirements for in-situ
measurement capacity, especially to contribute to the assessment of the turbulent features of the flows in the
highly energetic areas considered for the deployment of tidal turbines.
Presenting results of a measurement campaign conducted at two energetic sites, Thomson et al. (2012)
discuss the development of a set of metrics for describing the turbulence together with the utility of ADCP
measurements for observing turbulence at tidal energy sites These results provide some of the first realistic
conditions for estimating the fatigue loads and the performance of tidal turbines as only very few
measurement campaigns at energetic sites have been conducted and little data is available.
47
A comprehensive methodology for the validation of numerical models in two and three dimensions is
presented by Gunn and Stock-Williams (2013) and discussed. Especially they present novel extended
methods for the validation of 3D data and use Acoustic Doppler Profiler data for a case study validation of a
specific energetic site and finally provide interesting guidelines for model validation purpose.
6.
DESIGN AND OPERATIONAL ENVIRONMENT
6.1
Design
New designs and operational decisions must be assessed/made relative to recognised codes and standards,
for which the responsible authority, perhaps a classification society or the user himself, will depend on
the design and its application. To achieve recognition, an environment parameters climatology must be
demonstrated as robust and of adequate accuracy and consequently such codes and standards may lag
behind the state-of-the-art.
The majority of ocean-going ships are designed currently to the North Atlantic wave environment, which
is regarded as the most severe. It is interesting to note that recent investigations of global wind and wave
conditions are confirming the latter, see (Cardone and Cox, 2011, Cardone et al., 2014). The traditional
format of classification society rules is mainly prescriptive, without any transparent link to an overall safety
objective. In 1997 and 2001 IMO has developed Guidelines for use of the Formal Safety Assessment (FSA)
methodology in rule development which will provide risk-based goaloriented regulations (IMO, 1997,
IMO, 2001). Although environmental wave data and models are not explicitly used by classification society
rules for general ship design they are needed in rule calibration when FSA methodology is applied. For
some less typical designs, classification society rules require or recommend some type of dynamic load
analysis that makes also use of metocean data.
Classification rules, in fact, permit the design of ships for restricted service (in terms of geographical
zones and the maximum distance the ship will operate from a safe anchorage); in which case reduced design
loads apply. Many aspects of the design, approval and operation require a detailed knowledge of local
weather conditions. While in principle open to all ship types, the use of such restricted service is in practice
mainly confined to high speed vessels.
Unlike ship structures, offshore structures normally operate at fixed locations and often represent a
unique design. As a result, platform design and operational conditions need to be based on location specific
metocean climate. Note that Floating Production Storage and Offloading (FPSO) systems are designed for
the North Atlantic wave environment if location specific wave climate cannot be proved more appropriate.
Even though the same basic principles prevail for hydrodynamic loads on ships and offshore structures,
actual problems and methods for assessing these loads in the design stage are quite different. Further, to
some extent different wave models are used for defining design and operational conditions for these two
types of structures.
In the comparatively nascent field of operational analysis techniques, it is more frequently the
responsibility of the user to select a climatology that they feel is most suitable to the task. Such decisions
are taken based on a risk assessment.
6.1.1
Met-Ocean Data
Visual observations of waves collected from ships in normal service and summarized in the British
Maritime Technology Global Wave Statistics (GWS) atlas (BMT, 1986) are still used for ship design and
operations. The average wave climate of four ocean areas in the North Atlantic, with some correction
introduced due to inaccuracy of zero-crossing wave period (Bitner-Gregersen et al., 1995), is recommended
by the International Association of Classification Societies (IACS 2000) for ship design.
The visual BMT data represent a sufficiently long observation history to provide reliable global climatic
statistics over much of the global ocean. Wind speeds (Beaufort Scale) and directions, and wave heights in a
coarse code have been reported since 1854. Observations of wave height, period, and direction have been
collected from ships in normal service all over the world since 1949, and are made in accordance with
guidance notes from the World Meteorological Organisation (WMO, 2001, WMO, 2003). These data
include some bad weather avoidance as ships try not to sail into storms; today many ships receive weather
forecast from meteorological offices. Thus this database is probably biased towards lower wave heights, see
e.g. (Bitner-Gregersen et al., 2014d). Note also that the GWS Atlas was published in 1986 thus the last 28
years is missing which has impact on extreme wave heights; the 100-year Hs reported by Grigorieva and
Gulev (2006) is beyond 18-19 m in the North Atlantic.
The utility of visual observations depends on appropriate calibration versus accurate measurements of
the wave characteristics. BMT (1986) compared the GWS marginal distributions for wave heights and wave
periods with instrumental Shipborne Wave Recorder and National Oceanic and Atmospheric Administration
48
(NOAA) buoy data for different locations and concluded that the wave heights and periods for which
statistics were given corresponded to measured values. However, the accuracy of the GWS data has been
questioned in the literature since the 90-ties, especially concerning the wave period as discussed, e.g. by
Wing and Johnson (2010), Bitner-Gregersen et al. (2014b) and Bitner-Gregersen et al. (2014d).
Apart from uncertainties associated with the GWS data and missing the last 28 years, the limitation of
these data is lack of information about directional wave spectrum, as pointed out by Wing and Johnson
(2010) and Bitner-Gregersen et al. (2014b). Wing and Johnson (2010) have shown that wave directionality
have quite an unpredictable effect on the long term ship motions and loads.
The necessity of replacing that historic, (essentially subjective) observation based wave database for ship
design with instrumentally collected (objectively measured) data bases, or by a combination of numerical
and measured data, has become a subject of increasing discussion within classification societies in recent
years. This discussion has intensified because of routing systems with which some ships are equipped and
the climate change debate. Currently, two other sources of global metocean climate are available in addition
to the ship observations. These are data from numerical wave prediction models and satellite data (e.g. the
GlobWave database). Predictions of extreme metocean parameters based on these new wave databases have
shown large discrepancies making still difficult reaching firm conclusions, as discussed by Bitner-Gregersen
et al. (2013a) and Bitner-Gregersen et al. (2014d). Some descripancies between different data bases
predictions will need to be recognised and accepted but they need to be further documented.
Further, there is an ongoing discussion within the shipping industry how to account in ship design for
wave climate which sailing ships experience during their lifetime; can wave climate derived from ship
motions and marine radar be utilized in this process? The technology for deriving wave heights from marine
radars has not yet been demonstrated in a satisfactory approach. Such approach is being under development
within the JCOMM Wave measurement and Evaluation (WET) project (www.jcomm.info/WET). It is
interesting to note that the recent investigations of Cardone et al. (2014) based on GlobWave and
GROW2012 data show that a vessel is more exposed to encounter dangerous sea states along mid and high
latitude NH (Northern Hemisphere) routes.
The offshore industry uses location specific data in specification of design and operation criteria.
Traditionally instrumentally recorded data were regarded as superior to model derived data. However, due
to limited availability of measurements and improved hindcasts, the latter have become increasingly used in
design in the last decade. New improved global wave hindcasts (e.g. ERA-Interim, ERA-Clim, CFSR) are
continuously developed, see e.g. (Cardone et al., 2014). In the extra-tropics these hindcasts can be expected
to provide good estimates of wave climate, especially for the highest waves, whereas ship observations of
the highest waves are notoriously unreliable, and may be subject to some fair-weather bias. The hindcast
models are somewhat less reliable in the tropics, but for tropical storms the waves are less extreme and do
not define the design criteria for a sailing ship but will effect offshore structure design. Note that a coarse
resolution of a wave model may give up to a few meters lower Hs extremes than a high resolution.
Since the last reporting period, research organisations as well as the offshore industry have updated wind
and waves hindcast data sets for several basins, within proprietary joint industry projects (see Section 2.1.3
and 2.2.3). Increasing attention has also been given to the uncertainties in hindcasts, and in particular to
energy partitioning procedures used to separate wind sea and swell contributions (Kpogo-Nuwoklo et al.,
2014, Bitner-Gregersen et al., 2014b).
The shipping and offshore industries, e.g. (Bitner-Gregersen et al., 2013a, Vanem et al., 2014, BitnerGregersen et al., 2014d), (Hagen et al., 2013), follow research findings on climate change, but so far effects
of climate change on metocean conditions have not been introduced in the standards due to large
uncertainties related to climate change projections.
6.1.2
Design Environment
In the design process, international standards are followed to calculate ship structural strength and ship
stability during extreme events with a return period of 20/25 years; the Ultimate Limit State (ULS) check
corresponding to the maximum load carrying resistance. Checks in the Accidental Limit State, ALS,
(corresponding to the ability of the structure to resist accidental loads and to maintain integrity and
performance due to local damage or flooding) cover grounding, collision, and fire and explosion. An
extreme weather event check is not included in ALS.
Offshore structures (including FPSOs) follow a different approach to ship structures and are designed for
the 100-year return period (ULS). The Norwegian offshore standards, being now under updating, NORSOK
(2012) take into account extreme severe wave conditions by requiring that a 10000-year wave does not
endanger the structure integrity (ALS).
Further, the shipping industry is commonly using the linear regular and irregular waves as input to
numerical codes for calculations of ship loads and responses while the 2nd order irregular waves are currently
applied by the offshore industry when analysing structural loads and responses. Both linear and second order
49
wave models are not able to capture very steep waves such as rogue waves (called also abnormal or freak),
see Section 3.2. Recently, an increasing use of CFD tools in analysis of marine structures has taken place
requiring a proper description of sea states as well as extreme and very steep waves.
The prediction of horizontal velocities underneath measured irregular wave surface elevations is
addressed by Birknes et al. (2013) using a simple case of unidirectional waves in deep water. The results
from three commonly used methods for calculating the crest kinematics (two second order wave models and
the Wheeler stretching, see (DNV, 2014)) are compared with the model test results. All three methods show
a reasonable agreement with model test results although the second order models are clearly superior to the
Wheeler method. However, when wave breaking is present the second order model is expected to
underestimate the kinematics at the very top of the crest and the velocities can exceed the phase velocity. To
obtain satisfactory prediction of wave kinematics higher order potential theorys solutions (e.g. HOSM) and
CFD methods need still further exploration (see Section 3.2.1), particularly for very steep and breaking
waves.
Very steep and breaking waves are getting growing attention in the shipping and offshore industry
because their impact on loads and responses of marine structures. They have been addressed by the
EXTREME SEAS and ShortCresT projects, and in MARINTEK where combination in a consistent manner
of CFD and model tests has intensively been studied in the last years, e.g. (Pakozgi et al., 2012; Stansberg
et al. 2012). Commonly, extreme random wave events identified from numerical simulations or model tests
are isolated and modelled by CFD. This methodology is in development and attention to it will grow in the
future.
Platform decks cover a reasonably large area compared to the size of a wave crest and when waves
propagate, their crest heights change. Forristall (2015) investigated this change using measurements from
the MARIN wave basin and numerical linear simulations. The second order enhancement of crest is
accounted for by factoring the Gaussian maximum. Empirical fits to the simulations have been proposed
that can be used for most practical problems.
Stansberg (2012) recommends to explore further random wave groups as they are of importance for the
slowly varying motions of large floating structures.
Research efforts regarding refining models and estimation procedures of the long-term sea state
description continues, given particular focus to the associated uncertainties. see e.g. (Bitner-Gregersen et al.,
2014a, Bitner-Gregersen et al., 2014b, Bitner-Gregersen, 2015). Long-term distributions of sea state
parameters are representing an important input to loads and response calculations of marine structures. They
are also required for the level III reliability analysis (Madsen et al., 1986).
A review of joint long term probabilistic modelling of wind, waves, and current and sea water level and
associated uncertainties can be found in Bitner-Gregersen (2012) Bitner-Gregersen (2015) with particular
attention given to the Conditional Modelling Approach (CMA). It is pointed out that the Nataf model needs
to be use with care because it can give bias results. Recently attention has been given to use of multivariate
copulas in establishment of joint probabilities. Tao et al. (2013b) utilized bivariate normal copula and Frank
copula to construct joint distribution of two random variables; extreme wave height and concomitant wind
speed, and applied it to calculate the maximum base shear of the on-site fixed jacket platform. The results
show that the joint probability models constructed by the bivariate copulas result in lower design
environmental parameters due to consideration of correlation between random variables instead of assuming
extreme wind and waves occurring simultaneously. Based on a bivariate equivalent maximum entropy
distribution, Dong et al. (2013b) estimated joint design parameters of the concomitant wave height and wind
speed at a site in the Bohai Sea for the exploitation of marginal oil field. A systemic comparison of
approaches applied today for description of joint probabilities, including copulas models, using data from
several ocean regions is still lacking and needs attention. Note that copula models do not utilize the
complete probabilistic information obtained from simultaneous observations.
It is interesting to note that also metocean parameters describing the typhoon event and used in design
can be determined under multivariate extreme ocean environmental conditions, as illustrated for the
exploitation of marginal oil field by Dong et al. (2007).
Modelling of wind sea and swell remains an important topic for engineering applications. Several
conventional partitioning exist, see (ISSC, 2012). Kpogo-Nuwoklo et al. (2014) has proposed a new method
to identify temporal sequences of wave systems parameters, consistent with respect to the meteorological
events that are the sources of the phenomena. This method is based on the watershed algorithm which is
directly applied to the whole time-history of wave spectra. Using appropriate criteria, the identified events
are classified into swell or wind sea events. The approach is validated using field data from West Africa and
hindcast data from the Iroise Sea. The results show a good identification of wave systems events with a
good correlation between wind sea events and wind characteristics.
50
Effects associated with the variability of the probability of extremes as a function of season and of
direction are shown by Feld (2014). There are statistically significant differences between return values for
different directions and seasons.
Keef et al. (2013a) propose a variant of the conditional extremes model in which marginal transformation
to Laplace rather than Gumbel scale is performed. (Keef et al., 2013b) propose additional constraints within
the conditional extremes model formulation, particularly relevant for negatively associated variables. GilletChaulet et al. (2012) use the conditional extremes model to estimate joint extremes of large-scale indicators
for severe weather.
Jonathan et al. (2013) incorporated the effect of covariates in the conditional extremes model of
Heffernan and Tawn (2004), which has been further extended to involve a common general-purpose
penalised spline representation of model parameters with respect to multidimensional covariates, see
(Jonathan et al., 2014). An example of the application of the method for estimating conditional extremes
with covariate for hindcast storm peak significant wave height and associated spectral period in the northern
North Sea are reported by Ewans and Jonathan (2014). The objective is to model the distribution of spectral
period for large storm peak significant wave height as a function of storm direction. The study shows the
influence of longer fetches on extremes.
Muyau et al. (2014) found poor agreement between the current ISO wind gust factors and their
measurements. The measurements of wind profiles showed more shear than the ISO profiles. Further, the
ISO spectrum has not provided a good description of the spectra of the measured data. The extreme winds in
tropical areas are usually associated with squalls. A Joint Industry Project was initiated to examine squalls
of West Africa. The investigations carried out have shown significant differences between (ISO, 2012) gust
to mean wind speed ratios and wind speed coherence relationships and those derived for measurements
made by the JIP. Alternative relationships have been proposed from the squall data acquired during the JIP
(Santala et al., 2014). The DeepStar project has validated the equations for hurricane winds recommended
by the American Petroleum Institute (API, 2012). The recent findings have revealed that the present
industry standards for hurricane wind spectra, profiles, and gusts can be improved (Cooper et al., 2013);
revisions are planned to be adopted.
Dong et al. (2012) has shown that joint occurrence period of wind speed and wave height (or other
metocean parameters), important for design and marine operations, can be estimated based on both service
term and risk probability.
The marine industry uses commonly joint metocean models with the environmental contour concept due
to Winterstein et al. (1993), IFORM, (see also (DNV, 2014)) for specification of design criteria. The
concept and associated uncertainties have been recently discussed by Haver et al. (2013) using the North
Sea and the Gulf of Mexico as examples. The authors underline that the contour concept is an approximate
method. Further, they point out that for a broad range of problems, long term q-probability extremes can be
estimated by finding the worst sea state along the q-probability contour for a response considered, and by
estimating 0.9-percentile of the 3-hour extreme value for this sea state. In the North Sea/Norwegian Sea, the
choice of 0.9-percentile will often give a reasonable, but may not necessarily, a perfect estimate. For the
Gulf of Mexico q-probability extremes have been estimated by identifying the most unfavourable
combination of hurricane peak characteristics along the q-probability contour and then finding the 0.95percentile of the 30-minute extreme value for this hurricane peak event. The authors have stressed that they
have little experience with using the method to hurricane governed areas therefore the results for the Gulf of
Mexico should be regarded as an example.
An alternative approach for establishing the environmental contour lines in the original environmental
space has been suggested by Huseby et al. (2013) by utilizing Monte Carlo simulations of the joint
environmental probability. Comparison of this approach with the IFORM concept is presented by Vanem
and Bitner-Gregersen (2014) and discussed in view of engineering applications. The new procedure
although theoretically consistent is not able to estimate non-convex contours and therefore may give
unrealistic values of sea state parameters in some cases.
The joint metocean statistical models were originally developed for design purposes. Bitner-Gregersen
(2015) has proposed an approach allowing use of these models also for marine operations; for illustration of
its application see (Hagen et al., 2015). Hagen and Solland (2013) discusses how weather criteria for
platforms that are unmanned during storms can be calculated to ensure an acceptable safety for personnel.
The study briefly discussed how forecast uncertainty can be accounted in a consistent manner.
More recently, the concept of scenario of wave weather has been revisited (Degtyarev, 2005). When
applied to design, the key idea behind this concept is that an uncountable infinite set of wave conditions may
be replaced by a countable finite set of discrete situations (scenarios).
6.1.3
51
Design for Climate Change and Rogue Waves
Global warming and extreme weather events reported in the last years have attracted a lot of attention not
only in academia and media but also in the shipping and offshore industry. Three important questions are
in focus: will occurrence of extreme weather events increase in the future, which geographical locations
will be most affected, and to what degree will climate change affect future ship traffic and design of
marine structures? Observed and projected changes in wave conditions are expected to have the largest
effect on ship and offshore structure design and operations in comparison to other environmental
phenomena while for offshore platforms changes in sea water level are also crucial. Potential changes in
ocean current and their impact on design and marine operations need also attention.
It is also interesting to note that climate changes resulting in some ocean regions in growth of storm
activity (intensity, duration and fetch) and changes of storm tracks may result in secondary effects such as
increased frequency of occurrence of abnormal waves, also called rogue or freak waves, see e.g. (Cavaleri
et al., 2012), (Bitner-Gregersen and Toffoli, 2014).
To be able to design for climate change time-dependent statistical descriptions need to be adopted.
Statistical extreme value analysis, as currently used in the metocean community, has to be upgraded to take
into account the non-stationary character of current climate, in terms of both climate change trends and
natural variability cycles. These changes need to be incorporated in the risk based approach used currently
in design as proposed by e.g. (Bitner-Gregersen et al., 2013a, Bitner-Gregersen et al., 2014d).
At present climate change and rogue waves are not explicitly included in classification societies rules
and offshore standards due to lack of sufficient knowledge about uncertainties associated with climate
change projections and a consensus reached about probability of occurrence of rogue waves. Significant
uncertainties associated with climate change projections are remaining and further research is needed to
quantify them and to propose how to account for them in design. Further, open access to field wave data,
including rogue waves, and more detailed information about wave conditions when marine accidents occur
in accident databases are also called for.
The risk associated with climate change and rogue waves has been recognized, however, by the shipping
and offshore industry and an adaptation process to climate change and rogue waves has started. The oil
company STATOIL has already introduced an internal requirement accounting in a simplified way for
rogue waves; a structure shall not be put at risk even if it is hit by a wave crest height 10% larger than the
crest height predicted by the second order wave model, see (ISSC, 2013). This is to account for a number of
uncertainties and not only possible rogue wave developments. This requirement is now under discussion for
possible implementation in a revised version of the Norwegian standard NORSOK. How to account for
climate change is also discussed.
The marine industry has initiated several studies to quantify potential impact of climate change on
current design (e.g. (Bitner-Gregersen et al., 2013a, Vanem et al., 2014, Bitner-Gregersen et al., 2014d,
Hagen et al., 2013). Also the international project dedicated to rogue waves, EC project EXTREME SEAS
(Design for Ship Safety in Extreme Seas), and coordinated by Legacy DNV of Norway, and the JIP project
ShortCresT (Effect of ShortCresTedness on extreme wave impact) coordinated by MARIN in The
Netherlands, had been initiated and completed during the period of the ISSC 2015 I.1 Committee. The
ongoing Research Council of Norway (RCN) project ExWaCli (Extreme Waves and Climate Change
Accounting for uncertainties in design of marine structures), coordinated by Legacy DNV of Norway, aims
at understanding impact of climate change on wave conditions in the northern areas in the 21st century,
identifying uncertainties associated with the predicted changes and demonstrating their consequences for
design and operations of marine structures (Bitner-Gregersen et al., 2013b). There are also other research
activities dedicated to rogue waves and climate change going on within the marine industry but still not
publically available. How to account for climate change and rogue waves in current design practice for
tankers (and marine structures in general) is presented by Bitner-Gregersen et al. (2013a) and BitnerGregersen et al. (2014d) and for offshore platforms by Hagen et al. (2013).
The observed increase of ocean temperature will also be a challenge for the marine and renewable
energy industry in the future. Increase of marine growth leading to increase of loads on marine and
renewable energy structures maybe expected in some ocean regions. A lack of marine growth data for
design remains a problem. The photosynthesis occurs only down to about 100200 m, and sunlight
disappears altogether at 1000 m or less, while the ocean descends to a maximum depth of about 11 000 m.
Thus it is not expected seeing marine growth below 500 m or 1000 m. Until recently the deep sea was
largely unexplored, therefore more investigations dedicated to deep sea are needed.
It is interesting to mention that assessments of climate change hazards to electric power infrastructure
such as the Sandy hurricane and flooding of habours and towns have been carried out by the Classification
Society DNV GL (Quan Luna et al., 2014, Yates et al., 2014).
52
6.2
Operations
The reduction of emissions of greenhouse gas (GHG) has become an urgent global task for the prevention of
global warming. In order to reduce the GHG emissions from the international maritime sector ahead of the
other sectors, amendments to MARPOL ANNEX VI making Energy Efficiency Design Index (EEDI) and
Ship Efficiency Management Plan (SEEMP) mandatory were adopted at the 62nd session of Marine
Environment Protection Committee (MEPC 62) held in July 2011, and have entered into force on 1 January
2013.The EEDI is used as an index to assess the energy efficiency of a new ship. On the other hand, the
SEEMP is a management plan for implementing energy efficiency improvement measure (for example,
slow steaming, optimum ship routing, just in time arrival, speed optimization, optimum trim and appropriate
hull maintenance, etc.) during actual operation in an organized and efficient manner. The SEEMP is to be
prepared by the ship owner. IMO has issued guidelines to develop the SEEMP (Resolution MEPC.213(63),
(IMO, 2012)), and the guidelines require at least the following items to be described in the SEEMP:
(1) Energy efficiency improvement measures
(2) Monitoring procedure for energy efficiency
(3) Measureable goal for energy efficiency improvement
(4) Procedure for evaluating the energy efficiency improvement being implemented
SEEMP Guidelines require that energy efficiency is improved by repetitive implementation of the cycle
consisting of Planning, Implementing, Monitoring and Self-evaluation and Improvements. That is,
each ship has to implement the planned efficiency measures by periodic self-monitoring of the energy
efficiency (the fuel consumption, etc.) and evaluate these results so that they can be fed back to the next
efficiency improvement plan. However, the purpose of the SEEMP requirement is to only independently
promote the implementation of efficiency improvements. Therefore, the detailed content of the efficiency
improvement measures to be used or the results (goal attainment level of efficiency improvements) are not
looked into by the third party. In accordance with the SEEMP requirement, ship owners and operators are
required to operate the ships from a viewpoint of environment as well as safety. Especially, accurate
weather forecast is important for the eco-efficiently ship operations. This is because the optimum ship
routing depends on the weather forecast significantly.
At the early stage of deliberations on the EEDI regulation in IMO, there was concern that ships with
excessively small propulsion power would be constructed just for the purpose of improving the EEDI value.
Therefore, discussions on minimum propulsion power in adverse weather condition were started in IMO.
Consequently, for ships which comply with EEDI requirements, it was required that the installed propulsion
power shall not be less than the propulsion power needed to maintain the maneuverability of the ship in
adverse conditions as defined in guidelines developed by IMO. As the result of subsequent deliberations in
IMO, 2013 Interim Guidelines for Determining Minimum Propulsion Power to Maintain the
Maneuverability of Ships in Adverse Conditions were developed and adopted as Resolution
MEPC.232(65) (IMO, 2013). The 2013 interim minimum power guidelines are applicable only to bulk
carriers, tankers and combination carriers of 20,000DWT or above to which compliance with required EEDI
is required during phase 0 (from 2013 to 2014) of the EEDI implementation. In this context, the final
guidelines applicable to ships in phases 1, 2 and 3 are to be developed by IMO at a later stage. The
applicable ships are required to fulfil either of the two assessment levels in accordance with the 2013
Interim Minimum Propulsion Power Guidelines. If a ship does not satisfy the criteria of a level-1
assessment, a level-2 assessment is to be considered.
Table 2. Required minimum propulsion power.
Type of ship
Minimum Propulsion Power (kW)
Bulk Carrier
0.0687 x DWT + 2924.4
Tanker/Combination Carrier
0.0689 x DWT + 3253.0
For the level-1 assessment, minimum power lines for each ship type are calculated using formula as a
function of deadweight shown in Table 2. Installed propulsion power is not to be less than power
calculated using the formula.
The level-2 simplified assessment is an indirect assessment procedure based on an assumption that, in
adverse condition, if a ship has sufficient installed power to move with a certain advance speed in head
waves and wind, and if it is lower than the torque limit within the operating range of the installed engine,
the ship can also be expected to maintain course in waves and wind from any other direction. In the 2013
Interim Minimum Power Guidelines, adverse conditions used for the level-2 assessment are defined as
shown in Table 3.

Table 3. Adverse conditions.
Ship length
Significant wave height Peak wave period
Lpp (m)
(m)
(s)
Lpp200
4.0
200Lpp250
*
7.0 to 15.0
Lpp250
5.5
(* Linearly interpolated value depending on ships length)
53
Mean wind speed

(m/s)
15.7
*
19.0
The maneuverability in waves are stricter in coastal waters than in the open sea: while in the open sea, it
is sufficient to keep a favorable heading with respect to wind and waves, and some drifting with wind and
waves is acceptable, in coastal waters, due to navigational restrictions, the ship might need to keep a
prescribed track irrespective of the direction of waves and wind. On the other hand, ships are not supposed
to be in coastal areas in very severe weather conditions, and should leave to the open sea before the weather
conditions become too severe. Thus, the weather conditions used in the assessment procedure can be relaxed
in comparison with the worst possible weather conditions expected in unrestricted service. In order to
determine the adverse condition, comprehensive assessments were carried out using numerical simulations.
North-Atlantic scatter table from IACS Recommendation 34 (IACS, 2000) was used as a seaway climate
used for comprehensive assessments. Moreover, JONSWAP sea spectrum with the peak parameter of 3.3
was considered for coastal waters. The results of the comprehensive assessments were compared with the
results of the statistical approach (Level-1 assessment). Consequently, the adverse conditions used for level2 assessment were determined taking into consideration both results of the comprehensive assessments and
the statistical approach.
Hereafter, the adverse conditions specified in 2013 Interim Minimum Power Guidelines are validated
by new research projects on the minimum propulsion power such as the EC project SHOPERA (Energy
Efficient Safe SHip OPERAtion), (Papanikolaou et al., 2014) and a new Japanese R&D project
(MEPC67/INF.22, 2014).
6.2.1
Planning and executing marine operations
Operators are normally required to plan the efficiency operations in advance. In case of ships, as mentioned
above, the plans are developed in accordance with the SEEMP requirement. Studies of weather forecast,
hindcast and window analysis are important for planning the efficiency operations of offshore structures as
well as ships.
In order to operate and maintain offshore marine renewables, a device will have to be accessible for a
certain period of time. This will require a weather window consisting of a consecutive period of wave
heights low enough and long enough for the device to be accessed. OConnor et al. (2013) presented the
results of a weather window analysis of wave data from the west coast of Ireland and the Atlantic coast of
Portugal in order to quantify the levels of access to ocean energy renewables, which may be deployed there,
for operation and maintenance activities. The results indicate that the levels of access off Ireland and
Portugal are far below those observed at other marine renewable locations, and at the lower wave height
access limits, there are very few suitable weather windows and considerable winter waiting periods between
these windows. The implications of these low levels of access suggest that maintaining wave energy
converters, off the west coast, may not be feasible and devices will need to be brought ashore for operation
and maintenance activities.
Walker et al. (2013) conducted a similar weather window assessment with application to transit and
deployment operations in the southwest UK, applying a Weibull persistence model to a large wave model
database. The method, which allows computation of access and waiting times requires high quality and high
quantity datasets so as to cover a sufficiently large area to account also for transit times. The specific study
conducted in southwest UK also shows that the level of access to a local marine renewable testing site is
highly dependent to seasonal variability and can be highly reduced in winter.
Numerical forecasts of weather and oceanography are increasingly common in the field of ship operations
due to advances in computer science. However, in some situations, the accuracy of forecasts is too
unreliable to ensure safe operations. Sasa et al. (2013) presented current situation and difficulty of wave
forecast from viewpoint of ship management by using a nationwide questionnaire. They recommend
improving wave forecasts from the viewpoint of ship operations. As a result, Sasa et al. (2014) carried out
feasibility study for improving the forecasting of wave growth pattern from the viewpoint of safe ship
operation. It was pointed out that the further study is necessary in the future.
54
Oil offloading from Spread Mooring System (SMS) FPSO is usually done by means of a dynamically
positioned shuttle tanker (DPST) in tandem configuration. The ST receives the oil pumped by the FPSO
from a bow or stern offloading station, and the operation may take up to 3 days. In order to minimize the
risks associated with the operation, the shuttle tanker (ST) should be kept within a safety zone with respect
to the FPSO, which is usually given as a minimum distance between the two ships and an aperture angle
from the FPSO centerline. In order to guarantee the tanker position during the whole operation, the
operation must be performed with tankers provided with DP (dynamic positioning) systems. Since SMS
FPSOs may be not aligned to the environmental forces, keeping the shuttle tanker in position may be a hard
task for the DP system, depending on the environmental conditions. There are non-rare situations in which
the ST must be disconnected and the operation interrupted. Corra et al. (2013) presented a methodology
based on static calculation of DP capacity for evaluating the downtime of offloading operation in Brazilian
waters. Three generations of DP tankers applied in Brazilian waters were considered. Santos and Campos
Basin long-term (8-year) environmental conditions were used in the downtime calculation. The results
indicated that due to the large variation of wave-wind conditions along the year, both offloading stations are
indeed necessary, since the ST can avoid the conditions in which it is pushed towards the FPSO. The results
also indicated that incrementing the angle that defines the green-zone substantially decreases the offloading
downtime and the number of disconnections required. The risk analysis is beyond the scope of the present
work. The DP power specified for the ST generations II (for Campos Basin) and III (for Santos Basin) are
shown to be quite adequate, since it is demonstrated that increasing this power will not lead to a substantial
reduction in the downtime.
Also the EC project SAFE OFFLOAD (coordinated by Shell) was dedicated to specification of
operational criteria for limiting metocean conditions for LNG terminals. The project has shown that for the
specified target failure probability, the maximum allowable metocean conditions can be determined for a
given operational strategy. When developing acceptance criteria for LNG terminals (which are transferred to
operational criteria), planning as well as carrying out offloading operations both a threshold for significant
wave height (and/or associated metocean parameters) and a required weather window for carrying out
offloading represent important characteristics. They need to be included in a joint metocean description to
allow utilizing it for operational purposes (see (Hagen et al., 2015), (Bitner-Gregersen, 2015)). As
demonstrated by Hagen et al. (2015) wind-sea and swell govern the criteria for LNG terminals. Thus
particular attention should be given to data uncertainty related to estimation of these wave components as
well as models adopted to describe them.
The U.S Office of Naval Research has been developing an Environmental and Ship Motion Forecasting
(ESMF) system capable of real-time predictions of future ship motions that utilizes a Doppler radar to
determine the wave field surrounding the ship; nonlinear wave theory to propagate the wave field forward in
time; and ship seakeeping theory to predict the future ship motions. The ESMF system was designed to
provide real-time environmental data and wave and ship motion forecasts over three time scales: detailed
phase-resolved wave and ship motion time series over 30 seconds, envelopes of the wave and ship motion
time series over 5 minutes, and statistically-averaged parameters characterizing the sea state and ship motions
over 24-48 hours. An ESMF system prototype was tested aboard the R/V Melville during the two week sea
trial in September 2013. Alford et al. (2014) describes the ESMF system with special emphasis placed on the
techniques used to analyze and to predict the future waves at the ship, the real time computations of the ship
motions, and the preliminary results of the sea trials.
6.2.2
Northern Sea Route, Weather routing, Warning Criteria and Current
Changing climate gives potential opportunities for seasonal shipping on the Northern Sea Route, the
Northwest Passage and a potential for Transpolar Route, improving access to many offshore resources in the
Arctic region.
Using monthly and daily CCSM4 sea ice concentration and thickness simulations for different scenarios
covering different periods along the 21st century and considering various radiating forcing, Stephenson et al.
(2013) investigate the technical shipping accessibility for various navigation routes in the arctic. Their
projections, based on capabilities of Polar Class 3, Polar Class 6 and Open Water vessels show that new
areas of the Arctic will become accessible to these classes of vessels. In spite of regional discrepancies all
areas will see their period of access lengthening. For instance along the Northern Sea Route, July-October
navigation season length averages ~120, 113, and 103 days for PC3, PC6, and OW vessels, respectively by
late-century.
Smith and Stephenson (2013) show the ATAM-derived optimal September navigation routes for
hypothetical ships seeking to cross the Arctic Ocean between the North Atlantic (Rotterdam, The
Netherlands and St. Johns, Newfoundland) and the Pacific (Bering Strait) during consecutive years 2006
55
2015 (A and C) and 20402059 (B and D) as driven by ensemble-average GCM projections of sea ice
concentration and thickness assuming RCPs 4.5 (A and B; medium-low radiative forcing) and 8.5 (C and D;
high radiative forcing) climate change scenarios. Red lines indicate fastest available trans-Arctic routes for
PC6 ships; blue lines indicate fastest available transits for common open-water ships. Where overlap occurs,
line weights indicate the number of successful transits using the same navigation route. Dashed lines
indicate national 200-nm EEZ boundaries; white backdrops indicate period-average sea ice concentrations
in 20062015 (A and C) and 20402059 (B and D).
Figure 8.
Stephenson et al. (2013) used the Arctic Transportation Accessibility Model (Stephenson et al., 2011)
with individual and ensemble-averaged datasets of projected sea ice thickness and concentration from seven
climate models: the Australian Community Climate and Earth-System Simulator versions 1.0 and 1.3
(ACCESS1.0, ACCESS1.3), the Geophysical Fluid Dynamics Laboratory version CM3 (GFDL-CM3), the
Hadley Global Environment Model 2 Carbon Cycle (HadGEM2-CC), the Institute Pierre Simon Laplace
medium resolution coupled oceanatmosphere model (IPSL-CM5A-MR), the Max Planck Institute for
Meteorology Earth System Model (MPI-ESM-MR) and the National Center for Atmospheric Research
(NCAR) Community Climate System Model version 4 (CCSM4), assuming two different climate change
scenarios and two vessel classes, to assess future changes in September Arctic shipping potential (see Figure
8). They found that by midcentury, changing sea ice conditions will enable:
Expanded September navigability for common open-water ships crossing the Arctic along the Northern
Sea Route over the Russian Federation.
Robust new routes for moderately ice-strengthened (Polar Class 6) ships over the North Pole.
New routes through the Northwest Passage for both vessel classes.
The limitation of the Northern Sea Route not sufficiently discussed is the water depth on parts of the
route. This limitation may require that ships will need to sail farther from the coast where ice coverage is
larger.
The development of decision support systems remains a focus. Application of such systems require
collection of relevant data such metocean conditions and ship response, on board. The recently completed
EC project NavTronic has demonstrated that these types of data can also be used for self-learning.
Several authors have studied relations between spectral parameters and occurrence of extreme and
rogue waves and the topic is still under development. It has also been investigated in the EC EXTREME
SEAS project in collaboration with the the European Centre for Medium-Range Weather Forecast.
Recently an approach for coupling the wave spectral model with the nonlinear phase resolving model has
56
been proposed by Bitner-Gregersen et al. (2014d) for use in forecasting of these abnormal waves; the
results are promising.
6.2.3
Eco-Efficiency Ship Operation
In order to reduce the GHG emissions from international maritime sector, the International Maritime
Organization (IMO) requires to promote higher efficiency of ship operations using energy efficiency
improvement measures such as optimum ship routing and just-in-time speed operation of ships. As a result,
shipping companies is required to promote higher efficiency of ship operation and also to evaluate effect of
the operation in accordance with the SEEMP requirement. The purpose of classical weather routing is to
search the best route based on the existing weather/current forecast data from the aspect of safety first and
economy. On the other hand, for the reason of the SEEMP requirement as well as rising cost of fuel, latest
optimum ship routing is carried out based on various data such as weather/current forecast, operation
schedule, fuel consumption, speed, trim, ship characteristics and real time monitoring etc. in order to
achieve a good balance between safety and environment. The objective is not to avoid all adverse weather
but to find the best balance to minimize transit time and fuel consumption without placing the vessel at risk
to weather damage or crew injury. Therefore, there have been so many studies on the optimum ship routing
to achieve further eco-efficiency ship operation. To seek the good balance between safety and environment
is core challenges for todays ship owners and operators.
Lin and Fang (2013) proposed a ship weather-routing algorithm based on the composite influence of
dynamic forces, i.e. wind, wave and current forces, for determining the optimized transoceanic voyages. The
developed routing algorithm, three-dimensional modified isochrones (3DMI) method, utilized the recursive
forward technique and floating grid system for both the east- and west-bound ship routes in the North
Pacific Ocean. In order to achieve the goals of minimized fuel-consumption or the maximized-safety routes
for the transoceanic voyages, two sailing methods were applied as the prerequisite routes in the earth
coordinate systems. The proposed calculation was verified to be effective for the optimized sailings by
adjusting the weighting parameters in the objective functions.
Chu et al. (2013) assessed the impact of METOC ensemble forecast systems on optimal ship route.
Evaluation of a weather routing decision aid for operational fleet use and concept of operations were also
conducted for the USS Princeton guided missile cruiser (CG)-59 in a sea trial test following the 2012 Rim of
the Pacific exercises. They were able to assess the impact and sensitivity of the SVPDA modeling to
METOC input parameters. Environmental model uncertainties were quantified through ensemble modeling.
It was found that the SVPDA model was very sensitive to: location, direction, seasonal synoptic/mesoscale
weather, hull/propulsion type and condition, route length, specific model improvements, and ensemble
methods. The possibility of significant fuel cost reduction was also identified by utilizing the best ensemble
member with the maximum fuel-saving of 20%.
Recently Essig et al. (2013) described the development of the Smart Voyage Planning Decision Aid
(SVPDA) from discovery to possible implementation in the context of a Maritime Energy Portfolio Tool
focused on providing the necessary data needed for evaluating energy saving initiative implementation and
operational strategies that optimize mission capability, energy savings, and return on investment.
On-board measurement of fuel consumption of a ship has been carried out in a relatively severe sea
condition. In the full scale experiment, the ship traveled on several courses to investigate the change of fuel
consumption relative to the encounter wave angle. The result shows that the wave direction has a great
influence on the main engine horse power and fuel consumption, and also shows a possibility of fuel
efficiency prediction. In order to develop an eco-friendly navigation support system, Iseki and U.D. (2013)
applied results of Bayesian wave estimation to fuel efficiency prediction, since the Bayesian method does
not require wave measurements but needs only ship motion data as input and the method is suitable for onsite wave estimation. It was shown that the proposed concept of an eco-friendly navigation system is
effective and worth further investigation.
During these times of fluctuating freight rates and oversupply, selection of optimum speed will give an
operator a crucial advantage. Until the recession, the emphasis has always been on larger capacity and
higher speed. Now, design innovation and slow steaming are becoming important. Khor et al. (2012)
showed that an optimum speed can be selected at design stage through a combined revenue-cost and
propulsion analysis. New software was set up to assist speed optimization process for large container
ships. The software was used both propulsion and revenue-cost analysis with an ultra-large container ship
model of 20,000 TEU as a case study to derive the most profitable speed. The results showed that 19.5
knots is the optimum speed which is a departure from current trend of 25 knots but supporting the concept
of slow steaming.
57
Speed optimization schemes face challenges in daily vessel operations due to very strict constraints
imposed by itinerary and due to limited accuracy of available weather and sea current forecasts. Since fueloptimal routing is highly sensitive to constraints such as just-in-time arrival, one high speed leg can wipe
out the accumulated savings of an entire voyage. Ilus and Heikkinen (2012) presented an approach to
optimal speed estimation which is based on statistical route forecasts based on historical data measured on
specific routes. Simulation was used to provide alternative and improved energy efficient speed profiles.
The simulation of optimal speed profiles suggested possible 3 1% energy savings.
As of today vessels have and will in future have even more possibilities to affect the ships overall energy
balance. These possibilities are enabled for example with diesel electric configurations, waste heat recovery
units, trim variations, operational profiles, alternative fuels and hull cleaning schedules. Unfortunately all
this flexibility comes with certain problems as the system complexity grows beyond human understanding.
Finding and especially operating constantly at the optimum point is significantly more challenging with all
these variables. Even if one or two specific areas can be efficiently optimized by for example a dedicated
chief engineer, the full potential of comprehensive optimization is often left unused. Ignatius et al. (2014)
took an insight into a holistic performance management and optimization system for any type of vessel. It
not only takes in account energy efficiency but also the availability and safety of the vessel and fleet. The
system produces a comprehensive optimization and monitoring framework for overall energy efficiency of
basically any process, and gives clear decision support both to the users onboard and also for the
management ashore.
Shipping companies seek to promote higher efficiency of ships operation and also to evaluate effect of
the operation. Just-in-time speed operation holds the prospective effect of major greenhouse gas (GHG)
reduction. However, the methodology to evaluate the effectiveness by using these services has not
confirmed. Kano and Namie (2014) introduced the outline of Eco shipping support system for the
navigation plan an optimal speed voyage by taking into account wind, wave and current forecasts. And the
estimation methodologies of amount of GHG emission reductions by the speed planning from the System
and applicability to RORO ship were confirmed.
Development and decrease cost of onboard sensor technology, data collection systems and satellite
internet connections have opened new possibilities to collect extensive datasets of the performance related
information from vessels. Analyzing this performance related data is challenging because the performance
of a ship is affected by many complex factors such as wind, waves, currents, shallow water and often the
ships operating condition has significant impact to performance such as variety of displacements in
operation. Therefore normalization methods are required for making the performance readings measured
and collected from a ship comparable. Normalization in this means evaluating the effect of these different
factors affecting ships performance during the time of measurement and correcting then the result to given
baseline condition which can for examples be calm water, design draft and speed condition or the average
operation condition of the ship. Kariranta (2012) showed to utilize normalized performance data for ship
operations and design. The normalization was carried out by using hydrodynamic functions and statistical
approach. Direct commercial use of such normalized data could be sharing the profit of a shipping pool
based on the real performance of the ships in it and for example making virtual arrival calculations more
exact. It also gives good input for the design stage of the ship by allowing the designers to take a look on
real operation conditions that the ships are in for example for calculating the required sea margin more
accurately, use the virtual voyage as design criteria instead of only using one design speed and draft and
give feedback how different hull shapes react for wave loads.
In general, the speed power performance of ships is optimized for design speed and draught in
accordance with the contract condition. But, the contract condition may not be always the same as the
actual operating condition. Therefore, in order to reduce the fuel consumption practically, it is
necessary to optimize the performance under various conditions considering the actual voyage. This is
the reason that the trim optimization covering various operating profiles becomes the main issue in
reducing fuel oil consumption. Lee et al. (2014) carried out a numerical study to optimize trim
conditions through the computational evaluation system with variation in draught, ship speed and
voyage trim. The results of the trim optimization performed numerically are well-matched with the
towing test result. It is confirmed that the computational evaluation system is a useful and efficient tool
for trim optimization and the provided optimum trim will be able to contribute to fuel savings under the
operating conditions.
7.
CONCLUSIONS
The issue of metocean data ownership remains a general problem even though access to new databases has
become available since 2012 (e.g. MONET, (Quiniou-Ramus et al., 2013)) and awareness of its importance
58
has increased. Whilst the advantages of having data freely available to academia and industry are clear, the
commercial sensitivity of some data sets is recognised. An example of making data available without
compromising their confidentiality is the SIMORC URL database: http://www.simorc.org/, administered by
the University of Southampton.
A question raised today is: Are the metocean measurements actually the ground truth? An answer to it
still does not exist. There are several uncertainties related to field and remotely sensed data and complete
knowledge about them is lacking. The stationarity and homogeneity assumption of measurements is
obviously questionable and likely not valid in some circumstances. The topic requires attention.
Development and decrease cost of onboard sensor technology, data collection systems and satellite
internet connections have opened new possibilities to collect extensive datasets of the performance related
information from vessels, offshore platforms and renewable energy installations. These so-called Big Data
include information about load and response conditions of structures and related metocean climate. A
methodology for utilization this information is still under development and both academia and industry is
interested in it.
Accuracy of wind and waves hindcast databases for several ocean basins has improved since 2012
through improved understanding of the physics of metocean phenomena. Validation of these databases
against field measurements and remotely sensed data for significant wave beyond 14 m is still too
limited.
Many research efforts in the last decade have contributed to understanding of mechanisms generating
rogue waves and their detailed dynamic properties. The state of the art development on rogue waves are
well summarised at three Rogue Wave Workshops, held in 2000, 2004, and 2008 by Ifremer, and the
publications reviewed by the previous and present ISSC I.1 Committee. Consistency achieved between
numerical models and experimental data has been documented. Since 2012, the focus has been given on
forcing terms like wind, current and wave breaking that are not typically included. Several photos of rogue
waves observed in the nature have been collected in a book of Olagnon and Kerr (2015) which may
encourage to new research findings.
Simplified definitions of rogue waves such as the wave height and crest criteria are commonly applied.
Despite recent achievements, a consensus on the probability of occurrence of rogue waves has not been
reached yet. Such consensus, however, is essential for a systematic evaluation of possible revision of
classification society rules and offshore standards, which currently do not include rogue waves explicitly.
The EC EXTREME SEAS project has contributed to new findings on probability of occurrence of rogue
waves which need to be further explored. It is worth to mention that the oil company STATOIL has already
introduced an internal requirement accounting in a simplified way for rogue waves, see (ISSC, 2013). This
requirement is now under discussion for possible implementation in the revised version of the Norwegian
standard NORSOK.
Attention to directional effects, modelling of wind sea and swell, seasonality, spatial and non-stationary
statistics continues. New studies documenting the importance of these effects on extreme metocean statistics
have been carried during the period of the Committee.
With the increase of offshore wind energy installations, reliable forecasts of the order of hours or
minutes are also becoming increasingly important since the complex electrical networks are sensitive to
large fluctuations, which may occur at the onset of a storm. In addition, more information on the wind
profile in the lower atmospheric layer is needed for the design and analysis of these structures. Long-term
trends, not only in the occurrence of extreme events but also other statistical properties will remain an
important research topic in wind analysis in forthcoming years. Further, non-stationarity and nonhomogeneity of the wind field needs to be accounted for.
CFD (Computational Fluid Dynamics) methodology is getting increasing focus in modelling of
metocean phenomena but it is still in development. Attention to it will grow in the future.
The Fifth Assessment Report (IPCC, 2013) is confirming the conclusion of the Fourth Assessment
Report (IPCC, 2007); the observed climate changes are due to human activities. Climate change projections
show that ice, sea water level, wind, waves and ocean circulations will be affected by global warming but
this will be much regional dependent. Extreme value estimates of wind and waves needed for design work
may be more affected by climate changes than the average values although there are some examples where
they were less affected. Too little attention has been given in AR5 to wind and waves and too few
publications are written from the viewpoint of the designer, focusing often on too low return periods.
However, a significant development has taken place since AR4 in the increased use of quantitative statistical
measures simplifying synthesis and visualization of models performance.
On a positive note for the marine community is the emergence of potential opportunities for seasonal
shipping on the Northern Sea Route, the Northwest Passage and a potential Transpolar Route, improving
access to many offshore resources in the Arctic region. On the negative side increases of probability of
59
encountering icebergs, bergy bits and other ice formations in some areas maybe expected where they
were not previously experienced. The increased intensity of tropical cyclones has caused devastating
damage to the offshore industries in the Caribbean in the past 11 years; the link with the warming climate
is debatable but if so these effects would be anticipated to continue as warming continues. The observed
trends and projected climate changes indicate that significant impact on marine structure design may be
expected in some ocean regions. It is noted that there is large uncertainty associated with the climate
projections.
Awarness of importance of accounting for uncertainties associated with environmental description in risk
assessment of ship, offshore and renewable energy structures is continuously increasing within the marine
and renewable energy industry. Although several data, statistical and model uncertainties have been
reported during the period of the ISSC 2015 Committee I.1 a systematic investigation of them is still
lacking. To increase further awareness of a role uncertainties have in design work the 2nd ITTC-ISSC Joint
Workshop on uncertainty modelling took place 30 August 2014 in Copenhagen. The Worshop was
organized by the ITTC Seakeeping Committee with contribution from the ISSC 2015 Committee I.1, the
ISSC 2015 I.2 (Loads) Committee and the ITTC Ocean Engineering Committee.
Operational aspects of ships have got a lot of attention in the report as they are reflecting topics being
under discussion in IMO such as energy efficiency, ship maneuverability in weather conditions and of
reduction of pollution from ships.
The needs of the renewable energy industry as well as climate change have strong impact on research
directions within metocean description and this trend is expected to continue.
7.1
Advances
New metocean databases have been opened for the users. Access to wind and waves remote sensing
databases is continuously improving. Utilisation of wind and wave information collected by satellites in
wave models is increasing, particularly due to the GlobWave project initiated by the ESA in 2008.
Wave models have largely improved over the recent years thanks to new developments in
parameterization, introducing more consistent description of the physics based on observations, and
numerical choices, introduction of currents, coastal reflection, and bottom sediment (prime interest for
modeling in coastal waters) in the models. Most of these improvements were implemented in a new version
(4.18) of the code WaveWatch III that was released in March 2014 by NOAA/NCEP allowing the use of
unstructured grids and introducing new parameterizations for wave dissipation together with new
parameterizations for bottom friction including movable bed roughness. An attempt has also been made to
include quasi-resonance interaction in the wave model lacking today.
Due to development of computers wave frequency-directional wave spectra have started to be
archived by met-offices opening new possibilities for environmental modelling and design and operation
work.
The knowledge about extreme and rogue waves has advanced since 2012; the nonlinear dynamics of
surface gravity waves is now reasonably understood. The predictions made by theoretical and numerical
models compare well with experimental results. Progress has been taken place regarding inclusion of
forcing terms such as wind, current and wave breaking. Indirect evidence on the specific meteorological and
oceanographic conditions leading to the formation of rogue waves has been strengthened by new field
observations.
The first more systematic investigations of extreme and rogue waves and ship behaviour in these waves
were carried out in the EC ERXTREME SEAS project, while behaviour of offshore platforms in these
waves was investigated by the JIP Crest/ShortCrest project; both projects were completed in the period of
the Committee.
The importance of accounting for non-stationarity and non-homogeneity of environment data and
models has been demonstrated by examples. Also further progress has been made on development of spatial
and temporal models. Recognition of the need to consider covariates when performing extreme value
analysis has been shown.
The increased use of renewable energy sources, especially offshore wind energy, has triggered many
new research activities. With the limited number of profitable locations came a trend towards higher
structures up to some two hundred metres. However, it is not the wind energy industry alone that focuses on
more accurate wind data in lower layers of the atmosphere; both academia and others industrial sectors are
also interested in it. The trends in the development of new sensors and data acquisition techniques are
expected to continue.
The issue of the Fifth Assessment Report by the Intergovernmental Panel on Climate Change (IPCC,
2013) is an important milestone in climate change research. A variety of model predictions of metocean
phenomena is available now to the users. The Coordinated Ocean Wave Climate Project (COWCLIP) has
60
found that the variance of wave-climate projections associated with wave downscaling methodology
dominated other sources of variance within the projections, e.g. the climate scenario or climate model
uncertainties, but that needs further exploration.
7.2
Recommendations
The need for improving the availability, quality and reliability of metocean databases was reported by
the previous ISSC I.1 Committees. This situation is mostly unchanged, and any effort to address this
concern is recommended. Further utilisation of remote sensing data by the marine industry needs to
continue. A lack of marine growth data for design remains a problem. In a case of ocean temperatures
increase marine growth will be a challenge in some ocean regions.
A question Are the metocean measurements actually the ground truth? needs further exploration.
Whenever observations of metocean conditions are made, stationarity and ergodicity needs to be
addressed. A consistent methodology for handling Big Data is called for. Development of it will require
collaboration between academia and industry. Some initiatives for establishing such collaboration have
already taken place. A remaining question is: how to utilize Big Data in design and operation work?
Studies dedicated to including wave breaking and external forcing (wind) when modelling rogue
waves should continue. Detailed investigations of meteorological and oceanographic conditions in
which extreme and rogue waves occur together with analyses of field wave time series (uncertainty due
to sampling variability may be a problem here) are needed to reach a consensus about probability of
occurrence of rogue waves; this being mandatory for evaluation of possible revision of classification
society rules and offshore standards. Such studies should focus on metocean conditions during storm
growth and decade, crossing seas as well as presence of current. Further, a consensus needs to be
reached on how combining new information about extreme and rogue waves in design and operation
work. The effect of modulational instability (one of the mechanisms responsible for generation of rogue
waves) is gradually suppressed, when the wave energy spreading increases. This needs to be reflected
in possible revisions of rules and offshore standards.
Inclusion of nonlinear and rogue waves in commonly used codes for analyses of wave-structure
interaction is strongly recommended. This has been initiated in the EC EXTREME SEAS project for
ships and in the JIP CresT/ShortCrest project for offshore structures, but further investigation is still
called for. Rogue waves affect not only local loads but also global loads.
Focus needs to be given to properly accounting for directional effects in design, assuring
consistency between omnidirectional and directional criteria, to seasonality, spatial and non-stationary
statistics, and current profile as well as modelling of wind sea and swell for design and operational
purposes. Utilization of frequency-directional wave spectra, archived by met-offices today, in this work
is strongly recommended.
Studies contributing to a systematic quantification of uncertainties of metocean description should
continue both in academia and industry. It is also recommended to agree on common definitions of
uncertainties within academia and industry.
The 2015 ISSC I.1 Committee recognizes the significance of the IPCC (2013) findings and the
conclusions drawn by the Panel. There are still significant uncertainties associated with climate change
projections. Identification and reduction of these uncertainties as well as consistently combining them
requires attention. That is of crucial importance for the shipping, offshore, renewable energy and
coastal engineering industry. Both the mitigation as well as the adaptation process to climate change
should continue within the industry. The climate changes caused damages but also open new
opportunities for Arctic development and challenges when shipping the goods to main economic
centers. To take advantage of these opportunities new technologies will be required to safely operate in
polar ice environments which need to be based on reliable metocean and ice data, and models.
Attention needs be given not only to the Arctic regions but also to the Antarctic waters.
ACKNOWLEDGEMENTS
The authors would like to express their thanks to the 2015 Committee I.1 Liaison Prof. Carlos Guedes
Soares for following the Committee during development of the report and for organization of the first
Committee meeting. The Committee would like also thanks Prof. Pandeli Temarel, ISSC 2012 I.2
(Loads) Committee, Prof. Yonghwan Kim, ITTC Seakeeping Committee, and Prof. Qiu, Wei, ITTC
Ocean Engineering Committee, for collaboration and valuable discussions. Also the ISSC 2015 Arctic
Technology Committee V.6 is acknowledged for exchange of information on ice. The ISSC 2015
Committee I.1 Chairman is addressing her thanks to the ISSC 2015 Standing Committee for giving her
the opportunity to lead the Committee I.1 for the third time.
61
REFERENCES
Aarnes, O. J., Breivik, O. & Reistad, M. 2012. Wave Extremes in the Northeast Atlantic. Jouranl of Climate, 25,
15291543.
Agarwal, P. & Mcneill, S. Extreme directional and planar profiles for Kuroshio current using Inverse Form and
proper orthogonal decomposition. Proc. OMAE 2013, 914 June 2013 Nantes, France.
Aiki, H. & Greatbatch, R. J. 2012. Thickness-Weighted Mean Theory for the Effect of Surface Gravity Waves on
Mean Flows in the Upper Ocean. Journal of Physical Oceanography, 42(5), 725747.
Ailliot, P., Bessac, J., Monbet, V. & Pne, F. 2014. Non-homogeneous hidden Markov-switching models for wind
time series HAL, hal-00974716, hal.archives-ouvertes.fr.
Ailliot, P. & Monbet, V. 2012. Markov-switching autoregressive models for wind time series. Environmental
Modelling and Software, 92101.
Amurol, S., Ewans, K. & Sheikh, R. Measured wave spectra offshore Sabah & Sarawak, Malaysia Proceedings of
OTC-Asia, March 2628 2014 Kuala Lumpur, Malaysia.
Andrioni, M., Lima, J.A.M., Guerra, L.A., Ribeiro, E.O., Nunes, L.M.P., Ceccopieri, W., de Souza Rego, V. & de
Oliveira, S.M. 2012. Ocean Eddies Influence on Lula Field, Santos Basin, Brazil. Proceedings of the OMAE
2012 Conference, 16 July 2012b Rio de Janeiro, Brazil.
Aoyama, M., Tsumune, D., Uematsu, M., Kondo, F. & Hamajima, Y. 2012. Temporal variation of 134 Cs and 137
Cs activities in surface water at stations along the coastline near the Fukushima Dai-ichi Nuclear Power Plant
accident site, Japan. Geochemical Journal 46, 321325.
API (American Petroleum Institute). 2012. Derivation of metocean design and operating conditions, API RP 2MET,
First Edition.
Ardhuin, F., Chapron, B., Collard, F. 2009. Observation of swell dissipation across oceans. Geophysical Research
Letters, 36.
Ardhuin, F., Jenkins, A. D. & Belibassakis, K. A. 2008a. Comments on The Three-Dimensional Current and
Surface Wave Equations. .Journal of Physical Oceanography, 38 (6), 13401350.
Ardhuin, F., Rascle, N. & Belibassakis, K. A. 2008b. Explicit wave-averaged primitive equations using a
generalized Lagrangian mean. Ocean Modelling, 20(1), 3560.
Ardhuin, F., Rogers, E., Babanin, A. V., Filipot, J.-F., Magne, R., Roland, A., Van Der Westhuysen, A., Queffeulou,
P., Lefevre, J.-M., Aouf, L. & Collard, F. 2010. Semi-empirical dissipation source functions for wind-wave
models: Part I. Definition, calibration and validation. J. Phys. Oceanogr., 40(9), 19171941.
Ardhuin, F., Roland, A., Dumas, F., Bennis, A.-C., Sentchev, A., Forget, P., Wolf, J., Girard, F., Osuna, P. &
Benoit, M. 2012. Numerical wave modeling in conditions with strong currents: Dissipation, refraction, and
relative wind. J. Phys. Oceanogr., 42, 21012120.
Babanin, A. Swell Attenuation due to Wave-induced Turbulence. Proceedings of the OMAE 2012 Conference, 16
July 2012 Rio de Jainerio, Brazil.
Babanin, A. V. & Mcconochi, J. Wind measurements near the surface of waves. Proc. OMAE 2013, 914 June
2013 Nantes, France.
Bamber, J. L. & Aspinall, W. P. 2013. An expert judgement assessment of future sea level rise from the ice sheets.
Nature Clim. Change, 3, 424427.
Banari, A., Grilli, S. T. & Janssen, C. F. Two phase flow simulation with Lattice Boltzmann method: application to
wave breaking. Proc. OMAE 2013 Conference, 914 June 2013 Nantes, France.
Barrand, N. E., Vaughan, D. G., N., S., M., T., Kuipers Munneke, P., Van Den Broeke, M. R. & Hosking, J. S. 2013.
Trends in Antarctic Peninsula surface melting conditions from observations and regional climate modeling.
Journal of Geophysical Research: Earth, 118 315330.
Benetazzo, A., Carniel, S., Sclavo, M. & Bergamasco, A. 2013. Wave-current interaction: Effect on the wave field
in a semi-enclosed basin. Ocean Modelling 70, 152165.
Bennis, A.-C. & Ardhuin, F. 2011. Comments on 'The Depth-Dependent Current and Wave Interaction Equations: A
Revision. Journal of Physical Oceanography, 41(10), 20082012.
Birknes, J., Hagen, ., Johannessen, T.B. & Nestegrd, A., 2013. Second-Order Kinematics Underneath Irregular
Waves. Proceedings OMAE 2013, 914 June 2013, Nantes, France.
Bitner-Gregersen, E. M. 2012. Joint long-term models of met-ocean parameters. In: Carlos Guedes Soares, Y. G., N.
Fonseca, A.P. Texeira (ed.) Marine Technology and Engineering: CENTEC Anniversary Book. London, UK:
CRC Press, A. A. Balkema, Taylor and Francis.
Bitner-Gregersen, E. M. 2015. Joint met-ocean description for design and operations of marine structures. Applied
Ocean Research, In press.
Bitner-Gregersen, E. M., Bhattacharya, S. K., Chatjigeorgiou, Y. K., I., E., Ellermann, K., Ewans, K., G., H.,
Johnson, M. C., Ma, N., C., M., Nilva, A., I., R. & Waseda, T. 2014a. Recent developments of ocean
environmental description with focus on uncertainties. Ocean Engineering, 86 (2014), 2646.
Bitner-Gregersen, E. M., Cramer, E. H. & Korbijn, F. Environmental Description for Long-term Load Response of
Ship Structures. Proceedings of ISOPE-95 Conference, 1116 June 1995 The Hague, The Netherlands.
Bitner-Gregersen, E. M., Eide, L. I., Hrte, T. & Skjong, R. 2013a. Ship and Offshore Structure Design in Climate
Change Perspective, Springer.
62
Bitner-Gregersen, E. M., Eide, L. I., Hrte, T. & Vanem, E. Impact of Climate Change and Extreme waves on
Tanker Design. Proceedings of the SNAME Conference, 2325 October 2014d Houston, USA.
Bitner-Gregersen, E. M., Eide, L. I., Reistad, M. & Huseby Bang, A. ExWaCli: Extreme waves and climate change:
Accounting for uncertainties in design of marine structures. Proceedings of the13th International Workshop on
Wave Hindcasting and Forecasting, 27 October1 November 2013b Banff, Alberta, Canada.
Bitner-Gregersen, E. M., Ewans, K. C. & Johnson, M. C. 2014b. Some uncertainties associated with wind and
waves description and their importance for engineering applications. Ocean Engineering, 86 (2014), 1125.
Bitner-Gregersen, E. M., Fernandez, L., Lefevre, J.-M. & Toffoli, A. 2014c. The North Sea Andrea storm and
numerical simulations. Natural Hazards and Earth System Sciences, 14.
Bitner-Gregersen, E. M. & Hagen, . 1990. Uncertainties in Data for the Offshore Environment. Struct. Safety, 7, 1134.
Bitner-Gregersen, E. M. & Magnusson, A. K. 2014. Effect of intrinsic an sampling varaibility on wave parameters
and wave statistics. Ocean Dynamics, 64(11), 16431655.
Bitner-Gregersen, E. M. & Toffoli, A. 2012a. On the probability of occurrence of rogue waves. Natural Hazards
and Earth System Sciences, 12 751762.
Bitner-Gregersen, E. M. & Toffoli, A. A semi-empirical wave crest distribution of random directional wave fields.
Proceedings of the OMAE 2012 Conference, 16 July 2012, Rio de Janeiro, Brazil.
Bitner-Gregersen, E. M. & Toffoli, A. 2014. Probability of occurrence of rogue sea states and consequences for
design of marine structures. Ocean Dynamics, 64, 14571468.
Bitner-Gregersen, E. M. 2015. Joint met-ocean description for design and operations of marine structures. AOR,
http://dx.doi.org/10.1016/j.apor.2015.01.007
Bmt 1986. British Maritime Technology (Primary Contributors: Hogben N., Da Cunha L.F., Oliver H.N.) Global
Wave Statistics. Atlas. London, England: Unwin Brothers Limited.
Boudiere, E., Maisondieu, C., Ardhuin, F., Accensi, M., Pineau-Guillou, L. & Lepesqueur, J. 2013. A suitable
metocean hindcast database for the design of Marine energy converters. International Journal of Marine Energy,
34, e40e52.
Box, J. E. 2013. Greenland ice sheet mass balance reconstruction. Part II: Surface mass balance (18402010). J.
Climate, 26, 69746989.
Box, J. E. & Colgan, W. 2013. Greenland ice sheet mass balance reconstruction. Part III: Marine ice loss and total
mass balance (18402010). J. Clim., 26, 69907002.
Bracegirdle, T. J. & Stephenson, D. B. 2012. Higher precision estimates of regional polar warming by ensemble
regression of climate model projections. Clim. Dyn., 39, 2805282.
Bratland, A. K., Brresen, K. & Berntsen, P. I. B. Higher Order Wave-Current Elevations in Deep Water. 30th
International Conference on Ocean, Offshore and Arctic Engineering, 1924 June 2011 Rotterdam, The
Netherlands. 19.
Bricheno, L. M., Soret, A., Wolf, J., Jorba, O. & Baldasano, J. M. 2013. Effect of high-resolution meteorological
forcing on nearshore wave and current model performance. J. Atmos. Oceanic Technol., 30, 10211037.
Buchner, B. & Forristal, G. Are basin wave different from field waves? (First investigations during CREST JIP).
Proc. 31st Int. Conf. on Ocean, Offshore and Arctic Eng. OMAE 2012, 1-6 July 2012, Rio de Janeiro, Brasil.
Campbell, A. J., Bechle, A. J. & Wu, C. H. 2014. Observations of surface waves interacting with ice using stereo
imaging. J. Geophys. Res. Oceans, 119, 32663284.
Cardone, V. J., Callahan, B. T., Chen, H., Cox, A. T., Morrone, M. A. & Swail, V. R. 2014. Global distribution and
risk to shipping of very extreme sea states (VESS). International Journal of Climatology.
Cardone, V. J. & Cox, A. T. Modelling Very Extreme Sea States (VESS) in real and synthetic design level storms.
Proceedings of the OMAE2011 Conference, 2011 Rotterdam, The Netherlands.
Cartwright, D. E. & Longuet-Higgins, M. S. 1956. Statistical distribution of the maxima of a random function.
Proc., Roy. Soc. A,, 237, 212232.
Cavaleri, L., Bertotti, L., Torrisi, L., Bitner-Gregersen, E., Serio, M. & Onorato, M. 2012. Rogue waves in crossing
seas: The Louis Majesty accident. Journal of Geophysical Research, 117.
Ceccopieri, W. & Silveira, I.C.A. 2012. Is the Vertical Variability of the Ocean in Santos Bight, Brazil, Dominated
by the Western Boundary Current Meanders? Proceedings of the OMAE 2012 Conference, 16 July 2012, Rio de
Janeiro, Brazil.
Chabchoub, A., Hoffmann, N., Onorato, M., Slunyaev, A., Sergeeva, A., Pelinovsky, E. & Akhmediev, N. 2012.
Observation of a hierarchy of up to fifth-order rogue waves in a water tank. Phys Rev E 86.
Chalikov, D. & Babanin, A. V. 2012. Simulation of wave breaking in one-dimensional spectral environment. J.
Phys. Oceanogr., 42, 17451761.
Chawla, A., Spindler, D. & Tolman, H. L. 2013. Validation of a thirty year hindcast using the climate forecast
system reanalysis wind. Ocean modeling, 70, 189206.
Chawla, A., Tolman, H. L., Gerald, V., Spinler, D., Spindler, T., Alves, J.-H., Cao, D., Hanson, J. & Devaliere, E.M. 2012. A multigrid wave forecasting model: A new paradigm in operational wave forecasting. Weather and
Forecasting, 28.
Chella, M. A., Bihs, H., Kamath, A. & Muskulus, M. Numerical modelling of breaking waves over a reef with a levelset based numerical wave tank. Proceedings of the OMAE 2013 Conference, 914 June 2013 Nantes, France.
63
Chen, S. S., Price, J. F., Zhao, W., Donelan, M. A. & Walsh, E. J. 2007. The CBLAST-Hurricane program and the
next-generation fully coupled atmosphere-wave- ocean models for hurricane research and prediction. Bull. Amer.
Meteor. Soc., 88, 311317.
Chen, S. S., Zhao, W., Donelan, M. A. & Tolman, H. L. 2013. Directional wind-wave coupling in fully coupled
atmosphere-wave-ocean models: Results from CBLAST-Hurricane. J. Atmos. Sci., 70, 31983215.
Chen, Y.-Y. & Chen, H.-S. 2014. Lagrangian solution for irrotational progressive water waves propagating on a
uniform current: Part 1. Fifth-order analysis. Ocean Engineering, In press.
Chen, Y.-Y., Chen, H.-S., Lin, C.-Y. & Li, M.-S. 2012. Lagrangian Solution for an Irrotational Progressive Water
Wave Propagating on a Uniform Current. Journal of Atmospheric and Oceanic Technology, 30(4), 825845.
Chen, Y.-Y., Hsu, H.-C. & Chen, G.-Y. 2010. Lagrangian experiment and solution for irrotational finite-amplitude
progressive gravity waves at uniform depth. Fluid Dynamics Research, 42(4).
Cherneva, Z. & Guedes Soares, C. 2014. Timefrequency analysis of the sea state with the Andrea freak wave. .
Nat. Hazards Earth Syst. Sci., 14, 31433150.
Cherneva, Z., Tayfun, M. A. & Guedes Soares, C. 2013. Statistics of waves with different steepness simulated in a
wave basin. Ocean Engineering, 60 (2013), 186192.
Choi, B. H., Min, B. I., Kim, K. O. & Yuk, J. H. 2013. Wave-tide-surge coupled simulation for typhoon Maemi.
China Ocean Eng., 27(2) 141158.
Christopher, B., Smith, J. M., Kenney, A. & Jensen, R. 2013. STWAVE simulation of hurricane Ike; Model results
and comparison data. Coastal Eng., 73, 5870.
Christou, M. & Ewans, K. 2014. Field measurements of rogue water waves. J. Phys. Oceanogr., in press.
Chu, P. C., Miller, S. E. & Hansen, J. A. 2013. Fuel-saving ship route using the Navys ensemble meteorological
and oceanic forecasts. Journal of Defense Modeling and Simulation: Applications, Methodology, Technology,
201X, Vol XX(X), 116.
Clauss, G., Klein, M., Dudek, M. & Onorato, M. Application of Breather Solutions for the Investigation of
Wave/Structure Interaction in High Steep Waves. 31th OMAEInternational Conference on Ocean, Offshore and
Arctic Engineering, 16 July 2012 Rio de Janeiro, Brazil.
Comiso, J. C. & D.K., H. 2014. Climate trends in the Arctic as observed from space. WIREs Clim Change 2014, 5,
389409.
Cooper, C., Mitchell, T., Forristall, G. & Stear, J. 2013. DeepStar Metocean Studies: 15 years of Discovery. Marine
Technology Society Journal, 47(3), 1926.
Corra, D. C., Oliveira, A. C., De. Tannuri, E. A. & Sphaier, S. H. Comprehensive Dowtime Analysis of DPAssisted Offloading Operation of Spread Moored Platforms in BraziWian waters. 32nd International Conference
on Ocean, Offshore and Arctic Engineering OMAE 2013, 914 June 2013 Nantes, France.
De Winter, R. C., Sterl, A. & B.G., R. 2013. Wind extremes in the North Sea Basin under climate change: An
ensemble study of 12 CMIP5 GCMs. Journal of Geophysical Research: Atmospheres 118 16011612.
Dee, D. P., Uppala, S. M., Simmons, A. J., Berrisford, P., Poli, P., Kobayashi, S., Andrae, U., Balmaseda, M. A.,
Balsamo, G., Bauer, P., Bechtold, P., Beljaars, A. C. M., Van De Berg, L., Bidlot, J., Bormann, N., Delsol, C.,
Dragani, R., Fuentes, M., Geer, A. J., Haimberger, L., Healy, S. B., Hersbach, H., Hlm, E. V., Isaksen, L.,
Kllberg, P., Khler, M., Matricardi, M., Mcnally, A. P., Monge-Sanz, B. M., Morcrette, J. J., Park, B. K.,
Peubey, C., De Rosnay, P., Tavolato, C., Thpaut, J. N. & Vitart, F. 2011. The ERA-Interim reanalysis:
configuration and performance of the data assimilation system. Quarterly Journal of the Royal Meteorological
Society, 137., 553597.
Degtyarev, A. B. New approach to wave weather scenarios modelling. Proceedings of the 8th International Ship
Stability Workshop, 67 October 2005 Istanbul, Turkey.
Didenkulova, I., Nikolkina, I. & Pelinovsky, E. 2013c. Rogue waves in the basin of intermediate depth and the
possibility of their formation due to the modulational instability. JETP Letters 97, 194198.
Didenkulova, I., Pelinovsky, E. N. & Didenkulov, O. I. 2014. Run-up of long solitary waves of different polarities
on a plane beach. Izvestiya, Atmospheric and Oceanic Physics 50, 532538.
Dietrich, J. C., Tanaka, S., Westerink, J. J., Dawson, C. N., Luettich Jr., R. A., Zijlema, M., Holthuijsen, L. H.,
Smith, J. M. & Westerink, J. J. 2012. Performance of the unstructed-mesh, SWAN+ADCIRS model in computing
hurricane waves and surge. J. Scientific Computing 52, 468497.
Dietrich, J. C., Zijlema, M., Allier, P.-E., Holthuijsen, L. H., Booij, N., Meixner, J. D., Proft, J. K., Dawson, C. N.,
Bender, C. J., Naimaster, A., Smith, J. M. & Westerink, J. J. 2013. Limiters for spectral propagation velocities in
SWAN. Ocean Modelling 70, 85102.
Dietrich, J. C., Zijlema, M., Westerink, J. J., Holthuijsen, L. H., Dawson, C. N., Luettich, R. a. J., Jensen, R. E.,
Smith, J. M., Stelling, G. S. & Stone, G. W. 2011. Modeling hurricane waves and storm surge using integrallycoupled, scalable computations. Coast. Eng., 58, 4565.
Dnv, L. 2011. Design of offshore wind turbine structures.
Dnv, L. 2014. Environmental Conditions and Environmental Loads. In: Dnv, L. (ed.). Hvik, Norway.
Dobrynin, M., Murawsky, J. & Yang, S. 2012. Evolution of the global wind wave climate in CMIP5 experiments.
Geophysical Research Letters 39.
Donelan, M. A., Curcic, M., Chen, S. S. & Magnusson, A. K. 2012. Modeling waves and wind stress. J. Geophys.
Res. Oceans, 117.
64
Dong, S., Fan, D.-Q. & S.-S., T. 2012. Joint occurrence period of wind speed and wave height based on both service
term and risk probability. Journal of Ocean University of China, 11(4), 488494.
Dong, S., Liu, W., Zhang, L. & Guedes Soares, C. 2013a. Return value estimation of significant wave heights with
maximum entropy distribution. Journal of Offshore Mechanics and Arctic Engineering 135.
Dong, S., Tao, S. S., Lei, S. H. & Guedes Soares, C. 2013b. Parameter estimation of the maximum entropy
distribution of significant wave height. Journal of Coastal Research 29(3), 597604.
Dong, S., Wang, L. & Fu, X.-Y. Design parameter estimation under multivariate extreme ocean environmental
conditions in the Bohai Sea. Proc. of International Conference on Ocean, Offshore and Arctic Engineering, 2007
San Diego, California, USA.
Dong, S., Wang, N.N., Liu, W. & Guedes Soares, C. 2013c. Bivariate maximum entropy distribution of significant
wave height and peak period. Ocean Engineering, 59(1), 8699.
Dutour-Sikiric, M.-A. 2013. Regional Ocean Modelling System and Wind Wave Model. Ocean Modelling, 72, 5973.
Dutton, A. & Lambeck, K. 2012. Ice volume and sea level during the last interglacial. Science, 337, 216219.
Ewans, K. C. 2014. Temporal characteristics of swell and a wavelet-based test for swell stationarity. Applied Ocean
Research, In press.
Ewans, K. C. & Jonathan, P. 2014. Evaluating environmental joint extremes for the offshore industry. J. Marine
Systems, 130, 124130
Extreme-Seas 2013. Design for Ship Safety in Extreme Seas (EXTREME SEAS) Europe: EC Grant Agreement
no.:234175, http://cordis.europa.eu/result/rcn/55382_en.html, (see also http://www. mar.ist.utl.pt/ extremeseas/),
Coordinated by Legacy DNV, Norway. Contact: Elzbieta.Bitner-Gregersen@dnvgl.com
Ezersky, A., Abcha, N. & E., P. 2013a. Physical simulation of resonant wave run-up on a beach. Nonlinear
Processes in Geophysics 20, 3540.
Ezersky, A., Tiguercha, D. & Pelinovsky, E. 2013b. Resonance phenomena at the long wave run-up on the coast.
Natural Hazards and Earth System Sciences 13, 27452752.
Fan, Y., Held, I. M., Lin, S. J. & Wang, X. L. 2013. Ocean warming effect on surface gravity wave climate change
for the end of the 21st century. J. Clim., 26, 60466066.
Fan, Y., Lin, S.-J., Held, I. M., Yu, Z. & Tolman, H. L. 2012. Global ocean surface wave simulation using a coupled
atmosphere-wave model. J. Climate 25, 62336252.
Fedele, F., Benetazzo, A., Gallego, G., Shih, P., Yezzi, A., Barbariol, F. & Ardhuin, F. 2013. Spacetime
measurements of oceanic sea states. Ocean Modelling, 70, 103115.
Feld, G., Randell, D., Wu, Y., Ewans, K.C. & Jonathan.P. Estimation of storm peak and intra-storm directionalseasonal design conditions in the North Sea. Proc. 33nd Conf. Oshore Mech. Arct. Eng., 714 June 2014 San
Francisco, USA.
Fernandez, L., Onorato, M., Monbaliu, J. & Toffoli, A., 2014. Modulational instability and wave amplification in
finite water depth. Nat. Hazards Earth Syst. Sci., 14, 705711.
Fettweis, X., Franco, B., Tedesco, M., Van Angelen, J. H., Lenaerts, J. T. M., Van Den Broeke, M. R. & Gallee, H.
2013. Estimating Greenland ice sheet surface mass balance contribution to future sea level rise using the regional
atmospheric model MAR. Cryosphere, 7, 469489.
Forristall, G. Z. 2000. Wave Crest Distributions: Observations and Second-Order Theory. J. Phys. Ocean., 30, 931
1943.
Forristall, G. Z. Laboratory measurements compared to theory. Proceedings of the OMAE 2015 Conference, 31
May5 June 2015 St. Johns, Newfoundland, Canada.
Forristall, G. Z., Ewans, K., Olagnon, M. & Prevosto, M. The WesT Africa Swell Project (WASP). Proceedings of
the OMAE 2013 Conference, 914 June 2013 Nantes, France.
Franco, B., Fettweis, X. & Erpicum, M. 2013. Future projections of the Greenland ice sheet energy balance driving
the surface melt. Cryosphere, 7, 118.
Gehrels, W. R., Balouin, Y. & Certain, R. 2012. Nineteenth and twentieth century sea-level changes in Tasmania
and New Zealand. Earth Planet. Sci. Lett., 315, 94102.
Gemmrich, J. & Garrett, C. 2012. The Signature of Inertial and Tidal Currents in Offshore Wave Records. Journal
of Physical Oceanography, 42(6), 10511056.
Gibson, R., Christou, M. & Feld, G. 2014. The statistics of wave height and crest elevation during the December
2012 storm in the North Sea. Ocean Dynamics, 64, 13051317.
Gillet-Chaulet, F. & Al., E. 2012. Greenland ice sheet contribution to sea-level rise from a new-generation ice-sheet
model. Cryosphere, 6, 15611576.
Gramstad, O., Zeng, H., Trulsen, K. & Pedersen, G. K. 2013. Freak waves in weakly nonlinear unidirectional wave
trains over a sloping bottom in shallow water. Physics of Fluid, 25, 122103.
Gramstad, O. & Babanin, A. Implementing nonlinear term in third generation models. Proceedings of the
OMAE2014 Conference, 914 June 2014 Nantes, France.
Grigorieva, V. & Gulev, S., 2006. Extreme wind waves worldwide from the VOS data and their changes over the last
50 years. ftp://ftp.wmo.int/Documents/PublicWeb/amp/mmop/documents/JCOMM-TR/J-TR-34-9th-wavesworkshop/Papers/Gulev.pdf
Guedes Soares, C. 1984. Representation of double-peaked sea wave spectra. Ocean Engineering, 11, 185207.
Gunn, K. & Stock-Williams, C. 2012. Quantifying the global wave power resource. Renewable Energy, 44(2012),
296304.
65
Gunn, K. & Stock-Williams, C. 2013. On validating numerical hydrodynamic models of complex tidal flow.
International Journal of Marine Energy, 34, e82e97.
Hacket, E. E., Fullerton, A. M., Merrill, C. F. & Fu, T. C. 2014. Comparison of incoherent and coherent wave field
measurements using dual-polarized pulse-Doppler X-band radar. IEEE Transactions on Geosciences and Remote
Sensing In press.
Hagen, ., Falkenberg, E. & Bitner-Gregersen, E. M. 2015. Reliability Based Approach for Offloading Operation
Related to Motion of Two Side-by-Side Moored LNG Carriers. Applied Ocean Research, In press.
Hagen, ., Garr, L. & Friis-Hansen, P. DNV-ADAPT framework for risk-based adaptation: a test case for the
offshore industry. Proc. 11th International conference on Structural Safety and Reliability (ICOSSAR), June
2013 2013 New York, USA.
Hagen, . & Solland, G. On Safety and reliability for platforms that are unmanned during severe storms.
Proceedings of the OMAE 2013 Conference, 914 June 2013 Nantes, France.
Hanson, J. L. & Phillips, O. M. 2001. Automated analysis of ocean surface directional wave spectra. Journal of
Atmospheric and Oceanic Technology 18, 277293.
Hasanat Zaman, M. & Baddour, E. 2011. Interaction of waves with non-colinear currents. Ocean Engineering,
38(4), 541549.
Hashemi, M. R. & Neill, S. P. 2014. The role of tides in shelf-scale simulations of the wave energy Resource,
Renewable Energy 69.
Haver, S. & Anderson, O. J. Freak waves: Rare realizations of a typical population or typical realizations of a rare
population. Proceedings of the 10th International Offshore and Polar Engineering (ISOPE) Conference, 28 May
2 June 2000 Seattle, USA.
Haver, S., Bruserud, K. & Baarholm Sagli, G. Environmental contour method: An approximate method for
obtaining characteristic response extremes for design purpose. Proceed. the 13th International Workshop on
Wave Hindcasting and Forecasting & 4th Coastal Hazard Symposium, 27 Oct.1 Nov. 2013 Banff, Canada.
Heffernan, J. E. & Tawn, J. A. 2004. A conditional approach for multivariate extreme values. . J. R. Statist. Soc., B 66.
Hemer, M. A., Fan, Y., Mori, N., Semedo, A. & Wang, X. L. 2013. Projected changes in wave climate from a multimodel ensemble. Nature Climate Change 3 471476.
Hemer, M. A., Wang, X. L., Weisse, R. & Swail, V. R. 2012. Advancing Wind-Waves Climate Science: The
COWCLIP Project. Bulletin of the American Meteorological Society 93, 791796.
Hennig, J., Scharnke, J., Swan, C., Hagen, ., Ewans, K., Tromans, P. & Forristall, G. Z. Effects of shortcrestedness on extreme wave impactA summarey of findings from the Joint Industry project ShortCresT.
Proceedings of the OMAE 2015 Conference, 31 May5 June 2015 St. Johns, Newfoundland, Canada.
Holland, R., Bruneau, N., Enright, C., Losch, M., Kurtz, N. T. & Kwok, R. 2014. Modeled trends in Antarctic sea
ice thickness. Journal of Climate, 27, 37843801.
Hu, H. H. & Ma, N. Numerical simulation on nonlinear evolution of rogue waves on currents based on the NLS
equation. 30th International Conference on Ocean, Offshore and Arctic Engineering OMAE 2011, June 2011
Rotterdam, The Netherlands.
Huseby, A. B., Vanem, E. & Natvig, B. 2013. A new approach to environmental contours for ocean engineering
applications based on direct Monte Carlo simulations. Ocean Engineering, 60.
Husson, R., Ardhuin, F., Collard, F., B., C. & Balanche, A. 2013. Revealing forerunners on Envisats wave mode
ASAR using the Global Seismic Network. Geophysical Research Letter, 39.
Hutchings, J. K., Roberts, A., Geiger, C. A. & Richter-Menge, J. 2011. Spatial and temporal characterization of seaice deformation. Ann. Glaciol., 52, 360368.
Hwang, P. A., Toporkov, J. V., Sletten, M. A. & Steven, P. M. 2013. Mapping Surface Currents and Waves with
Interferometric Synthetic Aperture Radar in Coastal Waters: Observations of Wave Breaking in Swell-Dominant
Conditions. Journal Of Physical Oceanography, 43, 563582.
Iacs 2000. Standard Wave Data. In: Iacs (ed.) Recommended Practice.
Iafrati, A., Babanin, A. & Onorato, M. 2013. Modulational Instability,Wave Breaking, and Formation of LargeScale Dipoles in the Atmosphere. Physical Review Letters, 110.
Ignatius, J., Rsnen, J.-E., Tervo, K. & Ellis, T. A Comprehensive Performance Management Solution. 13th
International Conference on Computer and IT Applications in the Maritime Industries (COMPIT14), 2014
Redworth, UK. 558568.
Ilus, T. & Heikkinen, A. Challenges in Vessel Speed Optimization. 11th International Conference on Computer and
IT Applications in the Maritime Industries (COMPIT12), 2012 Liege: Belgium., 284296.
Imo 1997. Interim Guidelines for the Application of Formal Safety Assessment (FSA) to the IMO Rule Making
Process. Maritime Safety Committee, 68th session June 1997; and Marine Environment Protection Committee,
40th session, September 1997.
Imo 2001. Guidelines for Formal Safety Assessment for the IMO Rule Making Process. IMO/Marine Safety
Committee 74/WP.19.
Imo 2012. 2012 Guidelines for the development of a ship energy efficiency management plan (SEEMP). Resolution
MEPC.213(63).
Imo 2013. 2013 Interim guidelines for determining minimum propulsion power to maintain the manoeuvrability in
adverse conditions. Resolution MEPC.232(65).
66
Ipcc 2007. The Fourth Assessment Report: Climate Change (AR4): The AR4 Synthesis Report, the Working Group
I Report: The Physical Science Basis (ISBN 978 0521 88009-1 Hardback; 978 0521 70596-7 Paperback), the
Working Group II Report Impacts: Adaptation and Vulnerability, the Working Group III Report: Mitigation of
Climate Change. AR4. UN.
Ipcc 2012. The IPCC SREX: Summary for Policymakers. Report. SREX. UN.
Ipcc 2013. The Fifth Assessment Report: Climate Change (AR5): The AR5 Synthesis Report. AR5. UN.
Iseki, T. & U.D., N. Study on a Method for Estimationing Fuel Consumption in a Seaway. 32nd International
Conference on Ocean, Offshore and Arctic Engineering (OMAE2013), 914 June 2013 Nantes: France.
Iso 2012. Petroleum and natural gas industries specific requirements for offshore structurespart 1: metocean
design and operating conditions. International Organization for Standardization.
Issc. ISSC 2009 Committee I.1 Environment Report. Proceedings ISSC 2009, August 2009 Seoul, Korea. 1126.
Issc. ISSC 2012 Committee I.1 Environment Report ((Bitner-Gregersen, E.M. (chairman), Bhattacharya S. K.,
Chatjigeorgiouc, I.K, Eames, I., Ellermann, K. Ewans, K. Hermanski, G., Johnson, M.C., Ma, N., Maisondieu,
Ch, Nilva, A., Rychlik, I., Waseda, T). Proceedings ISSC 2012, 913 September 2012 Rostok, Germany.
Issc. ISSC 2012 Committee I.1 Environment Reply to Discussers (Bitner-Gregersen, E.M. (chairman), Bhattacharya
S. K., Chatjigeorgiouc, I.K, Eames, I., Ellermann, K. Ewans, K. Hermanski, G., Johnson, M.C., Ma, N.,
Maisondieu, Ch, Nilva, A., Rychlik, I., Waseda, T). Proceedings ISSC 2012 913 September 2013 Rostok,
Germany.
Jeans, G., Harrington-Missin, L., Calverley, M., Maisondieu, C., Herry, C. & Quiniou, V. Deepwater Current Profile
Data Sources for Riser Engineering Offshore West Africa. Proceedings of the 32nd International Conference on
Ocean, Offshore and Arctic Engineering, 914 June 2013 Nantes, France
Jeans, G. & Wade, I. Simulation of strong intertial currents to support preliminary engineering offshore Namibia.
Proceedings of the OMAE 2013 Conference, June 914 2013 Nantes, France.
Jensen, R. E., Swail, V. & Hesser, T. J. What is the ground truth? Proceedings of the 13th International Workshop
on Wave Hindcasting and Forecasting and 4th Coastal Hazards Workshop, 27 October1 November 2013 Banff,
Canada.
Jevrejeva, S., Moore, J.C & Grinsted, A. 2012b. Potential for bias in 21st century semiempirical sea level
projections. J. Geophys. Res., 117.
Jevrejeva, S., Moore, J. C. & Grinsted, A. 2012a. Sea level projections to AD 2500 with a new generation of climate
change scenarios. Global Planet. Change, 8081, 1420.
Johnson, J. T., Burkholder, R. J., Toporkov, J. V., Lyzenga, D. R. & Plant, W. J. 2009. A numerical study of the
retrieval of sea surface height profiles from low grazing angle radar data. IEEE Trans. Geosci. Remote Sens., 47,
16411650.
Jonathan, P., Ewans, K. C. & Randell, D. 2013. Joint modelling of environmental parameters for extreme sea states
incorporating covariate effects. Coastal Eng., 79, 2231.
Jonathan, P., Randell, D., Wu, Y. & Ewans, K. 2014. Return level estimation from non-stationary spatial data
exhibiting multidimensional covariate effects. Ocean Eng., In press.
Jones, O., Ewans, K. & Chuah, S. A Monte Carlo approach for estimatiing extreme currents in the Singapore Straits.
Proceedings of the OMAE 2013 Conference, 914 June Nantes, France.
Kano, T. & Namie, S. A Study on Estimation of GHG Emission for Speed Planning Operation Using Energy
Efficiency Index and Time-Series Monitoring Data. 13th International Conference on Computer and IT
Applications in the Maritime Industries (COMPIT14), 2014 Redworth, UK. 167180.
Kariranta, R.-J. Utilization of performance data collection and normalization for ship operations and design. 11th
Liege, Belgium. 438448.
Keef, C., Papastathopoulos, I. & Tawn, J. A. 2013a. Estimation of the conditional distribution of a vector variable
given that one of its components is large: additional constraints for the Heffernan and Tawn model. J. Mult.
Anal., 115, 396404.
Keef, C., Tawn, J. A. & Lamb, R. 2013b. Estimating the probability of widespread flood events. Environmetrics, 24,
1321.
Khor, Y. S., Dhlie, K. A., Konovessis, D. & Xiao, Q. Optimum Speed Analysis for Large Containerships. 11th
Liege:, Belgium. 121131.
Knutti, R., Furrer, R., Tebaldi, C., Cermak, J. & Meehl., G. A. 2010. Challenges in Combining Projections from
Multiple climate Models. Journal of Climate 23 27392758.
Kodaira, T., Waseda, T., Nakagawa, T., Isoguchi., O. & Miyazawa, Y. 2013. Measuring the Kuroshio Current
Around Miyake Islands, a Potential Site for Ocean-Current Power Generation. Int. J. Offshore Polar Eng., 23(4),
272278.
Kpogo-Nuwoklo, K. A., Olagnon, M. & GuedE, Z. Wave spectra partitioning and identification of wind sea and
swell events. Proceedings of the OMAE 2014 Conference, 813 June 2014 San Francisco, California, USA.
Krogstad, H. E., Barstow, S. F., Mathiesen, L. P., Lnseth, L., Magnusson, A. K. & Donelan, M. A. Extreme waves
in the long-term wave Measurements at Ekofisk. Proc. Rogue Waves 2008 Workshop, 2333 October 2008
Brest, France.
67
L., L., A., C. & G., S. 2013. Wave energy resource assessment in the Mediterranean, the Italian perspective.
Renewable Energy 50, 938949.
Lane, E. M., Restrepo, J. M. & Mcwilliams, J. C. 2007. WaveCurrent Interaction: A Comparison of RadiationStress and Vortex-Force Representations. Journal of Physical Oceanography, 37(5), 11221141.
Leckler, F., Ardhuin, F., Filipot, J.-F. & Mironov, A. 2013. Dissipation source terms and whitecap statistics. Ocean
Modelling, 70, 6274.
Lee, J. Y., C.S., S., Kim, H., Hong, C. & Seo, J. Development and Application of Trim Optimization and parametric
Study using an Evaluation System (SoLuTion) based on the RANS for Improvement of EEOI. 33nd International
Conference on Ocean, Offshore and Arctic Engineering (OMAE 2014), 813, June 2014 San Francisco, USA.
Li, J.-G. 2012. Propagation of ocean surface waves on a spherical multiple-cell grid. Journal of Computational
Physics, 231(24), 82628277.
Li, J.-X., Liu, D.-Y. & Liu, S.-X. 2012. Numerical Investigation of the Effect of Current on Wave Focusing. China
Ocean Engineering, 26(1), 3748.
Li, L., Gao, Z. & Moan, T. Joint environmental data at five European offshore sites for design of combined wind
and wave energy devices. Proceedings of the ASME 32nd International Conference on Ocean, Offshore and
Arctic Engineering (OMAE 2013), 914June 2013 Nantes, France.
Lin, P. & Liu, P. 1999. Internal Wave-Maker for Navier-Stokes Equations Models. Journal of Waterway, Port,
Coastal, and Ocean Engineering, 125(4), 207215.
Lin, Y. H. & Fang, M.-C. The ship-routing optimization based on the three-dimensional modified isochrones
method. 32nd International Conference on Ocean, Offshore and Arctic Engineering (OMAE2013), 16 July
2013 Nantes, France.
Lindsay, R., Wensnahan, M., Schweiger, A. & Zhang, J. 2013. Evaluation of Seven Different Atmospheric
Reanalysis Products in the Arctic. Journal of Climate, 27, 25882606.
Little, C. M., Oppenheimer & Urban, N. M. 2013. Upper bounds on twenty-first-century Antarctic ice loss assessed
using a probabilistic framework. Nature Clim. Change, 7, 654659.
Liu, B., Guan, C., Xie, L. & Zhao, D. 2012. An investigation of the effects of wave state and sea spray on an
idealized typhoon using an air-sea coupled modeling system. Advances in Atmospheric Sciences 29(2), 391406.
Ma, Y., Dong, G. & Ma, X. Investigations of the Effetcs on Benjamin-Feir Instability by Uniform Currents. 33nd
International Conference on Ocean, Offshore and Arctic Engineering, 914 June 2013 Nantes, France.
Ma, Y., Dong, G., Perlin, M., Ma, X., Wang, G. & Xu, J. 2010. Laboratory observations of wave evolution,
modulation and blocking due to spatially varying opposing currents. Journal of Fluid Mechanics, 661, 108129.
Madsen, H. O., Krenk, S. & Lind, N. C. 1986. Methods of Structural Safety, Prentice-Hall, Enlewood Cliffs.
Magnusson, A. K. & Donelan, M. A. 2013. The Andrea wave. Characteristics of a measured North Sea rogue wave.
JOMAE.
Mahlstein, I. & Knutti, R. 2012. September Arctic sea ice predicted to disappear for 2oC global warming above
present. J. Geophys. Res. Atmos., 117.
Markus, D., Hojjat, M., Wchner, R. & Bletzinger, K. U. 2013. A CFD approach to modeling wave-current
interaction. International Journal of Offshore and Polar Engineering, 23(1), 2932.
Massel, S. 2013. Ocean Surface Waves: Their Physics and Prediction, World Scientific.
Massel, S. & Przyborska, A. 2013. On the surface waves generation due to glacier calving. Oceanologia, MS No
40/2012.
Massonnet, F., Mathiot, P., Fichefet, T., Goosse, H., Konig Beatty, C., Vancoppenolle, M. & Lavergne, T. 2013. A
model reconstruction of the Antarctic sea ice thickness and volume changes over 19802008 using data
assimilationc. Ocean Modelling 64, 6775.
Masui, T. & Al., E. 2011. An emission pathway for stabilizing at 6 Wm-2 radiative forcing. Climatic Change 109,
5976.
Masumoto, Y., Miyazawa, Y., Sumune, D., Kobayashi, T., Estournel, C., Marsaleix, P., Lanerolle, L., Mehra, A. &
Garraffo, Z. D. 2012. Oceanic dispersion simulation of Cesium 137 from Fukushima Daiichi Nuclear Power Plan.
Elements, 8, 207212.
Mcmillan, M., Shepherd, A., Sundal, A., Briggs, K., Muir, A., Ridout, A., Hogg, A. & Wingham, D. 2012.
Increased ice losses from Antarctica detected by CryoSat-2. Geophys. Res. Lett., 41, 38993905.
Mcwilliams, J. C., Restrepo, J. M. & Lane, E. M. 2004. An asymptotic theory for the interaction of waves and
currents in coastal waters. Journal of Fluid Mechanics, 511, 135178.
Meinshausen, M. & Al., E. 2011. The RCP greenhouse gas concentrations and their extensions from 1765 to 2300.
Climatic Change 109, 213241.
Mellor, G. 2011. Reply. Journal of Physical Oceanograph, 41(10), 20132015.
Mellor, G. L. The Depth-Dependent Current and Wave Interaction Equations: A Revision. Journal of Physical
Oceanography, 38(11), 25872596.
Mepc67/Inf.22 2014. Japanese activity on "Minimum propulsion power to maintain the manoeuvrability of ships in
adverse conditions. MEPC 67/INF.22. Submitted by Japan to IMO.
Merrill, C. F., Geiser, J., Pfitsch, D., Fu, T. C., Terrill, E., Depaolo, T., Cook, T., Lenain, L., Richer, S. & Fullerton,
A. M. Simultaneous ship motions and ocean waves measured in the time domain. 30th Symposium on Navaly
Hydrodynamics, 27 Nov. 2014 Hobart, Australia.
68
Meylan, M., Bennetts, L. G. & Kohout, A. L. 2014. In situ measurements and analysis of ocean waves in the
Antarctic marginal ice zone. Geophys. Res. Lett., 41, 50465051.
Moreira, R. M. & Peregrine, D. H. 2012. Nonlinear interactions between deep-water waves and currents. Journal of
Fluid Mechanics, 691, 125.
Mori, N., Shimura, S., Nakajo, T., Yasuda, T. & (2012), A. H. M. Multi-model ensemble projection of future coastal
climate change Coastal Engineering Proceedings, 2012.
Mori, N., Shimura, T., Nakajo, S., Ysujio, D., Yasuda, T., Mase, H. & Suh, K. D. Projection of future wave climate
change and application to coastal structure design. Proceedings of Breakwater Conference, 2013 Edinburgh, UK.
Moss, R. H. & Al., E. 2010. The next generation of scenarios for climate change research and assessment. Nature,
463, 747756.
Muyau, J., Ewans, K. C. & Jonathan, P. Short-term variability of wind measurements in South China Sea. OTC24904. Proceedings OTC-Asia, 2628 Marchx 2014 Kuala Lumpur, Malaysia (submitted for publication).
Myrhaug, D. & Holmedal, L. E. 2014. Wave-induced current for long-crested and short-crested random waves.
Ocean Engineering, 81(0), 105110.
Naess, A. & Karpa, O. Statistics of extreme wind speeds and wave heights by the bivariate acer method Proceedings
of the 32nd International Conference on Ocean, Offshore and Arctic Engineering OMAE 2013, 914 June 2013
Nantes, France.
Nielsen, U. D. & Stredulinsky, D. C. 2012. Sea State Estimation from an Advancing ShipA Comparative Study
using Sea Trial Data. Applied Ocean Research, 34, 3344.
Nieto-Borge, J. C., Hessner, K. & Mata-Moya, D. D. 2008. Signal-to-noise ratio analysis to estimate ocean wave
heights from X-band marine radar image time series. IET Radar. Sonar and Navigation, 2(1), 3541.
Nikolkina, I. & Didenkulova, I. 2012. Catalogue of rogue waves reported in media in 20062010. Nat. Hazards
Earth Syst. Sci., 61.
Nitsure, S. P., Londhe, S. N. & Khare, K. C. 2012. Wave forecasts using wind information and genetic
programming. Ocean Engineering 54 6169.
Norsok 2012. Action and action effects. In: Directorate, T. N. O. (ed.). Norway.
Notz, D. & Marotzke, J. 2012. Observations reveal external driver for Arctic sea-ice retreat. Geophys. Res. Lett., 39.
O. Lecomte, 1 T. Fichefet, M. Vancoppenolle, F. Domine, 4, F. Massonnet, 1p. Mathiot, S. Morin & 2013.Barriat, P.
Y. 2013. On the formulation of snow thermal conductivity in large-scale sea ice models. Journal of Advances in
Modelling Earth Systems, 5, 116.
Oconnor, M. L., Lewis, T. & Dalton, G. Weather Window Analysis of Irish and Portuguese Wave Data with
Relevance to Operations and Maintenance of Marine Renewables. 32nd International Conference on Ocean,
Offshore and Arctic Engineering (OMAE 2013), 9-14 June 2013 Nantes, France.
Oberhagemann J, L. J., & El Moctar O., 2012. Prediction of ship response statistics in severe sea conditions using
RANSE. Proc. of ASME 2012 31th Int. Conf. on Ocean, Offshore and Arctic Engineering, 1-6 July 2012 Rio de
Janeiro, Brazil.
Ochi, M. K. & Hubble, E. N. On six-parameters wave Spectra. Proceedings of 15th Coastal Engineering
Conference, 1976. 301328.
Oh, S.-H. & Jeong, W.-M. 2013. Characteristics of high waves observed at multiple stations along the east coast of
Korea. Nat. Hazards Earth Syst. Sci., 13, 35033514.
Olagnon, M. & Kerr, J. 2015. Anatomie curiese des vagues sclrates. Editions Qu (in French).
Olagnon, M., Prevosto, M., Van Iseghem, S., Ewans, K. & Forristall, G. Z. 2004. WASPWest Africa Swell Project
Final report and Appendices.
Olagnon, M., Ewans, K., Forristall, G.Z. & Prevosto, M. West Africa swell spectral shapes. Proceedings of the
OMAE 2013 Conference, 914 June 2013 Nantes, France.
Onorato, M., Osborne, A., Serio, M., Cavaleri, L., Brandini, C. & Stansberg, C. 2006a. Extreme waves,
modulational instability and second order theory: wave flume experiments on irregular waves. European Journal
of MechanicsB/Fluids, 25, 586601.
Onorato, M. & Proment, D. 2012. Approximate rogue wave solutions of the forced and damped nonlinear
Schrdinger equation for water waves. Physics Letters A, 376 (2012), 30573059.
Onorato, M., Residori, S., Bortolozzo, U., Montina, A. & Arecchi, F. T. 2013. Rogue waves and their generating
mechanisms in different physical contexts. Physics Reports, 528, 4789.
Onorato, M. A., Osborne, A. & Serio, M. 2006b. Modulation instability in crossing sea states: A possible
mechanism for the formation of freak waves. Phys. Rev. Lett., 96.
Osborne, A. 2010. Non-linear Ocean Waves and the Inverse Scattering Transform, Academic Press.
Osborne, A. 2013. Classification of Rogue Wave Solutions of the Nonlinear Schroedinger Equation. EGU General
Assembly Geophysical Research 2013, NH5.2. Vienna, Austria.
Overland, J. E. & Wang, M. 2013. When will the summer arctic be nearly sea ice free? Geophys. Res. Lett.
Pakozdi, C., Kendon, T. E. and Stansberg, C. T., 2012. A numerical study of a focused wave packet near the surf
zone. Proceedings of the OMAE 2012 Conference, 16 July 2012, Rio de Janeiro, Brazil.
Panigrahi, J. K., Umesh, P. A., Padhy, C. P. & Swain, J. Nearshore propagation of cyclonic waves. Nat Hazards, 60,
605622.
69
Papanikolaou, A., Zaraphonitis, G., Bitner-Gregersen, E., Shigunov, V., El Moctar, O., Guedes Soares, C., Reddy,
D. N. & Sprenger, F. Energy Efficient Safe Ship Operation (SHOPERA). Proceedings RINA Conference, 2425
Sept. 2014 London, UK.
Peng, B., Ma, N. & Gu, X. C. Study on Influence of Wave-current Interaction on Prediction of Rouge Wave Taking
Viscous Effect into Account. 32nd International Conference on Ocean, Offshore and Arctic Engineering, 914
June 2013 Nantes, France.
Pereira, H.P.P., Ribeiro, C.E.P, Carvalho, F.N., Moraes, L.F. & Campos, R.M. 2014. Improvement of Directional
Wave Information and Quality Control Before Real-Time Telemetry of Heave-Pitch-Roll Metocean Buoys.
Proceedings OceansSt. John's, 2014, 1419 Sept. 2014, IEEE, St. Johns. NL, DOI:
10.1109/OCEANS.2014.7003276
Petrova, P. G. & Guedes Soares, C. 2014. Natural Hazards and Earth System Sciences. Natural Hazards and Earth
System Sciences, 14, 12071222.
Pinton, P. 2012. Very short-term probabilistic forecasting of wind power with generalized logit-Normal
distributions. Journal of the Royal Statistical Society, 61(4), 555576.
Plagge, A. M., Vandemark, D. & Chapron, B. 2012. Examining the Impact of Surface Currents on Satellite
Scatterometer and Altimeter Ocean Winds. Journal of Atmospheric And Oceanic Technology, 29, 17761793.
Pleskachevsky, A. L., Lehner, S. & Rosenthal, W. 2012. Storm observations by remote sensing and influences of
gustiness on ocean waves and on generation of rogue waves. Ocean Dynamics 62, 13351351.
Ponce De Leon, S., Gomez, J., Sanchez-Arcilla, A. & Guedes Soares, C. Comparison of extreme storms in the North
Atlantic and Mediterranean. 3rd IAHR Europe Congress, 1416 April 2014 Porto, Portugal.
Ponce De Leon, S. & Guedes Soares, C. 2014. Extreme wave parameters under North Atlantic extratropical
hurricanes. Ocean Modelling, 81, 7888.
Portilla, J., Sosa, J. & Cavaleri, L. 2013. Wave energy resources: Wave climate and exploitation. Renewable Energy,
57, 594605.
Prevosto, M., Ewans, K., Forristall, G. Z. & Olagnon, M. . Swell genesis, modelling and measurements in West
Africa. Proc. 32nd Int. Conf. on Offshore Mech. and Arctic Eng., 914 June 2013 Nantes, France.
Ribeiro, E.O., Ruchiga, T.S. & Lima, J.A.M. 2013. A Brazilian Northeast Coast Wave Data Comparison: Radar vs
Buoy. Proceedings OMAE 2013, Nantes, France.
Quan Luna, B., Garr, L. & Friis Hansen, P. 2014. Risk based adaptation of infrastructures to floods and storm
surges induced by climate change. Proc. EGU 2014, abstracts and presentations. Vienna, Austria.
Queffeulou, P. Merged Altimeter Wave Height Database. An Update. Proc. ESA Living Planet Symposium 2013,
913 September (ESA SP-722, December 2013) 2013 Edinburgh, UK.
Quiniou-Ramus, V., Estival, R., Venzac, P. & Cohuet, J.-B. Resl-time of weather and ocean stations: public-private
partnership on in-situ measurements in the Gulf of Guinea. Proceedings of the OMAE 2013 Conference, 914
June 2013 Nantes, France.
Rascle, N. & Ardhuin, F. 2013. A global wave parameter database for geophysical applications. Part 2: Model
validation with improved source term parameterization. Ocean Modelling 70, 174188.
Restrepo, J. M., Ramrez, J. M., Mcwilliams, J. C. & Banner, M. 2011. Multiscale Momentum Flux and Diffusion
due to Whitecapping in Wave--Current Interactions. Journal of Physical Oceanography, 41(5), 837856.
Riahi, K. & Al., E. 2011. RCP 8.5A scenario of comparatively high greenhouse gas emissions. Climatic Change
109, 3357.
Ribeiro, E.O., Ruchiga, T.S. & Lima, J.A.M. 2013. A Brazilian Northeast Coast Wave Data Comparison: Radar vs
Buoy. Proceedings OMAE 2013, 914 June 2013, Nantes, France.
Robinson, T. O., Eames, I. & Simons, R. 2013. Dense gravity currents moving beneath progressive free-surface
water waves. Journal of Fluid Mechanics, 725, 588610.
Roland, A. & F., A. 2014. On the developments of spectral wave models: numeric and parameterization for the
coastal ocean. Ocean Dynamics, 64 833846
Romero, L., Melville, W. K. & Kleiss, J. M. 2012. Spectral energy dissipation due to surface wave breaking. J.
Phys. Oceanogr., 42, 14211444.
Rusu, L., Bernardino, M. & Guedes Soares, C. 2011. Modelling the influence of currents on wave propagation at the
entrance of the Tagus estuary. Ocean Engineering, 38(10), 11741183.
Rusu, L. & Guedes Soares, C. 2011. Modelling the wavecurrent interactions in an offshore basin using the SWAN
model. Ocean Engineering, 38(1), 6376.
Rybkin, A., Pelinovsky, E. N. & Didenkulova, I. 2014. Nonlinear wave run-up in bays of arbitrary cross-section:
generalization of the Carrier-Greenspan approach. J. Fluid Mechanics 748, 416432.
Rychlik, I. & Mao, W. 2014. Probabilistic model for wind speed variability encountered by a vessel. Natural
Resources, In press.
Rychlik, I. & Mustedanagic, A. 2013. A spatial-temporal model for wind speeds variability Department of
Mathematical Sciences Division of Mathematical Statistics, Chalmers, University of Technology, University of
Gothenbourg.
Salvacao, N., Bernardino, M. & Guedes Soares, C. Validation of a regional atmospheric model for assessing the
offshore wind resources along the Portuguese coast. Proceedings of the32nd International Conference on Ocean,
Offshore and Arctic Engineering OMAE 2013, 914 June 2013 Nantes: France.
70
Santala, M. J., Calvery, M., Taws, S., Grant, H., Watson, A. & Jeans, G. Squall Wind Elevation/Gust Factors and
Squall Wind Coherence. Proceedings Offshore Technology Conferencex, 58 May 2014 Houston, Texas, USA.
Santoro, A., Guedes Soares, C. & Arena, F. Analysis of experimental results on the space evolution of wave group
in crossing sea, . Proceedings of the OMAE 2013 Conference, June 914 2013 Nantes, France.
Sasa, K., Chen, C., Shiotani, S., Ohsawa, T. & Terada, D. Numerical Analysis of Failed Forecasts of Waves under
Low Pressures from Viepoint of Ship Operation. 33nd International Conference on Ocean, Offshore and Arctic
Engineering (OMAE 2014), 813 June 2014 San Francisco, USA.
Sasa, K., Terada, D., Shiotani, S., Wakabayashi, N. & Ohsawa, T. Current Situation and Difficulty of Wave Forecast
from Viwepoint of Ship Management. 32nd International Conference on Ocean, Offshore and Arctic
Engineering (OMAE 2013), 914 June 2013 Nantes, France.
Schwendeman, M., Thomson, J. & Gemmrich, J. R. 2014. Wave breaking dissipation in a young wind sea. J. Phys.
Oceanogr., 44, 104127.
Semendo, A., Weisse, R., Behrens, A., Sterl, A., Bengstsson, L. & Gunther, H. 2013. Projection of Global Wave
Climate Change toward the End of the Twenty-First Century. Jounal of Climate, 36.
Sergeeva, A. & Slunyaev, A. 2013. Rogue waves, rogue events and extreme wave kinematics in spatio-temporal
fields of simulated sea states. Nat. Hazards Earth Syst. Sci., 13, 17591771.
Shrira, V. I. & Slunyaev, A. V. 2014a. Nonlinear dynamics of trapped waves on jet currents and rogue waves. Phys.
Rev., E 89, 15.
Shrira, V. I. & Slunyaev, A. V. 2014b. Trapped waves on jet currents: asymptotic modal approach. J. Fluid Mech.,
738, 65104.
Siadatmousavi, S. M., Jose, F. & Stone, G. W. 2012. On the importance of high frequency tail in third generation
wave models. Coastal Engineering 60, 248260.
Sillmann, J., V., Kharin, V., Zhang, X., Zwiers, F. W. & Bronaugh, D. 2013. Climate extremes indices in the CMIP5
multimodel ensemble: Part 1. Model evaluation in the present climate. J. Geophys. Res. Atmos., 118, 17161733.
Slunyaev, A., Clauss, G. F., Klein, M. & Onorato, M. 2013b. Simulations and experiments of short intense envelope
solitons of surface water waves. Phys. Fluids., arXiv:1302.4531.
Slunyaev, A., Pelinovsky, E. & Guedes Soares, C. 2014. Reconstruction of extreme events through numerical
simulations. Journal of Offshore Mechanics and Arctic Engineering, 136.
Slunyaev, A., Pelinovsky, E., Sergeeva, A., Chabchoub, A., Hoffmann, N., Onorato, M. & Akhmediev, N. 2013d.
Super rogue waves in simulations based on weakly nonlinear and fully nonlinear hydrodynamic equations. Phys.
Rev., E 88.
Slunyaev, A. V., Sergeeva, A. V. & Pelinovsky, E. N. 2012. Modelling of deep-water rogue waves: different
frameworks. In: Carlos Guedes Soares, Y. G., N. Fonseca, A.P. Texeira (ed.) CENTEC Anniversary Book.
Marine Technology and Engineering. london, UK: Taylor & Francis Group, 199216.
Slunyaev, A. V. & Shrira, V. I. 2013. On the highest non-breaking wave in a group: fully nonlinear water wave
breathers vs weakly nonlinear theory. J. Fluid Mech., 735, 203248
Smith, J. A. 2006. WaveCurrent Interactions in Finite Depth. Journal of Physical Oceanography, 36(7), 1403
1419.
Smith, L. C. & Stephenson, S. R. New Trans-Arctic Shipping Routes Navigable by Midcentury. PNAS, March 4
2013. E1191-E1195.
Smith, T. A., Chen, S., Campbell, T., Martin, P., Erick Rogers, W., Gaberek, S., Wang, D., Carroll, S. & Allard, R.
2013. Ocean-wave coupled modeling in COAMPS-TC: A study of Hurricane Ivan (2004). Ocean Modelling 69,
181194.
Stansberg, C. T. 2012. Random wave groups and nonlinear extreme platform responses in model testing. SOBENA
Journal of Marine Systems and Ocean Technology, 7(2).
Stansberg, C. T., Berget, K., Graczyk, M., Muthanna, C. & Pakozdi, C., 2012. Breaking wave kinematics and
resulting slamming pressures on a vertical column. Proceedings of the OMAE 2012 Conference, 16 July 2012,
Rio de Janeiro, Brazil.
Stefanakos, C. N., Schinas, O. & Eidnes, G. Application of fuzzy time series techniques in wind and wave data
forecasting. Proceedings of 33nd International Conference on Ocean, Offshore and Arctic Engineering OMAE
2014, 813 June 2014 San Francisco, USA.
Stephenson, D. B., Collins, M., Rougier, J. C. & Chandler, R. E. 2012. Statistical problems in the probabilistic
prediction of climate change. Environmetrics 23, 364372.
Stephenson, S. R., Smith, L. C., Brigham, L. W. & Agnew, J. A. 2013. Projected 21st-century changes to Arctic
marine access. Climatic Change.
Stephenson, S. R., Smith, L. C. & J.A., A. 2011. Divergent long-term trajectories of human access to the Arctic. Nat
Clim Change, :156160, :156160.
Stroeve, J. C., Serreze, M.C., Holland, M. M., Kay, J. E., Meier, W. & Barrett, A. P. 2012. The Arctics rapidly
shrinking sea ice cover: A research synthesis. Clim. Change, 110, 10051027.
Sumata, H., Lavergne, T., Girard-Ardhuin, F., Kimura, N., Schdi, M. A., Kauker, F., Karcher, M. & Gerdes, R.
2014. An intercomparison of Arctic ice drift products to deduce uncertainty estimates. J. Geophys. Res. Oceans,
119, 48874921.
Swan, C. & Latheef, M. 2014. Nonlinear wave amplification and the effect of wavebreaking., ShorTCresT Joint
Industry Report (Work Package 1.2). London, UK.
71
Tao, S. S., Dong, S., Wang, N. N. & Guedes Soares, C. 2013a. Estimation storm surge intensity with Poisson
bivairate maximum entropy distribution based on copulas. Natural Hazards, 68.
Tao, S. S., Dong, S. & Xu, Y. H. Design parameter estimation of wave height and wind speed with bivariate
copulas. The Proceedings of 32nd International Conference on Offshore Mechanics and Polar Engineering, 914
June 2013b Nanes, France.
Teles, M. J., Pires-Silva, A. A. & Benoit, M. 2013. Numerical modelling of wave current interactions at a local
scale. Ocean Modelling, 68(0), 7287.
Thomson, A. M. & Al., E. 2011. RCP4.5: a pathway for stabilization of radiative forcing by 2100. Climatic Change
109, 7794.
Thomson, J., Polagye, B., Durgesh, V. & Richmond, M. C. 2012. Measurements of Turbulence at Two Tidal Energy
Sites in Puget Sound, WA, IEEE. Journal Of Oceanic Engineering, 37 (3).
Toffoli, A., Bitner-Gregersen, E. M., Osborne, A. R., Serio, M., Monbaliu, J. & Onorato, M. 2011a. Extreme Waves
in Random Crossing Seas: Laboratory experiments and numerical simulations. Geophys. Res. Lett, 38, L06605.
Toffoli, A., Bitner-Gregersen, E. M., Osborne, A. R., Serio, M., Monbaliu, J. & Onorato, M. 2011b. Extreme Waves
in Random Crossing Seas: Laboratory experiments and numerical simulations. Geophys. Res. Lett., 38.
Toffoli, A., BitnerGregersen, E. M. & Onorato, M. Statistics of wave orbital velocity in deep water random
directional wave fields. Proceedings of the OMAE 2012 Conference, 26 July 2012 Rio de Janeiro, Brazil.
Toffoli, A., Fernandez, L., Monbaliu, J., Benoit, M., Gagnaire-Renou, E., Lefevre, J.M., Cavaleri, L., Proment, D.,
Pakozdi, C., Stansberg, C.T., Waseda, T. & Onorato, M., 2013a. Experimental evidence of the modulation of a
plane wave to oblique perturbations and generation of rogue wavse in finite water depth. Physics of Fluids, 25,
091701.
Toffoli, A., Waseda, T., Houtani, H., Kinoshita, T., Collins, K., Proment, D. & Onorato, M. 2013b. Excitation of
rogue waves in a variable medium: An experimental study on the interaction of water waves and currents.
Physical Review E, 87(5).
Tolman, H. L. 2013. A Generalized Multiple Discrete Interaction Approximation for resonant four-wave interactions
in wind wave models. Ocean Modelling 70, 1124.
Tolman, H. L. & Group, T. W. I. D. 2014. User manual and system documentation of WAVEWATCH III version
4.18.
Tolman, H. L. & Grumbine, R. W. 2013. Holistic genetic optimization of a Generalized Multiple Discrete
Interaction Approximation for wind waves. Ocean Modelling 70, 2537.
Torsethaugen, K. A two-peak wave spectral model. Proceedings of the 12th International Conference on Offshore
Mechanics and Arctic Engineering (OMAE 1993), 2024 June 1993 Glasgow, UK.
Torsethaugen, K. 1996. Model for Double Peaked Wave Spectrum. SINTEF Civil and Environmental Engineering.
Trondheim, Norway.
Touboul, J. & Pelinovsky, E. 2014. Bottom pressure distribution under a solitonic wave reflecting on a vertical wall.
European Journal of Mechanics B/Fluids 48, 1318.
Towe, R., Eastoe, E., Tawn, J., Wu, Y. & Jonathan, P. The extremal dependence of storm severity, wind speed and
surface level pressure in the Northern North Sea. Proceedings of the 32nd International Conference on Ocean,
Offshore and Arctic Engineering OMAE 2013, 914 June 2013 Nanes, France.
Trombe, P.-J., Pinson, P., Bovith, T., Cutululis, N. A., Draxl, C., Giebel, G., Hahmann, A., Jensen, N. E., J., B.P.,
Le, N. F., Madsen, H., Pedersen, L. B., Sommer, A. & Vincent, C. L. 2013. Weather radarsThe new eyes for
offshore wind farms. Wind Energy, 17(11), 17671787.
Tsai, C.-C., Hou, T.-H., Popinet, S. & Chao, Y. Y. 2013. Prediction of waves generated by tropical cyclones with a
quadtree-adaptive model. Coastal Engineering 77, 108119.
Tsumune, D., Tsubono, T., Aoyama, M. & Hirose, K. 2012. Distribution of oceanic (137) Cs from the Fukushima
Dai-ichi Nuclear Power Plant simulated numerically by a regional ocean model. Journal of Environmental
Radioactivity, 111, 100108.
Umeyama, M. 2009. Changes in Turbulent Flow Structure under Combined Wave-Current Motions. Journal of
Waterway, Port, Coastal and Ocean Engineering, 135(5), 213227.
Umeyama, M. 2011. Coupled PIV and PTV Measurements of Particle Velocities and Trajectories for Surface Waves
Following a Steady Current. Journal of Waterway, Port, Coastal and Ocean Engineering, 137(2), 8594.
Van Vuuren, D. & Al., E. 2011b. RCP2.6: exploring the possibility to keep global mean temperature increase below
2C. Climatic Change, 109, 95116.
Van Vuuren, D. P. & Al., E. 2011a. The representative concentration pathways: an overview. Climatic Change, 109,
531.
Vanem, E. & Bitner-Gregersen, E. M. Alternative environmental contours for marine structural designa
comparison study. Proceedings of the OMAE 2014 Conference, 815 June 2014 San Francisco, USA.
Vanem, E., Bitner-Gregersen, E. M., Eide, L. I., Garr, L. & Friis Hansen, P. Uncertainties of Climate Modeling and
Effects on Wave Induced Bending Moment. Proceedings of the SNAME Conference, 2325 October 2014
Houston, USA.
Vicinanza, D., Contestabile, P. & Ferrante, V. 2013. Wave energy potential in the north-west of Sardinia (Italy).
Renewable Energy, 50.
Walker, R. T., Vannieuwkoop-Mccall, J., Johanning, L. & Parkinson, R. J. 2013. Calculating weatherwindows:
Applicationtotransit,installation and the implications on deployment success. Ocean Engineering, 68, 88101.
72
Wang, M. & Overand, J. E. 2012. A sea ice free summer Arctic within 30 years: An update from CMIP5 models.
Geophysical Research Letters 39.
Wang, Y. & Jiang, X. 2012. Improvement and application of a saturation based wave dissipation function in SWAN
model. Acta Oceanol. Sin., 31(1), 2432.
Wang, Y., Tao, A. F., Zheng, J. H., Doong, D. J., Fan, J. & Peng, J. 2014. Preliminary investigation on the coastal
rogue waves of Jiangsu. Nat. Hazards Earth Syst. Sci..
Waseda, T., Hallerstig, M., Ozaki, K. & Tomita, H. 2011. Enhanced freak wave occurrence with narrow directional
spectrum in the North Sea. Geophysical Research Letters, 38.
Waseda, T., Sinchi, M., Kiyomatsu, K., Nishida, K., Takahashi, S., Asaumi, S., Kawai, Y., Tamura, H. &
Miyazawa, Y. 2014. Deep water observations of extreme waves with moored and free GPS buoys. Ocean
Dynamics, 64, 12691280.
Waseda, T. K., In, K., Kiyomatsu1, K., Tamura, H., Miyazawa, Y. & Iyama1, K. 2013. Predicting freakish sea state
with an operational third generation wave model. Nat. Hazards Earth Syst. Sci. Discuss., 1, 62576289.
West, B. J., Brueckner, K. A., Jand, R. S., Milder, D. M. & Milton, R. L. 1987. A new method for surface
hydrodynamics. J. Geophys. Res., 92, 1180311824.
Wijesekera, H. W., Wang, D. W., Teague, W. J., Jarosz, E., Rogers, W. E., Fribance, D. B. & Moum, J. N. 2013.
Surface Wave Effects on High-Frequency Currents over a Shelf Edge Bank. Journal of Physical Oceanography,
43(8), 16271647.
Willis, J. & Bonnefond, P. 2013. Report of the Ocean Surface Topography Science Team Meeting. In: Boulder, C.
(ed.). USA.
Wing, D. A. & Johnson, M. C. Ship operability predicted from long term directional wave records. In: Architects, R.
I. O. N., ed. The William Froude ConferenceAdvances in Theoretical and Applied Hydrodynamics, Past and Future, November 2010 Portsmouth, UK. 2425.
Winkelmann, R., Levermann, A., Martin, M. A. & Frieler, K. 2012. Increased future ice discharge from Antarctica
owing to higher snowfall. Nature, 492, 239242.
Winterstein, S. R., Ude, T. C., Cornell, C. A., Bjerager, P. & Haver, S. Environmental parameters for extreme
response: Inverse FORM with omission factors. Proceedings of ICOSSAR93, 1993 Innsbruck, Austria.
Wmo 2001. Guide to Marine Meteorological Services. Geneva, Switzerland: World Meteorological Organization.
Wmo 2003. Manual on the Global Observing System. Geneva, Switzerland: World Meteorological Organization.
Xiao, W. Y., Liu, Y., Wu, G. & Yue, D. K. P. 2013. Rogue wave occurrence and dynamics by direct simulations of
nonlinear wave-field evolution. J. Fluid Mech., 720, 357392.
Yan, K., Zou, Z. & Zhou, Y. Competition of class I and II instabilities in evolution of crescent waves. Proceedings
of the OMAE 2013 Conference, 914 June 2013 Nantes, France.
Yates, D., Quan Luna, B., Rasmussen, R., Bratcher, D., Garre, L., Chen, F., Tewari, M. & Friis-Hansen, P. 2014.
Stormy Weather: Assessing Climate Change Hazards to Electric Power Infrastructure: A Sandy Case Study.
Power and Energy Magazine, IEEE.
Young, I. R., Babanin, A. V. & Zieger, S. 2013. The decay rate of ocean swell observed by altimeter. Journal of
Physical Oceanography, 43, 23222333.
Young, I. R., Rosenthal, W. & Ziemer, F. 1985. A three-dimensional analysis of marine radar images for the
determination of ocean wave directionality and surface currents. Journal of Geophysical Research, 90(C1),
10491059.
Zappa G., Shaffrey, L. C., Hodges, K. I., Sansom, P. G. and Stephenson, D. B. A Multimodel Assessment of
Future Projections of North Atlantic and European Extratropical Cyclones in the CMIP5 Climate Models. J.
Climate, 26 (2013):58465862. doi: http://dx.doi.org/10.1175/JCLI-D-12-00573.1
Zhang, H., Cherneva, Z., Guedes Soares, C. & Onorato, M. Comparison of Distributions of Wave heights from
Nonlinear Schrdninger Equations Simulations and Laboratory Experiments. Proceedings OMAE 2013
Conference, 914 June 2013a Nantes, France.
Zhang, H., Guedes Soares, C. & Onorato, M. 2014a. Modelling of the Spatial Evolution of Extreme Laboratory
Wave Heights with the Nonlinear Schrdinger and Dysthe Equations. Ocean Engineering, 89, 19.
Zhang, J. S., Zhang, Y., Jeng, D. S., Liu, P. L. F. & Zhang, C. 2014b. Numerical simulation of wavecurrent
interaction using a RANS solver. Ocean Engineering, 75(0), 157164.
Zhang, L. & Xu, D. 2005. A new maximum entropy probability function for the surface elevation of nonlinear sea
waves. China Ocean Eng., 19(4), 637646.
Zijlema, M., Van Vledder, G. P. & Holthuijsen, L. H. 2012. Bottom friction and wind drag for wave models.
Coastal Engineering 65, 1926.
Zou, Z. L., Hu, P. C., Fang, K. Z. & Liu, Z. B. 2013. Boussinesq-type equations for wavecurrent interaction. Wave
Motion, 50(4), 655675.

710 SEPTEMBER 2015
CASCAIS, PORTUGAL
VOLUME 1
COMMITTEE I.2
LOADS
COMMITTEE MANDATE
Concern for environmental and operational loads from waves, wind, current, ice, slamming, sloshing,
weight distribution and operational factors. Consideration shall be given to deterministic and statistical load
predictions based on model experiments, full-scale measurements and theoretical methods. Uncertainties in
load estimations shall be highlighted. The committee is encouraged to cooperate with the corresponding
ITTC committee.
COMMITTEE MEMBERS
Chairman:
P. Temarel, UK
W. Bai, Singapore
A. Bruns, Germany
Q. Derbanne, France
D. Dessi, Italy
S. Dhavalikar, India
N. Fonseca, Portugal
T. Fukasawa, Japan
X. Gu, China
A. Nestegard, Norway
A. Papanikolaou, Greece
J. Parunov, Croatia
K.H. Song, Korea
S. Wang, USA
KEYWORDS
Cables/risers, fatigue, green water, ice loads, multi-bodies, parametric roll, rogue waves, slamming,
sloshing, uncertainty analysis, vortex induced vibrations, wave loads.
ISSC committee I.2: LOADS
74
CONTENTS
1.
INTRODUCTION ........................................................................................................................ 75
2.
COMPUTATION OF WAVE-INDUCED LOADS ................................................................... 75

2.1 Zero speed case .................................................................................................................. 75
2.1.1 Body wave interactions ..................................................................................... 75
2.1.2 Body-wave-current interactions ........................................................................... 79
2.1.3 Multibody interactions.......................................................................................... 79
2.2 Forward speed case ............................................................................................................ 80
2.3 Hydroelasticity Methods.................................................................................................... 83
2.4 Loads from abnormal waves ............................................................................................. 85
3.
SHIP STRUCTURES SPECIALIST TOPICS ......................................................................... 87

3.1 Slamming and Whipping ................................................................................................... 87
3.2 Sloshing .............................................................................................................................. 91
3.2.1 Analytical methods ............................................................................................... 91
3.2.2 Experimental investigations ................................................................................. 92
3.2.3 Numerical simulation ........................................................................................... 93
3.2.4 Sloshing with internal suppressing structures...................................................... 94
3.2.5 Sloshing and ship motions.................................................................................... 95
3.3 Green water ........................................................................................................................ 96
3.4 Experimental and full scale measurements ....................................................................... 99
3.5 Loads due to damage following collision/grounding ..................................................... 101
3.6 Weather routing and operational guidance ..................................................................... 102
4.
OFFSHORE STRUCTURES SPECIALIST TOPICS .............................................................. 104

4.1 Vortex-induced vibrations (VIV) and Vortex-induced motions (VIM)......................... 104
4.1.1 VIV...................................................................................................................... 104
4.1.2 VIM ..................................................................................................................... 106
4.2 Mooring Systems ............................................................................................................. 108
4.3 Lifting operations ............................................................................................................. 111
4.4 Wave-in-deck loads ......................................................................................................... 113
4.5 Floating Offshore Wind Turbines ................................................................................... 113
5.
PROBABILISTIC MODELLING OF LOADS ON SHIPS ..................................................... 115

5.1 Probabilistic methods....................................................................................................... 115
5.2 Equivalent design waves ................................................................................................. 117
5.3 Design load cases and ultimate strength ......................................................................... 119
6.
FATIGUE LOADS FOR SHIPS................................................................................................ 120
7.
UNCERTAINTY ANALYSIS .................................................................................................. 123

7.1 Load uncertainties ............................................................................................................ 123
7.2 Uncertainties in loading conditions ................................................................................. 124
8.
CONCLUSIONS ........................................................................................................................ 125
REFERENCES ................................................................................................................................... 128
1.
75
INTRODUCTION
The content of this committees report is informed by its mandate, the expertise of its membership. The
subject areas undertaken by specialist task committees of ISSC 2015: Arctic Technology (V.6),
Accidental Limit States (V.1), Offshore Renewable Energy (V.4), Natural Gas Storage and
Transportation (V.2), Risers and Pipelines (V.8), do have an impact on the committees mandate, hence
content of this report. Although ice loads have been entirely omitted, the subject areas of sloshing and
vortex-induced vibrations have been reviewed. In addition, making the best use of the committees
expertise there are sections for loads on ships following damage and floating offshore wind turbines.
The structure of this report follows along similar lines to that adopted in previous ISSC reports.
Nevertheless, it should be noted that (i) the versatility of computational methods to model more than one
phenomenon, e.g. sloshing and slamming, and (ii) the focus of investigations in dealing with as complete
fluid-structure interaction modelling as possible, e.g. risers and moorings and vessels, results in many
overlaps between the sections and subsections of this report. Evaluation of wave-induced loads for ships
is in Sections 2 and 3, with the former dealing with fundamental computations methods, hydroelasticity
and abnormal waves, and the latter on slamming, sloshing, green water, measurements and damaged
ships. Evaluation of wave- and current-induced loads on offshore structures is in Sections 2 and 4, with
the former focusing on generic computational methods and the latter on vortex-induced motions and
vibrations, mooring systems and wave-in-deck loads. Specialist offshore topics of floating wind turbines
and lifting operations are also in Section 4. Probabilistic methods, design waves and fatigue loads for
ships are in Sections 5 and 6. Finally weather routing and uncertainties in loads and loading conditions
are in Sections 3.6 and 7, respectively, with conclusive remarks in Section 8.
2.
COMPUTATION OF WAVE-INDUCED LOADS
Thanks to dramatic advances of computer science and technology during recent years, the numerical
wave tank has attracted great interest from researchers. Computational Fluid Dynamics (CFD) making
use of the Reynolds Averaged Navier Stokes (RANS) equations, although computationally intensive,
features significantly in the investigations. Nonlinear potential flow solutions continue to be developed
and used. Most of the numerical investigations deal with two-dimensional (2D) problems due to
constraints from computational resources and/or CPU time consumption associated with threedimensional (3D) modelling. This is particularly so when dealing with fully nonlinear modelling and
irregular waves.
2.1
2.1.1
Zero speed case

Body wave interactions
The problem of wave-structure interactions has been of great interest in both offshore industry and
academic research in ocean engineering for several decades. Due to the complexity of interactions
between water waves and offshore or coastal structures, model tests are traditionally considered as the
most reliable method to predict wave impacts on structures. Nonetheless, the cost and scale effects of
model tests are considered as the major limitations in the initial study where large numbers of cases might
be required for testing. Thanks to the great efforts made by the scientific community in the past few
decades, numerical models are becoming increasingly accurate and reliable with high efficiency, and
numerical simulations now are usually employed in the initial stage during the design of marine structures
and provide validation results for model tests. More efficient and accurate numerical modes are under
development both in industry and academia.
The accuracy of linear and second order potential flow models, dominantly utilized in the industry,
remain acceptable in practical engineering, in most circumstances. However, nonlinear effects and fluid
viscous effects may be significant in some particular situations such as wave trapping within an array of
bodies, wave resonance in a gap between side by side vessels. To address the higher order nonlinear and
viscous effects, various numerical models are recently developed to achieve better predictions of wave
impacts.
The nonlinearity due to the free surface is believed to affect the hydrodynamic responses of, hence
loads on, structures, particularly if wave trapping is induced around the structures at certain frequencies.
To consider higher order (beyond second order) nonlinear effects, fully nonlinear potential flow (FNPF)
models are considered as one of the effective methods in simulating wave-body interactions. Ducrozet et
al. (2012a) presented a modified higher order spectral (HOS) nonlinear potential model with a controlled
wavemaker. In the HOS method, free surface velocity potential is represented by finite number of spectral
basis functions, individually satisfying Laplaces equation and periodicity. The key point in the modified
model is wave generation by an additional potential which satisfies the no-flux condition on wavemaker
76
and nonlinear free surface conditions. Validation cases of 2D irregular waves and 3D focused waves
illustrate high accuracy with comparisons against experimental data. Meanwhile, a comparative study was
presented by Ducrozet et al. (2012b) who compared the finite difference based FNPF solver
OceanWave3D (Engsig-Karup et al., 2009) and the HOS model. Simulating the typical problem of highly
nonlinear waves propagating on a finite bottom of constant depth, they found that the HOS model is more
efficient than OceanWave3D, and the efficiency difference depends on required accuracy level and wave
steepness. In order to improve the efficiency of this finite difference based model, Engsig-Karup et al.
(2012) implemented a massively parallel and scalable algorithm utilizing modern graphics processing
units (GPUs). The significant improvement of computational efficiency shows the promising utilizations
of GPUs in the development of numerical models in the near future. Subsequently, applying the idea of
splitting the total potential in the fluid domain into an incident and a scattered field, Ducrozet et al., 2014
described a nonlinear decomposition method for solving the wave-structure interaction problems by
extending the solver OceanWave3D. Accordingly the actual water surface is separated into two
components, i.e. an incident and a scattered wave field. The incident wave field (elevation, velocity,
potential, etc.) can be prescribed explicitly, resulting in a new boundary value problem for only the
scattered wave field. In particular, the scattered flow potential satisfies the Laplace equation in the fluid
domain and the boundary conditions are in terms of the scattered field properties (unknown) and incident
field properties (known). At each time step of simulation, only the unknown scattered wave field is
solved, while the actual wave field is obtained by superposing the incident and scattered wave field. They
showed that one, in principle, could choose any wave model as the incident part; the only limitation on
the incident wave being the capability of the chosen wave model. In their simulations, they only
employed a nonlinear regular wave model, i.e. the stream function model. However, it is possible to
prescribe irregular or focused 2D and 3D wave fields. They validated the model by simulating the case of
nonlinear reflection on a vertical wall, with comparisons against other numerical results and experiments.
They also presented a study of wave shoaling on an uneven beach. Linear components of wave surfaces
obtained from harmonic analysis agree well with linear theory results. In addition, the efficiency,
accuracy and convergence properties of the model were studied to demonstrate the significant gains of
application of the splitting technique.
Meanwhile, the boundary element method (BEM) remains one of the widely used approaches to solve
FNPF model. Guerber et al. (2012) presented a 2D model with a freely or forced moving submerged
horizontal cylinder and solved the boundary value problem by a higher order BEM. The model was
simulated to represent the wave energy converters (WECs), and model accuracy was checked by
verifying the conservation of both volume and energy during the time marching of simulation. They
compared their nonlinear results against linear results in the cases of submerged heaving cylinders. The
agreement with linear results is great when they used a very small amplitude cylinder motion, while
discrepancies increase as the motion amplitude increases. Zhou et al. (2013) presented a similar higher
order 3D BEM model where a surface piercing cylinder was forced to move in the fluid domain. By
varying the amplitudes of the moving cylinder, the nonlinear effects were found to be stronger in the
cases of cylinders with rotational motions than those with translational motions. Discrepancies between
fully nonlinear, linear and second order simulation results were observed as expected while using large
moving amplitudes. They compared their results with previously validated fully nonlinear model by Bai
and Eatock Taylor (2006) and experimental data by Chaplin et al. (1999), obtaining good agreement.
To deal with structures with complex geometries, Abbasnia and Ghiasi (2014) developed a 2D FNPF
model based on higher order BEM utilizing the non-uniform rational B-spline curve, which could
precisely capture structure boundaries especially where large curvatures appear. They studied wave
diffraction around single, dual and an array of cylinders and again the nonlinearities were clearly
observed from the time history of free surface elevations. The comparisons of wave forces against
previous numerical results and experiments were only carried out for the case of a single cylinder.
Focusing on the higher order effects on wave responses due to nonlinear free surface conditions, Bai et al.
(2014a) employed a 3D fully nonlinear numerical wave tank (NWT) based on FNPF theory to study wave
interactions with an array of cylinders. A mixed Eulerien-Lagrangian (MEL) approach was applied to
update water surface through free surface boundary conditions which are described in the Lagrangian
form. Fast Fourier Transformation (FFT) was used to perform harmonic analysis of their fully nonlinear
results. They numerically demonstrated the existence of both first and second order near-trapping
phenomena associated with the cylinder array of a typical configuration experimentally and analytically
studied. A third harmonic influence at the second order near-trapped mode was found to considerably
contribute to the fully nonlinear responses. Harmonic analysis shows that their first order results agree
well with both experimental and analytical results, and that in the case of second order trapped mode their
second order components agree better with experiments than the analytical results by Malenica et al.
(1999). The third harmonic components of their results were reported to be very large, which however
77
cannot be captured in the second order analysis by Malenica et al. (1999). You and Faltinsen (2012)
investigated the interaction between waves and submerged body using a 3D fully nonlinear BEM
(HOBEM). An additional application is on the interaction between the waves and a moored ship and the
strong nonlinearity of ship motions and hydrodynamic loads were discussed.
One of the challenges for solving FNPF model using BEM is the long computational time required. A
conventional BEM needs a CPU operation in order of O(N2), where N is the number of unknowns on the
boundaries of the computational domain. Targeted at improving the computational efficiency, Shao and
Faltinsen (2014) recently developed a new 3D FNPF model based on harmonic polynomial cells. The
computational domain is discretized by harmonic polynomials such that the velocity potential at each
field point is interpolated by a set of harmonic polynomials. Assembling all the discretized unknowns will
result in operating with a sparse matrix which can be solved by any efficient matrix solver. This recently
developed FNPF solver was demonstrated to be more efficient than an existing Quadratic Boundary
Element Method (QBEM) and a Fast Multipole Accelerated QBEM (FMA-QBEM) by studying the
boundary value problem on a unit cube which has an analytical solution. To validate the model, the
authors presented the free surface elevations in the study of 3D sloshing in a rectangular box, with
comparisons against finite element (FE) method and experimental results. Excellent agreement was
demonstrated. Capability of capturing the higher order effects was also demonstrated in a case of wave
diffraction around a bottom-mounted cylinder where experimental results are available. Higher harmonic
wave forces on the cylinder compare well with existing FNPF model while some discrepancies were
observed with experimental data, especially for the second harmonic forces.
While nonlinear effects in principle can be considered in all the FNPF solvers, viscous effects,
however, remain another challenge which cannot be accounted within the framework of potential flow
theory. With the viscous effects considered, one needs to solve the classical Navier-Stokes (N-S)
equation. Peng et al. (2013) extended a NWT model based on a continuous direct-forcing immersed
boundary (IB) method with the combination of volume of fluid (VOF) method. They applied the
developed model to investigate the interactions between water waves and inclined-moored submerged
breakwaters. Viscous process in the flow field, such as flow separation and vortex generation, can be
captured and reproduced. Comparisons of wave elevations, mooring line forces and displacements of the
moored body with tests revealed a favorable agreement. A similar combined IB and VOF model based on
finite difference method was presented by Zhang et al. (2014) who performed various validation studies
including oscillating cylinder in fluid without a free surface, liquid sloshing in a tank, water exit and entry
of a horizontal cylinder, and a solitary wave over a submerged rectangular obstacle. Excellent agreement
with existing analytical, numerical and experimental results was demonstrated for all. With this model,
they also investigated a submerged and a semi-submerged ellipse rotating in a tank respectively, and
violent water splashing, fluid vortex and flow jets were well captured, although there are no comparisons
against numerical and experimental data. However, the numerical models of Peng et al. (2013) and Zhang
et al. (2014) are both in two dimensions. In modelling 3D wave-body interactions, Wang and Guedes
Soares (2014) presented a FE model to solve 3D N-S equations based on the Arbitrary-LagrangianEulerian (ALE) description, where the fluid is solved by using an Eulerian formulation while the structure
is discretized by a Lagrangian approach. The interface between solid bodies and fluids is also captured
utilizing the VOF method. They investigated the water impact on a buoy which was a representation of a
WEC and maximum pressures were shown to occur at the moment the buoy touched the water surface.
The authors investigated particularly the influence of fluid domain mesh density showing that it could
generate significant influence on the results. For example three different mesh densities were used for the
case of a hemisphere impacting on calm water surface, showing that the highest mesh density predictions
agree better with experimental measurements. This indicates that a high mesh density is required to
achieve a certain accuracy, leading to high computational efforts in the case of 3D modelling.
The open source CFD solver OpenFOAM was recently demonstrated by Higuera et al. (2013a),
(2013b), (2014)) to well capture the local flow characteristics around coastal structures. Specified
boundary conditions for realistic wave generation and active wave absorption were implemented to
demonstrate the capability of modelling a various type of nonlinear waves. Both 2D and 3D studies were
carried out, and with comparisons against experiments, the overall performance in terms of accuracy and
computational time is encouraging. A similar OpenFOAM model with further extension on wave
generation was presented by Chen et al. (2014), who illustrated the higher harmonic effects of wave runup and loading on bodies as well as the higher harmonics of focused wave groups. To generate wave
groups in the wave tank, they did not use a piston-like wave maker; instead the wave was generated via
the flux into the computational domain through a vertical wall. The volume fraction and horizontal and
vertical velocities at the wave inlet boundary faces were specified. They simulated wave impacts on a
semi-submerged horizontal circular cylinder and a vertical circular cylinder. Wave elevations and wave
forces were compared with tests where the agreement is good even for harmonic components up to 4th
78
order. For a semi-submersible platform with shallow draft, it is important to estimate wave run-ups
accurately in order to fulfil requirements of the positive air gap below the deck. Nam et al. (2013)
presented a simulation method which solves the two-phase N-S equations using finite volume
discretization with a VOF method. Under the shallow draft conditions, highly nonlinear wave run-ups
were observed at the front side of the first column. Wave amplifications due to nearly trapped mode
between two columns lead to high wave run-ups at the back side of the first column. Vertical suction
forces dramatically increase with the draft. Present calculations under both draft conditions show fairly
good agreement.
Wellens (2012) specifically developed a model named ComFLOW based on the N-S equations to
simulate extreme waves. A finite volume discretization of Cartesian grids was adopted in this model and
the free surface in ComFLOW was advected using an improved VOF algorithm. The extreme wave
impacts on the offshore structures were investigated in detail and an attempt was taken to efficiently
reduce the spurious reflection from boundaries as the domain size was limited due to CPU time cost.
stman et al. (2014) presented numerical simulations of green water events and wave impact on a FPSO.
The simulations are performed at model scale and the results are compared against experimental model
test results and show reasonable agreement. The commercial Star-CCM+ CFD software is used in the
simulations. The studied phenomena involve strong nonlinear effects with large variation in peak values
of the studied quantities. The variations were, however, found to be more pronounced in the CFD
simulations than in the experiments. Ideally a large amount of green water impacts should be computed in
order to be able to make a statistical comparison against model test results. Due to very long simulation
time it is concluded that use of RANS CFD tools for design optimizations of vessels exposed to green
water phenomena is, at present, not mature enough. However, it can be used to give valuable physical
insight in terms of visualization and understanding of complex nonlinear green water events.
Other than grid-based models, the gridless Smoothed Particle Hydrodynamics (SPH) method can also
handle complex local viscous effects. However the difficulty is the treatment of fluid-structure interface.
Bouscasse et al. (2013) presented a weakly compressible SPH solver which imposed a no-slip boundary
condition on the solid surface via a ghost-fluid technique. Specifically the solid domain boundary is
modelled by a set of imaginary particles, referred to as ghost particles, and the fluid fields (velocity,
pressure and internal energy) are extended on these ghost points through mirroring such that the global
loads on solid boundaries are evaluated through the momentum exchange between fluid and ghost
particles. They modeled a dolphin dropped into a water tank and an oscillating box in a water tank.
Comparisons were made with experiments in the case of a wave group interacting with a freely floating
box. The surface elevations, box motions in sway, heave and roll are close to the experimental results.
This technique was demonstrated to be capable of modelling complex interfaces between fluids and
bodies; yet their model is in two dimensions.
It is well known that a 3D CFD solver requires a large amount of computational effort for large scale
modelling such as wave interaction with marine structures. Although more details can be captured than a
potential flow solver, a full scale 3D CFD solver may not be a preferred choice considering CPU time.
Consequently, the combination of a FNPF solver and a CFD solver could be a promising strategy in order
both to achieve high efficiency and to capture detailed fluid viscous effects near the structures (Sriram et
al., 2014). Kristiansen and Faltinsen (2012) presented a domain decomposition model based on finite
volume method (FVM) which solves the outer domain by a potential flow solver and the inner domain by
an in-house CFD solver. Their 2D model was developed mainly to investigate the flow separation in way
of the barge bilge in the problem of gap resonance between two side-by-side barges and the free water
surface was modelled by the linearized potential solver which revealed to significantly improve
computational efficiency. The accuracy of the model was demonstrated by comparisons of gap surface
elevation against tests. Meanwhile, Sriram et al. (2014) presented a more complex hybrid model which
combines not a linear potential flow model but a FNPF solver and an N-S solver. Particularly their FNPF
solver is based on FE method and the N-S solver is based on the Improved Meshless Local Petrov
Galerkin method with Rankine source solution (IMLPG_R), which was previously presented by Sriram
and Ma (2012). The concept of the combination is based on the fact that a FNPF solver is more efficient
in modelling wave propagation while a CFD solver is more accurate in capturing local fluid viscous
effects. Comparisons with experiments demonstrated a very accurate and detailed capture of wave
breaking while improving the computational efficiency of the IMLPG_R based N-S solver. Again, their
model is in two dimensions. Up-to-date, Paulsen et al. (2014) presented a 3D FNPF solver combined with
an N-S/VOF solver where the FNPF solver is based on their existing code OceanWave3D and the NS/VOF solver is OpenFOAM. The computational domain was decomposed into an outer domain
governed by the FNPF solver and an inner domain described using the N-S/VOF solver. Numerical
computations of wave loads on surface piercing circular cylinders were carried out for the situations of
weakly nonlinear regular waves, phase-focused irregular waves, unidirectional irregular waves on a
79
sloping bed and multidirectional irregular waves on a sloping bed. Comparisons against experimental
tests of wave elevations and inline forces in all the cases demonstrate good agreements. The efficiency
achieved utilizing Message Passing Interface based parallel computation illustrates the potential of further
development and application of the combined model in three dimensions.
In general, potential flow model remains one of the dominated efficient solvers for numerical
simulation of wave-body interactions. Further development of 3D FNPF solves is still valuable.
Nonetheless, it could be promising to develop hybrid 3D models combining a potential flow and a CFD
solver.
2.1.2
Body-wave-current interactions
The effect of body-wave-current interactions for permanent, fixed or floating structures is primarily an
issue for predictions of air-gap and wave drift force calculations. By change of coordinate systems, the
hydrodynamic problem is equivalent to the forward speed seakeeping problem for a ship, with added
resistance and wave drift force being the equivalent force contributions. Within the last reviewing period
there has been a renewed awareness of the important effect of current on the wave drift force due to
several incidences of mooring failures where overload is considered to be the main cause (Kvitrud, 2014).
The fact that drift forces can be considerably enhanced in a combined wave-current environment has been
known since the mid-1980s. However, the mooring design industry does not seem to be sufficiently aware
of this. A reason for this is that available mooring design tools do not properly account for this effect. The
major effect of current for linear wave-body interactions is the relative motion between wave and
structure which is relevant for air-gap predictions, in particular the wave run-up in front of vertical
surfaces.
Monroy et al. (2012) studied the interaction effects of the steady flow past the floater on the wave
diffraction-radiation potential. The steady flow is taken as the double body flow. The time-harmonic
velocity potential is decomposed into a linear part and an interaction part taking into account the
interaction of the linear potential with the local steady flow at the free surface. The main difference in this
approach and earlier developments, e.g. Grue and Biberg (1993), is the use of a non-secular Greens
function. Results include the effect of current on first order surface elevation on fixed cylinder and drift
force on freely floating tanker, comparing with the simplified method by Aranha (1994) which has been
adopted by the industry. Stansberg et al. (2013) presented a discussion on the significance of the wavecurrent-structure interaction on wave drift forces, slow-drift motions and mooring line forces for offshore
structures. Observations from experiments on an FPSO and a semi-submersible show the important effect
of current on the surge motion and mooring line tensions.
The second order diffraction and radiation problem for a floating body with small forward speed has
been investigated by Shao and Faltinsen (2013). The main contribution of this work is a consistent and
robust method to handle the numerical difficulties associated with higher order derivatives in both the free
surface and body boundary conditions. This problem is due to the transfer of the body boundary condition
from the mean body position to the instantaneous position, resulting in the, so called, mj terms which do
not exist and cannot be evaluated at sharp corners. The new method takes advantage of formulating the
boundary values problem in a body-fixed coordinate system in the near field. Body motions then appear
in the free surface condition. However, when using the body-fixed system no higher order derivatives
appear. A time domain HOBEM based on cubic shape functions was used to solve the wave-current
interactions for the case of a bottom mounted and a vertically truncated cylinder. Shao and Helmers
(2014) further developed and verified this method by comparing with previous analytical studies.
2.1.3
Multibody interactions
Within the context of multibody interactions research focuses on the gap resonance effect and multibody
offshore systems. With reference to the former the 2D method by Kristiansen and Faltinsen (2012),
combining inner domain CFD and outer domain potential flow solutions, has already been discussed in
Section 2.1.1. It has been shown by Hong et al. (2013a), using linear potential theory, that the gap
resonance can appropriately be reduced by introducing an artificial damping (a wetted surface damping),
in the body boundary condition of two ships floating side-by-side in waves. Their numerical results have
been obtained using 9-node discontinuous HOBEM. They have shown that using an appropriate value of
the wetted surface damping, the unrealistically large first order, as well as time-mean second order,
numerical values due to the Helmholtz and higher mode resonances in the gap can be reduced without
altering numerical results at non-resonant frequencies. The appropriate wetted surface damping can be
obtained by comparing numerical and experimental results of wave elevation at the first gap resonance
frequency. Comparison of the numerical and experimental values of the sway drift forces on a FPSO and
a shuttle tanker suggests that the wetted surface damping parameter can be successfully used to predict
80
the first order hydrodynamic coefficients as well as the time-mean drift forces of two bodies floating sideby-side in close proximity.
Bunnik (2014) presented a time domain potential flow method which simulates the behavior of
multibody offshore systems with large horizontal relative motions in waves. When the relative motions
exceed a certain threshold value a new linear diffraction calculation is carried out. The time domain
simulation is then continued with a new set of hydrodynamic coefficients (added mass, damping, 1st and
2nd order wave forces). As a test case a berthing operation was simulated in which an LNG carrier was
pushed sideward towards a LNG FPSO in waves of various headings. These simulations showed that,
depending on the wave direction, the effect of the change in hydrodynamic interaction can be significant,
resulting in a considerable change in the time needed to complete the berth. Although the results so far
look logical and realistic, convergence of the results with respect to the amount of panels used and the
threshold value for updating the coefficients need to be studied.
2.2
Forward speed case
The problem of wave-induced loads on ships with forward speed is one of the most demanding in ship
hydrodynamics, especially when considering the excitation by moderate or large amplitude waves. In
theoretical/numerical approaches, this requires the proper consideration of the forward speed effects on
ship motions and loads and of a variety of nonlinearities related to the large amplitude ship motions and
to ship's actual wetted surface, as well to the change of the free surface of the incoming waves and their
interaction with the moving ship. Due to the complications of the above set problem, simplifications and
engineering solutions are often adopted. This implies that the exact nonlinear seakeeping and large
amplitude wave-induced loads problems may be still considered unsolved. A brief review of related
theoretical and numerical methods is outlined in this section.
Greco and Lugni (2012) developed a 3D seakeeping numerical solver to handle occurrence and effects
of water-on-deck and bottom slamming. It couples (A) the rigid ship motions with (B) the water flowing
along the deck and (C) bottom slamming events. Problem A is studied with a 3D weakly nonlinear
potential flow solver based on the weak-scatterer hypothesis. Problem B, and so local and global induced
green water loads, are investigated by assuming shallow water conditions onto the deck. Problem C is
examined through a Wagner type wedge impact analysis. For coupling between A and B, the external
seakeeping problem furnishes the initial and boundary conditions to the in-deck solver in terms of water
level and velocity along the deck profile; in return, the shallow water problem makes available to the
seakeeping solver the green water loads to be introduced as additional loads into the rigid motion
equations. For the coupling between A and C, the instantaneous ship configuration and its kinematic and
dynamic conditions with respect to the incident waves fix the parameters for the local impact problem; in
return, the slamming and water entry pressures are integrated on the vessel region of interest and
introduced as additional loads into the rigid motion equations. The developed solver has been applied to
the problems of a dam breaking inside a closed tank and to the wave-ship interaction problem
with/without water-on-deck occurrence for validation. Obtained results are compared with experimental
data. Subsequently, Greco et al. (2012) carried out experimental and numerical investigations on a patrol
boat at rest or travelling in head regular and irregular waves. In these studies motion RAOs, relative
motions and occurrence, features and loads of water-on-deck, bottom slamming and flare slamming
events, as well as added resistance in waves, were investigated. The analysis is systematic covering a
range of Froude numbers, wave length () to ship length (LPP) ratios and wave steepness values. The main
parameters that affect the global and local quantities are identified and possible issues in terms of, for
example, water-on-deck severity and structural consequences are determined. Different slamming
behaviors were identified, depending on the spatial location of the impact on the vessel, namely singlepeak, church-roof and double-peak behaviors. A bottom slamming criterion is assessed. The major
discrepancies with the experiments are discussed, and the importance of viscous hull damping and flare
impact for the most violent conditions is emphasized. Inclusion of these effects has improved the
numerical solution.
Hanninen et al. (2012) studied an interface capturing VOF solution for a passenger ship advancing in
steep (kA = 0.24, k: wave number, A: wave amplitude) and short waves (/L PP = 0.16), with the focus on
estimating quantitative uncertainties for the longitudinal distributions of the first-third harmonic wave
loads in the bow region. The computations were performed with the commercial flow solver ISIS-CFD.
The uncertainty distributions obtained reveal that even the uncertainty of the first harmonic wave load
varies significantly along the bow region. It is shown that the largest local uncertainties of the first
harmonic wave load relate to the differences in the local details of the propagating and deforming
encountered waves along the hull. The authors also discussed the challenges that were encountered in the
quantification of the uncertainties for this complex flow case. Hanninen et al. (2014) considered the
capability of their interface capturing method to predict local ship wave loads in short and steep waves
81
(kA = 0.24, /LPP = 0.16) by comparing with experimental results. The computations were performed
with an unstructured RANS solver that models free surface flows with a VOF method. It was shown
qualitatively that the solution behavior of the computed results is reasonable, even though the results can
depend on the location of the computational points within the surface area of a pressure sensor. The
agreement between computed and measured results is good at all the 10 locations on the bow of the ship.
The characteristics of the wave loads vary between the sensor locations. Impact-type as well as smoother
behavior of the loading is captured well by the numerical method. The work by Kim et al. (2012a) on
estimating the long term midships bending moments and Oberhagemann et al. (2012) on embedding time
domain field methods in extreme value predictions are discussed in Section 5.2.
Greco et al. (2013) developed a 3D domain decomposition strategy to deal with violent wave-ship
interactions involving water-on-deck and slamming occurrence. This couples a linear potential
seakeeping solver, in the outer domain, and Lagrangian markers for the body motion, in the inner domain
where slamming, water-on-deck and free surface fragmentation may occur, involving important flow
nonlinearities. The field solver combines an approximated projection method with a level set technique
for the free surface evolution. Main features of the weak and strong coupling algorithms are described
with special focus on the boundary conditions for the inner solver. Two ways of estimating the nonlinear
loads by the N-S method are investigated, on the basis of an extrapolation technique and an interpolation
marching cubes algorithm, respectively. The domain decomposition is applied for the case of a patrol boat
travelling in head regular waves and compared against water-on-deck experiments in terms of flow
evolution, body motions and pressure on the hull. The solver was successfully verified by comparison
against the linear potential flow solution in the case of incident waves with small steepness and validated
against model tests in the case of steeper waves.
Rajendran et al. ((2013), (2014)) presented a time domain numerical method based on strip theory.
They calculated the probability distributions of relative motions and bending moments of a cruise ship in
a set of extreme seas. Their approach includes two levels of complexity. The simpler one combines linear
radiation and diffraction forces with nonlinear Froude-Krylov forces, hydrostatic forces and shipping of
green water at the bow. Cummins Impulse Response Function (IRF) formulation is used to represent the
radiation forces. The second approach is a generalization of the first and the effects of body nonlinearity
are considered by a simplified method, namely the memory functions, infinite frequency added masses
and the radiation restoring coefficients are evaluated at each time instant as functions of the instantaneous
wetted surface. A similar procedure is used to calculate the diffraction forces. In their latter model, the
first order Froude Krylov pressures are replaced with a second order model. The 2nd order Froude-Krylov
pressures are integrated up to the exact wetted surface area for each time instant. The nonlinear radiation
and diffraction effects on the responses are analyzed by comparing the fully nonlinear results with the
numerical predictions assuming linear radiation and diffraction forces. The short term nonlinear responses
are represented by empirical probability distributions, obtained from time domain simulations, and the
quality of the predictions is assessed by comparing with model test data.
Seng and Juncher Jensen (2013) developed a new approach, which requires reduced computational
effort for the estimation of the short term statistical properties of the hull girder responses, as predicted by
a free surface CFD solver. The approach, known as the MCF (model corrector factor), is an efficient
alternative to the polynomial based response surface approach to the structural reliability analysis. The
concept is to apply a predictor (e.g. a strip theory) to determine the most probable response conditioned
wave sequence and the associated statistical properties. Thereafter, the predictor is applied to improve the
evaluation of the responses for only a few selected wave sequences. An algorithm is proposed to support
the selection process. To illustrate the process of transferring statistical properties to the CFD results (i.e.
the corrector), an application to a 9400 TEU post Panamax container vessel is given using a nonlinear
time domain strip theory as the predictor. The discussion focuses on the usage and the implicit
requirements of the MCF approach, especially when slamming induced responses are considered.
Finally, a series of research investigations referring to the seakeeping problem are also reviewed.
Though without direct reference to loads, these methods contain important methodology developments or
applications, and will allow the calculation of wave loads with some additional development work. Sun
and Faltinsen (2012) studied the steady and unsteady hydrodynamic problems of a semi-displacement
ship with round bilge at high forward speed with a numerical method based on 2D + t theory. The ship
was forced to oscillate in heave in the unsteady problem. No incident waves were present. The nonviscous flow separation from the round bilge of the ship hull was simulated. The pressure on the hull
surface was evaluated and the sectional hydrodynamic vertical forces were obtained. Good agreement
was achieved between the present calculations and the experiments, although some discrepancies near the
bow and stern were observed. The interaction between the local steady flow and unsteady flow are
automatically included and the nonlinearities in both steady and unsteady flow were considered. Shao and
Faltinsen (2012) presented an alternative formulation of the boundary value problem for linear seakeeping
82
and added resistance analysis based on a body-fixed coordinate system which does not involve the, so
called, mj terms in the traditional formulation when an inertial coordinate system is applied. Numerical
studies were carried out for a modified Wigley I hull, a Series 60 ship and the S175 container ship for
moderate forward speeds where it is thought appropriate to use the double body flow as basis flow.
Results for the forced heave and pitch oscillations, motion responses, and added resistance in head waves
show good agreement with experiments and other numerical studies.
He and Kashiwagi (2014) developed a time domain HOBEM for the 3D forward speed radiation and
diffraction problems. Extensive results, including the exciting forces, added mass and damping
coefcients, wave proles and wave patterns for blunt and slender Wigley hulls with forward speed, are
presented to validate the efciency of the proposed 3D time domain approach. Model tests of the
radiation and diffraction problems in a towing tank were also carried out. Computed numerical results
show good agreement with the corresponding experimental data and other numerical solutions.
Liu et al. (2014b) presented a 3D nonlinear time domain method for the prediction of ship responses in
waves based on the IRF concept. In this method, the wave excitation is decomposed into Froude-Krylov,
radiation and diffraction forces. Incident wave forces are calculated through direct integration of the
corresponding pressures over the instantaneous wetted surface. The radiation forces are obtained using
the added mass and damping coefficients calculated by a 3D frequency domain code and transformed into
the time domain by application of the IRF concept. Diffraction forces are obtained in a similar manner.
Solving the six coupled nonlinear integro-differential equations of motion by a time integration method,
motions in the six degrees of freedom (DoF) of the ship are obtained in the time domain. The validation
of this method was conducted through applications to the S-175, DTC and KVLCC2 hulls. Good
agreement was observed between the results of the present method, other numerical codes and available
experimental data, which confirm the capability of the developed numerical approach to deliver reliable
predictions.
Guo et al. (2012a) predicted the added resistance and ship motion of the KVLCC2 hull in head waves
using the ISIS-CFD flow solver. The numerical results are analyzed in terms of added resistance, ship
motions and wake flow. Both free to heave and pitch and fixed models are studied to investigate the
contribution to added resistance from ship motions at different wavelengths, and the results show that
ship motion-induced added resistance is negligible when the wavelength < 0.6 L PP. Comparison with
calculations based on strip theory and experimental results shows that RANS predicts the added
resistance better in all wavelengths.
Simonsen et al. (2013) investigated the KCS container ship in calm water and regular head waves by
means of experimental and CFD studies. The experimental study was conducted in FORCE Technologys
towing tank in Denmark, and the CFD study was conducted using the URANS codes CFDSHIP-IOWA
and Star-CCM+ as well as the potential theory code AEGIR. The wave conditions were chosen in order to
study response in waves under resonance and maximum exciting conditions for three forward speeds. In
the experiment, the heave and pitch motions and the resistance were measured together with wave
elevation of the incoming wave. The results show that the ship responds strongly when the resonance and
maximum exciting conditions are met. With respect to experimental uncertainty, the level for calm water
is comparable to PMM uncertainties for manoeuvering testing while the level is higher in waves.
Concerning the CFD results, the computation shows a very complex and time-varying flow pattern.
Comparison with experimental data shows that the computed motions and resistance in calm water are in
fair agreement with the measurement. In waves, the motions are still in fair agreement with measured
data, but larger differences are observed for the resistance.
It has been expected that the correct forward speed hydrodynamic coefficients of a surface ship
advancing in waves would be obtained if the 3D frequency domain forward-speed free surface Greens
function (Brard, 1948) and the 3D forward speed Greens integral equation presented by Hong (2000)
were jointly used. Hong et al. ((2013b), (2014)) showed that the line integral in the 3D forward speed
Greens integral equation can properly be discretized using 8-node inner collocation 2nd order BEM. Their
numerical results were obtained using Brards Green function expressed through complex exponential
integrals (Guevel et al., 1979). Predictions of the heave damping coefficients for the Wigley hull models
I, II and III show that the larger the water plane area is, namely Wigley II, the more significant the
waterline integral effect is. In this case, the free surface condition represented by the line integral, can
neither be linearized nor be split into steady and unsteady conditions. They compared the calculated
hydrodynamic coefficients to the experimental results by Journe (1992). They reported that their 3D
forward speed method including the exact line integral can successfully be used to obtain the forward
speed hydrodynamic coefficients of surface ships whose Length/Breadth ratios are greater than 5 when
the Froude number is not much greater than approximately 0.3.
2.3
83
Hydroelasticity Methods
Research continues in improvements to frequency domain methods, as well as 3D time domain methods,
the latter, by and large, involving coupling between linear and nonlinear BEM and FE analysis. Coupling
between RANS and FE solvers is investigated for slamming related problems, e.g. Piro and Maki (2013).
Das and Cheung (2012) presented a directly coupled hydroelastic approach in the frequency domain
with a rigorous treatment of the vessel forward speed. The formulation adopts a translating coordinate
system with the free surface boundary conditions accounting for the double-body flow around the vessel
and the radiation condition taking into account the Doppler shift of the scattered waves. A BEM, based on
the Rankine source formulation, describes the potential flow and the hydrodynamic pressure on the
vessel. A FE model relates the vessel response to the hydrodynamic pressure through a kinematic and a
dynamic boundary condition on the wetted hull surface. This direct coupling of the structural and
hydrodynamic systems leads to an equation of motion in terms of the nodal displacement of the finite
elements. The results are compared against predictions from a seakeeping model with forward speed and
the modal superposition method at zero speed. A parametric study of a Wigley hull shows that forward
speed introduces new resonance modes, which amplify the elastic and rigid body responses of the vessel
in waves.
Kashiwagi and Hara (2012) presented a 3D analysis method for ship hydroelastic problems in the
frequency domain, which combines the Rankine panel method for analyzing hydrodynamic responses and
the mode superposition method with 3D FEM for representing the structural deformation. The Rankine
panel method takes into account the forward speed and 3D effects in a rational way and also the effect of
steady disturbance flow on the free surface boundary condition. The natural frequencies of the dry elastic
modes and corresponding mode shapes are computed by a 3D FE method, and the amplitudes of these
elastic modes are determined by solving the coupled equations of motion. Numerical results are presented
for a 2m modified Wigley model travelling in regular head waves, verifying expected performance of the
developed calculation method.
Based on beam theory, Miao et al. (2012) developed a methodology for antisymmetric dynamic
behavior including, warping function as independent deformation and the influence of structural
discontinuity for open-deck ships using a range of theoretical models. The methodology was applied to
investigate the response of a container ship travelling in regular oblique waves. The predicted results, in
terms of dry hull characteristics and wave-induced loads, are compared with predictions using 3D linear
hydroelasticity analysis. It was stated that, although the predictions from the 2D antisymmetric analysis
are dependent on the theoretical model adopted, reasonably good agreement was achieved between 2D
and 3D hydroelastic predictions; thus justifying their use as a more computationally efficient alternative
to 3D models.
Senjanovi et al. (2012) presented an improved method of ship hydroelasticity analysis, based on the
modal superposition method and including structural, hydrostatic and hydrodynamic models. A beam
model is used for the structure comprising a reliable advanced thin-walled girder theory taking into
account shear influence on torsion as well as the contribution of bulkheads and the engine room structure,
as structural discontinuities affect the ship hull stiffness. Consistent restoring stiffness is included in the
analysis via the hydrostatic model. Added mass and hydrodynamics are determined based on the linear
radiation/diffraction theory. Also, the analysis of springing effect on the ship fatigue life is introduced
using the combination of the improved hydroelastic model and 3D FE substructure model. It is shown that
the improved beam hydroelastic model can be efficiently used in the assessment of stress concentrations
of ship structural details. Applicability of the developed theory is shown for the global hydroelastic
analysis of a 11400 TEU container ship, including stress concentration determination in the selected
structural details.
Vidic-Perunovic (2012) presented a finite water depth nonlinear hydroelastic strip theory, based on
relative motion, for calculation of a ships wave-induced vertical bending moment (VBM) in the
frequency domain and applied it to an analytical beam and a tanker. An increasing trend in the low
frequency part of the VBM was noted with decreasing water depth. When the water depth equaled
approximately four times the draught of the vessel, the load response shifted towards lower frequencies,
and the peak was significantly affected. For the high-frequency response, the magnitude of the springing
peak increases with decreasing water depth, primarily owing to an increase in the added mass of the ship
and the wave excitation force. Based on the analysis of the tanker, it was stated that the ships response in
different sea states increases substantially when the water depth is less than approximately five times the
draught, both in linear and nonlinear springing calculations, compared with the deep water calculation.
Based on the linear random vibration theory, Papaioannou et al. (2013) developed a framework for
stochastic hydroelastic analysis of very large floating structures (VLFS) subjected to multidirectional
irregular waves defined through a directional wave spectrum. The approach involves a discrete evaluation
of the relevant transfer matrices through a numerical solution of the fluidstructure interaction (FSI)
84
problem that combines BEM for the fluid potential and the FE method based on the Mindlin plate theory
for the plate response. Spectra of responses are obtained as well as extreme responses, assuming a
Gaussian input. The proposed method is applied to the stochastic analysis of a VLFS and the influence of
the mean wave angle on the standard deviation and extreme values of response quantities is demonstrated.
It is found that the hydroelastic behavior of VLFS is greatly affected when considering a directional wave
spectrum.
Chapchap et al. (2012) investigated the linear 3D radiation problem of a stationary floating flexible
structure undergoing forced oscillations in time domain. The method uses a MEL scheme on an
unstructured mesh. The Eulerian phase of the MEL scheme is solved using a constant BEM, in which
Rankine sources and dipoles are distributed over the boundaries (i.e. walls, free surface and body surface)
of the whole domain. Neumann and Dirichlet boundary conditions are enforced. In the Lagrangian phase,
the linear version of the kinematic and dynamic boundary conditions are explicitly integrated in time
allowing, at the next iteration, Dirichlet boundary conditions to be imposed on the free surface and the
evaluation of the time derivatives of the potential function. The unified symmetric, including both
motions and distortions, radiation problem is formulated and applied to the Wigley hull, treated as a
uniform Euler beam, undergoing forced rigid motions and distortions. Comparisons with 3D frequency
domain hydroelasticity predictions indicate that the implemented numerical schemes are working
reasonably well.
Kim and Kim (2012) presented numerical analysis of ship hydroelasticity based on a fully coupled
approach of linear BEM and FE method. For the analysis of FSI problems, a partitioned method is
applied. The fluid domain surrounding the flexible bodies is solved by a B-spline Rankine panel method
and the structures are modelled using a 3D FE method. The two methods are fully coupled in time domain
using an implicit iterative scheme. The computed natural frequencies and motion responses of a simple
barge and a segmented barge are validated through comparisons with experimental and other numerical
results. A large containership is also considered in order to investigate the accuracy of the method on real
ship application. Based on the computational results, the pros and cons of the approach are discussed,
showing a promising capability for complicated FSI problems. Kim et al. (2013c) applied this fully
coupled 3D BEM-FE method approach to two real ships, namely 6,500 TEU and 10,000 TEU
containerships. Kim et al. (2012b) dealt with the numerical springing analysis of containerships. Good
agreement is observed in linear responses. However, nonlinear springing responses based on weakly
nonlinear approach do not show good agreement with the experiments. In addition, fatigue damage is
calculated using this numerical method.
Yang et al. (2013) investigated the FSI for slamming impact phenomena and dynamic structural
response problem in containerships at an initial design stage using a direct analysis method. Slamming
impact pressures and dynamic structural response are studied using a commercial CFD program (STARCCM+) and a structure analysis program (ABAQUS), respectively. These two programs are coupled
using the co-simulation function of STAR-CCM+, called the one-way coupled scheme of FSI. Numerical
simulations are carried out for bow bottom and stern slamming impact loads of a containership in extreme
design wave conditions. Also, the one way coupled analysis has been applied to obtain the dynamic
structural responses of local structures.
Lee et al. (2012a) investigated nonlinear wave actions and wave induced global loads acting on a large
container ship. An analysis procedure was established to determine values of wave induced VBMs
considering the effects due to whipping suitable for design application. The analysis for predicting
structural capacities has been carried out by computing the ultimate longitudinal strength of the container
ship. Furthermore the assessment of the safety against failure due to excessive maximum loads is
accomplished in Ultimate Limit States.
Mirciu et al. (2012) investigated the hydroelastic responses of a large LPG carrier in irregular head
waves, based on Longuet-Higgins model with second order interference waves spectra. The LPG carrier
has a double hull structure with length over all 238.7 m, design speed 17 knots and two loading cases,
namely full cargo and ballast. The numerical analysis was carried out using the in-house code DYN,
which is based on the 2D linear frequency and nonlinear time domain hydroelasticity theories, the latter
using implicit integration procedures for the equations of motion. The resulting numerical hydroelastic
linear and nonlinear response include low frequency oscillations response and high frequency vibratory
response, taking into account springing and whipping phenomena induced by bottom and side slamming
forces. It was shown that the nonlinear hydroelastic analysis could predict the extreme wave induced
loads on the LPG hull structure.
In order to achieve safe and reasonable design with reference to whipping response of ultra-large
vessel, Kobayakawa et al. (2012) developed a hydroelastic response analysis system which can calculate
the stresses, including hydroelastic responses, on actual ship structures. The details of the system were
described and validation study of system in regular wave was carried out. It was concluded that the
85
system can simulate ship motions, hydrodynamic pressures and VBMs in regular head wave conditions
with sufficient accuracy.
Malenica and Derbanne (2012) discussed actual tools and methodologies used in the design of Ultra
Large Container Ships (ULCS), especially for the hydroelastic phenomena of springing and whipping. It
was concluded that the modeling of springing and whipping is still a challenge and that there is no fully
satisfactory numerical tool capable of dealing with these issues consistently.
Based on 3Dhydroelasticity theory of ships previously developed, including the effect of fluid
compressibility, Zou et al. (2012) further incorporated the hydroelastic and sonoelastic analysis methods
with the Greens function of the Pekeris ocean hydro-acoustic waveguide model, to produce a 3D
sonoelastic analysis method for ships in the ocean hydro-acoustic environment. The method was applied
to predict the sound radiations of a floating elastic spherical shell excited by a concentrated force and a
LNG ship excited by the propeller induced pulsating forces acting on the wetted bottom plate of the stern
in a shallow sea environment. The influences of the free surface and the sea bed on the generalized
hydrodynamic coefficients, the acoustic pressure distributions in space are illustrated and discussed.
2.4
Loads from abnormal waves
Research on abnormal waves and their consequences on marine structures continue on many fronts,
mainly focusing in new techniques of experimental generation of abnormal waves, experimentation of
abnormal wave-structure interactions and numerical modelling to assess the adequacy of the methods
used in representing the abnormal wave and its interaction with marine structures.
Bennett et al. (2012a) investigated how best to generate accurate, repeatable and controllable
unidirectional abnormal waves in an experimental test facility for the purpose of practically modelling the
response of a travelling vessel to abnormal waves. Three techniques, based on linear superposition
principles, for producing an abnormal wave from a sea spectrum were compared, namely NewWave,
Constrained NewWave and Optimized Sea. Each technique was tested for multiple sea spectra. The
experimentally generated waves were compared to both linear and second order wave theory, to assess
the adequacy of the former as it is used in the numerical models for predicting ships response and it is
important to ensure they replicate the abnormal wave record used during experiments. The advantages
and disadvantages of each abnormal wave generation technique are summarized in the paper. Results
showed that in order to allow the most realistic predictions of ship responses, inclusion of the response
history of the vessel, hence the random sea surrounding the abnormal wave is required, i.e. embedding the
abnormal wave in an irregular wave. Two of the techniques used allow this, namely Constrained
NewWave and Optimized Sea. The latter is the preferred option for realistic predictions, as it offers more
control over the abnormal wave shape and significant anomalies between the required wave and that
generated in the tank were observed for the Constrained NewWave technique.
Onorato et al. (2013) used a breather solution based on the one-dimensional nonlinear Schrodinger
equation to generate critical wave sequences containing abnormal waves in the tank of the Technical
University of Berlin. A breather solution describes the modulation of a slightly perturbed wave and is
considered as a suitable modelling tool for abnormal waves. The seakeeping tests were carried out using a
2.3m model of a chemical tanker. The investigators carried out measurements with and without the model
showing that the presence of ship did not alter the measurements of the generated wave. They also
obtained satisfactory agreement between theoretical location of freak wave and tank observations. The
amidships VBM and water-on-deck were amongst the measurements of interest for this committee. To
measure the former the model comprised of two segments connected with force transducers. A large
VBM was measured during the impact of the abnormal wave, exceeding what the authors referred to as
early stage design VMB, especially in the sagging condition by approximately 40%. Furthermore, the
water gauge on deck measured a water column of 10m full-scale during the impact. Although there are
traces of high frequency components in the VBM record shown in this paper, it is difficult to draw any
conclusions with reference to their origin. Nevertheless, this work emphasizes that abnormal waves have
a significant influence not only on global loads but local loads such as water on deck.
Bennett et al. (2013) investigated experimentally the influence of forward speed on the motions
experienced by a frigate encountering an abnormal wave and the implications that such encounters may
have for ship design. Long-crested abnormal waves were generated using the NewWave and Optimized
Sea techniques, thus allowing the influence of vessel response history on ship motions to be investigated.
Random irregular waves, not exhibiting abnormal wave characteristics, were also used as comparators.
The height of the abnormal wave tested in was defined relative to the sea state using the abnormality
index and relative to the encountering vessel using the length index developed during the course of this
research. Motion measurements are compared to prediction using 2D linear strip theory, and 3D partly
nonlinear seakeeping theory, both showing good overall agreement and the latter showing better
agreement with increasing forward and wave height. The severity of motions experienced by the frigate in
86
abnormal waves increases with forward speed, until the frigate appears to be tunneling through the
abnormal wave. Response history is a dominant contributor to ship motions, particularly at higher speeds
and it is, therefore, essential to include it. The unexpected nature of abnormal wave occurrence means that
it is important to consider additional factors such as likely operating condition and ship speed at the time
of encounter when considering implications for ship design.
Vasquez et al. (2013) carried out experimental and numerical investigations on a bulk carrier stationary
and travelling (Fn = 0.1) in irregular head waves with embedded abnormal waves. Two abnormal waves
were used based on existing records, e.g. the new year wave. A segmented model, with the segments
connected by force transducers, was used for the experiments and heave and pitch motions and amidships
VBM were recorded. The numerical predictions were obtained using a 2D rigid body time domain
method, including nonlinear hydrostatic and incident wave contributions, whilst the radiation component
is based on the mean wetted surface hydrodynamics in the frequency domain and accounted for through
IRFs. In addition the effect of green water, but not that of slamming, is included. The authors noted that
the nonlinear effects in the numerical predictions were small, attributed to the hull shape of the bulk
carrier. The overall agreement between experimental measurements and numerical predictions is good for
Fn = 0, but VBM is overestimated for Fn = 0.1, especially for the more extreme wave condition. The
authors observed that for the bulk carrier tested the influence of green water on the global loads was not
important. With reference to the effects of forward speed a maximum increase of 27% for the VBM was
observed for the more extreme case, although it is difficult to draw conclusions as only one non-zero
forward speed case was tested.
Bennett et al. (2012b) compared experimental measurements of rigid body motions and global waveinduced loads of a naval frigate in abnormal waves to predictions made using a 2D linear hydroelasticity
method. Experiments are conducted using a segmented, flexible backbone model of a typical naval frigate
constructed of four rigid segments attached to a uniform aluminum backbone. The model was designed to
match the natural frequency of the 2-node vertical bending moment. VBMs were measured at midships
and the aft and stern quarters. The 2D linear hydroelasticity model idealized the vessel as a non-uniform
Timoshenko beam and slamming was included via the Ochi-Motter impact slamming model. Qualitative
agreement is seen between experimental measurements and 2D linear hydroelasticity predictions of VBM
in abnormal sea states. However, the numerical predictions in general overpredict the maximum VBMs
experienced. It is thought that this may be due to no consideration being given to the nonlinear effects
when evaluating steady state components and future work should consider this. Furthermore, the
maximum VBM along the vessel occurs at the aft quarter in experiments and amidships in predictions.
Zhao and Hu (2012) carried out experimental and numerical investigations on a 2D box shaped body,
with a superstructure for green water simulation, in abnormal waves, focused and combined with a
regular wave. The aim of their investigation is to assess how the RANS CIP (Constrained Interpolation
Profile) -based method performs with extreme waves, ignoring turbulence and surface tension effects.
There is good agreement between numerical and experimental wave profiles. Motions in the vertical
plane, heave and pitch, and pressure on deck were measured in the experiments, with the longitudinal
displacement constrained. Good agreement between predicted motions and measurements is obtained for
the focused wave, but the predictions are smaller than the measurements in the more realistic case of the
combined wave. The method also underestimates the pressures on the deck, but the overall qualitative
agreement between predictions and measurements is satisfactory, showing that the method in question is
suitable in modelling highly nonlinear phenomena. Zhao et al. (2014d) continued these investigations by
refining their 2D numerical model by incorporating a more accurate free surface capturing method based
on VOF. They also allowed for the surge motion, both in the experiments and numerical investigations,
comparing the influence of constraining this motion. Only the focused wave was used in this work. When
the surge is constrained, the agreement between experimental and numerical predictions for the motions
shows some improvement, especially for the pitch motion. In the case when the box is free to surge, there
is underestimation of the pitch motion attributed by the authors to the increasing complexity of the flow.
With reference to the pressure on deck, the improved free surface capturing results in small improvements
in the predictions. Nevertheless, the pressure is underestimated, when surge is either constrained or not,
by comparison to the experimental measurements. The authors also investigated the influence of grid
density on their predictions using coarse, medium and fine grids. The effect of grid density is small on the
motions with the coarse grid providing closer predictions to the measurements. The on-deck pressure
predictions are influenced more by the grid density, with the middle grid showing the best overall
agreement, still underestimating, and the fine grid having issues with the first pressure peak when surge is
not constrained.
Hu et al. (2014) investigated the interactions between a freak wave and a beamlike structure using 2D
numerical modelling, equivalent to assuming that the beamlike structure is infinitely long in the
athwartships direction. Nevertheless, their investigations are of interest due to (i) coupling between a
87
RANS (VOF) solver and (linear) FE method, through iterating the fluid pressure field and velocity and
deformation of the structure at each time step; (ii) the successful generation of a, long-crested, freak wave
based on real measurements; and (iii) the use of different boundary conditions for the beam, including a
rigid beam. There are no comparisons of the predictions with either experimental measurements or other
numerical predictions. Their observation that the beam deformation affects the bottom impact pressure, in
the sense of pressure reducing with increasing deformation, is of interest. They also observed that there
were relatively small differences between dry and wet natural frequencies, probably due to the 2D
simplification of their modelling approach.
With ship encounters with abnormal waves occurring more frequently, it is important to assess whether
such encounters are causing ships to operate beyond the limits of the design rules. Bennett et al. (2014)
experimentally assessed the global loads experienced by a typical naval frigate in a range of abnormal sea
states generated using the Optimized Sea technique. Experiments are conducted using a segmented,
flexible backbone model (Bennett et al., 2012b) operating at its service speed of 18 knots full scale, in
head seas. Measured VBM values are compared to those obtained from the design rules for naval vessels
(Lloyds Register, Bureau Veritas and Det Norske Veritas), both with and without allowances for extreme
wave events. The presence of an abnormal wave event in a sea state increases the maximum VBM the
frigate experiences by up to 2.5 times compared to in an equivalent random sea. The severity of slams
experienced by the ship increases significantly with 40-60% of slams having a forefoot emergence of
greater than 25% the ship length. The authors recommend that, in order to have a sufficient safety margin,
design rules that account for extreme waves should be used; however, there were still some cases where
this did not appear sufficient and these require further investigation.
Likelihood of occurrence of abnormal waves in relation to recent climate changes is an extremely
important subject area affecting the consideration of abnormal waves when evaluating design loads.
Accordingly two fairly recent papers are singled out for inclusion in this report. Vanem et al. (2014) dealt
with uncertainties in the climate modeling and the effects of abnormal waves on the wave bending
moments of ships. This may significantly affect the safety level of existing and newly developed ships;
thus, it is an important area of concern. The uncertainties in the modeling of both the climate changes and
in the modeling of the wave statistics are significant and conclusions vague. Without claiming generality,
when comparing historical with simulated data, surprisingly the statistical mean and up to the 90%
statistical quintiles of the historical data of wave heights are higher than the simulated future data for
some selected scenarios. Also, the standard deviations are larger for the future projections, whilst the
statistical upper tails are longer. The interpretation of this behavior and the consequences/conclusions on
the way ahead are very crucial. In a further paper by the same research team, Bitner-Gregersen et al.
(2014) presented the impact of climate change and extreme waves on tanker design. The study indicates
that observed and projected changes in wave climate will have large impact on tanker design practice.
Necessary increase in partial safety factor(s) and/or revised specification of the characteristic wave
bending moment need to be further investigated.
3.
3.1
SHIP STRUCTURES SPECIALIST TOPICS

Slamming and Whipping
Impulsive slamming loads and the consequent vibratory response, namely whipping, continue to be the
focus of numerous investigations, as commercial fleets face increasingly rough seas due to climate change
and tight schedules. The modeling of water impacts and the calculation of whipping response along with
their statistical combination with wave loads remain crucial aspects in assessing ship structural strength.
Analysis of the water exit phase in slamming problems has been the focus of many investigations.
Korobkin (2013) used a potential flow formulation for describing the loads acting on 2D and
axisymmetric floating bodies that suddenly start moving upward at a constant acceleration. The unknown
size of the wetted area is determined by the condition that the speed of the contact points is proportional
to the local velocity of the flow. This condition provides a nonlinear Abel-type integral equation which is
solved explicitly. Predicted hydrodynamic forces are compared with the CFD results for a rigid wedge of
deadrise angle 10 and a circular cylinder which enters the water and then exits by Piro and Maki (2011)
and Tassin et al. (2013). Piro and Maki ((2011), (2012)) investigated the 2D slamming problem focusing
in particular on the exit phase of a rigid wedge of deadrise angle 10 which lifts at constant acceleration.
Their numerical approach for the air-water flow exploits a finite volume solver from the OpenFOAM
library and is applied to an ALE form of the N-S equations. This approach allows computing the solution
on moving and deforming meshes whereas the free surface is captured with a VOF technique. Conditions
of the impact are selected in such a way that the speed of the wedge becomes zero before the wedge is
completely wetted. Piro and Maki (2011) compared the computed slamming force with Wagner and Von
Karman predictions, which overestimated and underestimated the slamming peak, respectively, with
respect to the CFD approach. On the other hand the exit force time series from the CFD simulations by
88
Piro and Maki (2012) compared well with the experimental force measured by Tveitnes et al. (2008).
Furthermore, Piro and Maki (2013) coupled their CFD solver to a deformable structure, i.e. an elastic
wedge, using a modal representation that was obtained from a FE modelling of the wedge with beam and
plate elements. A linearized boundary condition is used on the common interface to eliminate mesh
deformation. In the case of a rigid body experiencing constant acceleration, a satisfactory agreement is
obtained during water entry with Wagner theory and, during water exit, with the von Karman model. In
the general case, accounting also for wedge flexibility, the numerical method is validated by comparing
the structural response with the theoretical results by Korobkin et al. (2006) for hydroelastic impact at
constant velocity. The hydroelastic effects were found to be important for a wide range of loading
conditions, depending on the imposed entry velocity variation, where the rigid quasi-static solution was
shown to underpredict the maximum deflection or stress and the rigid dynamic solution may underpredict
or overestimate them.
Tassin et al. ((2012a), (2013)) addressed the 3D problem by solving the 2D problem of a body with
time-varying shape (2D + t approach). Each 2D problem was studied by using the modified Logvinovich
model (MLM) during the entry stage and the von Karman model during the exit stage. This combined
MLM-von Karman model was applied to the vertical water entry and exit of the wedge studied by Piro
and Maki ((2011), (2013)). It was found that during the entry stage these two methods agree very well.
The agreement during the exit stage is in general good; however, the CFD results by Piro and Maki
predict a longer duration and a larger force peak of the exit phase than the von Karman method.
Renewed attention to the water impact problem was motivated also by the analysis of aircraft landing
on the water surface (ditching problem) as one of the goals of the SMAES FP7 funded project, as reported
by Iafrati et al. (2014) The experiments were performed in the new CNR-INSEAN facility that was
designed for performing water impact tests at the full-scale ditching speeds up to 180 km/h. The
instrumented elastic plates of the fuselage are attached to a nacelle which is pushed by a crossbow system
along inclined rails ending in the tank water. This is the first facility of its kind ever built for this purpose.
Accurate prediction of the negative forces in the rear part of the fuselage during its landing was shown to
be important in terms of the aircraft motion and the bending stresses in the fuselage. The ditching
problem was also studied by Hua et al. (2011). To simulate the fluid-solid interactions caused by low
speed ditching, the authors proposed a 3D dynamical structural model of the full-scale airplane including
the wing and the control surfaces but disregarding the plastic deformation during impact. The FE model
was coupled to an ALE fluid field model, and the computational model was then solved within LSDYNA nonlinear FE code. No comparisons were given with either experimental data or other numerical
methods.
The exact prediction of the wetted area affected by slamming requires some additional assumptions in
the approaches based on the solution of the Laplace equation and, in this respect, some authors pointed
out a similarity between the local flow close to the separation point in the problem of water exit and the
problem of oblique impact with separation, in particular in the cases where there is a relevant horizontal
speed. Reinhard et al. (2012b) implemented three different criteria for the rear separation point in blunt
body water impacts with large horizontal and freefall vertical speeds. It was shown that the choice of the
separation condition significantly changes the contact point limiting the wetted part of the body, the
applied loads and then the predicted motion of the body. Reinhard et al. (2012a) imposed Wagners
condition for the rear contact point to achieve a solution for the oblique hydroelastic impact of a plate into
the water. It was shown that the bending stresses in the plate may exceed the yield stress but ventilation
decreases the bending stresses compared with a non-ventilated plate. Reihnard et al. (2013) described in
detail the fully coupled model developed for representing the elastic plate impact onto the free surface of
an incompressible fluid. In this paper they assumed that the fluid is attached to the plate from the turnover
region to the trailing edge, so that any suction fluid forces contribute to the dynamics of the plate. The
structural equations are solved by modal superposition subjected to a hydrodynamic pressure field along
the wetted part of the vibrating plate for which an explicit formula is provided. The authors highlighted
that if one applies to the elastic plate the pressure determined from the problem of rigid plate impact, then
the bending stresses are overestimated. It was confirmed that the shape of the free surface, the
hydrodynamic pressure and the flow, are all sensitive to the plate vibrations.
Tassin et al. (2012b) developed a simplified method based on the displacement potential formulation
and the BEM for modelling the impact pressure field on 3D bodies of different shapes. Particular
attention is devoted to the analysis of the wetted surface and the predicted slamming force is compared
with existing results and with a series of impact tests carried out with a hydraulic machine. The
application of a desingularized variational numerical method to the vertical hydrodynamic impact
problem of axisymmetric bodies is addressed within the so-called generalized von Karman model by
Santos et al. (2012). The 3D body penetrating the free surface is represented at each time instant as a
double body built on the wetted part alone of the entire body. The method does not account for the
89
description of the free surface pile up, as in the von Karman theory, but added mass contribution to the
slamming force can be accurately computed for any 3D axisymmetric shape and were compared with
WAMIT in the case of a sphere and an oblate spheroid.
Water entry of sandwich panels was recently studied because of their potential higher efficacy in
absorbing slamming impacts with respect to panels made of homogeneous materials. Xiao and Batra
(2012) employed the BEM for computing the loads on several 2D rigid hull sections because previous
computations using LS-DYNA software (Das and Batra, 2011) exhibited fluid penetration at the solid
boundaries. The authors compared the slamming pressure predictions with analytical solutions, other
BEM results and experiments for straight wedges with deadrise angles ranging between 4and 45 and
curved bow sections, obtaining a satisfactory agreement overall. Subsequently Xiao and Batra (2014)
coupled this BEM code to the FE solver of a sandwich panel to study the delamination problem during
water entry of an elastic wedge with deadrise angles equal to 5, 10, and 14, respectively. The sandwich
panel was modelled with a layerwise third order shear and normal deformable plate/shell theory. At each
time step the velocity and acceleration of the body and its local deformations are updated according to the
condition that the hydrodynamic pressure varies within a prescribed tolerance; otherwise iterations in
solving the panel-fluid interface equations restart (two-way coupling). In particular, they found that the
consideration of geometric nonlinearities, namely nonlinear terms in the strain-displacement relations,
significantly increase the peak slamming pressure experienced by the panel. The presence of
delamination, studied including the cohesive zone model into a third order shear and normal deformable
plate/shell theory, reduces the pressure acting on the panel surface. Delamination affects then deformation
values compared to the case where the structure behaves linearly. No comparisons are presented against
experimental data or other numerical methods.
Abrahamsen and Faltinsen ((2012b), (2013)) considered the effect of air pocket entrapment in 2D
slamming phenomena caused by a free surface wave impinging a rigid wall. The air pocket during
compression due to slamming is represented as an underdamped mass-spring system. They also derived a
new analytical formula for the oscillation frequency of the entrapped air pocket, showing good agreement
with BEM and sloshing experiments (Abrahamsen and Faltinsen, 2011). It was shown that the Topliss et
al. (1992) formula is not accurate for large air pockets close to the free surface. The differences between
the boundary element solution and the analytical formula showed also that the results are sensitive to the
assumed shape of the air pocket. Abrahamsen and Faltinsen (2013) improved the scaling theory proposed
by Lundgren (1969) which involved only the pressure amplitude but not the rise time. The proposed
approach is based on fitting experimental data relative to pressure time history (Abrahamsen and
Faltinsen, 2011) with a simple mathematical model governed by two parameters, denoted as kinematic
parameters.
Lv and Grenestedt (2013) studied the sensitivity of the response of a hull bottom to a moving slamming
load. Several parameters such as bending stiffness or pressure travelling speed were used. Both the
structure, modelled as a linear Euler-Bernoulli beam, and the slamming load, using piecewise constant
representation, are simplified. This is a 2D one-way coupling model and these assumptions allowed the
authors to calculate analytically the dynamic solution without any dependence of the load on the
response. The authors pointed out the existence of slam load travelling speeds at which bending moments
and deflections are significantly amplified. Kim and Paik (2013) developed a design formula for
predicting permanent deflection of stiffened plates and grillages under impact pressure loads acting
uniformly in space and constant in time, i.e. square wave, on the wetted side of the plate. The method is
verified by a comparison with nonlinear FE methods of stiffened plate structures under impact loads. The
calculated permanent deflection at the center of the plate was slightly overestimated with respect to
ANSYS/LS-DYNA results. The largest difference, of the order of 10%, was reached in the case of small
plate thicknesses. Though the hydrodynamic slamming load is extremely simplified, the conceived
expressions allow inclusion of the slamming problem into the structural design process. To calculate the
added mass of elastic wedges impacting on the water, the discrete vortex method was proposed by Fu and
Qin (2014). The same geometrical approximation of Wagners theory is used and the structure is
represented by a modal decomposition. The numerical results relating to the added mass matrix in case of
small deadrise angles are in good agreement with analytical ones for the rigid wedge or with numerical
ones provided by Khabakhpasheva and Korobkin (2013) for the case with the deadrise angle set to 10.
Analysis of more complex geometrical configurations has motivated further efforts in implementing
sophisticated computational methods. Panciroli et al. ((2012), (2013)) experimentally and numerically
analyzed the water slamming of elastic wedges made of composite panels. In both papers the 2D
hydrodynamic load was computed using the SPH formulation available in LS-DYNA, accounting for
hydroelastic effect due to two-way coupling with the FE solver. In the first paper (2012) attention is
mainly devoted to assess the calculation of hydrodynamic loads using SPH through comparison with
strain responses recorded in free fall impact tests for various panel thickness, deadrise angle and impact
90
velocity values. Results are in acceptable agreement in the initial stage of the water entry in terms of
maximum strains, but accuracy suffers when the wetted zone becomes more fragmented at the late impact
stages. Panciroli et al. (2013) focus on capturing the hydroelastic effects on stresses in case of large
flexibility of the wedge panels. First it is shown that the experimental vertical acceleration of the free
falling body is dependent on the wedge configuration, which could be one of these in the impacting part:
cantilever plates, simply supported plates and plates clamped on both sides. In the case of wedges with
clamped panels, the solution obtained with the FE/SPH coupled model compared well with results
available in the open literature in terms of normalized pressure. Subsequently, the sensitivity of the
solution with respect to the ratio between the wetting time and the first natural period was analyzed and
was found to affect significantly the peak pressures. The BEM based on velocity potential and fully
nonlinear boundary conditions in the time domain was applied by Sun et al. (2015) to solve the problem
of a 2D wedge entering waves with gravity effect. The gravity effect was found to be relevant for the
pressure distribution as the impact time becomes comparable to the ratio of the entry speed to the gravity
acceleration. At the same time, impacting a wavy surface results in differences to the pressure distribution
on the wedge sides depending on the wave height and wavelength. Scolan and Korobkin ((2012), (2013))
developed a potential flow theory for an elliptic paraboloid entering vertically a liquid surface on which a
regular wave propagates. The Wagner problem is posed in terms of the displacement potential for
achieving an analytical solution. Comparisons with experiments and the MLM show good agreement in
predicting the initial rise of the vertical force at the early impact stage. Moore et al. ((2012), (2013))
extended the Wagner theory for the normal impact of 3D rigid bodies that are nearly parallel to the water
surface. They highlighted that the points at which the free surface turns over in the solution of the Wagner
model for the oblique impact of a 2D body are directly related to the turnover points in the equivalent
normal impact problem. This observation allowed the authors to discuss the limits of applicability of
Wagner theory to 2D oblique impact problems. Thus, for a symmetric body profile, they found that free
surface elevation can be considered symmetric after proper change of variables, but this property is not
transferred to the leading order velocity potential or pressure. Wang and Guedes Soares (2013) studied
the 2D water entry of a bow-flared section, previously used in drop tests by Aarsnes (1996), with different
heel angles using an explicit FE code in order to assess the influence of heel on the slamming pressures
and forces. The modeling technique of the FSI adopts the explicit FE code LS-DYNA based on a multimaterial ALE formulation and a penalty coupling method. The computed vertical force agreed well with
BEM predictions irrespective of the heel angle, even if a significantly lower peak value was found at the
maximum heel angle of 28.3. In general, experimental values were found to be equal or lower than the
computed ones with important discrepancies for hell angles larger than 10. Maximum values increased
with increasing heel angle and vertical impact speed for all the approaches considered. Pressure time
histories were also compared but relative trends showed less clear interpretations. Yang and Qiu (2012)
computed slamming forces on 2D and 3D bodies by solving N-S equations with a CIP-based finite
difference method. In the case of wedges the numerical predictions agreed well with BEM results and
experiments in terms of the pressure coefficient. They also investigated the free fall of a catamaran
segment and a finite length cylinder, for which experimental data in terms of penetration velocity and
maximum depth are available, respectively. The agreement was acceptable in both cases. A 3D Cartesian
cut cell free surface capturing method was developed by Hu et al. (2013) for water entry problems and
applied to rigid wedges and the generic Bobber heave type wave energy device. A high resolution
Riemann-based finite volume flow solver for the air-water fluid domain was also implemented and results
were compared with available experimental data. Nguyen et al. (2014) studied the water impact of various
3D geometries, namely a hemisphere, two cones and a free falling wedge, with an implicit algorithm
based on a dual-time pseudo-compressibility method. Flow fields of incompressible viscous fluids were
solved using unsteady RANS equations. A second order VOF interface tracking algorithm was developed
in a generalized curvilinear coordinate system to track the interface between the two phases in the
computational domain. Sensitivity analysis with respect the spatial grid resolution and the time step was
performed so as to assess the accuracy of the results. Free surface deformation, pressure coefficients,
impact velocities and vertical accelerations during impact are compared with available experimental data
and asymptotic theory, showing good agreement at a reasonable computational cost.
Several papers were devoted to including/coupling slamming load predictions into seakeeping codes.
Greco and Lugni (2012) employed the Ochi criterion for identifying the presence of water slams in the
time history of the ship motion provided by a 3D numerical solver. The predicted pressures were then
compared with experiments in Greco et al. (2012). A more complex approach for violent FSI, combining
a potential flow and a N-S solver within a domain decomposition strategy, was then proposed by Greco et
al. (2013) for handling slamming and water-on-deck phenomena at an acceptable computational cost. All
three papers are discussed in more detail in Section 2.2. A benchmark study performed by the ISSC 2012
Committee II.1 Dynamic Response was reported by Drummen and Holtmann (2014). Several approaches
91
for calculating the slamming loads in head regular waves, namely BEM over 2D sections and RANS
solvers provided by ANSYS-CFX, Star-CCM+ and Comet, were considered and their capability for
predicting the induced VBM of a flexible structure was compared. The reference experimental data were
provided from seakeeping tests of a scaled segmented model of a long ferry. The method that generally
performed better in predicting bow forces and amidships VBM was based on a fully coupled FSI
computation based on the Comet RANS solver for fluid dynamics and a Timoshenko beam for structural
representation. This paper is discussed in more detail in Section 7.1. Tuitman et al. (2013) pointed out the
need of using a nonlinear seakeeping model, including the slamming load computation, for improving the
stress analysis when using realistic 3D FE models. Slamming, Froude-Krylov force and hydrostatic
loads included nonlinearities, whilst the diffraction and radiation loads were kept linear to reduce the
computational effort. This analysis, though disregarding hydroelastic coupling, appears to provide
significant improvements in estimating the ultimate stress on structural details compared to linear 3D
BEM. A generalized Wagner model was exploited by Kim and Kim (2012) for the evaluation of
slamming loads within a FSI solver, also discussed in Section 2.3, coupling a 3D Rankine panel method
with structural models of different complexity, namely Timoshenko-Vlasov beam model and a 3D shellbased FE representation. The generalized Wagner model required dividing the bow part of the hull into
2D strips for which slamming pressures are computed and then applied to the structural models of a 6600
TEU containership and the scaled segmented model of a 10000 TEU ULC. Comparison of the different
approaches with experimental data from scaled model experiments showed the codes capability of
describing the slamming induced response. A practical procedure for computing the long term
distribution of combined wave and whipping bending moments of container ships was proposed by Corak
et al. (2013), discussed in more detail in Section 7.1.
3.2
Sloshing
Sloshing is the motion of free liquid surface inside a container and of practical importance with regard to
the safety of relevant transportation systems. So far, many researchers have devoted their efforts to
studying liquid sloshing problems. Over the course of the reporting period, newly proposed analytical
methods, designed experimental setups and developed numerical schemes have been applied to improve
the understanding of the sloshing phenomena. Due to the growing LNG market, hence, increasing size of
LNG ships and tanks, sloshing has become of increasing concern for designers. Large impact loads
induced by sloshing on tank walls may lead to structural damage and also have effects on ship motions
and loads.
3.2.1
Analytical methods
Based on their previous work, Faltinsen and Timokha (2012) derived a new Trefftz representation for the
sloshing velocity potential for the liquid sloshing problem in a 2D circular tank. This solution included a
modified Poisson integral depending on the fully continuous component of the strength functions and
terms that are proportional to the jumps of the strength function at the tank top. This work provides an
analytical presentation for the sloshing velocity potential that is applicable for higher tank fillings and has
a clear mathematical and physical treatment for the linear and nonlinear cases. Numerical experiments
with the present solution using different base functions were conducted. Agreement with others
published results is good, suggesting that this solution can approximate the natural sloshing modes
accurately for both lower and higher tank filling levels.
Ikeda et al. (2012) studied nonlinear liquid sloshing in a 3D square tank subjected to oblique horizontal
excitation with an improved model. In their theoretical analysis, besides the two predominant sloshing
modes, five higher sloshing modes were considered by applying Galerkin's method to derive the modal
equations of motion. Linear viscous terms were incorporated to consider the damping effects. Van der
Pol's method was applied to determine the frequency response curves for sloshing. Experiments were
conducted with various excitation frequencies and excitation angles. The quantitative agreement between
the theoretical results and the experimental data confirms the validity of the theoretical analysis. Since
damping effects were taken into account in their study the obtained frequency response curves, containing
both stable and unstable steady state solutions, for amplitudes and phase angles can clearly explain the
phenomena of nonlinear sloshing dynamics in real systems. In their study, it was also concluded that
higher sloshing modes play a significant role in improving the accuracy of the frequency response curves.
To theoretically study sloshing phenomena in a 2D rectangular tank in shallow water, i.e. low filling
conditions, Antuono et al. (2012) proposed a modal system starting with a set of Boussinesq-type
equations with a linearized dispersive term. Based on a spatial Fourier decomposition and capable of
representing a generic 2D motion of the tank, this system employs a spatial modulating term to include
the exciting forces. This leads to a formulation that is consistent with the conservation of momentum and
energy of the flow. Comparisons between experimental measurements and numerical simulations validate
92
the capability of the proposed model. The authors also pointed out the validation of the proposed model
confirms that depth-averaged equations can provide a good description of sloshing motions when the
water depth is shallow and the waves are non-breaking.
Faltinsen and Timokha (2013) were the first to construct an analytical method of nonlinear sloshing in
a spherical tank. Using analytically approximate natural sloshing modes and curvilinear spatial
coordinates, (i) general fully nonlinear modal equations, (ii) weakly nonlinear modal equations and (iii)
Moiseev-Narimanov modal system were derived, respectively, for the spherical shape. In their study, the
three systems play different roles. The first one is of auxiliary character while the second can be
considered as a basis for the third system. The latter was emphasized in this study to construct an
asymptotic time-periodic solution and classify the steady state wave regimes appearing as stable and
unstable planar waves and swirling for spherical tanks theoretically. The results were compared against
experimental data and good agreement was achieved for the liquid depth to radius ratios between 0.2 to 1.
Discrepancies were found for higher ratios (i.e. between 1 to 2) due to the secondary resonance. The
formulae for a square-based tank developed by Faltinsen et al. (2003) were extended by Takahara and
Kimura (2012) and Takahara et al. (2012), suggesting that the method for the square-based tanks might
have uses for appropriate engineering applications.
3.2.2
Experimental investigations
Ji et al. (2012) studied the nonlinear sloshing experimentally with the aid of the time history of wave
elevation at the sidewall, free surface profile captured by high speed camera and velocity vector flow field
obtained by PIV technique. In this study, non-resonant sloshing under the deep water filling condition
was considered in a harmonically oscillating (surge) rectangular tank with breadth of 60 mm, length of
300 mm and height of 180 mm. Based on the experiments, they found that four regimes of sloshing waves
can be categorized: mild 2D wave, strongly nonlinear 2D wave with hydraulic jump-like motion, 3D
wave with regular structure in the longitudinal direction, and 3D chaotic wave. In their study they focused
on a comprehensive discussion on strongly nonlinear 2D wave and they concluded that this off-resonant
sloshing problem can be characterized into a combination of three sloshing motions: (i). standing waves
during run-down process similar with linear sloshing, (ii) hydraulic jump along the vertical wall during
the run-up process and (iii) bore motion propagating from the sidewall to the interior fluid region
resembling a dam break during stationary process. The authors presented time histories of wave
elevations on the tank wall, wave profiles and velocity vectors at selected time instants to show the
evolution. These can be used to validate relevant numerical models.
To measure the impact pressure with accuracy during the sloshing process excited by large sinusoidal
sway motion, Pistani and Thiagarajan (2012) setup a sloshing experiment in a 2D rectangular tank. In
their study, the artificial pressure spike acquired by the transducer when measuring in the two-phase
environment was resolved by lowering the excitation voltage of the sensors. Details of the analysis of
characteristics of the pressure traces during the impact of the fluid and their location in the tank were also
discussed, for a 30% filling level only. In addition, they proposed a strategy for defining a threshold
pressure based on the transducer location and sloshing impact time.
Kim et al. (2013f) introduced the sloshing experimental facility at the Seoul National University
(SNU). Recently, SNU equipped a new experimental facility for sloshing model testing, and this paper
summarizes recent research and experimentation carried out. In particular, this paper focuses on the
experimental setup, post-processing of measured signal, and findings from recent sloshing experiments.
Their findings, being useful to those engaged in experimentation, are summarized as follows:
Integrated circuit piezoelectric (ICP) sensors are sensitive to the change of contact medium.
Temperature difference between the sensor and contact medium (i.e. gas or liquid) must be minimized
before sloshing experiment. Sensors of the same type, diameter and linearity can show different results.
Therefore, the pressure sensors need to be carefully chosen and applied to sloshing experiment.
The density ratio of liquid and gas should be carefully handled in sloshing experiment sand analysis.
The present experiment showed generally higher sloshing impact pressure with the higher density ratio
of gas and liquid applied. Thus, when sloshing model test is performed, the consideration of the density
ratio of liquid (LNG) and gas (NG) which satisfies that of actual LNG cargo is recommended in model
tank.
The current pressure-visual synchronized system may provide some important correlation between
impact pressure and instantaneous flow velocity. The current PIV system showed oscillatory velocity
field during sloshing impact occurrence. This type of observation is helpful for the development of
numerical simulation code and understanding the relationship between kinematics and dynamics during
sloshing.
Statistical properties of peak pressure are dependent on the size of the time window in sloshing
experiments for both regular and irregular motion excitations. In the case of sloshing experiments at
93
SNU, regular excitation required about 500 cycles for reasonable convergence and about 50-hour
motion excitations were suitable for irregular motion.
Zhao et al. (2014a) carried out scaled model tests for an FLNG section excited in roll by white noise
waves. Comparison studies of the vessel motions using fresh water and equivalent steel ballast weights
are conducted. The following conclusions are drawn from this experimental investigation: (i) the spectral
peak of the roll motions shifts to a lower frequency due to the changed center of gravity and the moment
of inertia for the reference filling conditions in the study; a spectral peak of the internal sloshing
oscillations is observed at the corresponding excitation frequency; (ii) sloshing at the head and the
following sides of the inner tank exhibits similar responses for the reference configuration in this study;
(iii) prominent RAO peaks of the internal sloshing oscillation have be observed for the excitation
frequencies that are equal to the natural frequencies of sloshing in different modes; (iv) a large RAO of
the internal sloshing at some excitation frequencies does not necessarily correspond to a large RAO of the
vessel motions. Only one filling condition was selected as the reference. Therefore, great care is required
when extending this studys conclusions directly to cases of translational motions and other filling levels.
3.2.3
Numerical simulation
To model the air pocket impact during sloshing in a 2D rectangular tank, Abrahamsen and Faltinsen
(2012a) used two numerical methods applied at different stages of the simulation. Before the wave
touches the roof, a new numerical method called the boundary-element finite-difference method was
developed to simulate the two-phase flows. The water domain was discretized by a BEM while the air
flow was modelled by a finite-difference method. The interface is tracked in space, satisfying fully
nonlinear boundary conditions and remains sharp during the simulation. After the wave touches the roof
and the air pocket is entrapped, a MEL method is used. The air pocket is then modelled by a polytropic
gas law, whilst the water flow is modelled by a potential flow theory. Comparison between numerical
result and experimental data shows that this numerical model overestimates the peak pressure inside the
air pocket by 17% which may be caused by errors in the initial air pocket volume or the 3D effects.
Godderidge et al. (2012b) developed a mathematical model using a phenomenological modelling
approach based on the pendulum equation. The equations were numerically solved by a variable-order
Adams-Bashforth-Moulton scheme. Damping characteristics were replicated using a first and third order
model. The use of modified impact potential makes the model capable of simulating nonlinear sloshing
with fluid impacts. Godderidge et al. (2012a) applied this model to evaluate its capability in simulating
sloshing in a rectangular tank at critical filling levels. Good agreement with other CFD method showing
that the rapid method is sufficiently accurate and suitable for the fast time assessment of sloshing.
Guo et al. (2012b) applied the newly developed finite volume particle (FVP) method based on the fully
Lagrangian frame work to simulate the sloshing flow in a 3D circular tank. Four sloshing cases were
studied numerically and experimentally. The first two cases are the typical sloshing for a single liquid
phase and studied the influence of the circular wall geometry in a sensitivity analysis. The free surface
behavior and hydrodynamic pressure are also reported. Sloshing motion and its time of occurrence were
compared with experimental data and generally good agreement was observed, while the maximum wave
height was overestimated by the present method due to the difference in the initial condition between
simulation and experiment. The second two cases simulated the sloshing with solid bodies to preliminary
verify the applicability of the FVP method. Good agreement against experiments indicates that the FVP
method can be applied to simulate 3D sloshing waves. Recently, Li et al. (2014b) employed a numerical
model based on Finite Volume method to simulate sloshing in a 2D rectangular tank with the tree-based
adaptive algorithm. The VOF scheme was adopted to capture the free surface. A series of sloshing
experiments under roll excitation with various excitation frequencies and filling levels were carried out,
obtaining pressure measurements to validate the numerical model. During their experiments, they found
that the 3D effect was more sensitive to the increase of the frequency than the filling levels. They also
reported that in lower excitation frequencies cases, standing wave dominates whilst travelling waves are
barely found. In all cases investigated the numerical predictions match the experimental data fairy well.
SPH is a popular mesh-free, Lagrangian particle method with attractive features in modelling free
surface flows. Shao et al. (2012) used a re-normalization approach to approximate density and applied a
corrective kernel gradient to achieve better accuracy with smooth pressure field. Their method is aimed at
simulating incompressible liquid sloshing, using the artificial compressibility technique. They also
incorporated the RANS model to consider the turbulence effects. A coupled dynamic solid boundary
treatment (SBT) algorithm was proposed to improve the accuracy near the solid boundary. Three sloshing
case excited considering horizontal and roll excitations, both without baffles, and horizontal excitations
with baffles in a rectangular tank were simulated, for a range of filling levels. Good agreement between
the present SPH model and the experimental measurements without baffles and other numerical
predictions validates the effectiveness of this numerical model. Based on this SBT algorithm, Chen et al.
94
(2013b) proposed further improvements to the SPH model by applying a smaller acting distance of
repulsive boundary particles. They also utilized Moving Least Squares method in the density reinitialization to obtain better pressure field. Based on the improved SBT, a new pressure measurement
method on solid walls was also given, which took a pressure correction term into account. A series of
experiments were carried out using a 2D square tank with 30% filling levels excited by sinusoidal roll
motions with four different excitation frequencies. The numerical predictions matched the experimental
data well and improved pressure results were observed with this numerical model showing less
fluctuations. Recently, Gotoh et al. (2014) proposed two schemes to improve the incompressible SPHbased methods used in the simulation of 2D violent sloshing flows with a particular focus on sloshing
induced impact pressure. They comprised a higher order Laplacian of Poisson Pressure Equation (PPE)
and an Error Compensating Source of PPE to minimize the instantaneous and accumulative projectionbased errors. According to the authors discussion, there are two error sources in this scheme which are
related to the instantaneous variation of particle density and the deviation of particle density from the
theoretical constant at a certain time step respectively. Comparison of predictions with available
experimental measurements for sloshing in a rectangular tank excited by sinusoidal sway and roll
motions, 16% and 33% filling levels, respectively, showed excellent agreement emphasizing the
enhancing effects of the proposed numerical model. In their study, the importance of dynamically
adjusted coefficients of error minimizing terms in the source term of PPE was highlighted, which is
shown to be superior relative to a scheme with constant coefficients.
3.2.4
Sloshing with internal suppressing structures
In recent studies, baffles are the most used devices to restrain the sloshing motions. Wu et al. (2012a)
developed a 2D numerical model to study the viscous fluid sloshing in tanks with baffles. The sloshing
motion is captured by the time independent finite difference scheme while to simulate the baffles, the
fictitious cell approach is applied. With the help of the stretch technique of the grid system, the local
phenomena of the flow motions around the baffle tip can be captured. An experimental investigation of
liquid sloshing in a rectangular tank with baffles was carried out, for surge motion and filling levels of
50% and 25%. Good agreement between the wave elevation measurements and the numerical results
indicate that their present numerical model can be applied to analyze the sloshing motion in 2D tanks with
internal structures. Jung et al. (2012) investigated the effect of the vertical baffle height on liquid sloshing
in a 3D rectangular tank, 70% filling level. The ANSYS FLUENT CFD code was applied solve the N-S
equations and the VOF method is adopted to track the sloshing motion. In their study, it was found that
the critical ratio of the baffle height to the initial liquid height is 0.3, beyond which the liquid cannot
reach the roof of the tank. When the ratio is 0.9, minimized sloshing wave was obtained. When the ratio is
larger than 1, the liquid could not go over the baffle and almost linear behavior was observed. To examine
the effectiveness of various baffle arrangements and provide test data for the numerical validation,
Akyildiz et al. (2013) designed an experimental setup to study the sloshing behavior with and without
baffles in a cylindrical tank with various filling levels under sinusoidal roll motion. In their study, it was
reported that ring baffle arrangements are very effective in reducing the sloshing loads. By contrast, Koh
et al. (2013) developed a constrained floating baffle (CFB). An improved consistent particle method
(CPM) which applies a combination of a zero-density-variation condition and a velocity-divergence-free
condition was employed to simulate the sloshing in a 2D tank with CFB. Experiments, in surge excitation,
were also conducted and good agreement with the numerical result validated the improved CPM. In
addition, it was also found that the CFB was effective in suppressing sloshing motions.
Apart from baffles, perforated screens are also commonly used sloshing-suppressing devices. Molin
and Remy (2013) conducted an extensive series of experiments to study the sloshing motion in a
rectangular tank subjected to forced sway and rolling motions with a perforated screen, 53% filling level
for both. In contrast to the experiments carried out by Faltinsen et al. (2011), they focused on measuring
hydrodynamic loads by force sensors fitted to the test bench with constant open ratio of the screen and
varied motion amplitudes. In their study, damping coefficients were found to be large over a wide
frequency range. Thus, they concluded that Tuned Liquid Dampers need not be accurately tuned and can
be effective over a wide range of excitation frequencies. Instead of vertical placed screens, Jin et al.
(2014) performed experiments to test the effect of a horizontal perforated plate on liquid sloshing in a 2D
62.5% filled tank subjected to sway oscillations with various excitation frequencies and amplitudes. It
was observed that the frequency, of the wave generated due to sloshing, was minimally influenced by the
plates whilst the wave amplitude was significantly reduced. In addition, they concluded that the plates
must be placed under and closer to the water surface from the bottom to offer better restraining
effects.
3.2.5
95
Sloshing and ship motions
Wang et al. (2012) investigated the interaction phenomenon of sloshing motions and the global wave
loads by seakeeping model tests of a self-propelled LNG ship with a liquid cargo tank, 30% filling level.
It was noted that the existence of liquid in the tank affected the natural rolling period of the ship. The
motion period of the liquid in the tank depends not only on the inner shape of the tank and the liquid
filling level, but also on the wave heading angle and ship speed. Sloshing induced tri-axial total forces
between the tank and the ship are different depending on wave length and travelling speed. The existence
of liquid sloshing in cargo tank might reduce the global wave loads of LNG ships to some degree.
Sloshing influence from more tanks on global responses of liquid cargo carrying ships in waves need
further investigation by experimental and numerical methods.
Mitra et al. (2012) developed a fully coupled model of nonlinear sloshing and ship motion. The fully
nonlinear sloshing motion was studied using a FE method, potential flow, and the nonlinear ship motion
was simulated using a hybrid marine control system. Simulated roll motions were compared with existing
results in waves. Fair agreement indicated that the proposed model was expected to be useful in
evaluating the coupling effect. In their study, the coupling model also allowed for the effects of wind,
wave and current. It was found that the wave height can have large effects on sloshing, whereas the
current velocities caused fairly marginal effects on the ship motions.
Tsai et al. (2013) carried out a numerical seakeeping-sloshing coupling analysis for an 8,000 TEU
container ship with some cargo holds partially ballasted, with sea water, to simulate a full load condition,
to design draft, required in sea trials. This procedure is unusual compared to the conventional ballast
condition in sea trials; the risk induced by resonance of ship motion and sloshing water in cargo holds has
to be checked in the design stage. A frequency domain model is utilized in the hydrodynamic numerical
code HydroSTAR considering high speed effect, and the analysis includes nonlinear seakeeping, linear
sloshing and finally the calculation of the coupled equations of motion. The additional resonant
frequencies, especially for the rolling motions, induced by the motion-sloshing coupling effect are
clarified and their corresponding operational scatter diagrams, for heading direction change based on
specific sea state, are provided. The structural strength of the ballasted cargo holds are to be assessed for
those critical rolling motions through a fully consistent quasi-static method by a hydro-structure
interaction numerical code HOMER, requiring hydro mesh and FE structure models. These results
confirmed that the induced resonant frequency will occur with full load condition and full ship speed,
especially for rolling motion, and may cause serious structural problems.
Recently, Zhao et al. (2014b) developed a 2D coupled model which brings together the nonlinear
sloshing flows and linear ship motions, both based on potential theory, in time domain. Experiments were
conducted to validate the numerical model. An FLNG section comprising an outer rectangular hull and an
inner rectangular tank was used in this study. The filling level was 20% in the tank. The hull was freely
floating in the wave tank subject to beam waves. It is similar to past experiments of a barge with a tank on
top. Good agreement was achieved for all cases presented. Based on the verified model, the coupling
between ship sway motion and internal sloshing was investigated. It was observed that the magnitude of
the internal sloshing is nonlinear with the incident wave height, both the internal sloshing and the ship
motion exhibited amplitude modulation phenomena and that the sloshing reduction effects on the global
motions at the natural frequency are caused by the fact that sloshing force was behind the wave force.
Kim et al. (2014) developed the multi-liquid moving particle semi-implicit (MPS) method. In
particular, a robust self-buoyancy model and surface tension model were developed. They also have a
new interface searching method for tracing interface particles in a reliable manner. The developed multiliquid MPS method is validated by comparison with three liquid-sloshing experiments by Molin et al.
(2012). The interfacial natural frequencies are simulated by a free decay test using the MPS program and
the results are compared with linear potential theory. The simulated results agree well against both
experimental and theoretical values. The authors observed that the interfacial elevations can significantly
increase when the excitation frequency is close to the natural frequencies of the respective interfaces.
Subsequently the verified multi-liquid MPS program is coupled to a vessel motion program in time
domain. At each time step, the excitations induced by the inner multi-liquid-tank sloshing are fed into the
vessel motion simulation and the resulting vessel motion is again fed back into the MPS module. For the
case of multiple liquid layers, there exist more than one sloshing natural frequency; hence, the relevant
physics can be much more complex compared to the single-liquid-tank case. From the numerical
simulations, the authors noted that the wash tank operation in head wave condition is much more robust
than that in beam wave condition. The vessel motions with the three-liquid cargo and the corresponding
solid cargo were also compared, and it was observed that the wash tank with liquid cargo can also
function as a beneficial anti-rolling device. Based on these comparisons it was concluded that the simple
approximation using equivalent rigid cargo and ignoring coupling with liquid sloshing can be unreliable
in the design stage. Since maximum vessel motions and maximum interface elevations can be checked
96
directly, the developed vessel motion/multi-liquid-sloshing coupled program can be used as a reliable
prediction tool for the safety and operability of a floating system with large multi-liquid tanks.
Paik et al. (2014) presented a new method for determining the design sloshing loads, peak pressure and
impulse, for FLNG. They also showed an example to validate the method. Thirty scenarios were selected
using the Latin hypercube sampling method and a series of numerical computations were undertaken on
FLNG cargo containment systems. Based on the results obtained from the study, they conclude that their
method is validated using applied examples; hence, feasible for application to FLNGs. The sloshing
probability calculated in the study is based on a limited amount of motion data. Thus, if field
measurement data were to be used with this method, the accuracy of the sloshing loads would be more
reliable and robust.
3.3
Green water
As the wave breaks and overtops the structure, the flow becomes multi-phased and chaotic as a large
aerated region is formed in the flow in the vicinity of the structure while water runs up on to the structure.
Thus, the so called green water is generated, which may cause significant damage to facilities and
equipment on the deck of the platform. In addition, green water can have strong impact on the stability of
marine structures depending on the amount of inundation and global momentum transfer. Green water
loads occur when an incoming wave significantly exceeds the free board and water rushes onto the deck.
Thus, any increase in the frequency of occurrence and severity of hurricanes implies that structures in the
ocean are at higher risk of exposure to extreme waves and green water loads.
Determination of green water loads can be categorized into three methods as approximate, numerical and
experimental (ISSC, 2012a). Approximate methods are simple and fast but less accurate and can be used at
early design stage. Numerical methods are suggested for more accuracy. Although in most numerical studies
conditions are more artificial. The major problem with this approach is excessive computational time.
Furthermore, it is not yet possible to obtain realistic results without reflection from the boundaries of the
domain. Experiments for green water events are still being carried out today. The objective of such
experiments can either be to determine extreme loads or it can be for the validation of numerical methods.
Although it is realized that scaled experiments violate scaling laws with respect to the effect of air entrapment,
it is still today the best, if not the only, method to arrive at a statistical distribution of extreme loads.
Despite the difficulties associated with analytical research, work in this area continued. Masoud (2013)
proposed theoretical calculation of 2D nonlinear wave loads on a horizontal deck of a coastal structure
located in water of finite depth, based on the Green-Naghdi theory of water waves or Eulers equations. It
appears that the girders do not have any influence on the vertical force, and only a small influence on the
horizontal force. The effect of formation of air pockets between the girders, in a model of an elevated
bridge deck, was studied by adding air pressure relief openings to the deck of the structure. It was found
that the entrapment of air pockets increases the vertical uplift force significantly. Results were compared
with available laboratory experiments and linear solutions showing a close agreement. This case can be
applied to ships with zero forward speed, e.g. moored.
Experimentally, the results of Lee et al. (2012b) for three different FPSO bow shapes in regular head
waves were analyzed and compared to each other. Based on these results, a database for CFD code
validation was built, such as peak pressure, and some design considerations were proposed. The
experiment also shows that pressure peaks were closely correlated with the incoming wave amplitude for
each wave length, which can be interpreted as a quadratic relationship between the pressure and velocity
of the flooded water on deck. Ariyarathne et al. (2012) examined impact pressure on the deck area of a
3D model structure in a laboratory and related the impact pressure to the measured velocity as well as the
void fraction on the deck. Xiao et al. (2014) carried out an experimental study to investigate the wave
run-up and green water along the broadside of the 300000 DWT FPSO with a tower yoke mooring system
in shallow water, considering non-collinear environments and different water depths. The impact of the
motions and the gap between floater and seabed on the relative wave elevations and wave run-up is
examined in details. It is shown that the impact of the gap is evident on all the wave elevations along the
broadside. Moreover, the impact of vertical motions on the wave elevations along the broadside is
relatively small and almost negligible for a large FPSO exposed to extreme waves in shallow water.
Buchner and van den Berg (2013) studied experimentally the nonlinear wave reflection along the side of
ships which can lead to green water on deck for moored and sailing ships, with a simplified test setup of a
thin plate at an angle to the wave direction. The observed processes with plunging and spilling breakers
close to the plate made clear that linear or second order models will not be able to predict this behavior
accurately. It was concluded that higher order methods, that could include wave breaking, or CFD are
necessary for a prediction of these important effects. These model tests can be used as important
validation material. The experimental and numerical investigations by Greco and Lugni (2012) and Greco
97
et al. (2012), which combine seakeeping and green water analyses, have already been discussed in
Section 2.2.
Lu et al. (2012) developed a numerical time domain simulation model with VOF technique to study the
green water phenomena and its impact loading on structures. Numerical simulations of green water
problems carried out in this study include: (i) green water overtopping a fixed 2D deck, (ii) green water
impact on a fixed 3D body without or with a vertical wall on the deck and (iii) green water impact on the
deck and deckhouse of a moving FPSO model. Numerical results are compared with experimental
measurements for both the water height on the deck and the impact pressure on the front wall of the
deckhouse for each case, and show fairly good agreement. Xu (2013) examined the case of an extreme
plunging wave impinging onto a horizontal deck, simulated by the enhanced, incompressible, SPH
method. A first order density smoothing procedure and a density diffusion term in the continuity equation
were introduced to smooth the pressure field. A series of simulations were carried out to verify the
stability and accuracy of the numerical model. These include a hydrostatic simulation, a bubble rise
problem and a dam break simulation. The benchmark studies showed that the enhanced SPH model
developed is able to capture details of complex wave-structure interactions including the flip-through
phenomenon. The simulations clearly show how the bifurcation of the impinging crest front could lead to
the formation of a strong jet (tongue in 3D) that eventually collapses onto the topside of the deck. The
remaining portion propagates forward under the deck, leading to higher pressures on the underside of the
deck. These results contribute to shedding more light on the dynamics of green water on the topside of a
deck in extreme waves. Zhao et al. (2014e) aimed their investigation on assessing the influence of the
DoF of a floating body on the maximum impact pressure due to green water on deck. In experiments the
DoF can be adjusted by a heaving rod and a restrained spring connecting the carriage and the guide rail.
Their numerical method combines a 2D CIP model and VOF-type tangent of hyperbola for interface
capturing/slope weighting (THINC/SW) approach. Experimental water surface elevations, body motions
and impact pressure were compared with the numerical results for different DoF cases, and the agreement
is satisfactory. Green water on deck and impact on the deckhouse is generated by the impingement of a
focused wave group on the floating structure. The results show that the peak impact pressure due to green
water decreases rapidly with increasing DoF.
Kim et al. (2013d) introduced hybrid CFD method for modelling and simulation of green water load on
deck. Their method involves a combined approach of linear seakeeping and nonlinear CFD analysis, both
using 3D modelling of the hull. Two large containerships were considered for analysis, for a range of
headings. Damage cases of container carrier in way of breakwater and main supporting members were
studied. Relative bow motion is considered as the dominant load parameter (DLP). An equivalent design
wave for this DLP is used for CFD simulations, modelled as 5th order Stokes wave. Analysis procedure
for the direct calculations of design pressure on breakwaters using advanced CFD method is provided in
this paper. Newly proposed coefficients for design pressure are compared with IACS UR S3 and found to
be reasonable, though on the higher side.
Ruggeri et al. (2013) evaluated numerically green water events in a box-shaped FPSO structure in
regular beam waves for a captive model. The potential flow method WAMIT was used for quick checks,
whereas CFD software StarCCM+ was used for detailed flow study. 2D and 3D approaches were
modelled in CFD. Free surface elevations amidships were compared with experimental results. It is
concluded that the 2D approach can be used only for few relatively long waves, whereas for the entire
wave range, 0.75 LPP to 1.5 LPP in this paper, the 3D approach is recommended.
Joga et al. (2014) extended application of CFD methodology to simulation of green water to find the
rate of water ingress into open holds of a container vessel. Two CFD solvers, StarCCM+ and ANSYSCFX, were used to model the full scale ship in waves and the effect of water ingress. Computations were
carried out only for beam waves and zero forward speed. Various heading and speed combinations were
not considered for comparison with experiments due to higher computational requirement. Significant
differences between the water ingress predictions of the two solvers and the experimental results were
observed, the latter extrapolated from model scale. STARCCM+ results underpredict the mass flow rate,
though relatively close to experimental rates. On the other hand, the mass flow rate predicted by ANSYS
CFX is much higher compared to the experiments. However, vessel motions computed using both solvers
were in line with experimental results. It is concluded that the differences in water ingress predictions are
mainly due to free surface capturing by the VOF method and much refined mesh requirement in the
region of open holds of the vessel.
Pakozdi et al. (2014) investigated the viability of a simplified coupled method between a potential
theory based green water engineering tool, Kinema3, and the commercial CFD tool Star-CCM+. Results
from a case study application on a large FPSO are validated against model test data. The case study
contains analyses of the FPSO in long-crested regular and irregular waves, both in fixed and moored
conditions. Three different headings are included. The approach for modelling green water events uses a
98
Finite Volume VOF method with a complex velocity inlet boundary condition. Thus, the Kinema3
engineering tool is used to generate simplified spatio-temporal inlet conditions from the relative wave
elevation and wave kinematics at the bulwark, based on linear potential theory combined with nonlinear
random wave kinematics. The VOF method is then used to model the detailed flow on deck, including
impact forces on deck structures; that is to say, that only the deck is modelled in STARCCM+. Kinema3
can also generate simplified estimates for the peak water height, velocity and impact force values,
assuming an extended dam-break approach together with a simplified, local 2D deck layout. Comparisons
with the Kinema3-STARCCM+ results show an overall fairly good agreement, although flow details on
deck can, of course, not be expected to be modeled that well. Comparisons of both aforementioned sets of
results to model test data show good agreement for the relative wave height, water height and impact
force level, in regular and irregular waves. Detailed time histories, including force rise-time, from the
coupled Kinema3-Star-CCM+ CFD simulation analysis are quite similar to the measured ones. The CPU
time consumption for the coupled simulation is moderate, compared to what would have been for a full
CFD simulation. Hence, the simplicity of the simulation setup, the quality of the results and the CPU time
efficiency makes this method a viable candidate for industrial use.
Considering the new offshore frontiers for oil exploration and production, in particular the Santos
Basin region, FPSOs will be exposed to more severe wave conditions. This scenario requires more careful
analysis with reference to the green water phenomenon. The complex physics involved in the water-ondeck flow implies several uncertainties regarding green water load analysis. Carvalho and Rossi (2014)
proposed a methodology, taking into account model tests, CFD simulations and analytical formulations,
to estimate the green water loads considering the water elevation above deck measured from experiments
or numerical tools. In order to accomplish this objective, CFD simulations with different solvers were run
for a benchmark case, showing that it is a suitable approach for a global result in dam-break cases. The
term global is used with reference to the pressure computed on block due to the dam break. Other results
include wave elevation at various locations. The CFD codes used were: ANSYS CFX, ANSYS Fluent,
StarCCM+, Edge CFD and Petrodem MPS. The exploratory CFD benchmark case results showed that
this technique is appropriate for complex free surface flows. Comparisons were carried out with
experimental measurements by Kleefsman (2005). Although a more comprehensive and systematic
evaluation should be done with respect to the differences between experiments and simulations, the
simulation results reproduced the most critical pressure time series with approximately 5% deviations.
Furthermore, CFD simulations using ANSYS CFX were carried out to simulate green water events for a
typical Santos basin FPSO, together with model tests. Using a combined approach with experiments, a
special boundary condition was calibrated to reproduce critical green water events for beam sea condition.
In this special boundary condition the inlet surface boundary condition was switched to an open boundary
at a particular time. This was done so as to avoid water accumulation in the deck area. Such calibration of
the model allows for a detailed flow analysis. Considering the promising results from the benchmark case,
this strategy can be used effectively in determining green water loads. Based on a critical event, it was
found that the classical dam-break approximation for green water analysis can be very useful for
estimating loads on structures, even in complex cases where appendages, in this case the riser balcony,
have a clear influence on the flow. In order to extend this formulation to an even wider range of
situations, the correction for objects having a vertical gap with respect to the platform deck was verified
against CFD simulations, showing good agreement and including a safe margin. The new formulation is
then combined with the local water elevation measurements, resulting in a green water loads evaluation
procedure. The load reduction of vane type protection barriers is addressed through CFD simulations and,
after a parametric study, the effectiveness of each barrier configuration is estimated. For the most efficient
protection arrangement, the final load was one third of the original load.
Schiller et al. (2014b) addressed a significant green water problem in the Santos Basin using Kinema3,
which is a simplified, yet robust, prediction tool of green water and wave impact on FPSOs in steep
nonlinear irregular waves. Given the multi-directional wave conditions in the Santos Basin, the authors
expanded this tools capabilities to consider arbitrary wave directions. Their study case was the FPSO-BR
model provided by Petrobras. Predictions were validated against results from model tests performed at
Marintek. A good agreement between the Kinema3 green water predictions and the model test records,
especially for head seas was observed. For quarter-beam and beam seas, a satisfactory agreement was
obtained with an adjustment of the empirical factor ef, which is a small non-linear contribution to the
relative wave height that is related to the amplification of the wave elevation due to interactions with the
vessel. For quarter-beam and beam seas, a lower ef factor is necessary, which can be interpreted as wave
dissipation processes. For beam seas, the low value of ef is also partially due to the larger importance of
roll motion, which may have a mean value and a low frequency component in reality. A progressive
decrease of ef with increasing heading angle was observed. The authors conclude that additional
verifications with CFD tools and model tests are necessary in order to clarify the relationship between ef,
99
wave heading and the nonlinear wave-structure interaction. The authors also showed that the application
of Kinema3 to arbitrary wave directions was a successful step in tackling new challenges associated with
the green water problem. One major challenge in the Santos Basin is the bi-directionality of sea states.
Although wave impact from the largest waves, coming from S-SW, may be minimized by heading the
vessel towards S-SW, the analysis demonstrated that the vessel may be subject to significant green water
from E-SE beam seas. The authors showed that beam seas produced the largest green water problem with
respect to occurrence, intensity and FPSO deck area affected by water impact.
3.4
Experimental and full scale measurements
In the last few decades, the investigation of complex FSI problems related to slamming phenomena has
motivated not only the need for reliable simulation, but also more challenging objectives for experiments
in model basins and for sea trials. In this section, therefore, recent research aimed at studying waveinduced high frequency response by means of experimental tests and full scale observation is discussed.
Measurements related to sloshing and green water are reviewed in Sections 3.2 and 3.3, respectively.
Lavroff et al. (2013) carried out an extensive experimental investigation of wet-deck slamming in
regular head waves using an elastically scaled model of a wave-piercer high speed catamaran. The
segmented model consisted of three parts with the forward part comprising the forward demi-hulls and
the center bow as a separate segment. The sophisticated design of the catamaran model allowed
measuring the slamming vertical force acting on the center bow and the vertical bending moment along
the demi-hulls in order to identify the critical conditions in the operational envelope of the vessel.
Slamming occurred in a range of the encounter frequency that was found to be relatively larger in the case
of the higher regular wave amplitudes and the induced loads were approximately proportional to the wave
amplitude squared. The longitudinal position of the slam impact on the center bow showed a small but
significant dependence on the wave height and frequency. Wu et al. (2012b) discussed the capability of
representing the bending full scale behavior with segmented models made of rigid segments and flexible
joints. Several aspects, such as the number of segments and the stiffness distribution, are considered for
accomplishing the hydroelastic scaling of a 13000 TEU containership. The authors found that 3-joint
segmented structural model provides a sufficiently accurate representation of the real ship, if the
hydroelastic effects depend mainly on the first vertical 2-node flexible mode. These hydroelastic effects
in the vertical bending moment appeared not to be sensitive to the stiffness distribution among the
flexible joints in a 3-joint segmented model. Hence, the authors concluded that the same joint stiffness
can be applied when adjusting the model to achieve the targeted natural frequency of the first flexible
mode. Zhu and Moan ((2013), (2014)) focused on the wave-induced nonlinear effects which appear in
VBM response of ULCS. Zhu and Moan (2013) focused on the explanation of the differences between
sagging and hogging VBM extreme values. Amongst well known sources of asymmetry, like
nonlinearities in the hydrodynamic forces and ship hull geometry, in particular, they considered the
nonlinear contribution to the overall loading due to the pressure field acting on the bow which affects
mainly the hogging VBM. Two container ships, 8600 TEU and 13000 TEU, were considered with
segmented model tests in regular and irregular head waves. Regular waves allowed the authors to clarify
the role of super harmonics in increasing or decreasing the hogging and sagging peaks. High, but rather
unrealistic, speeds for such ships in rough seas had a significant impact on the VBM asymmetry through
higher order harmonics and high frequency vibrations mainly more in sagging than hogging. Zhu and
Moan (2014) investigated the effect of regular and irregular oblique waves on extreme values of VBM for
the case of a 13000 TEU ULCS. The authors found that the extreme values for both sagging and hogging
were higher in oblique waves than in head waves. In particular, the second harmonics of the amidships
VBM affected the peaks in regular waves to an extent which showed to be dependent on the heading
angle.
Dessi and Ciappi (2013) investigated the slamming occurrence for a ferry cruising at high speeds with
an elastic segmented model tested in irregular long-crested waves. They observed that the tendency of the
slams grouping into sequences made of two or more events (clustering phenomenon) violated the
hypothesis of their statistical independence. Sequences of closely occurring impacts were found to
produce higher or lower global responses depending on the phasing with ship-beam oscillation excited by
the previous impact(s). The combination of high speeds and high sea states was found to play a relevant
role in inducing the slam clustering. However, a direct correlation between each slamming event and the
encountered wave height, or even between the slamming severity and the impact velocity, was not always
present. Time separation of slams inside a sequence was approximately equal to the natural pitching
period of the scaled ship. A new criterion based on the wavelet analysis of the amidships VBM was then
introduced to identify the occurrence of a slamming event, apart from the usual distinction between
bottom or bow-flare slamming events. A slightly different version using filtering and Hilbert transforms
for the analysis of transients is also presented by Dessi (2014a). Jacobi et al. (2014) applied a similar
100
criterion to the identification of slamming events during full scale trials on a wave-piercer high speed
catamaran. This criterion is based on the definition of a stress rate threshold, first proposed by Thomas
et al. (2003). The choice of the strain gauge location proved to be critical for the correct application of the
proposed slamming criterion: the strain gauge located in the centrebow archway and mounted on a
stiffener cut out at frame 64 was at the end selected, as it clearly presented vibrations uniquely associated
to slamming. To avoid false positive slamming events, the analysis of the stress time history needs to be
combined with a second criterion detecting the downwards decelerating bow motion. Using this criterion,
the number of slams per hour with a certain intensity can be determined and their dependence on speed,
wave height and heading analyzed. The influence of the ride control system in alleviating the slamming
stresses was found to be more relevant at high speeds. The existence of an overall trend, with a weak
association, between the relative vertical velocities and the stresses occurring after the slam impacts was
noted. However, the association is so weak that it cannot be used as a primary indicator of slam
occurrence and magnitude, similarly to the case of fast monuhulls as shown by Dessi and Ciappi (2013).
Andersen and Juncher Jensen (2014) investigated the effect of hydroelastic high frequency vibration on
the extreme hogging wave bending recorded onboard a 9400 TEU container carrier. The stress
measurements amidships were obtained using two long-base strain gauges mounted in the passageways,
port and starboard sides, just below deck approximately amidships. The wave field close to the ship could
be estimated with the onboard wave radar and the rigid body response was recorded from an inertial
motion unit. They found that the 2-node mode is, by and large, responsible for the ship vibrations even in
quartering seas, with no evident torsional contribution, and that the hydroelastic contributions may double
the stress levels at some instants. Thus, their analysis brought to attention the possibility that predicted
total wave hogging VBM, accounting also for whipping loads, may exceed slightly the design stress
values for that ship in extreme but realistic sea states. The combination of steady state wave loading with
loads associated to whipping and springing is also analyzed by Barhoumi and Storhaug (2013) in the case
of 8600 TEU container vessel, equipped with a comprehensive onboard hull monitoring system with 20
strain sensors for global and local hull response and monitored for approximately 4 years. On the basis of
the strain data collected at various ship sections, the authors found that the contribution of whipping to the
reduction of fatigue life and to extreme loading is important in particular zones of the ship structure.
Though fatigue appeared as a less relevant issue for the east Asia routes compared to the north Atlantic
ones, the effect of voluntary speed reduction and routing on the structural loads continues to be significant
for this type of vessels.
Iijima et al. (2014) carried out an experimental investigation on the post-ultimate strength collapse
behavior of a ships hull girder under whipping loads. The 1/100 scaled model comprises two rigid
bodies, a connecting deck hinge and a sacrificial specimen amidships which bent under the relative
rotation of the two segments. The mechanical properties of the model and, in particular, the ultimate
bending strength follow the similarity law proposed by Wada et al. (2010). The tests were conducted
keeping the model afloat on the water and the whipping loads were artificially reproduced by dropping an
object onto the ship section. It was observed that the sacrificial specimen collapsed less when the
whipping loads had a shorter duration, indicating that the collapse may not occur to a large extent under
impact loads. An original experimental approach of 3D water impact at constant speed was presented by
El Malki Alaoui et al. (2012). The hydrodynamic force acting on axisymmetric rigid bodies striking a
horizontal liquid surface at constant vertical speed was investigated using a hydraulic shock machine,
which allows carrying out impacts at high speeds with small deviations in the velocity. Comparisons
between theoretical model, numerical results, also discussed by Tassin et al. (2012b), and available
experimental measurements are reported in this paper showing an acceptable agreement.
Dessi (2014b) presented a technique based on a combination of proper orthogonal decomposition and
spline approximation for reconstructing the unsteady distribution of sectional loads due to waves along a
ship hull. To validate the proposed technique, the hydrodynamic lumped forces acting over hull segments
were initially made available by calculating numerically the sectional loads and then spatially integrating
these loads over the segment sections. The set of time-varying segment forces constituted the input data
from which the original sectional load distribution was successfully reconstructed. The robustness of this
technique was tested against the presence of noise in the input data. This approach was used by Dessi
(2013a) for obtaining the experimental slamming load distribution in regular wave tests with a segmented
hull. The time history of the forces acting on each segment was measured with load cells and the sectional
slamming force at each time step was extracted on the hull sections subjected to water impacts. Thus, a
greater detail in the spatial load distribution can be obtained from experiments with segmented models
and structural ship design can benefit from reliable load data with improved resolution.
Several papers dealt with signal processing techniques for highlighting some features of the ship
vibratory response. Kim et al. (2013e) proposed the use of the cross correlation between the wavelet
transforms of the ship response and an ideal and tailored IRF to extract the whipping part of the response.
101
The authors successfully applied this technique to a realistic signal obtained from the WISH-Flex
simulation code. The analysis of the ship vibratory response was also considered by Mariani and Dessi
(2012) for identifying the wet operational modes in terms of shape, frequency and relative energy
associated with each elastic mode, with proper orthogonal decomposition which does not require the
measurement of the exciting force. They compared seakeeping tests of an elastically scaled model in head
waves for different sea states, with/without slamming as the case may be, highlighting how energy is
captured mainly by the 2-node vertical bending mode. Damping estimation of the experimental whipping
mode using the random decrement technique (RDT) or the autocorrelation function (ACF) was
considered in a successive paper by Dessi (2013b), using the same dataset used by Mariani and Dessi
(2012). The damping values were found to be different depending on the particular signal processing
technique, RDT or ACF, used and on the choice of some parameters within the same technique. The
random decrement technique in combination with the wavelet transform was also applied by Kim and
Park (2014) to data obtained from elastic model tests for damping identification of the wet bending
modes. The authors explained the differences in modal damping estimation between still water conditions
and forward speed in waves with the presence of viscous effects.
Ikeda and Judge (2014) carried out a series of tow tank experiments performed on a 1.2 m model hull
in regular waves. Pressures on the bottom of the hull were measured on one side with traditional pointpressure sensors and on the other side with pressure mapping pads. The goal of this research is to study
the fundamental physics of the water impact of high speed planning hulls and to measure the slamming
loads and resulting motions of the craft upon reentry into the water after becoming partially airborne. A
set of towed model experiments was conducted in calm water, regular waves and irregular waves to
capture a sequence of individual impact events. The pressure signals from the pressure pads (providing
both spatial and temporal resolution) and the point-pressure measurements (high temporal resolution)
have been presented for individual slam events, allowing a deterministic approach to investigating high
speed planing craft wave slamming.
3.5
Loads due to damage following collision/grounding
One area that has become of great concern to the design and operation of ships is that of accidental
damage. Accidental damage to ships and, subsequent, flooding can occur in a number of ways, but
generally damages due to collision and grounding are of primary concern, particularly for ships at high
risk with respect to the loss of lives or pollution of the marine environment. Naturally, events of nonaccidental structural failures leading to ships loss are also of great concern for some ship types, e.g. the
loss of containership MOL Comfort in June 2013. The aim of this section of the report is to focus on the
load implications on the ship after the event, mainly on loads of hydrodynamic origin and not to replicate
part of the work undertaken by Committee V.1 Accidental limit states.
The IACS Harmonized Common Structural Rules (CSR-H) (IACS, 2013) are aimed at checking the
hull girder ultimate bending capacity in the damaged state for the seagoing condition to ensure that it
satisfies the residual strength checking criteria. Accordingly the VBM, MD, in hogging and sagging
conditions, to be considered in the ultimate strength check of the hull girder in the damaged state is
obtained as:
M D SD M SW D wD M wv
(1)
where MSW-D is the permissible still water bending moment, Mwv is the rule vertical wave bending moment
(VWBM), SD = 1.1 is the partial safety factor for the still water bending moment SWBM) in the damaged
condition and WD = 0.67 is the partial safety factor for the VWBM in the damaged condition. The general
characteristics of global loads on damaged ships are apparent from the partial safety factors included in
Eq.(1). The SWBM may increase with respect to permissible SWBM in intact condition, whilst the rule
VWBM is considerably reduced. The most dangerous situation is the flooding of ballast compartments in
the amidships region, causing increase of sagging SWBM in the vicinity of the damage. Downes et al.
(2007) performed a case study on an Aframax tanker showing that in the full load condition 10% of the
cases lead to an increase in sagging SWBM of 25% or more of the allowable SWBM. The literature
review of increases of SWBM used in structural reliability assessments of damaged oil tankers is
presented by Buri et al. (2012), where they found that SWBM in sagging may in some cases increase by
a factor of two compared to the intact SWBM.
Differences between wave load effects employed for intact and damaged ships arise because of the
effect of flooding on the calculation of hydrodynamic loads, as well as from different environmental
conditions and exposure times to be taken into account. Downes et al. (2007) suggested two practical
approaches for calculation of hydrodynamic loads on a damaged ship. The first is the added mass
approach where the seawater which floods into the vessel is assumed to become part of the vessels mass
and moves with the vessel. For calculating the hydrodynamic forces, the damage opening is assumed to
102
have negligible impact on the overall hydrodynamic properties of the hull. This approach should be
accurate for damage extents which are small compared to the size of the tanks which are breached. The
second approach is applicable for larger breaches, where the damaged tank and all of its mass are
removed from the vessel, as well as its surface area from the hydrodynamic model. The hydrodynamic
interaction between the waves and the structure of the opening needs to be modelled.
Reduced exposure time to environmental conditions after damage should be considered before the
ships is taken to a safe location. Prestileo et al. (2013) modelled the damaged ship by the added mass
approach with 24 hour of exposure to a sea with a heading of 210O and a speed of 5 knots. Reduced
exposure time is counteracted by the hypothesis that the damage has occurred because of adverse sea
conditions, so the sea state envisaged after the accident is not a normal but a severe one. The truncated
diagram was used by removing half of the complete scatter diagram characterized by lower wave height,
less than 8.5 m, and keeping only the most severe sea states. The aforementioned choice of truncation and
exposure time yield results in terms of average extreme VWBMs which are approximately 80% of the
corresponding value for the intact ship in her entire life.
Lee et al. (2012c) applied a computational tool based on a 2D linear method to predict the
hydrodynamic loads for a damaged warship. They obtained larger VWBM for damaged condition,
compared to the intact ship. The global dynamic wave-induced loads calculated using 2D linear method
are also compared to measurements. In head and stern quartering waves, the differences between the
computations and measurements of global dynamic wave-induced load response amplitudes are
reasonable. In general, however, linear strip theory overestimates measurements for both intact and
damaged ship conditions.
Stettler and Thomas (2012) presented a detailed flooding and structural engineering study to accurately
model, simulate and evaluate the progressive flooding, sinking and structural failure of the historic
Titanic following her collision with icebergs. Detailed computer flooding analysis models were
developed, and novel techniques for dynamic flooding simulation using the commercial software
GHSTM were developed and implemented. The resulting loads of these detailed flooding simulations
were applied to structural FE models using MAESTROTM for conducting detailed stress and hull failure
analyses. The technical documentation of the developed computer models, including sources of
information, assumptions, conventions and methods, as well as the technical discussion of simulation
results and the lessons learned are applicable to similar marine forensic analyses.
Spanos and Papanikolaou (2012) investigated the time dependence of survivability/floatability and
stability of ROPAX vessels, when sustaining side collision damage in beam irregular waves using a 3D
and 6 DoF nonlinear time domain simulation method for ships motions and flooding, where the flood
water mass is time-varying and oscillating in a coupled way with the ship motions. Conducted research
confirms that ROPAX ships characteristically capsize fast, when sustaining damage leading to capsizing.
A probabilistic analysis of the survival time after collision damage reveals that even for the most generic
damage conditions assumed, the survival time in the case of capsizing remains short, which is
characteristic of this type of ship design, exposing the typically large undivided deck to flooding in higher
waves.
Dankowski and Krger (2013) presented a progressive flooding method in still water which is based on
the calculated flux between the compartments using a modified Bernoulli equation. Large and partly
flooded openings are taken into account as well as optional air compression and flooding through
completely filled rooms. The method uses a typical damage opening based on the generated damage
cubes by a Monte Carlo simulation to perform a direct progressive flooding assessment for each critical
intermediate case. The combination of flooding calculations with a Monte Carlo method extends the
classical damage stability calculations to the time domain, which allows a more accurate estimation of the
overall safety level of a ship to withstand damage. Furthermore, this method is very useful at the early
design stage to identify critical intermediate stages of flooding. For the examples shown, a sudden capsize
during the flooding process is observed after the vertical centre of gravity is increased by a specific
threshold.
3.6
Weather routing and operational guidance
Most research with the aim of assisting the master in weather routing is focused on logistic optimization
by formulating the problem of minimum time or minimum fuel consumption with constraints. Some
efforts are aimed at weather routing taking into account seakeeping behaviour, loads and related measures
and the consequences of weather routing on these parameters. These form the subjects reviewed in this
section.
The research on weather routing can be divided into four types: (i) research with respect to a suitable
and efficient optimization scheme, (ii) simplification of complex hazards in a seaway intended for on-line
onboard operational guidance, (iii) monitoring of relevant measures onboard and (iv) a combination of the
103
former three types to form an applicable onboard decision support taking into account the seakeeping
behaviour and loads on a vessel.
The current state of the art in weather routing or operational guidance has not changed since the last
report of this committee (ISSC, 2012b) The seakeeping characteristics of a vessel are considered through
the RAOs. In combination with a standard wave spectrum and integrated sea state parameters, such as
significant wave height and average zero-crossing period, the linear spectral analysis predicts the
response of the vessel for all risk control options. A review of the current state of the art is published by
Papanikolaou et al. (2014), where the importance of including uncertainties in the analysis and the
consequential need for an efficient probabilistic method are emphasized. A simulation result is presented
by Delitala et al. (2010). A route optimization software was applied to the output of a limited area
weather model and to a wave model. They concluded that weather routing would improve ship
performance and would support ship masters virtually.
A number of papers are devoted to the development of optimum path finding systems with the use of
existing environmental data and mathematical models for ship response. Voluntary or involuntary speed
reduction is taken into account in some of them; however, little progress can be found with regard to the
load estimation. Lee et al. (2011) propose a multi-resolution planning method, which appears to be
suitable for including wave effects as well. Panigrahi et al. (2012) advocate the use of Dijkstras
algorithm for a better handling of the voluntary speed loss in the optimization procedure. By applying a
Bzier curve for the description of a ships route, Ishii et al. (2010) demonstrate that only a few variables
are needed to generate an optimum route. Though the model has only been tested to minimize transit
time, it is possible to include ship responses as well. Lin et al. (2013) propose a 3D modified isochrome
method to be applied for weather routing which is able to address safety concerns. Parametric rolling is
taken into account as a risk by Maki et al. (2011), where the real-coded genetic algorithm technique is
applied to globally search for the optimum route taking into account of the risk of parametric rolling as
one of its objective functions. They concluded that it is easy to obtain a route that reduces parametric
rolling probability using the proposed method.
More detailed investigations were carried out with reference to the load/response estimation as part of
the routing simulation. A decision-analytic approach is introduced by Nichols et al. (2014) illustrating this
approach using a detailed example of a ship in transit, where the goal is to minimize both transit time and
probability of structural failure. Assuming that the structural integrity of the ship is governed by the
aluminium stringers used in certain naval vessels for hull support, a simple model was used for predicting
future damage states and estimating the probability of component failure by means of the spectral
approach together with slamming impact force. The main conclusion is that a variable speed strategy
compared to a fixed speed strategy is superior and the approach can be made adaptive to incorporate new
information such as updated weather predictions. Dec and Frangopol (2013) set up a risk framework
incorporating strength elements. In this work the midship section capacity is modelled using FE method
including corrosion as a function of time and annual corrosion rate. The load effects are evaluated using
strip theory. The dynamic VBMs are accounted for by general approximations based on L PP and ships
breadth. The framework is applied to the Joint High-Speed Sealift travelling between two points through
a selected stormy sea weather map. In the case study the direct risk and reliability are precalculated such
that the optimization process takes about 30 minutes. Three representative solutions with different
normalized direct risk show that in relatively calm weather the optimum is the direct route, but in harsher
weather the optimum route avoids the storm. Mao et al. (2010) presented a simplified fatigue model
which can be used in ship routing. The simplified model is based on linear stress transfer functions, while
whipping is considered by introducing a modified mean stress upcrossing frequency (fz) based on the
wave encounter frequency. Measurements on board a 2800 TEU container vessel showed that f z is about
50% underestimated when using a linear approach. The modification of fz is aimed at reducing the
underestimation but accepting that the variability of the encounter frequency is transferred to the modified
fz. The comparison between the empirical distribution of the observed rainfall damage and the cumulated
distribution of fatigue damage using the simplified method shows good agreement. The error of the
accumulated damage based on the simplified method compared to the measurements is of the order of 1020%. Mao et al. (2012) applied this procedure to two case studies, a 2800 TEU container vessel and a
4400 TEU container vessel. For most of the voyages, the highest fatigue damage can be decreased by
50% in the most optimal case.
Song et al. (2013) present an example of operational guidance to help the crew avoid dangerous
situations by precalculations of roll motions using the IRF approach. The results are displayed in
occurrence maps of parametric roll as a function of speed and heading, for a range of GM values and sea
states. The authors address neither how the precalculations are efficiently presented to the crew nor the
required quality of the input parameters, such as sea state and loading condition.
104
4.
OFFSHORE STRUCTURES SPECIALIST TOPICS
The computational methods associated with the FSI of offshore-type structures have already been
reviewed in Section 2.1. This section focuses on two important areas: namely vortex-induced motions and
vibrations, and mooring systems. In addition, making use of the expertise in this Committee, a
background review of lifting operations and floating offshore wind turbines is carried out. There is also a
short section on wave-in-deck loads, due to differences with the green water problem of ships.
4.1
Vortex-induced vibrations (VIV) and Vortex-induced motions (VIM)
VIV and VIM are of concern in most offshore riser and platform designs when exposed to currents. The
load induced by VIV and VIM could result in collisions of risers, fatigue damage of risers and mooring
systems. Over the yeas substantial research work has been conducted to investigate, understand, predict,
and verify VIV and VIM (Sarpkaya, 2004). Much progress has been made both numerically and
experimentally toward the understanding of VIV and VIM. Nevertheless, there are still many challenges,
such as how to deal with shear, how to evaluate single-span model test experiments versus continuous
pipes, how to account for multi-modes and mode interference, how to suppress VIV at any Reynolds
number (Re), large Re gaps between the computable and measurable ranges of VIV etc.
In offshore applications, the industry continues to rely on a semi-empirical method for the assessment
of VIV-induced fatigue damage together with a relatively large safety factor, and rely on model tests for
the motions of a floating unit due to VIM. Clearly there is still a lot of work ahead, not only because
turbulence remains poorly understood but also because simulations are needed at a level that could be
used for industrial design. In offshore engineering, most VIV studies have been focused on risers and
VIM studies for spars, TLPs and deep draft semi-submersibles. This section covers the latest
developments on the prediction, verification and mitigation of VIV and VIM in offshore applications.
4.1.1
VIV
Offshore risers are pipes that normally run from the seabed to a floater transporting oil, gas, water etc.
Risers are susceptible to VIV, especially deep water risers due to their increased length which lowers the
natural frequency values, thereby lowering the magnitude of current required to excite VIV. VIV is
probably the single most important design issue for steel catenary risers for fatigue damage. Analysis
methods currently used in the industry for the prediction of riser VIV mainly include frequency domain
approaches and direct time domain simulations using CFD methods. One of the most important
milestones of VIV frequency domain approach development is the research work in late 80s (Vandiver
and Chung, 1989), and the software SHEAR7 was developed during that period using empirical models
and mode superposition to calculate the motions of the riser. This software has been updated over the
years to cover cylinders without uniform cross section, strakes and varying currents (Resvanis and
Vandiver, 2011). Other popular frequency domain software for VIV analysis on slender marine structures
are VIVANA and VIVA. Most current VIV analysis programs are limited to computing vibrations
orthogonal to the current velocity only. Although riser VIV was once considered a cross-current
vibration, it was found that the riser is actually oscillating in a figure 8-shape with an inline component.
With the right current condition, the inline amplitude could reach the same level as that of the cross-flow
component. Passano et al. (2012) reported the latest developments on VIVANA with its new inline
prediction model.
It has been reported that all these software show a conservative bias in predicting fatigue damage by
comparison with the data obtained from a number of full scale measurement campaigns in the fields.
Tognarelli and Winterstein (2014) stated that software error statistics show markedly different behaviour
in different regimes, e.g. for different levels of predicted damage. They suggested a non-parametric
approach, in which software error statistics are sorted and separately analysed for different predicted
damage levels. In a similar subject, Fontaine et al. (2013) presented a reliability based method which
accounts for uncertainties in S-N behaviour, metocean conditions and software prediction for the
assessment of VIV factor of safety for risers in the Gulf of Mexico.
Schiller et al. (2014a) investigated the effects of current profile, shear and directionality on the
development and excitation of VIV for a deep water tensioned riser by employing a semi-empirical
frequency domain approach. The study found that they all have an important effect on the development of
VIV for the riser model studied. It was also observed in this study that uniform profiles, in particular
those with high velocities, generate the largest VIV response of the riser. The findings are based on
numerical sensitivity studies and there is a need to further understand how complex current profiles in the
offshore regions affect VIV development in comparison to simpler profiles that are recurrent in model test
conditions.
While the frequency domain approach is still the leading method in practical engineering for the
fatigue damage assessments of risers due to VIV, there have been continuous efforts on VIV predictions
105
using CFD approach. For example, Nguyen and Temarel (2014) investigated numerical simulations of
flow past a circular cylinder in uniform cross-flow for a stationary cylinder, as well as forced and free
oscillations of a cylinder using the 2D CFD RANS code ANSYS Fluent, at Re = 10,000. Comparisons
with available experimental measurements and other numerical predictions, show that the essential
features of the FSI are captured in the simulations with a reasonable mesh density. The authors draw
attention to the influence of turbulence modelling on flow characteristics, such as lift and drag
coefficients. Zhao et al. (2013) studied VIV responses of a cylinder in the combined steady and
oscillatory flow by solving the 2D RANS equations, including k- turbulence model, using a PetrovGalerkin FE method. It was reported that the lock-in regime in the combined steady and oscillatory flow
is wider than either that in the pure steady flow or that in the pure oscillatory flow.
Holmes and Constantinides (2014) performed direct numerical simulations (DNS) of a riser VIV with
large buoyancy elements and a general lazy wave shape through the use of efficient meshing strategies
developed. A lazy wave riser is a riser with addition of buoyancy to provide this unique shape to reduce
the dynamic stresses at the touchdown zone. To validate the developed method, comparisons between
model tests and CFD model were made for the time history of displacement and RMS of the displacement
from the mean position at 51 accelerometer locations along the riser. The CFD results show good
agreement with the laboratory experiments. A general approach for building a CFD model with a length
to diameter ratio of 10,000 and a lazy wave shape riser was presented and demonstrated with an example.
Tofa et al. (2014) presented two DoF numerical simulations of the flow-induced vibration of two equal
diameter cylinders in tandem, with varying mass ratio, to investigate the effect of the upstream cylinder's
mass ratio on the vibration of the downstream cylinder. The shear stress transport detached eddy
turbulence model was used for the turbulent flow around the two cylinders. ANSYS-CFX software was
used for the solution of the coupled FSI problem. The motions of the two cylinders were limited to the
translational motions in the horizontal plane whilst other DoFs are fixed. It was reported that the upstream
cylinder's mass ratio has a significant effect on the VIV of the downstream cylinder. This study may
provide useful information on the assessment of potential risk of collision between closely spaced risers.
The numerical results of a single cylinder subjected to two DoF vibration were validated with available
experimental measurements.
Constantinides and Zhang (2014) presented a study on the VIV of a deep-water lazy wave riser using
the AcuSolve FE N-S CFD solver. AcuSolve is based on the Galerkin/Least-Squares formulation and uses
a fully coupled pressure/velocity iterative solver plus a generalized alpha method as a semi-discrete time
stepping algorithm. The RANS model was selected for the analysis. VIV assessment of the riser was
carried out using a CFD and semi-empirical software (VIVA, Shear7) approaches. The riser response at
touchdown point obtained from CFD is close to that from VIVA, but larger than that from Shear 7. In the
CFD simulation the lazy wave riser is modelled at full scale with actual structural models and the addition
of buoyancy modules. It is the industrys first full scale CFD simulation for such a riser system. Different
buoyancy region geometries are considered in the simulations. It was found that the selection of the
buoyancy configuration is important for controlling VIV and the aspect ratio L/D of the buoyancy module
is an important parameter for the risers response performance.
A large number of model tests have been conducted for VIV during this review period. Jain and
Modarres-Sadeghi (2013) presented experiments conducted on the study of VIV of a flexibly-mounted
rigid cylinder placed inclined to the oncoming flow with one DoF. Angles of inclination from 0 to 75
degrees in the subcritical Re range of 5004000 were considered. It was reported that the lock-in region
started at a larger reduced velocity as the angle of inclination was increased. Sanaati and Kato (2012)
performed experimental studies of the effects of axial applied tension on the vibration amplitude, the
suppression of vibration, hydrodynamic force coefficients and inline and cross-flow frequency responses
during VIV of a horizontally mounted flexible cylinder with a low mass ratio in the subcritical Re (=
100016000). The experimental results show that higher applied tensions could reduce the vibration
amplitude and result in narrower lock-in bandwidths.
Gu et al. (2013) investigated the dynamic response of a vertical flexible cylinder vibrating at low mode
numbers with combined x-y motion in a towing tank. The uniform flow was simulated by towing the
flexible cylinder along the tank in still water. It was reported that the experimental model setup
successfully simulated VIV phenomena and captured several useful characteristics of VIV, but the ratio
between the cross-flow and inline motion was considered high compared with previous findings. Possible
reasons could be the different characteristics of the riser model and experimental setup, as well as the
small Re due to limitations of size and technical characteristics of the testing facilities.
Bourdier and Chaplin (2012) carried out experimental investigations to assess the influence of endstops, placed on one or both sides of a beam connected to the spring-mounted frame supporting the
cylinder, limiting the cylinders motion through impacts with the end-stops on cross-flow VIV. Their
setup, without end-stops, was validated against existing experiments. They used a range of offsets for
106
placing the end stops. The impacts with the end-stops increase the stiffness of the system for particular
offset values. Features observed in their experiments, depending on offsets used included: no evidence of
lock-in, progressive growth in sub- and super-harmonics and chaotic motion. They also computed lift
coefficients. Chaplin and Batten (2014) carried out a series of experiments with multiple DoF spring
mounted cylinders. The carried out 3 sets of experiments (i) a cylinder with 1 DoF in both inline and
cross-flow directions, (ii) a cylinder with 1 DoF in cross-flow direction operating in the wake of
stationary cylinder, at a centre-to centre distance of 5 diameters, and (ii) a cylinder with 2 DoF in both
inline and cross-flow directions operating in the wake of stationary cylinder, all in uniform current. For
case (iii) a number of centre-to-centre and offset values were used. Their experiments show that,
predominantly, wake-induced (low frequency) vibrations are in-line whilst vortex-induced (high
frequency) vibrations are in the cross-flow directions.
Investigations were also focussed on mitigation of VIV of risers. It has been reported that properly
designed helical strakes are effective in the mitigation of VIV fatigue damage for many riser applications.
However, such strakes tend not to be applicable to offshore drilling riser applications due to increases
in drag force as well as workability problems for drilling operations. Taggart and Tognarelli (2008)
summarised studies on many VIV mitigation devices for drilling risers. It appears that the most effective
method is to fit fairings along the riser string. Fairings can also reduce the current drag on the riser and
extend the operability envelopes in strong current. The drawback is the additional cost of the fairings and
the additional time to fit and strip the fairings during riser deployment and retrieval. On a different VIV
mitigation method, Chen et al. (2013a) presented a study on utilizing a steady suction flow control
method to mitigate VIV of a circular cylinder. The experimental study was conducted in a wind tunnel
with a circular cylinder test model as a spring-mass system. The study includes the wind tunnel tests for a
circular cylinder with and without suction flow and the results comparison. Five suction velocities were
adopted to control the flow field surrounding the circular cylinder. The measurement results show good
VIV suppression performance and the performance is related to the suction velocity. This was the first
report of a project lasting for many years.
4.1.2
VIM
Floating platforms with cylindrical structures such as spars, TLPs, and semi-submersibles can be
susceptible to VIM when exposed to currents. The occurrence of lock-in is related to the natural periods
of horizontal motions of a platform, which is normally expressed in non-dimensional reduced velocity.
VIM is most prominent in spar platforms, where most of the industrial experience has been acquired
(API, 2008).
Spar platforms experience vortex-induced oscillations when their surge/sway or roll/pitch periods are
close to the Strouhal number. Gonalves et al. (2012) provided an overview of the influential aspects of
the VIM of spars and single-column platforms, such as current heading, external appendages of the hull,
concomitant presence of waves and currents, motion suppressors, draft condition, and external damping
due to the presence of risers and mooring systems. Field measurements of VIM response have been
recorded for 3 classic spars: Genesis, Hoover, and Neptune, and the measured data indicate inline motion
is much less than that of cross-flow motion (API, 2008). It is also reported that no significant VIM
responses have been recorded from the motion measurement systems installed on a number of truss spars.
Helical strakes have commonly been used on spars to reduce VIM. Strakes can ideally be 95% effective
in eliminating VIM. However, their effectiveness on spars depends on various factors, e.g. the exact
layout and size of the strakes, appurtenances, and current profiles.
Although it has been noted that under certain experimental conditions multi-column floaters, such as
TLPs and typical semi-submersibles, may experience limited VIM, there is no substantiated in-service
evidence of VIM. Recently the semi-submersibles has become a favourable choice as a wet-tree or drytree floating platform supporting steel catenary risers (SCRs), mainly due to its capability of quayside
topside integration and cost effectiveness. However, it is still a challenge for a conventional semisubmersible to support SCRs, particularly large ones, in harsh environments and relatively shallow waters
due to its large heave motion. Deep draft semi-submersible with smaller heave motion could be a
solution. Nevertheless, it has been reported that the deep draft (41m) semi-submersibles experienced VIM
of the platform in loop/eddy current, which could pose a serious problem for riser and mooring fatigue
life (Rijken and Leverette, 2009). Unlike spars, the studies on VIM mitigation for a deep draft
semisubmersible are limited. Several new semi-submersible designs have been developed to reduce heave
motion and VIM. Yu et al. (2013) presented a comprehensive overview of various technologies for drytree semi-submersibles.
Kyoung et al. (2013) presented Technips design of HVS (heave and VIM-suppressed) class semisubmersible. The new design is a modification of the conventional deep draft semi-submersibles through
the use of hull form optimization. The main characteristic feature of the HVS class semi-submersible is
107
the redistribution of displacement from the pontoon to the column through the use of a column step that
breaks up the vortex coherence along the column length and, in concert with the narrow pontoons,
reduces VIM. Xu et al. (2012) presented the VIM validation study of the HVS semi-submersible through
model tests and CFD simulations. In this study, the FE-based CFD software ACUSOLVE was used to
predict VIM response of the hull configurations tested. Detached eddy simulation (DES), RANS with
Spalart-Allmaras turbulence, and Large Eddy Simulation (LES) models were used. Model tests
demonstrated that the HVS semi-submersible design reduces VIM response by approximately 50%
compared to equivalent conventional semi-submersible design. CFD showed good correlation with the
model test results. However, the sensitivity of the VIM response on the scale effect was not included in
the study.
On a similar path, a new concept design of a paired-column semi-submersible (PC semi with 8
columns) has been developed by Houston Offshore Engineering to reduce VIM. The PC semisubmersible and a conventional semi-submersible are included in a project that will last for two years
(REPSEA, 2013). The project is sponsored by Research Partnership to Secure Energy for America
(RPSEA) and the scope of the project includes the testing design parameters for deep draft column
stabilized floaters to determine which have the most impact on VIM, and which VIM mitigation strategies
are preferred for deep draft semi-submersibles. A number of CFD simulations and model tests will be
carried out for different conditions. Software Acusolve, ANSYS Fluent and STAR-CCM+ are used for
the CFD simulations. The project is also to investigate the differences between CFD simulations of using
model scale and full scale, as well as damping effects on VIM. The project is to produce guidelines
on the use of CFD for VIM analysis and on the design of deep draft semi-submersibles with
minimum VIM.
On a relatively broader scope, MARIN has organized a VIM JIP (2013-2016) to study VIM of offshore
structures through model test, CFD simulations and prototype data evaluation. The JIP is to evaluate
differences between model tests and CFD on the one hand and field observations on the other, and to
provide guidance for VIM model testing and CFD studies.
Ma et al. (2013) presented the latest full scale VIM field measurements of a semi-submersible. It was
reported that the actual severity and persistence of VIM is much less than forecasted according to field
measurement data, whilst the results from CFD simulations, provided to the authors, and model tests are
in good agreement. The paper provides some potential factors that could contribute to the differences
between the model and full scale results, but the physical mechanisms behind the overprediction of VIM
in model scale are still not fully understood. Investigations of the VIM of the same semi-submersible
were also carried out by Wu et al. (2014), focusing on scale effect and damping effect due to mooring and
risers, since a typical model scale in a VIM model test is 1:50, leading to the model scale Re being two to
three orders of magnitude less than that at full scale. Furthermore, VIM model tests are typically
performed in a towing tank without including mooring and risers. In this study, a CFD model is first built
at model scale and the results are compared with model test data for validation. The CFD model is then
modified to investigate, respectively, the scaling effects and the effects of mooring- and riser-induced
hydrodynamic damping. The CFD analysis was performed using STAR-CCM+ and DES was selected as
the turbulence model to resolve the unsteady vortex shedding at high Re. The motions of the hull in surge,
sway and yaw is considered in the simulations and the equivalent damping effect was added to the
simulations. The CFD results from the model scale were compared with model tests results, whilst the
CFD results from full scale were compared with field measurement data. The results show a good
correlation between the model tests and CFD simulations performed at model scale. Correlation is also
found to be good between the CFD simulations performed at full scale and field measurements. The paper
concluded that the full scale CFD simulation, with proper meshing and boundary conditions, can
reproduce field VIM with confidence for headings that are critical for fatigue damage. The results also
show that the hydrodynamic damping due to mooring and risers can act to reduce VIM substantially.
Holmes (2014) carried out studies on the VIM of a spar platform in a sheared current. This study was
motivated in part by the problem of using flume tanks for the simulation of platform VIM in sheared
currents. The AcuSolveTM FE CFD solver was used for the simulations to examine the effect of
buoyancy forces, namely changes in density, on sheared current flows around offshore platforms. The
study shows the buoyancy forces could have a strong effect on VIM. On alternative VIM analysis
methods, Kondo (2011) reported a newly developed third order upwind FE scheme to solve the N-S
equations numerically. The method has been applied to the analysis of inline and cross-flow vibrations of
a 2D circular cylinder. The analysis results in good comparison with experimental results. Miyamura
et al. (2014) presented numerical simulations for VIV of a cylindrical structure using Lattice Boltzmann
method which is suited to parallel computation.
108
4.2
Mooring Systems
This section covers the latest developments on the methods and tools for mooring analysis, namely for
calculation of fatigue and extreme mooring line.
It is possible to solve the cable dynamic equations in the frequency domain and a good option is to
combine the nonlinear static solution with linearized dynamic solutions. The advantage is the
computational efficiency. However, the lines dynamic behaviour is inherently nonlinear and often the
linear approach is not accurate enough. Most of the current state of the art procedures use nonlinear time
domain solvers. There are basically two possibilities: the finite difference method and the FE method. The
second is more general since it is able to better represent the slender elements material properties. In fact,
most of the models developed and applied in recent years are based on FE models of the mooring lines
and risers.
Calculation of floating structures mooring line tensions involves the solution of the coupled dynamic
problem; that is to say, the motions of the floater, induced by waves, wind and current, are coupled to the
mooring system (and risers) responses. Different solutions have been proposed in the past and they can be
classified into frequency domain, time domain and hybrid frequency/time domain methods.
Frequency domain methods assume the mooring system forces are represented in the floater equations
of motion by linear stiffness, damping and inertial matrices. The latter should be derived about the
mooring system mean displaced position. The floater linear wave frequency dynamics and quadratic
difference frequency responses are calculated separately and the relevant statistics combined afterwards.
Frequency domain methods, e.g. Zimmerman et al. (2013), are efficient from a computational point of
view and can sometimes be applied for fatigue analysis and ultimate limit state analysis, if adequately
calibrated for those purposes. However, the linearity assumption is often not valid, especially for extreme
line tension predictions; hence, they should be used with discretion.
Hybrid frequency/time domain methods solve the wave frequency responses in the frequency domain,
assuming the related line vibrations are linear, and the low frequency responses, responsible for the large
deformations, are calculated by nonlinear time domain simulations (Low, 2011).
Most of the methods and analysis presented during the reporting period are based on time domain
methods. The objective is to include nonlinear effects, related to loadings on both the floater and the
mooring system and risers, in the analysis. There is a range of options available for solving the coupled
dynamic problem in the time domain, depending on the assumptions to: solve the wave-floater
hydrodynamic interactions, solve the mooring system dynamic problem, and couple the floater dynamics
and the mooring and risers systems. The different approaches can be grouped into partly coupled and
fully coupled methods. The mooring analysis methods can be generalized for multi-body interacting
floating structures by properly considering the hydrodynamic and mechanical coupling effects between
the bodies.
Partly coupled methods solve the time domain equations of floater motions coupled to a quasi-static
mooring system model. The hydrodynamic loading on the mooring lines can be considered in a simplified
way. This approach is valid if mooring line dynamics can be neglected. Zhao et al. (2014c) applied a
partly coupled time domain to study the dynamic behaviour of a side-by-side floating liquefied natural
gas production platform (FLNG) and a LNG carrier. The FLNG is moored to the sea bottom with a turret
system and the two vessels are coupled with linear hawsers and fenders. The numerical model is first
calibrated by comparisons with model tests. The calibration is performed in terms of linear damping
coefficients for each of the degrees of freedom. Comparisons with model test results, in terms of motions
and lines' tensions, show a reasonable agreement the dynamic behaviour characteristics identified in the
experimental results are represented by the numerical predictions, but some differences are observed in
terms of magnitudes. The calibrated numerical model is used to perform sensitivity studies to investigate
the pretension and stiffness of the hawsers effects on the side-by-side system performance. It was
concluded that the changing the two parameters indeed affects coupled responses between the two
vessels. In general, partly coupled methods can be used for screening a large number of environmental
conditions and select a small number for posterior fully-coupled analysis, such as the one by Aksnes et al.
(2013).
There are two options for fully coupled methods: (a) the mooring and riser systems dynamics are
solved at each time step using the floater motions at the fairleads and the floater equations of motion use
the mooring forces applied to the hull, such as the work by Kim et al. (2013b), Kim et al. (2013a) and
Aksnes et al. (2013); (b) the equations of floater motions and mooring lines/risers are set up and solved
simultaneously, such as the work by Yang et al. (2012) and Jacob et al. (2012a). Fully coupled methods
are the most accurate, but most time consuming as well. The use of fully coupled methods is justified by
the need to consider the flexible slender systems dynamic effects on the floater responses; hence, it is
important for the model to include the risers, not considered in this section. The risers damping effects on
the floater motions are often particularly important.
109
Kim et al. (2013a) applied a fully coupled method to analyse the dynamics of a side-by-side floating
storage re-gasification unit (FSRU) and LNG ship. Linear springs and dampers represent the mechanical
coupling between the two vessels. It appears that the hydrodynamic coupling between the two vessels,
which in principle should not be neglected, is not considered. The FSRU time domain equations of
motion include the mooring system force vector, obtained by solving the mooring system dynamic
problem with FE method. The two dynamic problems are solved iteratively at each time step. The authors
compare the coupled responses with and without links between the two vessels and it is concluded that
the linkage reduces the relative motions, which are critical for the offloading operation. However, the
FSRU roll motion increases which means the linkage system characteristics need to be carefully selected.
A fully coupled mooring analysis tool was used by Aksnes et al. (2013) to investigate the failure of two
leeward mooring lines of a floating storage unit operating at the North Sea. The analysis was carried out
for a set of North Sea 100 years return period storms. This study is a good example of the level of detail
which state of the art models can use to represent complex dynamic mooring line problems. The authors
concluded that a possible failure mechanism originates when, in certain conditions, a transverse wave
propagates upwards along the mooring line upper wire segment, which lead to a large curvature in the
wire near the socket. The repeated high stresses at this location eventually lead to failure. This wave
comes about when the link plate at the lower end of the upper wire segment impacts the sea bottom after
being lifted. For this reason, the seabed contact was modelled carefully in the numerical analysis. The FE
model of the mooring line makes use of beam elements for the upper wire segment to account for the
bending and torsional stiffness, whilst the remaining lower segments are modelled with bar elements,
considering only axial stiffness. The upper wire socket and the bending stiffener were modelled with
beam elements of appropriate stiffness. In order to capture the travelling waves strong dynamic
behaviour, the upper wire segment was modelled with very small elements, between 0.5 and 1.0 m, which
required very short time steps for the simulations.
Yang et al. (2012) presented a fully coupled time domain method for the analysis of moored floating
structures. The wave-floater hydrodynamic interactions are calculated directly in the time domain, up to
second order, by a HOBEM, as opposed to the more common approach of using retardation functions
obtained from frequency domain coefficients and second order forces calculated from quadratic transfer
functions. The mooring/risers dynamics are calculated by a FE method based on a global coordinate
system and the theory of slender rods. The equations of body and mooring lines/risers motions are solved
simultaneously at each time step. The method is first applied to a floating hemisphere connected to linear
springs for validation by comparisons with frequency domain results. Subsequently it is applied to
calculate the motion responses and mooring line tensions of a truss spar in regular waves. Fully coupled
and partly coupled results are compared leading to the conclusion that the mooring damping and inertial
effects are important for deep water.
Jacob et al. (2012a) compare two different formulations for the coupling between the floating
structure hydrodynamics and mooring lines and riser dynamics. The simpler is designated as weak
coupling whereby the coupling into the floater motion equations is established through forces applied at
the hull. At each time step, the hull equations of motion are integrated by an explicit method, while the
dynamics of each line are solved independently by a FE method with one, or more than one, time steps.
The authors claim that the lag between the lines forces calculation and their application to the hull may
lead to numerical inaccuracies, especially if the slender elements connecting to the hull are stiff. For
this reason they tested the strong coupling" formulation where all lines and risers are included in the FE
model and the 6 DoF of the hull are associated to a node of the FE model. The floating body equations
of motion are valid for large amplitude angular displacements by considering the complete
transformation relating the body and the global coordinate systems. Nevertheless, it appears that the
calculation of the hull hydrodynamic forces does not follow the large amplitude assumption. Both
formulations are applied to a semi-submersible subjected to an irregular sea state (H s = 7.2 m) and
uniform current, moored at a water depth of 1800 m, with 4 clusters of 4 lines and including 47 risers
and umbilicals. The authors conclude that the motion responses and fairlead mooring line tensio ns from
both formulations are similar. The weak coupling method is computationally more efficient for systems
with a large number of lines.
Fully coupled solutions are the most accurate for prediction of mooring line tensions; however, they are
very demanding from a computational point of view, especially since a large matrix of environmental
conditions needs to be considered for design or verification purposes. With the objective of improving
computational efficiency, Jacob et al. (2012b) implemented domain decomposition strategies for the
coupled analysis methods described by Jacob et al. (2012a). Parallel algorithms solve the equations of
floater motions and the equations for each line in several processors.
One interesting intermediate option which often produces accurate extreme line tension results is to
solve the floater dynamics with a partly coupled model as a first step, and use the resulting motions at the
110
fairleads as prescribed for a cable full dynamic analysis as a second step. The assumption is that the
dynamic effects of the mooring lines/risers do not influence the floater dynamics, or their influence can be
represented with known global damping and inertial coefficients. Along these lines Christiansen et al.
(2013) proposed a hybrid method which combines FE analysis and artificial neural networks (ANN). The
ANN is trained to predict the relationship between the loads on a floating structure and the resulting
tensions on a mooring line. The ANN learning is based on pre-generated training data, which should
cover a broad range of wave characteristics. In this study, the authors used a partly coupled analysis,
including a FE model of the mooring system, to generate the training data and demonstrate the procedure
with a floating structure moored with 18 lines. The hybrid method results are compared with the results
from direct dynamic simulations, in terms of accumulated yearly fatigue damage and fatigue life of a
mooring line, and the results are very similar. The computational time with the hybrid method is 2 orders
of magnitude lower. One should note, however, that some time is needed to setup and train the ANN.
Zimmerman et al. (2013) presented a coupled mooring and anchor translation analysis for a semisubmersible drilling unit. While the conventional mooring analysis considers the anchors as fixed points,
in fact many of the offshore floating units use drag embedment anchors, which translate when exposed to
loads greater than the previous highest load. This translation influences the mooring line tensions. The
author applied a frequency domain mooring analysis method. During the first step the environmental
static loads are calculated at the anchor and, if they are larger than the previous highest load, which could
be the installation test load, the anchor translates according to its holding capacity until equilibrium is
achieved. The dynamic analysis is carried out during the second step. Comparisons between the
conventional method and the present method show that the predicted maximum line tensions are up to
10% lower when the coupled model is used and, therefore, the anchors are allowed to translate.
Whilst existing numerical methods (FE methods) provide quite accurate mooring line dynamics,
including line tensions, for steel and chain catenaries, the uncertainty is larger for synthetic lines.
Synthetic fibre ropes, which have advantages for deep water moorings, are characterized by much more
complex nonlinear dynamic behaviour. This is related to the material properties, namely time dependent
characteristics, viscoelasticity, viscoplasticity and large stretch, which need to be taken into consideration.
Relevant experimental and theoretical/numerical work has been developed during the reporting period.
For example, Liu et al. (2014a) carried out an experimental investigation focused on the dynamic
behaviour of three types of synthetic fibre ropes under cyclic loading. These were polyester, aramid and
HMPE (high modulus polyethylene). It is concluded that the mean load has a strong influence on the
dynamic stiffness and the loading, but the strain amplitude and the number of cycles are important factors
as well. An empirical expression, which takes into account the aforementioned effects, was proposed to
estimate the dynamic stiffness evolution. However, several coefficients depend on the material properties
and must, in principle, be obtained for specific types of fibre ropes.
Huang et al. (2012) proposed a numerical model to represent the time dependent creep and recovery
behaviour of synthetic fibre ropes. The nonlinear constitutive model combines a viscoelastic theory with a
viscoplastic spring-dashpot-slider model. An identification method based on creep recovery experiments
is also proposed to estimate the model parameters. Comparisons of predicted viscoelastic and viscoplastic
strains for aramid and polyester fibres with experimental data show good agreement and the authors
suggest the method can be incorporated into existing codes and procedures for mooring analysis.
Accounting for the viscoelastic and viscoplastic properties of synthetic lines in deep water mooring
analysis was of the aim of the time domain model implemented by Kaasen et al. (2014). A time domain
nonlinear spring-dashpot model is implemented to represent, simultaneously, the recoverable
instantaneous elongation, the slow elongation response, the irreversible instantaneous elongation and the
long term elongation (creep). The lines are represented by finite elements, which means that a new
element incorporating the aforementioned characteristics was implemented. Data from tension-elongation
tests with polyester ropes was used to identify the model parameters. The new model was applied to the
mooring analysis of a platform subjected to storm conditions and it was possible to identify the increase
in ropes length and the consequent modification in mooring system characteristics as function of time.
A simple and practical engineering solution to account for the nonlinear and viscoelastic properties of
polyester has been to consider two types of axial stiffness, namely a static and a dynamic stiffness, run the
mooring simulation twice and take the largest extremes. This method tends to provide conservative
values. The dual stiffness method presented by Tahar and Sidarta (2014) goes one step further. The
authors proposed to use the static stiffness to calculate the mean platform motions, adjust the lines
length, and then use the dynamic stiffness to calculate the dynamic motions, all in a single fully coupled
mooring simulation. The method was applied with a fully coupled code for the mooring analysis of a
semi-submersible, using the 100-year wind and wave hurricane environment in the Gulf of Mexico. The
solutions using the static stiffness, dynamic stiffness and the dual stiffness are compared. The authors
111
conclude the new method can replace the previous procedure of running the simulations twice, for the
static and for the dynamic stiffnesses.
Large stretch is one additional aspect where synthetic fibre ropes differ from chain and wire lines.
Consistent FE models should take into consideration the large stretch characteristics. Webster et al.
(2012) extended the dynamic model for a slender rod without stretch to include large stretch. The authors
adopted a simple assumption regarding the stretch characteristics and include an existing model for the
viscoelastic behaviour of the large stretch elements. The method can be implemented into FE models, as
presented by Ma et al. (2014). They implemented the new element into a FE method and tested with a 3D
extensible catenary. Comparisons with analytical solutions validate this method.
4.3
Lifting operations
With growing interest towards increasingly unconventional offshore exploration and exploitation, the indepth understanding of nonlinear hydrodynamics involved in coupled multi-body offshore systems, such
as heavy lifting and subsea installations, is becoming immensely important, if we are to ensure safe and
profitable operation of such systems. It is for these reasons that this committees report contains an indepth review of developments in the modelling of loads due to offshore lifting operations. Offshore lifting
operations, in general, include lifting in air and subsea lifting. Engineering analysis for design verification
and lifting operation, normally, include calculations demonstrating the adequacy of the vessels stability
and station-keeping capability, either through dynamic positioning or mooring system, during heavy
weight lifting operations and calculations demonstrating strength of the crane and supporting structures.
The load for the strength assessment includes fixed loads (e.g. hull weight), and variable loads (e.g.
portable cranes and machinery), dynamic loads, wind loads, loads due to trim or list, snow/ice loads and
others depending on the area and nature of the operation. Modelling of the lifting appliance system should
take the following aspects into account: basic dynamic factors based on boom tip velocities (API, 2012),
buoyancy, added mass, drag, rope weight, resonance and sea bed suction ((ABS, 2014a), (ABS, 2014b)).
Various approaches for investigating the dynamics of crane vessels have been studied, especially
during the last two decades. A brief review of studies undertaken during earlier years were summarised
by Ellermann et al. (2002). Accordingly, all previous investigations use a common approach to define the
motion of the payload, where the excitation of the payload is taken simply as a prescribed motion of the
pivot point of the hoisting rope. However, the influence of the payload on the motion of the vessel is
neglected. This method might be applicable for the scenario where the payload to vessel ratio is very
small. The theoretical, experimental and numerical studies which have been documented in all these
contributions reflect the strong interest towards the analysis of offshore crane vessel dynamics. Among
them worth noting is the work by Wouts et al. (1992) aimed at monitoring two major offshore heavy lift
operations of the time. These operations involved lifting of jacket structures from the cargo barge using
large semi-submersible crane vessel (SSCV), lowering them in the water, upending and then setting on
the seabed. Among the observed results, motion and tension spectra were correlated with model test data,
as well as with data obtained from calculations based on the theoretical formulation of the wave spectra in
order to evaluate the significance of lift dynamics involved in the process. It was noted that the
calculation based results showed considerable deviations from the measured data, except for the
computation of natural periods. However, comparison of other statistical quantities is not performed,
because the recorded experimental data comprised of low, medium and high frequency components and
the numerical formulations of the time were not capable of capturing these details effectively.
The first simplified, complete linear computer program to analyse the equation of motion of a crane
vessel and its suspended load is attributed to Witz (1995) where the system was described by six DoF for
the vessel together with three orthogonal displacements of the lifting mass. In this model all external
forces were considered to be proportional to the motions of the floating crane and the cargo. Furthermore,
the motion of the floating crane, as well as the wire rope force, were linearized and the off-diagonal terms
of the added mass and damping matrices were neglected in the analysis. An improved model was
developed by Kral et al. (1996) which is a 3D linear model solved in the frequency domain. In their
model, the coefficients of the hydrodynamic forces were calculated as a function of the frequency and the
inverse Laplace transformation was used to transform these coefficients into the time domain. The
system, although considered a simple crane ship model with linear hydrodynamics, showed the full range
of nonlinear phenomena from period doubling to chaotic behaviour with quasi-static changes of the
hoisting rope length over 28 m. No comparison of the findings with experimental or other numerical
methods is performed as such data appear to be unavailable at that time.
Clauss and Vannahme (1999) performed experimental studies in order to shed further light on this
subject. Seakeeping tests were carried out using a Magnus crane barge model at a scale of 1:25 in regular
head waves to analyse the influence of nonlinear mooring forces. All model tests were performed in the
wave tank of the Berlin University of Technology. The experimental analysis showed that although the
112
dynamic behaviour of moored floating cranes can be treated as linear in some cases, the coupled system
of floating structure and swinging load shows distinctly nonlinear behaviour and parametric oscillations
under certain conditions, e.g. at larger motion amplitudes, which are indispensable for heavy lift
operations. Therefore, they recommended developing nonlinear numerical models in combination with a
bifurcation analysis and followed by model test verification for improved results. Ellermann et al. (2002)
and Ellermann and Kreuzer (2003) focussed their efforts on developing such models by adopting the
potential theory approach for the FSI and the main objective of their investigation was to detect nonlinear
phenomena, such as bifurcations and the existence of multiple attractors. Both theoretical and
experimental studies were undertaken to achieve these goals. In the mathematical modelling part, the state
space model was used where additional state variables needed to be defined in order to transform the
frequency-dependent hydrodynamic radiation forces into the time domain. Two mathematical models of
different levels of complexity were used to systematically determine the responses of the vessel-payload
system to periodic forcing of waves. One technique was the multiple scales method which allows for the
investigation of the nonlinear dynamical system in the frequency domain and which results in an
analytical solution. The other technique applies numerical path following methods to trace bifurcations of
periodic solutions. The comparison between experimental and numerical findings is rather limited,
probably due to the incapability of the numerical model to simulate in-depth the details of the
experimental setup. Nevertheless, comparisons of mooring line forces and the surge motion of the barge
with experimental measurements are agreeable, within the limitations of the simplifications of the
numerical model. However, quite a good agreement was achieved between the two computational
methods.
Among more recent studies, Cozijn et al. (2008) proposed a combined physical scale model test, time
domain computer simulations and full scale observation approach in order to determine the operational
limits of the offshore installation process. The complete analysis comprised hydrodynamic scale model
tests, time domain computer simulations and observations made during the actual installation offshore.
The model tests were carried out in MARINs offshore basin and the computer simulations were
performed using the commercial multi-body linear time domain simulation tool LIFSIM, developed by
MARIN. The model test results are first used to optimize the performance and accuracy of the simulation
model LIFSIM. The damping parameters are also calibrated using the model test data. A large number of
simulations are then performed to investigate the different environmental conditions, once reasonable
agreement between the model test and simulation tool are obtained. Apart from this study, Cha et al.
(2010) proposed a coupled model to simulate the dynamic response of cargo suspended by a floating
crane. They set up the dynamic equations of motion considering the 6 DoF floating crane, in regular
waves, and the 6 DoF cargo suspended in air based on multi-body system dynamics. The nonlinear
hydrostatic and the linear hydrodynamic forces were considered as external forces and only the motion
parameters of the crane barge and payload were calculated. To verify their equations of motion used, a
comparison of the motions of the floating crane and the heavy cargo, was carried out with predictions
from the Multi-Operational Structural Engineering Simulator (MOSES). This comparison shows similar
patterns with small differences in amplitude and phase of the heave and pitch motions. They also
performed parametric study to find the tension in the rope, connecting the cargo with the crane in barge,
under different wave amplitudes, frequencies and heading angles. However, results for the tension were
not compared with predictions from any other software. This model was extended by Park et al. (2011) to
provide a more realistic approach by considering a floating crane and a heavy block which are connected
using elastic booms and wire ropes. Extensive parametric studies were performed to highlight the
differences between the numerical results obtained with rigid and elastic booms. No comparisons with
other numerical and experimental results were performed due to the unavailability of such data. The main
focus of both of these studies, however, is to analyse the behaviour of the cargo suspended in air with
reference to the motion of the crane barge. It appears that mathematical model for the analysis of
nonlinearity involved in the underwater motion of the cargo/payload, i.e. nonlinear hydrodynamics, near
the free surface is not as yet available in open literature, probably due to the complexities of this problem.
On the other hand, a recent recommended practice report released by DNV (2010), on modelling and
analysis of marine operations, provides some simplified methods for obtaining simple conservative
estimation of forces acting on an object when lowered through the splash zone. In this report some
guidelines are also provided on how to create a model for the coupled dynamic motion analysis during
heavy lifts operations. In addition, a recent study performed by Bai et al. (2014b) describes the findings of
a fully nonlinear time domain analysis of a fully submerged cylindrical payload hanging from a rigid
cable and subject to wave actions. The numerical model is validated, against other experimental and
numerical results available, for wave interaction with submerged fixed horizontal cylinder. The model
provides very consistent results similar to other numerical models. The model also appears to be in good
agreement with experimental results up to the point where viscous effect start to play significant role,
113
Keulegan Carpenter number of 0.6. However, this fully nonlinear time domain model only considers the
wave interaction of a single submerged payload with one rotational degree of freedom without coupling
with the crane vessel. Hence, there exists ample scope to analyse and understand the coupled
hydrodynamic behaviour of a crane barge and its submerged payload subjected to nonlinear waves.
4.4
Wave-in-deck loads
Further investigations for obtaining reliable estimates of extreme loads on platform decks due to possible
wave in deck loading continued during the reporting period. Improved statistical methods combined with
advanced fluid flow simulations using CFD have given enhanced insight into this loading problem, which
is essential for the safe assessment of life extension of ageing platforms. Subsidence of seabed, updated
wave data bases and improved methods for prediction of kinematics of extreme wave crests may result in
considerable higher estimates of deck impact height with consequent impact loads higher than those used
in the design of the platforms. Previous simplified global methods only providing the maximum impact
force and detailed component approaches based on the momentum method providing time histories of
impact load have been compared with more accurate CFD simulations, mainly applying a VOF method
for free surface capturing.
Abdussamie et al. (2014) compared results obtained using a simple momentum method with a single
phase VOF method and experimental results for a fixed rectangular box. Regular undisturbed incoming
waves were assumed. It was found that the momentum method severely underestimates the magnitude of
horizontal wave-in-deck force whilst for the vertical force the upwards force is underestimated and the
downwards force ovepredicted during exit phase.
Scharnke et al. (2014) investigated experimentally wave-in-deck loads in regular and irregular waves
for a detailed model of a complex deck structure of an existing jacket platform. The model tests were
compared with a very simplified global design method based on impact height and kinematics of a fitted
undisturbed 5th order Stokes wave. They concluded that the simplified load model underestimates the
impact loads in both regular and irregular waves due to the underestimation of horizontal velocities in the
wave crest of the theoretical wave model used. From the measurements it was also found that wave
breaking during deck impact may increase the horizontal deck loads and decrease the vertical deck loads.
Iwanowski et al. (2014) reported results from VOF analyses for the same platform deck model used by
Scharnke et al. (2014), applying both single- and two-phase flow models. The computationally more
intensive two-phase flow model was used in order to quantify the possible effect of air pockets on the
wave impact loads. It was concluded that the two-phase flow CFD solution, in general, shows a better
agreement with the experimental results than the single-phase flow solution for regular waves. The
comparison for the irregular wave was not conclusive.
Lu et al. (2014) presented results from CFD analyses, using OpenFOAM, of wave impact loads on
platform deck and wave-bridge interactions. A two-phase flow model was used. Carefully focused
extreme irregular waves, derived from NewWave theory and Jonswap wave spectra, were applied. The
New Wave model was found to better approximate the real wave crest hitting the deck than a regular
design wave.
4.5
Floating Offshore Wind Turbines
The success of several Floating Offshore Wind Turbine (FOWT) demonstration projects, for example
Roddier et al. (2009) and Skaare et al. (2014), has shown great promise for this emerging new field. The
numerical tools used to model FOWTs leverage many decades of research supported by the offshore oil
and shipping industry, although wind turbines present new issues and complexities that involve a revisit
of offshore mechanics theory. Several methods and software have been developed to address unique
FOWT requirements, accomplished by coupling the aero-elastic behaviour of wind turbines with the
hydrodynamic and station-keeping characteristics of floating offshore platforms, such as those by
Jonkman and Buhl Jr. (2005), Larsen and Hansen (2007) and Bossanyi (2009). Important investigations in
specific areas are also worth noting. Rotor aerodynamic thrust is efficiently modelled using the blade
element momentum theory by Manwell et al. (2009). CFD activity is also applied toward capturing rotor
wake deficits by Churchfield et al. (2012). Many approaches are available to account for vibration issues
related to blade and tower flexibility, including variations of beam theory by Branner et al. (2012) and the
more rigorous FE model by Heege et al. (2007). Beam theory is a popular applied method to balance
computational speed with accuracy. A successful FOWT simulation method should entail coupling an
aero-structure-elastic method with an offshore hydrodynamic and station keeping method to yield
effective simulation tools, such as those recently developed by Thomassen et al. (2012), Cordle and
Jonkman (2011) and Duarte et al. (2013).
Several studies explored the process to couple a wind turbine simulation tool with existing
hydrodynamic/mooring programs to leverage validated toolsets and extend model capabilities. Cordle and
114
Jonkman (2011) provide a survey of the early programs used. Additional examples are given using
AQWA (Huijs et al., 2014), Charm3D (Bae et al., 2010), OrcaFlex (Masciola et al., 2011),
SIMO/RIFLEX+HAWC2 (Skaare et al., 2007), SIMPACK (Matha et al., 2012) and TimeFloat (Roddier
et al., 2009). Various techniques to implicitly and explicitly couple the rotor aerodynamics, tower
structural response, and platform hydrodynamics and expand model fidelity are discussed by Jonkman
(2013). These tools must provide sufficient accuracy to address issues raised in international design
standards, such as ISO 19900 for offshore platforms (ISO, 2002) and IEC 61400 series for wind turbines
(IEC, 2005, IEC, 2009). The significance of second order sum-difference hydrodynamic effects on
moored floating offshore systems is well documented in the marine industry. Findings by Roald et al.
(2013) reinforce this to suggest second order sum-frequency hydrodynamic effects may impart loads
targeting higher frequency bandwidth. This observation is based on global performance simulations. Xing
et al. (2012), considered high order hydrodynamic theory to show these frequencies can coincide with the
wind turbine drive train natural frequency to illustrate the importance of second order effects. The TLP
system explored by Bachynski and Moan (2013) discusses application of first and second over
hydrodynamic theory in a FOWT system, with added consideration towards the atmospheric modelling
needs required for the wind turbine. This study reveals an increase in tendon tension with second order
wave forces included, in line with observations experienced with conventional oil and gas TLPs.
Several of the proposed FOWT systems emulate geometries found in oil and gas platforms, but at
smaller scales. The examples provided by Roald et al. (2013), Xing et al. (2012) and Bae et al. (2010)
document instances where standard practice offshore simulation methodologies can be applied to advance
offshore wind technology. The characteristic distinguishing FOWT from traditional oil, gas and shipping
vessels is that wind turbines are designed to extract energy from the wind. Departing from traditional oil
and gas modelling practices, wind turbines require accurate 3D turbulent wind field representations to
capture the localized blade lift and drag forces. One common approach to model the torque and thrust
loads in the wind turbines rotor swept area is discussed by Ingram (2005). The premise of this approach
is to calculate the localized force and moments along the wind section, and integrate this force along the
blade span in a 3D wind field to produce the resulting time-varying torque and thrust loads. This is the
essence of the blade element momentum theory as discussed by Madsen et al. (2007). The ensuing torque
and thrust loads, plus the added motion of the floating support platform, lead to large tower-base and
blade-root loads, which has motivated a desire for analysis codes and techniques to model FOWT
systems, exemplified by Nielsen et al. (2006), Wayman et al. (2006) and Robertson and Jonkman (2011).
Matha et al. (2011) led a study summarizing gaps in FOWT simulation tools. Two issues addressed by
these authors, which have drawn the attention of the offshore community, include a FE mooring
representation and higher order hydrodynamics theory. In the time elapsed since this study was published,
research studies by Bae et al. (2010) and Hall et al. (2013) have addressed the mooring system
considerations, and high order hydrodynamics concerns were investigated, for example, by Li et al.
(2014a), Bachynski and Moan (2013) and Gueydon et al. (2014).
The role of FOWTs is to convert wind energy into electrical energy. This conversion results in large
surge and heeling offsets as energy is extracted from the ambient environment. Energy capture is made
more efficient through active control of the wind turbine blade pitch angles to regulate power generated.
However, there is concern that blade pitch control strategies may inadvertently lead to negative platform
pitch damping, which would lead to larger platform pitch offsets with the turbine operating (Nielsen et al.,
2006, Jonkman, 2008). Control logic also affects power performance and fatigue of wind turbine
components by damping the platform pitch response, as discussed by Skaare et al. (2007). Suzuki and
Sato (2007) conclude that land-based rotor-nacelle-assemblies (RNA) can be applied to offshore systems,
provided the platform motions are small. In light of the new environmental challenges facing FOWTs,
Robertson and Jonkman (2011) showed, through simulations, that the loads experienced in the RNA
components of offshore systems during normal operation are moderately larger than land-based
counterparts due to the platform global motion. This study also reveals that the tower base moment
significantly exceeds the magnitude found in comparable land-based systems. However, the conclusions
drawn apply to FOWT in normal operations regimes with small wave heights. It is generally observed
that high loads can be experienced in the FOWT RNA components occur during energy production
phases, but greater loads can be experienced in the floating system and mooring foundations when the
wave height is largest. Since there is a large degree of uncertainty on which environment conditions lead
to the largest loads in the integrated FOWT system, the design and analysis procedure usually resorts to
many simulations in a wide variety of sea states and wind conditions, discussed by Li et al. (2014a).
For the aforementioned reasons, accurate simulations to capture the control logic, aero-elastic
behaviour, structural response and hydrodynamic interactions are crucial to the design of FOWT systems.
Results were disseminated by Popko et al. (2012) as part of a multi-national effort under IEA
(International Energy Agency) Wind Task 30 to conduct code-to-code comparisons of various fixed-
115
bottom floating offshore systems and corroborate application of various theories. This project spawned
the OC4 Phase II effort, which focused on code-to-code comparisons of a semi-submersible FOWT unit
(Robertson et al., 2013). The purpose of both exercises was to observe if codes using/based on similar
theories converged to the same results. Future IEA efforts will focus on comparing numerical models
with experimental data, which will help identify recommended approaches for simulating floating wind
systems. Findings from early FOWT model experiments are summarised by Coulling et al. (2012) who
explored the requirements to scale a 1/50th floating semi-submersible platform and wind turbine rotor
against a full scale simulation. The paper discusses concerns related to blade scaling and matching the
rotor aerodynamic thrust at the rated wind speed through the Reynolds number. The platform itself
follows Froude scaling. Combining different scaling rules helps match experimental results with full scale
simulations. Numerical and experimental validation efforts were carried out by Adam et al. (2013) for a
TLP model in transit and operation cases. Adam et al. (2014) followed up with tests in scaled Baltic Sea
conditions. Both papers demonstrate the importance of combining representative wind and wave
conditions in the scaled model experiments.
Offshore wind is emerging as an alternative energy source. The Global Wind Energy Council (GWEC)
estimated an installed capacity of 7 GW 2013, which is projected to grow to 10 GW by 2015 (GWEC,
2014). A majority of the global installed capacity are bottom-founded designs, which first took root as
demonstration projects in the early 1990s. Floating offshore wind systems are following a similar path
and historical learning curve as their bottom-founded counterparts, with present floating wind
installations confined to a few demonstration projects (EWEA, 2013). Bottom-founded offshore wind
installations become less economical as water depth increases, allowing FOWTs to reach economic parity
(Musial and Ram, 2010). Wind turbine physics incentivise larger machines because generated power
capacity scales proportional to the rotor diameter squared (Manwell et al., 2009). These factors combine
to promote and encourage research in this area.
5.
PROBABILISTIC MODELLING OF LOADS ON SHIPS
The ship design process needs to ensure the integrity of the hull, where wave-induced loads contribute
significantly. These loads have to be representative of the environmental conditions encountered through
the entire life of the ship, which involves statistical representation of the environment and the consequent
loads, as well as computationally efficient methods simulating suitable design conditions. This section is,
therefore, structured with the aforementioned thoughts in mind.
5.1
Stochastic description of loads is required for fatigue spectral analysis and computation of extreme loads
for ship design, as well as for the application of ship structural reliability methods. For longitudinal
strength assessment of most ship types, basic load variables considered are VWBMs and SWBMs. Load
combination of these two components is also considered, accounting for the fact that their maximum
values do not occur simultaneously. SWBMs change from one voyage to another and they also vary
slowly within each individual journey because of the use of consumables. However, such variation within
a voyage is normally neglected in the load combination studies. Load combination of different
components of global wave loads, such as vertical and horizontal bending moments (HBM), and torsional
moments, and between global and local wave load components are also important (Mohammed et al.,
2012).
In recent years, attention, by and large, focussed on ULCSs, where vibratory loads become important.
For these flexible ships operating at high speed, the encounter frequency can overlap with the natural hull
girder vibration frequency and cause resonance, denoted as springing. This steady state resonant vibration
may arise from the two-node flexural mode or the coupled horizontal and torsional mode, the latter only
subject to relatively few investigations. Both springing and, transient vibratory, whipping responses are
relevant for the fatigue limit state check. For the ultimate limit state check whipping response is more
important since it can significantly increase the extreme rigid body VWBM. In recent years a substantial
number of studies dealt with stochastic description of combined rigid body and vibratory responses of
ULCSs. There appears to be a tendency to consider these two load components together, rather than to
study them separately and to combine them using load combination principles. This is partly due to the
ascendance in coupled hydrodynamic and structural analyses, providing results of reasonable/acceptable
accuracy, and partly due to the nature of model or full scale measurements. In these cases, rigid body and
vibratory loads are explicitly included in the analysis and there is no need to combine them. The
contribution of the vibratory response can be assessed either by performing analyses with and without
elastic effects or by separating low and high frequency signals.
Mao and Rychlik (2012) presented a simple approach for the prediction of extreme response, for
example, 100-year return stress, using Rices method combined with Wintersteins transformed Gaussian
116
model for stresses. The method requires description of long term variability of the standard deviation,
skewness, kurtosis, and zero upcrossing frequency of ship response. It is assumed that the parameters are
functions of encountered significant wave height, heading angle and ship speed. The accuracy of this
model is validated using the full scale measurements of a 2800 TEU container ship. The parameters in the
Winterstein transformation are given by analytical functions of significant wave height only. The
proposed method is also used to estimate extreme responses of a 4400 TEU container ship, for which no
measurements are available. Results were similar to the 2800 TEU ship. The presented investigation
shows that having accurate wave environment model is extremely important for the reliable estimation of
the extreme ship responses.
Lee et al. (2012a) presented an analysis procedure for determining values of wave-induced bending
moments, considering the effects due to whipping, suitable for design application. The design bending
moments due to whipping of the hull girder were determined by multiplying the design rule bending
moments by correction factors for hogging and sagging based on the Lloyds Registers guidance notes
(Lloyd's Register, 2011). The correction factors for hogging and sagging of a 13000 TEU container ship
were predicted by employing time domain nonlinear analysis and an Equivalent Design Sea state (EDS).
In the nonlinear hydroelastic analysis, the correction factor for hogging was 1.42 for the ship and is 50%
greater than the standard rule hogging correction factor of 0.94. The correction factor for sagging was
1.98 and is 60% greater than the standard rule sagging correction factor of 1.24. These nonlinear bending
moments are significantly greater than the standard IACS- based rule wave bending moments.
The effect of hull girder flexibility on the VWBM for ULCSs was analysed by Andersen and Juncher
Jensen (2012). A nonlinear time domain strip theory is used for the hydrodynamic analysis, whilst
slamming forces are determined by a standard momentum formulation. The hull flexibility is modelled as
a non-prismatic Timoshenko beam. The statistical analysis is carried out using the First Order Reliability
Method (FORM) supplemented with Monte Carlo simulations. Strip theory calculations are compared to
model tests in regular waves of different wave lengths, and good agreement is obtained for the longest of
the waves. For the shorter waves the agreement is less satisfactory. The discrepancy in the amplitudes of
the bending moment is explained by an underestimation of the effect of momentum slamming in the strip
theory applied.
Ogawa et al. (2012) examined the relationship between the occurrence probability of a slamming
induced vibration and sea state. These relationships were investigated based on the full scale
measurement data of two large container ships, which operate on the same sea route, but at different
periods. The effect of the variability of sea state on the occurrence probability of whipping-induced stress
on the hull was examined using the computation of long term prediction and wave hindcast data. The
probability of occurrence for high stress is different between the two container ships owing to the
difference of period for full scale measurement, though these container ships navigated the same sea
route. The stresses are, in general, same for the same wave condition. The difference of probability of
wave height has much effect on the whipping and the magnitude of induced stress. It is also clarified that
most of slamming induced vibration occurs in head or bow seas. The probability of occurrence for both
container ships is consistent with the actual sea state. It is also found that the probability based on the
IACS Recommendation 34 (IACS, 2001) is quite larger than the probability in the real sea state. It is
noted that the operational effect is not negligible for the evaluation of the occurrence probability of
slamming.
Andersen and Juncher Jensen (2014) analysed full scale measurements of the wave-induced amidships
VBM of a 9400 TEU container ship. The focus was on assessing the effect of the hydroelastic high
frequency vibration on the extreme hogging VWBM. In the extreme case, they noted that the high
frequency vibrations were of the same magnitude as the wave frequency, i.e. so called rigid body,
response. It was also noted that even though the ship is sailing in bow quartering seas, only the 2-node
vertical vibration mode is apparently excited. Following the extreme event analysis and verification, three
hours of strain measurements are used for establishing a Gumbel distribution for the extreme value
prediction. Extreme value predictions using the measured results indicate that the probability of
exceeding the rule design hogging VWBM by 50% could be of the order of 1.3% during three hours of
operation in a sea state with significant wave height around 8 m.
Teixeira et al. (2013) assessed the probabilistic characteristics of the load combination factors for
global still water and wave-induced VBMs of double-hull tankers. The calculations are performed based
on loading manuals of oil tankers representative of the range of application of the IACS Common
Structural Rules (CSR) design rules. Different load combination methods are used, including an analytical
method that provides the combined characteristic value of still water and wave-induced bending moments
based on the Poisson assumption for upcrossing events and using the first order reliability method in
combination with the point-crossing method. The mean value and the standard deviation of SWBM in one
random voyage are defined as 70% and 20% of the maximum value in the loading manual, respectively.
117
These values are used for statistical description of sagging bending moments in full load conditions and
hogging bending moments for the ship in ballast. For the wave-induced loads, a Weibull distribution was
adopted to describe the values of VWBM at a random point in time. A simplified approach for the
Weibull model was adopted, in which the shape parameter was assumed as 1.0 and the scale parameter
was defined from the IACS rule minimum value defined for design purposes. It was shown that the
average load combination factors of the tankers decrease rapidly when the design period increases from 1
to 10 years and then tend to stabilize. It was also shown that the mean voyage duration demonstrates a
strong influence on the load combination factors. From this analysis, average of load combination factors
of 0.84 and 0.81 were obtained, respectively, for large (LTK) and small (STK) tankers in full load and for
a reference time period of 20 years. In the ballast load condition, the average of load combination factors
reduced to 0.80 and 0.76 for large (LTK) and small (STK) tankers, respectively, which differ on the
probabilistic models of the voyage duration and time in port. These factors multiply the most probable
VWBM to be added to the most probable SWBM in order to get the most probable combined bending
moment.
Mohammed et al. (2012) presented a cross-spectral stochastic analysis methodology for the
determination of the combination of global wave-induced dynamic loads. The methodology considers the
use of bivariate probability density functions, for the cross-spectral probabilistic approach, or the covariances of two random variables with their associated derivatives, for the cross-spectral Hamiltons
method, and assumes only wave frequency hydrodynamic actions under steady forward speed conditions.
A 3D source distribution based on Greens function was applied on the panel model to predict the ship
motions and rigid body dynamic wave loads. The design extreme values of global wave-induced load
components and their combinations for a container ship progressing in irregular seaways are predicted
using these two cross-spectral methods together with the short and long term statistical formulations. It is
shown that, in general, both cross-spectral analysis methods can be employed to assess the effects of
loads in ship design and reliability analyses. However, the cross-spectral Hamiltons method predicts
slightly higher load combinations than the cross-spectral probabilistic approach.
5.2
Equivalent design waves
The concept of design waves is used in direct calculations to reduce the number of load cases to be
checked for yielding, buckling, ultimate strength or fatigue. A design wave is an equivalent wave or wave
group representing the long term response of the dominant load parameter under consideration. Hence,
the accuracy of the methodology highly depends on how the design parameters have been chosen. Several
types of EDWs exist. General practice in industry is to choose a regular wave for the EDW, the main
advantage being its simplicity. The EDW is then defined by the following parameters: frequency,
heading, amplitude and phase. However, use of irregular design waves is on the increase, as they are a
more accurate representation of irregular sea states. Response Conditioned Wave (RCW) or Most Likely
Response Wave (MLRW) are examples of irregular design waves. The notion is to include both the wave
spectrum and the ship response in the definition of the Design Wave. de Hauteclocque et al. (2012)
compared the stress response of a FSRU computed with a long term spectral approach and with a set of
equivalent design waves. The approach for selecting the heading of the applied EDW has been shown to
have a significant impact on the accuracy of this method, especially when a regular design wave is used to
represent short-crested sea states. The irregular design wave, that may include the directional spreading,
produces more robust results at the expense of negligible additional computation costs. It is shown that
using at least 10 EDWs, the accuracy (standard deviation of the ratio between the EDW stress and the
long term stress) is of the order of 6%.
Sarala et al. (2011) proposed a method to derive an EDW from a response based analysis (RBA) to
represent extreme loads on a weather-vaning FPSO. In this approach the combined effects of wind, windsea, current and swell are considered. RBA is based on three-hourly hindcast metocean data and uses
results of heading analysis directly. EDWs are then derived based on the spectral characteristics of each
response. This approach is compared with RAO based approach as generally applied in the industry,
where the regular design wave is defined only from the RAO characteristics. Six different EDWs are
derived based on RBA and RAO methods. It is concluded that the RBA approach provides more realistic
responses compared to the RAO based method. Deriving equivalent design waves using only the RAO
characteristics is found to give some non-conservative and unrealistic EDWs in some cases
The EDWs are often used to reduce the duration of the simulations to allow for computation of
nonlinear loads with more accurate but time consuming software; that is to say, a three hour sea state or
even 25 years life time may be reduced to only a few waves. Derbanne et al. (2012a) used EDWs to
compute the 25 years extreme bending moment of an ULCS including the whipping response. The
hydroelastic model couples a 3D hydrodynamic potential solver with 3D FE structural dynamics.
Slamming loads are computed using a 2D generalized Wagner approach. Results using several type of
118
EDWs are compared with a 53 hour long simulation of a Design Sea State. Regular and irregular design
waves are both providing a good estimation of the nonlinear bending moment. However, the method
based on an irregular sea state with increased wave height is found to be more accurate, by comparisons
with a very long simulation in the design sea state. The significant wave height of the design sea state is
increased in order to artificially increase the upcrossing rate of the design value and to save on
computation time.
Johannessen and Hagen (2012) were concerned with the accuracy of the estimation of wave induced
design responses from model tests in irregular sea states. They showed that for a highly nonlinear
response, such as wave-in-deck loading, it is difficult to have an accurate estimation of the extreme
response. By increasing the sea state significant amplitude, it is possible to observe the relevant response
more frequently, i.e. increased upcrossing rate; hence, reduce the number of model tests.
Juncher Jensen (2011) combined properties of the FORM analysis and Monte Carlo simulations to
better predict the upcrossing rate of a given response. With Monte Carlo simulations the necessary length
of the time domain simulations for very low upcrossing rates might be prohibitively long. Using a
property of the FORM reliability index, assumed to be valid in the Monte Carlo simulations, makes it
possible to increase the upcrossing rates and, thus, reduce the necessary length of the time domain
simulations by applying a larger load spectrum than relevant from a design point of view by increasing
the significant wave height. The mean upcrossing rate thus obtained can then afterwards be scaled down
to its actual value. In this paper the usefulness of this approach is investigated, considering problems
related to wave loads on marine structures. Here the load scale parameter is conveniently taken as the
square of the significant wave height.
Kim et al. (2012a) estimated the long term midships VWBM and impact-induced bending of two hull
forms, both experimentally and numerically. The experimental approach utilizes the lifetime maximum
loads analysis based on the Weibull analysis technique. The numerical methods include a series of
analyses based on Design Loads Generator (DLG), a tool that can construct an ensemble of short input
wave time series, the extreme responses of which follow the theoretical extreme value distribution of a
Gaussian random variable for a given exposure time. The exposure time associated with the distribution
becomes a good measure by which the associated non-Gaussian responses can be bound. Based on this
strategy, the design whipping response has been estimated. To show the responses of comparable
exposure time, limited Monte Carlo simulations of combined wave and whipping bending moments are
conducted numerically. The DLG approach has the potential to supplement or replace a typical lifetime
load analysis based on the combination of operational cells.
Oberhagemann et al. (2012) discussed ways to embed time domain field methods in extreme value
predictions, which require appropriate hydrodynamic codes capable of modelling all relevant
nonlinearities. Approaches are suggested that appear to give most reliable results. They rely on Monte
Carlo simulations, a reduction of parameter variations and extrapolation of exceedance rates over
significant wave height. The computational effort is large, yet it can be handled with modern computer
clusters. Further studies are recommended to gain more experience on the applicability of the
extrapolation over the significant wave height, as well as the uncertainty related to this extrapolation.
Finally, the discretization errors are a critical issue, which must be considered in combination
de Hauteclocque et al. (2013) evaluated the extreme nonlinear VBM, without whipping, for 70 ships (9
containerships, 19 tankers and 29 bulk carriers) using EDWs or long simulations in short-crested or longcrested sea states. Simulations are carried out in time domain, using linear potential theory with nonlinear
hydrostatic and Froude-Krylov forces. It is found that the regular design wave fails to predict the
nonlinear hogging bending moment correctly (mean error -19%, standard deviation 17%). The irregular
design wave, even without spreading, gives a good estimate both in hogging and sagging (standard
deviation of the error less than 5%).
Instead of defining the design wave from the linear response, Clauss et al. (2013) proposed a method
that combines a nonlinear strip theory solver with an optimization algorithm to find a critical wave
sequence corresponding to a predefined maximum response for the VBM, usually based on a rules value.
The obtained shape of the critical wave sequences has been reproduced in the wave tank and model tests
have been performed to validate the optimization results.
Seng and Juncher Jensen (2012) presented a study of slamming events in conditional waves, under the
assumption of rigid body motions. Based on a time domain nonlinear strip theory, most probable
conditional waves are generated to induce given short term extreme responses in sagging and hogging
VBM at amidships on two container ships. The results of the strip theory are then compared against the
results of free surface N-S/VOF CFD simulations, using OpenFOAM, under the same wave conditions.
When bow flare slamming occurs the strip theory overpredicts the slamming momentum resulting in
higher VBM compared to CFD results. The peak value of the VBM, however, occurs at approximately
119
the same time, implying that the more accurate CFD results can be used to correct the momentum
formulation in strip theory through a correction coefficient.
For the structural analysis, EDWs are used to define the few load cases for which structural checks, e.g.
yielding and buckling, will be performed. It is important to check that these few load cases are
representative of the whole life of the ship. Derbanne et al. (2013a) studied stress response of four
different types of structures, comprising conventional ships and unconventional floating structures, using
various design waves. Computations are carried out using a 3D linear hydrodynamic solver coupled with
a 3D FE solver. It has been shown that with a limited number of design waves, between5 and 10, the
difference between long term linear stress response and EDW stress response is of the order of 10%, for a
large range of floating structures. The selection of the governing parameters is, however, the key point of
the EDW method. If for a beamlike shape the usual design load parameters, such as VBM, HBM, vertical
acceleration at FP and roll angle, are giving good results, it is important for other type of floating
structures to define carefully the relevant governing parameters. For unconventional designs, when it is
impossible to assume a relationship between a load parameter and the stress response, it is even
sometimes necessary to define the EDWs from the linear stress response. In general, the irregular EDWs
are performing better than the regular ones, because they are more consistent with the definition of the
irregular sea states they are representing. They suffer, however, from a slightly higher bias in the response
(underestimate), which can be corrected with a safety factor. On a case by case basis, this approach can be
used to define a group of 5 to 10 design waves that represent the worst load cases for a given floating
structure. For the development of rule load cases, this approach has to be applied to a large number of
ships of different types, different size and different loading conditions. It has to be checked that the same
set of governing load parameters can be used for all these ships to define the worst load cases.
5.3
Design load cases and ultimate strength
Various design loads used in the strength assessment of ship structures have been introduced by
classification societies. Most of these design loads have been determined as standard loads. Hence, the
relationship between design loads and sea states actually encountered by ships appear to be weak.
Members of IACS developed some common unified requirements and recommendations. The most
significant examples are the recommendation n 34 (IACS, 2001) defining the wave statistics to be used
for the design load computations and the unified requirement S11 (IACS, 2010) providing formulations
for VBMs and shear forces.
However, when design loads are computed through direct computations, using the assumptions of
recommendation n 34, some important differences are found, compared to S11. Derbanne et al. (2013b)
compared the nonlinearity of the design VBM (hogging/sagging asymmetry) predicted by the S11
formulations and computed through a nonlinear seakeeping analysis using design waves and design sea
state for 56 different ships. Some important differences were found, especially for containerships, and
new formulations are proposed. Work is currently ongoing in IACS to modify the S11 formulations in
order to close the gap between direct computation results and rule formulations, and to keep the
consistency with the assumptions given in recommendation n 34.
The design loads are then combined to create some load cases, for which structural checks, e.g.
yielding, buckling and fatigue, are to be carried out. These load cases are based on the concept of the
aforementioned EDWs. Zhu and Shigemi ((2003), (2007)) developed practical methods for setting design
loads. A series of calculations were performed using the strip theory approach for a total of 27 tankers and
22 bulk carriers. 38 wave-induced load components such as ship motions, accelerations, sectional forces
and moments and hydrodynamic pressure were considered in this analysis. Dominant loads were
identified and regular design waves were used to define each load case, using wave encounter angle,
wave period and wave amplitude. Four prominent cases are defined as: (i) VBM (head sea), (ii) VBM
(following sea), (iii) Roll (beam sea) and (iv) Hydrodynamic pressure at waterline (beam sea).
Harmonized structural rules are developed by IACS wherein common structural rules of bulk carriers
and tankers are combined (IACS, 2006a, IACS, 2006b, IACS, 2013). Loads for strength and fatigue
assessment are based on recommendation 34 assumptions (IACS, 2001), namely
North Atlantic scatter diagram,

Sea states described by a Pierson-Moskowitz spectrum,
cos2 angular spreading for wave energy and
equal heading probability.
The rule load cases for extreme loads and fatigue loads have been defined based on the EDW concept.
For extreme loads, a design life of 25 years and a ship speed of 5 knots are considered. Seven EDWs have
been defined to generate the dynamic load cases for structural assessment:
120

HSM load cases: head sea EDWs that minimise and maximise the amidships VWBM, respectively;
HSA load cases: head sea EDWs that maximise and minimise the head sea vertical acceleration at FP,
respectively;
FSM load cases: following sea EDWs that minimise and maximise the amidshipsVWBM respectively;
BSR load cases: beam sea EDWs that minimise and maximise the roll motion;
BSP load cases: beam sea EDWs that maximise and minimise the hydrodynamic pressure at the
waterline amidships;
OST load cases: oblique sea EDWs that minimise and maximise the torsional moment at 0.25L from the
AP;
OSA load cases: oblique sea EDWs that maximise and minimise the pitch acceleration.
For fatigue assessment a ship speed of 75% of the design speed is considered. The stress history is
approximated by a two-parameter Weibull distribution. These parameters are scaling factor and shape
parameter. A probability level of 10-2 is selected for determination of the scaling factor, and the shape
factor is taken equal to 1.0. In previous rules, for double-hull oil tankers and bulk carriers, the reference
probability level was 10-4 (IACS, 2006a, IACS, 2006b) The advantages of the 10-2 probability level have
been explained by Derbanne et al. (2011) as follows: the most important contribution to fatigue damage
is due to the stress ranges corresponding to a probability approximately 10-2; if the design stress is directly
defined at 10-2, the influence of the shape factor on the total damage is nearly cancelled; hence, it can then
be taken as a constant. Moreover it has been shown that the EDW method can be used to compute the 10-2
stress.
Five EDWs have been defined to generate the dynamic load cases for structural assessment: HSM,
FSM, BSR, BSP and OST, as defined above. The load cases HSA and OSA, defined above, are not
considered for fatigue assessment since they are redundant with load case HSM.
The approach for combining other responses or subjected loads with the maximized load under the
EDW is obtained by the load combination factors (LCFs). The LCFs have been derived through
direct analysis for a significant number of oil tankers and bulk carriers covering ballast, full load
and intermediate loading conditions.
The hull girder strength is the most critical failure mode for the hull structure. As ships length
becomes longer, this strength is even more important. Hence, the hull girder ultimate strength check is
required for ships with length of more than 150m. The vertical hull girder ultimate bending capacity is to
be checked for hogging and sagging conditions, for the following design load scenarios:
For bulk carriers: design load scenario A, for seagoing, harbour/sheltered water and flooded
conditions.
For oil tankers: design load scenario A, for seagoing and harbour/sheltered water conditions, and
design load scenario B, for the operational seagoing homogeneous full load condition.
Partial safety factors to account for material, geometric and strength prediction uncertainties are
defined for the aforementioned load scenarios.
6.
FATIGUE LOADS FOR SHIPS
With increases in size of container ships, the contribution of high frequency vibrations to the fatigue
strength/damage of the hull girder has been extensively studied in recent years. The peak values of the
VBM and the resulting bending stress increase as a result of the high frequency vibrations due to
whipping and springing of the ship. This observation is based on analysis of full scale measurements. It is
not possible to distinguish springing from whipping in measurements; hence, most publications refer to
the high frequency vibration as whipping, since it is usually considered that springing occurs in relatively
calm seas. This section mainly deals with the effects of high frequency vibrations on the fatigue strength
of ships.
Andersen and Juncher Jensen (2014) analysed the full scale VWBM measurements of a 9400 TEU
container ship in bow quartering seas by using FFT and concluded that the high frequency vibrations
caused by impulsive loads are observed to be of the same magnitude as the rigid body response and, thus,
act so as to double the total VBM amidships in the extreme case, as discussed more extensively in Section
5.1. Storhaug et al. (2012) showed that this high frequency vibration contributed to increase the dynamic
extreme stresses on deck amidships by 22-33% in hogging and 27-32% in sagging by analysing the
maximum values every half hour. Their conclusions were based on the full scale measurements from two
LNG vessels operating in typical worldwide trading patterns for more than 5 years, excluding the TransPacific trade. Based on elastic backbone model test data of two ULCSs, 8600 TEU and 13000 TEU,
respectively, in regular and irregular head waves, Zhu and Moan (2013) showed that the high frequency
vibrations increase the sagging VBMs more significantly than the hogging VBMs, however, the hogging
121
peaks are increased more by the high frequency vibrations in relatively short waves because of the
behaviours of second and third harmonics.
After fatigue damage was identified on a large ore carrier, attributed to whipping, its effect on
longitudinal strength was further examined. Onboard measurements and model tests with flexible models
were performed, and it was shown that the contribution of the rigid body response to the total fatigue
damage was relatively large (ISSC, 2012b). Storhaug (2012) showed that the high frequency vibration
contributes to approximately 26% of the fatigue damage on a 2800 TEU container ship and 29% of the
fatigue damage on a 4400 TEU container ship, based on measurements in the North Atlantic for a few
years. Toyoda et al. (2012) applied the rainflow counting and Miners rule to the full scale measurement
data of two post Panamax container ships taken over a period of 2.5 years and showed that the fatigue
damage is increased 3.9 times due to the effect of whipping. It should be noted, however, that the absolute
value of fatigue damage is quite low, 10-13 level, say. Based on measurement campaigns onboard a
Panamax and a post Panamax container ships, the effect of high frequency hull girder vibrations on
fatigue damage was assessed by Rathje et al. (2012). The contribution of high frequency loads was found
to be 35% of total damage for the Panamax ship, while it was 57% for the post Panamax ship. They
concluded, however, that the screening of damage data bases for Panamax container ships indicated that
no significant amount of damage had occurred on similar ships, even for ships operating worldwide for
more than 20 years. Andersen and Juncher Jensen (2013) estimated the fatigue damage applying the
spectral analysis to the full scale stress measurements onboard the hull of a large container vessel, 9400
TEU, during several months of operation. They concluded that the spectral analysis show satisfactory
agreement with the results from rainflow counting. Fatigue damage was also evaluated using the model
test data and numerical simulation, and the effect of high frequency vibration due to whipping and/or
springing was discussed. Derbanne et al. (2012b) computed linear long term fatigue damage for 6
container ships based on bending moment RAOs. The conclusion is that for vertical bending damage the
hydrostatic restoring effect and the low frequency dynamic amplification effects are cancelling out each
other.
Wang et al. (2013) predicted springing responses of large ships using 3D hydroelastic theory and
model experiments. They showed that the structural fatigue damage induced by combined wave loads is
much more severe than that induced by rigid body loads when wave period is small. Choi et al. (2013)
numerically evaluated the hull-girder responses including the springing effect for an ULCS based on
coupling of Vlasov beam theory and time domain 3D Rankine panel method. They concluded that
hydroelastic effect is not reflected sufficiently in the numerical simulation for harsh sea conditions when
compared with the results of model tests. Koo et al. (2013) carried out the fatigue assessment of an
18000TEU container vessel using a spectral approach and taking into account the springing effect. They
concluded that the fatigue damage due to the springing effect has been between 24% and 64% of total
fatigue damage when carrying out a fatigue damage analysis; however, fatigue damage considering the
springing effects of case studies with measured data is much lower than results from the analysis.
Usually the rainflow counting method of stress cycles, considering the memory effect during elasticplastic material behaviour in notches, as well as the Palmgren-Miner rule, are applied to predict the
fatigue damage under variable amplitude loading, yielding the so called life curve compared to the S-N
curve. In the Palmgren-Miner rule, the damage due to small stress cycles below the knee point of the S-N
curve, i.e. fatigue limit, is usually considered assuming a modified slope exponent of the S-N curve
beyond the fatigue limit, i.e. the so called Haibachs correction. Another parameter affecting fatigue life
prediction is the bandwidth of the stress spectrum, which becomes wider in cases when the load process
contains several frequencies. A rainflow correction factor was introduced to consider bandwidth effects
on rainflow counting and damage summation for different spectra and bandwidth (Fricke and Paetzold,
2013).
(Fricke and Paetzold, 2012) performed fatigue tests using simplified stress histories with superimposed
stress cycles of different frequencies and concluded that almost the whole of the fatigue damage is caused
by the stress cycles induced by wave-frequency (rigid body load) which is enlarged by whipping, whereas
the additional small stress cycles can be neglected. Fricke and Paetzold (2013) also performed fatigue
tests of a welded detail with variable amplitude loading to investigate the effect of whipping stresses on
fatigue damage in ships and the suitability of the Palmgren-Miner rule for fatigue life assessment. Stress
histories comprising of two sinusoidal functions and a stress history obtained from onboard measurements
were applied. The traditional approach with rainflow counting and linear damage accumulation based on
the Palmgren-Miner rule is found to be suitable for ship-typical stress histories with low-frequency stress
cycles due to waves and high-frequency stress cycles due to whipping. Most of the fatigue damage is
caused by the low frequency stress cycles, enlarged by whipping, as long as the whipping stress
amplitudes are smaller than the wave-induced rigid body stress amplitudes. The contribution of the
additional small stress cycles due to whipping is rather small. Osawa et al. (2013) developed new simple
122
fatigue testing machines which can carry out fast and low cost fatigue tests of welded joints subject to
high frequency vibration superimposed on to wave loadings. Fatigue tests of out-of-plane gusset welded
joints subject to springing and whipping, superimposed on to wave loadings were carried out. They also
showed that the enlargement effect of the total stress range has a dominant influence on the fatigue
strength under the conditions chosen. However, they concluded that it is unclear whether similar results
can be found for the cases under various wave loadings with high frequency vibration.
On the other hand, Gotoh et al. (2012) performed fatigue tests of centre-cracked tensile specimens with
simplified stress cycles containing high cycle stresses with/without damping, which is attained by the
transient stresses being 5 times as high as the cyclic stress. Tests are also conducted with the regular low
cycle stresses representing the envelope of the simplified stress cycles. Numerical simulation of fatigue
crack propagation based on an advanced fracture mechanics approach using the RPG (Re-tensile Plastic
zone Generating) load criterion for fatigue crack propagation is conducted and an extracting procedure for
the effective loading sequence during random loading is proposed with the use of RPG load criterion. The
test results were compared with the traditional S-N approach using the more advanced RPG load criterion
which was superior to the ordinary cycle counting method such as rainflow method, because the load
sequence effects on fatigue life can be taken into account. It was found that the fatigue life was
underestimated in some cases, which were mainly attributed to irregular stress cycles (history) or
deceleration/delay in fatigue crack growth when small stress range cycles followed a large stress range
cycle, which is usual in the case of whipping. Furthermore, it was also confirmed from the experimental
results that the traditional fatigue life evaluation method based on the S-N curves approach with rain flow
counting is inappropriate under superposed loading conditions. Although the enlargement effect of stress
amplitude by high frequency component is a very simple way for the consideration of whipping stresses
in fatigue analyses, it is pointed out that uncertainties still exist regarding the real fatigue damage under
typical stress histories containing wave loads and high frequency vibrations.
It is accepted that the stress amplitude of a ships hull girder increases due to whipping and springing
vibrations, and that fatigue damage also increases as a consequence of the enlargement effect due to high
frequency vibration components. This fact is confirmed through comparisons with fatigue test data using
test pieces in several cases. Nevertheless, it is also an undeniable fact that damages attributed to such hull
girder vibrations were rarely reported for well-maintained, properly operated ships. One reason for such
occurrence may be attributed to the nature of the Palmgren-Miner rule. Fukasawa (2012) showed, from
numerical simulation results, that the number of stress cycles increases also in the zero-crossing counting
case, but the increase rate is lower than that corresponding to the local peak counting case. Derbanne et al.
(2011) and Fukasawa and Mukai (2013) showed in their numerical simulations that the dominant stress
range to the fatigue damage is that of relatively small amplitude whose occurrence probability is
approximately 10-2 according to Miners rule, and a certain stress range has a significant effect on the
fatigue damage.
Another reason may be attributed to the ship handling and the seaway conditions encountered. It is
quite usual nowadays to operate a container ship at a speed less than 20 knots to save fuel and the ship
can avoid rough weather due to advanced weather forecasting. The operational effect is not negligible for
the evaluation of the occurrence probability of slamming. Ogawa et al. (2012) investigated the effect of
different sea states on the occurrence probability of whipping-induced stress. Their work was based on
full scale measurements of container ships, and the long term predictions were conducted based on the
measured stresses and wave hindcast data. They showed that the difference of probability in wave height
has significant or large effect on whipping and the consequent magnitude of induced stress. They also
noted that most of slamming induced vibration occurs in head or bow seas. Storhaug (2012) measured
strains and wind speed and direction on 2 container vessels (2800 TEU and 4400 TEU) and showed that
fatigue damage is largest in head seas for the larger vessel and bow quartering seas for the smaller vessel,
and that the relative vibration related damage decreases from head to stern seas. Based on fatigue tests
and fatigue crack growth simulation, Sumi et al. (2013) stated that shipping routes and ship operation,
including influence of ship heading and speed, must be taken into consideration for the proper estimation
of probabilities. Prasetyo et al. (2012) conducted fatigue crack propagation based assessment of a surface
crack in a weld of a longitudinal stiffener using JWA (Japan Weather Association) hindcast data for the
north Pacific Ocean region. Load history is generated based on third generation storm model, which can
take into account fluctuations in storm duration. It was shown that the fatigue damage ratio of second
generation storm model, which can only take into account the relation between wave height and wave
period of the short term sea state, to third generation storm model is almost equal to the ratio of calculated
crack propagation life in each storm. The proper choice of wave environment and seagoing condition is
important in evaluating the effects of whipping and springing.
7.
123
UNCERTAINTY ANALYSIS
This section reviews only uncertainties related to wave-induced loads and loading conditions.
7.1
Load uncertainties
The subject of modelling and implementing uncertainties in maritime technology related areas is vast and
different approaches are used, mostly depending on the aim of the analyses. Thus, Papanikolaou et al.
(2014) presented recent advances in the treatment of uncertainties related to two engineering applications,
namely development of reliability-based code formats and development of modern decision support
systems (DSS) as guidance to a ships master. Whilst both applications of uncertainty modelling share
some common issues, there are also obvious differences. For example, in the former application, there is a
basic uncertainty of design scenario for ship operation, requiring appropriate selection of parameters and
calculation methods. Accordingly, the general practice has been to adopt the North Atlantic wave climate
as a reference condition because of its severity. The counter argument is that real operational scenarios
should be used in probabilistically based design. Even if a ship route, other than the North Atlantic one, is
known at the design stage, ship owners argue that the ship could be sold during her lifetime and possibly
be used in the North Atlantic trading routes (Guedes Soares, 1996).
Such uncertainty is not important for the second application, i.e. the development of DSS, where other
uncertainties dominate. One of the important uncertainties in DSS relates to the definition of threshold or
limiting values of seakeeping parameters that are used to predict the sustainable ship speed and heading in
certain sea conditions. General operability limiting criteria, such as RMS of vertical acceleration at FP,
probability of slamming, probability of green water and probability of propeller emergence are usually
defined as fixed values depending on ship size and type and then used for design purposes (Moan et al.,
2006). However, for probabilistically based DSS more refined seakeeping criteria need to be defined,
preferably as random variables, and then included in the limit state function for evaluation of different
seakeeping hazards (Papanikolaou et al., 2014).
Accuracy of observations or measurements relating to parameters of sea state in which a ship is sailing
is one of the dominating uncertainties in applications of DSS. Nielsen et al. (2013) produced sea state
estimates by three different means: the wave buoy analogy whereby one is relying on onboard response
measurements, a wave radar system, and an onboard system providing the instantaneous wave height. The
presented results show that for the given data, recorded on five different days of continuous operation, the
agreement between the estimated means is reasonable.
The principal global loads causing longitudinal hull bending of ship may be summarized as follows
(Bai, 2003): (i) still-water bending moments, (ii) rigid body (wave frequency or quasi-static) wave
bending moments and (iii) vibratory bending moments caused by whipping and springing. The still water
bending moments result from the longitudinal distribution of the cargo onboard and other controllable
factors which are changing for each departure condition. Although still water loads are deterministic in
nature, during a ships lifetime they fluctuate; hence, they can, and should, be considered as stochastic
process.
Rigid body VWBMs are obtained from standard seakeeping analysis. Primary uncertainties sources in
numerical seakeeping analysis, according to Kim and Hermansky (2014), are :
different mathematical modelling of (initial) boundary value problem,

different numerical modelling of the assumed mathematical model,
non-converged or inaccurate hull geometry modelling,
insufficient or incorrect knowledge regarding mass distribution, and
human, i.e., user error.
In addition, uncertainties at different stages of spectral analysis, which according to Hirdaris et al.
(2014) need to be also considered, are:
type of wave spectra,

the choice of wave scatter diagram,
method for prediction of long term extreme values and
the operation of the ship (human actions/factors).
As the aforementioned uncertainties relate to the linear hydrodynamic tools only, computational
uncertainty of nonlinear effects in wave loads, especially different sagging and hogging bending
moments, should be added to the total uncertainty.
Transient elastic vibrations of the ship hull girder due to whipping represent important contributions
to the VBM of large slender ships, especially ULCSs. Whipping vibrations increase both the extreme
and fatigue load effects occurring in ships. VBMs induced by the combined effect of wave bending
124
and whipping can double the magnitude of the former, as previously discussed in Section 6. Thus,
considerable efforts have been spent on predicting loads associated with slamming and consequent
structural response. Drummen and Holtmann (2014) presented results of benchmark study on
uncertainties in slamming and whipping undertaken in 2012 by ISSC. The ship used in this benchmark
study was 173m long RO/RO ferry. Surprisingly, they noted large discrepancies, even in the estimate
of the two lowest natural frequencies of the global flexural vibration modes. However, it is interesting
to note that overall results, especially for 3-node vibration, are in better agreement to the experiment
for wet than for dry modes. When participants applied different realistic, analytical, pulses to their
model, significant differences in fatigue loading of up to a factor of five were observed. Furthermore,
the computations considering the response to an impulse induced by a regular head wave showed
significant differences between the experiment, the different participants, and applied methods. The
computation of slamming loads was performed i) using 2D slamming sections and BEM, ii) hybrid
approach for slamming pressure computation, where initially vessel motions were computed using a
Rankine panel method and imposing these motions on a RANS model using ANSYS CFX, iii) as per
ii) but the motions were calculated using RANS solver StarCCM+, iv) fully coupled FSI computation
based on RANS solver Comet for fluid dynamics and Timoshenko beam structural representation. The
resulting fatigue damage compared to the experimentally measured loads was in the range of 0.42% to
507%, where experimental fatigue damage reads 100%. The authors concluded that the benchmark
emphasized the need to validate and gain personal experience with dedicated tools.
Corak et al. (2013) carried out a parametric study of the influence of container ship route and
operational and environmental restrictions on the long term distribution of combined, rigid-body and
vibratory bending moments. The problem is formulated in the frequency domain using standard
engineering tools for load computation: a 3D seakeeping code for the rigid body response and a beam FE
model for transient vibratory response. The von Karman approach, with correction for pile-up effect, is
employed for bow flare slamming load assessment. As expected, the IACS rule VWBM is largely
exceeded by the combined VBM for the North Atlantic sea environment. The choice of alternative
shipping routes can significantly reduce the long-term extreme combined VBM. Thus, combined VBMs
could be reduced by 6% in the case of the North Pacific and by 26% in the case of the Far East-North
European shipping route. Voluntary speed reduction has a considerable influence on the total VBM. By
employing the speed profile reduction from ship operability analysis, the combined VBM for the North
Atlantic route could be reduced by 21% for a constant ship speed of 15 knots.
Research related to the whipping effects of container ships has intensified after the accident of the
container ship MSC Napoli in the English Channel in 2007. Although whipping certainly occurred at the
time of the accident, it is still not clear how much it actually contributed to the hull girder failure
(Storhaug, 2009). Parunov et al. (2014), in their investigations of the MSC Napoli, proposed the structural
reliability approach, which enables identification of the relative importance of various pertinent variables
on the structural failure. Sensitivity analysis, which is regular part of the structural reliability study,
showed that the most important random variable at the time of the accident appears to be the modelling
uncertainty in the bending capacity (32%). Such large uncertainty is caused by the uncertainty in the
material yield strength and the modelling uncertainty of the progressive collapse analysis method used to
assess the hull girder vertical bending capacity. However, the overall importance of all variables related to
the uncertainty of the rigid body VWBM (37%) exceeds the importance of the strength uncertainty. The
overall importance of the whipping contribution (24%) is approximately 2/3 of the importance of the rigid
body VWBM.
The importance of uncertainties in design and analysis of large container ships was emphasized once
again due to the case of structural failure of the MOL Comfort container ship in the Indian Ocean in June
2013. The report issued by the Maritime Safety Committee (IMO, 2014a) reveals that the estimated total
global bending loading at the time of the accident was of the order of 9.4 GNm, whilst the estimated
bending moment capacity was 14.0 GNm. This indicates that the estimated load was approximately 67%
of the hull strength and the conditions for fracture could not be simulated. For this reason, the authors of
the report concluded that it would be necessary to consider uncertainties in further verification of both
load and strength related calculations.
7.2
Uncertainties in loading conditions
The uncertainty of the weight and weight distribution and their consequences have not been the subject of
scientific analysis in the usual journals, nor any ISSC committee. Nevertheless, these uncertainties have
been identified by industry as a considerable problem, and not only in the context of the recent ship
accidents of MOL COMFORT and MSC NAPOLI. According to the world shipping council some
customs, namely Ukraine, Poland and India, have investigated container weights. The IMO, in May 2014,
125
approved draft amendments to SOLAS chapter VI requiring mandatory verification of the gross mass of
containers (IMO, 2014b), which were approved in November 2014.
Within the European research project ADOPT (Nielsen and Michelsen, 2007) the variation of
deadloads corresponding to six months loading conditions from a RORO vessel were analysed. The
analysis showed deadloads up to 16% of the displacement, that is up to approximately 1300 tonnes, with
the dominant range from 200 to 900 tonnes. Deadload is the excess between the vessels calculated
deadweight, based on draught readings, and known weights such as cargo, fuel ballast, water etc. The
uncertainty of cargo weight and distribution for RORO, ROPax and heavy lift vessels is mentioned in the
report of the research project lashing@sea but it is concluded that since the accident rate has not
increased, it is not considered a problem for such ship types (Koning, 2009).
Cargo that is heavier than declared is not only a problem for the establishment of maximum allowable
loads on ships but also an important factor in the collapse of container stacks, loss of containers at sea,
stability failure and transportation on shore. One of the issues noted within the lashing@sea project was
that no centralized data base of shipping incidents of cargo loss and damages are kept (Koning, 2009).
This makes an analysis of the extent and severity of the uncertainty of cargo weight difficult.
Within the investigation of major structural failures of container ships, the uncertainty of cargo load
has been evaluated whenever possible. The accident reports of the MSC NAPOLI (MAIB, 2008) stated
that this ship often had large deadloads on completion of loading. The deadload on departure from
Antwerp, prior to the cracking, was about 1250 tonnes (approximately 1.6% of displacement). During the
removal of containers, the position of 700 containers on deck were compared with the position recorded
by the terminal operator. Of these units 53 (7%) were in the wrong position or misdeclared. It is generally
agreed within the container industry that up to 10% of containers loaded onto a vessel might not be in
their planned positions. Since the two parts of the MOL COMFORT sank, the contribution of the
uncertainty of weight and weight distribution cannot be accounted for. The effect of the uncertainty of
container cargo weight on the SWBM, based on the hull deflection obtained through draught
measurements at the time of departure, is difficult to assess (MAIB, 2008).
A submission to IMO (2012) summarizes some incidents due to, amongst others, misdeclaration of
container weights. For example,
Containership Deneb in Algeciras (CPIAIM, 2012): According to this report, the weight of the cargo
used for calculating the stability was 332 tonnes, approximately 9% of displacement, lower than the
calculated weight;
P&O Nedlloyd Genoa (MAIB, 2006): Two containers were overweight, but the results show that the
remaining containers were close to their declared weights.
Based on information obtained from the world shipping council it is known that the Ukrainian customs,
over a two week period in October 2012, weighed all packed containers discharged in Ukrainian ports.
56% of the containers had an actual weight greater than the weight stated in the manifest based on the
shippers declared weight as provided in the shipping instructions.
Based on current evidence, it is difficult to speculate on the consequences of the uncertainties in
loading conditions. In the case of the structural failure of MSC NAPOLI the overloading of the vessel, i.e.
the resulting deviation to the SWBM, would have been extremely small by comparison to the potential
variability of the wave loading. Therefore, the effect of the discrepancies alone would have been
insufficient to cause hull failure. Nevertheless, they would have contributed to the reduction of the safety
margin available (MAIB, 2008). Whether this conclusion holds true for other vessels cannot be
established due to poor and incomplete evidence.
8.
CONCLUSIONS
On Zero speed case: The potential flow model remains, in general, one of the dominated efficient solvers
for numerical simulation of wave-body interactions at zero speed, including the influence of a range of
nonlinearities. Therefore, further development of 3D FNPF solvers is still valuable. The versatility and
increased accuracy of N-S type solvers for capturing important characteristics of wave-body interactions
has been demonstrated. Nonetheless, there are very promising developments in 3D models combining
potential flow solver and N-S solvers, which should continue.
On Forward speed case: Potential flow solutions continue to be dominant also in the case where
forward speed is non-zero, albeit with inclusion of a range of nonlinearities. The focus in these
applications is the generation of practically applicable solution packages, capable of coupling with
phenomena, such as slamming, sloshing and green water. Use of N-S solvers focuses in the main on the
seakeeping problem with improved accuracy of predictions. Hybrid solutions for computational
efficiency, involving nonlinear potential flow and N-S solvers, are amongst the promising developments.
126
It is interesting to note that the state-of-the-art for either zero or forward speed cases is very similar, in spite
of differences in operational and environmental frameworks. Furthermore, the need for measurements to
validate computational methods is paramount for both cases.
On Sloshing: The majority of investigations into sloshing focus on numerical simulations using RANS
CFD methods and experimentation, 3D simulations and realistic excitations for both. Analytical
investigations can be useful in terms of providing simple formulae useful in design. Recent advances
include studies on sloshing with internal suppressing structures and investigation on the coupling between
sloshing and ship motions, both numerically and experimentally. It is expected that these studies will
continue in order to improve the accuracy of numerical simulations.
On Green water/wave-in-deck: The green water problem has been investigated experimentally and
numerically, but with very limited success. The issues arise due to the fact that green water is very
difficult to measure either under laboratory conditions or in the field, and very difficult to simulate
numerically. The problem, due its fast moving, multiphase and highly turbulent nature, is not very
amenable to accurate numerical simulation, although 3D models help to better understand and simulate
the mechanisms of green water impinging on deck. Green water load computation with various wave
headings is still a challenge. Qualitative benchmark studies are required to validate and improve
numerical approaches. The issues for the wave-in-deck problem are similar, but applicable to structures
fixed or moored in an extreme environment.
On Loads from abnormal waves: Research and early findings raise the some critical questions: should
we design future ships for abnormal waves or for the upper tails (essentially beyond 99% quintiles) of
wave statistics and will be these ships economically viable? Here we need to differentiate between
merchant and naval ships, as their mission profiles differ significantly. Also, the design problem of
offshore structures, which cannot avoid extreme environmental loads, is inherently different from that of
ships, for which operational measures and guidance to the master to avoid dangerous environmental loads
appear more efficient than an increase of design loads and safety factors.
On Hydroelasticity: The main progress in hydroelasticity analysis is in the development of BEM based
nonlinear FSI methods through coupling with FE models. Coupling between RANS and FE solvers,
currently used for slamming related problems, is also being applied to predict the rigid body and
springing related loads. The success of the latter will depend on the computational efficiencies made and
the improvements in accuracy obtained with reference to potential flow based predictions.
On Slamming: Prediction of slamming loads has recently covered some aspects that were of secondary
importance in the past with respect to the determination of the pressure peaks over the impacting bow:
water-exit phase, presence of both vertical and horizontal components in the water-entry velocity, oblique
or asymmetric impacts, wavy surface are some of them. Hydroelastic effects have been dealt with
different approaches ranging from semi-analytical methods to coupled finite element and RANS solvers
including 3D fluid-structure interaction and composite materials. Indeed, many of these studies have been
oriented toward analysis of containerships since their flexibility poses challenging questions to numerical
simulations.
On Measurements: Recent advances in measurement techniques and fabrication technology have
indeed broaden the capability to design experimental setups where scaled models can reproduce the fullscale elastic behavior and provide high quality data for code validation. Thus, much effort focused on the
assessment of physical models to be used in FSI tests and their mechanical and structural properties. The
analysis of transient phenomena, such as slamming, has benefited from the application of advanced signal
analysis techniques. Full scale monitoring of environmental conditions and hull structures has also greatly
improved, thus allowing for real ship response investigations at a reasonable cost, as well as assessing our
capability for modeling numerically and experimentally.
On Loads following damage: The motivation for increasing research on damaged ship is the
implementation of risk based methods in rule development, where residual strength of ships in different
foreseen damage situations is required in addition to that of the intact ship. The use of reliability based
methods demands realistic estimates of loads on damaged ship. Thus, the main driving force for further
research, is the development and understanding of both deterministic and probabilistic models for load
assessment in accidental situations.
127
On Weather routing: Development of optimization schemes and simplified mathematical models for
complex hazards, such as fatigue damage form the basis of the research into weather routing. The
necessity of embedding an uncertainty analysis into an operational guidance has been identified, but not
accounted for in most research. The reliability of the guidance, e.g. identifying and mitigating false
predictions, has not been an issue; neither the transmission of information gained to the master, the crew
and the ship operators to enable sound decisionmaking under uncertainties.
On VIV: Although industry practice continues with the use of semi-empirical methods, the pace of
development is picking up, involving both academic and industrial research. Operations in deeper waters,
possibility of collision in riser bundles and fatigue damage and relevant high safety factors used provide
the incentive for such developments in experimentation, simulation and validation studies.
On VIM: The technology dealing with VIM is advancing constantly especially using CFD simulations,
as well as focusing on suppression arrangements. However, in current industry practice model tests are
still required to derive VIM design criteria. A hybrid methodology that considers both full scale CFD
results and model test results may have potential use in the VIM design assessment instead of using
model test results only.
On Mooring Systems: Research focuses on a range of methods for coupling the dynamics of mooring
lines, floaters and risers and assessing the validity of partly, fully coupling methods, as well as frequency,
time and hybrid frequency/time domain approaches. In addition research into different materials, and their
properties, used in mooring lines is of great importance in terms of including in the dynamic simulations
and assessment of fatigue damage.
On Lifting Operations: The review of this subject clearly shows the progress achieved in the last few
decades in coupling the payload and lifting vessel dynamics. Current investigations focus on nonlinear
potential flow hydrodynamics and measurements for validation. There is room for improvement on the
nonlinear system and environment characteristics used.
On FOWT: Development of simulations tools for FOWT systems benefited from past experience by
combining wind turbine aerodynamic/elastic and offshore structure hydrodynamic and station-keeping
methods and tools. With the quest for higher energy recovery and the use of larger turbines, an increase in
experimental validation and refinements in simulation tools are expected.
On Probabilistic methods: These are, by and large, employed to study short and long term distributions
of ship responses. Load combination studies of still water and a range of wave loads are also performed.
Although a significant amount of research is dedicated to the design-oriented probabilistic methods for
prediction of extreme combined wave-induced and vibratory responses, we are of the opinion that
practical methods, applicable for conceptual and preliminary ship design, are still lacking.
On Design methods for ships: The focus of investigations has been on developing and verifying
equivalent design waves. The selection of the characteristics of such waves, such as increased sea state
significant wave height, aids in cutting down the amount of simulations required.
It is expected that the rule based approach and the direct calculation approach should give same answer
in terms of the acceptance of the particular ship structural design in the near future.
On Fatigue loads for ships: Crack initiation and propagation behaviours are nonlinear phenomena and
there may be some limitations in the application of Miners rule in terms of accuracy of estimated fatigue
damage. Further investigations are necessary to clarify the effect of hull girder vibration upon the fatigue
strength of a ship by accounting for stress histories which are effective in crack initiation and propagation.
Full scale measurements, model tests and numerical simulations, such as the aforementioned crack
propagation analyses, should be combined to clarify the fatigue life characteristics of a ship in waves.
On Uncertainties: The importance of uncertainties can be seen in comparative studies predicting ship
motion and structural loads, such as that on a large containership arranged by the 2nd ITTC-ISSC Joint
Workshop in 2014. We are of the opinion that systematic uncertainty assessment is one of the main issues
requiring resolution for the advancement of marine structural design. Development of sophisticated
computational tools should be followed by uncertainty analysis of their predictions, including operational
uncertainties as one of the main contributing factors in definition of design loads.
128
Final comment: The review in the reporting period indicates two specific trends for fluid-structure
interaction problems: (i) a systems approach which should be extended to include coupling between
propulsion/control devices, in ships, and mooring/riser systems, in marine structures, to enable a more
accurate and complete evaluation of loads; (ii) the pursuit of hybrid solutions for computational
efficiency, which should continue and expand towards data fusion, namely solution systems bringing
together a range of computational predictions and measurements, and making use of data-rich and datasparse analyses.
REFERENCES
Aarsnes, J. V. 1996. Drop test with ship sections-effect of roll angle. Report 603834.00.01, Norwegian Marine
Technology Research Institute, Trondheim.
Abbasnia, A. & Ghiasi, M. 2014. A fully nonlinear wave interaction with an array of submerged cylinders by
NURBS numerical wave tank and acceleration potential. Ships and Offshore Structures 9, 404417.
Abdussamie, N., Thomas, G., Amin, W. & Ojeda, R. 2014. Wave-in-deck forces on fixed horizontal decks of
offshore platforms. In Proc. 33rd Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE, San Francisco,
USA.
Abrahamsen, B. & Faltinsen, O. 2011. The effect of air leakage and heat exchange on the decay of entrapped air
pocket slamming oscillations. Physics of Fluids 22, 102107.
Abrahamsen, B. & Faltinsen, O. 2012a. A numerical model of an air pocket impact during sloshing. Applied Ocean
Research, 37, 5471.
Abrahamsen, B. & Faltinsen, O. 2013. Scaling of entrapped gas pocket slamming events at dissimilar Euler number.
Journal of Fluids and Structures 40, 246254.
Abrahamsen, B. & Faltinsen, O. M. 2012b. The natural frequency of the pressure oscillations inside a water-wave
entrapped air pocket on a rigid wall. Journal of Fluids and Structures 35, 200212.
ABS 2014a. Guide for Certification of Lifting Appliances. American Bureau of Shipping.
ABS 2014b. Rules for Building and Classing Offshore Support Vessels. American Bureau of Shipping.
Adam, F., Myland, T., Dahlhaus, F. & Gromann, J. 2014. Scale tests of the GICON-TLP for wind turbines. In
Proc. 33rd Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE, San Francisco, USA.
Adam, F., Steinke, C., Dahlhaus, F. & Gromann, J. 2013. GICON-TLP for wind turbinesValidation of
calculated results. In Proc. 32nd Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE, Nantes, France.
Aksnes, V., Nyb, T. & Lie, H. 2013. A possible failure mode for leeward mooring lines on a floating storage unit.
In Proc. 32nd Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE, Nantes, France.
Akyildiz, H., Erdem Unal, N. & Aksoy, H. 2013. An experimental investigation of the effects of the ring baffles on
liquid sloshing in a rigid cylindrical tank. Ocean Engineering 59, 190197.
Andersen, E. A. & Juncher Jensen, J. 2013. Hull girder fatigue damage estimations of a large container vessel by
spectral analysis. In Proc. 12th Int. Symp. on Practical Design of Ships and Other Floating Structures PRADS,
Changwon, Korea.
Andersen, I. M. A. & Juncher Jensen, J. 2012. On the effect of hull girder flexibility on the vertical wave bending
moment for ultra large container vessels. In Proc. 31st Int. Conf. on Ocean, Offshore and Arctic Engineering
OMAE, Rio de Janeiro, Brazil.
Andersen, I. M. A. & Juncher Jensen, J. 2014. Measurements in a container ship of wave induced hull girder stresses
in excess of design values. Marine Structures 37, 5885.
Antuono, M., Bouscasse, B., Colagrossi, A. & Lugni, C. 2012. Two-dimensional modal method for shallow-water
sloshing in rectangular basins. Journal of Fluid Mechanics, 700, 419440.
API 2008. Design and Analysis of Stationkeeping Systems for Floating Structures. API Recommended Practice 2SK.
3rd Edition. American Petroleum Institute.
API 2012. Offshore Pedestal-mounted Cranes. API Specification 2C American Petroleum Institute.
Aranha, J. A. P. 1994. A formula for wave damping in the drift of floating bodies. Journal of Fluid Mechanics, 275.
Ariyarathne, K., Chang, K. A. & Mercier, R. 2012. Green water impact pressure on a three-dimensional model
structure. Experimental Fluids 53, 18791894.
Bachynski, E. & Moan, T. 2013. Hydrodynamic modeling of tension leg platform wind turbines. In Proc. 32nd Int.
Conf. on Ocean, Offshore and Arctic Engineering OMAE, Nantes, France.
Bae, Y. H., Kim, M. H. & Shin, Y. S. 2010. Rotor-floater-mooring coupled dynamic analysis of mini TLP-type
offshore floating wind turbines. In Proc. 29th Int. Con. on Ocean, Offshore and Arctic Engineering.
Bai, W. & Eatock Taylor, R. 2006. Higher-order boundary element simulation of fully nonlinear wave radiation by
oscillating vertical cylinders. Applied Ocean Research 28, 247265.
Bai, W., Feng, X., Eatock Taylor, R. & Ang, K. K. 2014a. Fully nonlinear analysis of near-trapping phenomenon
around an array of cylinders. Applied Ocean Research 44, 7181.
Bai, W., Hannan, M. A. & Ang, K. K. 2014b. Numerical simulation of fully nonlinear wave interaction with
submerged structures: fixed or subjected to constrained motion. Journal of Fluids and Structures 49, 534553.
Bai, Y. 2003. Marine Structural Design. Elsevier.
Barhoumi, M. & Storhaug, G. 2013. Assessment of whipping and springing on a large container vessel. In Proc.
12th Int. Symp. on Practical Design of Ships and Other Floating Structures PRADS, Changwon, Korea.
129
Bennett, S. S., Hudson, D. A. & Temarel, P. 2012a. A comparison of abnormal wave generation techniques for
experimental modelling of abnormal wave-vessel interactions. Ocean Engineering 51, 3448.
Bennett, S. S., Hudson, D. A. & Temarel, P. 2013. The influence of forward speed on ship motions in abnormal
waves: Experimental measurements and numerical predictions. Journal of Fluids and Structures 39, 154172.
Bennett, S. S., Hudson, D. A. & Temarel, P. 2014. Global wave-induced loads in abnormal waves: Comparison
between experimental results and classification society rules. Journal of Fluids and Structures 49, 495515.
Bennett, S. S., Hudson, D. A., Temarel, P. & Price, W. G. 2012b. The influence of abnormal waves on global waveinduced loads. In Proc. 6th Int. Conf. on Hydroelasticity in Marine Technology, Tokyo, Japan.
Bitner-Gregersen, E., Hrte, T., Eide, L. & Vanem, E. 2014. Impact of climate change and extreme waves on tanker
design. SNAME Annual Meeting, Houston, USA.
Bossanyi, E. A. 2009. GH Bladed user manual. Garrad Hassan.
Bourdier, S. & Chaplin, J. R. 2012. Vortex-induced vibrations of a cylinder on elastic supports with end stops, Part
1: Experimental results. Journal of Fluids and Structures 29, 6278.
Bouscasse, B., Colagrossi, A., Marrone, S. & Antuono, M. 2013. Nonlinear water wave interaction with floating
bodies in SPH. Journal of Fluids and Structures 42, 112129.
Branner, K., Blasques, J. P. A. A., Kim, T., Fedorov, V., Berring, P., Bitsche, R. & Berggreen, C. 2012. Anisotropic
beam model for analysis and design of passive controlled wind turbine blades. DTU Wind Energy Report E0005.
Brard, R. 1948. Introduction l'tude thorique du tangage en marche. Bulletin de l'ATMA 47, 455479.
Buchner, B. & van den Berg, J. 2013. Non-linear wave reflection along the side of ships leading to green water on
deck. In Proc. 12th Int. Symp. on Practical Design of Ships and Other Floating Structures PRADS, Changwon,
Korea.
Bunnik, T. 2014. A simulation approach for large relative motions of multi-body offshore operations in waves. In
Proc. 33rd Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE, San Francisco, USA.
Buri, Z., Buani Primorac, B. & Parunov, J. 2012. Residual strength of damaged oil tanker in the Adriatic Sea. In
Proc. 17th Int. Conf. on Ships and Shipping Research NAV, Naples, Italy.
Carvalho, D. F. & Rossi, R. R. 2014. Green water loads determination for FPSOs exposed to beam sea conditions.
In Proc. 33rd Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE, San Francisco, USA.
Cha, J. H., Park, K. P. & Lee, K. Y. 2010. Dynamic response simulation of a heavy cargo suspended by a floating
crane based on multi-body system dynamics. Ocean Engineering 37, 12731291.
Chapchap, A. C., Miao, S. H., Temarel, P. & Hirdaris, S. E. 2012. Time domain hydroelasticity analysis: The threedimensional linear radiation problem. In Proc. 6th Int. Conf. on Hydroelasticity in Marine Technology, Tokyo,
Japan.
Chaplin, J. R. & Batten, W. M. 2014. Simultaneous wake-induced and vortex-induced vibrations of a cylinder with
two degrees of freedom in each direction. Journal of Offshore Mechanics and Arctic Engineering 136.
Chaplin, J. R., Retzler, C. H. & Rainey, R. C. T. 1999. Waves generated by a vertical cylinder moving in still waves.
In Proc. 14th Int. Wkshp. on Water Waves & Floating Bodies IWWWFB, Port Huron, USA.
Chen, L. F., Zang, J., Hillis, A. J., Morgan, G. C. J. & Plummer, A. R. 2014. Numerical investigation of wavestructure interaction using OpenFOAM. Ocean Engineering 88, 91109.
Chen, W. L., Xin, D. B., Xu, F., Li, H., Ou, J. P. & Hu, H. 2013a. Suppression of vortex-induced vibration of a
circular cylinder using suction-based flow control. Journal of Fluids and Structures 42, 2539.
Chen, Z., Zong, Z., Li, H. & Li, J. 2013b. An investigation into the pressure on solid walls in 2D sloshing using SPH
method. Ocean Engineering 69.
Choi, J. H., Jung, B. H. & Hwang, J. H. 2013. Evaluation of springing-induced fatigue damage for ultra-large
container carrier. In Proc. 23rd Int. Offshore and Polar Engineering Conference ISOPE, Alaska, USA.
Christiansen, N. H., Voie, P. E. T., Hgsberg, J. & Sdhl, N. 2013. Efficient mooring line fatigue analysis using a
hybrid method time domain simulation scheme. In Proc. 32nd Int. Conf. on Ocean, Offshore and Arctic
Engineering OMAE, Nantes, France.
Churchfield, M. J., Lee, S., Michalakes, J. & Moriarty, P. J. 2012. A numerical study of the effects of atmospheric
and wake turbulence on wind turbine dynamics. Journal of Turbulence 13.
Clauss, G. F., Klein, M., Guedes Soares, C. & Fonseca, N. 2013. Response based identification of critical wave
scenarios. Journal of Offshore Mechanics and Arctic Engineering 135.
Clauss, G. F. & Vannahme, M. 1999. An experimental study of the nonlinear dynamics of floating cranes. In Proc.
9th Int. Offshore and Polar Engineering Conference ISOPE, France.
Constantinides, Y. & Zhang, M. 2014. VIV assessment of deepwater lazy-wave risers. In Proc. 33rd Int. Conf. on
Ocean, Offshore and Arctic Engineering OMAE, San Francisco, USA.
Corak, M., Parunov, J. & Soares, C. G. 2013. Long-term prediction of combined wave and whipping bending
moments of container ships. Ships and Offshore Structures.
Cordle, A. & Jonkman, J. 2011. State of the art in floating wind turbine design tools. In Proc. 21st Int. Offshore and
Polar Engineering Conference ISOPE, Hawaii, USA.
Coulling, A. J., Goupee, A. J., Robertson, A., Jonkman, J. & Dagher, H. 2012. Validation of a FAST
semisubmersible floating wind turbine model with deep wind test data. Journal of Renewable and Sustainable
Energy 5.
Cozijn, J. L., van der Wal, R. J. & Dunlop, C. 2008. Model testing and complex numerical simulations for offshore
installation. In Proc. 18th Int. Offshore and Polar Engineering Conference ISOPE, Canada.
CPIAIM 2012. Investigation of the capsizing of merchant vessel DENEB at the Port of Algeciras on 11 June 2011.
Technical report A-20/2012 Comisin Permanente de Investigacin de Accidentes e Incidentes Martimos.
130
Dankowski, H. & Krger, S. 2013. Progressive flooding assessment of the intermediate damage cases as an
extension of a Monte-Carlo based damage stability method. In Proc. 12th Int. Symp. on Practical Design of Ships
and Other Floating Structures PRADS, Changwon, Korea.
Das, K. & Batra, R. C. 2011. Local water slamming impact on sandwich composite hulls. International Journal of
Multiphysics 6, 305339.
Das, S. & Cheung, K. F. 2012. Hydroelasticity of marine vessels advancing in a seaway. Journal of Fluids and
Structures 34, 271290.
de Hauteclocque, G., Derbanne, Q. & El Gharbaoui, A. 2012. Comparison of different design waves with spectral
analysis. In Proc. 31st Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE, Rio de Janeiro, Brazil.
de Hauteclocque, G., Derbanne, Q. & Mienahou, T. 2013. Nonlinearity of extreme vertical bending moment comparison of design wave approaches and short term approaches. In Proc. 32nd Int. Conf. on Ocean, Offshore
and Arctic Engineering OMAE, Nantes, France.
Dec, A. & Frangopol, D. M. 2013. Risk-informed optimal routing of ships considering different damage scenarios
and operational conditions. Reliability Engineering and System Safety 119, 126140.
Delitala, A. M. S., Gallino, S., Villa, L., Lagouvardos, K. & Drago, A. 2010. Weather routing in long-distance
Mediterranean routes. Theoretical and Applied Climatology 102(12), 125137.
Derbanne, Q., Bigot, F. & de Hautecloque, G. 2012a. Comparison of design wave approach and short term approach
with increased wave height in the evaluation of whipping induced bending moment. In Proc. 31st Int. Conf. on
Ocean, Offshore and Arctic Engineering OMAE, Rio de Janeiro, Brazil.
Derbanne, Q., de Hautecloque, G., El Gharbaoui, A. & de Belizal, P. 2013a. Design wave selection for strength
assessment of floating structures. In Proc. 12th Int. Symp. on Practical Design of Ships and Other Floating
Structures PRADS, Changwon, Korea.
Derbanne, Q., de Hautecloque, G. & Mienahou, T. 2013b. Nonlinearity of extreme vertical bending moment :
Discussion on the existing rule formulation. In Proc. 32nd Int. Conf. on Ocean, Offshore and Arctic Engineering
OMAE, Nantes, France.
Derbanne, Q., Rezende, F., de Hauteclocque, G. & Chen, X. B. 2011. Evaluation of rule-based fatigue design loads
associated at a new probability level. In Proc. 21st Int. Offshore and Polar Engineering Conference ISOPE,
Hawaii, USA.
Derbanne, Q., Sireta, F. X., Bigot, F. & de Hauteclocque, G. 2012b. Discussion on hydroelastic contribution to
fatigue damage of containerships. In Proc. 6th Int. Conf. on Hydroelasticity in Marine Technology, Tokyo, Japan.
Dessi, D. 2013a. Reconstruction of the experimental slamming force distribution based on POD. In Proc. 32nd Int.
Conf. on Ocean, Offshore and Arctic Engineering OMAE, Nantes, France.
Dessi, D. 2013b. Damping of ship global modes: Techniques and analysis. In Proc. 12th Int. Symp. on Practical
Design of Ships and Other Floating Structures PRADS, Changwon, Korea.
Dessi, D. 2014a. Whipping-based criterion for the identification of slamming events. International Journal of Naval
Architecture & Ocean Enginering 6(4).
Dessi, D. 2014b. Load field reconstruction with a combined POD and integral spline approximation technique.
Mechanical Systems and Signal Processing 2, 422467.
Dessi, D. & Ciappi, E. 2013. Slamming clustering on fast ships: From impact dynamics to global response analysis.
Ocean Engineering 62, 110122.
DNV 2010. Modelling and Analysis of Marine Operations (DNV-RP-H103). Det Norske Veritas recommended
practice report.
Downes, J., Moore, C., Incecik, A., Stumpf, E. & McGregor, J. 2007. A method for the quantitative assessment of
performance of alternative designs in the accidental condition. In Proc. 10th Int. Symp. on Practical Design of
Ships and Other Floating Structures PRADS, Texas, USA.
Drummen, I. & Holtmann, M. 2014. Benchmark study of slamming and whipping. Ocean Engineering 86, 310.
Duarte, T., Tomas, D., Matha, D., Sarmento, A. & Schuon, F. 2013. Verification of engineering modeling tools for
floating offshore wind turbines. In Proc. 32nd Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE,
Nantes, France.
Ducrozet, G., Bingham, H. B., Engsig-Karup, A. P., Bonnefoy, F. & Ferrant, P. 2012b. A comparative study of two
fast nonlinear free-surface water wave models. International Journal for Numerical Methods in Fluids 69, 1818
1834.
Ducrozet, G., Bonnefoy, F., Touze, D. L. & Ferrant, P. 2012a. A modified high-order spectral method for
wavemaker modeling in a numerical wave tank. European Journal of Mechanics B-Fluids 34, 1934.
Ducrozet, G., Engsig-Karup, A. P., Bingham, H. B. & Ferrant, P. 2014. A non-linear wave decomposition model for
efficient wave-structure interaction. Part A: Formulation, validations and analysis. Journal of Computational
Physics 257, 863883.
El Malki Alaoui, A., Neme, A., Tassin, A. & Jacques, N. 2012. Experimental study of coefficients during vertical
water entry of rigid shapes at constant speeds. Applied Ocean Research 37, 183197.
Ellermann, K. & Kreuzer, E. 2003. Nonlinear dynamics in the motion of floating cranes. Multibody System
Dynamics 9, 377387.
Ellermann, K., Kreuzer, E. & Markiewicz, M. 2002. Nonlinear dynamics of floating cranes. Nonlinear Dynamics
27, 107183.
Engsig-Karup, A. P., Bingham, H. B. & Lindberg, O. 2009. An efficient flexible-order model for 3D nonlinear
water waves. Journal of Computational Physics 228, 21002118.
131
Engsig-Karup, A. P., Madsen, M. G. & Glimberg, S. L. 2012. A massively parallel GPU-accelerated model for
analysis of fully nonlinear free surface waves. International Journal for Numerical Methods in Fluids 70, 2036.
EWEA 2013. Deep Water: The next Step for Offshore Wind Energy. Report. European Wind Energy Association.
Faltinsen, O. M., Firoozkoohi, R. & Timokha, A. N. 2011. Steady-state liquid sloshing in a rectangular tank with a
slat-type screen in the middle: Quasilinear modal analysis and experiments. Physics of Fluids 23, 042101.
Faltinsen, O. M., Rognebakke, O. F. & Timokha, A. N. 2003. Resonant three dimensional nonlinear sloshing in a
square base basin. Journal of Fluid Mechanics 487(1), 142.
Faltinsen, O. M. & Timokha, A. 2012. On sloshing modes in a circular tank. Journal of Fluid Mechanics 695, 467
477.
Faltinsen, O. M. & Timokha, A. 2013. Multimodal analysis of weakly nonlinear sloshing in a spherical tank.
Journal of Fluid Mechanics 719, 129164.
Fontaine, E. R., Rosen, J., Marcollo, H., Vandiver, J. K., Triantafyllou, M., Resvanis, T. L., Larsen, C. M.,
Tognarelli, M. A., Oakley, O. H., Constantinides, Y. & Johnstone, D. R. 2013. Using model test data to assess
VIV factor of safety for SCR and TTR in GOM. In Proc. 32nd Int. Conf. on Ocean, Offshore and Arctic
Fricke, W. & Paetzold, H. 2012. Experimental investigation of the effect of whipping stresses on the fatigue life of
ships. In Proc. Int. Marine Design Conference IMDC, Glasgow, UK.
Fricke, W. & Paetzold, H. 2013. Experimental investigations on fatigue damage of ship structures caused by
whipping stresses. In Proc. 12th Int. Symp. on Practical Design of Ships and Other Floating Structures PRADS,
Changwon, Korea.
Fu, X. & Qin, Z. 2014. Calculation of the added mass matrix of water impact of elastic wedges by the discrete
vortex method. Journal of Fluids and Structures 44, 316323.
Fukasawa, T. 2012. Some considerations on the effect of wave-induced vibrations upon hull-girder fatigue strength
of a Post-Panamax container ship. In Proc. 6th Int. Conf. on Hydroelasticity in Marine Technology, Tokyo,
Japan.
Fukasawa, T. & Mukai, K. 2013. On the effects of hull-girder vibration upon fatigue strength of a post-panamax
container ship disaggregated by short-term sea state. In Proc. 12th Int. Symp. on Practical Design of Ships and
Other Floating Structures PRADS, Changwon, Korea.
Godderidge, B., Turnock, S. R. & Tan, M. 2012a. Evaluation of a rapid method for the simulation of sloshing in
rectangular and octagonal containers at intermediate filling levels. Computers & Fluids 57, 124.
Godderidge, B., Turnock, S. R. & Tan, M. 2012b. A rapid method for the simulation of sloshing using a
mathematical model based on the pendulum equation. Computers & Fluids 57, 163171.
Gonalves, R. T., Rosetti, G. F., Franzini, G. R., Meneghini, J. R. & Fujarra, A. L. C. 2012. Two-degree-of-freedom
vortex-induced vibration of circular cylinders with very low aspect ratio and small mass ratio. Journal of Fluids
and Structures 39, 237257.
Gotoh, H., Khayyer, A., Ikari, H., Arikawa, T. & Shimosako, K. 2014. On enhancement of incompressible SPH
method for simulation of violent sloshing flows. Applied Ocean Research 46, 104115.
Gotoh, K., Matsuda, K. & Kitamura, O. 2012. Numerical simulation of fatigue crack propagation under superposed
loading histories with two different frequencies. In Proc. 6th Int. Conf. on Hydroelasticity in Marine Technology,
Tokyo, Japan.
Greco, M., Bouscasse, B. & Lugni, C. 2012. 3D seakeeping analysis with water on deck and slamming Part 2:
Experiments and physical investigation. Journal of Fluids and Structures 33, 148179.
Greco, M., Colicchio, G., Lugni, C. & Faltinsen, O. M. 2013. 3D domain decomposition for violent wave-ship
interactions. International Journal for Numerical Methods in Engineering 95, 661684.
Greco, M. & Lugni, C. 2012. 3D seakeeping analysis with water on deck and slamming Part 1: Numerical solver.
Grue, J. & Biberg, D. 1993. Wave forces on marine structures with small forward speed in water of restricted depth.
Applied Ocean Research 17.
Gu, J., Vitola, M., Coelho, J., Pinto, W., Duan, M. & Levi, C. 2013. An experimental investigation by towing tank
on VIV of a long flexible cylinder for deepwater riser application. Journal of Marine Science Technology 18,
358369.
Guedes Soares, C. 1996. On the definition of rule requirements for wave induced vertical bending moments. Marine
Guerber, E., Benoit, M., Grilli, S. T. & Buvat, C. 2012. A fully nonlinear implicit model for wave interactions with
submerged structures in forced or free motion. Engineering Analysis with Boundary Elements 36, 11511163.
Guevel, P., Bougis, J. & Hong, D. C. 1979. Formulation du problme des oscillations des corps flottants anims
d'une vitesse de route moyenne constante et sollicits par la houle. Summary of 4me Congrs Franais de
Mcanique, Nancy, France.
Gueydon, S., Duarte, T., Jonkman, J., Bayati, I. & Sarmento, A. 2014. Comparison of second order loads on a
semisubmersible floating wind turbine. In Proc. 33rd Int. Conf. on Ocean, Offshore and Arctic Engineering
OMAE, San Francisco, USA.
Guo, B. J., Steen, S. & Deng, G. B. 2012a. Seakeeping prediction of KVLCC2 in head waves with RANS. Applied
Ocean Research 35, 5667.
Guo, L. C., Zhang, S., Morita, K. & Fukuda, K. 2012b. Fundamental validation of the finite volume particle method
for 3D sloshing dynamics. International Journal for Numerical Methods in Fluids 68(1), 117.
GWEC 2014. Global wind reportannual market update 2013. Global Wind Energy Council.
132
Hall, M., Buckham, B. & Crawford, C. 2013. Evaluating the importance of mooring line model fidelity in floating
offshore wind turbine simulations. Wind Energy 17(12), 18351853.
Hanninen, S. K., Mikkola, T. & Matusiak, J. 2012. On the numerical accuracy of the wave load distribution on a
ship advancing in short and steep waves. Journal of Marine Science and Technology 17, 125138.
Hanninen, S. K., Mikkola, T. & Matusiak, J. 2014. Computational and experimental study on local ship loads in
short and steep waves. Journal of Marine Science and Technology 19, 103115.
He, G. & Kashiwagi, M. 2014. A time-domain higher-order boundary element method for 3D forward-speed
radiation and diffraction problems. Journal of Marine Science and Technology 19, 228244.
Heege, A., Betran, J. & Radovcic, Y. 2007. Fatigue load computation of wind turbine gearboxes by coupled finite
element, multibody system and aerodynamic analysis. Wind Energy 10(5), 395413.
Higuera, P., Lara, J. L. & Losada, I. J. 2013a. Realistic wave generation and active wave absorption for NavierStokes models application to OpenFOAM. Coastal Engineering 71, 102118.
Higuera, P., Lara, J. L. & Losada, I. J. 2013b. Simulating coastal engineering processes with OpenFOAM. Coastal
Engineering 71, 119134.
Higuera, P., Lara, J. L. & Losada, I. J. 2014. Three-dimensional interaction of waves and porous coastal structures
using OpenFOAM. Part : Formulation and validation. Coastal Engineering 83, 243258.
Hirdaris, S., Argiryiadis, K., Bai, W., Dessi, D., A, E., Fonseca, N., Gu, X., Hermundstad, O. A., Huijsmans, R.,
Iijima, K., Nielsen, U. D., Papanikolau, A., Parunov, J. & Incecik, A. 2014. Loads for use in the design of ships
and offshore structures. Ocean Engineering 78, 131174.
Holmes, S. 2014. Modeling vortex induced motion in sheared currents. In Proc. 33rd Int. Conf. on Ocean, Offshore
and Arctic Engineering OMAE, San Francisco, USA.
Holmes, S. & Constantinides, Y. 2014. CFD modeling of long risers with buoyancy modules and complex shapes.
Hong, D. C. 2000. Application of the improved Green integral equation to the radiation-diffraction problem for a
floating ocean structure in waves and current. International Journal of Ocean Engineering and Technology 3(1),
1422.
Hong, D. C., Hong, S. Y., Lee, G. J. & Shin, M. S. 2014. Influence of the waterline integral on the solution of the
frequency-domain forward-speed radiation-diffraction problem of a ship advancing in waves. In Proc. 33rd Int.
Conf. on Ocean, Offshore and Arctic Engineering OMAE, San Francisco, USA.
Hong, D. C., Hong, S. Y., Nam, B. W. & Hong, S. W. 2013a. Comparative numerical study of repulsive drift forces
and gap resonances between two vessels floating side-by-side in proximity in head seas using a discontinuous
HOBEM and a constant BEM with boundary matching formulation. Ocean Engineering 72, 331343.
Hong, D. C., Hong, S. Y., Sung, H. G. & Nam, B. W. 2013b. Influence of the waterline integral on the solution of
the time-domain forward-speed radiation-diffraction problem. In Proc. 12th Int. Symp. on Practical Design of
Ships and Other Floating Structures PRADS, Changwon, Korea.
Hu, Z., Tang, W., Xue, H. & Ren, S. 2014. Response of beams under the impact of freak waves. In Proc. 33rd Int.
Conf. on Ocean, Offshore and Arctic Engineering OMAE, San Francisco, USA.
Hu, Z. Z., Causon, D. M., Mingham, C. G. & Qian, L. 2013. A Cartesian cut cell free surface capturing method for
3D water. International Journal for Numerical Methods in Engineering 71, 12381259.
Hua, C., Fang, C. & Cheng, J. 2011. Simulation of fluid-solid interaction on water ditching of an airplane by ALE
method. Journal of Hydrodynamics 23, 637642.
Huang, W., Liu, H., Lian, Y. & Li, L. 2012. Modeling nonlinear creep and recovery behaviors of synthetic fiber
ropes for deepwater moorings. Applied Ocean Research 39, 113120.
Huijs, F., de Ridder, E. J. & Savenije, F. 2014. Comparison of model tests and coupled simulations for a semisubmersible floating wind turbine. In Proc. 32nd Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE,
Nantes, France.
IACS 2001. Recommendation 34: Standard Wave Data. International Association of Classification Societies.
IACS 2006a. Common Structural Rules for Bulk Carriers. International Association of Classification Societies.
IACS 2006b. Common Structural Rules for Double Hull Oil Tankers. International Association of Classification
Societies.
IACS 2010. Unified Requirement S11: Longitudinal Strength Standard. International Association of Classification
Societies.
IACS 2013. Common Structural rules for Bulk Carriers and Oil Tankers. International Association of Classification
Societies.
Iafrati, A., Grizzi, S., Siemann, M. & Bentez Montans, L. 2014. Experimental analysis of the water entry of a plate
at high horizontal speed. In Proc. 30th Symp. Naval Hydrodynamics, Tasmania, Australia.
IEC 2005. Wind Turbines - Part 1: Design Requirements. Report 614001, 3rd edition. International
Electrotechnical Commission.
IEC 2009. Wind Turbines - Part 3: Design Requirements for Offshore Wind Turbines. Report 614003, 1st edition.
International Electrotechnical Commission.
Iijima, K., Suzaki, Y. & Fujikubo, M. 2014. Scaled model tests for the post-ultimate strength collapse whipping
loads behaviour of a ship's hull girder under whipping loads. Ships and Offshore Structures.
Ikeda, C. M. & Judge, C. Q. 2014. Slamming impacts of hydrodynamically supported craft. In Proc. 33rd Int. Conf.
on Ocean, Offshore and Arctic Engineering OMAE, San Francisco, USA.
Ikeda, T., Ibrahim, R. A., Harata, Y. & Kuriyama, T. 2012. Nonlinear liquid sloshing in a square tank subjected to
obliquely horizontal excitation. Journal of Fluid Mechanics 700, 304328.
133
IMO 2012. Development of measures to prevent loss of containers - verification of container weight. Submission by
Denmark, Netherlands, the United States, BIMCO, the International Association of Port and Harbours (IAPH),
the International Chamber of Shipping (ICS), the International Transport Workers Federation (ITF) and the
World Shipping Council (WSC). International Maritime Organisation IMO DSC 17/INF.5. .
IMO 2014a. Interim report on large containership safety in response to the loss of MV MOL COMFORT.
International Maritime Organisation MSC 93/INF.14.
IMO 2014b. Maritime Safety Committee (MSC), 93rd session, 14 to 23 May 2014. International Maritime
Organisation.
Ingram, G. 2005. Wind turbine Blade Analysis using the Blade Element Momentum Method. Version 1.0. School of
Engineering, Durham University, UK.
Ishii, E., Kobayashi, E., Mizunoe, T. & Maki, A. 2010. Proposal of new-generation route optimization technique for
an oceangoing vessel. In Proc. of OCEANS, Seattle, USA.
ISO 2002. Petroleum and Natural Gas Industries - General Requirements for Offshore Structures. Report 19900.
International Standards Organization.
ISSC 2012a. Report of Committee V.7 Impulsive Pressure Loading and Response Assessment. In Proc. 18th Int.
Ships and Offshore Structures Congress, Rostock, Germany 2: 275 - 330.
ISSC 2012b. Report of Technical Committee I.2 Loads. In Proc. 18th Int. Ships and Offshore Structures Congress,
Rostock, Germany 1, 79150.
Iwanowski, B., Vestbstad, T. & Lefranc, M. 2014. Wave-in-deck load on a jacket platform: CFD calculations
compared with experiments. In Proc. 33rd Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE, San
Francisco, USA.
Jacob, B. P., Bahiense, R. A., Correa, F. N. & Jacovazzo, B. M. 2012a. Parallel implementations of coupled
formulations for the analysis of floating production systems, Part I: Coupling formulations. Ocean Engineering
55.
Jacob, B. P., Bahiense, R. A., Correa, F. N. & Jacovazzo, B. M. 2012b. Parallel implementations of coupled
formulations for the analysis of floating production systems, Part II: Domain decomposition strategies. Ocean
Jacobi, G., Thomas, G., Davis, M. R. & Davidson, G. 2014. An insight into the slamming behaviour of large highspeed catamarans through full-scale measurements. Journal of Marine Science and Technology 19, 1532.
Jain, A. & Modarres-Sadeghi, Y. 2013. Vortex-induced vibrations of a flexibly-mounted inclined cylinder. Journal
of Fluids and Structures 43, 2840.
Ji, Y. M., Shin, Y. S., Park, S. J. & Hyun, M. J. 2012. Experiments on non-resonant sloshing in a rectangular tank
with large amplitude lateral oscillation. Ocean Engineering 50, 1012.
Jin, H., Liu, Y. & Li, H. J. 2014. Experimental study on sloshing in a tank with an inner horizontal perforated plate.
Joga, R., Dhavalikar, J. S. S. & Kar, A. 2014. Numerical simulations to compute rate of water ingress into open
holds due to green waters. In Proc. 24th Int. Offshore and Polar Engineering Conference ISOPE, Busan, Korea.
Johannessen, T. B. & Hagen, . 2012. Estimating design levels for strongly nonlinear response. In Proc. 31st Int.
Conf. on Ocean, Offshore and Arctic Engineering OMAE, Rio de Janeiro, Brazil.
Jonkman, J. 2008. Influence of Control on the Pitch Damping of a Floating Wind Turbine. National Renewable
Energy Laboratory, Golden, CO, USA.
Jonkman, J. & Buhl Jr., M. L. 2005. FAST Users Guide. National Renewable Energy Laboratory, Golden CO,
USA.
Jonkman, J. M. 2013. The new modularization framework for the FAST wind turbine CAE tool. In 51st AIAA
Aerospace Sciences Meeting and 31st ASME Wind Energy Symp., Grapevine, Texas.
Journe, J. M. J. 1992. Experiments and Calculations on 4 Wigley Hull Forms in Head Waves. Technical report
0909, Ship Hydromechanics Lab. DUT, The Netherlands.
Juncher Jensen, J. 2011. Extreme value predictions using Monte Carlo simulations with artificially increased load
spectrum. Journal of Probabilistic Engineering Mechanics 26, 399404.
Jung, J., Yoon, H., Lee, C. & Shin, S. 2012. Effect of the vertical baffle height on the liquid sloshing in a threedimensional rectangular tank. Ocean Engineering 44, 7989.
Kaasen, K. E., Lie, H., Wu, J., Falkenberg, E., Ahjem, V. & Larsen, K. 2014. Development of time domain model
for synthetic rope mooring systems. In Proc. 33rd Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE,
San Francisco, USA.
Kashiwagi, M. & Hara, T. 2012. A method for ship hydroelasticitic analysis by means of Rankine panel method. In
Proc. 6th Int. Conf. on Hydroelasticity in Marine Technology, Tokyo, Japan.
Khabakhpasheva, T. I. & Korobkin, A. A. 2013. Elastic wedge impact onto a liquid surface: Wagners solution.
Kim, B. W., Hong, Y. S., Sung, H. G. & Hong, S. W. 2013a. Dynamic coupled multi-body analysis of FSRU and
mooring system. In Proc. 23rd Int. Offshore and Polar Engineering Conference ISOPE, Alaska, USA.
Kim, B. W., Sung, H. G., Kim, J. H. & Hong, S. Y. 2013b. Comparison of linear spring and nonlinear FEM methods
in dynamic coupled analysis of floating structure and mooring system. Journal of Fluids and Structures 42, 205
227.
Kim, D. H., Engle, A. H. & Troesch, A. W. 2012a. Estimates of long-term combined wave bending and whipping
for two alternative hull forms. Transactions Society of Naval Architects and Marine Engineers SNAME 130.
134
Kim, J. H., Kim, K. H., Jung, B. H., Choi , J. H. & Kim, Y. 2012b. Analysis on ship springing using fully-coupled
FSI models. In Proc. 6th Int. Conf. on Hydroelasticity in Marine Technology, Tokyo, Japan.
Kim, K. H., Bang, J. S., Kim, J. H., Kim, Y., Kim, S. J. & Kim, Y. 2013c. Fully coupled BEM-FEM analysis for
ship hydroelasticity in waves. Marine Structures 33, 7199.
Kim, K. H. & Kim, Y. 2012. Numerical analysis on ship hydroelasticity by using 3D Rankine panel method and 3D
finite element method. In Proc. 6th Int. Conf. on Hydroelasticity in Marine Technology, Tokyo, Japan.
Kim, K. S., Kim, M. H. & Park, J. C. 2014. Simulation of multi-liquid-layer sloshing with vessel motion by using
moving particle simulation. In Proc. 33rd Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE, San
Francisco, USA.
Kim, S., Kim, C. Y. & Cronin, D. 2013d. Green water impact loads on breakwaters of large container carriers. In
Proc. 12th Int. Symp. on Practical Design of Ships and Other Floating Structures PRADS, Changwon, Korea.
Kim, S. J. & Paik, J. K. 2013. Advanced method for ship structural design under slamming impact pressure loads. In
Proc. 12th Int. Symp. on Practical Design of Ships and Other Floating Structures PRADS, Changwon, Korea.
Kim, Y. & Hermansky, G. 2014. Uncertainties in seakeeping analysis and related loads and response procedures.
Kim, Y., Kim, J. H. & Kim, Y. 2013e. Whipping identification of a flexible ship using wavelet cross-correlation.
Kim, Y., Kim, S. Y., Ahn, Y. J. & Kim, K. H. 2013f. Experimental analysis on sloshing loads for LNG cargo
design. In Proc. 12th Int. Symp. on Practical Design of Ships and Other Floating Structures PRADS, Changwon,
Korea.
Kim, Y. & Park, S. G. 2014. Wet damping estimation of the scaled segmented hull model using the random
decrement technique. Ocean Engineering 75, 7180.
Kleefsman, K. M. T. 2005. Water Impact Loading on Offshore Structures: A Numerical Study. PhD Thesis,
University of Groningen.
Kobayakawa, H., Kusumoto, H., Nagashima, T. & Neki, I. 2012. Hydroelastic response analysis using unsteady
time domain analysis of ship motions. In Proc. 6th Int. Conf. on Hydroelasticity in Marine Technology, Tokyo,
Japan.
Koh, C., Luo, M., Gao, M. & Bai, W. 2013. Modelling of liquid sloshing with constrained floating baffle.
Computers & Structures 122, 270279.
Kondo, N. 2011. Three-dimensional computation for flow-induced vibrations in in-line and cross-flow directions of
a circular cylinder. International Journal for Numerical Methods in Fluids 70, 158185.
Koning, J. 2009. Clashing@sea, Executive Summary. SMIG07002 Report No: 1971720TM. Maritime Research
Institute, The Netherlands, Senter Novem.
Koo, J. B., Kim, B. J., Jang, K. B., Suh, Y. S. & Bigot, F. 2013. Fatigue assessment of the 18,000 TEU container
vessel considering the effect of springing. In Proc. 23rd Int. Offshore and Polar Engineering Conference ISOPE,
Alaska, USA.
Korobkin, A. A. 2013. A linearized model of water entry. Journal of Fluid Mechanics 737, 368386.
Korobkin, A. A., Guret, E. & Malenica, S. 2006. Hydroelastic coupling of beam finite element model with Wagner
theory of water impact. Journal of Fluids and Structures 22, 493504.
Kral, R., Kreuzer, E. & Wilmers, C. 1996. Nonlinear oscillations of a crane ship. Journal of Applied Mathematics
and Mechanics (Zeitschrift fr Angewandte Mathematik und Mechanik) 76(4), 58.
Kristiansen, T. & Faltinsen, O. M. 2012. Gap resonance analyzed by a new domain-decomposition method
combining potential and viscous flow draft. Applied Ocean Research 34, 198208.
Kvitrud, A. 2014. Lessons learned from Norwegian anchor line failures. In Proc. 33rd Int. Conf. on Ocean, Offshore
Kyoung, J., Yang, C. K., OSullivan, J. & Miliante, T. 2013. Validation of the HVS semisubmersible global
performance by model tests. In Proc. 32nd Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE, Nantes,
France.
Larsen, T. J. & Hansen, A. M. 2007. How 2 HAWC2, the Users manual. Ris National Laboratory, Roskilde.
Lavroff, J., Davis, M. R., Holloway, D. S. & Thomas, G. 2013. Wave slamming loads on wave-piercer catamarans
operating at high-speed determined by hydroelastic segmented model experiments. Marine Structures 33, 120
142.
Lee, H. H., Lim, H. J. & Rhee, S. H. 2012b. Experimental investigation of green water on the deck for a CFD
validation database. Ocean Engineering 42, 4760.
Lee, T., Chung, H. & Myung, H. 2011. Multi-resolution path planning for marine surface vehicle considering
environmental effects. In Proc. of OCEANS, Santander, Spain.
Lee, Y., Chan, H. S., Pu, Y., Incecik, A. & Dow, R. S. 2012c. Global wave loads on a damaged ship. Ships and
Offshore Structures 7(3), 237268.
Lee, Y., White, N., Wang, Z. & Park, J. B. 2012a. Whipping responses and whipping effects on design bending
moments of a large container ship. In Proc. 6th Int. Conf. on Hydroelasticity in Marine Technology, Tokyo,
Japan.
Li, A. J., Tang, B. Y. & Yeung, C. R. W. 2014a. Effects of second-order difference-frequency wave forces on a new
floating platform for an offshore wind turbine. Journal of Renewable and Sustainable Energy 6(3).
Li, H. T., Li, J., Zong, Z. & Chen, Z. 2014b. Numerical studies on sloshing in rectangular tanks using a tree-based
adaptive solver and experimental validation. Ocean Engineering 82, 2031.
135
Lin, Y. H., Fang, M. C. & Yeung, R. W. 2013. The optimization of ship weather-routing algorithm based on the
composite influence of multi-dynamic elements. Applied Ocean Research, 43, 184194.
Liu, H., Huang, W., Lian, Y. & Li, L. 2014a. An experimental investigation on nonlinear behaviors of synthetic
fiber ropes for deepwater moorings under cyclic loading. Applied Ocean Research 45, 2232.
Liu, S., Papanikolaou, A. & Zaraphonitis, G. 2014b. Time domain simulation of nonlinear ship motions using an
impulse response function method. In Proc. Int. Conf. on Maritime Technology ICMT.
Lloyd's Register 2011. Guidance Notes on the Assessment of Global Design Loads of large Container Ships and
Other Ships Prone to Whipping and Springing. Draft Version 1.1.
Low, Y. M. 2011. Extending a time/frequency domain hybrid method for riser fatigue analysis. Applied Ocean
Research 33, 7987.
Lu, H., Yang, C. & Loehner, R. 2012. Numerical studies of green water impact on fixed and moving bodies.
International Journal of Offshore and Polar Engineering 22(2), 123132.
Lu, X., Kumar, P., Bahuguni, A. & Wu, Y. 2014. A CFD study of focused extreme wave impact on decks of
offshore structures. In Proc. 33rd Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE, San Francisco,
USA.
Lundgren, H. 1969. Wave shock forces: An analysis of deformations and forces in the wave and the foundation. In
Proc. Research on Wave Action. Symp., Delft, The Netherlands.
Lv, J. & Grenestedt, J. L. 2013. Some analytical results for the initial phase of bottom slamming. Marine Structures
34, 88104.
Ma, G., Sun, L., Wang, H. & Ai, S. 2014. Implementation of a visco-elastic model into slender rod theory for deep
water polyester mooring line. In Proc. 33rd Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE, San
Francisco, USA.
Ma, W., Wu, G., Thompson, H., Prislin, I. & Maroju, S. 2013. Vortex induced motions of a column stabilized
floater. In Proc. of Deep Offshore Technology Int. Conf., Texas, USA.
Madsen, H. A., Mikkelsen, R., ye, S., Bak, C. & Johansen, J. 2007. A detailed investigation of the Blade Element
Momentum (BEM) model based on analytical and numerical results and proposal for modifications of the BEM
model. Journal of Physics: Conference Series 75(1).
MAIB 2006. Report on the Investigation of the Loss of Cargo Containers overboard from P&O Nedlloyd Genoa
North Atlantic Ocean on 27 January 2006. Report No 20/2006, Marine Accident Investigation Branch,
Southampton, United Kingdom.
MAIB 2008. Report on the Investigation of the Structural Failure of MSC Napoli, English Channel on 18 January
2007. Marine Accident Investigation Branch, Southampton, United Kingdom.
Maki, A., Akimoto, Y., Nagata, Y., Kobayashi, S., Kobayashi, E., Shiotani, S., Ohsawa, T. & Umeda, N. 2011. A
new weather-routing system that accounts for ship stability based on a real-coded genetic algorithm. Journal of
Marine Science and Technology 16(3), 311322.
Malenica, S. & Derbanne, Q. 2012. Hydroelastic issues in the design of ultra large container shipstulcs project. In
Proc. 6th Int. Conf. on Hydroelasticity in Marine Technology, Tokyo, Japan.
Malenica, S., Eatock Taylor, R. & Huang, J. B. 1999. Second order water wave diffraction by an array of vertical
cylinders. Journal of Fluid Mechanics 390, 349373.
Manwell, J. F., McGowan, J. G. & Rogers, A. L. 2009. Wind Energy Explains. John Wiley & Sons 2nd edition.
Mao, W., Li, Z., Ringsberg, J. W. & Rychlik, I. 2012. Application of a ship-routing fatigue model to case studies of
2800 TEU and 4400 TEU container vessels. Journal of Engineering for the Maritime Environment 226(3), 222
234.
Mao, W., Ringsberg, J. W., Rychlik, I. & Storhaug, G. 2010. Development of a fatigue model useful in ship routing
design. Journal of Ship Research 54(4), 281293.
Mao, W. & Rychlik, I. 2012. Estimation of extreme ship response. Journal of Ship Research 56(1), 2334.
Mariani, R. & Dessi, D. 2012. Analysis bending modes of floating structure by POD. Journal of Fluids and
Masciola, M., Robertson, A., Jonkman, J. & Driscoll, F. 2011. Investigation of a FAST-OrcaFlex Coupling Module
for integrating Turbine and Mooring Dynamics of Offshore Floating Wind Turbines. Preprint, National
Renewable Energy Laboratory.
Masoud, H. 2013. Nonlinear Wave Load on Decks of coastal Structures. PhD dissertation, University of Hawaii,
Manoa, USA.
Matha, D., Schlipf, M., Cordle, A., Pereia, R. & Jonkman, J. 2011. Challenges in simulation of aerodynamics,
hydrodynamics, and mooring-line dynamics of floating offshore wind turbines. In Proc. 21st Int. Offshore and
Polar Engineering Conference ISOPE, Hawaii, USA.
Matha, D., Wendt, F., Werner, M., Cheng, P. W. & Lutz, T. 2012. Aerodynamic inflow conditions on floating
offshore wind turbine blades for airfoil design purposes. In Proc. 22nd Int. Offshore and Polar Engineering
Conference ISOPE, Rhodes, Greece.
Miao, S. H., Temarel, P. & Hirdaris, S. E. 2012. The antisymmetric dynamic behaviour of a modern containership in
regular waves. In Proc. 6th Int. Conf. on Hydroelasticity in Marine Technology, Tokyo, Japan.
Mirciu, I., Rubanenco, I. & Domnisoru, L. 2012. On the numerical non-linear hydroelastic response of a LPG
100,000 cbm carrier in irregular head waves. In Proc. 6th Int. Conf. on Hydroelasticity in Marine Technology,
Tokyo, Japan.
Mitra, S., Wang, C., Reddy, J. & Khoo, B. 2012. A 3d fully coupled analysis of nonlinear sloshing and ship motion.
136
Miyamura, A., Hirabayashi, S. & Suzuki, H. 2014. Numerical simulation of vortex-induced motion with free surface
by lattice Boltzmann method. In Proc. 33rd Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE, San
Francisco, USA.
Moan, T., Shu, Z., Drummen, I. & Almashi, H. 2006. Comparative reliability analysis of ships - considering
different ship types and the effect of ship operations on loads. Transactions Society of Naval Architects and
Marine Engineers SNAME 114, 1654.
Mohammed, E. A., Chan, H. S. & Hirdaris, S. E. 2012. Global wave load combinations by cross-spectral methods.
Marine Structures 29, 131151.
Molin, B. & Remy, F. 2013. Experimental and numerical study of the sloshing motion in a rectangular tank with a
perforated screen. Journal of Fluids and Structures 43, 463480.
Molin, B., Remy, F., Audiffren, C. & Marcer, R. 2012. Experimental and numerical study of liquid sloshing in a
rectangular tank with three fluids. In Proc. 22nd Int. Offshore and Polar Engineering Conference ISOPE,
Rhodes, Greece.
Monroy, C., Giorgiutti , Y. & Chen, X. B. 2012. First and second order wave-current interactions for floating
bodies. In Proc. of 31st Int. Conf. on Ocean, Offshore and Arctic Engineering, Rio de Janeiro, Brazil.
Moore, M. R., Howison, S. D., Ockendon, J. R. & Oliver, J. M. 2012. Three-dimensional oblique water-entry
problems at small deadrise angles. Journal of Fluid Mechanics 711, 259280.
Moore, M. R., Howison, S. D., Ockendon, J. R. & Oliver, J. M. 2013. A note on oblique water entry. Journal of
Engineering Mathematics 81, 6774.
Musial, W. & Ram, B. 2010. Large-Scale Offshore Wind Power in the United States: Assessment of Oppportunities
and Barriers. National Renewable Energy Laboratory.
Nam, B. W., Kim, N. W., Kim, Y., Hong, S. Y. & Sung, H. G. 2013. Computation of nonlinear wave run-up around
a semi-submersible platform. In Proc. 12th Int. Symp. on Practical Design of Ships and Other Floating
Nguyen, L. T. T. & Temarel, P. 2014. Numerical simulation of an oscillating cylinder in cross-flow at a Reynolds
number of 10,000: Forced and free oscillations. In Proc. 33rd Int. Conf. on Ocean, Offshore and Arctic
Engineering OMAE, San Francisco, USA.
Nguyen, V. T., Vu, D. T., Park, W. G. & Jung, Y. R. 2014. Numerical analysis of water impact forces using a dualtime pseudo-compressibility method and volume-of-fluid interface tracking algorithm. Computer & Fluids 10,
1833.
Nichols, J. M., Fackler, P. L., Pacifici, K., Murphy, K. D. & Nichols, J. D. 2014. Reducing fatigue damage for ships
in transit through structured decision making. Marine Structures 38, 1843.
Nielsen, F. G., Hanson, T. D. & Skaare, B. 2006. Integrated dynamic analysis of floating offshore wind turbines. In
Int. Conf. on Offshore Mechanics and Arctic Engineering.
Nielsen, J. K. & Michelsen, J. 2007. Data Models for Ship Varying Data Project: ADOPT. Document ID: ADOPTWP4.2-DEL-20070321-FINAL-Data_Models_Ship_Varying_Data-FORCE Technology.
Nielsen, U. D., Andersen, I. M. V. & Koning, J. 2013. Comparisons of means for estimating sea states from an
advancing large container ship. In Proc. 12th Int. Symp. on Practical Design of Ships and Other Floating
Oberhagemann, J., Shigunov, V. & Moctar, O. E. 2012. Application of CFD in long-term extreme value analyses of
wave loads. Ship Technology Research 59, 422.
Ogawa, Y., Kitamura, O. & Toyoda, M. 2012. A study for the statistical characteristic of slamming induced
vibration of large container ship. In Proc. 6th Int. Conf. on Hydroelasticity in Marine Technology, Tokyo, Japan.
Onorato, M., Proment, D., Clauss, G. & Klein, M. 2013. Rogue waves: From nonlinear Schrdinger breather
solutions to seakeeping test. Public Library of Science PLOS One 8(2), e54629.
Osawa, N., Nakamura, T., Yamamoto, N. & Sawamura, J. 2013. Experimental study on high frequency effect on
fatigue by using the new high speed fatigue testing machine. In Proc. 12th Int. Symp. on Practical Design of
Ships and Other Floating Structures PRADS, Changwon, Korea.
stman, A., Pakozdi, C., Stansberg, C. T. & Carvalho, D. F. 2014. A fully nonlinear RANS-VOF numerical
wavetank applied in the analysis of green water on FPSO in waves. In Proc. 32nd Int. Conf. on Ocean, Offshore
Paik, J. K., Lee, S. E., Kim, B. J., Seo, J. K., Ha, Y. C., Matsumoto, T. & Kim, D. H. 2014. A new method for
determining the design sloshing loads for LNG FPSOs. In Proc. 33rd Int. Conf. on Ocean, Offshore and Arctic
Pakozdi, C., stman, A., Stansberg, C. T. & Carvalho, D. F. 2014. Green water on FPSO analyzed by a coupled
Potential-Flow-NS-VOF method. In Proc. 33rd Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE,
San Francisco, USA.
Panciroli, R., Abrate, S. & Minak, G. 2013. Dynamic response of flexible wedges entering the water. Composite
Panciroli, R., Abrate, S., Minak, G. & Zucchelli, A. 2012. Hydroelasticity in water-entry problems: Comparison
between experimental and SPH results. Composite Structures 94, 532539.
Panigrahi, J. K., Padhy, C. P., Sen, D., Swain, J. & Larsen, O. 2012. Optimal ship tracking on a navigation route
between two ports: A hydrodynamics approach. Journal of Marine Science and Technology 17(1), 5967.
Papaioannou, I., Gao, R. P., Rank, E. & Wang, C. M. 2013. Stochastic hydroelastic analysis of pontoon-type very
large floating structures considering directional wave spectrum. Probabilistic Engineering Mechanics 33, 2637.
137
Papanikolaou, A., Mohammed, E. A. & Hirdaris, S. E. 2014. Stochastic uncertainty modelling for ship design loads
and operational guidance. Ocean Engineering 86, 4757.
Park, K. P., Cha, J. H. & Lee, K. Y. 2011. Dynamic factor analysis considering elastic boom effects in heavy lifting
operations. Ocean Engineering 38, 11001113.
Parunov, J., Andri, J., orak, M. & Kitarovi, S. 2014. Structural reliability assessment of container ship at the
time of accident. Journal of Engineering for the Maritime Environment.
Passano, E., Larsen, C. M. & Lie, H. 2012. Comparison of calculated in-line vortex induced vibrations response to
model tests. In Proc. 31st Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE, Rio de Janeiro, Brazil.
Paulsen, B. T., Bredmose, H. & Bingham, H. B. 2014. An efficient domain decomposition strategy for wave loads
on surface piercing circular cylinders. Coastal Engineering 86, 5776.
Peng, W., Lee, K. H., Shin, S. H. & Mizutani, N. 2013. Numerical simulation of interactions between water waves
and inclined-moored submerged floating breakwaters. Coastal Engineering 82, 7687.
Piro, D. J. & Maki, K. 2011. Hydroelastic wedge entry and exit. In Proc. 11th Int. Conf. On Fast Sea
Transportation, Hawaii, USA.
Piro, D. J. & Maki, K. 2012. Water exit of a wedge-shaped body. In Proc. 27th Int. Wkshp. on Water Waves &
Floating Bodies IWWWFB, Copenhagen, Denmark.
Piro, D. J. & Maki, K. J. 2013. Hydroelastic analysis of bodies that enter and exit water. Journal of Fluids and
Pistani, F. & Thiagarajan, K. 2012. Experiments on non-resonant sloshing in a rectangular tank with large amplitude
lateral oscillation. Ocean Engineering 52, 6074.
Popko, W., Vorpahl, F., Zuga, A., Kohlmeier, M., Jonkman, J., Robertson, A., Larsen, T. J., Yde, A., Stertr, K.,
Okstad, K. M., Nichols, J., Nygaard, T. A., Gao, Z., Manolas, D., Kim, K., Yu, Q., Shi, W., Park, H., VsquezRojas, A., Dubois, J., Kaufer, D., Thomassen, P., de Ruiter, M. J., van der Zee, T., Peeringa, J. M., Zhiwen, H. &
von Waaden, H. 2012. Offshore Code Comparison Collaboration Continuation (OC4), Phase IResults of
coupled simulations of an offshore wind turbine with jacket support structure. In Proc. 22nd Int. Offshore and
Polar Engineering Conference ISOPE, Rhodes, Greece.
Prasetyo, F. A., Osawa, N. & Kobayashi, T. 2012. Study on preciseness of load history generation based on storm
model for fatigue assessment of ship structure members. In Proc. 22nd Int. Offshore and Polar Engineering
Conference ISOPE, Rhodes, Greece.
Prestileo, A., Rizzuto, E., Teixeira, A. P. & Guedes Soares, C. 2013. Bottom damage scenarios for the hull girder
structural assessment. Marine Structures 33, 3355.
Rajendran, S., Fonseca, N. & Guedes Soares, C. 2013. Estimation of short term probability distributions of wave
induced loads acting on a cruise vessel in extreme seas. In Proc. 32nd Int. Conf. on Ocean, Offshore and Arctic
Rajendran, S., Fonseca, N. & Guedes Soares, C. 2014. Prediction of ship responses in large amplitude waves using a
body nonlinear time domain method with 2nd order Froude-Krylov pressure. In Proc. 33rd Int. Conf. on Ocean,
Offshore and Arctic Engineering OMAE, San Francisco, USA.
Rathje, H., Kahl, A. & Schellin, T. E. 2012. Semi-empirical assessment of long-term high-frequency hull girder
response of containerships. In Proc. 22nd Int. Offshore and Polar Engineering Conference ISOPE, Rhodes
Greece.
Reihnard, M., Korobkin, A. A. & Cooker, M. J. 2013. Water-entry of an elastic plate at high horizontal speed.
Journal of Fluid Mechanics 724, 123153.
Reinhard, M., Korobkin, A. A. & Cooker, M. 2012a. Elastic plate impact into water at high horizontal speed with
early water detachment. In Proc. 6th Int. Conf. on Hydroelasticity in Marine Technology, Tokyo, Japan.
Reinhard, M., Korobkin, A. A. & Cooker, M. J. 2012b. The bounce of a blunt body from a water surface at high
horizontal speed. In Proc. 27th Int. Wkshp. on Water Waves & Floating Bodies IWWWFB, Copenhagen,
Denmark.
REPSEA 2013. Vortex induced Motion Study for Deep Draft Column Stabilized Floaters (S&ES). Project No.
11121540403, Research Partnership to Secure Energy for America.
Resvanis, T. L. & Vandiver, J. K. 2011. Modelling risers with partial strake coverage. In Proc. 30th Int. Conf. on
Ocean, Offshore and Arctic Engineering OMAE, Rotterdam, The Netherlands.
Rijken, O. & Leverette, S. 2009. Field measurements of vortex induced motions of a deep draft semisubmersible. In
Proc. 28th Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE, Hawaii, USA.
Roald, L., Jonkman, J., Robertson, A. & Chokani, N. 2013. The effect of second-order hydrodynamics on floating
offshore wind turbines. Energy Procedia 35, 253264.
Robertson, A. & Jonkman, J. 2011. Loads analysis of several offshore floating wind turbine concepts. In Proc. 30th
Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE, Rotterdam, The Netherlands.
Robertson, A., Jonkman, J., Musial, W., Vorpahl, F. & Popko, W. 2013. Offshore Code Comparison Collaboration,
Continuation: Phase II Results of a Floating Semi-submersible Wind System. National Renewable Energy
Laboratory, Golden, CO, USA.
Roddier, D., Cermelli, C. & Weinstein, A. 2009. WindFloat: A floating foundation for offshore wind turbinesPart
I: Design basis and qualification process. In Proc. 28th Int. Conf. on Ocean, Offshore and Arctic Engineering
OMAE, Hawaii, USA.
Ruggeri, F., Wata, R. A., Brisson, H., Mello, P. A., Sampaio Carvalho e Silva, C. M. P. & Vieira, D. P. 2013.
Numerical prediction of green water events in beam seas. In Proc. 12th Int. Symp. on Practical Design of Ships
and Other Floating Structures PRADS, Changwon, Korea.
138
Sanaati, B. & Kato, N. 2012. Vortex-induced vibration (VIV) dynamics of a tensioned flexible cylinder subjected to
uniform cross-flow. Journal of Marine Science Technology 18, 247261.
Santos, F. M., Casetta, L. & Pesce, C. P. 2012. Application of a variational method to the vertical hydrodynamic
impact of axisymmetric bodies. Applied Ocean Research 39, 7582.
Sarala, R., Hajiarab, M. & Bamford, R. 2011. Equivalent Design Wave approach for calculating site-specific
environmental loads on an FPSO. In Proc. 30th Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE,
Rotterdam, The Netherlands.
Sarpkaya, T. 2004. A critical review of the intrinsic nature of vortex-induced vibrations. Journal of Fluids and
Scharnke, J., Vestbstad, T., de Wilde, J. & S. Haver, S. 2014. Wave-in-deck impact load measurements on a fixed
platform deck. In Proc. 33rd Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE, San Francisco, USA.
Schiller, R. V., Caire, M., Affonso, P. H., Passnao, E. & Lie, H. 2014a. Vortex induced vibrations of deep water
risers: Sensitivity to current profile, shear and directionality. In Proc. 33rd Int. Conf. on Ocean, Offshore and
Arctic Engineering OMAE, San Francisco, USA.
Schiller, R. V., Pkozdi, C., Stansberg, C. T. & Carvalho, D. F. 2014b. Green water and wave impact on FPSOs in
Santos Basin: Challenges and prediction tools. In Proc. 33rd Int. Conf. on Ocean, Offshore and Arctic
Scolan, Y. M. & Korobkin, A. A. 2012. Hydrodynamic impact (Wagner) problem and Galins theorem. In Proc.
27th Int. Wkshp. on Water Waves & Floating Bodies IWWWFB, Copenhagen, Denmark.
Scolan, Y. M. & Korobkin, A. A. 2013. Hydrodynamic impact of three-dimensional bodies on waves. In Proc. 27th
Int. Wkshp. on Water Waves & Floating Bodies IWWWFB, Marseille, France.
Seng, S. & Juncher Jensen, J. 2012. Slamming simulations in a conditional wave. In Proc. 31st Int. Conf. on Ocean,
Offshore and Arctic Engineering OMAE, Rio de Janeiro, Brazil.
Seng, S. & Juncher Jensen, J. 2013. An application of a free surface CFD method in the short-term extreme response
analysis of ships. In Proc. 12th Int. Symp. on Practical Design of Ships and Other Floating Structures PRADS,
Changwon, Korea.
Senjanovi, I., Hadi, N. & Vladimir, N. 2012. Improved methodology of ship hydroelastic analysis. In Proc. 6th
Int. Conf. on Hydroelasticity in Marine Technology, Tokyo, Japan.
Shao, J., Li, H., Liu, G. & Liu, M. 2012. An improved SPH method for modeling liquid sloshing dynamics.
Computers & Structures 100, 1826.
Shao, Y. & Faltinsen, O. M. 2012. Linear seakeeping and added resistance analysis by means of body-fixed
coordinate system. Journal of Marine Science and Technology 17, 493510.
Shao, Y. & Helmers, J. 2014. Numerical analysis of second-order wave loads on large volume marine structures in a
current. In Proc. 33rd Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE, San Francisco, USA.
Shao, Y. L. & Faltinsen, O. M. 2013. Second-order diffraction and radiation of a floating body with small forward
speed. Journal of Offshore Mechanics and Arctic Engineering 135.
Shao, Y. L. & Faltinsen, O. M. 2014. A harmonic polynomial cell (HPCF) method for 3D laplace equation with
application in marine hydrodynamics. Journal of Computational Physics 274, 312332.
Simonsen, C. D., Otzen, J. F., Joncquez, S. & Stern, F. 2013. EFD and CFD for KCS heaving and pitching in regular
head waves. Journal of Marine Science and Technology 18, 435459.
Skaare, B., Hanson, T. D., Nielsen, F. G., Yttervik., R., Hansen, A. M., Thomsen, K. & Larsen, T. J. 2007. Integrated
dynamic analysis of floating offshore wind turbine. In Proc. European Wind Energy Conf., Milan, Italy.
Skaare, B., Nielsen, F. G., Hanson, T. D., Yttervik, R., Havmller, O. & Rekdal, A. 2014. Analysis of measurements
and simulations from the Hywind Demo floating wind turbine. Wind Energy.
Song, K. H., Kim, Y. & Park, D. M. 2013. Quantitative and qualitative analyses of parametric roll for ship design
and operational guidance. Journal of Engineering for the Maritime Environment 227(2), 117189.
Spanos, D. A. & Papanikolaou, A. D. 2012. On the time dependence of survivability of ROPAX ships. Journal of
Marine Science and Technology 17, 4046.
Sriram, V. & Ma, Q. W. 2012. Improved MLPG_R method for simulating 2D interaction between violent waves and
elastic structures. Journal of Computational Physics 231, 76507670.
Sriram, V., Ma, Q. W. & Schlurmann, T. 2014. A hybrid method for modelling two dimensional non-breaking and
breaking waves. Journal of Computational Physics 272, 429454.
Stansberg, C. T., Hermundstad, E. M., Hoff, J. R. & Baarholm, R. 2013. Wave drift forces and responses in current.
In 32nd International Conference on Ocean, Offshore and Arctic Engineering, Nantes, France.
Stettler, J. W. & Thomas, B. S. 2012. Flooding and structural forensic analysis of the sinking of the RMS Titanic.
Transactions Society of Naval Architects and Marine Engineers SNAME 120.
Storhaug, G. 2009. The 4400 TEU container vessel MSC Napoli broke its back, but did whipping contribute? In
Proc. 5th Int. Conf. on Hydroelasticity in Marine Technology, Southampton, UK.
Storhaug, G. 2012. The effect of heading on springing and whipping induced fatigue damage measured on container
vessels. In Proc. 6th Int. Conf. on Hydroelasticity in Marine Technology, Tokyo, Japan.
Storhaug, G., Pettersen, T. A., Oma, N. & Blomberg, B. 2012. The effect of wave induced vibrations on fatigue
loading and the safety margin against collapse on two LNG vessels. In Proc. 6th Int. Conf. on Hydroelasticity in
Marine Technology, Tokyo, Japan.
Sumi, Y., Yajima, H., Toyosada, M., Yoshikawa, T., Aihara, S., Gotoh, K., Ogawa, Y., Matsumoto, M., Hirota, K.,
Hirasawa, H., Toyoda, M. & Morikage, Y. 2013. Fracture control of extremely thick welded steel plates applied
to the deck structure of large container ships. Journal of Marine Science and Technology 18, 497514.
139
Sun, H. & Faltinsen, O. M. 2012. Hydrodynamic forces on a semi-displacement ship at high speed. Applied Ocean
Research 34, 6877.
Sun, S. Y., Sun, S. L. & Wu, G. X. 2015. Oblique water entry of a wedge into waves with gravity effect. Journal of
Fluids and Structures 52, 4964.
Suzuki, H. & Sato, A. 2007. Load on turbine blades induced by motion of floating platform and design requirement
for the platform. In Proc. 26th Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE, San Diego, USA.
Taggart, S. & Tognarelli, M. A. 2008. Offshore drilling riser VIV suppression devices Whats available to
operators? In Proc. 27th Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE, Estoril, Portugal.
Tahar, A. & Sidarta, D. 2014. Dual stiffness approach for polyester mooring line analysis in time domain
semisubmersible case. In Proc. 33rd Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE, San
Francisco, USA.
Takahara, H., Kensuke, H. & Ishida, T. 2012. Nonlinear liquid oscillation in a cylindrical tank with an eccentric core
barrel. Journal of Fluids and Structures 25, 120132.
Takahara, H. & Kimura, K. 2012. Frequency response of sloshing in an annular cylindrical tank subjected to
pitching excitation. Journal of Sound & Vibration 331(13), 31993212.
Tassin, A., Korobkin, A. A. & Cooker, M. J. 2012a. Modelling of the oblique impact of an elongated body by 2D + t
approach. In Proc. 27th Int. Wkshp. on Water Waves & Floating Bodies IWWWFB, Copenhagen, Denmark.
Tassin, A., N. Jacques, Alaoui, A. E. M., Neme, A. & Leble, B. 2012b. Hydrodynamic loads during water impact of
three-dimensional solids: Modelling and experiments. Journal of Fluids and Structures 40, 317336.
Tassin, A., Piro, D. J., Korobkin, A. A., Maki, K. & Cooker, M. J. 2013. Two-dimensional water entry and exit of a
body whose shape varies in time. Journal of Fluids and Structures 28, 211231.
Teixeira, A. P., Guedes Soares, C., Chen, N. Z. & Wang, G. 2013. Uncertainty analysis of load combination factors
for global longitudinal bending moments of double-hull tankers. Journal of Ship Research 57(1), 4258.
Thomas, G., Davis, M. R., Holloway, D. S. & Roberts, T. 2003. Transient slam response of large high speed
catamarans. In Proc. 7th Int. Conf. on Fast Sea Transportation, Ischia, Italy.
Thomassen, P. E., Bruheim, P. I., Suja, L. & Fryd, L. 2012. A novel tool for FEM analysis of offshore wind
turbines with innovative visualization techniques. In Proc. 22nd Int. Offshore and Polar Engineering Conference
ISOPE, Rhodes, Greece.
Tofa, M. M., Maimun, A., Ahmed, Y. M. & Jamie, S. 2014. Numerical study of the flow-induced vibration of two
equal-diameter cylinders in tandem with varying the mass ratio. In Proc. 33rd Int. Conf. on Ocean, Offshore and
Arctic Engineering OMAE, San Francisco, USA.
Tognarelli, M. A. & Winterstein, S. R. 2014. Fatigue of risers: calibrating reliability estimates from full-scale data.
Topliss, M. E., Cooker, M. J. & Peregrine, D. H. 1992. Pressure oscillations during wave impact on vertical walls.
In Proc. 23rd Int. Conf. on Coastal Engineering, Venice, Italy.
Toyoda, K., Matsumoto, T., Yamamoto, N. & Terai, K. 2012. Simplified fatigue assessment considering the
occurrence probability of hydroelastic response in actual sea state conditions. In Proc. 6th Int. Conf. on
Hydroelasticity in Marine Technology, Tokyo, Japan.
Tsai, S. C., Hsu, S. H., Chien, H. L., Chou, C. M., Malenica, M. & Chen, X. B. 2013. Numerical study on
seakeeping-sloshing coupling effect of container ship for sea trial purpose. In Proc. 12th Int. Symp. on Practical
Design of Ships and Other Floating Structures PRADS, Changwon, Korea.
Tuitman, J. T., Bosman, T. N. & Harmsen, E. 2013. Local structural response to seakeeping slamming loads. Marine
Tveitnes, T., Fairlie-Clark, A. C. & Varyani, K. 2008. An experimental investigation into the constant velocity water
entry of wedge-shaped sections. Ocean Engineering 35, 14631478.
Vandiver, J. & Chung, T. 1989. Hydrodynamic damping on flexible cylinders in sheared flow. Journal of Waterway,
Port, Coastal & Ocean Engineering 115(2), 154171.
Vanem, E., Bitner-Gregersen, E., Eide, L. I., Garre, L. & Friis Hansen, P. 2014. Uncertainties of climate modeling
and effects on wave induced bending moment. SNAME Annual Meeting, Houston.
Vasquez, G., Fonseca, N. & Guedes Soares, C. 2013. Experimental and numerical extreme motions and vertical
bending moments induced by abnormal waves on a bulk carrier. In Proc. 32nd Int. Conf. on Ocean, Offshore and
Arctic Engineering OMAE, Nantes, France.
Vidic-Perunovic, J. 2012. Ship springing response in finite water depth. Journal of Ship Research 56, 8090.
Wada, R., Iijima, K., Kimura, K., W, X. & Fujikubo, M. 2010. Development of a design methodology for a scaled
model for hydro-elastoplasticity of a hull girder in waves. In 4th PAAMES Meeting, Singapore.
Wang, S. & Guedes Soares, C. 2013. Slam induced loads on bow flared sections with various roll angles. Ocean
Wang, S. & Guedes Soares, C. 2014. Numerical study on the water impact of 3D bodies by an explicit finite element
method. Ocean Engineering 78, 7388.
Wang, X. L., Gu, X. K. & Hu, J. J. 2013. Investigation of springing influence on fatigue damage of ship structures.
In Proc. 12th Int. Symp. on Practical Design of Ships and Other Floating Structures PRADS, Changwon, Korea.
Wang, X. L., Gu, X. K., Hu, J. J. & Xu, C. 2012. A study of sloshing influence on wave induced responses of a LNG
ship by experimental method. In Proc. 6th Int. Conf. on Hydroelasticity in Marine Technology, Tokyo, Japan.
Wayman, E. N., Sclavounos, P. D., Butterfield, S., Jonkman, J. & Musial, W. 2006. Coupled dynamic modeling of
floating wind turbine systems. In Proc. Offshore Technology Conference OTC, Houston, USA.
140
Webster, W. C., Lambrakos, K., Kim, J. & Jing, X. 2012. Rod dynamics with large stretch. In Proc. 31st Int. Conf.
on Ocean, Offshore and Arctic Engineering OMAE, Rio de Janeiro, Brazil.
Wellens, P. 2012. Wave Simulation in truncated Domains for offshore Applications. Ph.D. thesis, Delft University of
Technology, The Netherlands.
Witz, J. A. 1995. Parametric excitation of crane loads in moderate sea states. Ocean Engineering 22(4), 411420.
Wouts, R., Coppens HeereMac, T. & van den Boom, H. J. J. 1992. Monitoring offshore lift dynamics. In Proc. 24th
Offshore Technology Conference OTC, Texas, USA.
Wu, C. H., Faltinsen, O. M. & Chen, B. F. 2012a. Numerical study of sloshing liquid in tanks with baffles by timeindependent finite difference and fictitious cell method. Computers & Fluids 63, 926.
Wu, G., Kim, J. W., Ma, W., Jang, H., Kramer, M. & OSullivan, J. 2014. Vortex-induced motions of a columnstabilized floater part II: CFD benchmark and prediction. In Proc. Deep Offshore Technology Conf., Aberdeen, UK.
Wu, M. K., Lehn, E. & Moan, T. 2012b. Design of segmented model for ship seakeeping tests with hydroelastic
effects. In Proc. 6th Int. Conf. on Hydroelasticity in Marine Technology, Tokyo, Japan.
Xiao, J. & Batra, R. C. 2012. Local slamming of curved rigid hulls. International Journal of Multiphysics 6, 305339.
Xiao, J. & Batra, R. C. 2014. Delamination in sandwich panels due to local water slamming loads. Journal of Fluids
and Structures 48, 122155.
Xiao, L. F., Tao, L. B., Yang, J. M. & Li, X. 2014. An experimental investigation on wave run-up along the
broadside of a single point moored FPSO exposed to oblique waves. Ocean Engineering 88, 8190.
Xing, Y. H., Karimirad, M. & Moan, T. 2012. Effect of spar-type floating wind turbine nacelle motion on drivetrain
dynamics. European Wind Energy Association annual event, Copenhagen, Denmark.
Xu, H. H. 2013. Numerical Simulation of breaking Wave Impact on the Structure. PhD dissertation, National
University of Singapore, Singapore.
Xu, Q., J. Kim, Bhaumik, T., OSullivan, J. & Ermon, J. 2012. Validation of HVS semisubmersible VIM
performance by model test and CFD. In Proc. 31st Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE,
Rio de Janeiro, Brazil.
Yang, J., Kim, S., Park, J. S., Jung, B. H. & Lee, T. 2013. Numerical analysis for slamming impact loads and
dynamic structural responses of a container ship. In Proc. 12th Int. Symp. on Practical Design of Ships and Other
Floating Structures PRADS, Changwon, Korea.
Yang, M., Teng, B., Ning, D. & Shi, Z. 2012. Coupled dynamic analysis for wave interaction with a truss spar and
its mooring line/riser system in time domain. Ocean Engineering 39, 7287.
Yang, Q. & Qiu, W. 2012. Numerical simulation of water impact for 2D and 3D bodies. Ocean Engineering 43, 8289.
You, J. & Faltinsen, O. M. 2012. A 3D fully nonlinear numerical wave tank with a moored floating body in shallow
water. In Proc. 22nd Int. Offshore and Polar Engineering Conference ISOPE, Rhodes, Greece.
Yu, H., Chen, Y. & Cui, Y. 2013. State of the art for dry tree semi technologies. Engineering Science 11(4), 9296.
Zhang, C., Lin, N. S., Tang, Y. H. & Zhao, C. B. 2014. A sharp interface immersed boundary/VOF model coupled
with wave generating and absorbing options for wave-structure interaction. Computers & Fluids 89, 214231.
Zhao, M., Kaja, K., Xiang, Y. & Yan, G. 2013. Vortex-induced vibration (VIV) of a circular cylinder in combined
steady and oscillatory flow. Ocean Engineering 73, 8395.
Zhao, W., Key, S., Yang, J., Tao, L. & White, D. 2014a. Research on the coupling effects between ship motions and
sloshing. In Proc. 33rd Int. Conf. on Ocean, Offshore and Arctic Engineering OMAE, San Francisco, USA.
Zhao, W., Yang, J., Hu, Z. & Tao, L. 2014b. Coupled analysis of nonlinear sloshing and ship motions. Applied
Ocean Research 47, 8597.
Zhao, W., Yang, J., Hu, Z. & Tao, L. 2014c. Prediction of hydrodynamic performance of an FLNG system in sideby-side offloading operation. Journal of Fluids and Structures 46, 89110.
Zhao, X. & Hu, C. 2012. Numerical and experimental study on a 2D floating body under extreme wave conditions.
Applied Ocean Research 35, 113.
Zhao, X., Ye, Z., Fu, Y. & Cao, F. 2014d. A CIP-based numerical simulation of freak wave impact on a floating
body. Ocean Engineering 87, 5063.
Zhao, X. Z., Ye, Z. T. & Fu, Y. N. 2014e. Green water loading on a floating structure with degree of freedom
effects. Journal of Marine Science and Technology 19(3), 320313.
Zhou, B. Z., Ning, D. Z., Teng, B. & Bai, W. 2013. New insight into the wave-induced nonlinear vertical load effects
of ultra-large container ships based on experiments. Journal of Marine Science and Technology 18, 87114.
Zhu, S. & Moan, T. 2013. New insight into the wave-induced nonlinear vertical load effects of ultra-large container
ships based on experiments. Journal of Marine Science and Technology 18, 87114.
Zhu, S. & Moan, T. 2014. Nonlinear effects from wave-induced maximum vertical bending moment on a flexible
ultra-large containership model in severe head and oblique seas. Marine Structures 35, 125.
Zhu, T. & Shigemi, T. 2003. Practical estimation methods of the design loads for primary structural members of
bulk carriers. Marine Structures 16, 489511.
Zhu, T. & Shigemi, T. 2007. Design loads used for direct strength assessment of merchant ship structures.
Transactions of the ASME 129, 120130.
Zimmerman, E. H., Garrity, R. & Nie, C. 2013. Mooring analysis utilizing a coupled mooring and anchor analysis
approach. In Proc. 23rd Int. Offshore and Polar Engineering Conference ISOPE, Alaska, USA.
Zou, M. S., Wu, Y. S., Wu, W. W., Ye, Y. L. & Tian, C. 2012. The three-dimensional hydroelasticity theory of ship
structures in acoustic fluid of shallow sea. In Proc. 6th Int. Conf. on Hydroelasticity in Marine Technology,
Tokyo, Japan.

7 10 SEPTEMBER 2015
CASCAIS, PORTUGAL
VOLUME 1
COMMITTEE II.1
QUASI-STATIC RESPONSE
COMMITTEE MANDATE
Concern for the quasi-static response of ships and offshore structures, as required for safety and
serviceability assessments. Attention shall be given to uncertainty of calculation models for use in reliability
methods, and to consider both exact and approximate methods for the determination of stresses appropriate
for different acceptance criteria.
COMMITTEE MEMBERS
Chairman:
J. W. Ringsberg, Sweden
B. Bohlmann, Germany
H-L. Chien, Taiwan
A. Constantinescu, Belgium
S. E. Heggelund, Norway
S. E. Hirdaris, UK
B. S. Jang, Korea
T. S. Koko, Canada
P. Lara, USA
S. Miyazaki, Japan
M. Sidari, Italy
B.R. van der Sluijs, Netherlands
M. Taczala, Poland
Z. Wan, China
A. Zamarin, Croatia
O. kland, Norway
KEYWORDS
Container ships, corrosion, direct calculations, experiments and testing, extreme load, fatigue assessment,
finite element analysis, IACS Common Structural Rules, IMO Goal-Based Standards, impact loads, loads
modelling, offshore structures, optimisation, probabilistic approach, quasi-static response, reliability
analysis, ship structures, strength assessment, stress response calculation, structural integrity, uncertainty
analysis.
142
ISSC committee II.1: QUASI-STATIC RESPONSE
CONTENTS
1.
INTRODUCTION .......................................................................................................................................... 144
2.
STRENGTH ASSESSMENT APPROACHES ............................................................................................ 144

2.1 Modelling of loads by quasi-static analysis ....................................................................................... 144
2.2 Response calculation .......................................................................................................................... 146
2.3 Reliability ........................................................................................................................................... 147
3.
CALCULATION PROCEDURES ................................................................................................................ 148

3.1 Taxonomy of engineering assessment methods................................................................................. 148
3.1.1 Simplified analysis (rule-based design) / first principles .................................................... 148
3.1.2 Direct calculations................................................................................................................ 148
3.1.3 Reliability analyses .............................................................................................................. 148
3.1.4 Optimisation-based analyses................................................................................................ 149
3.2 Design for production loads modelling.............................................................................................. 149
3.2.1 Rules versus rational based ship design............................................................................... 149
3.2.2 Direct simulations for global quasi-strength assessment .................................................... 149
3.2.3 Loads extracted from experiments and testing .................................................................... 151
3.2.4 Loads from seakeeping codes .............................................................................................. 152
3.3 Structural modelling ........................................................................................................................... 152
3.3.1 Finite element modelling ..................................................................................................... 152
3.3.2 Models for global and detailed analyses.............................................................................. 152
3.3.3 Composite structures ............................................................................................................ 153
3.4 Structural response assessment .......................................................................................................... 153
3.4.1 Buckling and ultimate strength ............................................................................................ 153
3.4.2 Fatigue strength .................................................................................................................... 154
3.4.3 Ship dynamics - vibrations .................................................................................................. 155
3.5 Validation of calculation results ......................................................................................................... 155
3.5.1 Model scale experiments and testing ................................................................................... 156
3.5.2 Full scale hull stress monitoring .......................................................................................... 160
4.
UNCERTAINTIES ASSOCIATED WITH RELIABILITY-BASED QUASI-STATIC RESPONSE

ASSESSMENT .............................................................................................................................................. 161
4.1 Uncertainties associated with loads.................................................................................................... 161
4.1.1 Still water and wave loads ................................................................................................... 161
4.1.2 Ice loads................................................................................................................................ 162
4.1.3 Combination factors ............................................................................................................. 162
4.2 Uncertainties in structural modelling ................................................................................................. 163
4.2.1 Corrosion .............................................................................................................................. 163
4.2.2 Structural characteristics ...................................................................................................... 164
4.2.3 Reliability and risk-based structural assessment ................................................................. 165
4.2.4 Methods and criteria............................................................................................................. 165
4.2.5 Structural capacity................................................................................................................ 166
4.3 Risk-based inspection, maintenance and repair ................................................................................. 167
4.3.1 Inspection ............................................................................................................................. 167
4.3.2 Maintenance and repair ........................................................................................................ 168
5.
SHIP STRUCTURES..................................................................................................................................... 169

5.1 Developments in international rules and regulations ......................................................................... 169
5.1.1 IMO Goal-Based Standards ................................................................................................. 169
5.1.2 IACS Common Structural Rules for Bulk Carriers and Oil Tankers.................................. 170
5.1.3 Development of structural design software systems ........................................................... 172
5.2 Special ship concepts.......................................................................................................................... 173
5.2.1 Service vessels for wind mills and offshore platforms........................................................ 173
5.2.2 Container ships ..................................................................................................................... 173
5.2.3 LNG / LPG tankers .............................................................................................................. 174

5.2.4
143
Other ship types.................................................................................................................... 175
6.
OFFSHORE STRUCTURES......................................................................................................................... 176

6.1 Types of analysis for various floating offshore structures................................................................. 176
6.2 Types of analysis for various fixed offshore structures ..................................................................... 179
6.3 Uncertainty, risk and reliability in offshore structural analysis ......................................................... 182
7.
BENCHMARK STUDY ................................................................................................................................ 184

7.1 Methodology....................................................................................................................................... 184
7.2 Simplified methods ............................................................................................................................. 186
7.3 Quasi-static linear FE analysis ........................................................................................................... 188
7.4 Nonlinear, transient dynamic FE analysis.......................................................................................... 188
7.5 Concluding remarks............................................................................................................................ 190
8.
CONCLUSIONS AND RECOMMENDATIONS ....................................................................................... 191
REFERENCES .................................................................................................................... ...... ........................... 192
144
1.
INTRODUCTION
Ships and floating offshore installations are large-scale, complex structures that are designed and built to
operate for long periods in an ever changing environment. For safe and sustainable structural design, the
design process should follow the limit-state-based design philosophy which encompasses serviceability,
ultimate strength, fatigue and accidental limit states. For engineering economy in the early design stage it
is common to employ quasi-static approaches for the evaluation of loads. It is then important to have
good understanding of the difference between quasi-static and dynamic response analyses as well as the
available engineering techniques (e.g. empirical approaches, direct analysis methods, reliability analysis,
etc.) and associated modelling procedures that can be applied for design assessment.
A convenient and useful computational tool for structural response analysis is the finite element (FE)
method which, if used properly, can model efficiently the complexity and interaction between components and parts of large structures and the maritime environment. Knowledge and treatment of uncertainties that may relate with the modelling of loads and the constitutive materials modelling, may have great
impact on the accuracy and reliability of results. Past reports by the ISSC Technical Committee II.1 have
presented comprehensive reviews of various strength assessment approaches (for example, see ISSC
2012). The present committee report is organised as follows.
Chapter 2 is a general introduction to strength assessment approaches for quasi-static response of ships
and offshore structures. Its objective is to give a brief overview on modelling of quasi-static loads and
subsequently elaborate the procedures and available technologies for the evaluation of associated responses within the context of reliability assessment.
Chapter 3 is the major part of the report and presents a comprehensive review of existing calculation
procedures. At first instance, different levels of analysis that relate to different design stages are discussed. Then, a review of direct assessment procedures for ships and offshore structures also including
composite applications is presented. Finally, recent studies and recommendations emerging from the application of Fluid Structure Interaction (FSI) methods for buckling, ultimate and fatigue strength assessment are discussed. Special emphasis is attributed to the importance of validation of numerical prediction
methods against model-scale experiments and full-scale monitoring.
Awareness of the risks that relate to uncertainties in modelling, analysis and assessment is crucial for
safe design. For this reason, Chapter 4 discusses the importance of uncertainties associated with reliability
based quasi-static response assessment. Methods and criteria for reliability and risk-based structural assessment are presented in the context of structural capacity methods. Finally, a review of recent work and
methods that consider existing and aged vessels with regard to risk-based inspection, maintenance and
repair is presented.
Chapter 5 focuses on ship structures and discusses developments in international rules and regulations,
followed by some specific ship types such as service vessels for wind mills and offshore platforms, container ships, and LNG/LPG tankers.
In Chapter 6, various types of floating and fixed offshore structures are reviewed. The chapter ends
with a more specific discussion and review of methods for uncertainty, risk and reliability analysis of
offshore structural analyses which are not covered by Chapter 4.
The committee has carried out a benchmark study where design against impact loads (slamming) was
studied for a free fall lifeboat case. The results from the study are presented in Chapter 7, followed by a
presentation of the conclusions and recommendations for future work and progress of the committees
work in Chapter 8.
2.
STRENGTH ASSESSMENT APPROACHES
A wide variety of strength assessment approaches that may be used for the design assessment and structural optimisation of ships and offshore structures are available. Simplified solutions although conservative are usually applied at preliminary stage. More detailed, yet time-consuming and precise methods are
usually applied for detailed design. Irrespective to their degree of fidelity available methods reflect three
main aspects namely: (1) modelling idealisations/assumptions, (2) the process of load derivation and application onto the model, and (3) uncertainty modelling and quantification. Each of these aspects is introduced in separate sections of this report.
2.1
Modelling of loads by quasi-static analysis
Within the reporting period the development of suitable methods for the simulation and evaluation of
quasi-static responses incorporating the influence of nonlinearities by multi-physics methods has proved
challenging, particularly within the context of industrial applications. This is primarily due to the lack of
unified validation studies or verification schemes that can limit the number of uncertainties related with
the computation of wave-induced dynamic loads (Hirdaris et al. 2014a and Hirdaris 2014b). From a mari-
145
time classification perspective, the main advances within the committees reporting relate to the implementation of mature (e.g. 3D linear seakeeping methods) or semi-mature (e.g. weakly nonlinear hydrodynamic) computational methods for the generation of the quasi-hydrodynamic pressures and their influence
on dynamic response. In the short-term understanding the modelling assumptions and the influence of
weak nonlinearities on quasi-dynamic response is considered as the obvious next step for design development. Medium to long-term use of multi-physics methods and their implementation on computational
robust solvers may assist with the following:
The idealisation of wave free-surface effects because of the nature of the free-surface boundary conditions and the nonlinear characteristics of the incident waves. Within this context it is now understood
that the time-dependent change of position and wetted surface of the ship in waves often cause nonlinear hydrostatic restoring forces and nonlinear force contributions in way of the free-surface intersection.
The idealisation of viscous force contributions that depends on the water velocity and introduces velocity-squared terms in the pressure equation.
The need to design for optimum sagging/hogging ratios of loads especially for the case of the latest
generation of ships that are long and slender (e.g. container ships).
Emerging multi-physics methods can be classified into three categories: (1) Rankine source/panel
method, (2) frequency domain Green function method, and (3) time domain transient Green function
method. In view of the importance of the subject and its impact on ship design, members of IACS Common Structural Rules for Bulk Carriers and Oil Tankers (CSR- BC&OT) recently compared 3D linear
seakeeping method programs (IACS 2013c). In this study it was shown that wave-induced ship motions
and loads obtained in way of different cross-sections and in different wave conditions are effective for
conventional ships. In a more recent paper, Hirdaris et al. (2014a) reviewed a large number of different
nonlinear methods for the forward speed problem. The methods reviewed are classified using six different
levels (see Figure 1):
level 1: linear,
level 2: Froude-Krylov nonlinear,
level 3: body nonlinear,
level 4: body exact - weak scattered,
level 5: fully nonlinear - smooth waves, and
level 6: fully nonlinear.
Figure 1. Level of idealisation for forward speed hydrodynamic solutions. Numbers 1 to 6 refer to levels 1 to 6 of
idealisation according to Hirdaris et al. (2014a).
In this range of methods one may distinguish between methods based on potential theory and those solving the Reynolds-Averaged-Navier-Stokes (RANS) equations (e.g. Stern et al. 2013). Within this group
there is also a large variety of methods ranging from linear theories to fully nonlinear methods. Between
these two extremes there are many partially nonlinear, or blended, methods, in which one aims at including the most important nonlinear effects. As techniques become more sophisticated assumptions become
more complex and uncertainties may vary. Equally, validation, computation time and complexity may be
an issue when we try to understand, simplify or validate the modelling assumptions (for example, see Ea
& Hoekstra 2013). It appears that within the context of quasi-static/dynamic response the use of weakly
nonlinear or fully nonlinear methods is feasible over the medium to long-term provided that validation
efforts are extended and modelling assumptions are well understood. Another example of this is illus-
146
trated in the work presented by Southall et al. (2014), where a prediction of impact pressures on a dropping wedge using Computational Fluid Dynamics (CFD) has been compared against experimental results.
2.2
Response calculation
The approaches where the uncertainty in loads and structural resistance are covered by one or more safety
factors are often referred to as the deterministic (or working stress design) approach, and as load and resistance factor design (or partial safety factor) approach. These methods are nowadays widely used for the
design of ships and offshore structures. Within these classes of calculations one could further distinguish
between more simplified, analytical or semi-analytical analysis and direct calculation methods, referring
either to the way the quasi-static loads are derived or to the complexity of the structural model used. More
and more, classification rules provide guidance on modelling and acceptance criteria based on either
beam theory or direct calculations using the FE method.
For example, traditionally the design of Ultra Large Container Ships includes mandatory items such as
quasi-static analyses of mid-hold, full-ship, and detailed fine mesh of local structures (LR 2013). In recent
years with the advances in technology and design innovation it is required to also perform additional
analyses of the influence of hull flexibility on dynamic response (LR 2014). Such methods specifically
address the impact of springing and whipping for the evaluation of the ultimate strength and fatigue life
of the hull structure. Plunging, bow and stern slamming analysis methods also emerge as the obvious
mean of confirming hull scantlings (Hirdaris et al. 2014a).
Within the reporting period literature review presents a trend in the development of procedures for the
idealisation and analysis of accidental limit states. For example, Nguyen et al. (2011) presented some of
the key idealisation parameters for dynamic grounding scenarios (e.g. seabed shape, the initial height over
the keel of the obstruction, the forces emerging from the penetration of the obstruction into the bottom of
the ship) and associated estimates of the damage to a ships bottom and seabed topology. In a similar type
of study Hong & Amdahl (2012) developed a simplified analytical method for the predictions from numerical simulations of structural performance of double bottom ship groundings over shoal seabed obstacles with large contact surfaces of trapezoidal cross-section. The method was verified against FE results
by Zhiqiang et al. (2011) and the verification was completed by comparing horizontal, vertical resistances
and the distortion energy between seven numerical simulation cases by their simplified analytical method.
Following a study by Khedmati et al. (2009) on the post-buckling behaviour and ultimate strength analysis of stiffened aluminium plates under combined axial compressive and lateral pressure loads, a set of
empirical formulations or equations was derived by Khedmati et al. (2010). The purpose of this work has
been to estimate the ultimate strength of stiffened plates under load combinations. The formulations were
verified against numerical results and they may be useful for the ultimate strength reliability analyses of
high speed aluminium made ship structures.
Yu et al. (2013c) introduced a theoretical model for the assessment of the structural performance of
stiffeners in way of the double bottom floor plating under a shoal grounding accidental scenario. Their
study introduced a set of numerical simulations that may be able to describe the progressive deformation
of a ships stiffened bottom structure because of the nonlinear effects of plasticity. They showed that the
preferred method for modelling such accidental scenario is to smear the stiffeners into the plating thickness. A comparison of the results from FE simulations indicated that the proposed analytical model accounts satisfactorily for the dominant deformation pattern.
Within the reporting period, FE analysis remains the principal approach for investigating structural response under accidental loading scenarios that may be associated with grounding, collision, ice structure
interaction or design for crashworthiness. For example, a few authors presented approaches where results
from detailed FE analysis were parametrically combined to produce simplified structural response methods (for example, see Ehlers 2009, Ehlers 2010, Ehlers & Tabri 2012, Tabri 2010). In these papers the
term engineering simplification reflects the derivation of combined numerical and semi-analytical techniques that could be used primarily for collision damage assessment procedures. Based on these developments it may be concluded that by using existing technology it is possible to apply an engineering procedure that considers the energy available for structural deformations as well as the damage extent for
various collision locations, striking angles and collision velocities. Such engineering approach that allows
for different accidental scenarios and load combinations may be significantly faster in comparison to a
fully coupled numerical simulation.
Liu et al. (2011a) proposed an ice material model that can be used in FE analysis of ship-iceberg collisions. This model was used further by Liu et al. (2011b) to perform large deformation, elasto-plastic FE
analysis of ship-iceberg collisions based on continuum mechanics. The authors concluded that the dissipated energy from an iceberg and foreship structure collision may be mainly determined by their masses
and the relative velocity between them. Hogstrm & Ringsberg (2012, 2013) compared the so-called
strength and ductile design approaches that may be used to improve the crashworthiness of ship struc-
147
tures. In their work the structure has been assessed against five criteria namely: (a) the intrusion depth, (b)
the energy dissipation, (c) the damage opening area after the collision, (d) the weight, and (e) the manufacturing cost. They concluded that there is potential for mitigating the consequences of a collision between ships by replacing conventional side-shell structures. The authors recommend that the global parameters of the collision case should be based on statistical analysis to establish the most likely values.
On the other hand, Koukounas & Samuelides (2013) addressed modelling aspects for the strength capacity of intact and damaged ship girders considering the criteria set by the IACS CSR-BC&OT. Zilakos et
al. (2013) proposed a methodology that combines CSR fatigue loadings with the FE method for the study
of cracks on actual marine structure subjected to high-cycle fatigue. The ultimate purpose of this work is
to establish a common base for evaluating different crack arrest technologies by developing a threecompartment model of a tanker. Accordingly, stress intensity factors were calculated for different geometries and loading conditions, and growth rate of cracks was calculated using the Paris-Erdogan law. It was
concluded that it is necessary to analyse fatigue cracks on a case by case basis due the variability in wave
loads, structural configurations and boundary conditions.
2.3
Reliability
Reliability can be defined as the ability of a structure to comply with given requirements under the operational conditions it may experience through its service life. Probabilistic analysis methods because of their
inherent variability are, in principle, capable to idealise the influence of such effects. In this sense reliability analysis may be used to measure the probability of structural failure by considering both the loads
acting on a vessel and the resistance (strength) of the structure.
This section of the report classifies strength assessment approaches as follows: (0) deterministic approach, (I) partial safety factor approach, (II) approximate reliability analysis, and (III) fully probabilistic
approach. Within the reporting period the first two approaches (0 and I) have been well used and implemented in design standards (see Chapters 3 to 6). Based on the allowable stresses design method (deterministic approach) the maximum acting stresses on a structure should not exceed the critical value of
material strength divided by a safety factor. The disadvantage of this method is that it relies on the evaluation of a suitable safety factor that at the moment may not necessarily consider load combinations or the
use of different materials. A more up-to-date method is the partial safety factor method (also known as the
limit-state method) that allows for design optimisation, load combinations and different materials (for
example, see BV (2014) and DNV (2011a)).
Class II and III approaches have the potential to provide a better indication of the structural reliability
at the expense of more information and computational effort. Examples justifying this statement can be
found e.g. in Zayed et al. (2013a) and Sobey et al. (2013). Taflanidis et al. (2013) addressed a simulationbased probabilistic framework for detailed estimation of the risk for tension leg platforms. Liu et al.
(2014) established the probability density equation based on a crack propagation rate model, and then
obtained the crack size probability density function that varies throughout the loading cycle.
In reliability-based design, the design value of the target reliability index can be derived by analytical
probabilistic processes. For example, Gaspar & Guedes Soares (2013) assessed hull girder reliability using a Monte Carlo-based simulation method. Silva et al. (2014) used FORM techniques to analyse the
ultimate strength reliability of a steel plate subjected to distributed and localized corrosion wastage allowance during the service life of a ship. Lus et al. (2009) studied the reliability of an accidentally grounded
Suezmax tanker with the objective to determine the influence of the damage on the ultimate moment of
the ship. In Deco et al. (2012) an efficient approach for the evaluation of ship reliability and redundancy
including the effects of corrosion for aging vessels was presented.
The concept of probabilistic analysis can be used to calibrate the values of the partial factors in the
Load and Resistance Factor Design (LRFD) methods. Faber et al. (2012) and Heredia-Zavoni et al.
(2012) introduced a generic framework for consequence assessment and risk analysis of Floating Production Storage and Offloading (FPSO) units for the purpose of establishing structural design criteria including the scenarios considered for the risk-based calibration of design codes. Another example of reliability
analysis to assess the implicit safety levels of the buckling strength requirements in IACS-CSR can be
found in Gaspar et al. (2011). In this study they considered three modes of collapse because of: (a) uniaxial buckling of the plating between stiffeners, (b) column buckling of stiffeners with attached plating, and
(c) lateral-torsional buckling or tripping of stiffeners. Using five oil tankers designs that represent the
range of application of the IACS-CSR design rules, they estimated the implicit safety level and variability
of each buckling strength requirement. It was concluded that probabilistic design methods are very important for the design of special cases or novel structures where previous experience does not exist.
148
3.
CALCULATION PROCEDURES
The degree of engineering complexity associated with different stages of the design process should be
aligned with the degree of fidelity in methods of analysis that are used throughout the design assessment.
For example, rapid assessment methods are needed in the early stages to assess global behaviour. More
detailed modelling and analysis techniques are required for detailed design. This chapter presents a range
of quasi-static engineering techniques that within the reporting period have been applied for design assessment. Special emphasis is attributed to the significance of qualifying the use of different technologies.
3.1
Taxonomy of engineering assessment methods
Analysis approaches range from first principles methods that can ensure that the structure is adequate at
preliminary stage, to ultimate strength methods used at the detailed design stage, reliability analysis
methods for lifetime prediction of loads, and optimisation analysis of the design for production loads
modelling.
3.1.1
Simplified analysis (rule-based design) / first principles
Simplified methods are preferable for the assessment of ship structures during the early stages of the design process. Laakso et al. (2013) presented an analytical method for calculating the fundamental frequency of cabin deck structures in passenger ships. The method considers the structure as a combination
of several structural members. Assumed mode shapes were created by using static bending equations.
Corresponding frequencies were then calculated by Rayleighs method. The method was validated against
fine mesh FE analysis for a variety of typical design space boundaries.
Paik et al. (2013) modified the Paik-Mansour method for the case of pure vertical bending moment.
The method is based on a credible bending stress distribution over the hull cross-section presumed at
the ultimate limit state. The accuracy of this method is demonstrated through comparison with computations obtained using more refined methods, such as the nonlinear FE method, the intelligent supersize FE
method (ALPS/HULL 2006), and idealised structural unit method used in the IACS-CSR. The comparisons showed that the modified Paik-Mansour formula was in very good agreement with both the nonlinear and intelligent supersize method and was shown to have good potential for the prediction of hull ultimate strength.
3.1.2
Direct calculations
Wilken et al. (2011) presented an efficient calculation method of Fluid Structure Interaction (FSI) in ship
vibration. Their approach is based on reduction methods for the hydrodynamic mass matrix and uses fast
solution methods for the exterior fluid problem when the velocity distributions of the shell are prescribed.
By projecting the vibration equation into a set of semi wet modes they combine a typical Lewis method
(Lewis 1929) with an advanced fast boundary element. The later yields accurate frequencies under forced
vibration conditions. The FE software package ABAQUS was utilised by Van den Abeele & Verleysen
(2013), who simulated the transient response of a subsea pipeline subjected to an underwater explosion.
In their approach an explicit dynamic solver was used to tackle acoustic pressure and structural response,
and to predict the behaviour of subsea pipelines exposed to an underwater explosion.
3.1.3
Reliability analyses
Hussein & Guedes Soares (2011) studied the ultimate strength and reliability of two single hull bulk carriers subjected to side collision and bottom grounding. The reliability calculations were carried out using
the COMREL software (Gollwitzer et al. 1988), considering the distribution of the extreme values of the
loads effects under different loading conditions. A sensitivity analysis was done to identify the importance of variables included in the used limit state function. Results showed that sagging is more critical
than hogging. Reliability assessment of the damaged ships showed that the homogenous loading conditions represent the safest condition, while the ballast loading condition and the alternative loading condition provided low reliability.
Material reliability was investigated by Yu & Karr (2011). In this study the statistical data for ship steel
strength suggested that best-fit probability density functions vary depending upon the type of steel and
the type of failure mode. The authors made use of several distributions such as the lognormal, Weibull, or
Gumbel to describe the yield stress, the ultimate stress and the failure strain. They selected several distribution functions for yield limit states of steels and studied the resulting reliability of a steel beam subjected to compressive loading conditions. In their approach the loading resultants were treated as random
variables. It was concluded that the probability of failure can vary by orders of magnitude for similar
nominal safety factors depending upon the type of steel or the yield function employed.
149
Kawamura & Miyazaki (2011) presented a method for structural optimisation with the main objective
function defined as construction costs. The method was tested on the hold frame of a bulk carrier. One of
the constraint conditions in the optimisation was the probability of failure in way of the hold frame. Life
cycle cost analysis considered both the risk of failure and the frequency of re-coating (as part of the maintenance plan). It was concluded that by considering risks associated with life cycle costs and life cycle
revenues CAPEX expenditure, safety and reliability are well balanced.
3.1.4
Optimisation-based analyses
Polic et al. (2011) identified the optimum shape for joint structures in terms of possible equivalent stress
levels by using FE analysis-based procedures and a genetic algorithm for fatigue life prediction. A similar
type of structural idelisation was considered by Yoon et al. (2011), who developed an optimisation
method considering the basic functions generated from Non-Uniform Rational B-Spline (NURBS). Numerical examples for design-dependent design problems were demonstrated to verify the effectiveness of
the proposed method.
Mao et al. (2014) focused on the dimensional optimisation study of the round-bilge craft taking the resistance as target. By multiple regression of the residual resistance, a fast method based on Froude theory
was set up to predict the resistance. They demonstrated that after optimisation of the principal dimensions, not only the resistance may be greatly reduced but the seakeeping indexes are also greatly improved.
Hu et al. (2013) presented a study where the magnitude of wind loads acting on a jack-up unit according to ABS rules (ABS 2010) for Mobile Offshore Drilling Units (MODU) was questioned in the context
of structure optimization. Following wind tunnel tests for a case study jack-up unit it was found that, in
comparison with the wind loads calculated according to the MODU rules, the measured wind loads were
lower. This discrepancy was explained by the aerodynamic interference between platform components,
such as wind shielding effects and acceleration flow effects that are not considered by the MODU rules.
3.2
Design for production loads modelling
3.2.1
Rules versus rational based ship design
For years, classification rules considered only still water and seaway loads. To this end, the use of empirical formulae for the determination of the scantlings of structural members and the suitable arrangement of
the ships principal structure has been fundamental. Quasi-static load assessment has been defined in way
of the so-called ultra-low frequency (still water) ship bending regime. This accounts for the differential
distribution between weight and buoyancy forces. On the other hand, quasi-dynamic low frequency bending has been considered to occur in way of frequencies associated with the natural seaway. Considering
that ship dynamics is primarily influenced by the time-dependent differential distribution between wave,
buoyancy and hydrodynamic effects, it accounts for the majority of stress reversals during a ships life
time (Hirdaris et al. 2010).
However, market dynamics as well as advances in maritime technology and innovation have led to the
development of new rules which are based to a larger extent on first principle approach. For instance,
IACS CSR-BC&OT for bulk carriers and oil tankers suggests uniform safety margins and rationalise design assessment methods and criteria (IACS 2013a, b and IACS 2015). Accordingly, dynamic loads consider the influence of the characteristic vessel motions, accelerations and, as applicable, internal tank
pressures. Dynamic loads associated with sloshing, local impact at the bottom forward, forward bow and
green water on deck are also specified.
3.2.2
Direct simulations for global quasi-strength assessment
The design of a ship involves formulating an accurate model of the ship to analyse its response internal
and external to its environment and the use of an optimisation method to determine system characteristics, while also fulfilling certain prescribed constraints on system characteristics and system response. A
traditional rationally-based design according to classification society strength assessment procedure comprises of the following four key steps.
1. Stochastic prediction of external loads by quasi-static methods.
2. Estimation of the limit values of load effects for all load conditions and cases.
3. Evaluation of the minimum required margins between load effects and their limit values on the basis of
a required degree of safety.
4. Analysis of the strength requirements based on design criteria that satisfy the design constraints.
150
For example, the above is well demonstrated by Payer & Schellin (2013) who presented an example for
the case of a 13,000-TEU modern container ship. Their analysis and discussion concentrated on aspects
related with the 3D overall FE idealisation of the ship structure and the mapping of quasi-hydrodynamic
loads for the prediction of the overall response. More detailed hydro-structural modelling issues related
with: (a) the application of quasi-hydrodynamic pressures, and (b) the combination of wave-induced dynamic loads, are discussed by Ma et al. (2014), Alfred Mohammed et al. (2012), and Papanikolaou et al.
(2014). These methods use quadratic programming to calculate equivalent nodal forces so that the resulting hull girder sectional loads match those calculated by seakeeping analyses, either by strip theory methods or 3D panel methods. To validate the approach 3D panel linear codes (e.g. MAESTRO-Wave), may
be useful. In the work by Alfred Mohammed et al. (2012) and Papanikolaou et al. (2014) a cross-spectral
stochastic analysis methodology for the determination of the combination of global wave-induced dynamic loads by taking into account uncertainties associated with the wave heading, the joint probabilities
of the wave environment and the correlations between different global wave-induced dynamic loads is
introduced. This methodology considers the use of bivariate probability density functions or the covariances of two random variables with their associated derivatives and assumes only rigid body hydrodynamic actions under steady forward speed conditions. The design extreme values of principal global
wave-induced load components and their combinations for a container ship progressing in irregular seaway were predicted using these two cross-spectral methods together with the short-term and long-term
statistical formulations. It is shown that in general terms both cross-spectral analysis methods can be employed to assess the effects of loads in ship design and reliability analysis. However, the cross-spectral
method predicted slightly higher load combinations than the cross-spectral probabilistic approach.
In all existing individual or common classification society quasi-dynamic strength assessment procedures the randomness of the ocean wave system is idealised in a statistical manner and in accordance with
IACS Recommendation 34 (IACS 2001). Accordingly, standard wave spectra and directional spreading
factors are used to describe the seaway mathematically and extreme lifetime loads are estimated from the
seaway distributions. However, individual classification society direct simulation procedures differ on the
basis of the quasi-design wave selection and evaluation of wave loads (see Table 1). Some classification
societies use stochastic analysis to decide for the effects of long-term statistical analysis on the equivalent
design wave loads (e.g. ABS 2006, BV 2006, DNV 2003, and KRS 2008). Others (e.g. LR 2013, 2014)
use the design wave loads of the rule formulae, in which the combined dynamic effects of wave and ship
motion are reflected, to apply equivalent static loads on the structural model. GL (2007) selects design
waves satisfying the rule design wave loads, using seakeeping analysis for the evaluation of motions and
employs an artificial static heeling for the additional torsional moment in the forward region of the ship.
The following key fluid structure idealisation aspects remain common in all classification society approaches.
1. The structural analysis is performed for the extreme lifetime loads and the global strength is assessed.
2. The dynamic loads including inertia forces, the hydrodynamic pressures and, where applicable, internal pressures of liquid cargo induced by ship motions are transferred to the structural model.
3. In those cases that the pressure distribution over the hydrodynamic model is too coarse to be used for
the structural analysis the pressure is interpolated linearly.
4. The unbalanced forces resulting from the difference between the hydrodynamic model and the structural model should be minimised.
5. The accelerations at the centres of elements, solid cargoes and liquid cargoes are calculated by the
combination of critical motions.
6. The loading conditions are determined based on the loading manual. The most severe cases are selected for the global analysis.
7. The FE models used are discrete and follow the arrangement of primary structure like decks, stringers, bulkheads webs and girders. Strength response is verified against strength criteria of yielding,
buckling, ultimate strength and fatigue strength. Since these criteria are the same in all cases, there is
a unified safety level of structure all over the ship.
8. The hydrodynamic model comprises the hull form and weight distribution. The weight of the ship
structure can be calculated from the global structural FE model using appropriate structural density.
The structural density sometimes needs to be tuned to achieve the total weight of the ship structure
taking account of local members, such as brackets, which are not contained in the global structural
model.
9. Transfer functions are obtained through hydrodynamic analysis. Short-term analyses are performed
for each irregular wave condition, namely modal period and significant wave height. For overall
strength evaluations, i.e. long-term analysis, the wave loads are imposed using waves a vessel may
encounter based on a probability of exceedance of 10-8 (IACS 2001).
151
10. The dynamic loads are represented by a series of load combination factors which represent the superposition of the various dynamic load components at a given point in time when the major dynamic
load component is being maximised.
11. For fatigue evaluations, representative characteristic loads are used to represent the large number of
modest fatigue-inducing fluctuating load ranges, which are based on a probability of exceedance of
10-4. Since fatigue calculation results are very sensitive to load and corresponding stress range applications, the most representative characteristic loads are applied which strive to eliminate any large
conservative assumptions. It should be noted that the follow-on fatigue calculation methods impose
safety margins later in the applied method and acceptance criteria itself; hence, imposing additional
conservatisms at the load determination stage is not necessary.
Table
1. Design wave selection and wave load evaluation.
_______________________________________________________________________________________________
Evaluation of wave loads
Classification society
Design wave selection
Rule
Stochastic analysis
Rule
Ship motion program
_______________________________________________________________________________________________
ABS
X
X
BV
X
X
DNV
X
X
GL
X
X
KRS
X
X
LR
X
X
_______________________________________________________________________________________________
3.2.3
Loads extracted from experiments and testing
The validation of loads on intact or damaged ship structures by model experiments and performancebased assessment is crucial within the context of design optimisation and design decision support. Within
the reporting period there seems to be a strong trend in the area of developing and validating performance-based methodologies for the design assessment of damaged ship hulls operating under harsh or extreme environmental conditions using quasi-static methods. For example, Saydam & Frangopol (2013)
presented a probabilistic framework for performance assessment of ship hulls under sudden damage accounting for different operational conditions. Their work considered the combined effects of sudden damage and progressive deterioration due to corrosion. The performance of ship hull was quantified in terms
of ship reliability and robustness. The longitudinal bending moment failure was evaluated within the context of quasi-static response and was considered as the limit state. The longitudinal bending moment capacities of the intact and damaged ship hulls were assessed using an optimisation-based version of the
incremental curvature method. In addition, aging effects on ship reliability were investigated. In a similar
type of study Begovic et al. (2013) presented a study on the prediction of motions of a frigate hull. Their
experimental campaign was conducted for two geosim models, 1:100 and 1:51 scales, at zero speed in
head, beam and quartering seas. All experimental results for the 1:51 model were presented as 1st and 2nd
order response amplitude operator, commenting on physical reasons for second-order response occurrences. The results show the changes in motion responses when a ship hull is in damaged condition. They
highlight the model scale effects and demonstrated the comparisons between the tests in which the model
may freely drift and those in which the mean position of the model is restrained.
More recently, Kukkanen & Matusiak (2014) presented numerical and experimental studies on nonlinear wave loads. The results from a nonlinear (transient) time domain Green function method were compared against experiments for a Wigley hull form in regular head waves. The influence of ship motions on
vertical shear forces and bending moments in regular and irregular head waves and calm water were
shown to be significant.
In response to wind farm vessels, Augener & Krger (2014) realised a study about the computation of
wave drift forces in regular and irregular waves for offshore wind farm installation vessels. In their study
the longitudinal and transversal drift forces and the yaw motion induced wave drift moment were calculated using potential flow theory. Augener & Hatecke (2014) analysed the seakeeping of an offshore wind
farm installation vessel during the jack-up process. The numerical computations were done with the
seakeeping code E4ROLLS (Petey 1988). Besides the limitations through the dynamic positioning capability of an offshore wind farm installation vessel, there are structural limits for the legs of the vessel. The
paper presents an analysis of the operational limits caused by the maximum acceptable impact loads on
the legs from the sea bottom during the installation or retrieval conditions of the jack-up vessel, due to the
motions of the vessel caused by the seaway. The design loads that were used were the axial and radial
impact forces acting on the bottom of one leg, while the impact itself resulted from the ships motion in a
seaway. The results showed that the operational limits of these vessels are not necessarily dependant on
impact loads.
152
3.2.4
Loads from seakeeping codes
A discussion for the different types of hydrodynamic idealisations was given in Section 2.1 of this report.
In addition, Peng et al. (2011) proposed an effective prediction method for the slamming loads of trimaran structures based on model tests and the similarity theory. To investigate the slamming load distribution and its relation to the impact velocity, the authors performed drop tests of a trimaran cross sectional
model. It was concluded that the peak pressure is proportional to the square of the impact speed and the
effects of nonlinear factors, such as air cushion and splashing, are quadratically declining as the speed of
impact increases.
3.3
Structural modelling
In recent years advances in computer technology enabled the development and implementation of improved FE modelling methods for the design of ships and offshore structures. This section presents a brief
overview and examples of studies dealing with FE modelling, models for global and detailed analyses,
and composite structures.
3.3.1
Finite element modelling
Element type
Within the reporting period considerable research efforts have been devoted into developing simple, robust, generalised and efficient element models. The use of conventional shell elements such as 4-noded or
8-noded quadratic shell elements is still broadly used. Rotation-free shell FE models have also been developed. In such idealisations the curvatures over an element are approximated in terms of deflection of
nodes in way of the adjacent element patches.
Avi et al. (2013) presented an equivalent shell FE that may be used for the assessment of ship global
and local static and vibratory response in the early stages of the design process. Their hybrid idealisation
comprises of three layer laminate elements that represent the plate, the stiffener web and the stiffener
flange. Accordingly, it accounts for in-plane membrane bending coupling and additionally shear stiffness.
Numerical comparisons with 3D fine mesh FE idealisations demonstrated good agreement.
To address the issue of localised response, Gannon (2013) used FE analysis to simulate the weld overlay procedure for corrosion repair of pressure vessels considering especially resulting residual stresses
and distortions. Their study investigated three different model types for the weld connection with changing element type, mesh configurations and material properties. It was demonstrated that the model with a
high mesh density in the vicinity of the weld had the greatest accuracy regarding maximum distortion and
residual stresses.
Model generation
In FE analysis a considerable portion of the total effort is still dedicated to model generation. In order to
speed up this process the shipbuilding industry and classification societies are investing in sophisticated
3D CAD systems with automatic mesh generators such as the Sesam GeniE software (DNV-GL 2015).
3.3.2
Models for global and detailed analyses
Objective of global FE analysis is to identify:
the stress levels on the plating of longitudinal hull girder structural members, and
the buckling capability of plates and stiffened panels.
Meier & Lehmann (2012) investigated the accuracy of different (global) coarse FE approaches for
modelling large openings such as holes in way of ships floor. They concluded that, in order to have a
more accurate behaviour under shear loads, it is necessary to represent the hole instead of reducing the
thickness.
Fricke et al. (2014) discussed the importance of FE mesh idealisation used to model walls with window
cut-outs. They concluded that modelling the correct shear stiffness of bulkheads in way of window cutouts is critical. The authors highlighted that in FE modelling the fidelity of the mesh is quite important.
Accordingly, rounded corners and/or insert plates should be modelled in detail unless shear stiffness is
pre-determined either by FE computation or an analytical calculation.
Fine mesh models may be used at the detailed analysis stage to determine (a) the stress concentration
factor of a complex structural geometry under a multi-axial loads or (b) hot spot stress in way of fatigue
corners. Refinement methods were evaluated by Fischer et al. (2011), who presented the computation of
the averaged strain energy in a cylindrical volume. The authors concluded that the most efficient approach to compute the Strain Energy Density (SED) is to perform a p-extension on a geometric mesh. If
153
FE computations are limited to low-order elements, geometric or radical mesh increase the efficiency
significantly. Although in 2D the SED can be also computed with h-refinement strategies, the efficiency
of high-order FE is very important. This is because 3D methods are not numerically efficient (Babuska et
al. 1981).
3.3.3
Traditionally the application of composite materials relates to the constant need for weight reduction.
Kunal & Surendran (2013) presented a work on the scope of using composites as major structural parts of
large commercial ships. This work focused on the evaluation of quasi-static and impact loads by FE
commercial analysis solvers (ANSYS and ABAQUS). Yang (2015) presented semi-analytical models that
can be used for prediction of post buckling response and ultimate strength of imperfect composite plates.
The models are based on a Rayleigh-Ritz method and can account for initial geometric imperfections,
post-buckling deformations, out-of-plane shear deformations, and failure and degradation models for
composites.
Jelovica & Romanoff (2013) carried out a theoretical investigation on the difference in the loadcarrying behaviour between a web-core sandwich plate, a stiffened plate and an isotropic plate. Their
study was carried out using two approaches, both solved with the FE method. The first method is based
purely on the deterministic 3D idealisation of a sandwich plate. The second is based on the principle of
the so-called equivalent single-layer theory of Reddy (2000). The results show that buckling loads of
sandwich plates range from 42% to 65% higher than those of stiffened plates. This is because a sandwich
plate is essentially a symmetrical structure for the cases where the coupling between in-plane and out-ofplane displacements does not exist. Furthermore, breadth-to-thickness ratio (representing local plate slenderness) is about two times lower in sandwich plates than in stiffened plates. Whereas the later prevents
local buckling, buckling loads may also reduce by increasing the transverse shear stiffness. Within the
reporting period another emerging field of application is the use of composite modelling idealisations for
the repair of damaged structures. For example, Avgoulas et al. (2013) studied the repairs in a cracked
structural component on an Aframax tanker. According to IACS CSR-BC&OT, cracks of different
lengths in two locations around the manhole of a hopper tank transverse web frame were evaluated. Different composite patch configurations were applied and studied for each crack case. Results showed that
stresses are significantly reduced. It was demonstrated that the reduction of stress intensity factor after
repair may range between 51% and 93% compared to the unpatched case.
3.4
Structural response assessment
3.4.1
Buckling and ultimate strength
Ozdemir & Ergin (2013) investigated the overall collapse of stiffened panels with imperfections under
compressive loads, by considering the influence of different stiffener type, web height and plate idealisations on ultimate strength. They concluded that: (1) for angle type stiffeners the stiffness of stiffener
components becomes more evident as the influence of nonlinear buckling becomes dominant; (2) when a
torsional deformation of a flat or deep tee bar panel structure becomes large it is difficult to predict the
stress in the stiffened plate with the beam-column model and it is necessary to consider the stiffener as a
plate structure. Yanagihara & Fujikubo (2013) derived an analytical solution based on the elastic large
deflection analysis to simulate the elastic behaviour of an isolated plate and stiffener under axial compression. The accuracy of their approach was validated by comparison with FE analyses and the buckling
software PULS introduced by DNV (2002). Santos Rizzo et al. (2014) developed a simplified model
based on nonlinear FE analyses to predict the ultimate loading capacity of stiffened panels under pure
shear stress. Their parametric model allowed for variations in geometry and initial imperfections. Results
compared to those calculated using DNV rules (DNV 2013b) and other studies (for example, see Santos
Rizzo et al. 2014 for details), were found to be in good agreement. It was confirmed that the ultimate
shear strength is sensitive to the magnitude and distribution shape of initial imperfections, except in cases
when the material yield shear stress is achieved.
Tekgoz et al. (2013) discussed the effect of residual stress on the ultimate strength assessment of stiffened panels. The ultimate strength was evaluated for three FE models accounting for different residual
stresses and boundary conditions in order to develop modified stress-strain curves which can be used
directly for assessing the ultimate strength of stiffened panels. It was concluded that residual stresses decrease the first yielding point of the structure response and initial geometrical imperfections may also lead
to strength reduction. While the structural capacity increases, the effect of residual stresses on the ultimate
strength decreases. Beznea & Chirica (2011a, b) presented an investigation of post-buckling behaviour
and estimation of global buckling ultimate strength of delaminated rectangular plates under shear. Their
study considered the influence of the position and geometry of elliptical delamination on the changes in
154
the buckling behaviour of composite ship deck plates. The authors made use of a delamination model out
of the software COSMOS/M, which represents damaged and undamaged part of the structure as layered
shell elements, separated by a hypothetical plane containing the delamination. The two sub-laminates
were modelled separately, with a surface to surface contact option, using 3-node shell composite element
(SHELL3L in COSMOS/M). FE analysis was thought to predict well global buckling loads of composite
plates on either side (i.e. sub-laminate) of the model.
Branner et al. (2013) worked on the numerical modelling of the buckling strength of composite laminates with delaminations. Nonlinear buckling and post-buckling analysis were carried out to predict the
critical buckling loads. Their innovative modelling approach made use of 20-noded solid elements for
every layer and results compared well with the multilayer shell element approach. Xu & Guedes Soares
(2011) conducted a numerical study to analyse the influences of boundary conditions and geometry on the
ultimate strength of stiffened panels under compression. Nine geometric configurations of stiffened panels were investigated including different boundary conditions for each case. It was concluded that
clamped boundary conditions on plate edges may slightly increase the panel ultimate strength.
Silva et al. (2011) investigated the effect of nonlinear randomly distributed non-uniform corrosion on
the ultimate strength of unstiffened rectangular plate that is subject to axial compressive loading. A series
of plate surfaces representing different degrees of corrosion, its location and ages were generated by a
Monte Carlo simulation. The results from nonlinear FE analyses were used in a regression analysis. Empirical formulae were derived to predict strength reduction because of corrosion with good accuracy to FE
analysis results. To address the importance of imperfections such as initial deflection and welding residual stresses, Li et al. (2013a) developed simplified simulation methods based on equivalent loads. In these
methods, the equivalent load of initial deflection is derived by coupling mathematical models describing
shell bending and the equivalent load of welding residual stress. Their study showed promising results of
the ultimate limit load in comparison to traditional direct calculation approaches and an experiment.
Wu et al. (2013) investigated the ultimate strength characteristics of a river-sea ship with large deck
openings. An experimental research has been carried out for a model ship hull which was designed on the
basis of the similitude theory. By means of the dimensional analysis, the similitude laws for both bending
and torsion were derived. The numerical analysis results of the ultimate bending strength under a desired
torque and the collapse behaviour at the ultimate limit state of the model ship hull from ABAQUS agreed
well with experimental tests. It was concluded that the influence of torsion on the ultimate vertical bending strength may be significant. A similar study by Pei et al. (2014) showed good agreement between the
test and the numerical models for the collapse behaviour and the ultimate strength.
Li et al. (2013b) presented a new approximate calculation method for the evaluation of ultimate bending moment of ship structures. For a typical stiffened panel a limit state analysis of strains and deflections
was carried out. The ultimate bending moments of example ship-sections evaluated using this method
showed relatively good agreement against experimental results, the simplified method introduced by
Smith (1977) and nonlinear FE analysis. Yamada (2014) evaluated the residual Ultimate Longitudinal
Strength (ULS) by comparing explicit FE analysis using LS-Dyna against Smith-type simplified analysis
(Smith 1977) using the code MARS2000 of BV (2000). It was concluded that in intact conditions, the
results for ULS are in good agreement. Residual ULS estimated by 3D FE analysis and that by the 2D
Smith-type simplified analysis showed fairly good agreement in case of a damaged condition where the
damage opening was included. However, in order to accurately predict residual ULS under high speed
collision it was recommended to use a 3D FE model of the struck vessel to calculate a realistic damage
opening size which thereafter is used in the following residual ULS analysis.
3.4.2
Fatigue strength
Within the reporting period research on fatigue strength using quasi-static methods concentrated on fatigue life predictions of cruciform joints with weld toe and weld root failure. For example, Fischer &
Fricke (2013) evaluated the initial crack shape and its influence on the crack propagation analysis when
using FE analysis. In this work different crack faces and shapes for weld toe and root failure arrangements have been simulated and compared against common approximate solutions. Sumi et al. (2013) investigated the effect of welding residual stress on fatigue crack growth lives of fillet weld with and without preloads. The as welded and after shakedown residual stress distributions were predicted by 3D uncoupled thermo-mechanical FE analysis. A crack propagation life of toe crack located at fillet welded
joints was predicted based on a crack opening and a closure simulation method. Huang & Xiao (2013)
investigated the fracture failure of longitudinal members including cracks. They employed the failure
assessment diagram methodology to assess the conditions of failure at the crack tip, establishing an analytical formulation of the crack-tip condition. In their approach, which was validated against FE analyses,
the material toughness is expressed in terms of crack-tip opening displacement. The failure stress of rep-
155
resentative cracked members is evaluated as a function of the crack length. This enables determining
critical crack lengths corresponding to the maximum stresses derived from extreme loads.
Pinheiro et al. (2013) studied the effect of deformations on the determination of Stress Concentration
Factors (SCFs) for the case of low speed collision of a supply vessel with an FPSOs side panel. Accordingly they developed an FE model to simulate the side structure and the striking bulbous bow and conducted a parametric study considering different penetration depths. Based on these results they developed
an analytical expression which gives SCFs as a function of the dimensions of the damage and the panel.
The SCFs obtained were used in a theoretical fatigue life study to estimate the duration of residual fatigue
life of a side panel collided by a supply vessel as a function of the damage magnitude. A similar type of
study was carried out by Pasqualino et al. (2013). Their FE model was used in a parametric study considering different penetration depths of the supply vessel on the side panel, resulting in different damage
magnitudes. The SCFs were estimated under the application of an in-plane compressive load in the panel,
reproducing the loading undergone by the vessel beam. Results were used to develop an analytical expression, with the aid of the Buckingham theorem, that linearly fits the numerical data and gives SCFs as
a function of dimensions of damage and panel.
Giuglea et al. (2011) developed an approach for fatigue assessment at early design stage using FE
analysis. To overcome the challenges due to limited information in the early design stage, generic structural elements and pre-defined fatigue critical details were chosen. This allows for the development of a
common approach for different ship types. Tanaka et al. (2011) adopted a shell-solid mixed analysis using semi-auto Rigid Body Element 3 (RBE3) in MSC.Nastran and the Virtual Crack Closure Method
(VCCM) for quadratic tetrahedral finite elements, to evaluate Stress Intensity Factors (SIFs) in surface
cracks of welded cruciform joints of a three cargo hold carrier. The shell-solid mixed analysis using the
semi-auto RBE3 connecting technique was performed on the ships structure. The SIFs were calculated
using the VCCM for the quadratic tetrahedral finite elements. The solutions were compared with the SIFs
calculated by the J-integral calculations of MSC.Marc. Thibaux & Cooreman (2013) determined SCFs for
tubular joints using solid quadratic elements. In this work results from computations were compared with
experiments and analytical expressions reported in literature and by ABS, DNV, the International Institute
of Welding (IIW), and Lloyds Register. It was concluded that whereas existing SCFs are not always
conservative in way of the crown toe of the weld chord they tend to be over-conservative at the weld
chord saddle.
3.4.3
Ship dynamics - vibrations
Boote et al. (2013) developed a detailed, global FE model for a 60 m super yacht with the aim to predict
the critical vibration areas. In their work, resonances from the propeller forces were considered to affect
the dynamic behaviour of the hull and superstructure. Comparisons of velocities, accelerations and frequencies to the maximum allowable levels from Bureau Veritas, Lloyds Register and RINA were presented for both global and local vibrations. It was concluded that the FE idealisation presented considers
most important parameters given by the classification societies. Mattioli et al. (2013) developed a new
method for establishing the load history at a site in case of metocean climate including combination of
several (up to 3 or 4) sea state components, such as those of main swell, secondary swell and wind sea.
Their method is applicable to both extreme conditions and fatigue assessment. They also showed that
long- and a short-term analysis with deterministic and probabilistic computation of vessel heading provide a reasonably conservative estimate of the vessel responses.
Rostami & Oskouei (2013) simulated the effect of seismic isolation (steel rubber bearings) installed on
offshore jacket platforms (between the topside and jacket structures) located in seismic regions, by using
the ANSYS FE software. A nonlinear FE model was applied for seismic assessment of isolated and unisolated platforms. Further on parameters for simulating isolators were defined. Their computational results
demonstrated a reduction in displacement from 20% to 73% when real earthquake excitations were considered (El Centro 1940, Kobe 1995 etc.)
3.5
Validation of calculation results
Enhancing safety at sea through the specification and quantification of uncertainties related with the description of the environment and predictions of loads and responses is currently one of the main concerns
of the shipping and offshore industry. These uncertainties play an important part in risk assessment for
the design and operation of marine and offshore structures. Whereas measured values are used in the
process of validating modelling techniques and associated assumptions both measurements and predictions in principle have errors associated with them. For example, high uncertainty of environmental description may lead to risks related with the seaworthiness of a ship. On the other hand uncertainties in the
modelling of loads and responses may adversely influence design assessment.
156
As part of the mandate of ISSC and ITTC to encourage cooperation in areas of mutual interest and in
order to address the subject matter of uncertainty qualification by the end of the last reporting period a
workshop was organised by the ISSC I.1 Committee on Environment, ISSC I.2 Committee on Loads as
well as the ITTC Seakeeping and Ocean Engineering Committees in 2012. Selected papers from this
workshop have been recently published (Hirdaris et al. 2014b). The topics discussed by the workshop
included:
model testing,
full-scale measurements,
load prediction techniques,
experimental validation techniques,
utilisation of satellite measurements,
extreme environmental phenomena,
risk assessment and mitigation, and
goal-based standardization.
In validating modelling techniques and associated assumptions, measurements, predictions and theoretical models are critical (Hirdaris et al. 2014b). Based on review by the committee the following sections discuss recent developments that relate with the evaluation or validation of quasi-static/dynamic
loads by model experiments and full scale testing. Special emphasis is attributed to methods that may be
used with the context of modern FSI idealisations, fatigue and ultimate strength assessment as well as
loads modelling under extreme environments (e.g. ice).
3.5.1
Model scale experiments and testing
Coupling of different simulation approaches allows for the simulation of complex systems. One interesting combination was to simulate rigid bodies moving in a fluid. For example, Beck et al. (2013) demonstrated that a pendulum moving in a water tank is a simple example of a complex system for which several effects have to be taken into account in order to reproduce the dynamic behaviour. In this work, the
simulation results for different immersion depths are compared against experimental data. The authors
demonstrated that when using a coupled simulation it is possible to merge several advantages of different
simulation techniques in one common simulation.
Bashir et al. (2013) presented results of towing tank tests carried out to predict the wave loads in regular wave conditions on a Deep-V hull form catamaran model. The experiments were carried out at the
Newcastle University towing tank using a segmented model of the universitys new research vessel, The
Princess Royal. The vessel is a twin hull with a Deep-V shape cross-section. The model, divided into
two parts at the cross-deck level, was fitted with a 5-axis load cell at the position of the vessels centre of
gravity in order to measure the motions response and wave loads due to the encountered waves. The longitudinal, side and vertical forces, along with the prying and yaw splitting moments were measured. The
results obtained were further compared with those from numerical predictions carried out using a 3D
panel method code based on potential flow theory that uses Greens function with the forward speed correction in the frequency domain. The results highlighted reasonable correlations between the measurements and the predictions as well as the need for a proper understanding of the response of the multihull
vessels to the wave-induced loads due to the nonlinearity that have been observed in the experimental
measurements of wave loads.
Sloshing
Within the reporting period research on sloshing loads has been ongoing due to the absence of fully validated unified methods. An overview of the actual status of engineering developments in this area is presented by Malenica & Kwon (2013). According to the authors the methodologies proposed by the classification societies for the practical design verification containment systems are based on the so-called
comparative approach which relies on the use of small scale model tests simulating intact ship operational
conditions and target ships (e.g. LR 2009). The former are used to deduce the conservative pressure scaling factor and the same scaling factor is applied to the target ship. The resulting pressure loading at full
scale is deduced and compared to the capacity of the containment system. Uncertainties relate with the
scaling factor which does not have clear rational justification since it reduces hydrodynamic phenomena
into a single number. Hence, challenges relate with: (a) how to scale the model test results, and (b) properly account for the structural elastic reactions considering that laboratory experiments are based on rigid
models.
Recent research work that attempts to shed some light into the afore-mentioned problems has been
carried out by Kim et al. (2012c, 2013a, 2013b, 2013c). The authors performed comparative study on
157
model-scale sloshing tests using a state-of-the-art 6 degree of freedom experimental rig as part of the International Joint Industry Project (JIP) PRESLO. This JIP was led by the Advanced Marine Engineering
Centre of Seoul National University in association with IACS Classification Societies (ABS, Class NK,
KRS, and Lloyds Register) and Korean major shipyards (DSME, HHI, SHI, and STX). Sloshing experiments were carried out using a state-of-the-art 6 degree of freedom experimental rig able to evaluate the
coupled influence of motions with sloshing loads under functional excitations corresponding to realistic
sea state conditions. Comparisons of the statistical quantities from 1:50 and 1:70 model scale tests demonstrated that uncertainties at model scale may relate with peak pressures, sensor types and sensing diameters. A simplified method for the pre-screening of sloshing severity has also been developed based on
comparisons between experimental and computational data.
Whipping and springing
Within the reporting period work on whipping and springing responses mostly relates to specific ship
types such as Ultra Large Container Ships (ULCS). This is mainly due to their slenderness, operational
speeds, and large bow-flare. Whereas the evaluation of the hydroelastic responses mostly aligns with the
remit of the ISSC Committee I.2 on Loads, incorporation of the influence of such loads and load effects
in the Rules for Ships implies the need to overall appreciate the influence of design innovation on design
margins. Within this context the committee considered important to reference two key developments supported by IACS Classification Societies and academics namely: (a) the International JIP WILS (Wave
Induced Loads on Ships) project initiated and led by MOERI (Maritime & Ocean Research Institute) (e.g.
Kim et al. 2014, Lee et al. 2012, Lee et al. 2014, Southall et al. 2014, Tiphine et al. 2014) and (b) the EU
FP7 project TULCS (Tools for Ultra Large Container Ships) (see IWWWFB 2012). Based on the committees consensus it is recommended that based on these studies further publications demonstrating systematically the departure of key design assessment criteria from the quasi-static/dynamic to the hydroelastic
region should be pursued further. For example, an experimental study on the bow-flare slamming loads of
a 10,000-TEU container ship was carried out by Hong et al. (2014a). In this study slamming loads, vertical bending moment, and torsional moment were measured in regular waves and irregular seaways and
for various speeds using a 1:60 scale model comprising of six segments connected by a U-shape steel
backbone. Strain gauges were installed to measure structural responses and fourteen load cells were distributed in way of the critical locations of the bow-flare area. It was concluded that: (1) the fluctuation in
the magnitudes of slams may be decreased when the impact forces are normalised by the instant surge
velocities, (2) as the ship speed is decreased critical slamming induced loads are observed in way of the
peak of the bow instead of the side cell bow waterline area, and (3) the magnitude of slamming loads is
proportional to the square of wave amplitudes.
Fatigue strength
The rapid enlargement of the size of container ships has also led to the application of extremely thick
plates in deck structures. This may grow concerns about the fracture toughness at the butt-weld with large
amount of heat input, and the arrest toughness of brittle crack propagation in the base metal of such thick
plates. Slamming induced whipping stresses might affect the fatigue crack propagation and the initiation
of brittle cracks in container ships. Based on results from a relevant JIP on this subject run by the Japan
Ship Technology Research Association (JSTRA), Sumi et al. (2013) presented two practical recommendations, that may prevent the brittle fracture of large container ships. Those are: (1) Ultrasonic Testing
(UT) during ship construction should be used to detect and remove harmful flaws of welding joints, and
may also confirm minimal fatigue growth in way of the welding flaws, and (2) sufficient brittle fracture
toughness of the welding joints for the prevention of brittle crack initiation.
Engineering experience suggests that brittle fracture may lead to large-scale hull fractures. Accordingly, it is important to ensure the hull structural integrity by preventing brittle cracks from being initiated
and also by providing proper means for arresting the brittle crack for those cases that crack propagation
occurs. Recent work by Kubo et al. (2012) attempted to shed some light into the subject by carrying out a
number of large scale and middle scale fracture tests for ULCS. They concluded that the required brittle
crack arrest toughness value should be more than 6,000 Nmm-3/2 to arrest brittle cracks in way of steel
plates with thickness exceeding 50 mm.
The structural longevity of marine structures where anti-corrosion measures are inadequate is another
area where understanding of uncertainties and their influence on quasi-static response may be important.
Common repair methods aim to primarily replace damaged steel structures. Alternative repair methods
suggest the use of patches made of composite materials. These methods account for the majority of the
problems faced by conventional renewal repairs. For example the paper by Karatzas et al. (2013) demonstrates an experimental study of artificially corroded steel plates repaired with composite patches. The
authors carried out tensile tests for eight specimens. The effect of aging was taken into consideration by
three different aging scenarios. Part of the experimental results was subsequently validated with the use of
158
numerical simulations that encompassed cohesive elements so as to simulate the debonding procedure.
Results showed that composite patch repairing was able to rehabilitate the defected steel plates and improve their load bearing capacity. The numerical results presented were found to be in good agreement
against experiments.
Ultimate strength
An experimental investigation into the collapse behaviour of a box-shape hull girder subjected to extreme
wave-induced loads was presented by Xu et al. (2012). Experiments were performed using a scaled tank
model and sacrificial specimens with circular pillar and trough shapes. Prior to the tank tests, static fourpoint-bending tests were conducted to detect the load-carrying capacity of the hull girder. It was shown
that the load-carrying capacity of a ship including reduction of the capacity after the ultimate strength can
be reproduced experimentally by employing the trough type specimens. It was also shown from the multiple collapse tests that the increase rate of collapse becomes higher once the load-carrying capacity enters
the reduction path while the increase rate is lower before reaching the ultimate strength.
Xu et al. (2013) presented a study on ultimate strength analysis of tanker hull. In order to obtain appropriate stress-strain curves for FE analyses, experiments on the collapse behaviour of stiffened panels were
carried out using geometrically similar scaling from small-scale specimen to full-scale stiffened panels.
The stiffened panel compressive ultimate strength test was designed according to geometrical scaling
laws so that the output of the test could be used as representative of the stiffened panels of the compressive zone of a tanker hull subjected to vertical bending moment. The ultimate strength of a case study
tanker hull was analysed by FE analysis using the experimentally developed master stress-strain curves
obtained by the beam tension test and the compressive test of the stiffened panel. The results from the
analysis were compared with results achieved by the progressive collapse method. Although not all parameters are scaled completely, for the full and smaller scale panels having the same collapse mode, the
difference in the ultimate strength is less than 12% which indicates that the similitude of stiffened panels
is still possible and practical.
In Wu et al. (2011), a structural model test method was introduced to study the ultimate strength of a high
speed trimaran model structure. They applied similarity theory with FE analysis as input to the structural
model design. To simulate the peak wave bending moment of the trimaran, three point bending was applied
by hydraulic jacks during the ultimate strength test. Load stress and load displacement curves of the model
and the global failure mode were obtained for the validation of FE analysis results.
Ice
The procedures to determine ice-induced global loads on a ship using full-scale data in accordance with
the method proposed by the Hydraulics Centre of the National Research Council of Canada (HNRC) are
described by Lee et al. (2013b). In this work 6 degree of freedom ship motions were found by processing
the measured linear accelerations and angular rates (using the commercial sensor named Motion Pak II)
under the assumption of rigid body motion. In addition, an algorithm and an analysis tool to estimate
global ice load based on LabVIEW using the ship motion was developed. Full scale data were acquired
while the ARAON rammed old ice floes in the high Arctic. Estimated ice impact forces for two representative events show 22-29 MN when ships operate in heavy ice conditions. In Ringsberg et al. (2014a), a
similar model was presented that uses recorded motions of a ship that collided with a heavy ice berg or
ice ridge. Results from predictions were found to be in good agreement with experimental measurements.
The other significant set of experiments includes improvements in lifeboat design. Full-scale field trials
of a conventional lifeboat in pack ice have yielded insights into the design and operation of evacuation
craft in ice. The multi-year trials program used an instrumented lifeboat to investigate design considerations such as powering and propulsion, hull form, manoeuvring, ice loads and ergonomics, as outlined by
Simes R & Veitch (2013). Operational issues that have been examined include ice management for
emergency evacuation, coxswain competence and training. Local loads were measured on an instrumented panel near the waterline on the port shoulder of a lifeboat during a field trials campaign that
spanned several days. This panel was subjected to impacts with ice during a series of transit tests in reasonably well-controlled field conditions. The lifeboat experienced impacts on other, un-instrumented
parts of the hull, which included the stem where the largest loads were likely experienced. Most of the
measured loads were well below the estimated ultimate strength of the lifeboat hull material. The highest
measured load was less than 40% of the estimated ultimate strength. By the end of the test program,
which included ramming and backing cycles, there was no evident damage to the hull. Hence, under the
ice conditions to which the lifeboat was exposed in the field campaign, the fiberglass structure was adequate. To increase knowledge about how ice loads affect the performance and safety of ships, Suominen
et al. (2013) manipulated data from a full scale ice trial data on S.A. Agulhas II that took place in the
Baltic Sea in 2012. These measured ice load data were compared with load prediction approaches given
by the Finnish-Swedish Ice Class Rules (FSICR) and the International Association of Classification Societies (IACS). For the bow areas, measurements and class predictions were found to be at the same
159
point of magnitude. This was not the case for the stern area during manoeuvres where load values from
Class Rules (FSICR 2010, IACS 2011) were found to be too low.
Collision
Knowledge about the level of structural damage after a collision is necessary for designing ships and offshore structures operating in ice-infested waters. An understanding of the physical processes during such
a collision is necessary to prevent (or limit) accidents, causing loss of life, the loss of a ship or environmental pollution. Accidental Collisions with Ice Masses (ACIM) laboratory tests are sensitive to structural design, i.e., the design of a structure that is flexible enough in relation to the ice mass. In such cases
both ice and structure should be able to deform during the collision event. Kim et al. (2013d) address
issues related to the planning of ACIM at laboratory scale with special emphasis on the choice of process
of ice manufacturing and ice mechanical properties, flexibility of impacted structure and scaling of the
experiment.
Schttelndreyer et al. (2013) reported on the current state of the SideColl JIP. Based on recent developments from this project for the case of a ship-ship collision, the struck vessel normally reacts as the
weak collision partner. The influence from uncertainties in material characteristics and collision scenariorelated parameters on the damage opening size and shape was investigated by Hogstrm & Ringsberg
(2012). In Hogstrm & Ringsberg (2013), four ship structures were assessed with regard to their crashworthiness. Most of them need a general redesign of the ship structure; hence it is difficult to implement
them only in critical parts of a ship. As a result, the Institute of Structural Mechanics of the University of
Rostock proposed an alternative stiffening system for double hull side. The basic idea of this design was
to connect the bulbs or face plates of two neighbouring profiles with a curved shell without changing the
conventional structure. This alternative stiffening system was shown to significantly increase the collision
resistance, but requires further investigations with respect to manufacturing as well as experimental and
numerical investigations which demonstrate the influence on collision safety. In addition, Buldgen et al.
(2013) presented a simplified analytical model of collision between two ships. By use of superelements they developed a new formulation for the estimation of impact resistance in way of inclined
ship side panels. They defined a super-element that may give the crushing resistance with respect to the
penetration of the striking ship. It was concluded that the numerical approach may lead to crushing forces
that are lower than those obtained using the FE software LS-Dyna. This could be attributed to the strain
hardening and the bending of the uppermost deck which are not taken into account by the numerical
model.
Figure 2. Results from FE analyses presented in Hogstrm & Ringsberg (2012) on how material characteristics
modelling, uncertainty in material parameters and striking bow stiffness influence the size and shape of
the struck vessels damage opening in the inner side shell.
160
Evaluation of ship collision safety is traditionally carried out with the striking ship assumed to be rigid.
This approach is in line with the approach of classic safety assessments (NORSOK 2004). In a recent
study Hogstrm & Ringsberg (2012) demonstrated that by systematic FE analyses the stiffness of the
striking and struck vessels may also have significant influence on the size and shape of the damage opening (see Figure 2). Hence, the safety level can be increased significantly if the design of bulbous bows
considers the influence of the aspect ratio of the bulb over the hull depth under the state of collision. In a
similar type of study Tautz et al. (2013) reported on collision experiments, with rigid and deformable
bulbous bows driven against double hull side structures, carried out in the test facility of the Institute for
Ship Structural Design and Analysis of TUHH in Hamburg, Germany. Their work may be considered as
good reference with regards to the level of maturity of numerical models against experimental results.
3.5.2
Full scale hull stress monitoring
Larger and larger ships are built and their ship motions and dynamic structural responses must be better
understood. Most new advanced ships have extensive data collection systems to be used for continuous
monitoring of engine and hull performance, for voyage performance evaluation etc. Information from
these systems is also a valuable source for the further development of quasi-static response design methods for future larger ships. Nielsen et al. (2011) recommended that such systems could be expanded to
include procedures for stress monitoring and for decision support, where the most critical wave-induced
ship extreme responses and fatigue damage accumulation can be estimated for hypothetical changes in
ship course and speed under real operational conditions.
During full scale measurements it is impossible to measure directly the hull girder loads. Bigot et al.
(2013) demonstrated that different procedures may be used to obtain the internal loads from strain measurements and the process of data processing can be carried out in two steps: (a) the distortion modes
should be incorporated in the structural model, and (b) a conversion matrix may be used to project the
measured values on these distortion modes and the internal loads may be obtained by recombination of
their modal values. The slamming behaviour of a large high-speed catamaran has been investigated
through the analysis of full-scale trials data and presented by Jacobi et al. (2013). This work presents sea
trials carried out by the US Navy in the North Sea and North Atlantic region on the 98 m wave piercer
catamaran, HSV-2 Swift, designed by Revolution Design Pty Ltd. and built by Incat Tasmania. For varying wave headings, vessel speeds and sea states the data records were interrogated to identify slam events.
An automatic slam identification algorithm was developed, considering the measured rate of change of
stress in the ships structure coupled with the vessels pitch motion. This helped with the identification of
slam occurrence rates over a range of operational conditions. The slam events have been further characterised by assessing the relative vertical velocity in way of slamming impacts. Since the ship was
equipped with a ride control system, its influence on the slam occurrence rates has also been assessed.
A slightly different approach was applied by Storhaug & Hareide (2013) who assessed the responses of
an ocean going vessel. Due to experience with whipping and springing, special attention to these effects
was also made during the design and approval. The vessel was consequently strengthened beyond the
minimum industry standard. It was concluded that since the measured fatigue life based on SCFs of 2.0
has been estimated to be well below the design life, special attention on cracks need to be taken from now
on if the trade remains the same in the future. However, no cracks have been identified so far during inspection. The maximum loading level has been higher than ever assessed by DNV before based on hull
monitoring data of blunt vessels. The rule of thumb value of 20% increase on extreme loading for blunt
vessels due to whipping has been exceeded. The wave bending moment according to IACS URS11 has
also been exceeded without whipping. The ultimate collapse strength has been assessed and compared to
the measured dynamic loading and allowable still water loading. When whipping is assumed fully effective to contribute to collapse, the safety margin is still above 1.0, but on the borderline of what is desirable. However, if the vessel had not been strengthened beyond the original design due to the concern of
whipping and springing, the safety margin would have been below 1.0. This may be the first documentation of a vessel that has been saved from breaking in two due to addressing springing and whipping properly during design.
Sder et al. (2012) presented a method for monitoring of racking-induced stresses in Ro-Ro ships.
Their approach assumed that racking stresses are mainly induced by roll and sway motions and therewith
related inertia and gravity forces. The method has been applied to full-scale measurement data from the
Wallenius Lines PCTC Mignon. Derived stresses showed good agreement with stresses derived from
strain gauge measurements, indicating that the method may be used as an alternative to conventional
strain-gauge-based monitoring. Motion-based stress monitoring has several potential areas of application
such as providing data for decision support, for live assistance and short-term route planning, structural
condition reports and for supplying feedback to the design process.
161
4.
UNCERTAINTIES ASSOCIATED WITH RELIABILITY-BASED QUASI-STATIC

RESPONSE ASSESSMENT
4.1
Uncertainties associated with loads
The uncertainties of loads and structural responses are important parts in the development and application
of the reliability-based design, assessment, inspection, maintenance, and repair methodology for marine
and structural engineering. Therefore, the studies on uncertainties and their assessment methods have
been and continue to be a large area of active research in marine structures; see Section 3.5 for a review
of work on validation of calculation results. This chapter focuses on the recent developments in uncertainties associated with reliability-based quasi-static response assessment since ISSC 2012.
4.1.1
Still water and wave loads
Chen et al. (2013) presented a rational reliability-based assessment procedure for hull girder ultimate
strength assessment of ship-shaped FPSOs. The stochastic model of still-water bending moment was established based on the loading conditions from the operation manual of FPSOs. A stochastic model was
proposed to represent the probabilistic characteristics of the extreme value of Vertical Wave-induced
Bending Moment (VWBM) based on its long-term distribution and the extreme value theory. The effects
of the return period of the VWBM, environment severity factor, and corrosion effects on hull girder reliability index were also investigated. Through sensitivity analysis it was concluded that the most uncertainties relate with the ultimate strength bending moment. Garr & Rizzuto (2012) developed a model for
the assessment of the still water bending moment of a tanker after collision. The position and extent of the
damage was modelled probabilistically. Results were compared in the light of the different features of the
input probabilistic models, and interpreted in terms of a ranking of the various situations according to
their probability of occurrence, the magnitude of the generated load effect and the spatial correspondence
between the maximum bending moment and the damaged area. The method was also used to provide
information for the selection of accident scenarios and design values for the maximum still water bending
moment.
Karmakar et al. (2013) estimated the long-term joint probability distribution of extreme loads for different types of offshore floating wind turbines by using the environmental contour method. In this work
the FAST code was used to simulate the wind conditions for various return periods and the design loads
of various floating wind turbine configurations. Rajendran et al. (2013) investigated the relative motion
and bending moment of a cruise vessel in extreme seas, looking in particular at the probability distribution functions of ship responses in irregular seas by comparing time domain simulations with model tests.
Accordingly, the quality of numerical predictions was assessed by validation of empirically derived probability distributions. Zhang et al. (2013b) employed the Weibull distribution fitted method and the stack
method to estimate the statistical value of wave loads induced by irregular incident waves and obtain the
long-term extreme value under a particular exceedance probability. Their work identified that the main
factors which may influence the long- or short-term prediction of quasi-static loads are the wave spectrum, the wave scatter diagram, the incident wave angle interval and last but not least the frequency interval. On the other hand, Si & Chen (2012) developed a method which could predict slamming pressure
directly for a vessel of unconventional hull form based on the probabilistic method proposed by Ochi
(1964). In this work long-term slamming pressures are predicted by a Monte Carlo method that makes use
of the instantaneous values of ship motion responses and slamming coefficients. Numerical results show
that slamming excitations in way of the peak and the trough of a wave will generate higher pressures. The
effect of relative velocity between the ship and the wave on the slamming coefficient varies with different
waves.
Abu Husain et al. (2013a) presented an efficient time simulation method to predict the probability distribution of the extreme values of quasi response based on the correlation between the response and its
corresponding linear response extreme values. Their method seems to be efficient for both low- and
high-intensity sea states. Wang et al. (2013b) investigated the suitability of three different probability
distributions for modelling of extreme responses by comparing them with empirical distributions obtained from extensive Monte Carlo time simulations. A mixed probability model consisting of both the
generalised extreme value and the generalised Pareto distributions was introduced to model the extreme
responses.
Mallahzadeh et al. (2013) proposed a very efficient time simulation procedure to predict the probability
distribution of the extreme responses based on a simple algebraic relationship between extreme responses
and their corresponding wet surface elevation. The case study demonstrated that accurate predictions of
the probability distribution of extreme responses can be obtained by a limited number (< 200) of short
response records (say, about 2-minutes each). A similar type of research was carried out by Kurian et al.
162
(2013). They developed the statistical models based on the collected data of jacket platforms which were
under construction at an ISO certified fabrication yard in Malaysia. The statistical analysis of the data was
performed and a Monte Carlo simulation technique was used to generate values for fundamental resistance variables using statistical distributions. The authors issued recommendations on the reliability
analysis of tubular members and joints. It is thought that their approach would be useful within the context of assessing the ultimate limit state design of jacket platforms.
4.1.2
Ice loads
Ji & Liu (2012) presented a state-of-the-art literature review on the area of ice loads on ships. They identified a number of strategic research directions related with: (a) the evaluation of pressures due to local and
global ice loads using numerical models, (b) the importance of comparing numerical approximations
against full-scale field observations and model tests, and (c) the need to understand the influence of operations such as manoeuvring and mooring on the global and local ice loads.
In a more focused type of work, Zhang et al. (2013a, 2014a) presented a mathematical model for calculating the geometric probability of ship-ice collision and carried out the studies of ship-ice collision loads
by using numerical simulation and model test. Taylor & Richard (2014) developed a probabilistic ice load
model to simulate ice loads during level ice interactions with a rigid structure based on the ensemble behaviour of empirical high pressure zones. The relationships between individual high pressure zones and
the spatial-temporal distribution of high pressure zones during an interaction were presented. Preliminary
results obtained from the proposed model were compared with the existing empirical models of local and
global ice loads acting on offshore structures. Zvyagin & Sazonov (2014) presented an ice load model
based on a stationary stochastic process with a lognormal distribution. In this work the simulation algorithms for the correlated lognormal processes was described in detail. It was concluded that the proposed
theoretical model and simulation algorithm provide fast way for probabilistic modelling stationary ice
loads process with lognormal distribution.
4.1.3
Combination factors
Mohammed et al. (2012) presented a cross-spectral stochastic analysis method for the combination of
global wave-induced dynamic loads by taking into account uncertainties associated with the wave heading, the joint probabilities of the wave environment and the correlations between different global waveinduced dynamic loads. The design extreme values of principal global wave-induced load components
and their combinations for a container ship progressing in irregular seaways were predicted by using two
cross-spectral methods together with the short-term and long-term statistical formulations. The comparative studies of cross-spectral probabilistic method and cross-spectral Hamiltons method showed that both
cross-spectral analysis methods can be employed to assess the effects of loads in ship design and reliability analysis. However, the cross-spectral Hamiltons method predicted slightly higher load combinations
than the cross-spectral probabilistic approach. Papanikolaou et al. (2014) presented recent advances in
modelling the combined hydrodynamic responses of ship structures using cross-spectral combination
methods and in implementing uncertainty models used for the development of modern decision support
systems as guidance to ships master.
Huang & Xiao (2013) performed a comparative study on the extreme values of the combined still water and wave load effects of oceangoing ships obtained by the Monte Carlo simulation and the theoretical methods. The obtained results show that there is very good agreement between the numerical simulation and theoretical methods. The empirical distribution of the combined extreme values simulated and
numerical theoretical distribution based on a load combination analysis can both be well fitted to an
analytical extreme value distribution model of Type II. Teixeira et al. (2013b) calculated the probabilistic characteristics of the load combination factors for still water and wave-induced longitudinal bending
moments of double hull tankers. The predictions of the different load combination methods were assessed based on the sample of five oil tankers adopted during the IACS-CSR design rules development
process. A parametric and an uncertainty propagation study were then performed to identify the range of
variation and the probabilistic models of the load combination factors that are applicable to double hull
tankers. The obtained results have shown that: (i) the average load combination factors of the tankers
decrease rapidly when the design period T decreases from 1 to 10 years, then tend to stabilize and that
the mean voyage duration demonstrates a strong influence on the load combination factors; (ii) the average load combination factors of still-water bending moment and wave-induced bending moment is 0.97
and 0.85 for double hull tankers in full load for the design period T=1 and 20 years respectively; (iii) the
average load combination factors of wave-induced bending moment is 0.84 and 0.81 for large and small
tanker in full load for a reference time period of 20 years, and reduce to 0.8 and 0.76 in ballast load,
respectively.
4.2
Uncertainties in structural modelling
4.2.1
Corrosion
163
Corrosion is a process of uncertain nature governed by many variables. Therefore, only probabilistic
models can describe the corrosion process itself and its effect on the strength of structural components.
Melchers (2012) presented a review of recent developments in the prediction of the likely future corrosion losses and of the maximum pit depth for steels exposed to marine environments. The review was
made for some typical applications, including marine corrosion of ship ballast tanks, corrosion of sheet
piling in harbours and corrosion of offshore platform mooring chains. A robust mathematical model based
on corrosion science principles was calibrated over a range of immersion conditions published in open
literature. These comparisons provided useful explanations for the effects of steel composition, water
velocity, depth of immersion and seawater salinity and it also facilitated new interpretations of data for
long-term pitting corrosion.
Jiang & Guedes Soares (2012) performed a series of nonlinear FE analyses on plates with partial and
through thickness corrosion pits. The effects of the radius, depth and location of corrosion pits and the
slenderness of plates on the ultimate capacity of mild steel rectangular plates under uniaxial compression were investigated. The results obtained show that the volume loss dominates the degradation of the
compressive capacity of pitted mild steel plates in addition to plate slenderness. Single side distributed
pits affect plates more than the double sided pits. Teixeira et al. (2013a) and Teixeira & Guedes Soares
(2013) proposed the probabilistic model for the ultimate strength of corroded plate. The spatial variability of corrosion wastage of plate was represented by a random field generated by using Monte Carlo
simulation.
The corrosion addition methods are widely adopted in structural design of individual classification societies, including CSR-BC&OT. Paik et al. (2013) investigated the historical trend in corrosion additions
in the structural design of ships and the effect of corrosion additions on the ultimate strength performance
of four double hull oil tanker structures, including Panamax, Aframax, Suezmax and VLCC. The ultimate
strength of hull girders was investigated in terms of the gross, half-corrosion margin deducted, and net
scantlings. The corrosion addition models specified by DNV Rules (DNV 2005) and CSR (IACS 2006)
were compared with a time dependent corrosion wastage model and the corrosion models suggested by
the Union of Greek Shipowners. Empirical formulas were proposed for the ultimate longitudinal strength
performance of double hull oil tankers for the different corrosion addition rules. In a follow up study Kim
et al. (2014) developed a Residual strength-Damage index (R-D) diagram by taking into account the timedependent corrosion wastage effects. The method was tested for an Aframax class double hull oil tanker
that had sustained grounding damages and suffered from corrosion at selected time intervals. The influence of those on net scantlings as specified by CSR was applied to gross scantlings using ALPS/HULL
(2006).
Kwon & Frangopol (2012b) analysed the reliability of ship hull girder structures by considering uncertainties in ultimate bending capacity and sea loads. The ultimate bending moment capacity was predicted
by time-variant random functions associated with corrosion and fatigue cracking. Still water and waveinduced bending moments were calculated using design-oriented and simplified direct methods considering different sea states as well as different ship operating speeds.
Mohd et al. (2013) developed a time-dependent corrosion wastage model for aging subsea gas pipeline.
The proposed empirical model gives simplification on the prediction of the pit depth of a gas pipeline at
any given age by manipulating the scale, shape and location parameter of the probability density distribution with respect to time. It was found that the pit depth of gas pipeline structure at subsea condition significantly increases with time and its corrosion progress can be modelled by 3-parameter Weibull distribution.
As an alternative to the traditional uniform corrosion model, Htun et al. (2013) introduced a random
field model based on the Karhunen-Loeve (K-L) expansion method, introduced by Ghanem & Spanos
(1991), to represent the stochastic properties of corroded plates. The stochastic properties of the minimum cross-sectional area of a corroded plate generated by the random field model were estimated as a
reference index. In a similar type of work Htun & Kawamura (2013) employed the random field model
based on K-L expansion method to generate the hypothetical corroded surfaces. The random characteristic of minimum cross sectional area of the plate with random field corrosion is calculated as a reference
index of strength reduction by using polynomial chaos expansion method as well as Monte Carlo simulation. The results showed that the strength reduction of the plate with random field corrosion was
smaller than the one obtained for uniform corrosion. Kim et al. (2012a) studied the ultimate longitudinal
strength of five different sizes of container ships to investigate the impact of considering aged corrosion
effects throughout a ships life. The motivation to the study was to show the necessity of corrosion
164
addition application to container ships and the relevant predictions regarding decreases in the longitudinal strength of new-building container ships after corrosion.
Mohd et al. (2014) assessed the reliability of ageing gas pipeline structures due to corrosion pits.
Measurement data for gas pipeline corrosion pit depth were collected and the statistical characteristics
were quantified by statistical analysis. The authors also reviewed available mathematical tools for the
reliability assessment of marine structures presented by Garbatov & Guedes Soares (2011). It was shown
that these models may allow for the existence of multiple cracks both in way of stiffeners and plating.
Within this context corrosion growth may be realistically presented as a time-dependent process. This
study recommended that greater emphasis should be given to the use of structural monitoring systems, not
only for decision support but also within the context of life cycle ship design and condition maintenance
schemes.
Saydam & Frangopol (2013) presented a probabilistic framework for performance assessment of ship
hulls under sudden damage (grounding and collision accidents) accounting for different operational conditions. Aging effects on ship reliability were investigated, i.e. the combined effects of sudden damage
and progressive deterioration due to corrosion. The performance of ship hull was quantified in terms of
ship reliability and robustness.
4.2.2
Structural characteristics
Akpan et al. (2012) presented a methodology for reliability assessment of damaged ships. The methodology recognises the existence of uncertainties when dealing with damaged ships and suggests a six-step
process, including: (i) definition of ship characteristics and operational profile, (ii) determination of damage size and scenarios by 2D/3D model representation of the damaged vessel, (iii) estimation of loads on
the damaged ships; (iv) estimation of the ultimate strength of damaged ship sections, (v) estimation of the
deterministic structural integrity of the damaged vessels, and (vi) estimation of the probabilistic reliability
of the damaged vessel. Ivanov (2013a) did a probabilistic calculation of the hull girder strength of a bulk
carrier. The calculation was based on a probabilistic representation of wave-induced and still water bending moment and resulted in the development of a probabilistic model of the hull girder section modulus
(elastic and plastic) including the influence of corrosion. In another paper (Ivanov 2013b), presented a
simplified method for the calculation of hull girder section modulus that was compared with his probabilistic approach when also accounting for the influence of corrosion.
Dec et al. (2012) presented a framework for the assessment of structural safety of ships under different operational conditions by evaluating performance indicators such as reliability and redundancy. Reliability and redundancy were based upon the evaluation of the flexural capacities associated with the ultimate hull girder failure and the failure of the first stiffened panel of a selected cross-section. Aging effects
due to corrosion were also investigated. The approach was applied to a naval catamaran vessel (Joint
High Speed Sealift) and results were presented in the form of polar representations of reliability and redundancy for different operational conditions. In extending this work Dec & Frangopol (2013) discussed
the development of a risk-informed decision tool for the optimal mission-oriented routing of ships. In this
work hull strength was computed by FE modelling of the midship section. The approach considered the
influence of different levels of damage and the response surface methodology was used to model the uncertainties associated with geometry and material properties under different operational conditions. Saydam & Frangopol (2013) further developed the methodology for performance assessment of ship hulls
that have undergone gross damage due to grounding or collision incidents. The combined effects of damage and progressive structural deterioration due to corrosion were also investigated and applied for the
case of an oil tanker. The variation of reliability and robustness indices for different operational conditions are again presented in polar plots with one half of the plot associated with performance in sagging
and the other half associated with performance in hogging. The methodology is used in a number of case
studies. In one of them, it is shown that some operational conditions result in significant reduction in the
performance of a damaged hull. In general, the worst performance is obtained under head sea, and the
effect of the sea state becomes more dominant when ship speed is increasing.
Zhu & Frangopol (2013) presented an approach for reducing the uncertainty in the performance assessment of ship structures by updating the wave-induced load effects with the data acquired from Structural Health Monitoring (SHM). The initial information on the wave-induced load effects was calculated
based on strip theory. Bayesian updating was used to estimate the parameters in the Rayleigh and Type I
extreme value distributions which were in turn used to model the peaks of wave-induced vertical bending
moment and the largest values of the peaks, respectively. Time-variant reliabilities before and after updating were evaluated. The study concluded that use of SHM data may reduce epistemic uncertainties.
Teixeira et al. (2013b) presented an approach for the assessment of the response of a structure with
random properties. Their approach considers the so-called level of certainty method that uses all of the
input random parameters and a more accurate first-order second moment sensitivity approaches to predict
165
the percentiles of the response. The ultimate strength of a corroded steel plate with random initial distortions and random material and geometrical properties predicted by semi-empirical design equations or by
means of nonlinear FE analyses was adopted as a case study to demonstrate the accuracy of the proposed
approach. The percentiles of the ultimate strength of the plates calculated using these simplified approaches were compared with the ones obtained by Monte Carlo simulation. Hull girder reliability assessment based on ultimate hull girder collapse of the midship cross section or its local failure due to
yielding or buckling of one of its structural elements was discussed by Gaspar & Guedes Soares (2013).
The geometric and material properties of the midship cross section elements were modelled as random
variables.
Zayed et al. (2013b) studied the lifetime reliability of ship hull structures subjected to corrosion degradation, including the effects of inspection and repair actions. The uncertainties in an inspection were accounted for by a practical probabilistic model. Accordingly ship loading uncertainties were modelled
based on the time ratio spent under each loading condition during the ships service life. Prestileo et al.
(2013) studied the hull girder reliability of a crude oil tanker with bottom damage. They examined a
number of possible flooding configurations, each one caused by a group of damage cases, characterised
by different location and extent. Static loads, wave loads and residual structural resistance were determined for each damage case, with the objective of obtaining a prediction for the probability of the hull
girders failure. A probabilistic Bayesian Network model was generated to deal with the interdependencies of different variables.
4.2.3
Reliability and risk-based structural assessment
Risk-based ship design is based on risk and reliability analysis as opposed to rule based ship design,
which is based on prescriptive, empirical rules (Papanikolaou 2009). When risk-based design is applied,
the safety of novel concepts can be quantified, even if they are not covered by formal design rules. Riskbased design can also be used for optimising an existing design with respect to safety. A pre-requisite is
that a consistent measure of safety is defined and can be quantified. When methods for risk and reliability
analysis are integrated, they can be used to balance requirements to safety and other design factors, e.g.
performance, life-cycle cost, and functionality. Risk-based methods gain more acceptance as decision
support tools in engineering applications. Today, all main elements of risk-based ship design and approval are being developed and early applications demonstrate their feasibility in practice.
4.2.4
Methods and criteria
An analytical method for the control of the strength of a grillage (gross panel) under unidirectional inplane axial load was proposed by Lokshin et al. (2013). The method was developed to: (i) calculate the
critical stiffness of transverse girders, (ii) calculate the maximum unidirectional in-plane compression
load when a structures scantlings are known, and (iii) calculate the required structures scantlings when
the unidirectional in-plane compression load is given. A probabilistic method for control of the strength
of ships deck structure under unidirectional in-plane axial load was proposed, which can be employed to
assess the effect of deterioration due to corrosion on the decks buckling strength. Faber et al. (2012)
presented a generic framework for consequence assessment and risk analysis of FPSO systems for the
purpose of establishing structural design criteria. Their work considered the output of the Joint Committee
on Structural Safety (JCSS) addressing the issue of system representation through exposure events, direct
and indirect failure consequences.
Patelli et al. (2012) presented and discussed efficient approaches for stochastic analysis and their applications to large FE models for the solution of realistic engineering problems. A general purpose software
was developed to provide integration between deterministic solvers for FE equations, efficient algorithms
for uncertainty management and high performance computing. It can be widely used for optimisation
analysis, life-cycle management, reliability and risk analysis, fatigue and fractures simulation, and robust
design. Mulder et al. (2012) presented a user-friendly method to model and visualise uncertainty in an
environment where little is known about the uncertainties involved. In this work uncertainty is defined at
particular nodes or elements of the FE model by using the interval or fuzzy logic method. Several case
studies were carried out to demonstrate the practical use of distance-based interpolation.
Zhu & Collette (2013) presented a new iterative inference algorithm for more accurate and efficient reliability evaluation of a dynamic Bayesian network-based deterioration process. The proposed algorithm
appeared to be robust and efficient in comparison to static discretisation procedures. Barltrop et al. (2012)
and Hifi & Barltrop (2014) have presented a methodology that has the potential to combine detailed
analysis of long-term experience from large numbers of ships with reliability and risk-based analysis
methods at both component and whole ship system levels. The methodology developed under the EU FP7
funded project RISPECT considers the influence of hydro-static and dynamic pressures, extreme events,
166
motions, fatigue induced global and local member forces, crack propagation, coating breakdown and corrosion modelling, structural reliability and fleet operational risk factors.
Li et al. (2013c) proposed acceptance criteria for fixed offshore platforms in the northern South China
sea under extreme storm events based on reliability analysis. The long-term distribution of extreme environmental loads was investigated. The role of ultimate strength analysis and reliability assessment in the
structural integrity management of fixed offshore structures was demonstrated by a pushover study for an
aging platform. Bai & Qian (2013) calculated one limit state equation of an unstiffened panel by using
MATLAB software and given the partial safety factors for the load and resistance factor design of the
panel under different reliability index levels. ani et al. (2013) presented an efficient method for ultimate
load capacity and safety calculation with inclusion of reliability and robustness-based design criteria,
including multi-criteria topology/geometry optimisation of the ship structural model and scantlings/material multi-criteria optimisation of structural panels. The method was verified with respect to
accuracy and speed on the box girder design and panel design with CalREL methods based on Monte
Carlo First Order Reliability Methods (FORM). Zavoni et al. (2012) presented applications of the framework for generic risk assessment for FPSO systems. The consequence models for structural risk assessment were developed for failure of hull girder sections, hull components and mooring lines using Bayesian probabilistic networks. The target reliability levels were estimated by: (a) optimising the expected
life-cycle costs, and (b) verifying compliance within the context of risk acceptance criteria.
4.2.5
Structural capacity
Yu et al. (2013b) proposed a method to calculate the relative angular deformation between arbitrary positions in the hull based on the simulation of irregular waves. The method was used in a case study where
spectral and reliability analysis methods were used to calculate the probability of peak response deformation beyond the critical value for the studied vessel. Ibekwe et al. (2013) carried out the hull girder ultimate strength reliability assessment of a damaged frigate by using the interactive-numerical method
(Ibekwe et al. 2011). A parametric analysis that considers different load cases obtained from experimental
wave measurements demonstrated the increased risk of failure of damaged ships. Pasqualini et al. (2013)
carried out a statistical analysis of the cross-correlation between geometrical parameters of a welded joint
and the spatial correlation along the weld. Jia & Moan (2012) studied the effect of sloshing in tanks on
the hull girder bending moments and the failure probability of an oil tanker in various damage conditions.
It is concluded that in certain tank resonance conditions, sloshing effects cannot be neglected.
Branner et al. (2013) presented a probabilistic approach to reliability assessment of fatigue critical
welded details in jacket support structures for offshore wind turbines according to ISO (2007) and DNV
(2011b). The fatigue stress cycles on the jacket members were computed by applying tower top loads
from an aeroelastic simulation with superimposed marine loads and in accordance to the IEC-61400-3
(IEC 2005) guidelines for operational conditions. The uncertainty in the fatigue stresses was taken into
account by using probabilistic S-N curves and a stochastic model to predict the failure in one chord/leg
intersection type weld. Differences in calculating and applying hot spot stresses in tubular welded joints
have been analysed and discussed. Mao (2014) proposed and validated an efficient spectral method for
the prediction of crack propagation in ship structures by taking into account the stress response amplitude
operators. A case study on the deck longitudinal stiffener of a 2,800-TEU container ship was carried out
to demonstrate the application of the method. The scatter of crack propagation associated with the wave
environment was also investigated. The results indicated the potential of crack inspection and maintenance optimisation to enable more efficient ship operation.
Cerkovnik et al. (2013) developed a method for the assessment of fitness for service of risers and flow
lines by using a Monte Carlo simulation. The authors evaluated the probability of failure based on incomplete in-line inspection data (from e.g. magnetic flux leakage to ultrasonic testing) and the statistical characterisation of other pertinent parameters. The case study results showed that these methods can be useful
in providing a basis for determination of fitness without the need to resort to arbitrary factors of safety. It
was concluded that using deterministic analysis may lead to conservative estimates in comparison to the
method developed. In fact a safety factor between 2.0 and 3.0 may be more appropriate where the target
annual probability of failure is of the order of 10-5.
Maes et al. (2013) carried out studies by the stationary and ductile probabilistic Failure Assessment
Diagrams. Based on this work it may be concluded that failure probabilities allow for an objective comparison of different cracks across a wide range of operating characteristics and therefore can be used to
establish a reliability-based integrity management plan. Mousavi & Gardoni (2014) presented a simplified
method for the reliability- and the integrity-based optimal design of engineering systems that directly
involves the system annual failure probability as a measure of system safety. By calculating the probability of system failure, their method could be used not only for risk-based decision-making but also for
structural inspection.
167
Faber et al. (2012) introduced a generic framework for consequence assessment and risk analysis of
FPSO systems based on the recent work by the Joint Committee on Structural Safety (JCSS) addressing
the issue of system representation through exposure events, direct and indirect failure consequences. The
scenarios considered for risk-based calibration of a design code safety format for FPSO systems were
outlined. Human factor related risk acceptance criteria have been determined based on the concept of the
Life Quality index. The consequence models were established by Bayesian Probabilistic Networks and
the target reliability indices were determined for structural design of FPSO components. Kawamura et al.
(2013) carried out life cycle structural optimisation of the midship section of a double hull tanker to
minimise construction costs and maximise life cycle benefits. Yang & Wang (2012) proposed a method
for fatigue reliability-based design optimisation of a bending stiffener, which may be employed to protect
the upper connection of umbilical/flexible risers against damage. Three meta models were constructed by
the optimum Latin Hypercube method (Stocki 2005) and the feasibility of the method was verified by a
case study of beam. The results demonstrated that the method is rational and improves the fatigue reliability of bending stiffeners. Yang et al. (2013a) applied a stochastic approach to the design of stiffened composite panels for which typical applications can be found in composite ship structures. A parametric study
was conducted using the Navier grillage theory (Blake et al. 2009) and FORM to investigate any detectable trend in the safety index with various design parameters. The reliability of the panels was found to be
sensitive to uncertainties in component thickness and fibre content. The fatigue reliability of fixed offshore platforms was investigated by Gholizad et al. (2012) by analysing different failure scenarios. In
order to evaluate the occurrence probability of a special scenario, it was divided into a finite number of
sub-scenarios and evaluated separately followed by a comparison of them. Based on the calculated values, the probability of occurrence was obtained for each scenario, and finally, the failure probability of
the entire system was calculated.
Yu et al. (2013a) conducted reliability analysis and reliability-based optimum design of thin-walled
structure as the stochastic structural system. The stochastic variables were categorised into two groups.
One is allowable stress and loading and the other is plate thickness and cross-sectional area of beam. The
failure probability of structural system was calculated considering beam cross-section area and plate
thickness are stochastic variables or not, separately. The optimum design for plane frame structure solves
the light weight problem of structure under constrain of reliability. Liu et al. (2014) performed the fatigue
life reliability analysis of submarine pressure shell butt weld by using the probability density evolution
method. The crack propagation growth probability density evolution equation was obtained by introducing extended state vector into crack propagation rate model with random initial conditions. The example
shows that results from the proposed method are in good agreement with the Monte Carlo method and
that the crack size probability density function evolves with loading cycle.
Present ship building rules follow the S-N curve approach to evaluate the fatigue damage at identified
locations while the use of fracture mechanics approach is yet to receive due attention. In Doshi & Vhanmane (2013), a methodology of evaluation of fatigue life of the longitudinal stiffener and transverse web
frame connection using fracture mechanics was demonstrated. The approach can be useful in application
to risk-based inspection of ship structures. This is because factors such as crack dimensions, crack growth
law parameters and applied loads are random in nature; hence, they were accounted for considering their
randomness. Examples of numerical results are presented for fatigue life of an oil tanker using the proposed method and IACS-CSR. The fatigue reliability of fixed offshore platforms was investigated by
Gholizad et al. (2012) by analysing different failure scenarios. In order to evaluate the occurrence of
probability of a special scenario, it was divided into a finite number of sub-scenarios and evaluated separately followed by a comparison of them. Based on the calculated values, the probability of occurrence
was obtained for each scenario, and finally, the failure probability of the entire system was calculated.
Another example study on the reliability-based investigation of fatigue damage was presented by
Srensen (2012). The required safety factors (namely Fatigue Design Factors - FDF values) were used for
fatigue design of steel substructures of offshore wind turbines. Design and limit state equations were formulated and stochastic models for the uncertain strength and load parameters were described. The effect
of possible inspections during the design lifetime was investigated. The results indicated that, for fatigue
critical details, where the fatigue load is dominated by wind load, FDF values equal to 2.5 are required.
Further, if the wave load is dominant, slightly larger FDF values are required.
4.3
Risk-based inspection, maintenance and repair
4.3.1
Inspection
Condition monitoring and maintenance management of marine structures are continuously being improved. The ever increased awareness of maritime safety and risks has contributed to the development of
risk- and reliability-based methods as a basis for inspection, maintenance and repair strategies. Hence,
168
marine structures can be operated safely and be maintained cost-efficiently in accordance with the established rules and regulations. This section presents a selection of recent work and methods that have contributed to reliability-based inspection, maintenance and repair, and their influence on uncertainty prediction of structural capacity and response. Ku et al. (2012) developed a systematic framework to assist operators in laying out inspection planning for semi-submersible structures using the structural reliability
analysis and risk assessment techniques. The inspection timing and frequency for various structural components can be determined based on the time-varying reliability index of the structural components under
consideration. Gud et al. (2013) presented a framework of risk-based inspection for a park of offshore
wind turbines taking into account the fact that only a sample of wind turbines is inspected at the scheduled inspection dates. The failure of a wind turbine is driven by the fatigue crack of one of its critical
structural detail and the uncertainties in the operation and maintenance costs due to the possible change of
maintenance strategies with respect to the number of wind turbines are considered. The numerical example of a wind farm shows that the proposed risk-based strategy significantly reduces the computational
effort required by the complete risk-based inspection analysis.
Guo et al. (2012) introduced a reliability-based procedure for effective inspection planning in order to
avoid the ultimate failure of deck plate of aging tankers. The time-variant failure probabilities of deck
plate were calculated by using the Latin Hypercube sampling method in the Monte Carlo simulation and
the time for thickness measurement of deck plate was predicted by comparing the calculation results with
target values. A total of 1,080 cases for 9 sample tankers were analysed to illustrate the procedure, including sensitivity and parametric studies. Frohbs & Lampe (2013) described some of the major aspects of
the methodology developed to estimate the risk to offshore pipelines and presented the results of a general
risk assessment of 22 aged offshore pipelines. The assessment reveals different governing threats and
failure modes for pipelines and risers, like ageing, free span or corrosion. The obtained results of the risk
assessment were used for the optimisation of the inspection intervals within the proposed risk-based inspection framework. It was concluded that the combination of remaining life time with index procedure
was able to cover all relevant failure modes. When only using the remaining life time approach threats
like impact damage, which is the reason for 30% of all pipeline damages, are not covered.
In order to optimise the inspection and repair schedules for structural components on ships, Vasconcelos de Farias & Netto (2012) conducted two case studies on corrosion damage of an FPSO due to its high
incidence among the observed damages. With the results from Bayesian inference, necessary subsidies
were developed for the next field inspection, establishing the basis for determining the regions where
survey is needed the most, and the corresponding inspection intervals. Goyet et al. (2013) presented a
study on risk-based inspection for offshore structures with the FPSO as an example. The study presented
strong insights about all the ingredients which have to be used to support the development of risk-based
inspection practices. In further, preserving the integrity of any system depends on predicting, assessing,
and preventing risks while any existing failure mechanism is mitigated. This depends on forecasts, technological innovations and finding appropriate solutions to prevent or deal with emergencies. Tammer &
Kaminski (2013) discussed an eight-step procedure for risk-based inspection (RBI) and its application for
safeguarding the hull integrity of floating structures (FPSOs), with primary focus on fatigue degradation
mechanism. The main steps included: (i) determination of asset integrity policy, (ii) functional decomposition and asset screening, (iii) assessment of probability of failure, (iv) assessment of consequences of
failure, (v) relative risk ranking, (vi) determination of inspection programme, (vii) implementation, data
evaluation and feedback, and (viii) conditional updating of the model. It is suggested that RBI be preferably used in combination with hull monitoring to gain the most of the benefit.
4.3.2
Maintenance and repair
A relevant problem nowadays that influences (decreases) the structural capacity with the operational period of marine structures to sustain loads is corrosion. In order to protect marine structures from corrosion
protective coatings are applied. The coating life is limited and hence it must be maintained or substituted
regularly following maintenance management programs. In Ulfvarson (2011), a state-of-the-art description concerning corrosion protection systems was presented which focused on the localised strain-induced
coating failures and some methods under development. Coatings and corrosion protection in ballast tanks,
quality requirements, coating breakdown and degradation resulting in corrosion were studied by Heyer et
al. (2013). Baere et al. (2013) presented an economical modelling approach which for a ships life-cycle
reduces the cost of ballast tank coating applied in double hull space ballast tanks of modern merchant
vessels. The proposed method was recommended to be used in inspection planning and evaluating the life
extension of existing offshore platforms. Examples of other studies where risk- and reliability-based procedures that can contribute to more efficient maintenance strategy and repair were presented by Guo et al.
(2012) who studied ultimate failure of deck plate by considering life degradation effects using the Latin
Hypercube sampling method in Monte Carlo Simulations. Kwon & Frangopol (2012a) estimated lifetime
169
maintenance costs due to fatigue damage based on reliability-based analysis. Obtained results show that
by tailoring the maintenance strategy to the design of the structure significant savings can be found in the
lifetime costs associated with the component. Temple & Collette (2013) developed a framework to estimate the lifetime maintenance cost of a naval structure based on an uncertain operational lifespan and an
associated maintenance schedule. The framework accounts for damage to the ships structure due to both
corrosion and fatigue. The maintenance schedule is optimised to minimise the total lifetime structural
costs for the ship based on global and local strength requirements.
5.
SHIP STRUCTURES
Rising oil prices and stronger emission regulations have influenced the trend in ship design to emphasize
on energy saving, slow speed operation, dual-fuel engine ship and LNG fuelled large commercial ships.
Furthermore, to reduce transportation cost of unitary container, designs of large container ships size are
still requested larger to increase the container capacity that a ship carries. Several 18,000-TEU container
ships were delivered from 2013 are now in service. For such designs, its more flexible hull and deflection
characteristics must be considered in the strength assessment, especially for whipping and springing phenomena that have a negative impact on the structural integrity. The design of 22,000-TEU container ships
which are approximately 450 meters in length will be the next challenge for designers and approval engineers. For bulk carriers and oil tankers, the CSR-BC&OT (IACS 2015) will enter into force on the projects contracted for construction on or after 1 July 2015. The important change in the direct strength
analysis, compared to the current rules, is that a cargo hold structural strength analysis will be requested
within the cargo hold region including the aft bulkhead of the aftmost cargo hold and the collision bulkhead. This also means that more accurate assessments are requested to ensure the structural safety of
commercial vessels.
Section 5.1 presents a historical background and an overview during the reporting period of the developments in international rules and regulations divided into sections for IMO Goal-Based Standards,
IACS-CSR and the development of structural design software systems. In Section 5.2, recent developments for ship concepts are highlighted with focus on container ships and LNG/LPG tankers.
5.1
Developments in international rules and regulations
5.1.1
IMO Goal-Based Standards
Goal-Based Standards (GBS) have been discussed by various ISSC committees over the last 10 years
(e.g. ISSC 2009, ISSC 2012). For this reason this section mostly focuses on key points and recent developments that have already impacted or will impact the implementation of unified or individual classification rules and design procedures relating with the prediction of loads on tankers and bulk carriers.
GBS were introduced in IMO at the 89th session of the Council in November 2002 through a proposal
by Bahamas and Greece (C 89/12/1), suggesting that IMO should play a larger role in determining the
standards to which new ships are built. The submission argued that the Organization should develop initial ship construction standards that would permit innovative designs. At the same time it should ensure
that ships are constructed in such a manner that, if properly maintained, they could remain safe for their
economic life. The IMO MSC 80 in May 2005 agreed on the basic principles of GBS as follows: GBS
are: (1) broad, over-arching safety, environmental and/or security standards that ships are required to
meet during their life cycle, (2) the required level to be achieved by the requirements applied by Classification Societies and other recognised organisations, administrations and IMO, (3) clear, demonstrable,
verifiable, long standing, implementable and achievable, irrespective of ship design and technology, and
(4) specific enough in order not to be open to differing interpretations. For bulk carriers and tankers the
GBS consist of five Tiers namely:
Tier I Goals: high-level objectives to be met.

Tier II Functional requirements: criteria to be satisfied in order to conform to the goals.
Tier III Verification of conformity: procedures for verifying that the rules and regulations for ship
design and construction conform to the goals and functional requirements.
Tier IV Rules and regulations for ship design and construction: detailed requirements developed by
IMO, national Administrations and/or recognised organizations and applied by national Administrations and/or recognised organizations acting on their behalf to the design and construction of a ship in
order to conform to the goals and functional requirements.
Tier V Industry practices and standards: industry standards, codes of practice and safety and quality
systems for shipbuilding, ship operation, maintenance, training, manning, etc., which may be incorporated into, or referenced in, the rules and regulations for the design and construction of a ship.
170
The GBS Tiers I to V constitute the IMO GBS, which became mandatory on 1 January 2012 under the
SOLAS Convention (new SOLAS regulation II-1/3-10), subsequent to the adoption of the following instruments at MSC 87 in May 2010:
New SOLAS regulation II-1/3-10 Goal-based ship construction standards for bulk carriers and oil
tankers (resolution MSC.290(87)).
International GBS for bulk carriers and oil tankers (resolution MSC.287(87)) (the Standards).
Guidelines for the verification of conformity with GBS for bulk carriers and oil tankers (resolution
MSC.296(87)) (the Verification Guidelines).
SOLAS regulation II-1/3-10 makes the GBS applicable to oil tankers and bulk carriers of 150 m in
length and above:
for which the building contract is placed on or after 1 July 2016,
in the absence of a building contract, the keels of which are laid or which are at a similar stage of construction on or after 1 July 2017, or,
the delivery of which is on or after 1 July 2020.
5.1.2
IACS Common Structural Rules for Bulk Carriers and Oil Tankers
IACS Common Structural Rules for Bulk Carriers and Oil Tankers (CSR-BC&OT) is applicable for selfpropelled oil tankers and bulk carriers with length greater than 150 m and 90 m, respectively, with unrestricted and worldwide navigation. The rules consist of two parts namely: (a) Part I that provides requirements common to both Double Hull Oil Tankers and Bulk Carriers, and (b) Part II that provides additional requirements applied to either ship type. The Consequence Assessment (CA) that was completed in
2014 (IACS 2014a, b, c) looked at scantling increases with no design changes and no design optimisation
was carried out to find increases as well as decreases. Some of the key output of the CA can be summarised as follows:
In general, CSR-BC&OT are more conservative than the CSR, for both oil tankers and bulk carriers.
CSR-BC&OT require that the cargo hold structural strength analysis is mandatory within the cargo
hold region including the aft bulkhead of the aftmost cargo hold and the collision bulkhead. The scope
covers five regions, including the midship, forward aft end, foremost and aftmost cargo holds (see Figure 3).The implementation of boundary conditions, loads and equilibrium, and permissible stresses, criteria for yielding, buckling and fine mesh have been harmonised.
The increase estimates calculated are approximations of the scantlings that may be needed to satisfy the
CSR-BC&OT requirements. In the computation of the estimates approximate methods have been used
to arrive at the scantling estimates, and the design have not been modified. Therefore, the final scantlings will not become apparent before new designs have been generated.
The original CSR net scantling approach is retained and the corrosion margin values have in general
been left unchanged. This is because corrosion data collected from bulk carriers and tankers have been
extensively re-analysed, verifying that existing corrosion margins are well on the conservative side.
The general tendency for oil tankers and bulk carriers is that the scantlings will increase when CSRBC&OT apply.
The newly introduced FE analysis for fore and aft cargo hold will generally not increase scantlings due
to the yielding assessment.
The additional class notation and assignment of the GRAB notation is intended to handle heavy grabs.
For bulk carriers the GRAB requirements lead to scantling increases in several holds. The FE buckling
requirements will generally increase scantlings for oil tankers and bulk carriers.
In general, fatigue analyses for both ship types give lower fatigue lives in the CSR-BC&OT than in
CSR.
With regards to the stress acceptance criteria, in most of the current classification rules criteria are implemented for the yielding check of local stress concentrations assessed by means of direct calculations and
modelled by fine meshes. However, due to differences in approach not all stress acceptance criteria are
the same. An overview of the stress criteria as defined in the main class rules is given in Table 2. It should
be noted that the table only contains an excerpt from the actual stress criteria and is therefore not complete. What can be observed is that there is a variety of mesh sizes to be used. Furthermore, stresses are in
most rules to be taken at the elements mid-plane. Only ABS (2014) explicitly allows incorporating plate
bending stresses in the analysis. The CSR-BC&OT stress acceptance criteria are identical to CSR-OT,
171
except that an additional 20% of stress may be allowed for details that comply with the fatigue assessment
criteria, which in other rules is not possible.
Figure 3. Definition of cargo hold regions for FE structural assessment.

Table
2. Stress criteria defined in main class rules (*DNV (CSA), see DNV (2013a)).
________________________________________________________________________________________________________
Rules
ABS
DNV
DNV* KRS
LR
BV
CSR-BC CSR-OT CSR-BC&OT
________________________________________________________________________________________________________
10-8
10-4
10-8
10-8
10-5
10-8
10-8
10-8
gross
net
net
gross
net
net
net
net
net
s-t
suitable t
s
1-1.5t 5050 s/4
5050
5050
15-800
15
800
22.5
50
200
50
50
382245
400
282
353
294/ 280
353/
353/
224
335
400
400
*Notes
(7)
(8)
(9)
(10)
(11)
(12)
(13)
Reference (6)
ABS
DNV
DNV
KRS
LR
BV
IACS
IACS
IACS
(2014) (2014b) (2013a) (2013) (2014) (2014) (2012a)
(2012b)
(2014a)
________________________________________________________________________________________________________
Probability (1)
Scantling (2)
Mesh (3)
Element length (4)
all (5)
Notes:
1. Probability of exceedance of load case for which the yielding check is performed.
2. Modelling of the scantlings, either by including (gross) or excluding (net) the corrosion margin of the plating.
3. Mesh size to be used in way of local stress concentrations. s indicates the general frame spacing, t plate thickness. Otherwise the element size is given in mm (e.g. 5050 indicates an element size of 50 by 50 mm).
4. Typical length of element side considering the mesh criteria and a typical spacing of 800 mm and typical plate thickness of
15 mm.
5. Allowable equivalent stress for mild steel in N/mm2.
6. Reference to the appropriate rules and regulations.
7. Non-tight structures, or tight structures with tertiary stresses accounted for.
8. For end bracket of stiffeners of unproven design.
9. Various general criteria depending on location.
10. Applicable to the stress analysis of hatch corner radii onboard container vessels.
11. The allowable stresses given are for elements respectively adjacent or not adjacent to a weld.
12. The allowable stresses given are for elements respectively adjacent or not adjacent to a weld and only for load cases where
static and dynamic loads are combined.
13. The stress criteria in the CSR-BC&OT are equal to those for CSR-OT, except that for details complying with fatigue assessment criteria stresses increased by a further 20% can be accepted.
Harmonised common structural rules introduce hull girder residual strength assessment of the hull in
damaged conditions. They provide a calculation formulation using an incremental and iterative Smithtype procedure. Various researches have proposed alternative methods for the calculation of the ultimate
strength of hulls with non-symmetric sections. FE codes have been also employed for the calculation of
residual strength. When using FE the main concerns of the users are: (a) the selection of mesh, (b) the
selection of the appropriate solver, i.e. implicit or explicit, and the parameters related to it, (c) the boundary conditions, and (d) how to apply the action, i.e. bending moments with or without external loads or
rotations of the end sections. An essential issue is also the description of the actual properties of the hull
including the effect of corrosion on the thickness of the structural members.
For evaluating different ship structural crack arrest technologies, Zilakos et al. (2013) proposed a
methodology that combines CSR fatigue loadings with the FE method for the study of cracks on actual
marine structure subjected to high-cycle (low-stress) fatigue. An Aframax tankers mid-hold model subjected to different fatigue scenarios were investigated. The proposed method is simple and based on rules
172
and regulations widely accepted in the ship industry. For the residual strength of hull in damaged conditions, CSR-BC&OT provides a calculation formulation using an incremental and iterative Smith-type
procedure (for details, see Zilakos et al. 2013). The FE method has been also employed for the calculation
of residual strength.
Samuelides et al. (2013) investigated the sensitivity of the FE analysis results from the modelling parameters. Guidance for the selection of modelling parameters so as to obtain convergence is provided.
Finally, the determination of the residual strength to a bulk carrier and a tanker subjected to collision
damage using ABAQUS/Explicit was presented.
Nam & Choung (2013) investigated the accuracy of average compressive strengths of stiffened panels
on ultimate longitudinal strengths of hull girder. Average compressive strengths based on simplified CSR
formulae and FE analysis were calculated for stiffened panels. Then hull girder ultimate strengths have
been calculated using the in-house software UMADS which is based on the progressive collapse method
with a new convergence criterion. It is concluded that average compressive strengths from FE analyses
induce conservative prediction of hull girder ultimate strength, because CSR formulae estimate less conservative shortening behaviours of stiffened panels than FE analysis results.
5.1.3
Development of structural design software systems
Structural design software systems including class rules calculation, FE assessment, combination of hydrodynamic analysis and structural analysis gradually become very important tools for ship designers and
classification societies. This is because of the maturity of FE technology and the advancement of the rule
requirements that now require the application of advanced direct calculation procedure for the prediction
of quasi-static loads. For ship structures that may be prone to structural damage because of environmental
loads that arise from impact loads (e.g. slamming, sloshing, green water on decks, etc.) most methods
account for the instantaneous peak pressure. With this in mind Kim & Paik (2013) developed an advanced
design formula that predicts the permanent deflection of plates and stiffened panels in the time domain.
Their work proposed the use of two parameters that can describe the impact loads namely peak pressure
and duration time. However, further comparisons with nonlinear FE methods of stiffened plate structures
under impact loads are required.
Ma et al. (2013) suggested that multi-objective optimisation methods can be used to determine the
Pareto optimal solutions of a stiffened panel based on the ALPS/ULSAP (2006) algorithm. Objective of
this work has been to solve a design problem by simultaneously minimising the weight/cost of a stiffened
panel, and by maximising its buckling and yielding stresses. Two methods, namely Pareto Simulated Annealing (PSA) and Ulungu Multi-Objective Simulated Annealing (UMOSA) were presented for a single
panel optimisation. The loads applied to the panel were assumed to be constant. An iterative procedure
was used to optimise stiffened panels. The numerical results showed that the proposed method is very
useful to perform ultimate strength based ship structural optimisation with multi-objectives, namely
minimisation of the structural weight and cost and maximisation of structural safety.
For large ship structural design, the Direct Load Analysis (DLA) for strength assessment of hull structure is adopted to strengthen the initial design based on rules or to satisfy the requirement for a specific
notation. However, this analysis is too complicated for designers because it includes quite a number of FE
analyses and corresponding parameters to be considered. To satisfy the requirements of the various classification societies and to consider the designers needs, Won et al. (2013) developed a process of the structural assessment using direct load analysis which was standardized to consider various rules and the calculation procedure is automated to reduce the calculation effort. The system is mainly composed of four
parts according to the general procedure of the dynamic load analysis such as the interface to ship motion,
load transfer module to the finite element model, global to local mapping module in the FE analysis and
fatigue life estimation modules.
Brindley et al. (2013) presented two approaches for the practical fatigue strength assessment of container carriers. Those reflect the industry demands for design and design assessment. In the former a
simplified stress concentration factor is defined to evaluate the hot spot stress range. This is directly compared to an allowable hot spot stress range/mean stress curve that may be used to assess the fatigue
strength. The latter records the hot spot stress directly from a local fine mesh model. Mean stress, material
strength and plate thickness effects are incorporated to evaluate an equivalent hot spot stress range. The
authors presented one possible concept concerning the influence of the so-called shakedown stress on
material strength effect and the expected S-N curve modification. Fukasawa & Mukai (2013) discussed
the effects of hull girder vibration on fatigue strength for a Post-Panamax container ship. In this study the
container ship under investigation was assumed to encounter 100,000 sea states according to the occurrence probability of the short-term sea state given in the IACS North Atlantic wave scatter diagram.
When the influence of hull flexibility was considered in the calculations the total cumulative fatigue damage factor doubled. They concluded that a stress range around 100 MPa may have a significant effect on
173
the fatigue damage in the ship. It is important to confirm whether this phenomenon can be observed or not
in the actual fatigue damage of a ship taking account of the nonlinearity in crack initiation/propagation
stages. When the smaller amplitude component of higher frequency vibration is superimposed on the
larger amplitude component of lower frequency vibration, the enlargement effect of the total amplitudes
of superposed stress may significantly affect fatigue life. However, the number of cycles of the small
amplitude component cannot be disregarded.
In a study by Jeong et al. (2013b), fluid coupling effects in tanks with stiffened plates were investigated by experimental measurements and FE numerical analysis. The results demonstrated that the differences of natural frequencies between in- and out-of-phase modes may increase depending on the fluid
depth. The reason is that the fluid effect on out-of-phase modes affects the mass of tank structures. The
influence from natural frequencies depending on water level in way of the adjacent tanks is small, while
the influence of mode shape can be significant. Yang et al. (2013b) estimated the slamming impact loads
and dynamic structural responses of container ships at an initial design stage using a direct analysis
method. The slamming impact pressures and dynamic structural response were studied using the commercial CFD program STAR-CCM+ and the structural analysis software ABAQUS. These two programs
were coupled using the co-simulation function of STAR-CCM+, known as the one-way coupled FSI
scheme. Numerical simulations were carried out for the bow bottom and the stern slamming impact loads
of a container ship in extreme design wave conditions.
5.2
Special ship concepts
5.2.1
Service vessels for wind mills and offshore platforms
Offshore structures are designed for certain accidental scenarios. This is done to guarantee a sustainable
level of equivalent safety. Accordingly, accident scenarios are defined to prevent that the consequences of
an accident are disproportional to the original cause. Within the reporting period a suitable example of
application is presented in Storheim & Amdahl (2014). They investigated the damage of offshore platforms that may be subject to ship collisions. In their analysis they considered bow and stern impacts of a
supply vessel having 7,500 tonnes displacement against the column of the floating platform, jacket legs
and braces. The effect of ship-platform interaction on the distribution of damage has been studied by
nonlinear shell FE simulations. For bow collisions the crushing behaviour and potential penetration of the
bulbous section into cargo tanks or void spaces of the floating platform was studied in detail. For the
jacket braces the study was focused on whether it is possible to penetrate into the ship bow without significant plastic bending or local denting. At post-processing stage the collision forces were compared
against the NORSOK (2004) standards. In a similar type of study by Notaro et al. (2013), they studied the
collision between the offshore installation and a modern 10,000 Offshore Service Vessel (OSV) with a
bulbous bow. The residual hull girder strength of the unit was assessed to identify potential threats to hull
integrity.
5.2.2
Container ships
The contribution from springing and whipping on fatigue and extreme loading may be significant for long
slender ships with large openings such as container vessels. Therefore, within the reporting period a number of studies based on available full-scale measurement data attempted to quantify the influence of hull
flexibility on the fatigue life of such vessels. For example, a study by Barhoumi & Storhaug (2013), utilised 4-year full-scale measurements to investigate the influence of whipping and springing effects, on an
8,600-TEU container ship. Their study demonstrated that fatigue damage initiates under head or beam
seas conditions. Whipping contributes significantly to the increase of the dynamic extreme stresses in
deck, hence doubling the dynamic extreme stress especially under hogging but also under sagging conditions in way of the aft quarter length of the vessel. Whipping-induced loads lead to amplification of the
dynamic stresses which may exceed the quasi-static IACS rule by up to 48% in hogging. A similar type
of study was conducted by Renaud et al. (2013) and Andersen & Jensen (2013) on a 9,400-TEU container
ship based on full-scale measurements over a 6 year period. These studies concluded that the inclusion of
the effects of hull girder flexibility may have a negative influence on the fatigue life. In particular the
work by Andersen & Jensen (2013) investigated the efficiency of spectral fatigue analysis methods to
capture the influence of non-Gaussian/narrow banded frequency components that emerge in way of hull
girder induced resonances. The authors suggest that such approaches are not accurate. Notwithstanding,
the agreement between the spectral methods and rain flow counting methods is generally good and the
narrow-band approximation seems to yield a fast and fair estimate of the fatigue damage.
In a recent study, Fricke & Paetzold (2013) carried out fatigue tests of a welded detail subject to variable amplitude loading. They investigated the effect of whipping stresses on fatigue damage and the suitability of the Palmgren-Miner rule for fatigue life assessment. Their study has been based on theoretically
174
derived/idealised sinusoidal stress histories and stress histories from onboard measurements. The results
showed that most of the fatigue damage is caused by the low frequent stress cycles, which may be
enlarged by whipping, as long as the whipping stress amplitudes are smaller than the wave-induced stress
amplitudes. The contribution of the additional small stress cycles due to whipping is rather small. This
validates a simplified approach proposed by the authors where only the enlarged wave-induced cycles are
considered by a correspondingly modified cumulative distribution of stress ranges.
Tsai et al. (2013) carried out a frequency domain weakly nonlinear seakeeping/linear sloshing coupled
analysis for an 8,000-TEU container ship under ballast and full load conditions. At first instance, their
analysis highlighted the influence of resonant frequencies that correspond to roll motions by scatter diagrams. Consequently, the structural strength in way of the boundary of ballasted cargo holds was assessed. The risk induced by resonance of ship motion and sloshing water in cargo holds were confirmed
in the design stage. A frequency domain model was utilised in a hydrodynamic numerical code HydroSTAR (see Tsai et al. 2013 for details), and the analysis included nonlinear seakeeping, linear sloshing,
and the coupling calculation of the motion equation. The additional resonant frequencies, especially for
the rolling motions, induced by the motion-sloshing coupling effect were clarified. The structural strength
of the boundary of the ballasted cargo holds was assessed for those critical rolling motions by the BV
numerical code HOMER (BV 2010). Further, Yang et al. (2013b) estimated the slamming impact loads
and dynamic structural responses of container ships at an initial design stage using a direct analysis
method as mentioned in Section 5.1.3.
5.2.3
LNG / LPG tankers
In recent years there has been increasing demand for energy efficiency, green shipping and reduced emissions. Considering that LNG fuel is considered as one of the most promising alternatives with low environmental footprint it is not a surprise that within the reporting period research and development efforts
concentrated mostly on understanding the influence of quasi-static loads on the global integrity of LNG
ships and their Cargo Containment Systems (CCS) as well as vessels that carry LNG as fuel.
Bang et al. (2013) developed the design of an enhanced 56,000 m3 IMO type B LNG cargo tank and assessed their design against DNV rules (DNV 2008, 2009). A study by Lee & Zhao (2013) revealed that
the maximum von Mises stresses of every CCS component exceeded the reference values in the horizontal and vertical CCS members, except for the top and bottom R-PUF in the horizontal CCS member. Even
though longer duration of shock pressure responses could be obtained using large water shooting water
model, response to shock and fluid flow may be more severe compared to those of small water shooting
models. Chun et al. (2013) evaluated the structural capacity of a corrugated membrane, against hull deformation, sloshing and thermal loading. Their study suggests the introduction of a failure criterion based
on the rupture strain of the material considering in way of the first CCS barrier. Paik et al. (2014) have
also studied sloshing loads in the storage tanks of LNG FPSOs and proposed a new method for the determination of nominal values of sloshing loads.
Hwang et al. (2013a, b) studied the structural responses due to sloshing loads of membrane type LNG
CCS. A three step numerical assessment is proposed which incorporate: (a) linear static, (b) transient
dynamic, and (c) localised FSI analyses. In a case study on a small-scale insulation box that represented a
LNG tank, the approach was compared against results obtained from small scale model tests. Considering
the uncertainty of sloshing impact loads, the results between the numerical simulations and the experiments show satisfactory agreement with regard to e.g. pressure history acting on the wall of the box.
Rudan et al. (2013) studied the crashworthiness of a typical LPG ship. Nonlinear FE analyses were
used in parametric studies to compare the collision resistance of a conventional side-shell structure design
with an innovative space-saving sandwich hull structure. The results show that the sandwich hull structure is more collision-resistant in terms of more energy-efficient absorption, however, the sandwich concept should be further studied. Ehlers (2013) proposed a particle swarm optimization-based procedure to
obtain a crashworthy ice classed LNG tanker. The procedure utilised a number of selected arctic materials, which further improved the collision resistance at sub-zero temperature. As a result, the LNG tanker
scantlings were optimised for local impact and compared to standard rules based concepts to identify the
potential gain in collision resistance. In Ehlers et al. (2013), the ultimate strength of an intact and damaged LNG vessel subjected to sub-zero temperature (SZT) due to cold climate was investigated. This
temperature influence is included by explicitly characterizing the material properties down to -90 degrees
Celsius. A ship collision simulation introduced damage to the hull girder and thereby allowed for a comparison between the ultimate strength of an intact and a damaged hull girder. A simplified method was
used to calculate the ultimate strength based on the individual panel contributions, both for the intact and
damaged conditions.
Low temperature problems in structural assessment of LNG carriers were studied by Biot et al. (2013).
In order to suggest a procedure for the structural design of a type C tank and its supporting structures,
175
they compared different approaches for numerical simulation of the interaction between tanks and ship
structures. The outcome of the study was an FE-based simulation procedure for heat transfer analysis
applicable on LNG carriers. Wang et al. (2013a) developed a procedure for thermal analysis and strength
evaluation of cargo tank structures in offshore FLNGs and LNG carriers. In the study, a heat transfer
analysis methodology was employed, and a computational tool was developed for application on hull and
tank structures of both membrane-type and independent Self-supporting Prismatic type B (SPB) LNG
vessels. Using this method, the temperature distribution and corresponding heat transfer coefficients
(HTCs) in the hull structure can be estimated so the appropriate steel grade can be selected for the inner
hull and the boil-off rate (BOR) can be calculated. FE results, from numerical simulations where the temperature range during operation was -170 to 20 degrees Celsius, were used for assessing the yielding and
buckling strength of a tank structure.
Jeong et al. (2013a) proposed a strength assessment procedure of Mark III CCS plates which incorporates FE analysis together with various strength-based failure criteria for composite materials. The
strength assessment in the study was performed within the initial failure state of a Mark III CCS plate.
Failure details like failure locations, loads and critical stress values can be identified by the proposed approach.
Storhaug et al. (2013) studied strain measurement data of structural members onboard two LNG vessels. The data has been collected for a period of 5 years which enables the analysis of whipping and
springing fatigue-related-effects of LNG carriers during different seasonal and weather conditions. The
two vessels have been sailing in the world wide trade except for the North Pacific area, and the time spent
in the North Atlantic is about 40%, which is higher than for typical LNG vessels. The vessel speed has
been relatively low in average and well below the service speed. The results show that whipping and
springing have resulted in a significant increase in accumulated fatigue damage fatigue damage in the two
vessels.
5.2.4
Other ship types
Passenger ships
Modern passenger ships and mega cruise liners have high and long superstructure that comprise of several
decks supported by pillars, longitudinal and transverse bulkheads on the hull and large openings. These
characteristics generate nonlinear strain distributions in way of the midship section and raise the demand
to understand the usefulness and practicality of applying advanced structural analysis techniques. Melk et
al. (2013) investigated the shear-induced secondary normal stresses in the balcony openings of modern
passenger vessels with narrow superstructures. The investigation was carried out using the FE method. In
their study two quasi-static loading schemes were considered: (a) a cosine shape loading simulating the
wave actions, and (b) a four-point bending load enabling deeper analysis on the shear-induced responses.
It was demonstrated that the shear-induced normal stresses have considerable effect to the overall stress
state around openings. Romanoff et al. (2013) presented a study of the interaction between the hull and
the superstructure for a passenger ship exposed to bending loads. An advanced beam theory model that
takes into account the influence of vertical and shear stiffness between various decks was employed and a
Pareto optimisation procedure was followed to establish the significance of different weight and vertical
centre of gravity distribution on overall quasi-static response. It was demonstrated that vertical and shear
coupling between different decks may significantly affect the hull girder response of passenger ships. The
authors concluded that the vertical bending moment is shared equally by the hull and the superstructure,
while an optimal vertical centre of gravity may considerably increase the share of load carried by the superstructure.
Korhonen et al. (2013) studied the influence of surface integrity on the fatigue strength of high strength
steel used in balcony openings of cruise ship structures. The fatigue test specimens, having a dog-bone
shape and yield strength of 355 MPa, 460 MPa or 690 MPa, were cut by plasma. After cutting, the specimens were treated by grinding or by grinding followed by sandblasting, i.e. using post-cutting treatments
suitable for shipyard conditions. The resulting surface roughness and hardness profile were measured.
Fatigue tests with load ratio 0.1 were carried out and the investigation demonstrated that post-cutting
treatments suitable for shipyard conditions can considerably increase the fatigue strength of high strength
steel.
Sailing yachts
Shimell et al. (2012) presented the Dream Symphony project which is a 4-mast staysail schooner. With a
length of 141 meters (463 feet), she will not only be the largest private sailing yacht ever constructed, but
also an all Glued Laminated Wood (GLT) construction, incorporating some composite materials and
steel. The requirement to use laminated wood as the main building material for a yacht of this size posed
several challenges. However, the first aim was to make sure that the structural arrangement provided suf-
176
ficient strength and stiffness to withstand the forces imposed by hydrodynamic and rigging loads. Due to
the unique nature of this project most calculations were done using first principles. An extensive material
testing program was also conducted to determine the material design properties and account for the very
specific nature and variability of wood. After an initial design using a two-dimensional approach, full FE
analysis of the boat was carried out; see Shimell et al. (2012) for details. By presenting the design methodology, specific aspects of the laminated wood construction and key findings of the FE analysis conducted, the study illustrates how modern design methods and tools can be applied to design such a unique
yacht.
Due to the increasing demand of methods and tools for the analysis of yacht behaviour in a realistic
environment and in particular the development of time domain approaches able to simulate yacht motions
in a seaway, a number of Dynamic Velocity Prediction Programs have been developed. For example,
Fossati & Muggiasca (2012) presented an opening simple model with the aim to reproduce unsteady sail
aerodynamics taking into account three dimensional effects and unsteady mainsail-jib interaction. The
authors assumed that the yacht design scale and its wave pattern are short compared with the time and
length scales of the wave motion. Consequently, it was modelled as a single point mass constrained to
move on a surface governed by the equations of wave motion and the equations of vessel motion were
derived.
The racing yacht design process, except the hull and appendages, involves the selection of sails, rigging and mast. Proper selection and scantling calculations of the mast and standing rigging is of crucial
importance as it is the backbone and connection of sail induced loads on the hull. Zamarin et al. (2013)
presented a novel methodology for the optimal selection of a mast and standing rigging based on the application of the Analytic Hierarchy Process (AHP) method on a 40ft racing yacht. Their approach consists
of three phases. The first phase identifies of possible design solutions. The second one specifies the best
design for stability configuration. The third phase is used for load calculations, scantlings determination,
and final approval of the project variables. Papantonatos et al. (2013) presented an experimental study on
the dynamic performance of a BOC-50-foot sailing yacht model. A scaled model of the hull form with the
keel-bulb configuration has been tested in the towing tank of the Laboratory for Ship and Marine Hydrodynamics of the National Technical University of Athens, Greece. During the tests the dynamic responses, as well as the added resistance were recorded. Results referring to the resistance, the side force,
the centre of gravity displacement, the pitch as well as the vertical accelerations of the model at the bow,
the centre of gravity and the stern were presented. Moreover, using a velocity prediction program, the
polar and the stability diagram of the tested sailing yacht were calculated. Useful conclusions about the
dynamic behaviour of the model were obtained.
6.
OFFSHORE STRUCTURES
Within the reporting period research efforts concentrated on the assessment of the consequences from
extreme, and accidental loads by advanced simulations such as nonlinear transient analysis, computational thermal-fluid dynamics, heat transfer analysis, gas dispersion simulation, thermal elasto-plastic
analysis, etc. Advances in the use of probabilistic methods for risk and uncertainty quantification have
also been significant.
6.1
Types of analysis for various floating offshore structures
Song et al. (2013) evaluated the effect of sloshing impact on the fatigue damage of an independent Type
B LNG by sloshing model tests utilising a 2D rectangular tank with internal structures. Impact pressures
were measured on stringers and a fatigue assessment was performed for a welded joint of the stringer on
the side wall. By means of the experimental data, a procedure for the fatigue assessment of structural
members inside such a tank was proposed and validated in a case study of a FLNG ship.
Pasqualino et al. (2013) and Pinheiro et al. (2013) developed a theoretical model for the evaluation of
the SCFs and their influence on the fatigue life of FPSO side panels that have been damaged by collision.
In this work a parametric study considering different damage magnitudes was carried out and the results
obtained were used to develop an analytical expression to provide SCFs as a function of the dimensions
of the damage.
High-frequency vibrations of Tension Leg Platforms (TLPs), commonly known as ringing and springing, have challenged TLP designers since the first full-scale TLP was installed in the North Sea in 1984.
In Muehlner et al. (2012), a nonlinear time-domain model is presented of a TLP that exhibits the ringing
and springing response of the vessel. The analysis model uses large displacement theory for the vessel
and tendons and a semi-empirical wave model based on a modified linear wave theory. Predictions of
vessel motions and tendon loads made with the analysis model were compared to model tests and were
found in good agreement with the measurements. The analysis model was also used to investigate
the fatigue damage in the tendons caused by the vessels high-frequency response. Tendon stress time
177
histories were computed for nine different unidirectional sea states. The analysis showed that highfrequency response was present in all sea states even though ringing occurred only in sea states with significant wave heights above 3 m. Tendon fatigue damage contribution from high-frequency loads were
found to be significant in every sea state.
Schmidt et al. (2014) performed a study on the fatigue reliability analysis for a brace-column connection of a semi-submersible hull. Three spectral methods were used to evaluate equivalent fatigue loads
and fatigue damage. These included the Rayleigh method, the Modified Rayleigh method with bandwidth
correction, and Dirliks method, to estimate fatigue damage using Miners rule. Uncertainties in the fatigue life assessment procedure were introduced in the analysis. The main sources of uncertainties in the
fatigue analysis were those in a proposed equivalent fatigue load estimate in the short term, in the S-N
curve parameters, and in the stress concentration factors. The results show that the frequency-domain
methods did not lead to significant differences in predicted fatigue loads and damage. One reason could
be that, for example, the cycle counting for Dirliks method is based on the rate of peak occurrences,
while the Rayleigh and Modified Rayleigh methods count cycles based on the zero up-crossing rates.
Ringsberg et al. (2014b) studied the importance of choosing a suitable fatigue calculation method for
the design of offshore structures. In their work they used two different methods, the structural hot spot
and effective notch stress method in a comparison study for two typical fillet welded joints. They found
that local weld parameters, such as weld size, may have a considerable effect on fatigue life. They also
concluded that wrong design criteria, including local weld parameters, may increase the chance of weld
root failure, meaning that a local method, such as the notch stress method should be used for calculating
the fatigue life for the critical area.
Gao & Low (2014) presented a new and efficient method for the simulation of long-term fatigue of
deepwater risers. The method makes use of a statistical sampling technique, so that the results from the
simulation of one sea state can also be used for other sea states. Application on a turret FPSO moored in
500 m water depth showed that the new method yields relative precise estimates of the fatigue damage,
when compared with the classical Monte Carlo approach.
In offshore structures, collision and wave impact are important issues in terms of structural integrity.
The residual strength of damaged structure as well as post-ultimate strength have been also treated as
important. Ning et al. (2013) suggested a numerical approach for the evaluation of the structural integrity
of a generic Spar hull in collision with a large supply vessel. Dynamic and nonlinear FE analysis were
implemented using ABAQUS/Explicit module respectively for two collision scenarios namely: (a) a realistic simulation where the impact kinetic energy governed by an initial impact speed and total mass of a
ship is gradually depleted during the collision, and (b) a simplified analytical method where the impact
speed of a ship bow throughout the collision is constant or the total impact energy is unlimited.
Venzon et al. (2013) introduced an FE methodology that can be used to verify the structural integrity of
damaged vertical webs of a semi-submersible, Mobile Offshore Drilling Unit (MODU). The methodology
isolated one of the damaged vertical webs and applied the boundary condition on the adjacent structure
where the web is located. The residual strength characteristics of damaged stiffened cylinders, namely,
ring-stiffened and stringer-stiffened cylinders subjected to combined axial compression and radial pressure were investigated using the FE software package ABAQUS (Cerik et al. 2014). The damage process
was explicitly simulated by means of quasi-static impact analysis in which cylinders were assumed to be
dented by a rigid indenter. Subsequently, a series of the FE analyses were carried out for a number of
damaged stiffened cylindrical shells. Closed-form expressions that may be useful for the prediction of the
residual strength of damaged stiffened cylinders were derived. Those may be useful for reliability-based
studies.
Amante et al. (2014) presented a study on residual compressive strength of dented FPSO side shell
panels. Their FE analysis was carried out in two steps. First, collision damage by a supply vessel was
imposed using the FE program ABAQUS/Explicit. After the indentation, a compressive load was applied
and then the panel residual compressive strength was calculated. Moreover, in general, perforated plates
containing cut-outs need to be carefully evaluated in terms of buckling failure. Park et al. (2013) carried
out a series of nonlinear FE analyses to evaluate the buckling/ultimate strength according to various geometry and loading conditions, namely axial compression and in-plane edge shear. Based on these results,
a simple formula was proposed that was found to give a reasonable estimate of ultimate strength of perforated plate with reinforcement under a variety of loading conditions (longitudinal/transverse compressive
load and edge shear load).
For a wave slap of FPSO, a methodology for practical structural assessment of offshore structures was
proposed by Moon et al. (2012). In this work the idealisation of impact loads was facilitated by the socalled pressure-impulse theory. Accordingly, time histories of impact pressures were generated along with
the pressure impulses predicted. Nonlinear transient structural analyses utilising the time series of impact
pressures led to the derivation of equivalent static pressure factors.
178
Hong et al. (2014b) studied the effects of steep wave-induced bow impact loads (so-called slapping
loads) on FPSOs using model tests. For a measurement of the pressure and impact force on the frontal
area, a bow-shaped panel was fabricated with the pressure and force sensors, and installed on the bow
starboard side of the model FPSO. It was concluded that the impact loads acting on the FPSO bow may
significantly increase with increasing steep waves (Hw/ > 1/16) and in this sense they should be considered in the evaluation of seakeeping performance and structural design assessment.
Zhang et al. (2012b) established a corrosion forecasting mathematical model for a 3000 m deep sea
semi-submersible platform. In their work the ultimate bearing capacities of a typical component and a
node were evaluated and the evolution laws of the ultimate bearing capacity under various failure modes
and various service years were summarised. It was concluded that as corrosion damage may have significant effect on ultimate strength, the corrosion damage factor should be taken into account in the design of
offshore platforms.
Nonlinear structural simulations for an aluminium helideck of a floating structure were investigated by
Koo et al. (2014). The aim of the study was to evaluate the buckling/ultimate strength of the developed
pancake under helicopter landing impact and to compare the structural safety and stability in accordance
with Eurocode 9 (CEN 2007). It was observed that the aluminium sandwich structure may be able to
withstand the external forces induced by helicopter landing. However, results of global yield and buckling strength assessment led to the conclusion that Eurocode 9 design assessment criteria are significantly
conservative.
Izadparast & Duggal (2013) studied the probability distribution of dynamic responses of turret-moored
FPSOs and evaluated the effects of nonlinearity on the response distribution and the extreme statistics.
For this purpose, sample data sets obtained from two experimental model tests studying the response of
typical external turret mooring systems designed for deepwater and shallow-water conditions were utilised. Special focus was attributed on the extreme statistics of the mooring leg tension and vessel horizontal offsets. Accordingly, the probability distributions of measured data were estimated using common
distribution models of linear and nonlinear random variables (Rayleigh, exponential, Stansberg, Weibull
and 3-parameter Rayleigh distributions).
Wang et al. (2014) proposed a computationally efficient methodology for the long-term extreme analysis of FPSO mooring systems by the use of Kriging metamodel (Simpson et al. 2001). The paper showed
that short-term extreme analysis over all sea states can be replaced by a long-term extreme analysis with a
Kriging metamodel. The later can improve the efficiency of the long-term extreme analysis significantly.
Seakeeping capability of a floating type offshore structure is directly connected with operational profit,
and the hull form is a dominant factor to the capability (e.g. column size, spacing and pontoon size) in the
case of semi-submersible type unit. Therefore, it is important to design optimal hull form which has the
best performance. In a study by Park & Jang (2013), a method for the hull form optimisation was presented. It consists of three modules: a panel generation module, a mass estimation module, and an analysis input generation module. In the first module, 10 design variables are defined to represent the hull form
such as the width, length, height of the pontoon. Once the values of those variables are determined, the
panel model for motion analysis is automatically generated. The second module, mass estimation module,
enables to estimate hull lightweight by surface and volume of the model. Finally, a linear hydrodynamic
motion analysis is performed to obtain response amplitude operator (RAO).
Construction practices of floating structures such as floating LNG terminals, container terminals, floating production storage offloading and floating breakwaters have increased worldwide. These types of
structures are exposed to severe ocean environment conditions such as wave, hydraulic pressure, and
impact loads. As a result, new types of floating structures with improved dynamic response and structural
performance compared to existing ones have been proposed. One example of such a structure was presented in Lee et al. (2013a) where a regular pontoon-type structure was used as a reference and further
developed to a hybrid-floating which has ballast compartments that includes separate space also for an
air-gap. In their paper, it was demonstrated in a case study of three floating structures (with the same
global dimensions) that the two analysed hybrid-floating structures had reduced motions compared to the
conventional pontoon structure. Further, the structural characteristics of the new concepts showed satisfying results with low stresses in the elastic regime.
A number of model tests and theoretical computational methods have been developed during the reporting period with the aim to evaluate the quasi-static response of offshore structures in those cases that
innovation impacts design choices. In the following paragraphs, some examples are presented. Qi et al.
(2014) presented the engineering analysis, computer simulation, model test validation, and mooring system optimisation for a Dry Tree Submersible (DTS). To verify the correctness of the numerical timedomain analysis results, DTS model test was done over a scale 1:60 in an ocean engineering basin. Heave
RAOs showed a significant benefit compared to traditional semi-submersible structure. In another investigation by Vant Veer & Vlasveld (2014), they studied the influence from green water on a twin-hull
179
FLNG concept. The green water problem was studied by model tests for a turret-moored midscale twinhull FLNG. Numerical simulations were carried out and compared against experiments. The numerical
results underestimated the freeboard exceedance. However, trends in the response over all sea states and
locations were well captured.
Kvaleid & Kvillum (2014) performed a nonlinear air-gap analysis of a semi-submersible and compared
the results with both linear analysis results and model test results. The representation of nonlinear platform motion was found to be vital for the accurate air-gap prediction. Results from the use of a 3D diffraction-radiation weakly nonlinear hydrodynamic approach resulted in a good agreement between the
model test and the nonlinear simulation air gap results in a severe sea state. It was concluded that sufficient modelling of hydrodynamic damping in severe sea states may be important alongside the consideration of viscous drag effects. Furthermore, in Cao & Wan (2014), a numerical study of the strongly nonlinear extreme wave interaction with a floating structure based on the open source package OpenFOAM was
presented. The numerical solver used in this study was based on the Navier-Stokes equations, and the
Volume of Fluid (VOF) method was employed to capture the complicated free surface. The results of
motion response of a floating body under extreme waves were much larger than that under regular waves.
Under extreme wave, the motion of floating body became large suddenly when the wave component focused at the location of floating body. The free surface showed strongly nonlinear characteristics. The
numerical results were found satisfactory when compared to experiments including motion responses and
green water phenomena.
Lu et al. (2012) presented a numerical time domain simulation model using a VOF technique to capture
the violent free-surface motion that may be used to study green water phenomena and its impact loading
on structures. The incompressible Euler/Navier-Stokes equations, written in an Arbitrary LagrangianEulerian (ALE) frame, were solved using projection schemes and an FE method on unstructured grids.
Numerical simulations of green water problems were carried out for green water overtopping a fixed 2D
deck, green water impact on a fixed 3D body without or with a vertical wall on the deck, and green water
impact on the deck and deckhouse of a moving FPSO model. The numerical results obtained using the
proposed models were compared with experimental measurements for each case. The agreement was
fairly good and it was concluded that proposed numerical model can be used for simulating green water
effects.
6.2
Types of analysis for various fixed offshore structures
Stress concentration factors for tubular joints were computed using solid quadratic elements and the results of the computations were compared with experiments reported in the literature and with expressions
reported in the literature and in design codes (Thibaux & Cooreman 2013). Kajolli et al. (2014) investigated the effect of loading sequence on fatigue life of a steel jacket. A new damage indicator based sequential law was proposed to estimate fatigue life of offshore steel structures. A verification of the new
damage model is conducted by comparing the theoretically predicted damage and fatigue life with experimentally observed damage and fatigue life respectively.
To model the joint flexibilities of the support structures of wind turbines, Tu & Vorpahl (2014) studied
the use of a superelement modelling approach which includes the idealisation of detailed joints. Their
results were compared to those obtained by a classic beam and shell idealisation for a generic 5 MW jacketed wind turbine. A static analysis including a reduced fatigue load case was conducted to pre-check the
different load bearing behaviour of modelling approaches. It was concluded that the mass difference was
more than 3% between the beam model and the shell model.
Probability distribution of extreme loads of offshore structures may be of great value for designs.
The major obstacle in establishing the probabilistic properties of response is due to the nonlinearities
of the wave load mechanism and/or the structural system which may lead to non-Gaussian distribution
for the response. The problem is further compounded by current and intermittent loading on members
in the splash zone, which have a significant effect on the statistical properties of extreme responses.
The Conventional (Monte Carlo) Time Simulation technique (CTS) is frequently used for predicting
the probability distribution of the extreme values of response. However, this technique suffers from
excessive sampling variability and hence a large number of simulated extreme responses (hundreds of
simulated response records) are required to reduce the sampling variability to acceptable levels. Three
different versions of a more Efficient Time Simulation technique (ETS) were compared by exposing a
test structure to sea states of different intensity (Abu Husain et al. 2013b). Further development of this
technique was made for more accurate estimates of the long-term probability distribution of the extreme response (Abu Husain et al. 2013a). Abu Husain et al. (2014) validated the ETS procedure
against results from CTS procedure by comparing probability distribution of extreme values of overturning moment from the two methods. Corresponding results for quasi-static responses were found to
be in very good agreement.
180
The suitability of the Generalised Extreme Value (GEV), and the Generalised Pareto (GP) distributions
for modelling of extreme responses, was presented in Wang et al. (2013b). They compared the GEV and
the GP distributions with empirical distributions derived from extensive Monte Carlo time simulations.
Mohd Zaki et al. (2013b) presented further development of the Finite-Memory Nonlinear System
(FMNS) method which is an identification technique to establish a relationship between the output and
input of some nonlinear systems. The computational effort using the FMNS method can be about 35 times
more efficient than a conventional time simulation procedure. In their study, the response of an offshore
structure exposed to Morisons wave loading was studied. The FMNS modelling technique was used to
model the drag-induced component of quasi-static response, the inertia-induced component of quasi-static
response, and the (total) quasi-static response of a case study offshore structure. The results from several
FMNS simulations were compared against conventional time simulation method with good agreement
between the two simulation methods.
Linear random wave theory is frequently used to simulate water particle kinematics at different
nodes of an offshore structure from a reference surface elevation record. However, it is well known
that linear random wave theory leads to water particle kinematics with exaggerated high-frequency
components in the vicinity of mean water level. To avoid this problem, Mohd Zaki et al. (2013a) suggested that empirical techniques such as Wheeler and vertical stretching methods may be used to provide a more realistic representation of the wave kinematics in the near surface zone. A fixed platform
in a water depth of 100 m was used as the case study. The structure was modelled as linear and its dynamic response was evaluated through mode superposition procedure. It was shown that the probability distributions of extreme responses based on the Wheeler and the vertical stretching methods can be
significantly different from each other, leading to uncertainty as to which method should be used in
design.
Baarholm et al. (2013) presented nonlinear time domain irregular wave simulations for the Kvitebjrn
jacket platform located in the North Sea. The aim was to quantify the Equivalent Dynamic Amplification
Factor (EDAF). It is the factor one has to multiply the q-probability quasi-static response with in order to
obtain an adequate estimate of the q-probability dynamic response. For each of the selected extreme sea
states, both quasi-static and dynamic response simulations were carried out for several wave realizations
using different seeds. Based on the quasi-static response and dynamic response, EDAFs were calculated
for different response measures in the jacket. These factors can subsequently be used in ultimate limit
state (ULS) and accidental limit state (ALS) analyses of the platform.
Due to increased estimates of extreme wave crest heights and subsidence of seabed, many fixed offshore structures experience problems due to negative air-gap effects. Scharnke et al. (2014) studied
wave-in-deck loads by model tests and simplified load models where the wave kinematics were estimated using the Stokes 5th order theory. Simplified models were found to underestimate the forces
measured in the model tests. Iwanowski et al. (2014) used an industrial CFD tool to study the wave-indeck loads obtained in model tests described by Scharnke et al. (2014). It was found that a careful
setup of CFD simulations could reproduce the measured wave-in-deck force, if the incoming wave was
close to regular/steady state excitations. For extreme events in irregular wave conditions, it was not
possible to reproduce the measured wave-in-deck loads when a Stokes 5th order theory was used for
the wave kinematics.
Lu et al. (2014) used a new wave maker based on New Wave theory to generate extreme waves at the
inlet boundary with the aim to analyse the wave-in-deck loading. Accurate prediction of water surface
elevation and wave impact force was reported using the proposed method. A unique advantage of the
wave generating strategy presented is that it can produce the required waves at a prescribed time and location, and thus reduce the total simulation time as compared to conventional CFD methods.
Abdussamie et al. (2014) studied the wave-in-deck force on the bottom plate of a rigidly mounted box
shaped structure that may be subject to unidirectional regular waves. An analytical momentum approach
recommended by classification societies and a VOF method were utilised to study the wave-in-deck load.
Numerical results were compared to results from model tests, and it was concluded that CFD techniques
can be used for the solution of wave-in-deck problems provided that a convergence study is done against
a representative tank experiment.
A Navier-Stokes code was employed in conjunction with the interface capturing level-set method for
the prediction of wave impact on a jack-up structure under hurricane wave conditions (Chen et al. 2013).
An overset grid system was employed to facilitate the simulation of complex flow around a generic jackup structure with 3 supporting legs and simplified topside equipment. Time-domain simulation of green
water and wave impact loads were performed for random 3D short-crested waves based on the directional
wave spectra of Hurricane Katrina.
Design loads from breaking wave impacts towards a GBS platform were studied by Oberlies et al.
(2014). Model tests of the structure in storm waves were executed to provide local wave impact load data
181
on the shaft of the GBS. An approach where probabilities of sea states and wave impacts were combined
into a joint probability distribution was used to derive yearly exceedance probabilities of wave impacts.
This method provided design values based on the probability of a particular response instead of the probability of a specific input (i.e., sea state return period).
An experimental and numerical evaluation of the impact response of ring-stiffened cylinders struck at
mid-bay with a mass having a rigid knife-edge indenter was studied by Cerik et al. (2014). The experiments aimed at reproducing a collision scenario involving offshore installations with supply vessels. Dynamic force-displacement curves and strain measurements were presented. The results were compared
with the nonlinear FE simulations performed using ABAQUS/Explicit software. It was observed that the
presence of ring-stiffeners has a significant effect on the resistance against denting and also permanent
deformation.
Wave breaking on structures induces a turbulent multi-phase flow and hence its characterisation represents a challenging task for both experimentalists and numerical modellers. Palemn-Arcos et al. (2014)
investigated the simulation of offshore wave-structure interaction using OpenFOAM software for a tension-leg platform with a scale ratio 1:168. The numerical results suggested that the vertical velocity cannot be ignored because it can exert a large upward vertical load to equipment and facilities close to the
frontal edge of the deck.
The ageing of offshore infrastructure presents a constant and growing challenge for operators. Ageing
is characterized by deterioration, change in operational conditions or accidental damages which, in the
severe operational environment offshore, can be significant. Nezamian & Altmann (2013) studied the
structural integrity of thirteen identified platforms under existing conditions as these platforms are either
nearing the end of their design life or have exceeded more than 50% of their design life. Information on
history, characteristic data, condition data and inspection results were collected to assess the current state
and to predict the future state of the facility for possible life extension. In-service integrity assessments,
pushover analyses, corrosion control and cathodic protection assessments and weight control reports were
completed to evaluate the integrity of these facilities for requalification to 2019 and life extension to
2030.
The existing knowledge on the structural integrity assessment of offshore platforms may benefit from
case studies on the life extension evaluations of aging structures. A case study for the structural integrity
assessment of an existing 8-legged aging drilling platform located in the Persian Gulf and now 42 years
old was proposed. The objective of the study has been to check whether the offshore structure in question
is fit for purpose for a life extension of 25 years beyond 2012 (Golpour et al. 2013).
Amdahl et al. (2012) and Amdahl & Storheim (2013) studied the scenario of broad side ship collision
with jacket legs in accordance with the NORSOK standards (NORSOK 2004). Accordingly, a jacket leg
and the shipside of a typical supply vessel were modelled, and various impact simulation scenarios were
carried out with LS-Dyna software. The resistance to denting of the jacket leg and the resistance to indentation of the ship side were compared against the NORSOK recommendations. The distribution of energy
dissipation and damage to the ship and the leg was studied for various leg thicknesses and two contact
positions. Travanca & Hao (2014) evaluated the consequences of bow impacts for different collision scenarios between ship and jacket structures. Two different vessels and two jacket designs (three and four
legs) were studied. The impact energy was in the range 59-74 MJ. Possible plastic deformation mechanisms were analysed and simplified approaches were considered for prediction in comparison with the
numerical results carried out by FE analysis.
Ice-induced vibrations caused by moving ice have been monitored on some of the platforms deployed
in north part of Bohai Sea of China. Zhang et al. (2012a) present field monitoring of ice-induced vibrations and full-scale measurements of ice force. The dynamic behaviour of ice-resistant jacket structure
and potential failure modes provoked by ice-induced vibrations were discussed. It was concluded that for
jacket structures, which are designed to satisfy criteria to resist a maximum ice force, the ice-induced
vibrations may increase the risk for e.g. fatigue failure. Moreover, Wang et al. (2012) presented a new
direct ice force measurement technology of jacket structures conducted in the Bohai Sea of China. In their
paper, a description of specially designed ice load panels to be mounted on any type of floating offshore
structure is presented followed by examples of measurements where the new ice force measurement panel
has been successfully used.
Oberlies et al. (2014) presented a methodology to estimate iceberg impact loading, as well as analyses
and design of exterior walls of the Hebron Gravity Based Structure to resist the 10,000-year return period
iceberg impact loading. The iceberg impact load on the Hebron Gravity Based Structure was calculated
using a probabilistic analysis including Type II uncertainty analysis with a logic tree. When subjected to
the 10,000-year iceberg impact, the Hebron Gravity Based Structure was designed to be highly utilised,
that is, the concrete and rebar were stressed to their specified strength. This was done by allowing internal
redistribution of elastic forces and by using nonlinear FE analyses.
182
Zhang et al. (2014b) investigated two vibration-based methods for detecting damage for tripod type
offshore wind turbine. These methods are the Modal Strain Energy (MSE) and the Modal Strain Energy
Decomposition (MSED). The results indicated MSE performs better for column member while MSED
does better for brace members. If the measured data is complete, damages can be localised easily in both
two damage locations.
Li et al. (2014) studied the impact between a wind turbine monopile and the transition piece during installation. The transition piece makes the link between the fixed monopile and the wind turbine tower.
Due to the random motion of the transition piece when installing from a floating vessel, vertical impacts
is a function of the limiting crane tip motion, environment condition, as well as the dynamic behaviour of
the installation vessel. The study gives indications about the maximum vertical impact velocity to be respected in such operations.
The optimisation of offshore wind mills structure becomes today an important field of investigation.
This task is generally done during the initial design stage, as for the ships structure. The technical solution chooses for the submerged part of wind turbines is mainly influenced by economic reasons. During
the last 3 years, it seems that the most economical solution is represented by the jack-up support types.
Schafhirt et al. (2014) demonstrated the feasibility of the genetic algorithms optimisation for the jack-up
support for wind turbines. A design and an analysis interface for the optimisation using genetic algorithms was developed. Using a simplified geometry and one load case, it was shown that the complexity
of the optimisation model was the same as for a realistic application.
6.3
Uncertainty, risk and reliability in offshore structural analysis
Sources of uncertainty for pile-founded fixed steel jacket platforms can be divided into three different
categories: (a) uncertainties associated with the soil-pile modelling parameters in clay soil, (b) the platform jacket structure modelling parameters, and (c) the uncertainties related to ground motion excitations
(El-Din & Kim 2013). The main uncertainty arises from the soil-pile interaction. El-Din & Kim (2013)
investigated the sensitivity of the seismic response parameters to the uncertain modelling variables of
pile-founded fixed steel jacket platforms using tornado diagrams, first-order and second-moment approaches, as well as static pushover analysis techniques.
Reliability study of axially loaded jacket piles in sand was presented by Ronold et al. (2012) aiming at
predicting their probability of failure in compression. A first-order reliability method (FORM) was used
for probabilistic modelling calculations. It was concluded that the new design methods need further validation against large scale test results in order to serve as consistent design methods. Alternatively, as a
minimum, they need to be accompanied with restrictions with respect to range of validity and extrapolation in use.
Khedmati et al. (2013) presented a concise reliability analysis of an offshore platform under fatigue
loading conditions. The fatigue analysis was based on the S-N approach considering fixed supports of the
piles and nonlinear simulation of soil reactions. A wide range of uncertainty parameters were included in
the fatigue reliability analysis, such as hydrodynamic coefficients, marine sediments, stress concentration
factors, stress intensity factors, and initial imperfections. The systematic procedure can be applied on
other types of offshore structures.
The effect of soil-structure interaction is also important issue in jack-up rig. Mirzadehniasar et al.
(2012) employed both the singular New Wave and multiple constrained New Waves to simulate random
sea states in order to investigate the nonlinear dynamic response and collapse mechanisms of a jack-up
platform subjected to extreme waves. Five different foundation types (pinned, fixed, and spudcans with
152 MN, 190 MN and 228 MN preloads) were investigated. The probability of collapse, the critical
areas dominating the collapse mechanisms, the maximum deck displacements and the variation of the
results for 100 randomly generated New Waves were discussed. It was also concluded that for assessment of the ultimate strength of the jack-up platform, neither pinned nor fixed supports can adequately
substitute for a more sophisticated model that tracks the load-displacement behaviour of the spudcan
foundations.
Dyanati & Huang (2014) developed an ultimate limit state function of a steel offshore platform based
on base shear capacity and demand, and calculated seismic reliability of the prototype structure against
collapse. A 3D FE model was used to calculate the values of limit state functions in terms of both capacity and demand.
A reliability-based methodology was presented in Li et al. (2013c) for an assessment of acceptance criteria of fixed offshore platforms in the northern South China Sea under extreme storm events. A structural
reliability method was proposed to quantify the probability of platform failure subjected to extreme
storms based on a long-term distribution of the extreme environmental load considering the joint occurrence of wave, current and wind.
ISSC committee II:1 QUASI-STATIC RESPONSE
183
There is research to mitigate computational burden of reliability analysis. Gholizad et al. (2012) investigated the fatigue reliability of fixed offshore platforms by analysing different failure scenarios. In order
to calculate the occurrence probability of each scenario, a massive reliability analysis should be done for
each of the corresponding sub-scenarios using an artificial transfer function instead of implementing time
consuming traditional methods. Based on the calculated values, the probability of occurrence was obtained for each scenario, and finally, the failure probability of the entire system was calculated.
Uncertainty coming from fabrication process was treated. Steel circular hollow sections have been extensively used in fixed offshore structures because of their high strength-to-weight ratio, non-directional
buckling and bending strength, and low wave resistance. In a tubular joint, circular hollow members are
connected by welding the prepared end of the brace members onto the surface of the chord member. The
tubular DKT-joint is the geometric configuration where the chord is welded with two outer braces and
one central brace. Lotfollahi-Yaghin & Ahmadi (2012) studied the results of FE analysis on the 81 steel
multi-planar tubular DKT-joints. These numerical results were used to determine the probability distribution of SCFs along the weld toe of the central brace under axial loads. The best fitted distribution was
determined and its parameters were calculated based on the maximum likelihood method. It was concluded that the proposed probabilistic density function can be used for fatigue reliability analysis of such
structures.
For a floating offshore structure, the random nature of the sea state is one of the main uncertainties in
loading. To more confidently simulate wave loads, all of the randomness of water surface should be taken
into account. Load history also plays an important role in the nonlinear dynamic response of structures.
Accordingly, an appropriate way to consider these effects is dynamic analysis of offshore platforms using
random time-domain generation of the sea surface over a long period of time. Chen et al. (2013) presented a rational reliability assessment procedure for hull girder ultimate strength assessment of shipshaped FPSOs. The hull girder ultimate strength of FPSOs was calculated by a rigorous progressive collapse analysis using Smiths method. The stochastic model of Still-Water Bending Moment (SWBM) was
established based on the loading conditions from the operation manual of FPSOs. A stochastic model for
the extreme value of Vertical Wave-Induced Bending Moment (VWBM) was proposed based on the
long-term distribution of VWBM and the extreme value theories. Hull girder reliability was measured by
a first-order reliability method.
Helmers et al. (2012) developed an efficient Monte Carlo method for the stochastic analysis of slamming loads on marine structures. The probability distribution of the maximum impact force during slamming was established for a given stationary sea state. The method was demonstrated by using a uniform
wedge and Wagners flat plate theory. When the observed data are limited, statistical estimates can be
used to supplement or even replace information based on the Bayesian approach. Vasconcelos de Farias
& Netto (2012) applied these concepts to the study of the structural integrity of the hulls of FPSO units.
Two case studies were conducted with a focus on corrosion damage due to its high incidence among the
observed damages.
Fire accidents have been recognised as a major hazard of offshore facilities in oil and gas industries
and many researches were focused on the consequence analysis to evaluate the severity of the accident.
Kim et al. (2012b) presented a practical procedure for the nonlinear structural consequence analysis of
structures under fire. The thermal and structural response analysis has been performed in this study using
a commercial nonlinear FE analysis code. The results of the structural response analysis were then compared to the experimental results. Jin & Jang (2013) proposed a new probabilistic fire risk assessment
procedure where structural cumulative failure frequency was calculated from CFD fire simulation and a
heat transfer analysis. Transient effect of hydrocarbon release was also properly reflected by applying an
effective method called snapshot. In Jin & Jang (2014) the new procedure of fire risk analysis was further discussed and demonstrated on a specific case.
As a conventional measure to mitigate structural failure under fire, Passive Fire Protection (PFP) is
widely used on main structural members. However, an excessive use of PFP can cause considerable cost
for material purchase, installation, inspection and maintenance and long installation time can be a risk
since the work should be done nearly at the last fabrication stage. Friebe & Jang (2013) presented a few
case studies on how different applications of PFP have influence on collapse time of an FPSO module
structure. A series of heat analysis and thermal elasto-plastic FE analysis were performed for different
PFP protections and the resultant collapse time and the amount of PFP were compared with each other.
Gas explosion has been categorized as an important issue of the design of offshore structures. Han et
al. (2012) presented practical considerations for the nonlinear dynamic structural analysis of offshore
structures under blast loadings from gas explosion accidents. Numerical investigations including modelling of blast loads and idealisation of structural materials and members have been conducted for the overall topside structures. As a design step for offshore structures under blast loadings, an applicable guidance
on the FE analysis was described in this study.
184
A practical procedure was presented for nonlinear structural response analysis of FPSO topside blast
wall under explosion loads (Sohn et al. 2012, 2013). Two methods were adopted for nonlinear structural
consequence analysis of FPSO topside blast walls. One has been computed based on the use of timedomain nonlinear FE analysis and the other was performed with single degree of freedom method based
on resistance function. The relationships between blast pressures versus impulse of FPSO topside blast
walls were developed.
Ahn et al. (2014) discussed methodology, conditions, and design consideration of dropped object
analysis on a panel/deck structure using dynamic FE analysis. Results from direct FE analyses were compared against those from the simplified energy method described in DNV (2010). Parameter sensitivity
analysis was carried out in order to study the influence from parameters and model uncertainty on energy
absorption in the panel/deck structure. It was concluded that results were influenced by the application of
failure criteria according to the rule requirements, application of material properties, dropping position,
condition of the object etc.
Ren et al. (2014) investigated the use of a Risk-Based Inspection (RBI) plan combined with a fatigue
failure analysis on a jacket platform located in the South China Sea. An RBI plan combines member failure consequence and probability and fatigue failure analysis in order to develop the final inspection plan.
The paper presented also the methodology to establish a practical and reliable inspection plan for such
offshore structures.
Valdman (2014) presented the results of risk assessment and a risk management analysis was performed as part of the environmental impact studies of offshore processing and transportation systems
designed for Arctic seas. It was noted that in spite of overall improvement in the safety levels of offshore
operations achieved over the last 10 to 20 years, the major accidents are still frequent and the safety barriers for efficient risk management and control remain to be an urgent issue.
7.
BENCHMARK STUDY
Design against impact loads (slamming) can be challenging, time consuming and involve complex calculations. Application of simplified, quasi-static calculation approaches will make the design process much
easier. The objective of this benchmark study was to evaluate the accuracy of such simplified approaches
within the context of quasi-static impact load analysis. During recent years, much attention has been directed towards the structural integrity of free fall lifeboats when they are subjected to impact loads. The
committee has received access to data, both structural drawings and data from drop tests (trajectory information, measured pressures, etc.); the committee acknowledges OLF, Schat-Harding, Statoil and
Marintek for the permission.
7.1
Methodology
The case study lifeboat is a Schat-Harding FF1000 model. The main particulars of the lifeboat are presented in Table 3. The hull is made of Chopped Strand Mat (CSM) laminate. The hull shape and structural drawings were provided by Schat-Harding. Model tests of the lifeboat were carried out by Marintek
using a 1:9 scale model, as illustrated in Figure 4. The skid angle was 50 degrees. Several force transducers for measuring of the pressure were installed in the model. Two of them were located on the bottom of
the lifeboat and they were used in the case study. The locations of these transducers are defined in Table 4
and shown in Figure 5. An FE model provided by DNV-GL is shown in Figure 6 and Table 5 presents the
material properties used in the FE model.
Table
3. Main particulars of case study lifeboat.
_____________________________________________________________________________
Parameter
Value
_____________________________________________________________________________
Overall length (Lpp)
12.57 m
Overall width (B)
3.34 m
Displacement (105%: light vessel + 64 persons 90 kg) 16.8 metric tonnes
Longitudinal centre of gravity (forward of stern)
5.29 m
Radius of gyration in pitch (% of LOA)
25%
_____________________________________________________________________________
Table
4. Definition of locations used in the benchmark study.
_______________________________________________________
Location
x (from stern) y
z
_______________________________________________________
S1
2.25 m
1.0 m
1.1 m
S2
8.20 m
1.0 m
1.1 m
_______________________________________________________
185
Figure 4. Model test at Marintek of a free fall lifeboat. (Left) Test setup and (right) body plan.
S1
S2
Figure 5. Location of the transducers S1 and S2 (see Table 4 for the exact locations).
Figure 6. Illustration of the full finite element model.

Table
5. Material properties.
________________________________________________________________________
Property
CSM
Structural foam
Buoyancy foam
________________________________________________________________________
Ex [MPa]
7500
42
5
7500
42
5
Ey [MPa]
5000
42
5
Ez [MPa]
0.32
0.3
0.4
xy [-]
0.32
0.3
0.4
xz = yz [-]
3000
Gxy [MPa]
Gxz = Gyz [MPa] 1500
1460
42
100
________________________________________________________________________
[kg/m3]
A high speed underwater video camera was used to record the underwater trajectory. It was used to estimate the relative velocity between the lifeboat bottom and the water surface. An example is shown in
Figure 7 from a test performed in irregular waves, Hs = 6.8 m, Tz = 6.8 s. This test was used as the basis
for the benchmark study.
186
Figure 7. Cross-sectional velocities for aft (upper thick solid, 0.3 m from stern) and fore (lower dashed,
12.3 m from stern) section. Solid lines and dashed/dotted lines represent the velocities relative to
the water (wave motion included) and to the air (earth-fixed coordinate system), respectively.
Zero velocity relative to water means that the section bottom is above sea surface.
7.2
Simplified methods
The space average pressure was found as p = 0.5CpV2, where V is the velocity taken from experimental
data and Cp is the pressure coefficient taken from DNV (2014a) as Cp = 2.5(tan())-1.1. This value represents the pressure over a broader area (i.e. several plate fields of a ship). The calculated pressure was
compared with the measured pressure presented in Figure 8. It was found that the calculated peak values
deviate significantly from the measured values. The main reason for these deviations may be due to the
influence of 3D effects, which can be prominent in the bow area (ABS 2011). Another factor that affects
the results may be the model test conditions. Model tests were performed in an irregular wave environment (Hs = 6.8 m, Tz = 6.8 s). This may complicate the calculation of impact pressure and can explain
some of the deviation.
Figure 8. (Left) Measured and (right) calculated pressures.
The maximum strain in the middle of the plate field was found by modelling a strip of the composite plate
as a beam, see Figure 9. The strain was calculated by two methods. The first was based on results from
experiments with stiffened panels. These experiments were reported by Faltinsen (1999) and were performed using aluminium and steel panels. The results of these experiments are summarised in Figure 10,
where the non-dimensional strain is plotted as a function of non-dimensional impact velocity.
187
Figure 9. Plate strip model of lifeboat bottom.
Figure 10. Maximum non-dimensional strain amplitude in the middle of a longitudinal stiffener (Faltinsen 1999).
For comparison, the response was also based on a pure static response combined with a dynamic amplification factor. The strain was found as = (Mt)/(2EI), where M = (pL2)/24. The dynamic amplification
factor was calculated from formulas in standard textbooks on structural dynamics (see e.g. Thomson
1972). The natural period of the plate strip was Tn = 1.3 s, whereas the duration of the pressure pulse was
t1 = 0.01 s. If it is assumed that the pressure pulse is triangular, the dynamic amplification factor is D =
t1/Tn = 0.0242. The hydrodynamic added mass can be estimated as A = 0.5R2. When it is included, the
natural period becomes 4.7 s and the dynamic amplification factor is reduced to 0.007.
Hydroelastic effects are important when tan() < 0.25V(L3/(EI))0.5 (DNV 2014a). For this case,
0.25V(L3/(EI))0.5 0.1, whereas tan(30) = 0.58, i.e. hydroelastic effects are not important. The results of
the simplified calculations are presented in Table 6 and Table 7. It is seen that when the effect of hydrodynamic added mass is included, the results from the simplified quasi-static method are quite close to the
results based on experimental data presented in Figure 10; see Heggelund et al. (2015) for additional
calculations and assessment of the results.
Table 6. Results based on experimental data.
_______________________________________________________________________________________________
Property
S1
S2
Comment
_______________________________________________________________________________________________
Velocity
17.6 m/s
12.8 m/s
From measurements (Figure 7)
Deadrise angle
30
40
Pressure
670 kPa
140 kPa
From measurements (Figure 8)
Pressure coefficient, Cp
4.2
1.7
Based on DNV (2014a)
Strain
1.07 s
0.540 s
Based on Figure 10
_______________________________________________________________________________________________
Table
7. Results from simplified quasi-static method.
_____________________________________________________________________________________________________
Property
Dry
Velocity
Deadrise angle
Pressure coefficient, Cp
17.6 m/s
30
4.6
Including hydrodynamic added mass
S1 (aft)
S2 (fore)
S1 (aft)
S2 (fore)
_____________________________________________________________________________________________________
12.8 m/s
40
3.0
17.6 m/s
30
4.6
12.8 m/s
40
3.0
188
Pressure
726 kPa
255 kPa
726 kPa
255 kPa
Dynamic amplification factor 0.02
0.02
0.007
0.007
Strain
3.57 s
1.28 s
0.97 s
0.35 s
_____________________________________________________________________________________________________
7.3
Quasi-static linear FE analysis
FE analysis was performed using the FE model shown in Figure 6. The pressure was applied on two strips
of the lifeboat as shown in Figure 11. Static pressure was assumed in the FE analysis. The impact pressure loads were represented by equivalent static pressures, which were applied separately on the specific
strips of S1 and S2. Inertia relief was applied as boundary condition. The analysis was performed as pure
quasi-static analysis, i.e. no dynamic factors were employed in the current results. Also 3D effects were
neglected in the analysis. This simplification was considered to be valid since the hull profiles of the investigated sections were rather constant.
The quasi-static analysis was performed by two parties in the committee using different analysis codes.
The results are presented in Table 8. It was found that stress/strain and deformation at these two points are
rather small, especially in comparison with the peak values in the respective sections. The results presented seem comparable with the results from the simplified quasi-static method (see Table 6) when hydrodynamic added mass is not included. The agreement is quite surprising since the dynamic amplification factor is very low and the effect of dynamics is not included in the FE analysis.
Table
8. Results from quasi-static FE analysis.
___________________________________________________________________________________________
SNU
Property
Chalmers
S1 (aft)
S2 (fore)
S1 (aft)
S2 (fore)
___________________________________________________________________________________________
Analysis code
ABAQUS/CAE 6.13-3
Patran/Nastran
Velocity
17.6 m/s
12.8 m/s
17.6 m/s
12.8 m/s
Pressure
726 kPa
255 kPa
726 kPa
255 kPa
Strain
3.92
s
1.49
s
3.42
s
1.45 s
___________________________________________________________________________________________
Figure 11. Pressure application to FE model. (Upper) Equivalent static pressure applied on
S1 and (lower) equivalent static pressure applied on S2.
7.4
Nonlinear, transient dynamic FE analysis
The analysis presented in Section 7.3 has been based on a linear, static analysis. A complete analysis need
to include nonlinear effects caused by large deformations (geometric nonlinearity). Also, due to the short
time duration, dynamic amplification should be included. An investigation was carried out based on a
transient dynamic analysis of a strip taken from the FE model. The position of the strip was at the fore
sensor and the width of the strip was 0.1 m. X-symmetric boundary conditions were imposed on both the
front and the rear sides in the longitudinal direction and y-symmetric at the transverse centreline. Additionally, the strip model was fixed at knuckles or corners where longitudinal structural members pass.
Thus, only the local plate response is included.
The model was loaded with uniform pressure on the bottom as shown in Figure 12. Different analyses
with different pulse durations, t, were also performed. The three shortest durations were comparable
189
with the duration from the experimental results (see Figure 8). For comparison with the quasi-static
analysis, a dynamic analysis with long pulse duration (9 seconds) was also performed. The analyses were
carried out using the commercial software LS-Dyna. All analyses included geometric nonlinearities, the
material behaviour is linear (purely elastic) and strain rate effects of the material were not included.
The results from the long duration analysis are shown in Figure 13. The strains at the upper and lower
surface as well as in the midplane of the plate are plotted. It is seen that the difference between these values is small. The maximum strain is 5.5 s and the membrane strain is 5.0 s. This means that the load is
carried mainly by membrane stress and that the effect of local bending is small. As the membrane stiffness dominates, the strain from the FE analysis is much less than the strain from the analytic plate strip
model (56 s).
X-strain (E-03)
Boundary conditions and loads on FE model

Pressure time histories
Figure 12. (Left) Strip model and (right) applied loads for nonlinear, transient dynamic analysis.
Element id
Sh-958 (lower)
Sh-958 (upper)
Sh-958 (mid)
Min=-1.0628e-05
Max=0.0053872
Time (s)
Figure 13. Response for long duration analysis (t = 9 s).
The same pressure was applied to the global FE model shown in Figure 11, but, the applied area was
extended in the longitudinal direction to be equivalent to the strip model. The resultant static strain
was 2.6 s, which is about half of the strip model. The reason seems to be that the longitudinal
stiffness of the plate between two neighbouring web sections is taken into account in the global FE
model.
The strain from the short pulse duration run (t = 0.05 s) is shown in Figure 14. Also here, the bending
stress is negligible. Results for all pulse durations are summarized in Table 9. It is clearly seen that the
190
stiffness of the structure is so large that the dynamic response is almost the same as for the static analysis,
i.e. the dynamic amplification factor is close to 1 and the response is quasi-static.
03
Element id
A Sh -1167492 (lower)
B Sh-1167492 (upper)
C Sh-1167492 (mid)
strain (E
Min= -0.00052951
Max=0.005436
Time (s)
Figure 14. Transient dynamic response for t = 0.05 s.

Table
9. Maximum strain from transient dynamic analysis with pmax = 273 kPa.
____________________________
t [s]
Strain [s]
____________________________
9
5.53
0.1
5.57
0.05
5.45
0.01
8.16
____________________________
7.5
Concluding remarks
Quasi-static prediction of structural response to impact loads was investigated based on basic mechanics,
experimental results for steel and aluminium panels and quasi-static FE analysis. Slamming loads on the
bottom of a free fall lifeboat during water entry was selected as the case study. It was found that results
from simple beam theory combined with basic formulas for dynamic response assessment was in good
agreement with the experimental results. It was found that hydroelastic effects were not important for this
case.
The peak pressure calculated with the parametric formulas in DNV (2014a) deviates from the measured
values. The main reason for this is believed to be due to 3D effects. The strain calculated by simple beam
theory combined with basic formulas for dynamic response assessment was in good agreement with experimental results for steel and aluminium panels as long as the hydrodynamic added mass was included.
Based on simple beam theory, it was found that the dynamic amplification factor was very small, approximately 0.01.
The strain calculated by nonlinear, transient dynamic FE analysis showed that the geometric nonlinearities were large so that the loads were carried mainly by membrane stress and that the effect of local
bending was small. The effect of plate bending was of secondary importance. The additional stiffness was
so large that the response was almost quasi-static, i.e. the dynamic amplification factor was close to 1.
The maximum strain was of the same order of magnitude as the results from the simplified calculations,
though.
ISSC committee II.1 QUASI-STATIC RESPONSE
191
Finally, linear FE analysis was also performed using inertia relief. The results were comparable to the
results from the methods discussed above. However, this was believed to be a coincidence since neither
the dynamic response or geometric nonlinearities were included.
8.
CONCLUSIONS AND RECOMMENDATIONS
The committee reviewed recent works concerning topics identified by the committee mandate. The report presents a summary of current publications that are relevant to quasi-static analysis methods applied
to ships and offshore structures. The summary presents a general introduction to strength assessment
approaches followed by descriptions of progresses of calculation procedures, uncertainties associated
with reliability-based quasi-static response assessment, ship structures, offshore structures, and a
benchmark study of slamming loads on the bottom of a free fall lifeboat during water entry. The following paragraphs highlight conclusions and observations from the literature review and the benchmark
study.
Research developments find their way more and more into industry standards and classification rules
(e.g. IACS CSR-BC&OT). It should be the aim of classification societies to explicitly provide the possibility for applying rational, first principle-based analysis in the design of ships and offshore structures.
The analysis methods utilised for ships and offshore structures vary depending on the design stages of
construction. This report provided an overview of these methods and the work being carried out to enhance these approaches. It touched on rapid first principles assessment methods for early stages, to design
stages where more detailed modelling and analysis techniques are required to support production. It also
addressed the evaluation of local and global behaviour utilising FE techniques. FE model methodology
and assessments have become an integral tool in design of marine structures. This tool is used to assess
various loading conditions encountered by ships and offshore structures, understand the impact from
structural degradation, assess wave loadings, evaluate ice loading conditions, and address damage due to
collisions. Further, for evaluation and validation of the calculated results, we will continue in the future
rather have to rely on thoughtful experiments, either on model or full-scale, as they observed the appearance of a highly complex dynamic character whose accuracy solutions are not only linked to the increasing advancement of technology (e.g. computer performance), but above all for thinking in detecting neglected and/or forgotten effects.
The topic of structural reliability analysis is of special interest for quasi-static response of ships
and offshore structures as this is related to uncertainties in quasi-static calculation models. The uncertainties in the environmental loads and structural resistance of marine structures can significantly
affect the structural performance and safety. It is important that we understand the extent to which
uncertainties associated with quasi-static response analysis influence design considerations and structural reliability. In the report, recent studies on uncertainties associated with the reliability-based
quasi-static assessment of ships and offshore structures were reviewed. The focuses were on the uncertainties of loads and load combination factors, probabilistic modelling of corrosion and its effects
on structural strength, reliability-based design and assessment method, and risk-based inspection,
maintenance and repair.
In the ship structures chapter, a review of developments in international rules and regulations was presented. A review of recent developments of special ship concepts such as container ships, LNG/LPG vessels, service vessels for wind mills and offshore platforms, passenger vessel and sailing yachts in relation
to quasi-static response was provided.
For an offshore structure, structural integrity is the most critical issue since it is operated at a fixed site
for its life time and a regular inspection in a dry dock is not feasible like a ship. There has been a lot of
research on wave impact loads under harsh condition such as bottom slamming, wave run-up, wave slap
on bow structure, and green water. For a simulation of the strongly nonlinear extreme wave interaction
with a floating structure, computational fluid dynamics (CFD) based on the Navier-Stokes equations and
the Volume of Fluid (VOF) method or 3D diffraction-radiation panel based codes have been employed.
Model tests are accompanied to verify those analyses. Structural integrity is assessed by a nonlinear transient structural analysis using the time series of impact pressures.
For an investigation of progressive collision damage characteristics for an offshore structure in collision with a supply vessel, numerical approaches based on dynamic and nonlinear FE analysis have been
proposed. The spectral approach in frequency domain has been widely used for a prediction of extreme
load for linear system. However, they are not valid for nonlinear environmental load such as Morison
force imposed on jacket structure. There have been many researches to treat the nonlinearity in efficient
way in place of Monte Carlo based time simulation technique such as efficient time simulation technique
and finite-memory nonlinear system.
192
Probabilistic approaches have been adopted to handle with uncertainties existing in the assessment of
fatigue strength or ultimate strength. For a fixed platform, the uncertainties mainly exist in soil-pile modelling parameters, marine sediments and seismic loadings. In a floating offshore structure, the uncertainty
arises from the structure itself such as material property, initial imperfections, residual welding stress,
etc., as well as the randomness of environmental load.
In the benchmark study presented in Chapter 7, quasi-static prediction of structural response to impact
loads was investigated based on basic mechanics, experimental results for steel and aluminium panels and
quasi-static FE analysis. Slamming loads on the bottom of a free fall lifeboat during water entry was selected as a case. It was found that results from simple beam theory combined with basic formulas for dynamic response assessment was in good agreement with experimental results for steel and aluminium
panels as long as the hydrodynamic added mass was included. It was found that hydroelastic effects are
not important for this case.
Future recommendations of topics for review are:
advance methods for mesh generation of FE models and new FE techniques,

advance methods to account for corrosion and fatigue in assessing structural strength,
uncertainties of internal loads and load effects on structural strength,
reliability-based life-cycle design,
risk-based inspection, maintenance and repair,
development of new rules and regulations by regulatory bodies, and
structural aspects of specialised ships and offshore structures.
REFERENCES
Abdussamie, N., Giles, T., Walid, A. & Ojeda, R. 2014. Wave-in-deck forces on fixed horizontal decks of offshore
platforms. In Proceedings of the 33rd International Conference on Ocean Offshore and Artic Engineering
(OMAE2014), San Francisco, CA, USA, 813 June 2014. (OMAE2014-23629).
ABS 2006. Guide for safehull-dynamic loading approach for vessels. American Bureau of Shipping, Houston, TX,
USA.
ABS 2010. ABS rules for building and classing mobile offshore drilling units, Part 3 - Hull construction and
equipment, January 2010. American Bureau of Shipping, Houston, TX, USA.
ABS 2011. Guide for slamming loads and strength assessment for vessels. American Bureau of Shipping, Houston,
TX, USA.
ABS 2014. Guide for safehull-dynamic loading approach for vessels. American Bureau of Shipping, Houston, TX,
USA.
Abu Husain, M.K., Mohd Zaki, N.I. & Najafian, G. 2013a. Efficient derivation of the probability distribution of the
extreme values of offshore structural response: comparison of three different methods. In Proceedings of the
32nd International Conference on Offshore Mechanics and Artic Engineering (OMAE2013), Nantes, France, 914 June 2013. (OMAE2013-10590).
Abu Husain, M.K., Mohd Zaki, N.I. & Najafian, G. 2014. Prediction of extreme values of offshore structural
response by an efficient time simulation technique. In Proceedings of the 33rd International Conference on
Ocean Offshore and Artic Engineering (OMAE2014), San Francisco, CA, USA, 813 June 2014.
(OMAE2014-23126).
Abu Husain, M.K., Mohd Zaki, N.I., Lambert, L., Wang, Y. & Najafian, G. 2013b. Long-term probability distribution of extreme offshore structural response via an efficient time simulation method. In Proceedings of the 32nd
International Conference on Offshore Mechanics and Artic Engineering (OMAE2013), Nantes, France, 9-14
June 2013. (OMAE2013-10602).
Ahn, K.H., Chun, M.S., Han, S.M., Jang, K.B. & Suh, Y.S. 2014. Dropped object analysis using non-linear dynamic FE analysis. In Proceedings of the 33rd International Conference on Ocean Offshore and Artic Engineering
(OMAE2014), San Francisco, CA, USA, 8-13 June 2014. (OMAE2014-23333).
Akpan, U.O., Koko, T.S., Yuen, B.K., Shahin, K.O. & Smith, M. 2012. Probabilistic reliability of ships with gross
damage. In Proceedings of the 6th International ASRANet Conference for Integrating Structural Analysis, Risk
and Reliability, London, UK, 24 July 2012.
Alfred Mohammed, E., Chan, H.S. & Hirdaris, S.E. 2012. Global wave load combinations by cross-spectral methods. Marine Structures 29(1): 131151.
ALPS/HULL 2006. A computer program for progressive collapse analysis of ship hulls (version 2006.3). Ship and
Offshore Structural Mechanics Laboratory, Pusan National University, Korea.
ALPS/ULSAP 2006. A computer program for ultimate limit state assessment for stiffened panels (version 2006.3).
Proteus Engineering, Stevensville, MD, USA.
Amante, D.A., Estefen, S.F. & Chujutalli, J.A. 2014. Residual compressive strength of dented FPSO side shell panel. In Proceedings of the 33rd International Conference on Ocean Offshore and Artic Engineering (OMAE2014),
San Francisco, CA, USA, 8-13 June 2014. (OMAE2014-24059).
193
Amdahl, J. & Storheim, M. 2013. Design of offshore installations against ship collision based on interaction
analysis. In C-S. Lee & S-H. Van (eds), Proceedings of the 12th International Symposium on Practical Design of Ships and Other Floating Structures (PRADS2013), Changwon, Korea, 20-25 October 2013. pp.
352358.
Amdahl, J., Watan, R., Hu, Z. & Holmas, T. 2012. Broad side ship collision with jacket legs: examination of NORSOK N-400 analysis procedure. In Proceedings of the 31st International Conference on Offshore Mechanics and
Artic Engineering (OMAE2012), Rio de Janeiro, Brazil, 16 July 2012. (OMAE2012-84266).
Andersen, I.M.V. & Jensen, J.J. 2013. Hull girder fatigue damage estimations of a large container vessel by spectral
analysis. In C-S. Lee & S-H. Van (eds), Proceedings of the 12th International Symposium on Practical Design of
Ships and Other Floating Structures (PRADS2013), Changwon, Korea, 20-25 October 2013. pp. 557565.
Augener, P.H. & Hatecke, H. 2014. Sea-keeping analysis of an offshore wind farm installation vessel during the
jack-up process. In Proceedings of the 33rd International Conference on Ocean Offshore and Artic Engineering
Augener, P.H. & Krger, S. 2014. Computation of drift forces for dynamic positioning within the very early
design stage of offshore wind farm installation vessels. In Proceedings of the 33rd International Conference
on Ocean Offshore and Artic Engineering (OMAE2014), San Francisco, CA, USA, 8-13 June 2014.
(OMAE2014-23074).
Avgoulas, E.I., Karatzas, V.A., Zilakos, I.K. & Tsouvalis, N.G. 2013. Numerical analysis of cracked marine structures repaired with composite patches. In C. Guedes Soares & J. Romanoff (eds), Analysis and Design of Marine
Structures; Proceedings of the 4th International Conference on Marine Structures (MARSTRUCT2013), Espoo,
Finland, 25-27 March 2013. London: CRC Press. pp. 367375.
Avi, E., Niemel, A., Lilleme, I. & Romanoff, J. 2013. Equivalent shell element for ship structural design. In C.
Guedes Soares & J. Romanoff (eds), Analysis and Design of Marine Structures; Proceedings of the 4th International Conference on Marine Structures (MARSTRUCT2013), Espoo, Finland, 2527 March 2013. London:
CRC Press. pp. 459468.
Baarholm, G.S., Johansen, A., Birknes, J. & Haver, S. 2013. Estimation of equivalent dynamic amplification factor
(EDAF) on a jacket structure. In Proceedings of the 32nd International Conference on Offshore Mechanics and
Artic Engineering (OMAE2013), Nantes, France, 9-14 June 2013. (OMAE2013-10085).
Babuska, I., Szano, B.A. & Katz, I.N. 1981. The p-version of the finite element method. SIAM Journal of Numerical Analysis 18(3): 515545.
Baere, K.D., Verstraelen, H., Rigo, P., Passel, S.V., Lenaerts, S. & Potters, G. 2013. Reducing the cost of ballast
tank corrosion: an economic modeling approach. Marine Structures 32(1): 136152.
Bai, Y. & Qian, M. 2013. Reliability-based LRFD criteria for unstiffened panel of ship structures. In Proceedings of
the 32nd International Conference on Offshore Mechanics and Artic Engineering (OMAE2013), Nantes, France,
9-14 June 2013. (OMAE2013-11116).
Barltrop, N., Xu, L. & Hifi, N. 2012. The RISPECT inspection planning tool. In Proceedings of the 6th International ASRANet Conference for Integrating Structural Analysis, Risk and Reliability, London, UK, 2-4 July 2012.
Bang, J.K., Lee, C.H., Jeong, T.S. & Heo, J.H. 2013. Development of a new IMO Type B LNG cargo tank. In CS. Lee & S-H. Van (eds), Proceedings of the 12th International Symposium on Practical Design of Ships and
Other Floating Structures (PRADS2013), Changwon, Korea, 20-25 October 2013. pp. 416421.
Barhoumi, M. & Storhaug, G. 2013. Assessment of whipping and springing on a large container vessel. In C-S. Lee
& S-H. Van (eds), Proceedings of the 12th International Symposium on Practical Design of Ships and Other
Floating Structures (PRADS2013), Changwon, Korea, 20-25 October 2013. pp. 179-188.
Bashir, M.B., Tao, L., Atlar, M. & Dow, R.S. 2013. Experimental and numerical investigation of the wave-induced
loads on a deep-V catamaran in regular waves. In Proceedings of the 32nd International Conference on Offshore
Mechanics and Artic Engineering (OMAE2013), Nantes, France, 9-14 June 2013. (OMAE2013-10213).
Beck, F., Fleissner, F. & Eberhard, P. 2013. Simulation of a pendulum with fluid interaction and experimental validation. In B. Brinkmann & P. Wriggers (eds), Proceedings of the 5th International Conference on Computational
Methods in Marine Engineering (MARINE2013), Hamburg, Germany, 21-23 May 2013. pp. 479490.
Begovic, E., Mortola, G., Incecik, A. & Day, A.H. 2013. Experimental assessment of intact and damaged ship motions in head, beam and quartering seas. Ocean Engineering 72(1): 209226.
Beznea, E.F. & Chirica, I. 2011a. Buckling analysis of composite delaminated ship plates under shearing. In C.
Guedes Soares & W. Fricke (eds), Advances in Marine Structures; Proceedings of the 3rd International Conference on Marine Structures (MARSTRUCT2011), Hamburg, Germany, 28-30 March 2011. London: CRC Press.
pp. 8592.
Beznea, E.F. & Chirica, I. 2011b. Shear buckling analysis of composite plates with cut-outs. In C. Guedes Soares &
W. Fricke (eds), Advances in Marine Structures; Proceedings of the 3rd International Conference on Marine
Structures (MARSTRUCT2011), Hamburg, Germany, 28-30 March 2011. London: CRC Press. pp. 9398.
Bigot, F., Derbanne, Q. & Baudin, E. 2013. A review of strains to internal loads conversion methods. In C-S. Lee &
S-H. Van (eds), Proceedings of the 12th International Symposium on Practical Design of Ships and Other Floating Structures (PRADS2013), Changwon, Korea, 20-25 October 2013. pp. 259266.
Biot, M., Mantini, N. & Moro, L. 2013. Thermal and strength analysis of LNG tanks and their supporting structures
in the early design stage. In C. Guedes Soares & J. Romanoff (eds), Analysis and Design of Marine Structures;
Proceedings of the 4th International Conference on Marine Structures (MARSTRUCT2013), Espoo, Finland, 2527 March 2013. London: CRC Press. pp. 469476.
194
Blake, J.I.R., Shenoi, R.A., Das, O. & Yang, N. 2009. The application of reliability methods in the design of stiffened FRP composite panels for marine vessels. Ships and Offshore Structures 4(3): 287297.
Boote, D., Pais, T. & Dellepiane, S. 2013. Vibration of superyacht structures: comfort rules and predictive calculations. In C. Guedes Soares & J. Romanoff (eds), Analysis and Design of Marine Structures; Proceedings of the
4th International Conference on Marine Structures (MARSTRUCT2013), Espoo, Finland, 25-27 March 2013.
London: CRC Press. pp. 3744.
Branner, K., Toft, H.S., Haselbach, P., Natarajan, A. & Srensen, J.D. 2013. Reliability assessment of fatigue critical welded details in wind turbine jacket support structures. In Proceedings of the 32nd International Conference
on Offshore Mechanics and Artic Engineering (OMAE2012), Nantes, France, 914 June 2013. (OMAE201311398).
Brindley, S.M., Matsumoto, T. & Sasaki, S. 2013. A practical method for the fatigue strength assessment of
container carriers. In C-S. Lee & S-H. Van (eds), Proceedings of the 12th International Symposium on Practical
Design of Ships and Other Floating Structures (PRADS2013), Changwon, Korea, 20-25 October 2013. pp.
376383.
Buldgen, L., Le Sourne, H. & Rigo, P. 2013. A new super-element for estimating the collision resistance of an inclined ship side. In C. Guedes Soares & J. Romanoff (eds), Analysis and Design of Marine Structures; Proceedings of the 4th International Conference on Marine Structures (MARSTRUCT2013), Espoo, Finland, 25-27
March 2013. London: CRC Press. pp. 137145.
BV 2000. MARS2000 Users guide. Bureau Veritas, Neuilly-sur-Seine, France.
BV 2006. Guidelines for structural analysis of container carrier - DT1-Hull Section. Bureau Veritas, Neuilly-surSeine, France.
BV 2010. HOMER Users manual. Bureau Veritas, Neuilly-sur-Seine, France.
BV 2014. Rules for the classification of steel ships. Report NR467, July 2014. Bureau Veritas, Neuilly-sur-Seine,
France.
Cao, H. & Wan, D. 2014. Numerical study of motion response of floating body in extreme wave. In Proceedings of
the 24th International Ocean and Polar Engineering Conference (ISOPE2014), Busan, Korea, 15-20 June 2014.
pp. 599606.
CEN 2007. Eurocode 9: Design of aluminium structures Part 1-1: General structural rules. EN 1999-1-1:2007.
European Committee for Standardization, Brussels, Belgium.
Cerik, B.C., Shin H.K. & Cho, S.R. 2014. Experimental and numerical investigations on the impact response of
ring-stiffened cylindrical shells. In Proceedings of the 24th International Ocean and Polar Engineering Conference (ISOPE2014), Busan, Korea, 15-20 June 2014. pp. 932939.
Cerkovnik, M., Saldana, D. & Yang, T. 2013. Determining fitness for service of SCRs and flowlines with probabilistic tools. In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering
(OMAE2013), Nantes, France, 9-14 June 2013. (OMAE2013-11019).
Chen, N.Z., Wang, G. & Bond, J. 2013. Hull girder reliability assessment of FPSO. In C. Guedes Soares &
J. Romanoff (eds), Analysis and Design of Marine Structures; Proceedings of the 4th International Conference
on Marine Structures (MARSTRUCT2013), Espoo, Finland, 25-27 March 2013. London: CRC Press. pp.
535543.
Chun, M.S., Yoon, J.W., Suh, Y.S., Rodda, K.K., Han, S.M. & Kim, S.C. 2013. A structural capacity evaluation for
corrugated membrane structure of LNG cargo containment system. In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering (OMAE2013), Nantes, France, 9-14 June 2013.
(OMAE2013-10493).
Dec, A. & Frangopol, D.M. 2013. Risk-informed optimal routing of ships considering different damage scenarios
and operational conditions. Reliability Engineering and System Safety 119(1): 126140.
Dec, A., Frangopol, D.M. & Zhu, B. 2012. Reliability and redundancy assessment of ships under different operational conditions. Engineering Structures 42(1): 457471.
DNV 2002. Buckling strength of plated structures. Recommended practice DNV-RP-C201. Det Norske Veritas,
Hvik, Norway.
DNV 2003. Classification Notes No. 31.6: Strength analysis of hull structure in container ships. Det Norske Veritas,
Hvik, Norway.
DNV 2005. Rules for classification of ships Hull structural design ship with length 100 meters and above. Det
Norske Veritas, Hvik, Norway.
DNV 2008. Rules for classification of ships Liquefied gas carriers (Part 5, Chapter 5). Det Norske Veritas, Hvik,
Norway.
DNV 2009. Rules for classification of ships Hull structural design ship with length 100 meters and above (Part 3,
Chapter 1). Det Norske Veritas, Hvik, Norway.
DNV 2010. Design against accidental loads. Recommended practice DNV-RP-C204. Det Norske Veritas, Hvik,
Norway.
DNV 2011a. Design of offshore structures: General (LRFS methods). Det Norske Veritas, Hvik, Norway.
DNV 2011b. Design of offshore wind turbine structures. Offshore standard DNV-OS-J101. Det Norske Veritas,
Hvik, Norway.
DNV 2013a. Classification Notes No. 34.1 (January 2013), CSA: Direct analysis of ship structures. Det Norske
Veritas, Hvik, Norway.
DNV 2013b. Fabrication and testing of offshore structures. Offshore standard DNV-OS-401. Det Norske Veritas,
Hvik, Norway.
195
DNV 2014a. Environmental conditions and environmental loads. Recommended Practice DNV-RP-C205. Det
Norske Veritas, Hvik, Norway.
DNV 2014b. Rules for ships, hull structural design ships with length 100 metres and above (January 2014, Part 3,
Ch 1, Sec 12). Det Norske Veritas, Hvik, Norway.
DNV-GL 2015. Sesam software - https://www.dnvgl.com/software/index.html. [Accessed: 2015-04-19].
dos Santos Rizzo, N.A., do Amaral Amante, D. & Estefen, S.F. 2014. Ultimate shear strength of stiffened panels for
offshore structures. In Proceedings of the 33rd International Conference on Ocean Offshore and Artic Engineering (OMAE2014), San Francisco, CA, USA, 8-13 June 2014. (OMAE2014-23155).
Doshi, K. & Vhanmane, S. 2013. Probabilistic fracture mechanics based fatigue evaluation of ship structural details.
Ocean Engineering 61(1): 26-38.
Dyanati, M. & Huang, Q. 2014. Seismic reliability of a fixed offshore platform against collapse. In Proceedings of
the 33rd International Conference on Ocean Offshore and Artic Engineering (OMAE2014), San Francisco, CA,
USA, 8-13 June 2014. (OMAE2014-24137).
Ea, L. & Hoekstra, M. 2013. Verification and validation for marine applications of CFD. International Shipbuilding Progress 60(1): 107141.
Ehlers, S. 2009. Material relation to assess the crashworthiness of ship structures. Doctoral Dissertation, Helsinki
University of Technology, Espoo, Finland.
Ehlers, S. 2010. The influence of the material relation on the accuracy of collision simulations. Marine Structures
23(1): 462474.
Ehlers, S. 2013. A particle swarm optimization-based procedure to obtain a crashworthy ice-classed LNG tanker. In
J. Amdahl, S. Ehlers & B.J. Leira (eds), Proceedings of the 6th International Conference on Collision and
Grounding of Ship and Offshore Structures (ICCGS2013), Trondheim, Norway, 17-19 June 2013. London: CRC
Press. pp. 233240.
Ehlers, S. & Tabri, K. 2012. A combined numerical and semi-analytical collision damage assessment procedure.
Marine Structures 28(1): 101119.
Ehlers, S., Benson, S. and Misirlis, K. 2013. Ultimate strength of an intact and damaged LNG vessel subjected to
sub-zero temperature. In J. Amdahl, S. Ehlers & B.J. Leira (eds), Proceedings of the 6th International Conference
on Collision and Grounding of Ship and Offshore Structures (ICCGS2013), Trondheim, Norway, 17-19 June
2013. London: CRC Press. pp. 289296.
El-Din, M.N. & Kim, J.K. 2013. Sensitivity analysis of pile-founded fixed steel jacket platforms. In Proceedings of
9-14 June 2013. (OMAE2013-10029).
Faber, M.H., Straub, D., Heredia-Zavoni, E. & Montes-Iturrizaga, R. 2012. Risk assessment for structural
design criteria of FPSO systems. Part I: Generic models and acceptance criteria. Marine Structures 28(1): 120
133.
Faltinsen, O.M. 1999. Water entry of a wedge by hydroelastic orthotropic plate theory. Journal of Ship Research
43(3): 180193.
Farias, B.V. & Netto, T.A. 2012. FPSO hull structural integrity evaluation via Bayesian updating of inspection data.
Ocean Engineering 56(1): 1019.
FSICR 2010. Finnish-Swedish Ice Class Rules 2010, TRAFI/31298/03.04.01.00/2010. Transport Safety Agency,
Helsinki, Finland.
Fischer, C. & Fricke, W. 2013. Realistic fatigue life prediction of weld toe and weld root failure in load carrying
cruciform joints by crack propagation analysis. In C. Guedes Soares & J. Romanoff (eds), Analysis and Design
of Marine Structures; Proceedings of the 4th International Conference on Marine Structures (MARSTRUCT2013), Espoo, Finland, 25-27 March 2013. London: CRC Press. pp. 241248.
Fischer, C., Dster, A. & Fricke, W. 2011. Different finite element refinement strategies for the computation of the
strain energy density in a welded joint. In C. Guedes Soares & W. Fricke (eds), Advances in Marine Structures;
Proceedings of the 3rd International Conference on Marine Structures (MARSTRUCT2011), Hamburg,
Germany, 28-30 March 2011. London: CRC Press. pp. 289294.
Fossati, F. & Muggiasca, S. 2012. Motions of a sailing yacht in large waves: an opening simple instationary modelling approach. In Proceedings of the 22nd International Symposium on Yacht Design and Yacht Construction,
Amsterdam, Netherlands, 12-13 November 2012.
whipping stresses. In C-S. Lee & S-H. Van (eds), Proceedings of the 12th International Symposium on Practical
465470.
Fricke, W., Gerlach, B. & Guiard, M. 2014. Effect of window openings on the stiffness of walls and bulkhead in
ships. In J.K. Paik (ed.), Proceedings of the 7th International Conference on Thin-Walled Structures
(ICTWS2014), Busan, Korea, 28 September - 2 October 2014. (ICTWS2014-0104).
Friebe, M. & Jang, B.S. 2013. Study on the application of passive fire protection on topside structures. In C-S.
Lee & S-H. Van (eds), Proceedings of the 12th International Symposium on Practical Design of Ships and Other
Floating Structures (PRADS2013), Changwon, Korea, 20-25 October 2013. pp. 785792.
Frohbs, G.S. & Lampe, J. 2013. Risk based inspection for aged offshore pipelines. In Proceedings of the 32nd
International Conference on Offshore Mechanics and Artic Engineering (OMAE2013), Nantes, France, 9-14
June 2013. (OMAE2013-11119).
196
Fukasawa, T. & Mukai, K. 2013. On the effects of hull-girder vibration upon fatigue strength of a post-Panamax
container ship disaggregated by short-term sea state. In C-S. Lee & S-H. Van (eds), Proceedings of the 12th
International Symposium on Practical Design of Ships and Other Floating Structures (PRADS2013), Changwon,
Korea, 20-25 October 2013. pp. 479485.
Gannon, L. 2013. Simulation of the weld overlay procedure for corrosion repair of pressure vessels. In C. Guedes
Soares & J. Romanoff (eds), Analysis and Design of Marine Structures; Proceedings of the 4th International
Conference on Marine Structures (MARSTRUCT2013), Espoo, Finland, 25-27 March 2013. London: CRC
Press. pp. 423431.
Gao, Y. & Low, Y.M. 2014. Long-term fatigue analysis of deepwater risers in the time domain. In Proceedings of
the 33rd International Conference on Ocean Offshore and Artic Engineering (OMAE2014), San Francisco, CA,
USA, 8-13 June 2014. (OMAE2014-24664).
Garbatov, Y. & Guedes Soares, C. 2011. Reliability based maintenance of marine structures. In C. Guedes Soares,
Y. Garbatov, N. Fonseca & A.P. Teixeira (eds), Marine Technology and Engineering. London: CRC Press. pp.
11011120.
Garr, L. & Rizzuto, E. 2012. Bayesian networks for probabilistic modelling of still water bending moment for sidedamaged tankers. Ships and Offshore Structures 7(3): 269283.
Gaspar, B. & Guedes Soares, C. 2013. Hull girder reliability using a Monte Carlo based simulation method. Probabilistic Engineering Mechanics 31(1): 6675.
Gaspar, B., Teixeira, A.P., Guedes Soares, C. & Wang, G. 2011. Assessment of IACS-CSR implicit safety levels
for buckling strength of stiffened panels for double hull tankers. Marine Structures 24(1): 478502.
Ghanem, R.G. & Spanos, P.D. 1991. A stochastic finite element: a spectral approach (Rev. Ed). New York: Springer Verlag.
Gholizad, A., Golafshani, A.A. & Akrami, V. 2012. Structural reliability of offshore platforms considering fatigue
damage and different failure scenarios. Ocean Engineering 46(1): 18.
Giuglea, S., Chirica, A., Chirica, I. & Beznea, E.F. 2011. Fatigue estimation of a ship structural detail. In C. Guedes
Soares & W. Fricke (eds), Advances in Marine Structures; Proceedings of the 3rd International Conference on
Marine Structures (MARSTRUCT2011), Hamburg, Germany, 28-30 March 2011. London: CRC Press. pp. 303
309.
GL 2007. Global strength analysis. Germanischer Lloyd, Hamburg, Germany.
Gollwitzer, S., Abdo, T. & Rackwitz, R. 1988. FORM-program manual. RCP GmbH, Munich, Germany.
Golpour, H., Zeinoddini, M., Khalili, H., Golbaz, A.Y.Y., Adl, M.R., Ranjbar, P., Saberi, R., Thibaux, P.H.M.N. &
Cooreman, S. 2013. Structural integrity assessment of aging fixed steel offshore jacket platforms: a Persian Gulf
case study. In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering
Goyet, J., Boutillier, V. & Rouhan, A. 2013. Risk based inspection for offshore structures. Ships and Offshore
Structures 8(3-4): 303318.
Gud, Z., Bigourdan, B., Rouhan, A., Goyet, J. & Renard, P. 2013. Framework for risk-based inspection of an
offshore wind farm. In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering (OMAE2013), Nantes, France, 9-14 June 2013. (OMAE2013-10799).
Guo, J., Wang, G., Perakis, A.N. & Ivanov, L. 2012. A study on reliability-based inspection planning application
to deck plate thickness measurement of aging tankers. Marine Structures 25(1): 85106.
Han, D.S., Kim, G.S., Sim, W.S., Jang, Y.S. & Shin, H.S. 2012. Practical considerations for the structural analysis
of offshore topside structures under gas explosion accidents. In Proceedings of the 31st International Conference
on Offshore Mechanics and Artic Engineering (OMAE2012), Rio de Janeiro, Brazil, 1-6 July 2012.
(OMAE2012-83667).
Heggelund, S.E., Li, Z., Jang, B.S. & Ringsberg, J.W. 2015. Quasi-static assessment of response to slamming impact of free fall lifeboats. In Proceedings of the 34th International Conference on Ocean, Offshore and Artic Engineering (OMAE2015), St. Johns, Canada, 31 May-5 June 2015. (OMAE2015-41810).
Helmers, J.B., Sun, H., Landet, T. & Driveklepp, T. 2012. Stochastic analysis of impact loads on marine structures.
In Proceedings of the 31st International Conference on Offshore Mechanics and Artic Engineering
(OMAE2012), Rio de Janeiro, Brazil, 1-6 July 2012. (OMAE2012-83849).
Heredia-Zavoni, E., Montes-Iturrizaga, R., Faber, M.H. & Straub, D. 2012. Risk assessment for structural design
criteria of FPSO systems. Part II: consequence models and applications to determination of target reliabilities.
Heyer, A., DSouza, F., Morales, C.F.L., Ferrari, G., Mol, J.M.C. & De Wit, J.H.W. 2013. Ship ballast tanks - a
review from microbial corrosion and electrochemical point of view. Ocean Engineering 70(1): 188200.
Hifi, N. & Barltrop, N. 2014. Risk-based inspection system for collection and investigating of structural data to
improve design and inspection planning. In Proceedings of Transport Research Arena 2014, Paris, France, 14-17
April 2014.
Hirdaris, S.E., White, N.J., Angoshtari, N., Johnson, M., Lee, Y. & Bakkers, N. 2010. Wave loads and flexible
fluid-structure interactions: current developments and future directions. Ships and Offshore Structures 5(4): 307
325.
Hirdaris, S.E., Bai, W., Dessi, D., Ergin, A., Gu, X., Hermundstad, O.A., Huijsmans, R., Iijima, K., Nielsen, U.D.,
Parunov, J., Fonseca, N., Papanikolaou, A., Argyriadis, K. & Incecik, A. 2014a. Loads for use in the design of
ships and offshore structures. Ocean Engineering 78(1): 131174.
197
Hirdaris, S.E. 2014b. Special issue on uncertainty modelling for ships and offshore structures. Ocean Engineering
86(1): 12.
Hogstrm, P. & Ringsberg, J.W. 2012. An extensive study of a ships survivability after collision a
parameter study of material characteristics, non-linear FEA and damage stability analyses. Marine Structures
27(1): 128.
Hogstrm, P. & Ringsberg, J.W. 2013. Assessment of the crashworthiness of a selection of innovative ship structures. Ocean Engineering 59(1): 5872.
Hong, L. & Amdahl, J. 2012. Rapid assessment of ship grounding over large contact surfaces. Ships and Offshore
Structures 7(1): 519.
Hong, S.Y., Kim, K.H., Kim, B.W. & Kim, Y.S. 2014a. Experimental study on the bow-flare slamming of a 10,000
TEU containership. In Proceedings of the 24th International Ocean and Polar Engineering Conference
(ISOPE2014), Busan, Korea, 15-20 June 2014. pp. 816823.
Hong, S.K., Lew, J.M., Jung, D.W., Kim, H.T., Lee, D.Y. & Seo, J.S. 2014b. A study on the impact load acting on
an FPSO bow by steep waves. In Proceedings of the 33rd International Conference on Ocean Offshore and Artic
Engineering (OMAE2014), San Francisco, CA, USA, 8-13 June 2014. (OMAE2014-23833).
Htun, M.M. & Kawamura, Y. 2013. A study on the method to estimate the stochastic property of the strength of
corroded plate by response surface method with polynomial chaos expansion. In C-S. Lee & S-H. Van (eds),
Proceedings of the 12th International Symposium on Practical Design of Ships and Other Floating Structures
(PRADS2013), Changwon, Korea, 20-25 October 2013. pp. 731738.
Htun, M.M., Kawamura Y. & Ajiki, M. 2013. A study on random field model for representation of corroded surface. In C. Guedes Soares & J. Romanoff (eds), Analysis and Design of Marine Structures; Proceedings of the
Hu, A.K., Lin, Y. & Jiang, W. 2013. The wind tunnel experiment study of wind load on jack-up drilling unit. In
Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering (OMAE2013),
Nantes, France, 9-14 June 2013. (OMAE2013-10194).
Huang, W.B. & Xiao, Y. 2013. Monte Carlo prediction of the extreme values of the combined load effect
and semi-probabilistic design of ocean-going ships. In Proceedings of the 32nd International Conference on
Offshore Mechanics and Artic Engineering (OMAE2013), Nantes, France, 9-14 June 2013. (OMAE201310787).
Hussein, A.W. & Guedes Soares, C. 2011. Reliability assessment of intact and damaged bulk carriers. In C. Guedes
690.
Hwang, S.Y., Lee, J.H., Nho, I.S., Kim, S.C. & Kim, Y. 2013a. Sloshing assessment procedure using structural
responses for membrane type LNG CCS. In C-S. Lee & S-H. Van (eds), Proceedings of the 12th International
Symposium on Practical Design of Ships and Other Floating Structures (PRADS2013), Changwon, Korea, 20-25
October 2013. pp. 912918.
Hwang, S.Y., Lee, J.H., Kim, S.C., Nho, I.S., Jang, B.S. & Kim, Y. 2013b. Structural response analysis of LNG
CCS experiencing the sloshing impact determined by both convolution and fluid Structure interaction method. In
Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering (OMAE2013),
IACS 2001. Unified Recommendation R.34. International Association of Classification Societies, London,
UK.
IACS 2006. Common Structural Rules for double hull oil tankers. International Association of Classification Societies, London, UK.
IACS 2011. Requirements concerning Polar Class. Structural requirements for Polar Class ships IACS Req.
2006/Rev. 2. International Association of Classification Societies, London, UK.
IACS 2012a. Common Structural Rules for bulk carriers, consolidated effective as of 1 July 2012 with Corrigendas
approved on 16 July 2012 and 21 November 2012. International Association of Classification Societies, London,
UK.
IACS 2012b. Common Structural Rules for oil tankers, July 2012. International Association of Classification Societies, London, UK.
IACS 2013a. Draft Harmonised Common Structural Rules for industry review. International Association of Classification Societies, London, UK.
IACS 2013b. Technical background documents for draft Harmonised Common Structural Rules. International Association of Classification Societies, London, UK.
IACS 2013c. Equivalent design wave definition of extreme loads, technical background documents for draft Harmonised Common Structural Rules. International Association of Classification Societies, London, UK.
IACS 2014a. Common Structural Rules for Bulk Carriers and Oil Tankers. International Association of Classification Societies, London, UK.
IACS 2014b. Consequence Assessment (CA) summary report for Bulk Carriers and Oil Tankers. International Association of Classification Societies, London, UK.
IACS 2014c. Consequence Assessment (CA) individual reports for Bulk Carriers and Oil Tankers. International
Association of Classification Societies, London, UK.
198
IACS 2015. Common Structural Rules for Bulk Carriers and Oil Tankers. International Association of Classification Societies, London, UK.
Ibekwe, A.U., Pu, Y.C. & Dow, R.S. 2011. Reliability of cracked structures by an interactive finite element-based
method. In Proceedings of the 21st International Offshore and Polar Engineering Conference (ISOPE2011),
Maui, Hawaii, USA, 19-24 June 2011. pp. 741747.
Ibekwe, A.U., Pu, Y.C., Ham, W.L. & Dow, R.S. 2013. Progressive collapse analysis and reliability of a damaged
frigate. In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering
IEC 2005. Wind turbines, Part1: design requirements for wind turbines. IEC International Standard 61400-1. International Electrotechnical Commission, Geneva, Switzerland.
ISO 2007. Petroleum and natural gas industries - fixed steel offshore structures. ISO 19902. International Organization for Standardization, Geneva, Switzerland.
ISSC 2009. Report of the Technical Committee I.2 on Loads, Volume 1, pp. 127210. In C.D. Jang & S.Y. Hong
(eds), Proceedings of the 17th International Ship and Offshore Structures Congress (ISSC2009), Seoul, Korea,
16-21 August, 2009. Seoul: Seoul National University.
ISSC 2012. Report of the Technical Committee I.2 on Loads, Volume 1, pp. 79150. In W. Fricke & R. Bronsart
(eds), Proceedings of the 18th International Ship and Offshore Structures Congress, Rostock, Germany, 9-13
September, 2012. Hamburg: Schiffahrtsverlag Hansa GmbH & Co. KG.
Ivanov, L.D. 2013a. Reliability estimation of ships hull girder in probabilistic terms when ultimate strength is used
as a failure mode. Ships and Offshore Structures 8(2): 141153.
Ivanov, L.D. 2013b. Semi-probabilistic approach for assessment of the requirement for minimum hull girder section
modulus. Ships and Offshore Structures 8(1): 4554.
Iwanowski, B., Vestbostad, T. & Lefranc, M. 2014. Wave-in-deck load on a jacket platform CFD calculations compared with experiments. In Proceedings of the 33rd International Conference on Ocean Offshore and Artic Engineering (OMAE2014), San Francisco, CA, USA, 8-13 June 2014. (OMAE2014-23434).
IWWWFB 2012. International Workshop on Water Waves and Floating Bodies - http://www.iwwwfb.org. [Accessed: 2015-04-19]
Izadparast, A.H. & Duggal, A.S. 2013. Empirical estimation of probability distribution of extreme responses of
turret moored FPSOs. In Proceedings of the 23rd International Ocean and Polar Engineering Conference
(ISOPE2013), Anchorage, USA, 30 June-5 July 2013. pp. 877882.
Jacobi, G., Thomas, G., Davis, M.R. & Davidson, G. 2013. An insight into the slamming behavior of
large high-speed catamarans through full-scale measurements. Journal of Marine Science and Technology 19(1):
1532.
Jelovica, J. & Romanoff, J. 2013. Comparison of load-carrying behavior between web-core sandwich, stiffened and
isotropic plate. In C. Guedes Soares & J. Romanoff (eds), Analysis and Design of Marine Structures; Proceedings of the 4th International Conference on Marine Structures (MARSTRUCT2013), Espoo, Finland, 25-27
Jeong, H.K. & Yang, Y.S. 2013a. Strength assessment of LNG CCS plate using failure criteria for composite
materials. In C-S. Lee & S-H. Van (eds), Proceedings of the 12th International Symposium on Practical
Design of Ships and Other Floating Structures (PRADS2013), Changwon, Korea, 20-25 October 2013. pp. 919
926.
Jeong, W.I., Lee, B.M., Kwon, J.H. & Kim, M.S. 2013b. A study on vibration characteristic of stiffened plates with
fluid coupling effect in a tank. In C-S. Lee & S-H. Van (eds), Proceedings of the 12th International Symposium
on Practical Design of Ships and Other Floating Structures (PRADS2013), Changwon, Korea, 20-25 October
2013. pp. 840-846.
Ji, S. & Liu, S. 2012. Interaction between sea ice/iceberg and ship structures: a review. Advances in Polar Science
23(4): 187195.
Jia, H. & Moan, T. 2012. The effect of sloshing in tanks on the hull girder bending moments and structural reliability of damaged vessels. Journal of Ship Research 56(1): 4862.
Jiang, X. & Guedes Soares, C. 2012. Ultimate capacity of rectangular plates with partial depth pits under uniaxial
loads. Marine Structures 26(1): 2741.
Jin, Y. & Jang, B.S. 2013. A study on probabilistic scenario based fire risk analysis for FPSO structure. In Proceedings of the 27th Asian-Pacific Technical Exchange and Advisory Meeting on Marine Structures, Keelung, Taiwan, 9-12 September 2013.
Jin, Y. & Jang, B.S. 2014. Study on probabilistic scenario based fire risk analysis for FPSO topside structure. In
Proceedings of the 33rd International Conference on Ocean Offshore and Artic Engineering (OMAE2014), San
Francisco, CA, USA, 8-13 June 2014. (OMAE2014-23903).
Kajolli, R., Siriwardane, S.C. & Gudmestad, O.T. 2014. Fatigue life estimation of an offshore steel jacket considering the loading sequence effect. In Proceedings of the 33rd International Conference on Ocean Offshore and Artic Engineering (OMAE2014), San Francisco, CA, USA, 8-13 June 2014. (OMAE2014-23903).
Karatzas, V.A., Kotsidis, E. & Tsouvalis, N.G. 2013. An experimental and numerical study of corroded steel plates
repaired with composite patches. In C. Guedes Soares & J. Romanoff (eds), Analysis and Design of Marine
Karmakar, D., Bagbanci, H. & Guedes Soares, C. 2013. Joint probability distribution of extreme loads of spar and
semisubmersible floating wind turbines using the environmental contour method. In Proceedings of the 32nd In-
199
ternational Conference on Offshore Mechanics and Artic Engineering (OMAE2013), Nantes, France, 9-14 June
2013. (OMAE2013-11269).
Kawamura, Y. & Miyazaki, M. 2011. Structural optimization of the hold frame of a bulk carrier considering lifecycle risk. In C. Guedes Soares & W. Fricke (eds), Advances in Marine Structures; Proceedings of the 3rd International Conference on Marine Structures (MARSTRUCT2011), Hamburg, Germany, 28-30 March 2011. London: CRC Press. pp. 691698.
Kawamura, Y., Ohba, Y. & Kaede, Y. 2013. Lifecycle structural optimization of mid-ship of double hull tanker
based on holistic risk evaluation. In C. Guedes Soares & J. Romanoff (eds), Analysis and Design of Marine
Khedmati, M.R., Zareei, M.R. & Rigo, P. 2009. Sensitivity analysis on the elastic buckling and ultimate strength of
continuous stiffened aluminium plates under combined in-plane compression and lateral pressure. Thin-Walled
Structures 47(11): 12321245.
Khedmati, M.R., Zareei, M.R. & Rigo, P. 2010. Empirical formulations for estimation of ultimate strength of continuous stiffened aluminium plates under combined in-plane compression and lateral pressure. Thin-Walled Structures 48(1): 274289.
Khedmati, M.R., Rigo, P., Amrane, A. & Nazari, M. 2013. Assessment of fatigue reliability for jacket-type
offshore platforms considering dynamic behavior. In Proceedings of the 32nd International Conference on
Kim, S.Y., Kim, Y. & Kim, K.H. 2013a. Statistical analysis of sloshing-induced random impact pressures. In Proceedings of the Institution of Mechanical Engineers, Part M: Journal of Engineering for the Maritime Environment 228(3): 235248.
Kim, Y., Kim, S.Y., Ahn, Y. & Kim, K.H. 2013b. Experimental analysis on sloshing loads for LNG cargo design.
In C-S. Lee & S-H. Van (eds), Proceedings of the 12th International Symposium on Practical Design of Ships
and Other Floating Structures (PRADS2013), Changwon, Korea, 20-25 October 2013. pp. 881891.
Kim, Y., Kim, S.Y., Kim, J. & Lee, J.H. 2013c. Comparison of sloshing pressures in 2D and 3D tanks. International
Journal of Ocean System Engineering 3(4): 225230.
Kim, D.K., Kim, B.J., Seo, J.K., Kim, H.B., Zhang, X. & Paik, J.K. 2014. Time-dependent residual ultimate longitudinal strength - grounding damage index (R-D) diagram. Ocean Engineering 76(1): 163171.
Kim, D.K., Park, D.K., Kim, H.B., Seo, J.K., Kim, B.J., Paik, J.K. & Kim, M.S. 2012a. The necessity of applying
the common corrosion addition rule to container ships in terms of ultimate longitudinal strength. Ocean Engineering 49(1): 4355.
Kim, E., Storheim, M., Amdahl, J., Lset, J. & von Bock und Polach, R. 2013d. Drop tests of ice blocks on stiffened panels with different structural flexibility. In J. Amdahl, S. Ehlers & B.J. Leira (eds), Proceedings of the
6th International Conference on Collision and Grounding of Ship and Offshore Structures (ICCGS2013), Trondheim, Norway, 17-19 June 2013. London: CRC Press. pp. 241250.
Kim, J.H., Kim, D.C., Kim, C.K., Islam, M.D.S. & Paik, J.K. 2012b. Nonlinear structural consequence analysis of
FPSO topsides under fire. In Proceedings of the 31st International Conference on Offshore Mechanics and Artic
Engineering (OMAE2012), Rio de Janeiro, Brazil, 1-6 July 2012. (OMAE2012-83295).
Kim, S.J. & Paik, J.K. 2013. Advanced method for ship structural design under slamming impact pressure
loads. In C-S. Lee & S-H. Van (eds), Proceedings of the 12th International Symposium on Practical Design
of Ships and Other Floating Structures (PRADS2013), Changwon, Korea, 20-25 October 2013. pp. 172176.
Kim, S.Y., Kim, K.H. & Kim, Y. 2012c. Comparative study on model-scale sloshing tests. Journal of Marine
Science and Technology 17(1): 4758.
Koo, J.B., Park, J.S., Ha, Y.S., Jang, K.B. & Suh, Y.S. 2014. Nonlinear structural response analysis for aluminum
helideck. In Proceedings of the 24th International Ocean and Polar Engineering Conference (ISOPE2014),
Busan, Korea, 15-20 June 2014. pp. 977984.
Korhonen, E., Remes, H., Romanoff, J., Niemel, A., Hiltunen, P. & Kontkanen. T. 2013. Influence of surface integrity on the fatigue strength of high strength steel in balcony openings of cruise ship structures. In C. Guedes
Soares & J. Romanoff (eds), Analysis and Design of Marine Structures; Proceedings of the 4th International
Conference on Marine Structures (MARSTRUCT2013), Espoo, Finland, 25-27 March 2013. London: CRC
Press. pp. 255-261.
Koukounas, D. & Samuelides, M.S. 2013. Modeling aspects of strength capacity of intact and damaged ship girders.
In C. Guedes Soares & J. Romanoff (eds), Analysis and Design of Marine Structures; Proceedings of the 4th International Conference on Marine Structures (MARSTRUCT2013), Espoo, Finland, 25-27 March 2013. London:
KRS 2008. Guidance for the direct strength assessment. Korean Register of Shipping, Busan, Korea.
KRS 2013. Rules for the Classification of steel ships, Pt 3, Annex 3-2. Korean Register of Shipping, Busan, Korea.
Ku, A., Nietmann, B., Krzonkalla, V., Wang, X., Chen, J.Y. & Howser, R. 2012. Structural reliability applications
in risk-based inspection plans for semi-submersible floating structures. In Proceedings of the Deep
Offshore Technology International Conference & Exhibition 2012 (DOT2012), Perth, Australia, 27-29 November 2012.
Kubo, A., Kiyosue, T., Kawabata, T., Yajima, H. Hirota, K., Inoue, T., Tani, T., Toyoda, M., Handa, T. &
Yamaguchi, Y. 2012. Experimental study on brittle crack propagation behavior with large scale structural com-
200
ponent model test. In Proceedings of the 22nd International Offshore and Polar Engineering Conference
(ISOPE2012), Rhodes, Greece, 17-22 June 2012. pp. 3643.
Kukkanen, T. & Matusiak, J. 2014. Nonlinear hull girder loads of a RoPax ship. Ocean Engineering 75(1):
114.
Kunal, K. & Surendran, S. 2013. On the scope of using composites as major structural parts of large commercial
ships. In C. Guedes Soares & J. Romanoff (eds), Analysis and Design of Marine Structures; Proceedings of the
Kurian, V.J., Nizamani, Z. & Liew, M.S. 2013. Modelling for uncertainties in resistance for jacket platforms in
Malaysia. Research Journal of Applied Sciences, Engineering and Technology 5(3): 750759.
Kvaleid, J.O.V. & Kvillum, T. 2014. Nonlinear air gap analyses of a semi-submersible compared with linear analyses and model tests. In Proceedings of the 33rd International Conference on Ocean Offshore and Artic Engineering (OMAE2014), San Francisco, CA, USA, 8-13 June 2014. (OMAE2014-24044).
Kwon, K. & Frangopol, D.M. 2012a. Fatigue life assessment and lifetime management of aluminum ships using
life-cycle cost optimization. Journal of Ship Research 56(2): 91105.
Kwon, K. & Frangopol, D.M. 2012b. System reliability of ship hull structures under corrosion and fatigue. Journal
of Ship Research 56(4): 234251.
Laakso, A., Romanoff, J., Remes, H. & Niemel, A. 2013. An analytical method for cabin deck fundamental frequency. In C. Guedes Soares & J. Romanoff (eds), Analysis and Design of Marine Structures; Proceedings of the
Lee, D.H., Jeong, Y.J., Park, M.S. & You Y.J. 2013a. Structural performance of the optimum floating structure for
reduced motion. In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering (OMAE2013), Nantes, France, 9-14 June 2013. (OMAE2013-10697).
Lee, J., Lee, C., Kim, Y. & Lew, J-M. 2013b. Determination of global ice loads on the ship using the measured fullscale motion data. In C-S. Lee & S-H. Van (eds), Proceedings of the 12th International Symposium on Practical
10261031.
Lee, S.G. & Zhao, T. 2013. Structural safety assessment of LNGC MARK III membrane type CCS under sloshing
impact loading. In C-S. Lee & S-H. Van (eds), Proceedings of the 12th International Symposium on Practical
911.
Lee, Y., White, N., Wang, Z., Zhang, S. & Hirdaris, S.E. 2012. Comparison of springing and whipping responses of
model tests with predicted nonlinear hydroelastic analyses. International Journal of Offshore and Polar Engineering 22(3): 1-8.
Lee,Y., White, N., Wang, Z., Tong, J., Xiao, Y. & Li, Q. 2014. Springing loads and fatigue assessment on large
container ships. In Proceedings of the 24th International Ocean and Polar Engineering Conference (ISOPE2014),
Busan, Korea, 15-20 June 2014. pp. 798805.
Lewis, F.M. 1929. The inertia of the water surrounding a vibrating ship. Transactions of the SNAME 37(1): 120.
Li, C., Gao, Z., Moan, T. & Lu, N. 2014. Numerical simulation of transition piece - monopile impact during offshore wind turbine. In Proceedings of the 24th International Ocean and Polar Engineering Conference
Li, C., Ren, H., Qin, J., Zhao, X., Zeng, J. & Xu, D. 2013a. The simplified simulation method of initial imperfection
in ultimate strength FEM analysis. In Proceedings of the 32nd International Conference on Offshore Mechanics
and Artic Engineering (OMAE2013), Nantes, France, 9-14 June 2013. (OMAE2013-11016).
Li, C., Ren, H., Zhao, Z., Zhao, X., Zeng, J. & Xu, D. 2013b. An approximate calculation approach for ultimate
strength of ship hull girder. In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering (OMAE2013), Nantes, France, 9-14 June 2013. (OMAE2013-11023).
Li, L., Li, P. & Liu, Y. 2013c. Structural reliability based design and assessment acceptance criteria development
for fixed offshore platforms in south sea under extreme storm conditions. In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering (OMAE2013), Nantes, France, 9-14 June 2013.
(OMAE2013-11476).
Liu, Y., Yi, H. & Chen, L. 2014. Submarine pressure hull butt weld fatigue life reliability prediction method. Marine Structures 36(1): 5164.
Liu, Z., Amdahl, J. & Lset, S. 2011a. Plasticity based material modelling of ice and its application to ship-iceberg.
Cold Regions Science and Technology 65(1): 326333.
Liu, Z., Amdahl, J. & Lset, S. 2011b. Integrated numerical analysis of an iceberg collision with a foreship structure. Marine Structures 24(1): 337395.
Lokshin, A.Z., Ivanov, L.D., & Mishkevich, V. 2013. Deterministic and probabilistic assessment of the critical
buckling strength of ships grillages (gross panels). Ships and Offshore Structures 8(5): 536564.
Lotfollahi-Yaghin, M.A. & Ahmadi, H. 2012. Derivation of the probability function governing the distribution of
SCFS in two-planer CHS DKT connections of offshore jacket structures. In Proceedings of the 31st International
Conference on Offshore Mechanics and Artic Engineering (OMAE2012), Rio de Janeiro, Brazil, 1-6 July 2012.
(OMAE2012-83971).
LR 2009. Sloshing assessment guidance document for membrane tank LNG operations, ShipRight design and construction. Lloyds Register, London, UK.
201
LR 2013. ShipRight design and construction, analysis of the primary structure of container ships guidance on
direct calculations. Lloyds Register, London, UK.
LR 2014. ShipRight design and construction, guidance notes on the assessment of global design loads of large container ships and other ships prone to whipping and springing. Lloyds Register, London, UK.
Lu, H., Yang, C. & Lohner, R. 2012. Numerical studies of green water impact on fixed and moving bodies. International Journal of Offshore and Polar Engineering 22(1): 123132.
Lu, X., Kumar, P., Bahuguni, A. & Wu, Y. 2014. A CFD study of focused extreme wave impact on decks of offshore structures. In Proceedings of the 33rd International Conference on Ocean Offshore and Artic Engineering
Lus, R.M., Teixeira, A.P. & Guedes Soares, C. 2009. Longitudinal strength reliability of a tanker hull accidentally
grounded. Structural Safety 31(1): 224233.
Ma, M., Hughes, O. & Paik, J.K. 2013. Ultimate strength based stiffened panel design using multi-objective optimization methods and its application to ship structures. In C-S. Lee & S-H. Van (eds), Proceedings of the 12th International Symposium on Practical Design of Ships and Other Floating Structures (PRADS2013), Changwon,
Korea, 20-25 October 2013. pp. 295306.
Ma, M., Zhao, C. & Hughes, O. 2014. A practical method to apply hull girder sectional loads to full-ship 3D finiteelement models using quadratic programming. Ships and Offshore Structures 9(3): 257265.
Maes, M.A., Coile, P.R.V., Caspeele, R., Gao, M. & Dann, M.R. 2013. Probabilistic FAD and ductile tearing assessment. In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering
Malenica, . & Kwon, S.H. 2013. An overview of the hydro-structure interactions during sloshing impacts in the
tanks of LNG carriers. Shipbuilding 64(1): 2230.
Mallahzadeh, H., Wang, Y., Husain, M.K.A., Zaki, N.I.M. & Najafian, G. 2013. Efficient derivation of the probability distribution of extreme responses due to random wave loading from the probability distribution of extreme
surface elevations. In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering (OMAE2013), Nantes, France, 9-14 June 2013. (OMAE2013-10917).
Mao, L., Zheng, Y., He, W. & Gan, L. 2014. Main dimension optimization for scheme design of round-bilge craft.
In Proceedings of the 24th International Ocean and Polar Engineering Conference (ISOPE2014), Busan, Korea,
15-20 June 2014. pp. 917920.
Mao, W. 2014. Development of a spectral method and a statistical wave model for crack propagation prediction in
ship structures. Journal of Ship Research 58(2): 106116.
Mattioli, M., Drago, M., Quondamatteo, F. & Bruschi, R. 2013. Load history for SSFU under multimodal wave
spectra. In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering
Meier, C. & Lehmann, E. 2012. FE coarse net modelling of floors with openings. Ship & Offshore 6(1): 32-35.
Melchers, R.E. 2012. Modeling and prediction of long-term corrosion of steel in marine environments. International
Journal of Offshore and Polar Engineering 22(4): 257263.
Melk, K., Romanoff, J., Remes, H., Varsta, P., Naar, H. & Niemel, A. 2013. Shear response of prismatic passenger
ship hull-girders. In C. Guedes Soares & J. Romanoff (eds), Analysis and Design of Marine Structures; Proceedings of the 4th International Conference on Marine Structures (MARSTRUCT2013), Espoo, Finland, 25-27
March 2013. London: CRC Press. pp. 477-483.
Mirzadehniasar, J., Kimiaei, M. & Cassidy, M.J. 2012. Nonlinear dynamic analysis of jack-up platforms exposed to
extreme random waves. In Proceedings of the 31st International Conference on Offshore Mechanics and Artic
Engineering (OMAE2012), Rio de Janeiro, Brazil, 1-6 July 2012. (OMAE2012-83786).
Mohammed, E.A., Chan, H.S. & Hirdaris, S.E. 2012. Global wave load combinations by cross-spectral methods.
Mohd, M.H., Kim, D.K., Kim D.W. & Paik, J.K. 2013. Probabilistic pit depth corrosion model of subsea gas pipeline. In C. Guedes Soares & J. Romanoff (eds), Analysis and Design of Marine Structures; Proceedings of the 4th
International Conference on Marine Structures (MARSTRUCT2013), Espoo, Finland, 25-27 March 2013. London: CRC Press. pp. 565572.
Mohd, M.H., Kim, D.K., Kim, D.W. & Paik, J.K. 2014. A time-variant corrosion wastage model for subsea gas
pipelines. Ships and Offshore Structures 9(2): 161176.
Mohd Zaki, N.I., Husain, M.K. & Najafian, G. 2013a. Comparison of extreme responses from wheeler and vertical
stretching methods. In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering (OMAE2013), Nantes, France, 9-14 June 2013. (OMAE2013-10594).
Mohd Zaki, N.I., Husain, M.K. & Najafian, G. 2013b. Finite-memory nonlinear system modelling of offshore structural response accounting for extreme values residues. In Proceedings of the 32nd International Conference on
Moon, J.S., Park, T.H., Sim, W.S. & Shin, H.S. 2012. Practical structural assessment of offshore structures for wave
slap. In Proceedings of the 31st International Conference on Offshore Mechanics and Artic Engineering
(OMAE2012), Rio de Janeiro, Brazil, 1-6 July 2012. (OMAE2012-83620).
Mousavi, M.E. & Gardoni, P. 2014. A simplified method for reliability- and integrity-based design of engineering
systems and its application to offshore mooring systems. Marine Structures 36(1): 88104.
202
Muehlner, E., Banumurthy, S. & Murray, J. 2012. Effect of high-frequency response on TLP tendon fatigue. In
Proceedings of the 31st International Conference on Offshore Mechanics and Artic Engineering (OMAE2012),
Rio de Janeiro, Brazil, 1-6 July 2012. (OMAE2012-83264).
Mulder, W.D., Moens, D. & Vandepitte, D. 2012. Modeling uncertainty in the context of finite element models with
distance-based interpolation. In Proceedings of the 1st International Symposium on Uncertainty Quantification
and Stochastic Modeling, Maresias, So Sebastio, SP, Brazil, 26 February-2 March 2012.
Nam, J.M. & Choung, J. 2013. Assessment of average compressive strengths effect of stiffened panels on hull girder ultimate longitudinal strengths. In C-S. Lee & S-H. Van (eds), Proceedings of the 12th International Symposium on Practical Design of Ships and Other Floating Structures (PRADS2013), Changwon, Korea, 2025 October 2013. pp. 10201025.
Nezamian, A. & Altmann, J. 2013. An oil field structural integrity assessment for re-qualification and life extension.
In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering
Nguyen, T.-H., Amdahl, J., Leira, B.J. & Garr, L. 2011. Understanding ship-grounding events. Marine Structures
24(1): 551569.
Nielsen, U.D., Jensen, J.J., Pedersen, P.T. & Ito, Y. 2011. Onboard monitoring of fatigue damage rates in the hull
girder. Marine Structures 24(1): 182206.
Ning, X., Zhang B.L. & Tallavajhula, S. 2013. Structural integrity of a spar in collision with a large supply vessel.
In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering
NORSOK 2004. NORSOK Standard N-004 (Rev. 2, October 2004). Design of steel structures, Appendix A, Design
against accidental actions. Standards Norway, Lysaker, Norway.
Notaro, G., stvold, T.K., Steen, E. & Oma, N. 2013. Collision damages and residual hull girder strength of a ship
shaped FPSO. In C-S. Lee & S-H. Van (eds), Proceedings of the 12th International Symposium on Practical Design of Ships and Other Floating Structures (PRADS2013), Changwon, Korea, 20-25 October 2013. pp. 1011
1019.
Oberlies, R., Khalifa, J., Huang, J., Hetland, S., Younan, A., Overstake, M. & Slocum, S. 2014. Determination of
wave impact loads for the Hebron Gravity Based Structure (GBS). In Proceedings of the 33rd International Conference on Ocean Offshore and Artic Engineering (OMAE2014), San Francisco, CA, USA, 8-13 June 2014.
(OMAE2014-23503).
Ochi, M.K. 1964. Prediction of occurrence and severity of ship slamming at sea. In Proceedings of the 5th Symposium on Naval Hydrodynamics: Ship Motions and Drag Reduction, Bergen, Norway, 10-12 September 1964. pp.
545596.
Ozdemir, M. & Ergin, A. 2013. Nonlinear buckling behavior of stiffened panels. In C. Guedes Soares &
J. Romanoff (eds), Analysis and Design of Marine Structures; Proceedings of the 4th International Conference
on Marine Structures (MARSTRUCT2013), Espoo, Finland, 25-27 March 2013. London: CRC Press. pp. 301
308.
Paik, J.K., Li, C.G., Zhang, X. & Kim, D.K. 2013. Ultimate longitudinal strength performance of tankers associated
with industry corrosion addition practices. In C-S. Lee & S-H. Van (eds), Proceedings of the 12th International
Symposium on Practical Design of Ships and Other Floating Structures (PRADS2013), Changwon, Korea, 20-25
October 2013. pp. 11811188.
Paik, J.K., Lee, S.E., Kim, B.J., Seo, J.K., Ha, Y.C., Matsumoto, T. & Kim, D.H. 2014. A new method for
determining the design sloshing loads for LNG FPSOS. In Proceedings of the 33rd International Conference on
Ocean Offshore and Artic Engineering (OMAE2014), San Francisco, CA, USA, 8-13 June 2014. (OMAE201423730).
Palemn-Arcos, L., Torres-Freyermuth, A., Chang, K.A., Pastrana-Maldonado, D. & Salles, P. 2014. Modeling
wave-structure interaction and its implications in offshore structure stability. In Proceedings of the 24th
International Ocean and Polar Engineering Conference (ISOPE2014), Busan, Korea, 15-20 June 2014. pp. 613
617.
Papanikolaou, A. (ed.) 2009. Risk-based ship design - methods, tools and applications. Berlin: Springer-Verlag.
Papanikolaou, A., Mohammed, E.A., & Hirdaris, S.E. 2014. Stochastic uncertainty modelling for ship design loads
and operational guidance. Ocean Engineering 86(1): 4757.
Papantonatos, G., Liarokapis, D.E., Sfakianaki, K.N., Trahanas, J. & Grigoropoulos, G. 2013. Experimental investigation of a BOC-50 sailing yacht model in regular and random head waves. In: C. Guedes Soares & F. Lpez
Pea (eds.), Developments in Maritime Transportation and Exploitation of Sea Resources Developments in Maritime Transportation and Exploitation of Sea Resources; Proceedings of the 15th International Congress of the International Maritime Association of the Mediterranean (IMAM2013), A Corua, Spain, 14-17 October 2013.
Park, J.S., Chum, M.S., Jeon, S.B., Kim, B.J., Jang, K.B., Suh, Y.S., Kim, J.H. & Lee, J.M. 2013. Advanced guideline to predict the ultimate strength of perforated plate with edge reinforcement. In C-S. Lee & S-H. Van (eds),
Proceedings of the 12th International Symposium on Practical Design of Ships and Other Floating Structures
(PRADS2013), Changwon, Korea, 20-25 October 2013. pp. 10831086.
Park, Y.M. & Jang, B.S. 2013. Hull form optimization procedure of semisubmersible floating production unit considering seakeeping capability and structural weight. In Proceedings of the 23th International Ocean and Polar
Engineering Conference (ISOPE2013), Anchorage, Alaska, USA, 30 June-5 July 2013. pp. 808813.
203
Pasqualino, I., Pinheiro, B. & Ferreira, C. 2013. Stress concentration factors of damaged side panels under cyclic
loads. In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering
Pasqualini, O., Schoefs, F., Chevreuil, M. & Cazuguel, M. 2013. Measurements and statistical analysis of fillet
weld geometrical parameters for probabilistic modelling of the fatigue capacity. Marine Structures 34(1): 226
248.
Patelli, E., Panayirci, H.M., Broggi, M., Goller, B., Beaurepaire, P., Pradlwarter, H.J. & Scheller, G. 2012. General
purpose software for efficient uncertainty management of large finite element models. Finite Elements in Analysis and Design 51(1): 3148.
Payer, H.G. & Schellin, T.E. 2013. A class societys view on rationally based ship structural design. Ships and
Offshore Structures 8(3-4): 319336.
Pei, Z., Wu, W. & Wu, H. 2014. Ultimate strength research on river-sea-going ship with large hatch opening. In
Proceedings of the 24th International Ocean and Polar Engineering Conference (ISOPE2014), Busan, Korea, 1520 June 2014. pp. 685692.
Peng, S., Sun, H., Yue, J. & Wu, W. 2011. Experimental study on slamming load of the trimaran. In C. Guedes
501.
Petey, F. 1988. Ermittlung der Kentersicherheit lecker Schiffe im Seegang aus Bewegungssimulationen. Bericht
Nr.: 487, Institut fr Schiffbau der Universitt Hamburg, Hamburg, Germany.
Pinheiro, B.C., Pasqualino, I.P. & Ferreira, C.F.C. 2013. Numerical study of stress concentration factors in damaged
FPSO side panels under in-plane compression loads. In C. Guedes Soares & J. Romanoff (eds), Analysis and Design of Marine Structures; Proceedings of the 4th International Conference on Marine Structures (MARSTRUCT2013), Espoo, Finland, 25-27 March 2013. London: CRC Press. pp. 263271.
Polic, D., Frank, D., Klanac, A. & Ehlers, S. 2011. Finite element-based shape optimization of an asymmetric steel
sandwich panel joint. In C. Guedes Soares & W. Fricke (eds), Advances in Marine Structures; Proceedings of the
3rd International Conference on Marine Structures (MARSTRUCT2011), Hamburg, Germany, 28-30 March
2011. London: CRC Press. pp. 619623.
Prestileo, A., Rizzuto, E., Teixeira, A.P. & Guedes Soares, C. 2013. Bottom damage scenarios for the hull girder
structural assessment. Marine Structures 33(1): 3355.
Qi, X.L., Lv, G.S., Luo, Y., Zhu, W.Q. & Zhang, Y.G. 2014. Model test and simulation of a new concept of dry tree
semi-submersible. In Proceedings of the 24th International Ocean and Polar Engineering Conference
Rajendran, S., Fonseca, N. & Guedes Soares, C. 2013. Estimation of short term probability distributions of wave
induced loads acting on a cruise vessel in extreme seas. In Proceedings of the 32nd International Conference on
Reddy, J.N. 2000. Mechanics of laminated composite plates and shells Theory and analysis, second edition. Boca
Raton: CRC Press.
Ren, H., Wang, Y., Li, C., Yu, J. & Cui, B. 2014. Risk based inspection plan combined with fatigue failure analysis
for jacket platform. In Proceedings of the 24th International Ocean and Polar Engineering Conference
Renaud, M., De Lorgeril, E., Boutillier, J.B. & Gerard, L. 2013. Fatigue and weather on ultra large
containerships. In C-S. Lee & S-H. Van (eds), Proceedings of the 12th International Symposium on Practical Design of Ships and Other Floating Structures (PRADS2013), Changwon, Korea, 20-25 October 2013. pp. 384
394.
Ringsberg, J.W., Nordqvist, P. & Broman, M. 2014a. Development of a model for global response of
ship hull during ramming of heavy ice features. In Proceedings of the 33rd International Conference on
Ringsberg, J.W., Djavit, D.E., Strande, E. & Anvari, M. 2014b. Fatigue failure analysis of fillet welded joints used
in offshore structures. In Proceedings of the 33rd International Conference on Ocean Offshore and Artic Engineering (OMAE2014), San Francisco, CA, USA, 8-13 June 2014. (OMAE2014-23166).
Romanoff, J., Remes, H., Varsta, P., Jelovica, J., Klanac, A., Niemel, A., Bralic, S. & Naar, H. 2013. Hullsuperstructure interaction in optimised passenger ships. Ships and Offshore Structures 8(6): 612620.
Ronold, K.O., Bergan, J., Amesen, K. & Guttormsen, T.R. 2012. Reliability of axially loaded jacket pile. In Proceedings of the 31st International Conference on Offshore Mechanics and Artic Engineering (OMAE2012), Rio
de Janeiro, Brazil, 1-6 July 2012. (OMAE2012-84162).
Rostami, H. & Oskouei, A.V. 2013. The use of seismic isolation with the method of finite element to reduce vibrations caused by the earthquake in the offshore platform. In C. Guedes Soares & J. Romanoff (eds), Analysis and
Design of Marine Structures; Proceedings of the 4th International Conference on Marine Structures (MARSTRUCT2013), Espoo, Finland, 25-27 March 2013. London: CRC Press. pp. 6166.
Rudan, S., Aic, B. & Vii, I. 2013. Crashworthiness study of LPG ship with type C tanks. In J. Amdahl,
S. Ehlers & B.J. Leira (eds), Proceedings of the 6th International Conference on Collision and Grounding of
Ship and Offshore Structures (ICCGS2013), Trondheim, Norway, 17-19 June 2013. London: CRC Press. pp.
331337.
204
Samuelides, M.S., Koukounas, D. & Pollalis, C. 2013. Residual strength of damaged ships hull. In C-S. Lee &
Saydam, D. & Frangopol, D.M. 2013. Performance assessment of damaged ship hulls. Ocean Engineering 68(1):
6576.
Schafhirt, S., Zwick, D. & Muskulus, M. 2014. Reanalysis of jacket support structure for computer-aided optimization of offshore wind turbines with a genetic algorithm. In Proceedings of the 24th International Ocean and Polar
Engineering Conference (ISOPE2014), Busan, Korea, 15-20 June 2014. pp. 234241.
Scharnke, J., Vestbostad, T., De Wilde, J. & Haver, S. 2014. Wave-in-deck forces on fixed horizontal decks of
offshore platforms. In Proceedings of the 33rd International Conference on Ocean Offshore and Artic Engineering (OMAE2014), San Francisco, CA, USA, 8-13 June 2014. (OMAE2014-23629).
Schmidt, D., Manuel, L., Nguyen, H.H., Sagrilo, L.V.S. & De Lima, E.C.P. 2014. Fatigue reliability assessment
for brace-column details in a semi-submersible hull. In Proceedings of the 33rd International Conference on
Schttelndreyer, M., Tautz, I., Fricke, W., Werner, B., Daske, C., Heyer, H. & Sander, M. 2013. Experimental and
numerical investigations of an alternative stiffening system for ship side structures to increase collision safety. In
C. Guedes Soares & J. Romanoff (eds), Analysis and Design of Marine Structures; Proceedings of the 4th International Conference on Marine Structures (MARSTRUCT2013), Espoo, Finland, 25-27 March 2013. London:
Shimell, A., Tison, T. & Ten Have, T.P. 2012. Structural design of s/y dream symphony: the largest wooden ship
ever built. In Proceedings of the 22nd International Symposium on Yacht Design and Yacht Construction, Amsterdam, The Netherlands, 12-13 November 2012.
Si, H.R. & Chen, Z. 2012. Research on probabilistic prediction method of ship bottom slamming. Shipbuilding of
China 53(1): 9-17. (In Chinese.)
Silva, J.E., Garbatov, Y. & Guedes Soares, C. 2011. Ultimate strength assessment of ageing steel plates subjected to
random non-uniform corrosion wastage. In C. Guedes Soares & W. Fricke (eds), Advances in Marine Structures;
Proceedings of the 3rd International Conference on Marine Structures (MARSTRUCT2011), Hamburg, Germany, 28-30 March 2011. London: CRC Press. pp. 213220.
Silva, J.E., Garbatov, Y. & Guedes Soares, C. 2014. Reliability assessment of a steel plate subjected to distributed
and localized corrosion wastage. Engineering Structures 59(1): 1320.
Simes R, A.J. & Veitch, B. 2013. Evacuation in ice: ice loads on a lifeboat during field trials. In Proceedings of
9-14 June 2013. (OMAE2013-10689).
Simpson, T.W., Peplinski, J.D., Koch, P.N. & Allen, J.K. 2001. Metamodels for computer-based engineering design: survey and recommendations. Engineering with Computers 17(2): 129150.
Smith, C.S. 1977. Influence of local compression failure on ultimate longitudinal strength of a ships hull. In Proceedings of the International Symposium on Practical Design in Shipbuilding (PRADS77), Tokyo, Japan, October 1977.
Sobey, A.J., Blake, J.I.R. & Shenoi, R.A. 2013. Monte Carlo reliability analysis of tophat stiffened composite plate
structures under out of plane loading. Reliability Engineering and System Safety 110(1): 4149.
Sohn, J.M., Kim, B.H., Paik, J.K. & Schleyer, G. 2012. Nonlinear structural consequence analysis of blast
wall structures under hydrocarbon explosive loads. In Proceedings of the 31st International Conference on Offshore Mechanics and Artic Engineering (OMAE2012), Rio de Janeiro, Brazil, 1-6 July 2012. (OMAE201283521).
Sohn, J.M., Kim, S.J., Kim, B.H. & Paik, J.K. 2013. Nonlinear structural consequence analysis of FPSO topside
balst walls. Ocean Engineering 60(1): 149162.
Song, M.J., Ahn, Y.J., Kim, S.Y., Kim, Y.H., Jung, H.J. & Kim, J.H. 2013. Effects of sloshing loads on fatigue
strength of independent Type B LNG tank. International Journal of Offshore and Polar Engineering 23(1): 112
119.
Southall, N.R., Lee, Y., Johnson, M.C., Hirdaris, S.E. & White, N.J. 2014. Towards a pragmatic method for prediction of whipping: wedge impact simulations using OpenFOAM. In Proceedings of the 24th International Ocean
and Polar Engineering Conference (ISOPE2014), Busan, Korea, 15-20 June 2014. pp. 806815.
Stern, F., Yang, J., Wang, Z., Hosseini, H.S., Mousaviraad, M., Bhushan, S. & Xing, T. 2013. Computational ship
hydrodynamics: nowadays and way forward. International Shipbuilding Progress 60(1): 3105.
Stocki, R. 2005. A method to improve design reliability using optimal Latin hypercube sampling. Computer Assisted Mechanics and Engineering Sciences 12(4): 393411.
Storhaug, G. & Hareide, O.J. 2013. Assessment of hull monitoring measurements for a large blunt vessel. In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering (OMAE2013),
Storhaug, G., Oma, N., Blomberg, B. & Hirota, K. 2013. The effect of measured whipping and springing on LNG
vessels. In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering
Storheim, M. & Amdahl, J. 2014. Design of offshore structures against accidental ship collision. Marine Structures
37(1): 135172.
205
Sumi, Y., Yajima, H., Toyosada, M., Yoshikawa, T., Aihara, S., Gotoh, K., Ogawa, Y., Matsumoto, T., Hirota, K.,
to the deck structure of large container ships. Journal of Marine Science and Technology 18(4): 497514.
Suominen, M., Kujala, P., von Bock und Polach, R. & Kiviranta, J. 2013. Measured ice loads and design ice loads.
In C. Guedes Soares & J. Romanoff (eds), Analysis and Design of Marine Structures; Proceedings of the 4th International Conference on Marine Structures (MARSTRUCT2013), Espoo, Finland, 25-27 March 2013. London:
Srensen, J.D. 2012. Reliability-based calibration of fatigue safety factors for offshore wind turbines. International
Sder, C.J., Rosn, A. & Palmquist, M. 2012. Motion based monitoring of racking stresses in Ro-Ro ships. Ships
and Offshore structures 7(4): 389-398.
Tabri, K. 2010. Dynamics of ship collisions. Doctoral dissertation, Aalto University, Espoo, Finland.
Taflanidis, A.A., Vetter, C. & Loukogeorgaki, E. 2013. Impact of modeling and excitation uncertainties on operational and structural reliability of tension leg platforms. Applied Ocean Research 43(1): 131147.
Tammer, M., & Kaminski, M.L. 2013. Fatigue oriented risk based inspection and srtuctural health monitoring of
FPSOs. In Proceedings of the 23th International Ocean and Polar Engineering Conference (ISOPE2013), Anchorage, Alaska, USA, 30 June-5 July 2013. pp. 438449.
Tanaka, S., Okada, H. & Okazawa, S. 2011. Three-dimensional fracture mechanics analyses of surface cracks at
welded joints in ship structure. In C. Guedes Soares & W. Fricke (eds), Advances in Marine Structures; Proceedings of the 3rd International Conference on Marine Structures (MARSTRUCT2011), Hamburg, Germany, 28-30
Tautz, I., Schttelndreyer, M., Lehmann, E. & Fricke, W. 2013. Collision tests with rigid and deformable bulbous
bows driven against double hull side structures. In J. Amdahl, S. Ehlers & B.J. Leira (eds), Proceedings of the
6th International Conference on Collision and Grounding of Ship and Offshore Structures (ICCGS2013), Trondheim, Norway, 17-19 June 2013. London: CRC Press. pp. 93100.
Taylor, R.S. & Richard, M. 2014. Development of a probabilistic ice load model based on empirical descriptions of
high pressure zone attributes. In Proceedings of the 33rd International Conference on Ocean Offshore and Artic
Teixeira, A.P. & Guedes Soares, C. 2013. Probabilistic modeling of the ultimate strength of ship plates with nonuniform corrosion. Journal of Marine Science and Technology 18(1): 115132.
Teixeira, A.P., Ivanov, L.D. & Guedes Soares, C. 2013a. Assessment of characteristic values of the ultimate
strength of the corroded steel plates with initial imperfections. Engineering Structures 56(1): 517527.
Teixeira, A.P., Guedes Soares, C., Chen, N.Z. & Wang, G. 2013b. Uncertainty analysis of load combination factors
for global longitudinal bending moments of double-hull tankers. Journal of Ship Research 57(1): 4258.
Tekgoz, M., Garbatov, Y. & Guedes Soares, C. 2013. Finite element modeling of the ultimate strength of stiffened
plates with residual stresses. In C. Guedes Soares & J. Romanoff (eds), Analysis and Design of Marine Structures; Proceedings of the 4th International Conference on Marine Structures (MARSTRUCT2013), Espoo, Finland, 25-27 March 2013. London: CRC Press. pp. 309317.
Temple, D.W. & Collette, M. 2013. Optimum lifetime maintenance schedule for naval vessels subjected to fatigue
and corrosion. In C-S. Lee & S-H. Van (eds), Proceedings of the 12th International Symposium on Practical Design of Ships and Other Floating Structures (PRADS2013), Changwon, Korea, 20-25 October 2013. pp. 395
402.
Thibaux, P. & Cooreman S. 2013. Computation of stress concentration factors for tubular joints. In Proceedings of
9-14 June 2013. (OMAE2013-10934).
Thomson, W.R. 1972. Theory of vibration with applications. Englewood Cliffs (N.J.): Prentice Hall.
Tiphine, E., Bigot, F., De-Lauzon, J., Sireta F.X., Chung Y.S. & Malenica, S. 2014. Comparisons of experimental
and numerical results for global hydroelastic response of container ship within the WILS III JIP. In Proceedings
of the 24th International Ocean and Polar Engineering Conference (ISOPE2014), Busan, Korea, 15-20 June
2014. pp. 764773.
Travanca, J. & Hao, H. 2014. Numerical evaluation of energy absorption in ship-offshore fixed platform collisions.
In Proceedings of the 33rd International Conference on Ocean Offshore and Artic Engineering (OMAE2014),
San Francisco, CA, USA, 8-13 June 2014. (OMAE2014-24118).
Tsai, S.C., Hsu, S.H., Chien, H.L., Chou, C.M., Malenica, . & Chen, X.B. 2013. Numerical study on seakeepingsloshing coupling effect of container ship for sea trial purpose. In C-S. Lee & S-H. Van (eds), Proceedings of the
12th International Symposium on Practical Design of Ships and Other Floating Structures (PRADS2013),
Changwon, Korea, 20-25 October 2013. pp. 509514.
Tu, Y. & Vorpahl, F. 2014. Influence of superelement support structure modeling on the loads on an offshore wind
turbine with a jacket support structure. In Proceedings of the 24th International Ocean and Polar Engineering
Conference (ISOPE2014), Busan, Korea, 15-20 June 2014. pp. 342349.
Ulfvarson, A. 2011. Corrosion protection of steel ships - localized high strain failures of protective coating. In C.
Guedes Soares, Y. Garbatov, N. Fonseca & A.P. Teixeira (eds), Marine Technology and Engineering. London:
Valdman, N.A. 2014. Methodical approaches and results of safety analysis for offshore transport & technological
systems. In Proceedings of the 24th International Ocean and Polar Engineering Conference (ISOPE2014), Busan,
Korea, 15-20 June 2014. pp. 571578.
206
Van den Abeele, F. & Verleysen, P. 2013. Finite element analysis of subsea pipelines subjected to underwater explosion. In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering
Vant Veer, R. & Vlasveld, E. 2014. Green water phenomena on a twin-hull FLNG concept. In Proceedings of the
33rd International Conference on Ocean Offshore and Artic Engineering (OMAE2014), San Francisco, CA,
USA, 8-13 June 2014. (OMAE2014-23915).
Wang, A., Huang, S. & Barltrop, N. 2014. Long term extreme analysis of FPSO mooring systems based on Kriging
metamodel. In Proceedings of the 33rd International Conference on Ocean Offshore and Artic Engineering
Wang, B., Chen, Y., Shin, Y.P. & Wang, X. 2013a. Thermal analysis and strength evaluation of cargo tanks in
offshore FLNGs and LNG carriers. In Proceedings of the 32nd International Conference on Offshore Mechanics
and Artic Engineering (OMAE2013), Nantes, France, 9-14 June 2013. (OMAE2013-11448).
Wang, Y., Yue, Q. & Bi, X. 2012. Ice force measurement technology of jacket platform in Bohai Sea. International
Wang, Y., Mallahzadeh, H., Husain, M.K., Mohd Zaki, N.I. & Najafian, G. 2013b. Probabilistic modelling of extreme offshore structural response due to random wave loading. In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering (OMAE2013), Nantes, France, 9-14 June 2013.
(OMAE2013-10905).
Vasconcelos De Farias, B. & Netto, T.A. 2012. FPSO hull structural integrity of inspection data. Ocean Engineering
56(1): 1019.
Venzon, R.Z. & Tancredi, T.P. & de Andrade, B.L.R. 2013. Hull optimization of semisubmersible with seakeeping
criteria evaluated with neural network response surface. In C-S. Lee & S-H. Van (eds), Proceedings of the 12th
Korea, 20-25 October 2013. pp. 944952.
Wilken, M., Menk, A., Voss, H. & Cabos, C. 2011. Efficient calculation of fluid structure interaction in ship vibration. In C. Guedes Soares & W. Fricke (eds), Advances in Marine Structures; Proceedings of the 3rd International Conference on Marine Structures (MARSTRUCT2011), Hamburg, Germany, 28-30 March 2011. London:
Won, S.I., Choi, B.K. & Kim, D.S. 2013. Development of the integrated system for DLA of the hull structure. In CS. Lee & S-H. Van (eds), Proceedings of the 12th International Symposium on Practical Design of Ships and
Other Floating Structures (PRADS2013), Changwon, Korea, 20-25 October 2013. pp. 272279.
Wu, H., Wu, W., Gan, J. & Sun, H. 2013. Ultimate strength analysis of a river-sea ship under combined action of
torsion and bending. In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering (OMAE2013), Nantes, France, 9-14 June 2013. (OMAE2013-11007).
Wu, W., Xu, S., Liu, W., Li, X. & Yue, J. 2011. Study on the ultimate strength of high speed trimaran. In C. Guedes
Marine Structures (MARSTRUCT2011), Hamburg, Germany, 28-30 March 2011. London: CRC Press. pp. 521526.
Xu, M. & Guedes Soares, C. 2011. Numerical study of the effect of geometry and boundary conditions on the collapse behavior of short stiffened panels. In C. Guedes Soares & W. Fricke (eds), Advances in Marine Structures;
Proceedings of the 3rd International Conference on Marine Structures (MARSTRUCT2011), Hamburg, Germany, 28-30 March 2011. London: CRC Press. pp. 229237.
Xu, M., Garbatov, Y. & Guedes Soares, C. 2013. Ultimate strength assessment of a tanker hull based on experimentally developed master curves. Journal of Marine Science Applications 12(1): 127139.
Xu, W., Iijima, K., Wada, R. & Fujikubo, M. 2012. Experimental behavior of the post-ultimate strength behaviour
of a ships hull girder in waves. Journal of Marine Science and Application 11(1): 3443.
Yamada, Y. 2014. Numerical study on the residual ultimate strength of hull girder of a bulk carrier after ship-ship
collision. In Proceedings of the 33rd International Conference on Ocean Offshore and Artic Engineering
Yanagihara, D. & Fujikubo, M. 2013. Elastic buckling and elasto-plastic collapse behaviors with torsion of a longitudinal stiffener under axial compression. In C. Guedes Soares & J. Romanoff (eds), Analysis and Design of Marine Structures; Proceedings of the 4th International Conference on Marine Structures (MARSTRUCT2013), Espoo, Finland, 25-27 March 2013. London: CRC Press. pp. 319328.
Yang, H. & Wang, A. 2012. Fatigue reliability based design optimization of bending stiffener. Journal of Ship Research 56(2): 120128.
Yang, N., Das, P.K., Blake, J.I.R., Sobey, A.J. & Shenoi, R.A. 2013a. The application of reliability methods in the
design of tophat stiffened composite panels under in-plane loading. Marine Structures 32(1): 6883.
Yang, J.H., Kim, S.G., Park, J.S., Jung, B.H. & Lee, T.I. 2013b. Numerical analysis for slamming impact loads
and dynamic structural responses of a containership. In C-S. Lee & S-H. Van (eds), Proceedings of the 12th
Korea, 20-25 October 2013. pp. 612618.
Yang, Q.J. (2015). Semi-analytical models for predicting postbuckling response and ultimate strength of imperfect
composite plates. Doctoral dissertation, University of Oslo, Oslo, Norway.
Yoon, M., Koo, B.-Y. & Cho, S. 2011. Isogeometric shape optimization of design-dependent structures. In C.
Guedes Soares & W. Fricke (eds), Advances in Marine Structures; Proceedings of the 3rd International Confe-
207
rence on Marine Structures (MARSTRUCT2011), Hamburg, Germany, 28-30 March 2011. London: CRC Press.
pp. 642647.
Yu, B. & Karr, D.G. 2011. Reliability analysis of marine structural components using statistical data of steel
strength. In C. Guedes Soares & W. Fricke (eds), Advances in Marine Structures; Proceedings of the 3rd International Conference on Marine Structures (MARSTRUCT2011), Hamburg, Germany, 28-30 March 2011. London:
Yu, C., Shin, H. & Lee, J. 2013a. Reliability analysis and reliability-based optimum design of stochastic
structural system. In C-S. Lee & S-H. Van (eds), Proceedings of the 12th International Symposium on Practical
504.
Yu, P., Ren, H., Feng, G. & Zhao, X. 2013b. Deformation analysis and reliability assessment of the ship hull in
irregular waves. In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering (OMAE2013), Nantes, France, 9-14 June 2013. (OMAE2013-10478).
Yu, Z., Hu, Z. & Wang, G. 2013c. Plastic mechanism analysis of structural performances for stiffeners on bottom
floor plating during shoal grounding accident. In C. Guedes Soares & J. Romanoff (eds), Analysis and Design of
Marine Structures; Proceedings of the 4th International Conference on Marine Structures (MARSTRUCT2013),
Espoo, Finland, 25-27 March 2013. London: CRC Press. pp. 219229.
Zamarin, A., Matulja, T. & Hadjina, M. 2013. Methodology for optimal mast and standing rigging selection of a
racing yacht using AHP and FEM. Shipbuilding 64(1): 1121.
ani, V., Piri, K. & Kitarovi, S. 2013. Reliability and robustness based design attributes for multicriteria decision making. In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering
Zavoni, E.H., Iturrizaga, R.M., Faber, M.H. & Straub, D. 2012. Risk assessment for structural design criteria of
FPSO systems. Part II: Consequence models and applications to determination of target reliabilities. Marine
Structures 28(1): 5066.
Zayed, A., Garbatov, Y. & Guedes Soares, C. 2013a. Time variant reliability assessment of ship structures with fast
integration techniques. Probabilistic Engineering Mechanics 32(1): 93102.
Zayed, A., Garbatov, Y. & Guedes Soares, C. 2013b. Reliability of ship hulls subjected to corrosion and maintenance. Structural Safety 43(1): 111.
Zhang, D., Yue, Q., Wang, Y. & Huang, Y. 2012a. Failure modes analyses of ice-resist jacket platforms in Bohai
Sea. In Proceedings of the 22nd International Offshore and Polar Engineering Conference (ISOPE2012), Rhodes,
Greece, 17-22 June 2012. pp. 13531357.
Zhang, J., Shen, Z., Ji, C. & Yin, Q. 2012b. The influence research of ultimate strength to deepwater
semisubmersible platforms structure under corrosion damage. In Proceedings of the 31st International Conference on Offshore Mechanics and Artic Engineering (OMAE2012), Rio de Janeiro, Brazil, 1-6 July 2012.
(OMAE2012-83549).
Zhang, J., Wan, Z.Q., Yin, Q. & Zhang, M.R. 2013a. Calculation method research of geometric collision probability
for ship and ice. In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering (OMAE2013), Nantes, France, 9-14 June 2013. (OMAE2013-10291).
Zhang, J., Wan, Z.Q. & Chen, C. 2014a. Research on structure dynamic response of bulbous bow in ship-ice collision load. Journal of Ship Mechanics 18(12): 106-114. (In Chinese.)
Zhang, M., Li, Y. & Ding, W. 2014b. Damage detection for tripod-type offshore wind turbine structures using modal strain energy methods. In Proceedings of the 24th International Ocean and Polar Engineering Conference
Zhang, Y.H., Ren, H.L., Li, H. & Li, X.Y. 2013b. The research of statistic prediction for wave loads based on the
three dimensional method. In Proceedings of the 32nd International Conference on Offshore Mechanics and Artic Engineering (OMAE2013), Nantes, France, 9-14 June 2013. (OMAE2013-10494).
Zhiqiang, H., Amdahl, J. & Lin, H. 2011. Verification of a simplified analytical method for predictions of ship
groundings over large contact surfaces by numerical simulations. Marine Structures 24(4): 436458.
Zhu, B. & Frangopol, D.M. 2013. Reliability assessment of ship structures using Bayesian updating. Engineering
Structures 56(1): 18361847.
Zhu, J. & Collette, M. 2013. Reliability based structural failure modeling through Bayesian networks. In C-S. Lee &
Zilakos, I.K., Karatzas, V.A., Chatzidouros, E.V., Papazoglou, V.J. & Tsouval, N.G. 2013. A FE based
numerical tool for crack assessment in ship structures employing the CSR loading scheme. In C. Guedes Soares
& J. Romanoff (eds), Analysis and Design of Marine Structures; Proceedings of the 4th International Conference
on Marine Structures (MARSTRUCT2013), Espoo, Finland, 25-27 March 2013. London: CRC Press. pp. 281
288.
Zvyagin, P. & Sazonov, K. 2014. Analysis and probabilistic modeling of the stationary ice loads stochastic process
with lognormal distribution. In Proceedings of the 33rd International Conference on Ocean Offshore and Artic

710 SEPTEMBER 2015
CASCAIS, PORTUGAL
VOLUME 1
COMMITTEE II.2
DYNAMIC RESPONSE
COMMITTEE MANDATE
Concern for the dynamic structural response of ship and floating offshore structures as required for safety
and serviceability assessments, including habitability. This should include steady state, transient and random
responses. Attention shall be given to dynamic responses resulting from environmental, machinery and
propeller excitation. Uncertainties associated with modelling should be highlighted.
COMMITTEE MEMBERS
Chairman:
D.S. Cho, Korea

S. Brizzolara, USA
I. Chirica, Romania
A. Dster, Germany
A. Ergin, Turkey
O.A. Hermundstad, Norway
M. Holtmann, Germany
C. Hung, China (Taiwan)
A. Ivaldi, Italy
C. Ji, China
W.H. Joo, Korea
B. Leira, Norway
S. Malenica, France
Y. Ogawa, Japan
M.A. Vaz, Brazil
A. Vredeveldt, Netherlands
Y. Xiong, UK
D. Zhan, Canada
KEYWORDS
Dynamic response, springing, slamming, whipping, hydroelastic response, vibration, natural frequency,
noise, underwater noise, sloshing impact, blast, underwater explosion, ice, wind, wave, current, internal
flow, vortex, damping, excitation, propeller, machinery, numerical model, model test, full-scale
measurement, monitoring, resonance, fatigue damage, assessment, acceptance criteria, countermeasures,
uncertainty.
210
ISSC committee II.2: DYNAMIC RESPONSE
CONTENTS
1.
INTRODUCTION
211
2.
SHIP STRUCTURES
2.1
2.1.1
Wave-induced vibration
2.1.2
Ice-induced vibration
2.2
2.2.1
Propeller-induced vibration
2.2.2
Machinery-induced vibration
2.2.3
Numerical and analytical vibration studies of ship structures
2.3
Noise
2.3.1
Interior noise
2.3.2
Air radiated noise
2.3.3
Underwater radiated noise
2.4
Sloshing impact
2.4.1
Experimental approaches
2.4.2
Numerical modelling
2.4.3
CCS structural response
2.4.4
Current approaches for sloshing assessment
2.5
2.5.1
Air blast
2.5.2
Underwater explosion
2.6
2.7
Monitoring
2.7.1
Hull structural monitoring system
2.7.2
New sensors technology and application
2.7.3
New full scale monitoring campaigns and related studies
2.8
Uncertainties
2.9
2.9.1
Habitability
2.9.2
Underwater noise
2.9.3
Others
211
211
211
219
220
220
220
221
222
222
224
224
227
227
228
229
229
229
229
230
232
234
234
234
236
239
241
241
242
242
3.
OFFSHORE STRUCTURES
3.1
Vibration
3.1.1
Wind-induced vibration
3.1.2
3.1.3
Vortex-induced motion
3.1.4
Internal flow-induced vibration
3.1.5
3.2
3.3
Noise
3.3.1
Analysis of underwater noise by pile-driving
3.3.2
Measurement and mitigation of underwater noise
3.3.3
Equipment noise
3.4
Blast
3.5
3.6
Uncertainties
3.7
243
243
243
244
245
246
246
249
249
250
250
250
251
252
253
254
4.
CONCLUSION
254
REFERENCES
257
1.
211
INTRODUCTION
In the recent years major challenges of marine industry, as observed by the 2009 and 2012 II.2 Dynamic
Response committee, were originated from energy and environment issues: to enhance energy efficiency
in shipping, to explore and produce resources in deeper sea and arctic region, to reduce adverse impact on
environment, especially as regards the emission of greenhouse gases and noise, and to utilize the Northern
Sea Route in commerce. These have motivated the following trends:
increase in ship sizes to benefit from economy of scale,

modification in ship propulsion system including energy-saving devices mainly aiming at reduction
of fuel consumption and use of cleaner fuels instead of bunker-C oil,
development of larger vessels for shipping in ice-covered and/or ice-infested waters,
development of offshore structures for deeper water and Arctic Ocean mining against extreme
environmental loads,
production of offshore renewable energy, especially through offshore wind farms,
safer shipping and ocean mining against environmental, operational or accidental loads,
introduction of new mandatory or stricter regulations, standards and rules to protect environment or to
reduce adverse impact on human beings and ocean wild life.
These trends were also reflected in the active research and their publications concerning the dynamic
response of ships and offshore structures in the review period. This report is divided into two Chapters to
account for the difference of dynamic responses between ship structures and offshore structures. The
Chapter on ship structures was subdivided into 9 Sections that range from environment-induced vibration
(wave- and ice-induced vibration) to standards and acceptance criteria. Special attention is paid to the
Sections for environment-induced vibration, noise, sloshing impact and monitoring since they have been
main research topics of the past years. The Sections for sloshing impact and monitoring to avoid
repetition introduce a few literatures related to offshore structures because their technologies can be
applicable for both ship and offshore structures. The Chapter on offshore structures was subdivided into 7
Sections. The Section on vibrations treats dynamic response to environmental excitations (wind, wave,
vortex, ice) and operational excitation (internal flow), in which the 2015 II.2 Committee has reviewed
vortex-induced motion of large floating offshore structure instead of vortex-induced vibration of slender
structure since a new special V.8 Risers and Pipelines committee in ISSC 2015 was mandated to covering
their dynamic response. This report cites and reviews 503 references of which 418 references are
published from 2012 to 2014.
2.
SHIP STRUCTURES
Dynamic response of ship structures can be caused by environmental condition such as wave, ice,
machinery, propeller, and impact loads due to sloshing, air blast, under water explosion and so forth.
Many researches have been carried out mainly from the viewpoint of structure excitation load and
structural response. Aspects of them are various in type of ship. In this period, many researches in terms
of dynamic response have been mainly focused on unconventional ships and large ships.
2.1
2.1.1
Wave-induced vibration of ships can occur in two different forms, denoted as springing and whipping.
Whereas springing is a resonance phenomenon, where the wave loads excite one of the lowest natural
frequencies of the hull girder, whipping is a transient phenomenon, where rapidly changing wave loads
excite one or more vibration modes. In both cases, the fundamental 2-node vertical mode is the most
important, although higher bending modes as well as torsion modes may also be excited. Also,
researchers differentiate between linear springing, where the encounter frequency of seaway components
with short wave lengths is in resonance with the natural frequency of the basic hull girder mode, and
nonlinear or sum frequency springing, where the periodic vibration excitation forces act with a higher
order of the wave encounter frequency. The former can be predicted by linear hydrodynamic theories,
while the latter depends on second order hydrodynamic effects; e.g., superposition of different wave
systems. Whipping responses may arise due to different impulsive loads, such as underwater explosions
(see e.g. Liu et al. 2012), but in this section we consider whipping due to wave loads only.
In the last decade we have seen an increasing concern for springing and whipping in very large
containerships (VLCS) and ultra-large containerships (ULCS). These ships have very pronounced bow
flare, high speeds and relatively low natural frequencies in the lowest bending and torsion modes.
212
There seems to be a common agreement that the springing and whipping vibrations can contribute
significantly to the fatigue damage of large containerships. This is supported by a number of full-scale
measurements.
It is also evident that the high-frequency component of the vertical bending moment due to whipping
can be about as large as the wave-frequency component due to the continuous wave-loads, and that the
total bending moment can sometimes exceed the rule design values.
However, the importance of the high-frequency oscillations in the bending moment for the collapse
mechanism of the hull girder is not yet fully investigated. It is not clear how much the limited energy
associated with the high-frequency whipping vibrations contribute to the global hull girder collapse.
Iijima et al. (2011), Iijima & Fujikubo (2012), Xu et al. (2012) and Iijima et al. (2013) found through
experiments that the hull girder collapse mechanism will not have time to develop significantly before the
whipping load starts to decrease, and that the extent of the collapse under whipping is much smaller than
under conventional wave loads. Hence, they state that the risk of collapse in an extreme slamming event
is much smaller than that under wave-frequency loads with the same amplitude.
Jiang & White (2012) studied stiffened panels under dynamic compression loading by a non-linear
finite element code and they found that the ultimate strength under dynamic load is higher than under a
static load; partly due to strain-rate effects and partly due to inertia effects.
Some questions have also been raised to the estimated importance of the springing and whipping
vibrations for the fatigue damage based on rainflow counting and the Miner-Palmgren rule. Kahl et al.
(2013) comment that the fatigue damage rates estimated based on full-scale measurements on
containerships subjected to wave-induced vibrations are not reflected in the observed damages on the
fleet in service. Hence, fatigue tests were undertaken with stress histories simulating wave-frequency and
high-frequency stresses (Fricke & Paetzold 2012) and with stress histories from full-scale measurements
(Fricke & Paetzold 2013, 2014, Kahl et al. 2014). They found that most of the fatigue damage is caused
by the wave-frequency stress cycles, enlarged by whipping, as long as the whipping stress amplitudes are
smaller than the wave-frequency stress amplitudes. The contribution from the additional small stress
cycles due to whipping is rather small. A similar conclusion was also reached by Fukasawa & Mukai
(2013).
Full-scale measurements
A large number of full-scale measurement campaigns have been reported recently. Analysis of these data
gives information on the contribution of ship vibrations to fatigue and extreme loads for various ship
types, as well as information on the importance of different parameters such as ship length, speed, bow
flare angle, wave and wind conditions and trade area. Some recently reported measurements are:
2800 TEU containership (Storhaug 2012)

4400 TEU containership (Storhaug 2012)
4600 TEU containership (Kahl et al. 2013)
278 m containership (Nielsen et al. 2011)
281 m containership (Toyoda et al. 2012)
284 m containership (Toyoda et al. 2012, Ogawa et al. 2012)
302 m containership (Ogawa et al. 2012)
8000 TEU containership (Koo et al. 2011)
8600 TEU containership (Barhoumi & Storhaug 2013)
9300 TEU containership (Koning & Kapsenberg 2012)
9400 TEU containership (Renaud et al. 2013, Andersen & Jensen 2013, 2014)
14,000 TEU containership (Kahl et al. 2013)
98 m wave-piercer catamaran (Jacobi et al. 2014)
119 m Cutter monohull (Drummen et al. 2014)
274 m LNG (Storhaug et al. 2012, 2013)
Large blunt vessel (Storhaug & Hareide 2013)
Many of these papers report that, for large containerships, high-frequency vibrations contribute
significantly to the fatigue damage (typically 3050%) and quite a few of them report that the IACS rule
bending moments are exceeded; some even in relatively moderate conditions (e.g. Barhoumi & Storhaug
2013). Storhaug et al. (2013) reported that vibrations contribute in the range 3050% to the fatigue
damage also for a 274 m LNG vessel with lower flare angle than a containership. Also for a blunt ship,
213
the contribution from high-frequency vibrations in fatigue was in the same order (Storhaug & Hareide
2013), and whipping stresses in excess of the rule-value were reported.
Andersen & Jensen (2014) found that even if the ship was sailing in bow quartering seas, only the
2-node mode seemed to be excited. This was also observed by Koning & Kapsenberg (2012), and by
Storhaug & Moe (2007). Koning and Kapsenberg suggest that it can be a consequence of large damping
in torsion due to the container stacks. The structural damping in steel is low, and a significant contribution
to the total structural damping comes from the cargo. Typical damping is in the range 13% of critical
damping (Storhaug & Moe 2007). A method for estimation of damping from full-scale measurements
using Proper Orthogonal Decomposition (POD) is described by Dessi (2013).
Model tests
Model tests described in the recent literature mainly concern segmented models with an elastic backbone
(BB), or with flexible hinges. The use of fully flexible models has not been reported during the last few
years. A major portion of the models represent large containerships, but some other vessel types have also
been studied. Some of the model tests referenced in the recent literature are:
130 m River-Sea link, Segmented BB (Peng et al. 2014)

290 m academic design, Segmented BB (Iijima et al. 2009)
4400 TEU containership, Hinged (Storhaug 2007)
5000 TEU containership, Segmented BB (Takaoka et al. 2012)
8600 TEU containership, Hinged (Storhaug et al. 2010a)
9300 TEU containership, Segmented BB (Kobayakawa et al. 2012)
9400 TEU containership, Segmented BB (Maron 2012)
10,000 TEU containership, Segmented BB (Kim, B.W. et al. 2014)
13,000 TEU containership, Hinged (Storhaug et al. 2010b)
13,000 TEU containership, Segmented BB (Takaoka et al. 2012)
112 m high speed catamaran (Lavroff 2009, Davidson et al. 2013, Thomas et al. 2011)
45 m high speed monohull, Hinged (Wu & Stambaugh 2013)
113 m monohull (frigate), Segmented BB (Bennett et al. 2014b)
125 m monohull (cutter) (Drummen et al. 2014)
260 m monohull (JHSS), Segmented BB (Piro et al. 2012)
333 m VLCC, Segmented BB (Wang, X. et al. 2014)
High speed ferry, Segmented BB (Dessi & Ciappi 2013)
Practically all these models were instrumented for vertical hull girder responses, such as the Vertical
Bending Moment (VBM). Some of the containership tests also focus on asymmetric responses in oblique
seas. This introduces challenges in the model design, since the shear center of the open ship sections is
located below the hull bottom. Kim, B.W. et al. (2014) tested a 10,000 TEU containership segmented
backbone model with two different backbone cross sections: U-shaped and H-shaped. Whereas the
vertical responses were not influenced by the cross section types, the measured torsional responses were
significantly influenced. Another challenge with torsion measurements is how to derive the torsional
moments at stations along the hull from the measured strains in the backbone. Kim, B.W. et al. (2014)
investigated different loading modes used in converting strains to torsional moments. Hong et al. (2011)
reported, for the same model (with H-backbone), that direct use of torsion strains instead of torsion
moments may give a more accurate interpretation of the torsion response. Tiphine et al. (2014) report
from the same model (with U-backbone), and they demonstrate that different longitudinal distributions of
the torsional moment can give rise to the same strains at a particular cross section. Hence, they
recommend and validate a new method (Bigot et al. 2011) using a base of distortion modes to take all
strain gauges into account simultaneously.
The models with flexible hinges have the advantage that the hinges can be made with adjustable
stiffness. The problem of connecting the backbone to the hull segments without disturbing the
measurements is also avoided. However, modelling torsion with a low shear center and appropriate
coupling with horizontal bending seems difficult with hinged models. It was shown by Wu et al. (2012)
that for a monohull it is sufficient to use 3 hinges if the hydroelastic effects mainly originate from the
2-node vertical mode. It was also shown that hydroelastic effects in the vertical bending moment are not
sensitive to the stiffness distribution among the flexible hinges, so the same stiffness may be used for all 3
hinges. This conclusion was also made by Takaoka et al. (2012) based on tests with segmented backbone
models of 5000 TEU and 13,000 TEU containerships.
214
Methods for assessment of mode shapes and associated damping during model and full-scale tests are
presented and discussed in Mariani & Dessi (2012), Dessi & DOrazio (2012), Dessi (2013). A new
system for measuring motions and hull girder deformations during springing and whipping model tests
was presented by Bennett et al. (2014a), while Kim, Y.I. et al. (2013) described a method for identifying
whipping events using wavelet cross-correlation.
Zhu & Moan (2013a) analysed the test data from the 8600 TEU and the 13,000 TEU hinged models
(Storhaug et al., 2010a, b) in head seas. They found that while the total dynamic sagging moment
amidships is significantly larger than the total dynamic hogging moment at a forward speed of 25 knots,
the difference is insignificant at a speed of 15 knots. Nonlinear wave forces that increase the sagging
moment, such as bow flare slamming, increase more rapidly with the speed than do the forces which
increase the hogging moment. Hence, with a realistically moderate speed in severe seas, accounting for
voluntary speed reduction, the difference between the dynamic sagging and hogging moments may be
lower than reflected by current computer tools and rule formulas. Zhu & Moan (2013b) discuss possible
reasons for the relatively high nonlinear hogging moments observed in some model tests and full-scale
measurements, and they suspect that it may be due to the suction force in the bow during water exit. It is
well known that there is a downwards added mass force during water exit, and this force increases if the
hull has a relatively wide flat bottom in the foreship. It was observed by Zhu et al. (2011) that for a model
with such a flat forward bottom area, the dynamic hogging moment was approximately 30% higher than
the sagging moment. The difference disappeared when the bow was replaced by a version with smaller
flat bottom and larger flare (Zhu & Moan 2012).
Zhu & Moan (2013b) analysed the test data from the 13,000 TEU hinged model in oblique seas
(Storhaug et al. 2010b). It was found that the nonlinearities causing ship vibrations are more pronounced
in long crested seas than in short crested seas. Hence, model tests in long crested seas may slightly
exaggerate the effect of the high-frequency vibrations.
It was observed by Zhu & Moan (2012) that whipping vibrations in torsion are sensitive to bow shape.
When a bow with larger flare was introduced, the torsional whipping increased significantly. From the
same model tests, Zhu et al. (2011) found that the hydrodynamic damping in the torsion modes is almost
negligible compared to that in the vertical and horizontal bending modes. Torsional vibrations were most
pronounced in waves from 30 degrees off the bow.
Based on the tests of 4400 TEU, 8600 TEU and 13,000 TEU hinged models, Storhaug (2014)
concluded that bow flare angle is the main factor for the importance of whipping, whereas vessel size is
less important. Similar findings were reported by Takaoka et al. (2012) from tests with 5000 TEU and
13,000 TEU ship models. Storhaug (2014) also found that extrapolating 20-year design values from a
Weibull-fit of the model test data gave dynamic hogging moments that were well above the IACS URS11
design moments. Voluntary speed reduction will influence the results and should be further documented.
Piro et al. (2012) observed whipping responses in head irregular waves that reached almost the same
amplitudes as the wave-induced bending moment. The tests were done with a segmented backbone model
of a 260 m monohull. The backbone had non-uniform stiffness along the length.
Tests of a segmented monohull with a non-uniform-stiffness backbone were also reported by Dessi &
Ciappi (2013). They found that the slams were grouped into clusters, and thus violating the hypothesis of
mutual independence between successive impacts that form the basis of many statistical models. Hence,
within a cluster, the whipping oscillations are not damped out before the next slam occurs.
Bennett et al. (2012, 2014b) used a segmented flexible backbone model with uniform stiffness of a 113
m frigate to measure hull girder bending moments in head seas and study the effect of abnormal waves.
Thomas et al. (2011, 2012) and Lavroff et al. (2013) reported on model tests with a segmented flexiblehinge model of a 112 m high-speed wave-piercer catamaran in head regular and irregular waves.
Hong et al. (2014) studied bow flare slamming forces and pressures from experiments with a 10,000
TEU containership. They concluded that slamming pressures increase strongly with ship forward speed,
and that they are sensitive to variations in the instant surge velocity.
Numerical methods
Hydroelastic response calculations involve a structural model of the ship, a hydrodynamic model of the
fluid and a method for coupling the two models to ensure that the interaction effects are properly
captured. The following paragraphs give a summary of the status and recent developments in these
methods.
The structural model is usually a 3D Finite Element Model (FEM) or, for monohull, a beam model.
The most common beam model is based on the Timoshenko theory, but for cases involving torsion and
215
warping, more advanced models are applied (e.g. Vlasov theory). Senjanovic et al. (2012a, b) present a
new theory for thin-walled girders taking into account shear influence on torsion as well as the
contribution from structural discontinuities, such as bulkheads and the engine room structure, to the hull
stiffness. A new structural model that accounts for warping and structural discontinuities is presented and
compared with alternative methods in Miao et al. (2012). Alternative methods for structural modelling are
also discussed in Kim, J.H. et al. (2013a).
A common method for fluid-structure coupling is the modal approach. An aggregate of the natural
modes of the dry structure is used to represent the global structural deformations. To avoid convergence
problems it may be useful to apply the hybrid method presented by Wu & Moan (2005). Alternatively, the
pressure distribution from the fluid model is transferred to the structural model (together with
accelerations) and the hull deformations are transferred from the structural model to the fluid model for
every time-step. Comparison of different structural modelling methods and coupling methods used with a
RANSE-solver for the fluid are presented by Oberhagemann & El Moctar (2012) and Oberhagemann et
al. (2012a). Based on the modal approach Senjanovi et al. (2012a, b, 2014a, b) performed linear
springing analyses of 7800 TEU and 11,400 TEU container ships combining advanced beam model, with
the above mentioned improvements and 3D potential flow code. Firstly, they compared their results with
rigid body calculations, and secondly a comparison with fully coupled 3D FEM-BEM (Boundary Element
Method) from Bureau Veritas tool is done. They reported good correlation at the global assessment level.
However, for stress concentration assessment and fatigue analysis some improvements are still necessary.
Lee, Y.W. et al. (2014) investigated methods to determine the springing response for fatigue
assessment. Fluid Structure Interaction (FSI) models are used to investigate nonlinear wave actions and
wave induced global loads acting on large container ships. Time domain simulation techniques in critical
wave frequencies are employed to investigate effects of springing on design bending moments of a large
container ship based on the Lloyds Register 2014 Rule requirements for container ships that mandate
springing fatigue analysis for large container ships. It was found that the fatigue life is reduced due to the
inclusion of springing effects on a large container ship. The analysis indicated that the effects of springing
for the sample container ship accounted for approximately 20 per cent of the predicted fatigue life for the
components considered. Hence the fatigue life calculated by standard methods which do not include
springing effects have to be multiplied by 0.8 to obtain an estimate of the fatigue life including springing.
Kashiwagi & Hara (2012) presented a computer code for the analysis of ship hydroelastic problems,
based on the Rankine panel method in the frequency domain and the mode superposition method for
representing the elastic deformation of a ship. 3D FEM commercial code is used to compute the dry
eigen-frequencies and corresponding elastic mode shapes. As a numerical example, a 2-meter modified
Wigley model advancing in head waves at Fn = 0.2 is considered, and verification has been made by
presenting the wave pattern on the free surface generated by the forced oscillation with elastic modes and
by confirming the convergence in the amplitudes of elastic modes as the mode number increases.
Kim, J.H. et al. (2012) performed springing analysis of a VLCS based on hybrid BEM-FEM method,
investigating influence of different beam modelling and structural damping on the results. They compared
motion RAO and the load signal time-histories with experimental data, providing discussion on
discrepancies.
Kim, K.H. et al. (2013) analysed hydroelastic response of two real container ships in head seas, i.e.
6500 TEU and 10,000 TEU, respectively, using partitioned method, where the fluid domain surrounding a
flexible body is solved using a B-spline Rankine panel method, and the structural domain is handled with
a three-dimensional finite element method. The two distinct methods are fully coupled in the time domain
by using an implicit iterative scheme. The numerical results of natural frequency and the motion
responses of simple and segmented barges are computed to validate the present method through
comparisons with experimental and numerical results.
Proper assessment of vessel motions is a prerequisite for correct slamming and whipping calculations.
Hence, the developments in seakeeping methods described in the report of the Loads committee are also
relevant for whipping. In whipping calculations two main types of fluid methods are applied: (i) boundary
methods, where the unknowns are on the boundaries of the different domains and (ii) field methods,
where the unknowns are distributed throughout the volume of the domains.
When using the boundary methods, like strip theory or 3D panel methods, violent local flow
phenomena, such as slamming is treated by separate methods and the associated forces are added to the
forces obtained from the global boundary method. For the field methods, like RANSE-solvers, the violent
local flow phenomena are in principle implicitly included in the overall method, but these phenomena
may still need special attention regarding grid and time-step resolution.
216
Since slamming in the bow is the main source of wave-induced whipping responses, it is important to
assess the slamming forces in a sufficiently accurate and practical manner. Water entry is often studied by
2D methods, since 3D calculations add significant complexity the problem.
The most common method for calculation of the slamming forces has been the von Karman momentum
approach, where the forces are obtained as the rate of change of the high-frequency added mass as the
bow enters the water. The method is normally applied to ship sections in a two-dimensional manner, as
recently exemplified in Wu et al. (2012), Wu & Stambaugh (2013) and Rubanenco et al. (2013). The
method may also be applied in three dimensions, as in orak et al. (2013), where they approximate the
bow of a containership with a half-cone, which in turn is represented by a half-disc for which an analytic
added mass formula is used (Jensen & Pedersen 2009).
In the von Karman approach, the pile-up of water around the body is neglected, and this is known to
give too low slamming forces. Hence, the two-dimensional Generalized Wagner Method (GWM), first
introduced by Zhao et al. (1996), has become popular. In contrast to the momentum methods, the GWM
also gives the pressure distribution on the hull section and it can therefore be used in local slamming
analyses as well as in whipping assessment. Application of the GWM in head seas whipping analysis of
large containerships is presented by Kim, J.H. et al. (2013a, b) and Kim & Kim (2014). They also do
calculations with a 2D von Karman based momentum method, where the added mass is calculated
analytically by approximating the section with a wedge. From analyses of 6500 TEU, 10,000 TEU and
18,000 TEU containerships they concluded that both these 2D slamming methods give similar whipping
responses, and that they compare reasonably well with experiments for low and moderate ship speed. For
a speed above 20 knots, the methods tend to overestimate the whipping responses. It was assumed that
this could be attributed to the exclusion of 3D flow effects.
In some recent whipping calculations (e.g. Tiphine et al. 2014) the so-called Modified Logvinovich
Model (MLM) has been used. The MLM (Korobkin & Malenica 2005) is a simplification of the GWM in
the sense that it includes the water pile-up, but the pressure is calculated on a flat-plate extending between
the two body-water intersection points, rather than on the exact body surface. Hence, the MLM is
primarily applicable to blunt sections.
Ogawa & Takagi (2012) use the Displacement Potential Approach (Takagi & Ogawa 2007,
Ogawa & Takagi 2009) for slamming force calculation on a large containership. From comparison with
measurements in regular head waves they concluded that the agreement was good for the whipping
responses. Lee, Y.W. et al. (2012) use Generalized Momentum Theory (Tuitman 2008) in whipping
analysis of a 13,000 TEU containership, but no validation is presented. This latter theory is also used by
Lee et al. (2011) on the WILS-II 10,000 TEU containership in head regular and irregular waves, and
reasonable agreement with measurements was reported.
A 2D slamming method is expected to give too high slamming loads, since the flow is restricted to a
2D plane. This was pointed out as a possible reason for overestimation of whipping responses by Kim,
J.H. et al. (2013a, b) and Kim & Kim (2014) when using a 2D GWM. On the other hand, a von Karman
momentum approach will underestimate the slamming loads since the water pile-up is neglected. This
was pointed out as a possible reason for underestimation of whipping responses by Andersen & Jensen
(2012) when using a 2D momentum method. Since the effects of the 2D approximation and the von
Karman approximation act in opposite directions, one could have hoped that a 2D von Karman method
would give reasonable results for practical applications. However, this is apparently not the case in
general. Another alternative is to use the GWM and then apply correction factors for 3D effects based on
geometric considerations (e.g. Hermundstad & Moan 2005), but there are many uncertainties in using
such correction factors.
Another aspect related to whipping is that a 2D slamming analysis for a vertical ship section will
produce only vertical slamming forces, while a bow slamming event would also give forces in the
ships longitudinal direction. Wu & Stambaugh (2013) made an empirical correction to the 2D bow
slamming forces on a high speed monohull, so that the slamming force got a component in the
longitudinal direction, and this improved the agreement between the measured and calculated vertical
bending moments. A more rational correction could have been obtained with a 2D GWM, since the
pressure distribution is then available and the longitudinal force could have been found from
integration on the 3D geometry of a slice of the ship. Again there are uncertainties associated with
such corrections.
Another way of adjusting the 2D slamming methods to better account for the realistic flow pattern and
the effects of forward speed, is to use 2D sections that are inclined towards aft rather than being vertical.
This is used by Bigot et al. (2011) with a 2D GWM, and reasonable agreement for whipping in head seas
217
is obtained for the WILS-II 10,000 TEU containership. The same method is used by Tuitman et al.
(2013), and they also do analyses in oblique waves where the sections are rotated about a vertical axis, so
that they become parallel with the wave crests. The method is applied to a 144 m frigate, but no validation
is presented. Tilting the 2D sections in order to better capture the direction of the main flow during
slamming is an engineering approach, but it is difficult to determine what is the most appropriate tilt
angles (see e.g. Hermundstad et al. 2002). Hence, this is yet another approach associated with
uncertainties.
To overcome the problems associated with the 2D methods, efforts are made to develop practical 3D
boundary methods for slamming. Introducing the third dimension into e.g. the generalized Wagner
formulation, adds considerable complexity to the problem. One example is the work of Chezhian (2003).
Recently however, Tassin et al. (2012) have presented some promising results for a 3D generalized
Wagner formulation.
Due to the difficulties of generalizing the boundary methods for slamming from 2D to 3D, it is
tempting to resolve to a field method, like e.g. a Finite Volume Method (FVM) solving the RANS
equations. Hence, there are many papers on field methods applied to bow slamming. Another advantage
of these methods is that they can generally handle flow separation from e.g. bulbous bows during violent
slamming events.
Many studies of 2D problems, with the purpose of validating field methods, have been presented.
Wang & Soares (2013) compared results from the FEM code LS-DYNA with the 2D ship-section drop
tests of Aarsnes (1996). The results from LS-DYNA do not appear to compare better with experiments
than do those from a BEM based on GWM. Yoshikawa & Maeda (2013) applied LS-DYNA to water
entry of a rigid and an elastic wedge. Good agreement with a Wagner solution for the rigid wedge was
documented, but no other validation was presented.
Veen & Gourlay (2012) use a 2D Smoothed Particle Hydrodynamics (SPH) method and compare with
the 2D drop tests of Aarsnes (1996), and quite good agreement was reported. They also applied the
method to a hull section with a relative velocity taken from seakeeping experiments with a 120 m
monohull by Hermundstad & Moan (2005), and the agreement was good. The method was also combined
with a strip theory and pressures were compared to tests by Ochi (1958), but then the agreement was less
favourable. It was recommended to extend the SPH method to 3D and to use it for slamming calculations
together with a less resource demanding method for ship motion calculations.
Southall et al. (2014) applied a FVM-based RANSE-solver (based on OpenFOAM) to wedge impacts
at different angles and compared with model tests (MOERI 2013). The Volume of Fluid Method (VOF) is
used for free surface capturing. The intention is to use the method together with ship motion software to
predict whipping. However, some discrepancies with the 2D wedge tests were found.
Another application of OpenFOAM software is presented by Piro & Maki (2011, 2013). The method is
applied to a rigid and an elastic 2D wedge. Good agreement with a Wagner solution for the rigid wedge
during water entry is documented. They also study water exit and it is shown that large forces occur in
this phase. It can be important to account for these water exit forces in practical whipping analysis in
order to accurately predict the hogging moments. A 2D method for water exit was presented by Korobkin
(2012), and results agreed well with those from OpenFOAM (Piro & Maki 2011).
Rahaman et al. (2012) use a FVM-VOF-based RANSE-solver (WISDAM-X) to study the bow flare
slamming on a containership. Good agreement is found when comparing with results for a 2D section
from Zhao et al. (1996). In the 3D containership case, results are compared with 2D analyses with the
commercial code FLUENT, and differences are documented. This may be due to 3D effects.
An overset grid method with a level set function for free surface capturing (FANS) is applied by Lee,
S.K. et al. (2012) and by Chen & Chen (2014) to study bow and stern slamming of a containership. Yang
et al. (2013) use the commercial FVM-VOF-based RANSE-solver STAR-CCM+ together with FEM to
analyse slamming pressures and structural responses in the bow of a containership. In these papers, no
validation is presented.
A common approach in whipping analysis is to first calculate the linear hydrodynamic forces in
frequency domain using a 2D strip theory formulation or a 3D panel method. Next, the linear frequencydomain results are transferred into time-domain, as proposed by Cummins (1962). The motion equations
are then integrated in time, and nonlinear modifications to the forces are added to the linear force vector
at each time-step. Nonlinear hydrostatic forces and Froude-Krylov forces are normally included, since
this is straight forward. Moreover, slamming forces must be included. An overview of different
refinements in the use of boundary methods in hydroelastic analyses is given by Malenica & Derbanne
(2012).
218
Tiphine et al. (2014) analyse the 10,000 TEU WILS III-JIP containership model (U-backbone) by
using a 3D BEM method combined with a 3D FEM model of the structure. Coupling is performed by the
modal approach. The frequency-domain hydrodynamic solution is transformed to time-domain and
nonlinear Froude-Krylov forces are added. Slamming forces are calculated by the 2D MLM. Quite good
agreement with experimental VBM whipping responses in head and oblique regular waves is obtained,
while the agreement for the torsional whipping responses is less favourable. A similar approach is used
for the same containership by Bigot et al. (2011), except that they used 2D GWM for slamming. Some
limited comparisons are made with earlier model tests (with H-backbone) and reasonable agreement is
found for VBM whipping in head and oblique regular and irregular waves. Torsion whipping in oblique
irregular waves shows less agreement. More validation is needed to conclude on the performance of this
method.
The same 10,000 TEU containership is analysed by Lee et al. (2011) using a similar method, but with a
Timoshenko beam for the structure and a Generalized Momentum Theory for slamming. Reasonable
agreement for VBM whipping in head regular and irregular waves was demonstrated for a few sample
time-windows. The same method is used by Lee, Y.W. et al. (2012) for a 13,000 TEU containership in
head seas without comparing with measurements.
Kim, J.H. et al. (2013a, b) and Kim & Kim (2014) use a method similar to that of Bigot et al. (2011)
in the analysis of the same containership. They found quite good agreement for the whipping VBM in
head waves, but they concluded that the whipping responses at the highest speed (20.3 knots) were
slightly overestimated when using the 2D GWM. A simplified 2D von Karman based momentum
formulation for slamming with wedge-approximation of the section gave slightly better agreement. It was
suspected that 3D effects reduced the slamming loads in the experiment, as discussed above.
Wu & Stambaugh (2013) apply the nonlinear strip theory code WINSIR to a 25 m high speed
monohull in head seas. A beam model is used for the structure, and coupling is performed by the hybrid
modal approach (Wu & Moan 2005). The frequency-domain solution is transformed to time-domain and
nonlinear modifications to the hydrostatic and Froude-Krylov forces, as well as 2D von Karman
momentum-based slamming loads are added. Comparisons with model tests of a 25 m high-speed
monohull in irregular waves showed that the measured and simulated standard deviations of the wavefrequency and the high-frequency part of the VBM agreed well. However, it was found that the theory
generally underestimated the extreme sagging moments, while it overestimated the extreme hogging
moments in various sea-states. One possible explanation was that the longitudinal component of the bow
slamming force is neglected in the 2D slamming calculations. By using an empirical adjustment of the
slamming force, the simulated sagging moments increased and compared better with experiments.
Field methods have become increasingly popular in recent years, and e.g. FVM-VOF-based RANSEsolvers are applied to slamming and whipping problems by several research groups.
Ship motion and whipping analyses with RANSE-solvers have been performed in regular and in
irregular long crested waves. The main focus has been on head seas (e.g. Oberhagemann & El Moctar
2012, Oberhagemann et al. 2012a, Seng et al. 2012) , but recently Oberhagemann et al. (2012b) present
results from oblique regular waves for the 10,000 TEU ULCS tested in WILS-II. Results are promising,
but there is little severe slamming and whipping in the cases studied. An exception is Bertram et al.
(2011) that show a time series including a few whipping events in head waves. They also discuss the
challenges related to calculation of the bow flare slamming forces with the VOF-method due to the
transition between fluids with two different densities. Forces during water exit however, are not
discussed, although the simulations include water exit events.
Instead of using a RANSE-solver for the complete whipping problem, one may use a simpler method,
e.g. strip theory, for ship motion calculations and then apply the RANSE-solver to the local bow
slamming problem only (e.g. Rahaman & Akimoto 2012).
An alternative field method is presented by Mutsuda et al. (2012). This is a coupling of the SPH
particle method and the Constrained Interpolation Profile (CIP) grid method, and it is partly motivated by
the challenges of accurately capturing violent free surface flows, like slamming, with the VOF-method.
Some comparisons with model tests where a segmented backbone model is dropped into calm water show
quite good agreement in hull girder load effects. The method is also applied to simulate an elastic ship in
waves involving bow slamming and wave breaking, but without comparison with measurements.
Practical assessment of extreme hull girder load effects including whipping
The available computer tools for assessment of wave-induced load effects differ significantly in
computational efficiency. The linear frequency-domain strip theories are efficient and can be used to
219
estimate load effects in a large number of combinations of loading condition, speed, wave heading and
sea-state. The nonlinear time-domain boundary methods require higher resources and can be used to
generate long time-series in a number of selected conditions. Finally, some of the most refined methods,
like the RANSE-solvers, can only be used to study a few events covering a few wave cycles. There are
several recent papers focusing on ways of using the various simulation methods in a practical design
process.
A review of alternative ways of using a combination of fast and computationally demanding simulation
tools is given by Schellin et al. (2013).
Oberhagemann et al. (2012c) use two different methods, the coefficient of contribution method
(Baarholm & Moan 2000) and the response conditioned wave episodes method, together with two
computer codes; a linear zero-speed 3D BEM and a RANSE-solver, in order to obtain extreme load
effects in ships. A method based on extrapolation of response upcrossing rates in sea-states is applied by
Oberhagemann et al. (2013) to study hogging moments in two containerships. It is based on analysis of
available Monte Carlo simulation results and the application of a RANSE-solver. Another method based
on upcrossing rates is presented in Mao & Rychlik (2012).
Derbanne et al. (2012) and De Hauteclocque et al. (2012) discuss and compare different ways of
establishing design waves. They conclude that the static regular design wave gives very poor results for
the effect of nonlinear Froude-Krylov loads, and it is unable to compute whipping responses. Methods
based on a dynamic response on a regular design wave or a response conditioned wave give better results,
even for extreme whipping responses. The latter of these two is not much more demanding to apply, and
it is therefore recommended even though it does not give significantly better results. It is reported that
much more accurate results are obtained when the simulations are done in an irregular sea-state with an
increased wave height. Increasing the wave height decreases the return period of a given linear extreme,
and hence decreases the required duration of the simulation. Using simulations in artificially increased
wave heights is also discussed by Jensen (2010).
2.1.2
Compared to operation in open water, vibrations of vessel operating in ice conditions are significantly
increased because of the ship hull and propellers interacting with ice. These increased vibrations may
cause fatigue damage of structures, equipment failure and also create unfavourable conditions for crew
habitation. Belov & Spiridonov (2012) analysed features of vibrations recorded on icebreakers and icegoing ships and the reasons for its amplifications, and presented estimation methods for excitation forces
and vibration levels in ice conditions. Some methods of reducing hull vibration and the effects of
vibration on the crew were proposed in the paper.
Pressure distribution and structural response in ship-ice interaction are two important topics in model
tests and numerical simulation of ship-ice studies.
Kujala & Arughadhoss (2012) conducted a series of model tests for two different ship models to
measure ice-induced pressure distribution using I-Scan 210 tactile sensors which were placed in bow,
bow shoulder, midship, and aft of the models. A statistical method was used to analyse ice crushing
pressure in level ice and compared with full-scale measurements. It was found that the line load shape as
a function of the nominal load width and load level was comparable with the full scale data, and the
crushing pressure can be compared with full scale data.
Sawamura (2013) predicted distribution of ice pressure along the waterline when a ship advanced into
level ice based on his numerical code for ship maneuvering in level ice. The structural response of a
stiffened panel under the calculated ice pressure distribution was analysed by a linear elastic FEM. He
pointed out that the shape and the location of the ice pressure area had strong influence on the structural
response.
Quinton et al. (2012) conducted a numerical study of ship structural response by using two realistic
ice load models (crushable-foam ice model and 4D ice pressure model) in LS-DYNA software. In the
simulation, these two realistic moving ice load models generated ice pressure distributions in
accordance with actual laboratory and field observations and coupled the normal and tangential
components of the moving loads. Simulation results showed that these realistic models agreed with the
findings of the previous work, that is, moving loads cause substantially more damage than stationary
loads.
Gagnon & Wang (2012) conducted numerical simulations of a tanker collision with a bergy bit. The
ship response in the ice contact part was predicted. In the simulation, the major portion of the vessel was
treated as a rigid body and a portion of its bow where the ice contact occurred was modelled as a typical
ship grillage that could deform and sustain damage. A load measurement from the laboratory tests
220
compared reasonably well with a rough estimate from the simulation. The numerical techniques should
make simulating a wide variety of ice interactions with vessels feasible.
2.2
2.2.1
Propeller-induced vibration
For the propeller-induced vibrations, the excitation forces are transmitted into the ship via the shaft line
and in the form of pressure pulses acting on the ship hull. The shaft line forces are mainly responsible
from the vibrations of shaft lines, and the propeller-induced pressure fluctuations from the vibrations of
ship structures.
Van Esch et al. (2013) used CFD to compute the hydrodynamic coefficients for torsional and axial
vibrations for a Wageningen B-series of ship propellers in open-water condition. In their method of
analysis, the coefficients for added mass and quasi-steady fluid induced damping are determined by a
proper selection of the value of the reduced frequency. It was shown that the unsteady vortices in the
wake of the propeller have an effect on both magnitude and the phase of the fluid induced fluctuations in
thrust and torque.
Zhang, G. et al. (2014) introduced a transfer matrix method for a propulsion shafting system in order to
describe its dynamic behaviour. Using hydrodynamic lubrication theory and small perturbation method,
the axial stiffness and oil damping were deduced and the foundation stiffness was estimated by finite
element method. Based upon these values, the Campbell diagram describing natural frequencies in terms
of shafting rotating speed were obtained. The effect on the 1st natural frequency of considerable
variations in thrust bearing stiffness was investigated. It was found that the amplitude of variation of the
1st natural frequency in range of low rotating speeds was great.
Lee, K.H. et al. (2014) developed a numerical prediction method for estimating the hull pressure
fluctuation induced by propeller sheet cavitation. In this study, the combined hydrodynamic and hydroacoustic method was employed to predict the pressure fluctuation caused by a volume variation in the
propeller sheet cavitation. The developed numerical method and findings are useful sources for predicting
the hull pressure fluctuations induced by a propeller at the design stage.
Song et al. (2014) investigated a performance of a periodic isolator to reduce the vibration and noise
radiation of an underwater vehicle caused by propeller forces. A simplified physical model of the
underwater vehicle is developed first. The isolation performance of the periodic isolator is investigated
and compared with the traditional homogenous isolator. In the study, an integrated isolation device was
proposed.
Wei & Wang (2013) investigated the unsteady hydrodynamics of the excitations from a 5-bladed
propeller at two rotating speeds running in the wake of a small-scaled submarine and the behaviour of the
submarines structure and acoustic responses under the propeller excitations. The propulsion is simulated
using computational fluid dynamics, so as to obtain the transient responses of the propeller excitations.
Finally, the structure and acoustic responses are predicted using a finite element/boundary element model
in the frequency domain.
Lee et al. (2013) studied the correlations between the propeller cavitation induced pressure fluctuation
measurement in a cavitation tunnel and full-scale data. Two major considerations were attempted for the
conventional test technique. One is to immerse the model ship deeper than its scaled design draught so
that the boundary layer effect can be minimized in the area of propeller disk. The second is to avoid the
resonance frequency range of model ship by adjusting propeller RPM. The approaches taken in this study
would provide a systematic method for a full scale correlation enhancement.
2.2.2
Machinery-induced vibration
The trend in shipping towards larger and more fuel efficient ships with lower speeds led to new engine
developments with ultra-long stroke and low revolution rate, e.g. like the MAN G-type engines.
Compared with the former engines like MANs S-type engine, the new G-type engines offer a higher
power at lower engine speeds. As expected, this changes the main engines external forces and moments
transmitted to the ship hull. The increase in guide force moment between the G and S type can be a factor
of 1.25 to 1.5, depending on ratings.
In MAN Diesel & Turbo (2014), a study with special attention towards the structural vibrations related
to this excitation source is given. Both global hull vibration response and local main engine vibration
performance was measured and analysed for a 64,000 dwt bulk carrier equipped with a 5G60ME-C9
engine. For this class the 5S60ME-C8 type engine was a widely applied for propulsion plant. Compared
with an average vibration level for an S-type engine, it is stated that the hull and engine vibration
221
performance has been improved for the G-type engine based on a comparison of the 5th order vibration
levels at the same percent of MCR speed (100% G-type 77rpm at 8500kW; 100% S-type 105rpm at
11,300 kW). A possible explanation might be that the disadvantage of higher forces is compensated by
lower rpm and also possibly higher safety margin to engines H-type natural frequency.
Engine excited vibration of slow running two stroke engines is mainly transmitted into the hull
structure via their foundations and top bracings. Top bracings are widely used to control H-mode
resonance of large two stroke main engines and to reduce the vibration of the engine itself in transverse
and sometimes also in longitudinal direction. Two types of top bracings have been widely used: the
Mechanical type Top Bracing (MTB) with a friction connection and the Hydraulic type Top Bracing
(HTB). When MTBs are installed, it is relatively easy to predict vibration characteristics through modal
analyses with the assumption of linear axial stiffness. However, because a HTB includes mechanical,
hydraulic, and pneumatic valve and chamber systems, the nonlinearity of the axial stiffness makes it
difficult to quantitatively estimate the detuning effect.
Lee et al. (2010) carried out dynamic compression tests of HTB. They proved that the axial stiffness of
HTB, which shows high nonlinearity according to exciting amplitudes, is independent of exciting
frequencies. Jin et al. (2010) tried to prove the force-carrying mechanism by deriving the equivalent
compressive stiffness of the HTB. Choung (2013) reports that the complex operation mechanism can be
idealized by a piecewise linear stiffness curve. However, if the excitation force at the running speed range
does not exceed the static setting force the operating mechanism will stay within the static load stage so
that the HTB behaves like the MTB. The setting force is expected to be exceeded in case of resonance
condition only. In this case, the governing load stage will be shifted to the light load stage, and thus the
natural frequency of the system will be changed. Since the aim of the HTB is to detune the system a high
setting force may be preferable. It should be noted that the dynamic behaviour of the HTB will strongly
vary depending on its individual design being different for different suppliers. None of the publications
addresses the damping effect of hydraulic top bracings.
Behaviour of passive resilient mountings changes according to the frequency of the exciting force and
may be not accurately predicted by numerical simulations. However these data are needed for appropriate
selection with respect to structureborne noise and vibration. In order to achieve the dynamic transfer
properties of resilient mountings for medium speed marine diesel engines, laboratory tests are to be
carried out. Moro et al. (2013) describe the theoretical background for evaluating the interaction between
the resilient mounting and the diesel engine foundation along with the fundamentals for acquiring data on
the dynamic behaviour of resilient mountings by means of laboratory tests.
2.2.3
Numerical and analytical vibration studies of ship structures
Due to increased awareness of comfort issues it would be beneficial to begin vibration analyses already in
the earliest design phase. However, limited time and resources prevents usage of detailed finite elements
analyses. Method suitable for conceptual design should produce reliable and useful information in basis
of limited knowledge about the structure.
Laakso et al. (2013) introduce an analytical method for calculating the fundamental frequency of an
orthotropic stiffened cabin deck. The fundamental mode is assumed to be either transversal or
longitudinal global mode, or local deck plate mode. Shapes of the global modes are approximated by
applying Newtons laws of motion, and static beam and plate theories. The shape approximations include
local deformations of the deck plate and its stiffeners. Rayleighs method is used to calculate
corresponding eigen-frequencies of the approximated mode shapes. The presented method is validated by
finite elements method. Sufficient accuracy is obtained for structural analyses in conceptual design.
Furthermore, the paper shows that the effect of local deformations is significant in certain cases.
Brubak et al. (2013) present an approximate, semi-analytical computational model to solve the
eigenvalue problem based on the Rayleigh-Ritz method for plates subjected to in-plane loading. The
model may handle complex plate geometries, by using inclined stiffeners to enclose irregular plate
shapes. Relatively high numerical accuracy is achieved with low computational efforts.
Cho et al. (2014b) present a procedure for the vibration analysis of stiffened panels with arbitrary edge
constraints. It is based on the assumed mode method, where natural frequencies and modes are
determined by solving an eigenvalue problem of a multi-degree-of-freedom system matrix equation
derived by using Lagranges equations of motion. The Mindlin thick plate theory is applied for a plate,
while the effect of stiffeners having the properties of Timoshenko beams is accounted for by adding their
strain and kinetic energies to the corresponding plate energies. A comparison of results with those
obtained by the finite element method is provided and good agreement is achieved. An extension of this
222
procedure for vibration analysis of plates with openings and arbitrary edge constraints is presented by
Cho et al. (2014c) as well as the consideration of contact with fluid on one side by Cho et al. (2014a).
Senjanovi et al. (2013) deal with simplified geometric stiffness formulation which has some
advantages in hydroelastic analysis comparing to the consistent mass matrix determined by the shape
functions of all degrees of freedom used for construction of conventional stiffness matrix, or with a
lumped mass matrix related to deflection degrees of freedom.
Jeong et al. (2013) investigated the fluid coupling effects in tank with stiffened plates by measurement
and numerical analysis through FEM applying the Helmholtz method used by MSC.Nastran. The natural
frequency in first mode of the measured is near in phase mode of FEM analysis. The analysis results show
that the effect of added mass parameter increases according to the fluid depth and the increase is steady at
some extend depth. Also the natural frequency of out-of-phase mode decreases lower than that of inphase mode because the out-ofphase mode makes the fluid movement restrained more and the inertial
force larger.
Neto et al. (2012) analysed the influence of the kinetic energy of the fluid adjacent to the hull of a
tanker ship in its vertical vibration frequencies and compared them with experimental measurements
obtained during sea-trials. The calculations were performed by the added mass formulations from
different referenced approximations like Burrill, Todd, Kumai and Lewis/Landweber. The results showed
that the theoretical formulations for obtaining added mass provide reliable results. The formulation of
Kumai for this case was the most suitable, probably because it is a specific formula for oil vessels. The
other formulations also showed comparatively close to the measurement results, although they are more
indicated for general cargo ships.
Generators can be a serious vibration source, giving the supporting structure is an important role in
ensuring an acceptable vibration level e.g. at the crews working and livings spaces. Daifuku et al. (2014)
studied on a bulk carrier the optimization of the reinforcement configuration of the engine room using
topology optimization to improve the anti-vibration characteristics. The vibrations were reduced by about
10% while the total weight was reduced by 8% of the original structure.
2.3
Noise
Traditionally, ship noise control is intended as protection of crew and passengers and improvement of
habitability and comfort on board, as well as reduction of underwater noise for navy vessels. As
already pointed out in the last ISSC 2012 report, the recent trend in this field is to regard noise as a large
spectrum problem, thus including also the protection of the external environment intended as coastal
environment (ports, coastal aerial) and underwater wildlife.
As showed by many researchers, the actual status of the research in those different directions is not
consistent and reflects the different histories that every field had in the past.
The interior noise has been studied since the first vessel has been fitted with an engine, initially with
regard to the protection of the seafarers and later to establish comfort criteria for both passengers and
crew. A complete set of design tools and regulations are available and so broadly used that today arises
the question if they are still valid or need to be refreshed.
The exterior airborne noise is, on the opposite side, the newest field of application and suffers of a big
lack in design tools and rules. It has been recognized to be a problem only recently and efforts are being
made to establish a common basis to deal with it.
Underwater Radiated Noise (URN) has been studied since many years mainly within the scope of
military application and consequently many results are not public. The civil application of studies in URN
is mainly addressed to the protection of the wildlife fauna from the noise pollution and so also the
research method and scopes are rather different.
2.3.1
Interior noise
The interior noise is maybe the oldest field of application and has a complete set of regulations, which are
nowadays put under discussion to prove their response to the actual status of the technological
development.
Analysis methods and applications
In Seiler & Holbach (2013a, b) the A-weighted sound pressure level is put under discussion to check if it
is still a suitable parameter for the judgment of the overall noise comfort on board. With the help of audio
records on a real ship and with a listening test of 30 people, they tested different psychoacoustic
parameters like loudness, sharpness and other more, to find out whether more sophisticated indicators can
223
better describe the noise situation of a vessel. The result was that, even if some parameters can be taken
into account, the A-weighted sound pressure level confirms its validity.
A similar interrogative is studied in Badino et al. (2012a) where a large overview among the existing
regulations and limits for the different impact fields of ship noise is presented. With regard to the internal
noise, several methods and well established normatives are present. However, there are still some aspects
of the characterization of the internal noise that are unclear and for which a more detailed description is
missing (e.g. tonal components of noise). Also the well-known overall dB(A) level is put under
discussion with regard to its capability to describe the noise phenomena on board. The authors also
presented a wide study on the existing regulation framework for the noise emitted from ships, both
radiated inside and outside the ship, as well as underwater. They argue that the actual degree of
knowledge (and consequently of active regulations) of the different impacts of the noise emitted is quite
different and reflects the different amount of time elapsed since their effects have been recognized to be
negative. Airborne noise transmitted and radiated inside the ship has been broadly studied since many
years and from different points of view, ranging from the safety and health in the crews working areas to
the comfort of passengers on board. Regulation on the matter are well established and constantly updated.
On the other side, Badino et al. (2012f) presented psychoacoustic criteria taken from the civil
engineering field, and particularly the Room Criteria Mark II as highly valuable to describe the acoustic
comfort on board due to their improved capability to describe side effects like spectrum characteristics,
low frequency annoyance, etc.
Turan et al. (2011) have performed a comparative study regarding the effectiveness of IMO 468 among
other normative, confirming the need for an update of the IMO rule. Noise measurements have been
carried out on six ships of similar characteristics (oil and chemical tankers, LOA ranging between 106
and 148 m) during sea trials and the results analysed with IMO 468 and EU directive 2003/10/EC on
noise exposure of workers finding that ships easily fulfilling the IMO limits are not compliant with the
EU rule.
Stritzelberger et al. (2013) proposed an alternative method to the classical FEA/SEA (Finite Element
Analysis/Statistical Energy Analysis) based numerical approaches, namely the wave based Energy Finite
Element Method (EFEM); the method is based on a finite element mesh and, due to the nature of the
solution, the mesh refinement is not dependent on the frequency of interest, which makes the method
capable of handle big sized/high frequency models. They also highlight the necessity to model correctly
the nature of the coupling between structural elements as the biggest limit of the method today.
Boroditsky & Fischer (2012) presented an algorithm based on the Smiths transfer function prediction
procedure, relating the vibration response to an airborne noise excitation. The authors discussed
especially the mechanism of structural excitation below the structure resonance, where a multimodal
approach is not possible.
Seok et al. (2011) presented a case study regarding an excessive noise level inside the engine control
room of a LNG carrier, due to a steam dumping line excited by a control valve. The authors conducted a
comparison between measured noise levels on board and a structural analysis of the steam dumping line,
sorting out that the structureborne noise generated by an airborne noise excitation caused by the steam
valve as the root of the problem.
Marchesini & Piana (2012) studied an improvement to an existing luxury yacht in order to obtain a
higher class comfort, showing how intensity measurements can be effective in problem solving related
problems.
Performance of acoustic insulation components
The improvement of the acoustic behaviour of insulation panels and partitions walls is still a major trend,
as pointed out by many research works. Kim, S.R. et al. (2012) performed a comparative study of a
prefabricated cabin mock-up made of sandwich panels, whose insulation characteristics had been
previously measured in laboratory. They explained the difference in the Weighted Sound Reduction Index
RW of almost 9 dB with both the construction gaps and irregularities of a real construction, but also with
the flanking paths. With the help of structureborne noise measurements and SEA analysis they identify
the transmission of the steel floor is the biggest responsible of the radiation in the receiving room.
Cho (2013) presents an experimental and analytical investigation of the low frequency vibro-acoustic
characteristics of a massive floating floor with resilient layer, leading to the conclusion that the highest
transmissibility of the system is when the bending modal frequency of the floating floor plate matches the
in-situ natural frequency of the isolator which is different from what calculated with the method of ISO
90521 (1989). The study indicates that some contributory factors such as plate dimension and location of
impact may be added to the assessment factor of 63 Hz.
Patraquim et al. (2011) analysed the influence of the lining materials usually present between a
perforated panel and the mineral wool beneath. This material, usually a thin fabric sheet, is found to have
224
a large influence on the resulting acoustic absorption of the system. Materials with high air-flow
resistivity lower the absorption of the panels, particularly at mid frequencies. Comparison between
measured data and theoretically calculated data were found in good agreement in the case of small
resistivity.
In Borelli et al. (2013) a study of the effect of different perforated facing panel sheet in sound
absorption panels is experimentally investigated and the different behaviour due to different combinations
of percent of open area and air gap between insulation material and facing panel are discussed.
Experimental data shows good agreement with literature, showing that for ratios perforated/total area
greater than 20% the effect of the perforations doesnt vary anymore and on the other side, air gaps seems
to have little influence, at least in the range considered.
2.3.2
Air radiated noise
The study of the noise emitted from the ship in the external space, regarded as noise pollution affecting
coastal inhabited areas and harbours, is a relative young field of application. As a consequence of that,
on one side there is a big lack of normative regarding the protection of such areas and methods and
parameters to be used to assess the radiated noise, on the other a lot of scientific works are in progress.
In particular, the ongoing European project SILENV (Ship Innovative Solutions to Reduce Noise) is
largely described and presented in Badino et al. (2013). The aim of the project is to define a sort of green
label in which noise target for every topic and guidelines to reach them are provided. The project is
organized in 5 Work Packages (WP) that are briefly described within the article: (i) noise related needs,
(ii) noise measurements, (iii) solutions, (iv) models and (v) green label requirements.
Within the framework of the SILENV project, again Badino et al. (2012c) propose an assessment
criterion for the airborne noise emitted by sea-going ships. Such method has been based on similar ones
already existing for similar purposes such as the ISO 3746 (2010a) norm and has been adapted to take
into consideration the dimension of the measured object and the particular ambient condition.
In another study, Badino et al. (2012b) presented the European Directive 2002/49/EC of which aim is
to assess and to control the environmental noise due to the main sources, including ports and port
activities. The Directive prescribes the development of a Noise Strategic Mapping (NSM) by
superimposing the contribution of every single source. The authors proposed a method to achieve this
goal. Moreover, limits to be applied to the port noise pollution are proposed.
Badino et al. (2012e) observe that the noise emitted from a ship has different impacts on the different
receivers that are affected by this noise: underwater with regard to the marine fauna as well as for military
purposes, inside the vessel for the comfort and safety of crew and passengers and outside the vessel in
coastal inhabited areas. They analysed the relevant normative framework that regulates such different
fields, observing that the state of the art of the regulation is very different and being very different the
time elapsed since those fields have been studied.
Di Bella & Remigi (2013b) propose a method to evaluate, with a retrofit procedure, the sound power
level of a cruise ship starting from sound pressure level measurements. The spectrum obtained can be
used as a point source in a mathematical model to further analyse the noise impact of a ship on a coastal
urban area. Di Bella & Remigi (2013a) conducted a study on the noise pollution over the historical city of
Venice due to the cruise ships sailing in the internal waters of the city or moored at harbour. Even if the
ambient situation of Venice is not simple, due to the heavy traffic (from both large cruise ships as well as
public water transportation), they demonstrated that an accurate and precise planning of such activities
can help to control effectively the total amount of noise affecting the urban population. Moreover, they
propose a buffer zone, in analogy with what existing for railways, where target noise limits should be
fulfilled.
Draganchev et al. (2012) conducted a comparative study on the noise emitted from three different
merchant ships while moored. Besides the differences between the three ships, they found out that for
every ship the main noise source was the ventilation fans (engine and cargo rooms) and that the response
on the environment was strongly dependent on the presence of surrounding reflecting surfaces, including
the container carried on the main deck of one of the three ships. A numerical simulation of that has shown
good agreement with the experimental data.
Badino et al. (2012d) made a comparison between measured and calculated data for two different
ships. They used a commercial available software to model ship geometry and surrounding ambient and
obtained good correlation with the measured data, although they pointed out that the correct modelling of
the ship geometry is a critical point since it has a rather large effect on the results.
2.3.3
Underwater radiated noise
The underwater radiated noise has been studied since a very long time for military applications: as a
consequence of that, there is lack of publication since the results of those studies cannot often be
225
published being covered by secret. The study of underwater radiated noise as a civil application is more
recent and is focused mainly in the protection of the marine environment from noise pollution.
In this sense, there is a fundamental difference with the studies developed in military application: the
requirement for a naval vessel to comply with is always clear, being represented by the state of the
art of the detection devices sensitivity. Being the progress in the technology of those systems always
running, there is virtually no limit in lowering the level of noise radiated by a ship: the more silent, the
better.
In civil applications, where time schedules and involved costs for a construction are normally tighter
than in a naval vessel, the question of how to tune up correctly the noise emission arises. How much a
ship must be silenced depends upon the target threshold level which is represented by the level of
annoyance/pain of the marine fauna.
Many of the most recent studies have focused to establish the correct limit to be taken into account and
in finding a correlation between such limit and the behaviour of the marine fauna.
Regarding the noise sources on a ship, the study of the propellers behaviour confirms to be a major
issue, both from the point of view of the numerical idealization and of the development of new materials.
The interaction of propeller and hull has been regarded as well as secondary noise source and studied in
many works.
Shipping noise and marine life
Scheer & Ritter (2013) have conducted a study on the underwater radiated noise of different kind of
ferries on their routes out of the Canary Islands (Spain) and compared the result with the ambient noise
and the audibility threshold of some marine mammals. They calculated the theoretical time available to
the cetacean to escape from an incoming vessel, that appears to be in most cases long enough for the
animal to move, thus coming to the conclusion that other factors, like the audiogram of the cetacean, need
deeply to be investigated
Allen et al. (2012) analysed noise emitted from several ships cruising in the Gulf of Maine and
compared those with ships routes and hearing threshold of whales, concluding that mysticetes should be
physically capable to detect ships. Nevertheless, many collisions between marine mammals and ships
occur and they can be explained with several effects like shadow zones or background noise especially in
shallow waters. The authors recommend that future studies focus on accurate 3D modelling of the
acoustic environment as well as on hearing capability of mysticetes.
Erbe et al. (2012) developed a simple method to derive a large scale noise map of the oceans, matching
AIS data from ships with marine wildlife distribution maps thus providing a tool to establish areas where
the underwater noise is - or is not- a problem.
Starting from a long term observation and measurement of ships traffic, McKenna et al. (2012)
propose a method to calculate Sound Exposure Level (SEL) for different kind of ships. The equation is a
simple and useful tool to input shipping noise in modelling the marine environment.
Baudin et al. (2014) describe the European research project AQUO (Achieve QUiter Oceans by
shipping noise footprint reduction) and propose to use a description of the oceans environment as noise
mapping in order to establish the effect of the anthropogenic sound on marine life.
Numerical methods
Salio (2013) presented a study where semi-empirical methods and BEM numerical simulation of a marine
propeller are compared to a series of measurements taken on different ships (cruise/patrol vessel/
megayacht). Results show that the reliability of such methods is not fully satisfactory even if they are
commonly used in the early design stage practice.
Peters et al. (2014) investigated the effect of the internal mass distribution on the radiated sound
power of a submerged hull due to external excitation. They used a fully coupled FE/BE analysis of a
submarine, modelled as a stiffened cylinder with hemispherical ends, and investigated the response of
such model to different distribution of the internal masses, taking into account the relevant damping
ratio, since the masses are often resiliently mounted. The obtained results showed a significant
contribution of such internal distribution of dumped masses on the total acoustic radiation of the
structure.
Pereira & Cordioli (2011) presented a comparison between a procedure based on a hybrid FE-SEA
method and an uncoupled BEM analysis for the prediction of the structural and acoustical response of a
submerged structure, where fluid impedance must be taken into account. They observed that the results of
the hybrid FE-SEA method are poor when used to calculate the structural response but are in good
agreement when concerning the acoustic dynamic stiffness.
Wei & Wang (2013) made a study on a scaled submarine model, pointing out as the axial forces,
normally taken in less consideration in such studies, can have a big contribution in the excitation of the
226
submarine hull, in both breathing and bending modes, contributing a lot to the total noise emitted from the
hull. However, the authors point out the fact that the model used in such study is a small scale model,
with no internal machinery and ballast tanks, excited from an unskewed propeller, being that slightly
different from the real applications.
Lee, K.H. et al. (2014) presented an inversion procedure for sheet cavitation propeller noise based on a
nonlinear optimization algorithm and acoustic BEM. The cavitation noise is modelled by mean of single,
double and triple monopole sources. Inversion is performed with an ASSA (Adaptive Simplex Simulated
Annealing) algorithm. The procedure enables the inversion of the source strength and positions using data
measured above the propeller. The double monopole model shows the best accuracy and the result
obtained by an acoustic BEM model are in good agreement with the measured data.
URN due to hydrodynamic forces of propellers
Paik et al. (2013) have investigated the behaviour of three kind of flexible propeller, measuring also the
emitted noise, reporting a good acoustic behaviour in case of glass fiber propeller construction. Bertetta
et al. (2012) carried out a CPP propeller optimization with a cavitating panel code coupled with an
optimization tool, at two very different design points with different pitches. The procedure showed good
results not only in the cavitation behaviour of the propeller but also with a significant reduction of the
emitted noise. They concluded that if the same approach is used to a simpler, one working point problem,
the improvement should be even better.
Takinac & Taralp (2013) have developed an empirical model of broadband noise for marine
propellers, based on a common formulation from Brown (1976). In addition to that, they modulated the
obtained spectrum to obtain a realistic audible signal to be used in a sonar simulator for training purposes.
The results of such empirical prediction model are found to be in good agreement with the available
experimental data.
Korkut & Atlar (2012) studied the effect of foul release paint on propellers, finding a beneficial effect
with regard to the noise emitted in non-cavitating conditions while an increase of noise in fully developed
cavitation condition.
Lee, J.H. et al. (2014) proposed a single nozzle air injection system to reduce the propeller cavitation
inducing hull excitation pressure. They found a significant degree of reduction even outside the air bubble
layer that cannot be explained with a simple air-cushion insulation effect. They developed a theoretical
model to describe the scattering behaviour of a simplified air bubble thus explaining such reduction with
a destructive interference effect.
In order to override the difficulties related to an underwater noise measurement, Jeon & Joo (2014)
developed a method to predict the underwater noise radiated by a propeller by mean of onboard
measurements. The method, which takes place from the measurements of the transfer function of the hull
in a dry dock, has been verified in a tank test experiment and applied to the prediction of real ships, with
good correlation.
URN due to other sources
Kellett et al. (2013) presented a study based on a CFD approach predicting URN of an LNG carrier,
compared with measured data. They pointed out how some modelling parameter such as the presence of
the free surface or the way to model the propeller can influence the accuracy of the result, thus giving
indications to designers in building their models.
Merz et al. (2013) studied the effectiveness of the active control of the sound power radiated from a
submarine hull excited by propeller forces. The control is achieved by two different systems and can be
implemented with a resonance changer, thus reaching the maximum performance. A similar study, based
on the effectiveness of a resonance changer, is described in Caresta & Kessissoglou (2011) who propose a
dynamic model of a submarine hull coupled to a propeller shafting system. A resonance changer is also
included in the system and the steady state response of the hull under harmonic excitation from the
propeller is calculated. The introduction of the resonance changer reduces the radiated sound pressure
level in both narrow and broadband frequency ranges significantly.
Roth et al. (2013) have conducted a comparative study of the noise emitted from an ice-breaker in
standard operation vs. during ice-breaking operation. They found that ice-breaking operation will increase
the noise emitted up to 10 dB, especially in backing and ramming maneuvers. They recommended an
improvement of the study of the underwater noise propagation in iced sea, as well as the development of
measurement protocols being the increase of arctic traffic a consolidated trend.
Traverso & Trucco (2014) presented a method to measure the underwater noise emitted from a ship,
measured from the ships bow. The method has the advantage to measure the radiated noise directly close
to its origin, avoiding the uncertainty connected to the calculation of the transmission loss.
2.4
227
Sloshing impact
This section is devoted to the dynamic structural response of the Cargo Containment System (CCS) inside
the membrane type LNG tanks of different floating units (ships, FPSOs, etc.). Membrane type LNG
tanks are currently dominating the LNG market and the correct assessment methodology for verification
of structural integrity of both CCS and the associated hull structure is critical for their design.
Unfortunately it appears that both the reliable deterministic models of hydro-structure interactions during
the sloshing impacts as well as the overall rational methodology for determination of the representative
design conditions are still missing, at least from the direct calculation point of view. The classical
engineering methods (laboratory tests, numerical simulations and full scale measurements) show to be of
limited use for several reasons, namely: scaling issues for small scale model tests, numerical inaccuracies
and huge CPU time requirements for numerical methods and inexistence of proper sensors and risks
associated with the instrumentation at full scale. In addition to the difficulties in evaluating the hydrostructure interactions during the deterministic sloshing impacts conditions, the consistent design
methodology, which should combine the different deterministic results in order to calculate the overall
safety index is also missing. The overview of the different issues related to the direct assessment
methodology for sloshing was discussed in Malenica & Diebold (2013).
During the last 3 years lot of effort has been done for better understanding of this problem but
unfortunately, and in spite of all the efforts, we can say that, at least from direct assessment methodology
point of view, the problem still remains open and relatively simple design methods are exclusively in use
today. Very important activities related to better understanding of the hydro-structure interactions during
the sloshing impacts were initiated by Gaztransport & Technigaz (GTT) who owns the patent of the two
most important CCSs i.e. MarkIII and NO96. The main part of the results of these investigations was
presented at the different ISOPE Conferences. These investigations concern all the aspects of the problem
(experimental, numerical full-scale, methodology) and very good interaction of GTT with other research
groups allowed to make significant progress in this field.
2.4.1
Experimental approaches
There has been lot of work in the last years regarding the improvements of the small scale model tests
procedures (Baudin et al. 2012, 2013, Diebold & Baudin 2014, Kayal & Berthon 2013, Kimmoun et al.
2012, Mehl et al. 2013, 2014, Neugebauer et al. 2014, Song et al. 2013, Wei et al. 2012, 2014, Loysel
et al. 2012, 2013, Choi et al. 2012, Kim, S.Y. et al. 2013, Kim, Y. et al. 2013, Souto-Iglesias et al. 2011).
In particular a so called ISOPE sloshing benchmark study was organized by GTT and several institutions
participated (Loysel et al. 2012, 2013). The principle of these benchmarks was to compare the
measurements, especially impact pressures, from sloshing model tests performed by different laboratories
involving the same nominal input conditions which were chosen as simple and as controlled as possible.
In spite of the relatively simple experimental conditions (2D tank and simple excitations) the overall
conclusions of these comparisons are not very encouraging at least when the impact pressure
measurements are concerned. However, the results for more stable quantities (events rate, global forces,
etc.) shows relatively good agreement.
The sensitivity of the model test results to the type of sensors which is used was investigated by Ahn
et al. (2013), Baudin et al. (2012) and Razzak et al. (2013). The conclusions from these investigations are
rather pessimistic and large differences in measuring pressures are obtained using the different sensors.
Statistical post processing of the pressure measurements represents the challenge in itself. The effects
of the test duration and the sampling rate were investigated by many researchers (Bulian et al. 2012,
Dematteo & Ratouis 2013, Diebold et al. 2013, Fillon et al. 2012, 2013) and the conclusions are not very
encouraging concerning the evaluation of the maximum pressures neither concerning their time evolution.
It is important to understand that even if we would have acceptable pressure measurements we could
not use them directly in the design process because of the important scaling problems. Indeed, due to a
very complex physics involved during the sloshing impacts there is no single scaling factor but this
factor(s) depends on the type of impact and often the combined scaling should be used for single impact.
All this means that the difference between the full scale values of the pressure, obtained by applying
different scaling laws, may be significant. Different aspects of the scaling of the small scale model tests
were considered in numerous publications (Ahn et al. 2012, Karimi et al. 2013, 2014). The scaling issues
are investigated by means of performing the small scale model tests at different scales typically 1:10, 1:20
and 1:40. In addition, since the scaling does not depend on the scale only but also on the types of fluids
which are employed, some authors performed the model tests with the fluid and gas of different density.
In particular the care was taken to properly scale the density ratio in between the liquid and gaseous
phases since some theoretical investigations indicated the importance of this effect for pressure
measurements of certain impact types.
228
It is necessary to mention another physical effect which is extremely difficult to model at small scale
and which concerns the hydroelasticity. Indeed, when the typical temporal duration of the local impact
loading is comparable with natural periods of the surrounding structure hydroelasticity matters and can
significantly affect the structural response. Due to the chaotic nature of the sloshing loading, these
situations can occur very often so that hydroelasticity effects cannot be ignored. In Bardazzi et al. (2012a,
b) and Lugni et al. (2012, 2013) these effects were investigated both experimentally and numerically for
the flip through type of impact situation. One of the main conclusions of these investigations was that the
hydroelasticity plays a critical role in the evolution of the structural response and that the maximum
structural response occurs during the fully coupled hydroelastic interaction regime. This means that the
pressure measurements using the rigid tank structure, which is the common practice nowadays, cannot be
used directly for design and should be supplemented by the evaluation of the dynamic amplification of
the response due to hydroelasticity. In their conclusions the authors also indicates the huge numerical
difficulties to correctly reproduce the structural response numerically.
Important effort to classify the different impacts according to dominant physical characteristics was
made by GTT research team (Lafeber et al. 2012a) who introduced the different Elementary Loading
Processes (ELP), combination of which can eventually cover any type of impact conditions. One of the
interesting results from these investigations was the development of the improved Bagnold model for
impacts with the air cavity, which allowed for more realistic scaling of the associated pressures in the air
pocket (Ancellin et al. 2012, Brosset et al. 2013). These investigations were mainly applied to the quasi
full scale model tests by a so called Sloshel project (Lafeber et al. 2012b, Pasquier & Berthon 2012) and
their application to small scale model tests was not considered yet. This work is still in progress and no
conclusive results were obtained yet.
Finally it is also important to mention that the small scale sloshing measurements are performed using
the flat tank wall surfaces while in the real, full scale, situation the tank walls are not flat and contains the
corrugations in the case of MarkIII and raised edges for NO96. Knowing the sensitivity of the pressure
measurements to small local geometrical changes, we can easily imagine how difficult it would be to take
these effects into account when scaling the pressure.
2.4.2
Numerical modelling
All sort of numerical methods for sloshing simulations were proposed in the past: Boundary Element
Method (BEM), Constrained Interpolation Profile method (CIP), Finite Difference Method (FDM), Finite
Element Method (FEM), Finite Volume Method (FVM), Level-Set method (LS), Marker-and-Cell method
(MAC), Moving Particle Semi-implicit method (MPS), Smoothed Particle Hydrodynamics method
(SPH), Volume-of-Fluid method (VOF) and others. In spite of the recent improvements of these methods
the results are still far from satisfactory (Bai et al. 2013, Baeten 2012, Cao et al. 2013, Colicchio et al.
2013, Costes et al. 2014, De Chowdhury & Sannasiraj 2013, Fossa et al. 2012, Gazzola & Diebold 2013,
Hashimoto & Le Touz 2014, He & Kashiwagi 2014, Hwang et al. 2012, Hwang, J.O. et al. 2014, tibar
et al. 2012, Jeong et al. 2012, Lindberg et al. 2012, Luppes et al. 2013, Rafiee et al. 2013, Xue et al. 2014,
Zhang & Wan 2012, Zhang, Y. et al. 2014, Zheng et al. 2013, Zhu et al. 2014). This is not only because of
the prohibitive CPU time requirements but also because of the complexity of the physical phenomena
which are involved (violent free surface deformations, hydroelasticity, phase transition, compressibility,
3D effects, low temperature, etc.). However, it should be noted that these comments concern the
evaluation of the local pressures and the local structural responses and the modelling of the global
sloshing behaviour in terms of the quantities such as overall forces on the tank, global evolution of the
free surface, global and local kinetic energy, impact occurrences, etc. shows relatively good results for
almost all methods. This means that CFD can be safely used for the evaluation of the sloshing severity
and for the identification of the critical sloshing conditions in a qualitative manner. An interesting
example of using the OpenFOAM software was presented in Gazzola & Diebold (2013) where a simple
method was proposed to identify the severity and number of impacts for given tank motion.
The reasons for relatively poor quality of the CFD simulations, when evaluating the local impact
pressures are similar to those associated with the pressure measurements at small scale. Indeed, the
maximum impact pressures are usually extremely localized in space and in time which makes their
numerical evaluation or measurement extremely difficult. This is probably one of the main reasons for
very particular statistical properties of the measured and/or calculated impact pressures. Indeed the
quantities such as the global forces and even local fluid velocities show the typical statistical behaviour
with Weibull shape parameter greater than 1, while the local pressures have this parameter less than 1
leading to very extreme long term values. In addition the associated confidence interval for pressures is
much larger which makes the direct use of the measured pressures for design purposes almost impossible
(Diebold et al. 2013).
229
Due to the difficulties associated with the simultaneous direct evaluation of the local pressures together
with the global sloshing behaviour, the actual research seems to be more oriented to a kind of hybrid
approach where the problem is subdivided into global and local parts. Within this approach the
representative impact conditions are identified by either the model tests or CFD calculations, and they are
subsequently simplified and treated using the dedicated local hydro-structure interaction models. This
idea was first introduced by Korobkin & Malenica (2006) and very simple impact situations were
proposed together with the corresponding semi-analytical hydro-structure interaction models. Recently
this approach was expanded to the more realistic impact situations and the combined potential flow and
CFD approaches were proposed for local hydro-structure interactions. In this context an important work
was initiated by GTT and the coupling of the nonlinear potential flow model for wave propagation with
the complex CFD model, based either on SPH or VOF technique, for impact simulations, are under
development (Guilcher et al. 2012, 2014, Costes et al. 2013, Scolan et al. 2014).
2.4.3
CCS structural response
In addition to the complex hydrodynamic loading which occurs during the sloshing impacts a complex
structure of CCS introduce the additional huge difficulties in the evaluation of the structural response.
Indeed, the CCS is composed of the plywood, foam, perlite, special steel, triplex, invar, F and is attached
to the hull structure by the resin rope and special couplers in the case of NO96. Several publications
addressed this problem during the last 3 years, Chun et al. (2013), Dobashi & Usami (2012), Hwang, S.Y.
et al. (2013a, b, 2014), Jeong & Yang (2013, 2014), Hwang, S.Y. et al. (2014), 2014, Kim, D.H. et al.
2014, Kim, J.M. et al. (2013), Kim & Paik (2013), Lee, D.J. et al. (2014), Lee & Zhao (2013), Lee, S.G.
et al. (2012), Nho et al. (2012), Paik et al. (2014), Wang, B. et al. (2012), but it seems that there is still no
fully reliable numerical model and we can say that the numerical modelling of the CCS structure still
remains big challenge which makes the solution of the fully coupled hydro-structure interaction model
almost impossible for the time being.
2.4.4
Current approaches for sloshing assessment
Due to the technical difficulties discussed above it is clear that the direct calculation approach for
sloshing assessment is still not ready. That is why the current procedures of Classification Societies for
sloshing assessment are manly based on the so called comparative approach (e.g. see Bureau Veritas
2011, Det Norske Veritas 2014). Philosophy of this approach is relatively simple and consists in
comparing the loading and capacity of the new design with the reference ship which has never sustained
damages due to sloshing impact loads. The capacity of the containment system is evaluated using the
nonlinear dynamic finite element model for different loading scenarios leading to the identified failure
modes (bending of the cover plate, buckling of the vertical bulkhead, residual compression of the foam,
etc.). As already indicated, CCS is a very complex structure composed of different materials connected
together by special procedures and the representativity of the classical finite element models need to be
checked very carefully. The loading scenarios are chosen in such a way to cover the typical loading
patterns observed in small scale model tests. This means that the different combinations of pressure
amplitudes, time histories and loaded surfaces have to be considered. Once the capacity curve identified
the last step is the determination of the scaling factor. The reference vessel is used for that purpose. The
scaling factor is chosen so that the scaled design load curve of the reference vessel is tangent to the
corresponding capacity curve. Finally, to assess the target vessel, the small scale design loads are scaled
by the same factor obtained previously and by the safety factor, and are compared to the capacity of the
target vessel.
2.5
2.5.1
Air blast
Previous ISSC committee V.I report (2012) provided a benchmark study for shock response of plate panel
to blast load. Figure 1 shows a typical pressure pulse curve; in which tm represents rising time, tdur is the
duration time and tterm is the termination time of the timepressure history curve. The idealized blast load
with triangle shape can be applied to the shock response analysis of structures.
Longre et al. (2013) dealt with the numerical simulation of the dynamic failure of a ship structure steel
plate under near-field air-blast loading. Various energetic levels of air-blast loading, involving variable
explosive mass and charge-plate distance were tested. The results show that the equivalent plastic strain
in the highly deformed area can reach 1.3, the maximum equivalent plastic strain rate 3700 s1 and the
temperature value 470C. These values are greater than the maximum values of strain rate and
temperature considered for the material thermo mechanical characterization.
230
Figure 1. Dynamic pressure pulse generated by the PPLR and idealized pressure (Sohn et al. 2013)
Alireza & Ahmad (2014) investigated the dynamic response of corroded plates subjected to blast loads.
The results show that a reduction in the dynamic load carrying capacity of 8% can occur due to corrosion
in plates with 50% degree of pitting. Corrosion pattern has significant effect on reduction of load carrying
capacity. Maximum reduction occurs when pits are concentrated at the center of plate.
Internal blast
Several major international design methods of Explosion Containment Vessels (ECVs) refer to the items
of pressure vessel codes and standards, where the fracture mechanics analysis of pressurized components
should be performed to prevent the occurrence of brittle fracture. Ma et al. (2013) proposed a ratedependent failure criterion to account for Adiabatic Shear Band (ASB) propagation, and a finite element
analysis of a cylindrical containment vessel with different size of cracks was performed. The failure
assessments based on ASB mode and Failure Assessment Diagram (FAD) method were conducted,
respectively. The results show that the strain rate on representative point is not influenced by local crack
size, which is only related with load conditions. The assessment result based on FAD method is in
agreement with that of ASB at low or intermediate strain-rate (lower than 85 s1). When strain rate
exceeds 85 s1, the assessment results based on two methods became different. For single-used ECV,
where plastic deformation is allowable, FAD assessment may lead to conservative result, thus assessment
based on ASB damage mode is recommended to give more comprehensive estimations. For the multiused ECV, the two methods incline to a consistent result.
Lee, S.G. et al. (2014) developed a shock analysis technique of blast hardened bulkhead under internal
blast using multi-material arbitrary Lagrangian Eulerian formulation and fluid-structure interaction
technique. Shock response analysis for a part chamber model was carried out under internal blast. The
calculated pressure and acceleration response, as well as damage and deformation configuration of
bulkhead have good agreement with test results.
Weapons attack on ship structures
The damage effect of weapons on ship structures mainly comes from the blast wave and fragments. When
stiffened plates subjected to the loading of blast wave and fragments, it can be easily penetrated by
fragments. The perforations on the stiffened plate act as the crack initiation locations due to the stress
concentrations. Soon afterwards, blast wave in confined space would aggravate the deformation of the
weakened stiffened plates and large vent appears. Blast wave propagates through the vent and exerts on
adjacent cabins. Kong et al. (2014) conducted experimental investigation and numerical simulations of a
cased charge exploded inside a multi-layer protective structure, the synergistic effect of blast and
fragment loadings were considered. The liquid cabin plays an important role in the enhancement of the
anti-explosion capacity of the multi-layer protective structure. The liquid in the cabin can effectively
absorb the energy of high speed fragments and prevent them from punching into internal cabins.
2.5.2
Underwater explosion
In the UNDerwater EXplosion (UNDEX), various combination of charge weight (W) and standoff
distance (R), can generate various pressure versus time curves and its related energy density (energy pass
through unit area) of shock wave. For a far field UNDEX, the shock wave can be taken as plan wave, the
energy flux is proportional to W/R2. And the shock factor ( SF = C W R ) is widely used to study the
shock resistance of a hull plate and also used as attack severity index. If the standoff is smaller than
the dimensions of target, e.g. ship length, the spherical wave effect should be taken into consideration. If
the standoff smaller than twice the radius of the explosion gas bubble, the UNDEX are referred to as
close-proximity explosions. At these ranges, the dynamics of the gas bubble are influenced by the
231
geometry of the target (in the case of a rigid target), and possibly also by the deformation of the target (in
the case of a responding target). Riley et al. (2012) describes the experimental results and Chinook (an
Eulerian computational fluid dynamics code). The predictions were compared with analytical and
similitude-based models of bubble growth. It was also shown that a hemispherical close-proximity
analytical model was found to predict the first bubble period for initial standoffs within one bubble radius
somewhat better than the free field models.
Response of ship hull subjected to close-in UNDEX
The bubble load in a close-in UNDEX can cause the ship hull global response and local response. For
global response analysis the ship hull can be simplified as a hull girder model, for the local response the
3D model of surface hull is needed. The vertical response of ship hull subjected to an UNDEX bubble is
mainly the global response, which is composed of rigid body motion and elastic deformation. Large local
vertical response appears at some locations when the ship model is subjected to the bubble load.
Zhang & Zong (2012) investigated the elastic and plastic response, resonance mechanism in hull girder
whipping response to an underwater bubble. The results show that the hull girder sustains significant elastic
response of low-order mode. When the bubble load is large enough, the bending moment in the mid-ship
may exceed the ultimate bending moment. And a single plastic hinge will form at center of hull girder.
Zhang, N. et al. (2014) used a procedure which couples the FEM with Doubly Asymptotic
Approximation (DAA) method to study the transient response of ship hull structures subjected to an
UNDEX bubble. The numerical results show that besides global whipping response, the ship hull also
sustains severe local response in different directions subjected to UNDEX bubble jetting. Severe local
transverse and longitudinal response also exist at some special location of the ship model. Large local
transverse responses take place at side plates.
Zong et al. (2013) analysed the impact response of full-scale surface ship (Length Breadth Draft =
55 m 9.5 m 2.5 m) subjected to non-contact close-in UNDEX. Three damage modes of whole ship are
specified according to the standoff distance, shown in Figure 2. For an explosive with equivalent 400 kg
TNT, the responses of the analysed ship under different standoff (R) are:
a) R > 10 m: damage mode I: the hull plate suffer dishing damage. This is familiar with far field
UNDEX. Especially when the deformation appears at shell plate and superstructures.
b) 6.5 m < R < 10 m: damage mode II: Buckled-frame deformation.
c) R < 6.5 m: damage mode III: bottom-indented mode. The part of hull close to the charge indent
rapidity when the shock hit the ship. Damage takes place on the local parts of ship, local crack and
fracture will appear.
Wang, H. et al. (2014) investigated the dynamic response of a ship-type box girder model (Length
Breadth Depth Draft = 2.8 m 0.8 0.3 m 0.04 m) subjected to close-in non- contact UNDEX with
experimental test and numerical analysis. The results reveal that global longitudinal strength collapse
combined with the local wrinkling, the plastic deformation of hull between bulkhead and plastic
deformation of local bottom hull are the mainly three damage modes.
Figure 2. Damage mode of ship hull subjected to close-in UNDEX (Zong et al. 2013)
232
If the UNDEX bubble effects are not incorporated in the analysis, the damage severity will be
underestimated. As the pulsation periods are close to the two-node bending vibration of test model, the
resonant result cause large amplitude whipping displacement. The ratio R/Rmax (standoff to maximum
radium of gas bubble) has important effects on the response and damage modes of structures. Experimental
data and numerical analysis reveal the structural collapse modes are different when R/Rmax increases.
Shock resistance
When naval ship is attacked by an UNDEX, the ship can be severely damaged by shock waves and gas
bubble pulse; such an attack can put the crew in danger and possibly destroy the ship. Preventing damage
to ships is of great interest in naval ship design. Elastomers, or rubber-like materials, and core sandwich
structure are often used to mitigate damage caused by impulsive or impact loads because of their low
stiffness and high damping characteristics.
Kim & Shin (2013) investigated the rubber coated plate and sandwich structure with core. The results
indicate that the rubber coating effectively mitigates plate stress by impact loading in an elastic regime
and, in a plastic regime, the rubber seemed not to be effective. The sandwich structure is useful in an
impact environment. The highest priority factor to consider in constructing a sandwich structure was the
number of cells. Core height and core thickness were secondary considerations.
Xiao et al. (2014) investigated the rubber protective coatings with different structures under
compression load and water blast shock wave. It is concluded that when under dynamic compressive load,
the cell topology played an important role at high speed, and when under UNDEX, the honeycomb
coatings can improve the shock resistance significantly at the initial stage. Although structural absorbed
energy has a significant contribution in the shock resistance, soft coating can significantly reduce the total
incident impulse at the initial fluid-structure interaction stage.
Schiffer & Tagarielli (2014b) performed a laboratory-scale Fluid-Structure Interaction (FSI)
experiments and finite element simulations to examine the one-dimensional blast response of doublewalled hulls, consisting of two skins sandwiching a layer of water. It is found that for the outer skin,
cavitation processes initiate in the water very close to the front face sheet of the sandwich hull and result
in a dramatic reduction of the impulse imparted to the inner hull (60% compared to the case of a
monolithic outer skin). Schiffer & Tagarielli (2014a) have constructed and validated theoretical models
for the dynamic deflection of fully clamped, circular elastic composite plates loaded by planar,
exponentially decaying underwater shock waves. Design charts are constructed and used to determine
plate designs which maximize the resistance to underwater blast for a given mass.
Avachat & Zhou (2012) analysed the response of sandwich composite structures to underwater blast
loading. Results reveal a significant difference between the responses of air-backed and waterbacked/submerged structures. In general, thick and low-density cores provide superior blast mitigation
and failure resistance.
Shallow water bottom contour effects
In most deep water UNDEX events, the contour of the ocean bottom is trivial as the bottom reflection of
the shockwave is of minimal magnitude. This assumption may not be true for littoral waters. Walters
et al. (2013) used Lagrangian solid bottom modelling approach to compare with the current bottom
modelling technique to determine its validity and potential benefits. Five different bottom contours and
one flat bottom model were simulated with a Floating Shock Platform (FSP) serving as the ship model. A
parametric study was conducted to determine the effect that contoured bottom profiles on the response of
a ship subjected to an UNDEX event in littoral waters. The initial analysis of the FSP response showed
only slight differences between the various contour models. This was caused by the buffer created by the
bulk cavitation zone. The effect was specific only to the particular geometry selection. Modifications of
the charge size, target separation, or bottom depth could diminish the effect, but to what extent needs is
left for future study.
2.6
With increases in ship size and speed, shipboard vibration becomes a significant concern in the design
and construction of vessels. In the theoretical prediction and numerical simulation of dynamic ship
responses, the damping characteristics of ships and their surrounding fluid are key issue and major
reason for the inaccuracies encountered. Various methods are available for modelling ship vibration
damping.
The total damping associated with overall ship hull structure vibration is generally considered as a
combination of the following components:
Structural damping;
Cargo damping;
233
Water friction;
Pressure wave generation;
Surface wave generation.
For the forced vibration analysis, it is assumed that the effects due to structural damping, cargo
damping, water friction and pressure wave generation can be lumped together. The effect of surface wave
generation needs only to be considered for very low frequencies of vibration. This effect is generally
neglected. For simplification, a constant damping coefficient of 1.5 percent of the critical damping is
suggested by ABS Guidance Notes on Ship Vibration (updated in 2014) for the entire range of propeller
rpm and main engine orders. Otherwise, more detail frequency-dependent damping coefficients may be
used, if applicable. In the forced vibration analysis of ship structures by Yucel & Arpaci (2013), the total
damping associated with overall ship hull structure vibration was considered as a combination of the
several damping components. The propeller-induced forced vibration responses of local ship structures
were determined using finite element method.
For hydrodynamic hull damping, Kim & Park (2014) used a random decrement technique together with
continuous wavelet transform to estimate the wet damping of a segmented hull model. The 16 sea states
were grouped together based on the speed of the ship to determine the possible influence of the ship speed
on the damping ratio. The wet damping ratios for each sea state group, as well as precise wet natural
frequencies, were estimated using a continuous wavelet transform.
Based on Statistical Energy Analysis method, the vibro-acoustic characteristics of a composite cabin
segment are investigated by Pang et al. (2014), in which the influence of different constraint damping
structure parameters and laying positions on the ship vibration radiation are analysed.
Various countermeasures for vibration damping and control have been reported. Vibration isolation
using passive isolators is widely used in marine applications through different configurations, such as
single-stage, two-stage and floating raft isolation system. By using hybrid genetic algorithm, modelling
and optimization of floating raft systems in submarines under different objectives was reported by Huang
et al. (2012), in which different vibro-acoustic objectives were considered and vibro-acoustic behaviour
of the fluid-loaded cylindrical structure is investigated. Applying the power flow mode theory, Xiong
(2014) recently proposed a new Power Flow Mode Dynamic Topology Optimization (PFMTOP)
approach to topologically optimise systems damping material distributions achieving enhanced vibration
suppression capability. This approach can maximize the energy dissipation for a given volume of the
material to achieve minimum power flow response. Kang et al. (2012) used structural vibration at
specified positions level as design objective for topology optimization by distributing a given amount of
damping material. Teng et al. (2014) proposed the method of determining the adhesion position of the
damping material, which is applicable to the vibration damping of ship plate based on the Bi-directional
Evolutionary Structural Optimization (BESO). In this method, the needed amount of damping material is
taken as the constraint condition, and the maximization of one natural vibration frequency of the structure
is taken as the target function.
Low-frequency passive vibration isolation is challenging and nonlinear power flow behaviour is less
understood. Yang et al. (2012) investigated a nonlinear isolation system with a Negative Stiffness
Mechanism (NSM) using nonlinear power flow approach because of the traditional force or displacement
transmissibility is not valid in nonlinear domain. Adding NSM can greatly enlarge the frequency band for
effective vibration isolation. By connecting a negative stiffness corrector to a linear isolator, a HighStatic-Low-Dynamic Stiffness (HSLDS) isolator can be formed and investigated by Huang et al. (2014).
The HSLDS isolator with quasi-zero stiffness characteristics can offer the lowest resonance frequency
provided that there is only stiffness or load imperfection.
Regarding to active vibration control, experimental study for an active floating raft vibration isolation
system was conducted based on a flexible hull structure by Zhou et al. (2013). An analytical study of
active structural acoustic control of an elastic cylindrical shell coupled to a two-stage vibration isolation
system is reported by Ma et al. (2014) to minimize the vibratory power transmitted to the foundation and
the acoustic power radiated from the supporting shell. Ahuja & Gupta (2014) studied a simplified semiactive floating raft vibration isolation system with the objective of mitigating the acoustic signature of a
warship by minimizing the transmission of forces, resulting from operation of on board machinery, to the
foundation.
It is worth noting that a newer type of smart Magneto-Rheological Elastomer (MRE) materials have
attracted much attention in recent year due to their stiffness and damping properties can be changed
instantly responding to applied external magnetic field. Research on MRE smart materials for adaptive
vibration control has attracted EU FP7 funding-ADAM4EVE project to develop such adaptive materials
and structures for vibration damping. Application of MREs to a two-stage vibration isolation system was
examined by Zhu et al. (2012) with the developed non-linear mathematical model of MREs for damping
and stiffness based on experimental characterizations.
234
2.7
Monitoring
2.7.1
Hull structural monitoring system
The technology for Hull Structural Monitoring System (HSMS) for ships and offshore structures is
rapidly growing, in the wake of the advancements in new sensors, large data handling and wireless
communication networks. Results of long monitoring periods of ship stress more frequently appear in the
literature. The most relevant and newest results are presented later on in this section.
Systems configurations range from the well consolidated systems to monitor hull girder bending moment,
to complex experimental systems based on fiber-optic sensors and wireless data transmission, used to
investigate dynamic responses to local loads. The general high level architecture of a modern HSMS system
acquires heterogeneous signals coming from various sensors distributed in the ship and processes the signals
in real time by local processing unit. Then each Peripheral (local) Processing Unit (PPU) communicates data
via a wired (Ethernet) network to the Main Processing Unit (MPU) which also acquires other types of
signals, coming from wave radar, ship speed log, etc. and processes them all in real time.
The current trend is to substitute the wired copper-based network communication between the PPU and
the MPU with wireless or fiber-optic cables. Prospected future developments in the area of HSMS by
Cusano & La Marca (2014), pursued by others, aim to transform the interaction between the user onboard
and the HSMS system from a passive experience, i.e. the user decides to read and analyse the (real-time)
results of HSMS measurements, to an active role of the HSMS which will be able to supply active
guidance to the board to minimize the predicted ship structural response based on the real time sensors
data. The new systems would be able to incorporate virtual sensors, i.e. information on stresses of
structural details obtained by real time numerical prediction.
As often in naval architecture, two different fields of application are generally discerned: merchant
ships and naval vessels. The HSMS systems used in the navy vessels often capitalize on systems
developed in the merchant ship world, enhancing them with modification. The research in the two fields
is driven by several motivations which are shared at different level of importance by the two market
sectors: improved safety, reduced life costs and better characterization of hull loads and remaining fatigue
life aimed a more effective strategic fleet planning.
A recent report from Department of Defence (2013), presents the architecture and main components of
a modern HSMS with particular attention to configurations and features of interest for application in navy
ships. After a general introduction on the static and dynamic loads acting on a hull structure, they reported
a detailed panoramic on a large series of HSMS commercially available for installation on merchant ships
and navy vessels. The list of available systems is also reported with their main specifications including
integrated sensors types and number.
2.7.2
New sensors technology and application
This paragraph is dedicated to the review of the state of the art and recent developments in sensors
technology and systems for data transmission and acquisition on board of ships. Few references to other
application sectors which may have a relevant impact on the development in the shipping world were also
included. Knowledge of other systems used for monitoring other types of structures can bring inspiration
and accelerate the development also in the maritime world. Interesting indications can come from the
aeronautical engineering community. A good review of the state of the art in this specialized field
complemented by interesting propositions of new technologies can be found in Brigman (2012), who
analysed different systems: from those based on vibration based monitoring, to fiber optic sensors, and high
frequency wave propagation techniques including acoustic emission, ultrasonic, Lamb waves, piezoelectric
and Micro-Electro-Mechanical System (MEMS) actuator/sensors. The research study of Brigman terminates
with a critical discussion on the challenges and restrictions facing implementation of proposed structural
health monitoring in commercial aviation (as opposed to military), some of these issues and remedies are
shared by the maritime community which introduces additional challenges, such as the much harsher and
stochastic environment and the complexity of structural detail and diversity of environmental loads.
Again, a good review of the state of the art of sensors technology for marine applications can be found
in the report of Department of Defence (2013). Strain gauges confirm to be the primary sensor utilized by
HSMS and these are broadly separated in Long Base Strain Gauges (LBSGs) and Short Base Strain
Gauges (SBSGs). New types of strain sensors are based on fiber optic technology. Table 1, extracted from
their report, gives a list of various sensors used in modern HSMS systems and cross correlate them with
parameters of interest and type of derived information that can be obtained.
Optic-based sensors
Optical-based sensors for monitoring the dynamic behaviour of structures are gaining popularity in this
period as they offer a series of advantages over conventional copper-wired based sensors. First the
235
electro-magnetic compatibility: sensors and cables do not induce Electro-Magnetic Field (EMF) and
uninfluenced by any EMF. This is particularly important to avoid interference with other wired-based
vital systems of the ship, such as the ship automation system or combat system in military ships.
Secondly they are intrinsically safe for potential application in explosive environments, avoiding the need
for Zener barriers required by copper based sensors. Additionally this type of sensors is light, can be
easily embedded into composite structures and they achieve high gains in noise to signal ratio in
transmission and do not suffer from any drift which typically affects copper wired system.
Qiu et al. (2013) reviewed the latest research of the Fiber Bragg Grating (FBG)-based Structural Health
Monitoring (SHM) technique for composite structures in various fields of applications, including ocean
vehicles and offshore structures but also civil engineering, aviation and aerospace. The rapid development
of the network promotes the expansion of the optical fiber telecommunication industry, which has
substantially driven down the cost of optical components, making optical fiber sensors, particularly FBG
sensors, more economically viable. The installation issues and standards of optic sensors for various types
of measurements, including structure monitoring, are well addressed in the NATO technical report RTO
AG160 (2012).
Wireless sensors
Wireless data transmission is also a recent development for HSMS, developed first for monitoring large
number of hotspots in static civil structures (Lynch & Loh 2006, Swartz et al. 2012) report on a prototype
hybrid (wireless/fiber optic) system installed on the high speed littoral combat vessel FSF1 Sea Fighter,
an aluminum catamaran and it is cited as an option for the Hull Monitoring System (HMS). The system is
a hybrid, where strain monitoring is provided through Narada wireless data acquisition and control units
linked to a ship-board fiber-optic data network. The Narada units are low-cost wireless devices developed
at the University of Michigan for the routine monitoring of large, complex structures such as buildings
and bridges.
Table 1. Sensors typology and support information as a function of the monitoring parameter in modern Hull
Structural Monitoring Systems (Department of Defence 2013).
Monitoring Parameter
Cargo loading
Hull Girder Vertical
Bending
Hull Girder Lateral
Bending
Hull Girder Torsion
Temperature
Slamming
Green Seas over Bow
Whipping
Hot-spot stress
Cross deck structures stress
Side frame loads
Hatch Corners stress
Sensors
Main deck LBSGs or
SBSGs, port & starboard
Bilge keel strain gauges

Themocouple or fiberoptic
temperature sensor
Accelerometers, pressure
transducer
High speed data
acquisition
Local strain gauges
(SBSG)
Wet deck slamming

(multihull)
Local accelerometers and

strain gauges (SBSG)
Navigation data
Ship systems
Wave Environment
Wave radar
Motions sensors
Supporting Information
Section properties
Operational profile
Structural analysis
Post-voyage analysis
Limit state analysis
(FEA)
Thermal coefficients
Structural analysis (FEA)
Sea keeping analysis

Weather forecasts
Launch and recovery safe
to operate limits
Potential Outcomes
Overload alarms
Hull girder load history
Hull girder fatigue
indicator
Improved overload alarm
(buckling risk) and hull
girder fatigue
Temperature
compensation
Thermal history
Slam avoidance warning
and event history
Improved hull girder
fatigue indicator
Local stress / overload
monitoring
Detailed structural load
history (e.g. multihull
spreading loads) and
fatigue indicator
Wet deck slam warning
and slam event history
Improved fatigue
indicator
Weather routing
Economic routing
Operator guidance for offboard systems
Improved routing
Improved load / fatigue
assessment
236
For the hybrid prototype system, a total of 20 Narada wireless units were installed throughout the ship
and interfaced to single or tri-axial accelerometers and foil strain gauges; a total of 28 sensor channels (8
strain, 20 acceleration) were added. These are sampled at high rates of 100 to 1000 Hz to capture
slamming events and the wireless monitoring system had to be divided into three separate networks to
avoid wireless bandwidth issues. Result of the wi-fi sensors installation on the vessel were quite
successful and efficient: the data lost by wireless communication were few and moreover compensated by
the greater advantage of reduced time and cost of installation with respect to wired systems.
As reported by Department of Defence (2013) , in spite of its rapid expansion in different engineering
applications, wireless data transmission is still not considered a proven technology for navy application as
the long term reliability of the wireless data acquisition and control units in this environment has not been
established. Bennett et al. (2014a) document the application of multiple wi-fi sensors with synchronized
video for measuring the loads and motions of a multi-segmented ship model tested in a wave tank versus
traditional wired sensors. The same accuracy of results is achieved, but the practical installation
advantages and flexibility make the wi-fi systems very attractive for future full-scale installations.
Interesting opportunities will come in the future from the combination of MEMS technology with
wireless data transmission. One of the first applications has been recently documented by Xiong et al.
(2013) and it is related to the measurement of atmospheric pressure in harsh environment (high
temperature pressure vessels). The use of various types of MEMS sensors is expected to expand in the
future also on board of ships. A singular example of recent application on board of a fishing vessel is
shortly described by Chen et al. (2013).
Acoustic emission sensors
The use of Acoustic Emission Testing (AET) systems targeted at potential fatigue hot spots areas has
been recently suggested as a viable approach for active detection of structural failures. The principle
behind these sensors is the measurement of the acoustic energy emission released during plastic
deformations of metals, providing a method to detect crack and monitor fatigue damage also in marine
structures (Rogers & Carlton 2010). These sensors have been successfully used to detect flaws in offshore
structures (Ternowchek 2012, Wang et al. 2010, Anastasopoulos et al. 2009) and are expected to rapidly
expand in ship HSMS, especially if combined with fiber-optic multiplexed data transmission or wireless
data transmission (to avoid the large number of wire cables otherwise needed).
Lee, A.K. et al. (2014) report results obtained from a joint development project between ABS,
MISTRAS and a global container transportation company. This pilot project positively verified the
viability of acoustic emission technology as a screening tool for surveys and inspection planning.
Specifically, in-service AET data were collected from container ships during voyages through the Pacific
and Atlantic Oceans and provided useful information on the more critical structural details of the ship.
The paper also establishes a standard AET procedure, as 1) test plan, 2) AET system installation and
checks, 3) data acquisition and analysis, 4) reporting documentation, and 5) follow-up inspection. A
problem encountered by the authors during the monitoring operation is the filtering of the background
noise; the identification and elimination of the noise made by water slamming was particularly difficult
and in some cases impossible to eliminate.
Rogers & Stambaugh (2014) describe the application of in-service Acoustic Emission (AE) and strain
monitoring for locating stable propagating fatigue cracks in ship hull structures. A new fracture
mechanics approach to fatigue damage assessment and crack life prediction is introduced as the basis for
the interpretation of results. Provisional results for potential fatigue sensitive structural details are
reported as obtained during a measuring campaign performed on board of the USCG Bertholf cutter. The
authors underlie the physics of stable fatigue crack growth in metals and the acoustic emission produced
by the associated micro-fracture events. New numerical models to simulate the fatigue crack growth and
the associated acoustic emission, incorporating the latest developments in our understanding of the
mechanics of fracture on an atomic scale, are described. Examples are given of in-service global AE
surveillance of ship hull structural details. When combined with strain monitoring at suitable locations in
the same structural area, the estimated crack growth from the AE measurements can be modelled as a
function of the strain energy input to the platform (cumulative cyclic loading) to predict the crack
propagation life.
2.7.3
New full scale monitoring campaigns and related studies
Applications of HSMS on board of merchant ships have increased and studies of long operating periods
offer interesting insights in particular about the relevance of high frequency loads on fatigue life of ship
structures and on the response of the ship hull girder to different environmental conditions in general.
Often, results derived from sea measurement campaigns, are compared with theoretical predictions based
on different level of fidelity, ranging from the rules empirical formulations on the lower end, to
237
seakeeping predictions including slamming induced whipping on the other end. In exceptional cases of
larger research projects, grouping classification societies, ship hydromechanics labs and industry, model
scale test are also performed and results obtained in model scale are verified with the measurements taken
by the HSMS at sea.
Frangopol et al. (2013) propose an integrated life-cycle framework to address maintenance of
reliability of naval ships structures, in which they envisage the use of structural health monitoring to
provide a powerful and necessary mechanism to reduce uncertainty, calibrate, and improve structural
assessment and performance prediction models. Various applications are worked out with their method
integrated with HSMS data, among which a US Navy high speed catamaran and a tanker. Their analysis
framework is aimed ultimately to serve the US Navy to optimize inspection planning.
Storhaug & Hareide (2013) report an interesting analysis of a three year long measuring campaign of
deck strains acquired on a blunt merchant ship. The characteristics of the ship have not been released
for confidential reason, not even the size, but possibly the so called blunt ship which has been
measured in ballast and cargo conditions should be generally representative of tankers or bulk carriers.
The ship was recently designed and built for a longer target life than usual in unrestricted service and
consequently strengthened beyond the minimum industry standards. A basic HSMS system compliant
with DNV hull monitoring rules was installed onboard to collect and process data from: GPS for
position, course and speed; gyro for ship heading; wind sensor for wind speed and heading; two strains
sensors on the deck amidships, one on port, one on starboard side. Data were collected in a period of
three years. The measured fatigue life based on a stress concentration factor of 2.0 has been estimated
to be well below the design life. It is found that whipping and springing loads contribution to the
fatigue life is about 42%, highlight the importance of these high frequencies dynamic responses for the
lifecycle design of blunt ships. No particular difference was noted in between the cargo and ballast
conditions, in terms of vibration contribution to the total fatigue damage. The stresses measured during
three extreme weather events encountered by the ship (in Northern Pacific) have been generally higher
than the IACS rule wave bending stress level. In the worst storm, actually the structure of the vessel
could have collapsed without the extra-margins on strength used for the design to allow for the
unknown whipping contribution.
The conclusions of this study in terms on the effect of whipping and springing on the fatigue life of
ships are completely opposite to those drawn in the monitoring study of USCG cutters, later on described.
The reasons have to do principally with the different extreme weather conditions encountered by the two
vessels during their operation, but also to the different hull shape (very slender with deep-V sections at
bow as opposed to blunt hull with presumably flat bottom also at the bow) as it is expected to have a
considerable influence on slamming events and loads.
These conclusions are confirmed by another analysis study of HSMS measurements made by Storhaug
et al. (2013) on a LNG carrier. Although the considered ship is less blunt than the previous one (with a
slenderness midway between tankers and container carriers), the relative importance of whipping and
springing on the total accumulated fatigue damage is still around 40% (32% in ballast and 45% in cargo
conditions). The analysed data refer to a period of 5 years of unrestricted service on worldwide trade
routes (40% time in the North Atlantic) and information from wind (simple anemometer) and waves
(radar) was also included. As for the previous vessel, head and bow quartering seas are responsible for the
most part of the fatigue damage. The measured stresses in Beaufort 6, corresponding to sea state 3-4,
contribute the most to the fatigue damage. The information about the encountered sea states permit the
comparison between numerical and measured wave damage which in this case is surprisingly good: only
3% deviation. This confirms that the measurement of the incoming sea state is very useful information for
through life cycle and maintenance planning.
Kahl et al. (2013) report the results of a monitoring campaign performed on two container carriers: a
4600 TEU Panamax (L = 275m, B = 32.25m, T = 12m, CB = 0.672, V = 23 knots) and a post-Panamax
(L = 356m, B = 51.2, T = 13.5m, V = 26 knots). The period of observation is shorter, corresponding to a
single round the world route. Weather conditions encountered by the Panamax included severe weather,
while the post-Panamax sailed in milder sea states. For this reason prediction of the fatigue life of ship
structure was also repeated superimposing the high frequency load contribution derived from the
measurement on the long term linear seakeeping predictions (frequency domain, 3D panel method) of
rigid body motions and hull bending moment in waves.
On both ships, global loads were monitored by strain gauges placed on three different stations along
the ship. Also, additional strain gauges were arranged in the same sections in such a way to enable the
separation of vertical and horizontal bending moment and torsion. The strain data were complemented by
ship motion measurements obtained from an inertial platform installed in the deck house and
accelerometers at the bow and stern and by measurement of the directional sea state via an X-band radar
scanner mounted on the foremast.
238
Also in this case, by means of spectral analysis of the monitored stresses and a standard rain flow
counting method, the accumulated fatigue damage was quantified from the data collected on the two
ships. Low pass filters were applied to distinguish the effect of high frequency loads (whipping and
springing) on the overall fatigue accumulated damage. On the Panamax ship, the measured highfrequency loads contributed 32% to the total fatigue damage on the route in the North Atlantic and 41%
on the PAX route that excluded the North Atlantic and the North Pacific. To represent typical world-wide
service routes, these measured high-frequency contributions of the Panamax ship were utilized to
extrapolate the high-frequency response of the post-Panamax ship operating in severe seaways. For the
PAX route, this resulted in a high-frequency contribution of 51%. Indeed this estimate of the quantitative
contribution of high frequency response to the total fatigue budget of the ship obtained by Kahl et al.
(2013) is significant and in the same order of magnitude of those found by Storhaug & Hareide (2013) for
the blunt ship or by Storhaug et al. (2013) for the LNG carrier. The fatigue life prediction of the two ships
was estimated by an empirical method based on the linear seakeeping predictions based on zero speed
green functions with forward speed corrections (Papanikolaou & Schellin 1992). According to this
method, the high frequency hull girder response contribution derived from full-scale measurements is
added on top of the rigid moment to derive the fatigue life for the total long term response on the basis of
a GWS (Global Wave Statistics) annual directional scatter diagrams relative to the areas touched by the
ship route weighted with their probability of occurrence over the whole ship life. According to this semiempirical method, for the Panamax ship the contribution of high-frequency loads was found to go down
to 35% of total damage (instead of 41% straightly derived from the relatively short time measurements at
sea), while for the post-Panamax ship the contribution goes up to 57% (instead of the 51%).
The long term seakeeping predictions on wave scatter diagrams have been used also to predict the
maximum wave bending moment encountered by both ships in their lives. In both cases the predicted
values exceed the envelope curve representing the average rule-based value of the hogging and sagging
wave-induced VBM (Germanischer Lloyd 2012) applied for fatigue strength checks. For the Panamax
ship the rule curve is exceeded by 30%; for the post-Panamax ship, by 20%.
At this point, it is worth noting that there is neither common recognized procedure nor even a
recommendation at regulatory level to include high frequency dynamic response loads on the fatigue life
prediction of ships for design. The research in this field is still trying to quantify the uncertainties level
behind different theoretical or semi-empirical procedures, i.e. different seakeeping predictions methods,
different sea state statistical characterizations, different hull structural models, etc.
The panoramic given on recent studies about HSMS application and data elaboration ends on fast
military vessels. In this case, the first results of large campaign monitoring structural response initiated by
the United States Coast Guard (USCG) have been recently presented at the Symposium of the Ship
Structural Committee. Some years ago, USCG initiated a project to assess fatigue design approaches for
its new National Security Cutters (NSC). The results reported by Drummen et al. (2014) have been
obtained from the long term monitoring campaign on NSC USCGC Bertholf: a 120 meters long, 4500 ton
full load, cutter. The unit was instrumented with a rich combination of sensors and tested along almost 9
months. Sensors included: 24 LBSGs whose locations on the ship is given in Figure 3, 73 unidirectional
strain gauges, 26 accelerometers, 28 fatigue damage sensors, and a wave radar.
The full scale measurements were complemented by numerical calculations and model tests in waves
on a self-propelled self-steering segmented hull built in scale 1:25 at MARIN. The model was composed
of six segmented rigid segments connected by a flexible backbone. Very good correlations (maximum
error in the order of 10%) are shown between the results of the measured bending moments in model tests
and at sea. Based on results from the model tests, it was concluded that weakly nonlinear effects do not
contribute significantly to fatigue damage.
For sea states with a significant wave height smaller than about 4m, whipping loads have a limited
contribution to the fatigue budget consumption. This result is completely opposite to what found on the
merchant ships previously mentioned. This may be due to the different type of ship (deep-V bow
sections), different structural dimensioning philosophy and particular severity of the wave climate
experienced.
Using the same data acquired at sea on the same ship, another study presented by Hageman et al.
(2014) was just dedicated to the comparison between theoretical/numerical methods for the prediction of
the fatigue life of a naval unit with respect to the measurements. The scatter of the results obtained from
four different seakeeping codes presented in the study of Hageman et al. (2014) well express this concept.
As also remarked in the previously mentioned paper of Drummen et al. (2014), the first perhaps biggest
uncertainty affecting long term predictions of maximum stress levels on ship hulls and in parallel the ship
structure fatigue life, is the statistics of the sea state in the operational area of the vessel: data acquired by
monitored ships at sea often show deviations from the GWS, which seems to be the main source of sea
state statistics taken in the design phase. The noted deviation could be due to a general change of
239
environmental conditions or to the pro-active decision of captains that try to avoid extreme sea states by
changing the route. The relevance of this human factor is addressed in the study of Stambaugh et al.
(2014), which presents a detailed analysis of the uncertainty deriving from various factors in the
assessment of the fatigue life of a ship. First principle considerations formulated in this work are general
and can serve as a guideline also for other studies based on HSMS data; their quantitative validity though
only applies to the particular case of the USC cutter on which measurement data were collected. The
relative weight of various factors influencing the uncertainty level in the prediction procedure of the ship
fatigue life, found by Stambaugh et al. (2014) on the USCG Bertholf cutter, is given as follows:
wave environment: 50%

wave spreading: 20%
number of days at sea: 18%
loads prediction: 8%
fatigue calculation (rainflow counting, stress definition): 4%.
Figure 3. Locations (indicated with dots) of the LBSGs installed on USCGC Bertholf cutter (Drummen et al. 2014).
They encourage the use of the method proposed by Sikora et al. (1983) and Sieve et al. (2000) or similar
spectral fatigue approaches in the early design since they may produce significant returns of investment
when compared to modifications required late in the design process, unplanned ship repairs and
sustainment cost. By comparison with values derived from the measurement campaign, they also
conclude once again that improvements are recommended for the Sikora et al. (1983) approach to fatigue
damage from impact loading and whipping response. Their conclusion appears very well supported by all
the other studies referenced in this section.
2.8
Uncertainties
In structural dynamic systems, there are various sources of uncertainty. The sources of the uncertainties
include but not limited to (a) parametric uncertainty - e.g., uncertainty in geometric parameters,
manufacturing tolerance, damping coefficient, strength of the materials involved; (b) model inadequacy
arising from the lack of knowledge about the model which is a-priori unknown; (c) experimental
uncertainty and unknown error percolate into the model when they are calibrated against experimental
results; (d) computational uncertainty e.g. machine precession, error tolerance in finite element analysis,
and (e) model uncertainty genuine randomness in the model.
Uncertainties generally appear to be random phenomena. Some of them such as material property
variations may be modelled in terms of probabilities, and others such as geometry variations are difficult
to quantify. There are two basic approaches to introduce uncertainties into structural dynamic analyses.
The parametric approach identifies model parameters individually and treats them as random variables
within the structural models. The nonparametric approach considers the overall effects of all the
uncertainties on the structural dynamical behaviours without specific analyses of individual uncertainties.
Characterizing dynamic properties of structures with uncertainty is an important task that provides critical
predictive information for structural design, assessment, and control.
Various methods are developed and applied for vibration analysis with uncertainties. The book entitled
Probabilistic Methods for Structural Design edited by Guedes Soares (2012) covered the recent
advances on modelling of uncertainty, prediction of the strength of components, load modelling and
combination, assessment of structural systems, stochastic finite elements and design consideration.
Considering ship operational and environmental uncertainties, procedures for the short term and the long
term prediction of wave-induced and whipping bending moments are developed by orak (2013).The
problem was formulated in the frequency domain using standard engineering tools for the load
240
computation: a seakeeping code for the rigid-body response and a beam finite element model for the
transient vibratory response.
DiazDelaO et al. (2013) applied Gaussian process emulators to alleviate the cost of characterizing the
random response of a structure subjected to vibration. The effectiveness of the method was demonstrated
by performing uncertainty analysis of the frequency response of a non-proportionally damped plate made
of a carbon fibre/epoxy composite material. Xia & Tang (2013) presented Gaussian process regression as
an efficient approach to analyse structural dynamics, especially characterizing structural responses under
uncertainty. This method uses only relatively small samples to make predictions about the collective
response characteristics, and computational costs thus can be saved remarkably. A two-stage hierarchical
modelling strategy has been adopted for Gaussian processes to bypass the difficulty in numerical
integration involved in a full Bayesian framework.
Regarding the vibration of a structure with uncertain properties a novel method, referred to as the
Stochastic Reduced Order Model (SROM) method, is proposed by Grigoriu (2013) for finding statistics
of the state of linear dynamic systems with random properties subjected to random noise. The method is
conceptually simple, computationally efficient and non-intrusive in the sense that it uses existing solvers
for deterministic differential equations to find state properties. Choi et al. (2014) investigated the sound
radiation from a vibrating plate having uncertain dynamic properties. Estimates are developed for the
reverberant vibration field in the uncertain plate subjected to a point-excitation, and for the ensemble
average of pressure from the direct field and from the reverberant field, leading to an estimate of the
average sound intensity.
Accurate prediction of the vibro-acoustic response of a structural system with uncertain properties is an
important issue in the design of engineering structures which are sensitive to manufacturing
imperfections. The difficulty of the problem is mainly embodied in two aspects, i.e. modelling approach
and uncertainty description. Existing literature provides two typical approaches including Finite Element
(FE) method and Statistical Energy Analysis (SEA) to model the system. In general, FE is the most
common technique in engineering practice with low frequency vibration while conversely SEA is
especially developed to deal with high frequency vibration. However, the mid-frequency problems are
not avoidable for the two methods.
As far as the description of system uncertainties is concerned, either parametric or non-parametric
models can be employed. Specific physical properties of the system are considered to be uncertain in the
parametric model, whereas the non-parametric model concentrates on the effects of uncertainty at a
higher level by using some form of random matrix theories.
Cicirello & Langley (2013) proposed a hybrid FE-SEA method, in which some components are
assumed to be deterministic, modelled by FE and other components to be highly random modelled by
SEA employing a non-parametric model of uncertainty. The coupling between the FE and SEA
components is affected by using the diffuse field reciprocity relation, and the resulting method can
adopted to yield both the ensemble mean and the variance of the response.
However, the division of the system into deterministic and random components is not always
appropriate in the case that some components may contain a degree of randomness, even though they
cannot be appropriately modelled as SEA subsystems. In order to address the problem, Cicirello &
Langley (2014) extended the hybrid method by applying parametric uncertainty models to components
that are not highly random, thus allowing an enhanced description of these components in the midfrequency range.
In nonlinear domain, the stochastic multi-dimensional harmonic balance method is proposed by
Didier et al. (2013) in order to solve dynamical problems with non-regular non-linearities in
presence of uncertainties. The quasi-periodic stochastic dynamic response is evaluated considering
uncertainties in linear and nonlinear parts of the mechanical system. The problem of optimal sampleddata vibration control for nonlinear systems with time delays and uncertainties is considered by Lei
(2013). The time-domain response of a randomly parameterized structural dynamic system is
investigated by Kundu & Adhikari (2014) with a polynomial chaos expansion approach and a
stochastic Krylov subspace projection, which has been proposed here. The simulations have been performed
for different degrees of variability of the input randomness and different dimensions of the input
stochastic space and compared with the direct Monte-Carlo simulations for accuracy and computational
efficiency.
To control vibrations for structures with uncertainty, Seigler & Hoagg (2013) presented a controller for
uncertain structures that are minimum phase and potentially subjected to unknown-and-unmeasured
disturbances. The controller is applied to structures modelled by finite-dimensional vector second-order
systems with unknown and arbitrarily large order. With the purpose of simplifying the Nonlinear Optimal
Vibration Control (NOVC) design, the original time-delay sampled-data system is converted into a
241
discrete-time non-delayed system first, as well as the nonlinear and uncertain terms is treated as external
excitations. A new procedure on random uncertainty modelling is presented by Gan et al. (2014) for
vibration analysis of a straight pipe conveying fluid when the pipe is fixed at both ends based on the
stochastically nonlinear dynamic theory and the Galerkin method.
For practical application in ships, a special issue themed in Uncertainty Modelling for Ships and
Offshore Structures (UMSOS) is published in Ocean Engineering 2014 based on the context of the joint
ISSC-ITTC UMSOS. Papanikolaou et al. (2014), Kim & Hermansky (2014) and Qiu et al. (2014)
reported recent advances in modelling the combined hydrodynamic responses of ship structures using
cross-spectral combination methods and in implementing uncertainty models used for the development of
modern decision support systems as guidance to ships master. Kim & Hermansky (2014) discussed
uncertainties in seakeeping analysis and the related ITTC procedures for loads and responses in waves.
The inherent variability and epistemic uncertainties associated with wind and wave data, including model
tests, are discussed and their consequences on specification of design criteria are illustrated by examples.
The authors conclude that following a well-established verification and validation process is important in
order to understand the error sources and the degrees of uncertainty and accuracy of both computational
predictions and model tests.
Ship structures are subjected to various deteriorating mechanisms throughout their service life. This
deterioration is highly uncertain and can adversely affect the performance and safety of the vessel.
Deteriorating mechanisms affecting ship structures and their prediction models under uncertainty are
reported by Frangopol & Soliman (2014) in the Handbook: Damage to Ship Structures under Uncertainty:
Evaluation and Prediction (2014).
orak et al. (2014) give an outline on how to derive wave time traces from a given sea state, which can
be used for such response calculations. The crux of their approach lies in explicitly addressing the
correlation between the wave bending moments and the whipping induced bending moments through a
probabilistic approach. It may be worthwhile to investigate whether the approach followed by the authors
can be used to deal with uncertainty of dynamic structural response as well.
2.9
This section focuses on noise, vibration, and shock acceptance criteria and procedures for their
measurement. International standards with regard to habitability, underwater noise radiation, and shock
test for ships are reviewed.
2.9.1
Habitability
IMO recognized the need to establish mandatory noise level limits for on board living and working
spaces. In 2012 the Maritime Safety Committee (MSC) adopted resolution MSC.337 (91) which contains
a new code with mandatory and recommendatory provisions to prevent seafarers from hazardous noise
levels and to provide standards for an acceptable environment (IMO 2012a) . Newly adopted SOLAS
regulation II1/13-2 as contained in resolution MSC.338(91), IMO (2012b), will apply the Code to ships
of 1,600 GT and above
with a building contract placed on or after 1 July 2014; or

in the absence of a building contract, the keel of which is laid or which is at a similar stage of
construction on or after 1 January 2015; or
delivered on or after 1 July 2018.
For applicable ships smaller 10,000 GT the limits remain the same as specified in IMO Resolution
A.468 (XII) , whereas for ships greater or equal 10,000 GT noise level limits in accommodation areas are
decreased by 5 dB. Also stricter requirements are settled for the measurement companies and
instrumentation. Measurements shall be conducted under simulated port conditions and at normal service
speed at no less than 80% of the Maximum Continuous Rating (MCR). For thrusters, measurements shall
be made at 40% thruster power.
On the 20th August 2013 the ILO MLC (2006) came into force setting minimum standards, for the
health, safety, and welfare of seafarers. The regulations are subject to the implementation into national
laws. The convention specifies requirements with respect to preventing the risk of exposure to hazardous
levels of noise and vibration and is calling for a decent ambient. The limits for noise levels defined in
IMO Resolution A.468 (XII) (1982) are commonly understood as satisfactory for compliance with the
noise aspects of the convention. Because MLC 2006 does not define limit values for vibration exposure,
this should be addressed by the national legislations. Most of the flag states that ratified the convention so
far did not concretise the convention in this respect. Recently the Flag of Antigua and Barbuda
encouraged both existing and new vessels joining the registry to comply with guidance such as that
242
contained in GL Guidelines for Compliance with MLC 2006 Noise and Vibration Requirements and also
the MSC.337(91).
GLs guidelines (2013) specify quantitative vibration and noise levels and define the process and
requirements for the approval of the corresponding service suppliers. Since some competent authorities
especially in the EU define vibration limits in way of daily exposure action and limit values acceleration
to ISO 26311, a simplified procedure to determine the daily vibration exposure based on a state of the art
sea trial measurement acceleration ISO 6954 is introduced.
Guidance for complying with the conventions requirements is also provided in the Guidance Notes of
ABS (2010) and BV Guidance Note (2012). Both give assessment criteria and measurement methodology
for obtaining a voluntary class notation.
IACS (2013) published its human element recommendations for structural design of lighting,
ventilation, vibration, noise, access and egress arrangements. The objectives of this recommendation are
to summarize information for human element and ergonomics during the structural design and
arrangement of ships e.g. to reduce noise and vibration in manned spaces. All four standards IACS, ABS,
BV and GL refer to the IMO Resolution A.468 (XII) or MSC.337(91) with respect to the applicable noise
limits and define a vibration acceptance criteria of 6mm/s maximum overall frequency-weighted r.m.s.
velocity values.
One of the aims of the EU funded SILENV (2012) project was to define a set of new pre-normative
requirements able to enforce a significant reduction of the noise and vibration impact of shipping
activities. The requirements for internal noise were set on the basis of an analysis of existing
requirements, a collection of existing and new data of noise measurements, and a correlation of noise
measurement results with the human perception, acquired through a large number of questionnaires,
distributed on board ships. Compared to MSC.337(91) the limits are significantly lower, e.g. for cabins a
limit of 50dB(A) is proposed. The same is true for the proposed vibration limits compared to ISO 6954
with limit values of 1 to 3 mm/s maximum overall frequency-weighted r.m.s. velocity values.
Also requirements for noise emissions in air aiming at ensuring living conditions for inhabitants of the
areas close to shipping activities were specified by the SILENV project. 75dB(A) at 25m was proposes as
radiation limit from sailing ship and 70dB(A) at101m for moored ships at quay.
2.9.2
Underwater noise
IMO has published MEPC Circ. 833 Guidelines for the reduction of underwater noise from commercial
shipping to address adverse impacts on marine life which sets out advice on design and operational
solutions that may be adopted to reduce underwater radiated noise (IMO 2014).
ISO (2012a,b) has developed the ISO/PAS 172081 Acoustics Quantities and procedures for
description and measurement of underwater sound from ships Part 1: General requirements for
measurements in deep water and ISO/DIS 16554.3 Ship and marine technology Measurement and
reporting of underwater sound radiated from merchant ships deep-water measurement.
Italian class society RINA has combined these standards into the DOLPHIN notation. The notation to be
published in 2014 will give requirements on instrumentation, site and procedures to carry out the
measurements, and will describe the information and post-processing activities necessary for reporting. Limits
both for when the ship is underway and quiet ship operational modes are established. By this RINA is the
second class beside DNV (Det Norske Veritas 2010) publishing underwater noise rules.
In a similar matter the SILENV (2012) project introduced pre-normative limits also aiming at reducing
the impact of shipping on marine mammals. The limits are based on the present state of the art, represented
by the most silent existing commercial vessels. Two curves are provided for commercial ships,
corresponding to the design speed transit, while the quite condition is related to a reduced speed,
particular studied in order to minimize the acoustic impact e.g. of cruise ships in protected areas.
EU is funding two ongoing underwater noise projects, SONIC and AQUO. With respect to prenormative standardization SONICs objective is to develop test measurement techniques for determining
the noise footprint at trial. Whereas AQUOs objective is to deliver practical guidelines for the ship
design and shipping control and regulation.
2.9.3
Others
The common German Dutch shock standard BV 0230 / D5050-0599 was revised in 2013. The main
alterations are
the standard now contains ship type specific, operational status and shock zone dependent shock
response spectra,
four independent load cases shall be considered in shock analysis, one in each of the three ships axes
and one ellipsoid loading,
243
for mast modules still individual loads shall be derived by full ship FEM calculations,
for the propulsion system a dynamic shock calculation shall be conducted based on shock loads to be
provided by the naval authorities. The simulation model needs to consist of the shaft, bearings, struts
and propeller,
for rigidly mounted equipment the NRL-Sum-Method is now also applicable,
the permissible stresses were revised.
ISO published ISO 20283-4 (2012a) focusing on the measurement and evaluation of vibration of the
ship propulsion machinery.
Figure 4. SILENV limits for the underwater noise radiated from commercial ships (SILENV 2012).
3.
OFFSHORE STRUCTURES
3.1
Vibration
This section covers vibration in offshore structures due to several environmental sources such as wind
(3.1.1), wave (3.1.2), ocean currents (vortex shedding, 3.1.3) and ice (3.1.5), as well as internal flowinduced vibration, which results from operational conditions (3.1.4). Even though they are treated herein
as separate physical mechanisms, they are frequently concurrent, therefore should be treated altogether.
The excitation mechanism phenomena are complex in nature and often the structure response exhibits
nonlinearities. The current state of knowledge for the vibration assessment and analysis of offshore
structures are discussed herein.
3.1.1
Wind-induced vibration
Owing to the characteristics of the jack-up drilling unit, the environment loads are important factors to the
structure safety. According to the MODU rule, the projected area method is used to calculate wind load.
In Hu et al. (2013) a 400ft jack-up is studied as an example. Considering the Reynolds number effects, the
wind loads of jack-up both in normal drilling and severe storm conditions are obtained in the experiment.
Compared with the results which are calculated according to MODU rule, it is found that the
experimental results give lower response levels. The effect of aerodynamic interference between the
different parts of the structure is hence believed to be important.
In Jameel et al. (2014) a spar platform which is subjected to both wave and wind loading is analysed,
and the corresponding steady-state response behaviour is studied. The wind loading which is acting on the
exposed part of the platform is decomposed into mean and fluctuating wind forces. The spar responses in
surge, heave and pitch along with mooring top tension are computed. The spar platform is found to
experience a significant lateral offset in the wave direction due to the wind loading. Inclusion of wind
loading also leads to an increase of the standard deviations of the surge and heave responses, in particular
for the latter. The mooring line tension increases for wind loading but the corresponding increase of the
dynamic tension fluctuations is very modest.
For floating offshore wind turbines, the simultaneous action of waves and wind are usually considered.
This applies both for model and full-scale testing as well as for numerical response analyses.
Accordingly, it is not always easy to identify the wind-induced versus wave-induced components.
Furthermore, it is not always possible to make a distinction between low-frequency versus high-frequency
response components. In the following a brief review of model tests and numerical analysis of floating
wind turbines is given.
Such floating platforms can be grouped into three categories: spar, semi-submersible and tension leg
platforms. Each has its benefits and drawbacks. Turbine designs require that pitch angle during operation
244
is below 10o, and with the large mass and thrust loadings inherent to these devices these low motions can
be difficult to achieve. A very informative review of different concepts for floating wind turbines is given
by NichollsLee et al. (2014).
Offshore wind turbines are designed and analysed using comprehensive simulation tools (or codes) that
account for the coupled dynamics of the wind inflow, aerodynamics, elasticity, and controls of the
turbine, along with the incident waves, sea current, hydrodynamics, mooring dynamics, and foundation
dynamics of the support structure. Comparison of computed results obtained by application of different
computer codes is hence important. This topic is addressed e.g. by Robertson et al. (2014) in relation to a
5-MW wind turbine installed on a floating semisubmersible in a water depth of 200m.
In a similar manner, verification of computed response with corresponding results from model tests and
full-scale measurements is a very important issue; see e.g. Robertson et al. (2013) in relation to the test
campaign which was performed for the DeepCwind Consortium. Model testing techniques are discussed
e.g. by Kimball et al. (2014). An example of a comparison between the resulting DeepCwind test data and
corresponding numerical calculations is provided by Koo et al. (2014).
Design verification of various floating wind turbine concepts by model testing is addressed in a number
of recent papers; see e.g. Adam et al. (2014) for testing of the GICON concept which is based on a
Tension Leg Platform (TLP) floater. Testing of the GustoMSC concept with a semi-submersible floater is
presented in Huijs et al. (2014). Testing of the VolturnUS concept is described in Young et al. (2014).
This turbine was installed in May 2013 and has a tower which is made of composite materials; see Lim et
al. (2013). Comparative testing of different floater types (i.e. TLP, spar and semi-submersible) was
described by Nihei et al. (2014).
A comparison between full-scale measurements and numerical analyses is provided by Utsunomiya
et al. (2014). The floating foundation is of the spar type. It carries a 100kW wind turbine of the downwind type, with a rotor diameter of 22m and a hub-height of 23.3m. The floating foundation consists of an
upper part made of steel and a lower part which is made of prestressed concrete segments. The turbine
was installed at a site about 1km offshore from Kabashima Island, Goto city in the Nagasaki prefecture in
2012. Since then, field measurement had been made until its removal in June 2013.
Full-scale measurements for the same turbine subjected to extreme wind excitation levels which were
experienced during the typhoon Sanba (and other somewhat weaker typhoons) are reported by
Utsunomiya et al. (2013).
Jia (2014) calculates wind-induced fatigue damage of offshore structures employing nonlinear time
domain dynamic analysis. He investigates the role of drag coefficient, comparing Norsok and DNV
specifications, showing that the former is more conservative. He also addresses the impact of time step
and duration and flare boom connections stiffness on the response. Results for static and dynamic
analyses (non-structural elements inertias are considered) are compared and he identifies that it is
important to take into account the contribution of secondary components such as flare and vent lines
when assessing the fatigue damage. He also investigates the role of the structure self-weight and
discussed the non-Gaussian nature of the statistical distribution of the local responses. Finally he
emphasizes that the fatigue methodology presented can be extended to other offshore tubular structures
exposed to wind excitation.
3.1.2
Wave load is one of the important environmental loads to offshore platform design, dynamic action of
waves on the ocean platform play a significant role in platform design to the deep regional development.
Offshore structures are designed to resist continual wave loading which may lead to significant fatigue
damage on individual structural members, and other types of loads due to severe storms, corrosion, fire
and explosion etc. The associated pressure pulsations excite the piping and/or equipment, which may
result in serious damage (Nakamura et al. 2014).
To date, the current design philosophy for the prediction of motions and wave-induced loads has been
driven by empirical or first-principles calculation procedures based on well-proven applications such as
ship motion prediction programs. Some of the recent advances in the assessment of loads for ships and
offshore structures with the aim to draw the overall technological landscape available for further
understanding, validation and implementation by the academic and industrial communities (Hirdaris et al.
2014). By artificially introducing a proper time-delay into control channel, a delayed H controller is
designed to attenuate the wave-induced vibration of the offshore platform and thereby improve the
control performance of the system (Zhang & Tang 2013).
Zhang et al. (2010) reviewed progress in the evaluation of wave forces on both slender and stocky
offshore structures in detail, with emphasis on the correlation between analytical/theoretical solutions and
full-scale measurements. Hydrodynamic impact and statistical aspects of wave-induced motions and loads
are briefly considered.
245
Recently on a previously unknown phenomenon in wave-induced vibration of the flexible structures with
large aspect ratio (length to diameter) was discovered in field experiments, which shows that traveling wave
rather than standing wave dominates the response of wave-induced vibration (Wu et al. 2010).
The wave-induced pressure is termed the push-and-draw pressure. At present, this push-and-draw
pressure is analysed using the potential theory developed for dissipative wave radiation problems (Anami
et al. 2012).
In both the mono- and multi-frequency types of response, the flow excites the structural vibrations
within the lock-in region and damps the structural motions in the non-lock-in region. The multi-frequency
character of the response impacts both the lock-in phenomenon and the fluidstructure energy transfer
(Bourguet et al. 2012).
A more practical approach entitled Modified Endurance Wave Analysis (MEWA) considering the
random and probabilistic nature of wave loading and utilizing optimal time duration is introduced.
MEWA can be a time-saving and also reliable method both in design and assessment of offshore
platforms (Diznab et al. 2014). The quantification of uncertainties may be challenging in model tests and
numerical simulations of ocean and offshore structures (Qiu et al. 2014). After considering more practical
factors, it is expected to be more applicable in structural damage assessment (Wang 2013).
The stochastic dynamic analysis of structure with and without damper is first presented in frequency
domain for parametric study on the performance of both the dampers to control wave induced vibration
response. The optimum performance of Tuned Liquid Column Damper (TLCD) and Tuned Liquid
Column Ball Damper (TLCBD) systems is further investigated to study the effectiveness of a particular
damper system over the other (Chatterjee & Chakraborty 2014).
3.1.3
Vortex-induced motion
Floating offshore platforms which are subjected to a uniform current may oscillate laterally across the
current if the vortex shedding frequency is in the vicinity of the natural surge and sway frequency. This
phenomenon is called Vortex Induced Motion, or VIM. Understanding VIM is of utmost importance to
the oil and gas industry, because especially spar platforms are very prone to it, with potential motion
amplitudes being of the order of the spar diameter. These types of motions have a significant impact on
both the mooring and the riser fatigue design. In particular, Steel Catenary Risers (SCRs) suspended from
the floater can be sensitive to VIM-induced fatigue at their mud line touchdown points. Strakes are
typically on the hull of the spar platform in order to mitigate or reduce the amplitude of motions to a more
manageable level.
At present, it seems that a widely applied approach for spar platforms is to calibrate numerical
calculations based on Computational Fluid Dynamics (CFD) analysis to experimental data obtained from
laboratory testing. Subsequently, a large number of parametric studies can be performed purely by
application of the numerical model. Benchmark studies have been performed in order to improve
confidence in CFD, see e.g. Halkyard et al. (2005, 2006).
Simulations of spar geometries with increasing detail are typically performed starting from a hull with
strakes, adding pipes and mooring components and also incorporating truss structures if such are present.
The effects of each component can then be studied and the corresponding effect on VIM response can be
quantified. It is generally found that quite accurate predictions are possible and that it is practical to
incorporate CFD into the design process. Screening and the generation of response curves can be done
during initial design and more targeted assessments during detail design, see Oakley & Constantinides
(2013).
In Lefevre et al. (2013) a set of VIM CFD simulations for a spar hard tank with appurtenances is
described together with comparison against a high quality scaled model test. The test data showed
considerable sensitivity to heading angle relative to the incident flow as well as to the magnitude of the
reduced velocity. The simulated VIM induced sway motion was compared against the model test data for
different reduced velocities and spar headings. Agreement between CFD and model test (VIM-induced)
sway motion was within 9% over the full range of parameters. Guidelines were provided for meshing and
selection of time step/solver settings.
The time variation of the experimental added mass coefficient and also the natural frequency of a
truss spar were investigated by Zhang, H. et al. (2012). The mean added mass coefficient as well as
the time-variable added mass coefficient was calculated based on measurements obtained from model
test.
In addition to spar platforms, it has also been observed that e.g. deep draft semisubmersibles are
subjected to VIM motions; see e.g. Gonalves et al. (2012a, b) and Zou et al. (2013). The impact is not as
significant as for the spar design, because the column diameters are generally smaller, and the hull
footprint larger. However, the understanding of the maximum motion characteristics is of critical
importance. This applies in particular to semisubmersibles with dry tree applications.
246
Although the potential for VIM of multi-column floating platforms such as semi-submersibles and
TLPs is well-acknowledged, the industry guidelines for design with respect to VIM are not
comprehensive and more research effort seems to be required.
Application of model tests combined with CFD analysis is also found to be feasible methods to
investigate VIM for the present type of platforms. Subsequent to the model tests and preliminary CFD
simulations, further CFD analyses can be carried out using improved simulation techniques by means of
commercial software. Good agreement between model test results and CFD calculations for VIM of a
multi-column floating platform was reported by Tan et al. (2013). Sensitivity of CFD results to the
modelling assumptions such as mesh size and density, time-step size and different turbulence models was
also presented.
In Rijken (2014) the effect of the mass ratio and the effect of performing physical experiments at
model scale are examined through CFD analyses for the hull of a semisubmersible platform. The
objectives of the CFD analyses are to focus on the VIM phenomenon itself and to compare response
magnitudes, while giving less importance to the hull details and absolute response magnitudes. Responses
for various cross-sectional shapes of the columns are examined which comprise square, rectangular and
five-sided geometries.
A comprehensive evaluation of the experimental investigations during the past decade on the VIM was
performed by Fujarra et al. (2012). One of the observations was that the effect on VIM due to the
coexistence of current and wave-excitation still deserves a better understanding. It was recommended to
perform a comprehensive investigation with consideration of a wide variety of wave characteristics. The
intention should be to propose better procedures on how to consider wave-excitation and VIM during the
first stages of design. Hence, excessively conservative predictions can be avoided.
This issue of VIM in the presence of waves superposed on a stationary current was investigated for a
large-volume semi-submersible platform by Gonalves et al. (2012a); see also Gonalves et al. (2012b).
The VIM model tests were performed both with regular waves and also different irregular wave
conditions which were characterized by spectral density functions. Significantly different behaviour was
observed for regular versus irregular waves. For regular waves, motion amplitudes in the transverse
direction were strongly reduced and no VIM was observed. However, for the case of irregular waves, the
amplitudes decreased slightly (as compared to the case with only current being present) but a periodic
motion which is characteristic for VIM was still observed.
VIM for other types of platforms than traditional semisubmersibles is also considered. In the already
mentioned work by Zou et al. (2013), results for VIM towing tests of a Paired-Column Semisubmersible
(PC Semi) platform are reported. The PC Semi configuration is different from a conventional Deep Draft
Semi (DD Semi) in three aspects, 1) 8 columns versus the more traditional 4 columns; 2) rectangular
column shape versus square column shape; 3) larger column slenderness ratio. The influence on VIM
from key design parameters, such as gap distance between the inner column and the outer column,
platform draft and mooring stiffness were investigated systemically.
Model testing and CFD simulation are used in order to investigate VIM behaviour of a TLP platform
by Tan et al. (2014). A mono-column hull type was investigated by Saito et al. (2012).
3.1.4
Internal flow-induced vibration
Offshore floating production units contain a large number of pipes and equipment to process the oil and
gas, however vibration due to internal flow has not been explicitly reported in applied publications.
A dynamic analysis of elastic cylindrical shells subjected to annular gas flow is developed by
Bochkarev & Matveenko (2013). The rotating fluid is assumed compressible and described by a potential
theory. A semi-analytical variant of the finite element method is employed and the shell stability is
analysed for different boundary conditions, geometrical and physical parameters. The effect of the outer
shell elasticity on the hydro-elastic stability is investigated.
Chang & Modarres-Sadeghi (2014) develop a numerical solution to investigate the Hopf bifurcation
stability condition of a cantilevered pipe conveying fluid subject to small displacement periodic base
excitation. The pipe may experience flow-induced planar or non-planar oscillations when flow exceeds
the critical velocity. The cantilevered pipe may experience 2 and 3D quasi-periodic and chaotic
oscillations at high velocities. The problem is governed by three-dimensional nonlinear equations which
are discretized using Galerkin technique and the resulting set of algebraic equations is solved by a finite
difference method. The authors advocate that the numerical predicted results qualitatively agree with
previous experimental work.
3.1.5
This part of the review was divided into theoretical analysis, numerical simulation and model tests and
field measurement.
247
Theoretical analysis:
Guo & Yue (2011) derived a criterion to predict the occurrence of ice induced self-excited vibration
based on its physical mechanism. The criterion was expressed as a relationship among the structures
parameters (stiffness, damping ratio, and diameter etc.) and ice properties (thickness, moving speed,
and compressive strength). Some measured data from full scale and model tests were used to validate
the criterion. Meanwhile Guo (2012a), from the engineering perspective, analysed two ISO 19906
(2010b) standard criteria (damping criterion and velocity criterion) of Ice Induced Vibration (IIV)
using data from two full scale structures, and found that the damping criterion was conservative and
velocity criterion was reasonable. Guo (2012b) developed a simple spectral model to simulate
continuous crushing ice loads based on ice load time series measured on the Norstromsgrund
lighthouse. The model had only two input parameters and it was relatively easy to estimate ice-induced
dynamic response of slender structures. A simulation example was also presented in the paper.
Considering that the uncertainties of ISO 19906 standard of ice induced frequency lock-in vibration
remained, Guo (2013) developed an approach to estimate the highest ice velocity causing frequency
lock-in vibration. He predicted the possible range of dynamic ice load, and estimated the structural
response if the dynamic portion of ice load was given.
Yue & Guo (2012) and Wang & Yue (2013), based on the observation on a prototype offshore
structure, proposed a physical mechanism to explain the ice induced self-excited vibration where a
single vibration cycle was divided into two phases: the loading phase and the unloading phase. It was
assumed that the compressive strain rate in the ice sheet was close to ductile-brittle transition during
the loading phase. An approximate quantitative analysis and verification was made for the proposed
mechanism.
Yan (2013), based on the analysis of ice failure process observed in the field and ice load and
structure response data, described the ice failure of lock-in vibration as a ductile damage collapse
failure process. A method of using an ice breaking length was presented to estimate the vibration
magnitude.
The number of vibration cycles caused by frequency locked-in ice load is one of the key parameters for
fatigue evaluation. Bjerks et al. (2014) proposed a new method to estimate the number of vibration
cycles and applied it to the Norstromsgrund lighthouse. Comparing with other methods, the authors
showed that all available methods overestimate the number of vibration cycles significantly.
IIV can influence the crew members both physiologically and psychologically when working and
living in a vibrating environment for long periods of time. Zhang, D. et al. (2012) conducted a study
which was based on data monitored on the platforms. Human feeling and dynamic response of the
pipeline on the platforms caused by deck vibration were evaluated. By comparing with the Chinese
standard Reduced Comfort Boundary and Evaluation Criteria for Human Exposure to Whole-body
Vibration evaluation results for the JZ20-2 platforms showed that serious IIV made work efficiency
degrade, and conventional IIV exceeded comfort degradation boundary. IIV can also lead to the
associated accidents to pipelines, such as fatigue fracture of the pipes and loosening of flanges.
Metrikine (2011) compared three induced vibrations: vortex-induced vibration of deep water riser, IIV of
flexible offshore platform, and pedestrian-induced vibration of bridge. He attempted to analyse
synchronization phenomenon in these three vibrations and tried to help cross-fertilize the research in the fields.
Numerical modelling:
Nandan et al. (2011) used Maattanens model to predict Steady-State Ice Induced Vibration (SS-IIV) of
the Norstromsgrund lighthouse. A closed-form stability contour based on eigenvalue analysis was
developed to define a boundary for ice-structure conditions conductive to SS-IIV, which can be used to
guide structure design away from SS-IIV at an early stage of the design process.
Hendrikse et al. (2011) investigated frequency lock-in vibration of a generalized beam with a
modification of the Matlock-Sodhi-Huang strip model. The study was tried to answer three interesting
questions: (i) can IIV occur at high ice sheet velocities? (ii) what are the conditions for IIV to occur at a
higher natural frequency of the structure? (iii) can an initially aperiodic ice loading cause IIV? And
further, as a part of Joint Industry Project (JIP), Hendrikse & Metrikine (2013) studied an effect of
friction in ice-structure interaction and to answer a question: why were modelled ice loads off more than
100% from measured load for a slender cylindrical structure. A Coulomb friction law was implemented in
the simulation and results showed the maximum increase in ice load was in the range of 15% when taking
into account the friction.
Shkhinek et al. (2013) developed a 2-D numerical solution of IIV using the discrete element method.
The modelling of the ice was based on PFC2D-Particle Flow Code. All the results were qualitatively and,
in some cases, quantitatively consistent with data from model tests and field measurements.
248
Popko (2014) simulated the dynamic response of the Norstromsgrund lighthouse in ice. A 3D
numerical model of the lighthouse was setup based on the solid elements in Abaqus software. And the
eigen-frequency response of the numerical model was tuned to correspond with the structural response of
the full-scale lighthouse.
Wind turbine dynamic behaviour in ice is also an interesting study. Hetmanczyk et al. (2011)
simulated wind turbine dynamic behaviour in ice by using the OnWind simulation software where
dynamic ice loads were calculated by an empirical ice model. Influences of ice velocity, ice thickness,
and damping effects of ice-structure and air-structure interaction were taken into account. By using
commercial software Abaqus, Jussila & Heinonen (2012) also predicted ice-structure interaction
response in the time domain for a cylindrical substructure and a conical substructure of a wind turbine.
Comparison of the results between cylindrical and conical structures showed that the conical structure
can reduce IIV.
Model test and field measurement:
Currently model test and field measurement is still the most important tool and very active in the
investigation of IIV even though a higher cost is needed.
Karna et al. (2013) introduced a JIP entitled Ice Induced Vibration sponsored by some offshore oil
companies for development and validation of models of IIV. The background, progress and some
validation findings were presented. The JIP activity includes workshop, reviewing experience from
laboratory to full scale, model development into working software, and validation of the software. Results
from the first phase of this project, completed in June 2012, were given in detail.
In 2011 NTNU also started a project Deciphering Ice Induced Vibration (DIIV). The objectives of
the project were to design and manufacture an adaptable test set-up, to conduct scale-model test, and
analyse results. The program included a series tests with ice properties, ice velocity, structure waterline,
surface roughness, structure compliance, natural frequencies, etc. Three papers related this project were
published in IAHR2012. Maattanen et al. (2012) introduced the measurement project, test rig design,
instrumentation, experiment program, and an overall view of achieved results. Nord & Maattanen (2012)
described an indirect approach, which was based on a frequency response function, to measure dynamic
ice load. Hendrikse et al. (2012) presented the forced vibration experiments in order to identify the added
mass and added damping in dynamic ice-structure interaction.
Yap (2011) and Yap & Palmer (2013) replicated various IIV modes in a model test from low to high
indentation velocities in order to investigate the mechanics of the Steady-State Self-Excited vibration
(SSSE). Analysis of the test results together with a collection of existing full-scale structure and model
test data was made and used to formulate a hypothetical framework of the mechanics of dynamic level
icestructure interaction. A dimensionless parameter was proposed to predict the occurrence of SSSE. A
suggestion was made to discard Froude scaling and to adopt Cauchy scaling and replica modelling in the
simulation of SSSE.
Bjerks et al. (2012) used Norstromsgrund lighthouse full-scale data from winter 2002 to 2003 to study
a question: whether IIVs were more frequent in specific time periods? It was found that the IIV seemed to
occur more frequently when the ice cover was warmer which corresponds with relative high air
temperatures and significantly higher drift speed of the ice. Bjerks et al. (2013) also analysed the
Norstromsgrund lighthouse field test data, which occurred on 30 March 2003 and was one of the harshest
vibrations during the entire measuring program 1999-2003. He concluded that spatial synchronization of
local loads was the reason for triggering of frequency locked-in vibration.
Palmer & Bjerkas (2013) analysed the field record data on nine load panels of the Norstromsgrund
lighthouse, and found synchronization phenomenon of the nine loads in frequency and phase when the
IIV began. An idealized model was used to explain the synchronization.
Gagnon (2011) introduced a high-speed imaging method (up to 30,000 images/s) to observe ice
crushing against a plate in laboratory-scale experiments. Lock-in vibration was observed at two different
frequencies, which was a frequency of the pump/actuator system (350 Hz) and a flexural resonant
frequency of the acrylic platen (900 Hz), by means of image analysis and ice load record. It also pointed
out that these observations were relevant to large-scale ice-structure interaction. Focused on an icecrushing induced vibration of Molikpaq (May 12, 1986), Gagnon (2012a, b) proposed a relationship
between the spalling frequency and ice sheet speed, and the lock-in occurred when the spalling frequency
was in the vicinity of the resonant frequency of the structure-ice system. He made an explanation for the
event in terms of ice spalling frequency, spalling mode, and variation of the effective mass and effective
spring of the structure-ice system.
Tian & Huang (2012) investigated a series of model tests on new compliant four-leg model where each
leg of the model was installed as ice breaking cone. It was found that the ice sheet broke nonsimultaneously before each single cone, and the structure was easy to be excited in nonlinear resonance.
249
Lan et al. (2011) introduced a measurement of IIV of the JZ9-3W HPB platform in Bohai Sea, China
with a remotely controllable system. According to the field test data and vibration mode of the platform
the effect of anti-ice induced vibration structures were analysed, and ice resistant proposals were
presented also.
Wang & Yue (2012) conducted a field measurement of a fixed ice-breaking cone which was installed
on the vertical riser of the JZ9-3E platform in the Bohai Sea, China in order to reduce IIV. The field
measurements and observations showed that the fixed ice-breaking cone was successful to reduce the
vibration.
Nord et al. (2013) investigated force identification methods in the frequency domain when a flexible
structure was acted on by an ice sheet and its response was measured from a set of non-collocated
measurement transducers. An example of the experimental result was discussed.
Huang et al. (2013a, b) conducted model tests to investigate the dynamic ice loads on two types of
four-legged jacket platform (cylindrical and conical models). The comparison showed that ice load and
structure response levels were significantly lowered by introducing a cone fixed to the platform legs, but
amplifications of the foundation reactions were higher in the cone test than those in the pile test.
Ziemer & Evers (2014) conducted a series of model tests in ice for a compliant cylindrical structure.
The IIVs were observed and were categorized into four different types of vibration according to their
characteristics: random, straight, circular and periodic vibrations. It was found that the same ice
conditions can lead to both periodic and straight vibrations, meanings that the loading frequency can be
doubled or even multiplied by three without visible change of outer circumstances.
3.2
Very Large Floating Structures (VLFS) enable the creation of land from the sea without the need for a
massive amount of fill materials. These kinds of structures have been gradually appearing in many parts
of the world for applications such as floating bridges, floating piers, floating performance stages, floating
airports, and floating storage facilities (Wang et al. 2008). Pontoon-type VLFS have large horizontal
dimensions and a relatively small depth. Due to this small depth to length ratio, VLFS are often modelled
as an elastic plate under wave action. Within the hydroelastic analysis, which can be carried out either in
the frequency or time domain (Wang & Tay 2011), the interaction of the fluid and the structure is taken
into account.
In Gao et al. (2011) the hydroelastic response of a VLFS is considered by modelling the structure as a
thick Mindlin plate and the water as an inviscid and incompressible fluid with an irrotational flow. The
modal expansion method proposed by Newman (1994) is applied to decouple the fluid-structure
interaction problem and the hydroelastic analysis is carried out in the frequency domain. The Laplace
equation is solved with the boundary element method and the finite element method is applied to compute
the deformation of the floating plate. In this way the influence of flexible line connections on the
hydroelastic response can be studied, demonstrating that hinge and semi-rigid line connections allow
reducing the hydroelastic response and the corresponding stress resultants in the VLFS. In Yoon et al.
(2014), a numerical procedure for the hydroelastic analysis of floating plates with hinges is presented and
validated by experiments. The influence of the hinge connections on the maximum bending moment is
studied for different structural and wave conditions. It is demonstrated that the hinge connection has a
strong effect on the hydroelastic response of the VLFS.
Papaioannou et al. (2013) present a stochastic hydroelastic analysis approach for VLFS assuming that
the surface waves can be described by a directional wave spectrum. The hydroelastic analysis is carried
out in the frequency domain, taking advantage of the modal expansion method. The fluid potential is
solved by the boundary element method and the Mindlin plate theory is discretized with the finite element
method. Applying the linear random vibration theory, the stochastic hydroelastic analysis of VLFS is
studied for multidirectional irregular waves. The influence of the mean wave angle on the standard
deviation and extreme values of the deflection and stress resultants is found to be very strong.
The numerical and experimental analysis of a hydroelastic response of a VLFS edged with a pair of
submerged horizontal plates in the time-domain is presented by Cheng et al. (2014). To this end, a direct
time domain modal expansion method is applied to the fluid-structure interaction problem. Based on the
numerical simulation and experiments the response-reduction efficiency of a VLFS equipped with
submerged horizontal plates is studied. The measured data and the simulation results are found to agree
well. Applying the simulation approach, different variants of submerged horizontal plates can be
investigated and an improved design of the anti-motion plate can be suggested.
3.3
Noise
Noise and underwater sound due to exploration, construction, transport, drilling, and production is
important for offshore activities. Identification of pertinent mechanisms, multiple potential noise
250
sources, and noise paths from vessels and drilling platforms are necessary for implementation of
effective treatments. Hence, the subject of noise can be divided into two major branches in which
specific sources and noise makers are identified, and possible treatments for those sources,
mechanisms, and paths are considered.
With the rising number of offshore activity, the noise emission into the sea, caused by pile driving for
the foundations of the wind turbines and offshore structures, becomes an issue of great importance.
Considering the immense costs and high logistic complexity of offshore tests, there is a strong demand for
numerical models, which are able to calculate the resulting underwater sound of pile driving activities.
Consequently, study for noise mitigation measures such as bubble curtain was carried out.
3.3.1
Analysis of underwater noise by pile-driving
Numerical study for underwater noise by pile-driving for wind turbines was studied by Lippert et al.
(2012) and Tsouvalas & Metrikine (2013, 2014). In the study of Lippert et al. (2012), a 2D modelling
approach for the acoustic near field based on the finite element method is able to correctly reproduce the
general characteristics of the underwater sound pressure field due to pile driving. Besides a qualitatively
proper representation of the developing pressure wave at the pile and the corresponding reflections at both
seabed and water surface, also a feasible magnitude for the absolute peak values of the sound pressure
could be achieved.
Tsouvalas & Metrikine (2013, 2014) studied a linear semi-analytical formulation of the coupled vibroacoustics of a complete pilewatersoil interaction model. The pile is described by a high order thin shell
theory whereas both water and soil are modelled as three-dimensional continua. Results indicate that the
near-field response in the water column consists mainly of pressure conical waves generated by the
supersonic compression waves in the pile excited by the impact hammer. The soil response is dominated
by shear waves. The Scholte waves are also generated at the watersea bed interface which can produce
pressure fluctuations in the water column that are particularly significant close to the sea floor. The
effects of soil elasticity and pile size are thoroughly investigated and their influence on the generated
pressure levels is highlighted.
3.3.2
Measurement and mitigation of underwater noise
Marine impact piling is a significant source of low-frequency impulsive noise and Robinson et al. (2012)
described methodologies developed for measurement of marine piling including estimation of the energy
source level. Measurement was made during construction of an offshore wind farm involving piles of
typically 5 m in diameter driven by hammers with typical strike energies of around 1000 kJ. Acoustic
data were recorded using hydrophones deployed from a vessel, allowing the transmission loss to be
confirmed empirically.
In order to effectively measure the underwater noise by offshore oil production vessels, Erbe et al.
(2013) performed the underwater acoustic recordings of six Floating Production Storage and Offloading
(FPSO) vessels. Monopole source spectra were computed for use in environmental impact assessments of
underwater noise. Given that operations on the FPSOs varied over the period of recording, and were
sometimes unknown, the authors present a statistical approach to noise level estimation. No significant or
consistent aspect dependence was found for the six FPSOs. Noise levels did not scale with FPSO size or
power.
Kuhn et al. (2012) have developed a new underwater piling noise mitigation system, Hydro Sound
Dampers (HSD). The HSD is based on the theories of dispersion, dissipation and resonance effects for
elastic balloons. One of the main advantages of the HSD is that the mitigation can be pre-adjusted to a
pre-defined frequency range, as marine mammals are sensitive only for a certain sound frequency range.
The results of a small scale tests and a full scale test were promising a reduction of 12 dB up to 20 dB.
The research findings concerning the shape and the material of the HSD were also presented.
Bohne et al. (2014) have performed numerical modelling of a bubble curtain to reduce the underwater
radiated noise from pile driving and Wochner et al. (2014) developed a new prototype open-ended
resonator design and tested for the purpose of incorporating arrays of the resonators into an underwater
noise abatement system. Individual resonators were designed to have a resonance frequency near 100 Hz
in order to reduce the low frequency noise by pile driving and drilling.
3.3.3
Equipment noise
Decoupling between vibrating machinery and ship & offshore structures produces both a significant
reduction in the vibration power transmitted to the hull and a reflection of some vibration energy back
to the machinery. Moro & Biot (2013) carried out the experimental test of ISO 108461~5 standards
that has been applied in order to achieve the dynamic response of a very large resilient mounting
specifically designed for a medium-speed marine diesel engine. The results obtained by the experiment
251
show that experiments are able to give useful information which, without complex data manipulation,
clearly identify dynamic behaviour of the resilient elements in both the vertical and horizontal
direction.
Wang & Shen (2014) developed the active vibration isolating system which can effectively reduce the
low-frequency noise transmission of the diesel engine. The technical problems are discussed to meet the
engineering application requirements when the active vibration plant is used for ships and offshore
structures, such as the control of installation size and weight, engineering reliability, the configuration of
active and passive vibration isolation components, anti-shock measures, electromagnetic shielding
measures and external interfaces.
Machinery resiliently supported by rubber or metal isolators constitutes a most common mounting
system. But at high frequencies, standing wave resonances occur in isolators and isolation effect would
be diminished. Xu et al. (2014) carried out the experimental study for engine mounting system that
about 30 dB attenuation of vibration acceleration can be obtained at most of frequencies above 1 kHz
especially for the modern variable speed electric motor. The study applied the two-stage mounting
system with an intermediate mass which is smaller and lighter than conventional mass. The
experimental results showed that over 45 dB isolation effects can be obtained at the switching and
harmonic frequencies.
Pipe or duct elements like orifices and constrictions are widely used in pipe or HVAC system and
are often responsible of turbulent flow noise and whistling. The characterization requires the
development of new techniques to describe the dynamics of the noise sources and the flow-acoustic
interaction. Sovardi et al. (2014) studied the identification of noise sources in internal ducted flows
using LES (Large Eddy Simulation) and SI (System Identification) in order to characterize
simultaneously both the acoustic passive scattering and the active noise generation of an orifice placed
in a duct or pipe.
3.4
Blast
The topsides of offshore platforms are the most likely areas to be exposed to hazards such as
hydrocarbon explosions. Profiled barriers are being increasingly used as blast walls in offshore topside
modules to provide a safety barrier for personnel and critical equipment. The corrugated blast walls
are one of the common passive protection systems. Most existing stainless steel blast walls are rated to
a pressure of approximately 1 bar. However, some joint industry projects have shown the possibility
of blast-induced overpressures as high as 4 bars. The blast walls can be designed using the Single
Degree Of Freedom (SDOF) method as recommended in the design guidance and a time-domain finite
element commercial software should be used for predicting the response of blast wall panel by the
Technical Note 5 (TN5) issued by the Fire and Blast Information Group (FABIG 1999, Fischer &
Hring 2009).
Sohn et al. (2013) conducted the structural response analysis of FPSO topside blast wall under
explosion loads, in which the computed based time-domain nonlinear finite element analysis and single
degree of freedom method based on resistance function were adopted. The results show that both methods
have more similar results at larger deflection domain than smaller deflection domain and results of the
SDOF method are less conservative than the finite element analysis results. Nwankwo et al. (2013)
examined a blast wall partially-retrofitted by Carbon Fiber Reinforced Plastic (CFRP) patches in the
central region for understanding the effect of a composite patch on the blast resistance of profiled blast
walls. The analysis results show that the strengthened scheme was able to absorb more blast energy than
the un-strengthened scheme and an average reduction of 33% in the maximum displacement was
observed in the inelastic response.
Sandwich panels are widely used in various fields because such panels have lower density, easier
fabrication method and higher strength and blast resistance compared with monolithic plates. The
dynamic response of a sandwich structure depend upon many parameters, including the properties of the
skins, the compressive/shear moduli of the core, the strength of the core, as well as the strength of the
bond between the skin and core. In many cases, the strength of the sandwich structure is controlled by the
failure characteristics of the core material and the skin-core interface.
Balkan & Mecitolu (2014) investigated the dynamic behaviour of a viscoelastic sandwich composite
plate subjected to the non-uniform blast load by theoretical and experimental study. Parametric studies
show that increasing the thickness of the core layer is more effective than the increase of face layer
thickness. In order to observe a good structural vibration damping, increasing the thickness of the core
layer is recommended. Guan et al. (2014) conducted the comparison of the stitched and unstitched
sandwich panels suggested that for a given impulse, the stitched laminates exhibited a slightly superior
blast resistance. Results show that the through-thickness stitching does not play a significant role in
enhancing the blast response of the sandwich panels.
252
Li et al. (2014) investigated dynamic response of corrugated aluminum sandwich panels under air
blast loadings experimentally and numerically. Parametric studies show that the residual deflections of
the face sheets can be effectively reduced by increasing the thickness, the yield stress and the contact
area of face sheets and core. Jing et al. (2014) experimentally investigated the deformation/failure
modes and blast resistance of cylindrical sandwich shells comprising two aluminum face-sheets and an
aluminum foam core, subjected to air blast loading. Various failure modes - indentation or tearing of
the front face-sheet, collapse of the core, severe inelastic deformation or tearing of the rear face-sheet,
and failure between the face-sheets and foam core, were observed. The findings are useful for
validating theoretical predictions, as well as to guide application of cellular metal sandwich structures
for blast protection purposes.
3.5
Methods for vibration control of structures can generally be classified into three main categories: (i)
passive control, (ii) semi-active control (i.e. tunable over time) and (iii) active control. While some of the
methods that can be applied for this purpose are the same for land-based and offshore structures,
additional options naturally arise for offshore structures. This is e.g. due to presence of thrusters for many
floating vessels and the possibility of increasing the hydrodynamic damping and not only the structural (and
possibly the aerodynamic) damping as for land-based structures. At the same time, the additional sources of
hydrodynamic excitation clearly also represent challenges.
This is exemplified by the struggle to mitigate vortex-induced vibrations for offshore structures.
Passive control methods are frequently used to control vortex shedding or the flow around the structure.
They are used widely for VIV suppression (see e.g. Zdravkovich 1981, 1997, 2003, Bearman &
Brankovi 2004, Assi et al. 2009, 2010). Passive control of aerodynamic and hydrodynamic devices are
according to the classical text by Zdravkovich (1981) divided into three categories: (i) surface protrusion,
(ii) shrouds, (iii) wake stabilizers. The passive control devices in the first two categories disrupt the
boundary layer on the surface of the structure. The wake stabilizers hinder the two shear layers, thus
weakening the vortex shedding process.
Helical strakes represent a much applied option for category (i) devices, i.e. surface protrusions. In Resvanis
et al. (2014) the effects of strake coverage and marine growth on flexible cylinder VIV was investigated. The
results show that even small bare sections (missing strakes) can lead to significant VIV response.
Results from two testing programs on long cylinders towed at high Reynolds numbers to assess the
performance of helical strakes with differing conditions along the cylinder length are given by Allen &
Liapis (2014b). Similarly, the performance of fairings was studied by Allen & Liapis (2014a). The results
show that the coverage length, density, and location of the helical strakes or fairings have a substantial
effect on both the local and global response of the tubular.
Ng et al. (2014) considered the effect of fairings for VIV suppression in relation to tandem risers by
means of scaled model tests. Application of parallel and oblique plates for suppression of VIV was
considered by Assi & Franco (2013).
Although helical strakes tend to reduce the VIV amplitudes, there are also cases with interaction
between several cylinders where the opposite may occur. Freire et al. (2013) performed a study on how an
upstream cylinder fitted with helical strakes can induce higher vibrations than bare cylinders in relation to
a second cylinder mounted downstream.
Park et al. (2013a) investigated Passive Turbulence Control (PTC) in relation to flow-induced motions
of a circular cylinder. The efficiency of different layouts of partial surface covering by means of
longitudinal strips was compared. Furthermore, in Park et al. (2013b) suppression of flow-induced
motions of two cylinders in tandem using surface roughness is studied experimentally.
Numerical studies are frequently employed as a supplement to experimental methods. Corson et al.
(2014) studied the application of CFD to predict the hydrodynamic performance of different fairing
designs. The VIV suppression effect for a finned cylinder was investigated by means of CFD in Wang, Y.
et al. (2014). The numerical results show that the cylinder with fins can significantly modify the vortex
shedding and synchronization process implying reduced vibration levels. Through numerical simulations,
the flow past a circular section cylinder with a conic disturbance is investigated at subcritical Reynolds
numbers by Lin et al. (2014).
Passive control of the vortex shedding process is also the basic mechanism underlying increase of roll
damping for floating vessels by introduction of bilge keels. This is a widely applied scheme for floating
production systems, and recent papers dealing with this subject are e.g. those by De Oliveira & Fernandes
(2012), Minnick et al. (2012), Tom et al. (2012). The possibility of applying U-tanks for increase of roll
damping was considered by E Silva et al. (2012).
Different methods for passive control of various types offshore structures are considered in a number
of studies. Bottom-fixed platforms are addressed e.g. by Lee, C.H. et al. (2014), TLPs by Chandrasekaran
253
et al. (2014) and Very Large Scale Structures (VLSFs) by Wang, C.M. et al. (2012). Barges for support of
floating wind turbines were studied by Shadman & Akbarpour (2012a, b).
Active control schemes implemented as Dynamic Positioning (DP) systems are much applied for
floating vessels. These are based on the application of counteracting forces (that sometimes can be
represented as additional stiffness and damping) for floating vessels by means of thrusters. This is in
order to control the motion levels, primarily those associated with slow-drift motions, see e.g. the
classical texts by Balchen et al. (1980) and Saelid et al. (1983). Here, just a brief summary is given of
some of the recent developments of such control algorithms.
Sliding mode control algorithms are investigated by De Sousa & Tannuri (2013). Two robust control
techniques were applied and compared: first order Sliding Mode Control (SMC) and Higher Order Sliding
Mode control (HOSM). To validate the simulated controller, experimental tests were performed
considering a small-scale model of a DP tanker.
Different algorithms for DP heading control were compared by Miyazaki et al. (2013). The paper
addresses the analysis of the final equilibrium heading of several published weathervane control strategies
for two different DP vessels. The computed headings are also compared to the exact minimum power
heading.
DP thrust allocation algorithms which also take into account the physical limitations of each thruster
such as the maximum thrust (saturation), the maximum rate of turn (azimuth) and the maximum rate of
change of RPM were addressed by Arditti et al. (2014).
The very few full-scale DP operations in the Arctic have demonstrated the need for improvements in
DP systems for ice-covered waters. An example of a controller which is addressing ice-related challenges
is presented by Kerkeni et al. (2013).
3.6
Uncertainties
The uncertainties depend on the phenomenon that is studied and on the time and spatial scales of interest.
Obisesan (2012) divided the uncertainty of offshore structures into the following two major categories:
Epistemic: The uncertainty induced by the lack of knowledge, and it is predictable.

Aleatory: The system has an intrinsic random or stochastic nature and it is not predictable.
The epistemic systems have uncertainty that may be reduced upon additional information. Uncertainty
in a structural member may be reduced by measurement of the element behaviour. Aleatory uncertainty
assumes that an underlying Probability Density Function (PDF) exists and is the square of the wave
function in quantum mechanics and also, the PDF is a fundamental property of the system. In most
engineering systems, the PDF is obtained from historic data and represents both epistemic and aleatory
uncertainties. Thus, the precise form of a PDF can only be assumed. On the other hand, interval methods
play an important role in quantifying epistemic uncertainty.
During the modelling data obtained from oceanographic instruments it is necessary to identify
uncertainties. The condition assessment of offshore structures is very challenging due to the presence of
various uncertainties in the geometric properties and material and load characteristics. Complex
interactions do exist between different uncertainties and the significance of extreme value effects in the
reliability analysis. The model of these uncertainties is used in limit states formulated to meet design
criteria for structural members.
The results of reliability and uncertainty analysis provide rational assertion on how more attention
should be given to the ultimate capacity of structural members and the representation of least sensitive
random variables with deterministic values. The uncertainty quantification and also structural dynamic
analysis help in identifying structural components that could lead to collapse of offshore structures
and also revealing the true state of structures. This information eventually contributes knowledge to
risk assessment and improve on the safety of personnel and properties operating on offshore
structures.
In Wang & Li (2013) the wave-induced vibration of offshore platform with Magneto-Rheological
(MR) damper is presented. The model of the platform coupled with MR damper is established where the
external wave force is approximated with a white noise via a designed filter. A lot of control strategies
have been investigated and illustrated to be effective for structural vibration mitigation. The authors
considers that full state feedback is required for an effective control which need a lot of sensors, and
uncertainty in measuring state variables cant be considered.
In Rezanejad & Guedes Soares (2013) the performance of floating oscillating water column in finite
water depth is analysed based on the linearized water wave theory in the two dimensional Cartesian
coordinate systems. It is assumed that the floating oscillating water column is very large and has
negligible motions and is therefore considered as fixed. The free surface inside the chambers is
254
modelled as a non-plane wave surface. The boundary integral equation method is employed to solve the
associated boundary value problem and the multi region concept is used to overcome the numerical
errors due to the assumption of thin barriers. The numerical results are compared with the existing
experimental results and analytical results and the accuracy of the numerical model is evaluated. The
effects of the geometrical dimensions of the floating oscillating water column and harbour attached to
the device are investigated.
3.7
Revision works for dynamic ice action and responses

Dynamic ice actions are addressed in ISO 19906 Clause A8.2.6. However, the ISO provisions are not
enough to conduct a fully dynamic design of an offshore structure under ice action. Especially,
information on floating structures is limited and only generalities are offered in the normative part A8 or
A13 involving check lists and general recommendations for design, but no guidance on induced ice
actions, including ice scenarios is offered. In addition, ISO 19906 does not provide implicit or explicit
guidance for the design and operation of arctic pipelines.
Therefore, through Barents 2020 project (Det Norske Veritas 2012), which was supported by Russian,
Norwegian authorities, and international oil and gas industry, an additional guidance document was
developed for design against ice loads on stationary floating structures. With respect to dynamic response,
these were focused on a consistent methodology for application of the recommended analytical ISO
formulas to derive the global ice load as a function of the floaters response and table over calculation
methods for ice loads with original references and applicability ranges.
Furthermore, by Joint Industry Project (JIP), Det Norske Veritas (DNV) is developing DnV-RP-C209
(2009) which is a new recommended practice for arctic structures and pipelines. The focus of this rule is
on describing methodologies for qualifying assessment tools and in generating rational characteristic
values to describe the governing ice regime during the life time of the pipeline system. Particular attention
is being paid to ice gouging and optimized pipeline burial depth, where assessment methodologies and
tools are very much at the cutting edge of technological development. However, DNV had not published
this recommended practice and it is expected to be released in the near future.
Revision of comfort and health notation
BV notation of comfort and health on board (Bureau Veritas 2013) for noise and vibration was revised to
enhance its practicability, which is applicable to all offshore units. In the notation, the procedures for
measurement of airborne sound insulation index between rooms and calculation of reverberation time in
space was more specifically described.
4.
CONCLUSION
An overview of the technical literature related to dynamic response of ship structures over the past three
years clearly indicates continuation of concern for environment-induced vibrations, which was also an
important issue in the previous reporting period. This is the true not only for springing and whipping, but
for ice-induced vibrations too. Springing and whipping remain to be important topic, which is particularly
consequence of increased number of very large ships having benefit of economies of scales. A large
number of full-scale measurement campaigns, mostly related to large container ships, have been reported.
Reasonably, the same vessel types are mostly considered in a number of reported model tests and
numerical calculations. Hence, it is obvious that all available methods i.e. full-scale measurements, model
tests and numerical methods were combined to assess wave effects on ship structures. Most of full-scale
measurements offered information on the influence of high-frequency vibrations on fatigue. Also, some
interesting observations on total structural damping are reported. Model tests mainly concern segmented
models with an elastic backbone (BB), or with flexible hinges, while the use of fully flexible models has
not been reported. Since mostly container ship models were investigated, challenges related to torsional
response assessment as a consequence of shear center below the keel arise irrespective on the segmented
model type selection. From the references related to numerical calculations, slamming and whipping are
recognized as important subjects. In whipping calculations boundary methods and field methods are
distinguished. Still, the most common method for calculation of the slamming forces has been the von
Karman momentum approach, while recently Generalized Wagner Method (GWM) has become very
attractive, or Modified Logvinovich Model (MLM) as its simplification for blunt sections.
In case of ice-induced vibrations, both ship hull and propeller interactions were investigated. Important
topics like pressure distribution and structural response in ship-ice interaction are dealt with by model
tests and numerically. Some estimation methods for excitation forces and vibration levels are offered
based on records from icebreakers and ice-going ships. However, general conclusion on this topic is that
255
additional effort is necessary to improve numerical techniques and related methodologies for reliable
prediction of ice-induced vibrations. Hence, it is expected that this topic will attract many studies in near
future.
Different aspects of propeller-induced vibrations are considered in the reporting period in rather low
number of publications: application of transfer matrix method for shafting line dynamic response
assessment, prediction of hull pressure fluctuation induced by propeller sheet cavitation, performance
analysis of a periodic isolator to reduce the vibration and noise radiation influenced by propeller forces,
etc. Also, application of Computational Fluid Dynamics (CFD) tools for some particular problems is
noted. In spite of new developments of engines with ultra-long stroke and low revolution rate, which
are recognized to have better efficiency but at the same time having increased external forces and
moments transmitted to the ship hull, there is lack of references related to the machinery-induced
vibrations in the reporting period. However, some references related to testing of widely used friction
connection and hydraulic type top bracings to control H-mode resonance and to reduce the vibration of
the engine itself are reported.
In the context of numerical and analytical vibration studies, limited time and resources are
recognized as drawbacks of finite elements, and therefore some simplified and approximate solutions
are offered for vibration and buckling problems of isotropic and orthotropic plate structures as
constitutive members of a ship structure, aiming at both modelling facilitation and simplification of
finite element formulations.
A set of interior noise regulations is being put under discussion to check its compliance with actual
technological developments. A-weighted sound pressure levels are confirmed to reliably indicate noise
situation on board, although there were several studies aiming to develop more sophisticated indicators.
However, it seems that some improvements of IMO A.468 (XII) are necessary. Alternative numerical
methods to FEA/SEA based approaches are considered, but still correct modelling of nature of the
coupling between structural elements remains to be the biggest limit of the available methods.
Improvement of acoustic behaviour of insulation panels and partitioned walls is recognized as a trend,
and extensive studies, mainly experimental ones, were conducted to enhance insulation performance by
varying materials, panel constitutions, etc. Within the air radiated noise, which is relative new field of
investigation, a big lack of normative regarding the protection of coastal inhabited areas and harbours is
recognized. In that field most of publications originated from the findings achieved EU project
SILENV. Many of the most recent studies in the field of underwater radiated noise are focused to
establish its correct limit to be taken into account and in finding a correlation between such limit and
the behaviour of the marine fauna. Still, underwater noise for military and civil applications is
necessary to be distinguished. As expected, within the former one there are no references due to
confidentiality issues, but within the latter one there are many studies particularly considering
propellers behaviours as a major source. Underwater radiated noise measurements are reported from
several places worldwide, considering influence of different ship types on marine life. Also, within the
ongoing EU project AQUO a description of the oceans environment as noise mapping is proposed in
order to establish the effect of the anthropogenic sound on marine life.
Sloshing induced impacts are very important in the design of a ship tank and the CCS. Many physical
effects have to be considered at the same time: gas cushion, liquid compressibility, boiling of liquid
cargoes, aeration, thermal exchange, hydroelasticity. When analysing sloshing impacts, one must always
have the structural response in mind. This implies that the fluid (liquid, gas) flow must be solved
simultaneously with the dynamic elastic structural reaction.
It is common in tank design to do model experiments for sloshing-induced impact effects by means of
forced oscillation tests. However, the scaling of the model-test results represents a challenge due to the
many physical effects that may matter. This concerns not only the maximum values of the pressure but
also their time evolution. The relationship between the temporal characteristics of the load and the
structural response is nonlinear and dependent on the impact characteristics related to the natural periods
of the structure. Therefore, the effect of scaling the pressure time histories may only be assessed by
analysing the dynamic response of the containment system.
On numerical side, it appears that the correct numerical modelling of hydro-structure interactions
during the sloshing impacts inside the LNG tanks is still beyond the state of the art and there is still no
rational direct calculation procedure to be used for design verification of the CCS. The quasi full scale
model tests and intermediate scale model tests are believed to bring more light into this difficult problem.
In particular they could be used for detailed validation of the hybrid approach discussed in Section 2.4.2,
which, once validated, could be put into the rational design methodology based on direct calculation
approach. The full scale measurements and monitoring of the real LNG ships would be extremely helpful
for better understanding of the way how the CCS is suffering in reality. How to perform these full scale
measurements is another complex question. In any case, the actual situation is that, for the design
256
verification of CCS, we still rely on the so called comparative approach. However, it is very important to
mention that, in spite of all the imperfections of the comparative approach, the overall safety record of
LNG floating units is excellent and only few incidents were experienced.
A benchmark study for shock response of panels to the blast loading is provided by the previous ISSC
committee V.I report (2012), and in the reporting period several numerical investigations are reported on
different issues as for instance dynamic failure of ship structure steel plate under near-field air-blast
loading, effects of blast loading on response of corroded plates, weapons attack on ship structures
(investigated also numerically), etc. Response of ship structures to the underwater explosions was
mainly investigated utilizing finite element method in combination with different techniques. Shock
resistance performance of different materials like composites, rubber coated plates and sandwich
structures with core, is investigated both numerically and experimentally. Numerical studies dealing
with the effect of bottom reflection of the shockwave in shallow water require further investigation to
achieve general conclusion.
In the forced vibration analysis of ship structures, damping is still mainly accounted for in a simplified
way, i.e. by lumping all its components together through a constant damping coefficient specified as a
percentage of critical damping.
Hull structural monitoring systems for ships and offshore structures are basically the same.
Advancements in new sensors technologies, large data handling and wireless communication networks
are recognized in the reporting period. A general trend is to replace classical communication between the
peripheral and main processing units with wireless or fiber-optic cables. Extensive review of new sensors
technology and application particularly and related to optic-based, wireless and acoustic emission sensors
is provided.
Among different sources of uncertainties, special attention was paid to ship operational and
environmental uncertainty assessment.
Concern for ship habitability improvement is evident in the reporting period not only from the needs to
establish mandatory noise level limits initiated by IMO, but also from guidelines and standards issued by
IACS and related classification societies, referring to relevant IMO resolutions. Two ongoing EU
underwater noise related projects (SONIC and AQUO) are expected to enhance standardization in
measurement techniques for the noise footprint at trials and to deliver practical guidelines for the ship
design and shipping control and regulation.
Vibration in offshore structures due to environmental and operational loads continues to be a major
concern for design. Often several physical mechanisms occur simultaneously and cannot be treated
separately. The excitation mechanisms are complex and the structural response frequently exhibits
nonlinearities.
Wind-induced loads impact the safety of jack-up drilling units. The aerodynamic coupling between
parts of the structure may be significant. For spar platforms, wind loading amplifies the standard
deviations for surge and heave motions. The mooring line tension increases with wind loading but the
dynamic tension fluctuations are modest.
Floating offshore wind turbines can be grouped into three major categories: spar, semi-submersible and
tension leg platforms. Comprehensive simulation tools account for the coupled dynamics considering all
relevant loading components: wind inflow, aerodynamics, elasticity and controls of the turbine, along
with incident waves, marine currents, mooring and foundation dynamics of the support structure. For such
complex analyses it is important to compare and validate numerical results from different codes with
small and full scale tests.
Dynamic action from waves is important for offshore platform design. Emphasis has been placed on
correlating results from analytical/theoretical solutions with small and full-scale measurements. However,
in the motion prediction and wave induced loads assessment, still ship motion prediction programs are
mainly used.
Vortex-induced motion occurs in floating offshore platforms subjected to uniform currents if the vortex
shedding frequency and natural surge and sway frequencies are close. These motions particularly affect
spar platforms, having also impact on mooring and riser fatigue design. It is important to calibrate
numerical results from CFD with experimental laboratory data. Benchmark studies have improved
confidence in CFD analysis. Deep draft semisubmersibles may also be subjected to VIM, however with
lower impact on the design as compared to spar platforms. VIM in multi-column floating platforms is
recognized, but industry guidelines are not comprehensive.
Although offshore floating production units contain a large number of pipes and equipment to process
the oil and gas, it is difficult to find references explicitly reporting internal flow induced vibration.
Ice-induced self-excited vibrations depend on the parameters that characterize the structure as well as
the ice properties. Relatively large number of measured data from full scale and model tests has been
documented. Ice-induced vibrations can influence the crew both physiologically and psychologically.
257
Model test and field measurement are very important in order to complement and validate numerical
analyses. Dynamic ice actions addressed in ISO 19906 are not sufficient for a full dynamic design of an
offshore platform and arctic pipelines. DNV is working on a recommended practice to cover this
deficiency.
Very Large Floating Structures (VLFS) have been considered for applications such as floating bridges,
piers, airports and storage facilities. Pontoon-type VLFS have small depth as compared to their horizontal
dimensions, therefore they may be modelled as an elastic plate under wave action. Bearing in mind their
large dimensions, and consequently lower natural frequencies that can fall into the range of ordinary sea
spectrum, hydroelastic analysis is recommended to be carried out both in the frequency and time domain.
Noise is important for offshore exploration and production activities. Identification of mechanisms,
potential sources and paths are required. Numerical simulation of underwater noise by pile-driving for
wind turbines reproduces the general characteristics of the sound pressure field. Marine impact piling is a
significant source of low-frequency impulsive noise. Methodologies have been developed for estimation
of the energy source level. Underwater noise has been measured in six Floating Production Storage and
Offloading (FPSO) vessels.
Decoupling between vibrating machinery and the hull of ocean structures produces reduction in the
transmitted and reflected vibration power. Active vibration isolating systems can effectively reduce lowfrequency noise transmission in diesel engines.
In the context of blast loading, it is worthy to mention that offshore platforms topsides are the most
exposed areas for hydrocarbon explosions. Barriers are being increasingly used in topside modules to
safeguard personnel and critical equipment. Corrugated blast walls are commonly employed in passive
protection systems. Sandwich panels are light, easy to fabricate and offer high strength and blast
resistance.
Vibration control methods can be classified into passive, semi-active and active. Hydrodynamic
excitation mechanisms represent challenges to offshore structures. To reduce vortex-induced vibrations
passive control methods are used to control vortex shedding.
Sources of uncertainties generally need to be taken into account during the design process. These will
typically be strongly dependent on time and spatial scales of interest. Uncertainties can in general be
classified as (i) Epistemic: induced by lack of knowledge and (ii) Aleatory: intrinsically random or
stochastic in nature. In practical assessment both uncertainty types are represented by probability density
functions, with assumed form originating from historic data.
Within the contribution to standard and acceptance criteria improvement, some new guidances for
design against ice loads and notations of comfort and health onboard are already offered in the reporting
period, and also some recommended practices are expected to be released by the relevant subjects in the
near future.
Generally, a significant contribution is achieved both in understanding of physical phenomena related
to dynamic response of ships and offshore structures and their numerical and experimental modelling and
monitoring. However, it is reasonable to expect that the trends in shipping and offshore industry
mentioned in the Introduction will continue in the near future. Therefore, additional efforts of all relevant
subjects are necessary to increase reliability and accuracy of available dynamic response analysis
techniques.
REFERENCES
Aarsnes, J.V. 1996. Drop test with ship sections Effect of roll angle. Report 603834.00.01. Trondheim, Norway:
Norwegian Marine Technology Research Institute.
Adam, F., Myland, T., Dahlhaus, F. & Gromann, J. 2014. Scale tests of the GICON-TLP for wind turbines. The
American Society of Mechanical Engineers (ASME) Conference Proceedings Ocean, Offshore and Arctic
Engineering (OMAE), 813 June 2014, San Francisco, USA.
Ahn, Y.G., Kim, K.H., Kim, S.Y., Lee, S.W., Kim, Y.H. & Lee, J.H. 2013. Experimental study on the effects of
pressure sensors and time window in violent sloshing pressure Measurement. International Offshore and Polar
Engineering Conference (ISOPE), 30 June - 5 July 2013, Anchorage, USA.
Ahn, Y.J., Kim, S.Y., Kim, K.H., Lee, S.W., Kim, Y.H. & Park, J.J. 2012. Study on the effect of density ratio of
liquid and gas in sloshing experiment. The International Workshop on Water Waves and Floating Bodies
(IWWWFB), 2225 April 2012, Copenhagen, Denmark.
Ahuja, A. & Gupta, A. 2014. Fuzzy logic controlled semi-active floating raft vibration isolation system. Universal
Journal of Engineering, 2, 142147.
Alireza, E.M. & Ahmad, R.R. 2014. Deformation behaviour of corroded plates subjected to blast loading. Ships and
Offshore Structures, DOI: 10.1080/17445302.2014.889371.
Allen, D.W. & Liapis, S. 2014a. The effect of coverage length and density on the performance of fairings. The
258
Allen, D.W. & Liapis, S. 2014b. The effect of coverage length and density on the performance of helical strakes.
The American Society of Mechanical Engineers (ASME) Conference Proceedings Ocean, Offshore and Arctic
Allen, J.K., Peterson, M.L., Sharrard, G.V., Wright, D.L. & Todd, S.K. 2012. Radiated noise from commercial ships
in the Gulf of Maine: Implications for whale/vessel collisions. The Journal of the Acoustical Society of America,
132, 229235.
American Bureau of Shipping 2010. ABS guide for compliance with the ILO maritime labour convention, 2006 title
3 requirements, Houston, USA, American Bureau of Shipping.
American Bureau of Shipping 2014. Updated guidance notes on ship vibration, Houston, USA, American Bureau of
Shipping.
Anami, K., Ishii, N. & Knisely, C.W. 2012. Added mass and wave radiation damping for flowinduced rotational
vibrations of skinplates of hydraulic gates. Journal of Fluids and Structures, 35, 213228.
Anastasopoulos, A., Kourousis, D., Botten, S. & Wang, G. 2009. Acoustic emission monitoring for detecting
structural defects in vessels and offshore structures. Ships and Offshore Structures, 4, 363372.
Ancellin, M., Brosset, L. & Ghidaglia, J.M. 2012. Influence of phase transition on sloshing impact pressures
described by a generalized Bagnolds model. International Offshore and Polar Engineering Conference (ISOPE),
1722 June 2012, Rhodes, Greece.
Andersen, I.M.V. & Jensen, J.J. 2012. On the effect of hull girder flexibility on the vertical wave bending
moment for ultra large container vessels. The American Society of Mechanical Engineers (ASME)
Conference Proceedings Ocean, Offshore and Arctic Engineering (OMAE), 16 July 2012, Rio de Janeiro,
Brazil.
Andersen, I.M.V. & Jensen, J.J. 2013. Hull girder fatigue damage estimations of a large container vessel by spectral
analysis. The International Symposium on Practical Design of Ships and Other Floating Structures, 2025
October 2013, Changwon, Korea.
Andersen, I.M.V. & Jensen, J.J. 2014. Measurements in a container ship of wave-induced hull girder stresses in
excess of design values. Marine Structures, 37, 5485.
Arditti, F., Cozijn, J.L. & Van Daalen, E.F.G. 2014. An advanced thrust allocation algorithm for DP applications,
taking into account interaction effects and physical limitations. The American Society of Mechanical Engineers
(ASME) Conference Proceedings Ocean, Offshore and Arctic Engineering (OMAE), 813 June 2014, San
Francisco, USA.
Assi, G.R., Bearman, P. & Kitney, N. 2009. Low drag solutions for suppressing vortex-induced vibration of circular
cylinders. Journal of Fluids and Structures, 25, 666675.
Assi, G.R.S., Bearman, P.W., Kitney, N. & Tognarelli, M.A. 2010. Suppression of wake-induced vibration of
tandem cylinders with free-to-rotate control plates. Journal of Fluids and Structures, 26, 10451057.
Assi, G.R.S. & Franco, G.S. 2013. Experimental investigation on the stability of parallel and oblique
plates as suppressors of vortex-induced vibration. The American Society of Mechanical Engineers
(ASME) Conference Proceedings Ocean, Offshore and Arctic Engineering (OMAE), 914 June 2013, Nantes,
France.
Avachat, S. & Zhou, M. 2012. Effect of core density on deformation and failure in sandwich composites subjected
to underwater impulsive loads. The International Journal of Multiphysics, 6, 241266.
Baarholm, G.S. & Moan, T. 2000. Estimation of nonlinear longterm extremes of hull girder loads in ships. Marine
Structures, 13, 495516.
Badino, A., Borelli, D., Gaggero, T., Rizzuto, E. & Schenone, C. 2012a. Acoustic impact of ships: Noiserelated
needs, quantification and justification. In: Rizzuto, E.S., C.G (ed.) Sustainable Maritime Transportation and
Exploitation of Sea Resources. London, UK: Taylor & Francis Group.
Badino, A., Borelli, D., Gaggero, T., Rizzuto, E. & Schenone, C. 2012b. Analysis of airborne noise emitted from
ships. In: Rizzuto, E.S., C.G (ed.) Sustainable Maritime Transportation and Exploitation of Sea Resources.
London, UK: Taylor & Francis Group.
Badino, A., Borelli, D., Gaggero, T., Rizzuto, E. & Schenone, C. 2012c. Control of airborne noise emissions from
ships. The International Conference on Advances and Challenges in Marine Noise and Vibration (MARNAV), 5
7 September 2012, Glasgow, UK.
Badino, A., Borelli, D., Gaggero, T., Rizzuto, E. & Schenone, C. 2012d. Modelling the outdoor noise propagation
for different ship types. The International Conference on Ships and Shipping Research, 1719 October 2012,
Naples, Italy.
Badino, A., Borelli, D., Gaggero, T., Rizzuto, E. & Schenone, C. 2012e. Noise emitted from ships: impact inside
and outside the vessels. Procedia-Social and Behavioral Sciences, 48, 868879.
Badino, A., Borelli, D., Gaggero, T., Rizzuto, E. & Schenone, C. 2012f. Normative framework for ship noise:
Present and situation and future trends. Noise Control Engineering Journal, 60, 740762.
Badino, A., Borelli, D., Gaggero, T., Rizzuto, E. & Schenone, C. 2013. The EU research effort towards the control
of noise emissions from ships: the SILENV Project (7FP). The Italian Annual Conference on Acoustics and
German Annual Conference on Acoustics (AIADAGA), 1821 March 2013, Merano, Italy.
Baeten, A. 2012. Finite particle based elastic structure dynamics modeling for LNG sloshing analysis. International
Offshore and Polar Engineering Conference (ISOPE), 1722 June 2012, Rhodes, Greece.
Bai, Z., Zhao, J., Zhang, W. & Wang, W. 2013. 3D SPH numerical investigation for the sloshing impact in LNG
tank. The American Society of Mechanical Engineers (ASME) Conference Proceedings Ocean, Offshore and
Arctic Engineering (OMAE), 914 June 2013, Nantes, France.
259
Balchen, J.G., Jenssen, N.A., Mathisen, E. & Slid, S. 1980. A dynamic positioning system based on Kalman
filtering and optimal control. Modeling, Identification and Control, , 1, 135163.
Balkan, D. & Mecitolu, Z. 2014. Nonlinear dynamic behavior of viscoelastic sandwich composite
plates under non-uniform blast load: Theory and experiment. International Journal of Impact Engineering, 72,
85104.
Bardazzi, A., Lugni, C., Faltinsen, O.M. & Graziani, G. 2012a. Hydroelastic study of the impact phenomena in
sloshing flow. The International Conference on Hydroelasticity in Marine Technology, 1921 September 2012,
Tokyo, Japan.
Bardazzi, A., Lugni, C., Faltinsen, O.M., Graziani, G., Greco, M. & Colicchio, G. 2012b. Waveimpact in a
sloshing tank: Hydroelastic challenges. The International Workshop on Water Waves and Floating Bodies
(IWWWFB), 2225 April 2012, Copenhagen, Denmark.
Barhoumi, M. & Storhaug, G. 2013. Assessment of whipping and springing on a large container vessel. The
International Symposium on Practical Design of Ships and Other Floating Structures (PRADS), 2025 October
2013, Changwon, Korea.
Baudin, E., Diebold, L., Pettinotti, B., Rousset, J.M. & Weber, M. 2013. Bureau veritas sloshing model tests & CFD
calculations within ISOPE benchmark. International Offshore and Polar Engineering Conference (ISOPE), 30
June - 5 July 2013, Anchorage, USA.
Baudin, E., Fillon, B. & Diebold, L. 2012. An alternate impact measurement for the sloshing impact
pressures. International Offshore and Polar Engineering Conference (ISOPE), 1722 June 2012, Rhodes,
Greece.
Baudin, E., Folegot, T., Andre, M., Audoly, C., Rizzuto, E. & Benedetti, L. 2014. Noise footprint: A proposal within
the framework of FP7 AQUO project to define a goal based approach towards the reduction of underwater
radiated noise from shipping. The American Society of Mechanical Engineers (ASME) Conference Proceedings
Ocean, Offshore and Arctic Engineering (OMAE), 813 June 2014, San Francisco, USA.
Bearman, P. & Brankovi, M. 2004. Experimental studies of passive control of vortex-induced vibration. European
Journal of Mechanics-B/Fluids, 23, 915.
Belov, I.M. & Spiridonov, N.N. 2012. Features of ship vibration in ice operation conditions. International Offshore
and Polar Engineering Conference (ISOPE), 1722 June 2012, Rhodes, Greece.
Bennett, S., Hudson, D.A., Temarel, P. & Price, W. 2012. The influence of abnormal waves on global wave-induced
loads. The International Conference on Hydroelasticity in Marine Technology, 1921 September 2012, Tokyo,
Japan.
Bennett, S.S., Brooks, C.J., Winden, B., Taunton, D.J., Forrester, A.I.J., Turnock, S.R. & Hudson, D.A.
2014a. Measurement of ship hydroelastic response using multiple wireless sensor nodes. Ocean Engineering, 79,
6780.
Bennett, S.S., Hudson, D.A. & Temarel, P. 2014b. Global wave-induced loads in abnormal waves:
Comparison between experimental results and classification society rules. Journal of Fluids and Structures, 49,
498515.
Bertetta, D., Brizzolara, S., Gaggero, S., Viviani, M. & Savio, L. 2012. CPP propeller cavitation and noise
optimization at different pitches with panel code and validation by cavitation tunnel measurements. Ocean
Engineering, 53, 177195.
Bertram, V., El Moctar, O., Oberhagemann, J. & Schellin, T.E. 2011. Free surface RANS method for hull girder
springing and whipping. The International Community for Maritime and Ocean Professionals (SNAME) Annual
Meeting, 1618 November 2011, Houston, USA.
Bigot, F., Derbanne, Q., Sireta, F.X., Malenica, S. & Tuitman, J.T. 2011. Global hydroelastic ship response
comparison of numerical model and WILS model tests. International Offshore and Polar Engineering
Conference (ISOPE), 1924 June 2011, Maui, USA.
Bjerks, M., Alsos, H.S. & Wsj, K. 2014. Estimates of the number of vibration cycles from frequency locked-in
ice loads. The American Society of Mechanical Engineers (ASME) Conference Proceedings Ocean, Offshore
and Arctic Engineering (OMAE), 813 June 2014, San Francisco, USA.
Bjerks, M., Lny, C. & Grtner, A. 2012. Seasonal variations in the occurrence of ice induced vibration of a
bottom fixed structure. International Offshore and Polar Engineering Conference (ISOPE), 1722 June 2012,
Rhodes, Greece.
Bjerks, M., Meese, A. & Alsos, H. 2013. Ice induce vibration Observations of a full scale lockin event. The
23rd International Offshore and Polar Engineering, 30 June - 5 July 2013, Anchorage, USA.
Bochkarev, S.A. & Matveenko, V.P. 2013. Stability of a cylindrical shell subject to an annular flow of rotating fluid.
Journal of Sound and Vibration, 332, 42104222.
Bohne, T., DiazCereceda, C., Criessmann, T. & Rolfes, R. 2014. Numerical modeling of a bubble curtain. The 2nd
International Conference and Exhibition on Underwater Acoustics, 2227 June 2014, Rhodes, Greece.
Borelli, D., Schenone, C. & Pittaluga, I. 2013. Analysis of sound absorption behaviour of polyester fibre material
faced with perforated panels. The International Congress on Acoustics (ICA), 27 June 2013, Montreal, Canada.
Boroditsky, L. & Fischer, R. 2012. Ship hull structure response excitation on airborne noise excitation. The
International Congress on Sound and Vibration (ICSV), 812 July 2012, Vilnius, Lithuania.
Bourguet, R., Lucor, D. & Triantafyllou, M.S. 2012. Mono-and multi-frequency vortex-induced vibrations of a long
tensioned beam in shear flow. Journal of Fluids and Structures, 32, 5264.
Brigman, N.N.A. 2012. Structural health monitoring in commercial aviation. MS. thesis, Massachusetts Institute of
Technology.
260
Brosset, L., Ghidaglia, J.M., Guilcher, P.M. & Le Tarnec, L. 2013. Generalized Bagnold model. International
Offshore and Polar Engineering Conference (ISOPE), 30 June - 5 July 2013, Anchorage, USA.
Brown, N.A. 1976. Cavitation noise problems and solutions. The International Symposium on Shipboard Acoustics
(ISSA), 610 September 1976, Noordwijkehout, Netherlands.
Brubak, L., Hellesland, J. & Hareide, O.J. 2013. Vibration analysis of plates with arbitrary stiffener arrangements
using a semianalytical approach. The International Conference on Computational Methods in Marine
Engineering MARINE 2013, 2931 May 2013, Hamburg, Germany.
Bulian, G., Botia-Vera, E., Castellana, F., Mas-Soler, J. & Souto-Iglesias, A. 2012. Repeatability and practical
ergodicity of 2D sloshing experiments. International Offshore and Polar Engineering Conference (ISOPE), 17
Bureau Veritas 2011. BV guidance note NI 554, Design sloshing loads for LNG membrane tanks, Paris, France,
Bureau Veritas.
Bureau Veritas 2012. BV guidance note NI 577, Accommodation design and construction of crew accommodation in
respect of title 3 of maritime labour convention 2006, Paris, France, Bureau Veritas.
Bureau Veritas 2013. Rules for the classification of offshore units Part C Facilities Chapter 5 Comfort and
health on board, Paris, France, Bureau Veritas.
Cao, Y., Zhang, F. & Liapis, S. 2013. A computer code for fast simulations of liquid tank sloshing. International
Caresta, M. & Kessissoglou, N.J. 2011. Reduction of hull-radiated noise using vibroacoustic optimization of the
propulsion system. Journal of Ship Research, 55, 149162.
Chandrasekaran, S., Kumar, D. & Ramanathan, R. 2014. Response control of TLP using tuned mass dampers. The
Chang, G.H. & Modarres-Sadeghi, Y. 2014. Flow-induced oscillations of a cantilevered pipe conveying fluid with
base excitation. Journal of Sound and Vibration, 333, 42654280.
Chatterjee, T. & Chakraborty, S. 2014. Vibration mitigation of structures subjected to random wave forces by liquid
column dampers. Ocean Engineering, 87, 151161.
Chen, C.M., Yu, L.J., Cao, S.Q., Ling, P.L. & Yang, J.S. 2013. The application of wireless sensor network in ocean
fishing vessel. Applied Mechanics and Materials, 401, 16321635.
Chen, C.R. & Chen, H.C. 2014. CFD Simulation of extreme slamming on a containership in random waves.
International Offshore and Polar Engineering Conference (ISOPE), 1520 June 2014, Busan, Korea.
Cheng, Y., Zhai, G.J. & Ou, J. 2014. Time-domain numerical and experimental analysis of hydroelastic response
of a very large floating structure edged with a pair of submerged horizontal plates. Marine Structures, 39, 198
224.
Chezhian, M. 2003. Threedimensional analysis of slamming. Ph.D. thesis, Norwegian University of Science and
Technology.
Cho, D.S., Kim, B.H., Vladimir, N. & Choi, T. 2014a. Natural vibration analysis of vertical rectangular plates and
stiffened panels in contact with fluid on one side. The Institution of Mechanical Engineers, Part M: Journal of
Engineering for the Maritime Environment, DOI: 10.1177/1475090214533955.
Cho, D.S., Vladimir, N. & Choi, T. 2014b. Natural vibration analysis of stiffened panels with arbitrary edge
constraints using the assumed mode method. The Institution of Mechanical Engineers, Part M: Journal of
Engineering for the Maritime Environment, DOI: 10.1177/1475090214521179.
Cho, D.S., Vladimir, N. & Choi, T.M. 2014c. Approximate natural vibration analysis of rectangular plates with
openings using assumed mode method. International Journal of Naval Architecture and Ocean Engineering, 5,
478491.
Cho, T.J. 2013. Vibro-acoustic characteristics of floating floor system: The influence of frequency-matched
resonance on low frequency impact sound. Journal of Sound and Vibration, 332, 3342.
Choi, H.I., Park, J.H., Kwon, S.H., Lee, K.H., Lee, S.B. & Yang, Y.J. 2012. An experimental study on
hydroelasticity in sloshing. The American Society of Mechanical Engineers (ASME) Conference Proceedings
Ocean, Offshore and Arctic Engineering (OMAE), 16 July 2012, Rio de Janeiro, Brazil.
Choi, W.J., Woodhouse, J. & Langley, R.S. 2014. Sound radiation from a vibrating plate with uncertainty. Journal
of Sound and Vibration, 333, 39663980.
Choung, J.M. 2013. Effective arrangement of hydraulic top bracing based on idealized load carrying mechanism.
Ocean Engineering, 58, 252259.
Chun, M.S., Yoon, J.W., Suh, Y.S., Rodda, K.K., Han, S.M., Kim, S.C., Park, T.Y. & Jang, K.B. 2013. A structural
capacity evaluation for corrugated membrane structure of LNG cargo containment system. The American Society
of Mechanical Engineers (ASME) Conference Proceedings Ocean, Offshore and Arctic Engineering (OMAE),
914 June 2013, Nantes, France.
Cicirello, A. & Langley, R.S. 2013. The vibro-acoustic analysis of built-up systems using a hybrid method with
parametric and non-parametric uncertainties. Journal of Sound and Vibration, 332, 21652178.
Cicirello, A. & Langley, R.S. 2014. Efficient parametric uncertainty analysis within the hybrid finite
element/statistical energy analysis method. Journal of Sound and Vibration, 333, 16981717.
Colicchio, G., Greco, M. & Faltinsen, O.M. 2013. A numerical strategy for gas cavity-body interactions from
acoustic to incompressible liquid phases. The International Workshop on Water Waves and Floating Bodies
(IWWWFB), 710 April 2013, Marseille, France.
261
orak, M. 2013. Probabilistic combination of wave and whipping bending moments on ship structures. Ph.D. thesis,
University of Zagreb.
orak, M., Parunov, J. & Guedes Soares, C. 2013. Long-term prediction of combined wave and whipping bending
moments of container ships. The International Conference on Marine Structures (MARSTRUCT), 2527 March
2013, Espoo, Finland.
orak, M., Parunov, J. & Guedes Soares, C. 2014. Short-term probabilistic combination of wave and whipping
bending moments. The International Conference on Maritime Technology and Engineering, 1517 October 2014,
Lisbon, Portugal.
Corson, D., Cosgrove, S. & Constantinides, Y. 2014. Application of CFD to predict the hydrodynamic performance
of offshore fairing designs. The American Society of Mechanical Engineers (ASME) Conference Proceedings
Costes, J., Dias, F., Ghidaglia, J.M. & Mrabet, A. 2013. Simulation of breaking wave impacts on a flat rigid
wall by a 2D parallel finite volume solver with two compressible fluids and an advanced free surface
reconstruction. International Offshore and Polar Engineering Conference (ISOPE), 30 June - 5 July 2013,
Anchorage, USA.
Costes, J., Ghidaglia, J.M. & Mrabet, A.A. 2014. On the simulation of liquid impacts on a flat rigid wall by a 2D
parallel finite volume solver. International Offshore and Polar Engineering Conference (ISOPE), 1520 June
2014, Busan, Korea.
Cummins, W.E. 1962. The impulse response function and ship motions, Washington, USA, DTIC Document.
Cusano, G. & La Marca, S. 2014. Evaluation and forecasting of elapsed fatigue life of ship structures by analysing
data from full scale ship structural monitoring. The Ship Structure Committee Symposium: Vessel Safety &
Longevity through Ship Structure Research, 1820 May 2014, Linthicum, USA.
Daifuku, M., Takezawa, A., Kitamura, M., Terashita, H. & Ohtsuki, Y. 2014. Smooth operators iron out engine
room vibrations. The Naval Architect, February 2014, 6264.
Davidson, G., Lavroff, J., Davis, M.R., Thomas, G.A., Binns, J.R., Freizer, S. & Roberts, T. 2013. Non-linear
behaviour in the seakeeping of high speed craft. International Conference on Fast Sea Transportation (FAST), 2
5 December 2013, Amsterdam, Netherlands.
De Chowdhury, S. & Sannasiraj, S.A. 2013. Numerical simulation of 2D wave impact pressure using SPH
schemes. International Offshore and Polar Engineering Conference (ISOPE), 30 June - 5 July 2013, Anchorage,
USA.
De Hauteclocque, G., Derbanne, Q. & El-Gharbaoui, A. 2012. Comparison of different equivalent design waves
with spectral analysis. The American Society of Mechanical Engineers (ASME) Conference Proceedings Ocean,
Offshore and Arctic Engineering (OMAE), 16 July 2012, Rio de Janeiro, Brazil.
De Oliveira, A.C. & Fernandes, A.C. 2012. An empirical nonlinear model to estimate FPSO with extended bilge
keel roll linear equivalent damping in extreme seas. The American Society of Mechanical Engineers (ASME)
Brazil.
De Sousa, T.F. & Tannuri, E.A. 2013. Experimental analysis of higher-order sliding mode control applied to
dynamic positioning systems. The American Society of Mechanical Engineers (ASME) Conference Proceedings
Ocean, Offshore and Arctic Engineering (OMAE), 914 June 2013, Nantes, France.
Dematteo, A. & Ratouis, A. 2013. Stochastic simulation of the pressure field induced by sloshing
impacts. International Offshore and Polar Engineering Conference (ISOPE), 30 June - 5 July 2013, Anchorage,
USA.
Derbanne, Q., Bigot, F. & De Hauteclocque, G. 2012. Comparison of design wave approach and short term
approach with increased wave height in the evaluation of whipping induced bending moment. The American
Society of Mechanical Engineers (ASME) Conference Proceedings Ocean, Offshore and Arctic Engineering
(OMAE), 16 July 2012, Rio de Janeiro, Brazil.
Dessi, D. 2013. Damping of ship global modes: Techniques and analysis. The International Symposium on Practical
Design of Ships and Other Floating Structures (PRADS), 2025 October 2013, Changwon,
Korea.
Dessi, D. & Ciappi, E. 2013. Slamming clustering on fast ships: From impact dynamics to global response analysis.
Dessi, D. & Dorazio, D. 2012. Modal parameters of ship global modes: Modelscale and fullscale estimations.
The International Conference on Hydroelasticity in Marine Technology, 1921 September 2012, Tokyo,
Japan.
Det Norske Veritas. 2009. DNVRPC209: Joint industry research to meet arctic challenges [Online].
Hvik, Norway: Det Norske Veritas. Available: http://www.dnv.com/industry/oil_gas/publications/updates/
arctic_update/2009/01_2009/JointindustryresearchtomeetArcticchallenges.asp [Accessed 8 Jan 2015].
Det Norske Veritas 2010. Silent class notation, in: Rules for classification of ships Newbuildings, Hvik, Norway,
Det Norske Veritas.
Det Norske Veritas 2012. Barents 2020 assessment of international standards for safe exploration, production and
transportation of oil and gas in the Barents Sea: Final report phase 4. 20120690. Oslo: Norway: Det Norske
Veritas.
Det Norske Veritas 2014. Classification note 30.9, sloshing analysis of LNG membrane tanks, Oslo: Norway, Det
Norske Veritas.
262
Di Bella, A. & Remigi, F. 2013a. Evaluation and control of cruise ships noise in urban areas. The International
Congress on Sound and Vibration (ICSV), 711 July 2013, Bangkok, Thailand.
Di Bella, A. & Remigi, F. 2013b. Prediction of noise of moored ships. The International Congress on Acoustics
(ICA), 27 June 2013, Montreal, Canada.
Diazdelao, F.A., Adhikari, S., Flores, E.I.S. & Friswell, M.I. 2013. Stochastic structural dynamic analysis using
Gaussian process emulators. Computers and Structures, 120, 2432.
Didier, J., Sinou, J.J. & Faverjon, B. 2013. Nonlinear vibrations of a mechanical system with non-regular
nonlinearities and uncertainties. Communications in Nonlinear Science and Numerical Simulation, 18, 3250
3270.
Diebold, L. & Baudin, E. 2014. Bureau Veritas sloshing model tests & CFD calculations for ISOPE sloshing
benchmark. International Offshore and Polar Engineering Conference (ISOPE), 1520 June 2014, Busan,
Korea.
Diebold, L., Derbanne, Q. & Gazzola, T. 2013. Statistical behavior of global and local sloshing key
parameters. International Offshore and Polar Engineering Conference (ISOPE), 30 June - 5 July 2013,
Anchorage, USA.
Diznab, M.A., Mohajernassab, S., Seif, M.S., Tabeshpour, M.R. & Mehdigholi, H. 2014. Assessment of
offshore structures under extreme wave conditions by modified endurance wave analysis. Marine Structures, 39,
5069.
Dobashi, H. & Usami, A. 2012. Dynamic amplification factor of NO96 insulation structures of membrane system.
International Offshore and Polar Engineering Conference (ISOPE), 1722 June 2012, Rhodes, Greece.
Draganchev, H., Valchev, S., Pirovsky, C., Georgiev, M. & Mihaylov, B. 2012. Experimental and theoretical
research of noise emitted by merchant ships in port. The International Congress on Sound and Vibration (ICSV),
812 July 2012, Vilnius, Lithuania.
Drummen, I., Schiere, M., Dallinga, R. & Stambaugh, K. 2014. Full and model scale testing of a new class of US
coast guard cutter. The Ship Structure Committee Symposium: Vessel Safety & Longevity through Ship Structure
Research, 1820 May 2014, Linthicum, USA.
E Silva, S.R., Vsquez, G., Soares, C.G. & Marn, A. 2012. The stabilizing effects of U-Tanks as passive and
controlled anti-rolling devices. The American Society of Mechanical Engineers (ASME) Conference Proceedings
Ocean, Offshore and Arctic Engineering (OMAE), 16 July 2012, Rio de Janeiro, Brazil.
Erbe, C., Macgillivray, A. & Williams, R. 2012. Mapping cumulative noise from shipping to inform marine spatial
planning. The Journal of the Acoustical Society of America, 132, 423428.
Erbe, C., Mccauley, R., Mcpherson, C. & Gavrilov, A. 2013. Underwater noise from offshore oil production vessels.
The Journal of the Acoustical Society of America, 133, 465470.
European Parliament 2002. Directive EU/2002/49/EC relating to the assessment and management of environmental
noise, Brussel, Belgium, European Parliament.
European Parliament 2003. Directive EU/2003/10/EC on the minimum health and safety requirements
regarding the exposure of workers to the risks arising from physical agents (noise), Brussel, Belgium, European
Parliament.
Fillon, B., Henry, J., Baudin, E., Diebold, L., Parmentier, G. & Derbanne, Q. 2012. Influence of sampling rates on
sloshing pressures prediction. International Offshore and Polar Engineering Conference (ISOPE), 1722 June
2012, Rhodes, Greece.
Fillon, B., Henry, J., Diebold, L. & Derbanne, Q. 2013. Extreme values theory applied to sloshing pressure
peaks. International Offshore and Polar Engineering Conference (ISOPE), 30 June - 5 July 2013, Anchorage,
USA.
Fire and Blast Information Group 1999. Design guide for stainless steel blast walls: FABIG technical note 5,
London, UK, Steel Construction Institute.
Fischer, K. & Hring, I. 2009. SDOF response model parameters from dynamic blast loading experiments.
Engineering Structures, 31, 16771686.
Fossa, M., Rizzo, C.M., Tani, G. & Viviani, M. 2012. Simulations of a sloshing experiment by FEM CFD and FEM
FSI approaches. International Offshore and Polar Engineering Conference (ISOPE), 1722 June 2012, Rhodes,
Greece.
Frangopol, Dan, Bocchini, Dec, A., Kim, Kwon, Okasha, Nader & Saydam, D. 2013. Integrated life-cycle
framework for maintenance, monitoring, and reliability of naval ship structures. Naval Engineers Journal, 124,
8999.
Frangopol, D.M. & Soliman, M. 2014. Damage to ship structures under uncertainty: Evaluation and prediction. In:
Voyiadgis, G.Z. (ed.) Handbook of Damage Mechanics. New York, USA: Springer.
Freire, C.M., Korkischko, I. & Meneghini, J.R. 2013. Interference effect in the flow around two aligned cylinders
and the reduction of a helical strakes suppressor effectiveness. The American Society of Mechanical Engineers
France.
Fricke, W. & Paetzold, H. 2012. Experimental investigation of the effect of whipping stresses on the fatigue life of
ships. The International Marine Design Conference, 811 May 2012, Glasgow, Scotland.
whipping stresses. The International Symposium on Practical Design of Ships and Other Floating Structures
(PRADS), 2025 October 2013, Changwon, Korea.
263
Fricke, W. & Paetzold, H. 2014. Effect of whipping stresses on the fatigue damage of ship structures. Welding in the
World, 58, 261268.
Fujarra, A.L.C., Rosetti, G.F., De Wilde, J. & Gonalves, R.T. 2012. State-of-art on vortex-induced motion: A
comprehensive survey after more than one decade of experimental investigation. The American Society of
Mechanical Engineers (ASME) Conference Proceedings Ocean, Offshore and Arctic Engineering (OMAE), 16
July 2012, Rio de Janeiro, Brazil.
Fukasawa, T. & Mukai, K. 2013. On the effects of hullgirder vibration upon fatigue strength of a postPanamax
container ship dis-aggregated by shortterm sea state. The International Symposium on Practical Design of Ships
and Other Floating Structures (PRADS), 2025 October 2013, Changwon, Korea.
Gagnon, R.E. 2011. An inside look at ice-crushing induced vibration and lockin. The 21st International Conference
on Port and Ocean Engineering under Arctic Conditions, POAC11, 1014 July 2011, Montreal, Canada.
Gagnon, R.E. 2012a. An explanation for the Molikpaq May 21, 1986 event. Cold Regions Science and Technology,
82, 7593.
Gagnon, R.E. 2012b. An explanation of icecrushing induced vibration lockin using the Molikpaq may 12, 1986
event as a test case. The 21st International Association for HydroEnvironment Engineering and Research
(IAHR) International Symposium on Ice, 1115 June 2012, Dalian, China.
Gagnon, R.E. & Wang, J. 2012. Numerical simulations of a tanker collision with a bergy bit incorporating
hydrodynamics, a validated ice model and damage to the vessel. Cold Regions Science and Technology, 81, 26
35.
Gan, C.B., Guo, S.Q., Lei, H. & Yang, S.X. 2014. Random uncertainty modeling and vibration analysis of a straight
pipe conveying fluid. Nonlinear Dynamics, 77, 503519.
Gao, R.P., Tay, Z.Y., Wang, C.M. & Koh, C.G. 2011. Hydroelastic response of very large floating structure with a
flexible line connection. Ocean Engineering, 38, 19571966.
Gazzola, T. & Diebold, L. 2013. Dynamic probes: An on-the-fly CFD plug-in for sloshing impact
assessment. International Offshore and Polar Engineering Conference (ISOPE), 30 June - 5 July 2013,
Anchorage, USA.
Germanischer Lloyd 2012. I Ship technology, part I Seagoing ships, chapter I Hull structures. Rules for
classification and construction. Hamburg, Germany: Germanischer Lloyd.
Germanischer Lloyd Aktiengesellschaft 2013. GL guidelines for compliance with MLC 2006 noise and vibration
requirements rules for classification and construction, Hamburg, Germany, Germanischer Lloyd
Aktiengesellschaft.
Gonalves, R.T., Franzini, G.R., Rosetti, G.F., Fujarra, A.L.C. & Nishimoto, K. 2012a. Analysis methodology for
vortexinduced motions (VIM) of a monocolumn platform applying the HilbertHuang transform method.
Journal of Offshore Mechanics and Arctic Engineering, 134, 011103.
Gonalves, R.T., Rosetti, G.F., Fujarra, A.L.C. & Nishimoto, K. 2012b. An overview of relevant aspects on VIM of
spar and monocolumn platforms. Journal of Offshore Mechanics and Arctic Engineering, 134, 014501.
Grigoriu, M. 2013. Solution of linear dynamic systems with uncertain properties by stochastic reduced order
models. Probabilistic Engineering Mechanics, 34, 168176.
Guan, Z.W., Aktas, A., Potluri, P., Cantwell, W.J., Langdon, G. & Nurick, G.N. 2014. The blast resistance of
stitched sandwich panels. International Journal of Impact Engineering, 65, 137145.
Guedes Soares, C. 2012. Probabilistic methods for structural design (solid mechanics and its application), New
York, USA, Springer.
Guilcher, P.M., Brosset, L., Couty, N. & Le Touz, D. 2012. Simulations of breaking wave impacts on a rigid wall
at two different scales with a two phase fluid compressible SPH model. International Offshore and Polar
Engineering Conference (ISOPE), 1722 June 2012, Rhodes, Greece.
Guilcher, P.M., Jus, Y., Couty, N., Brosset, L., Scolan, Y.M. & Le Touz, D. 2014. 2D simulations of breaking
wave impacts on a flat rigid wall part 1: Influence of the wave shape. International Offshore and Polar
Engineering Conference (ISOPE), 1520 June 2014, Busan, Korea.
Guo, F. 2012a. Reanalysis of ice induced steady state vibration from an engineering perspective. The 21st
International Association for HydroEnvironment Engineering and Research (IAHR) International Symposium
on Ice, 1115 June 2012, Dalian, China.
Guo, F. 2012b. A spectral model for simulating continuous crushing ice load. The 21st International Association for
HydroEnvironment Engineering and Research (IAHR) International Symposium on Ice, 1115 June 2012,
Dalian, China.
Guo, F. 2013. Analysis of the key parameters in ice induced frequency lockin. The 22nd International Conference
on Port and Ocean Engineering under Arctic Conditions, POAC13, 913 June 2013, Espoo, Finland.
Guo, F. & Yue, Q. 2011. Criterion of ice induced self excited vibration. The 21st International Conference on Port
and Ocean Engineering under Arctic Conditions, POAC11, 1014 July 2011, Montreal, Canada.
Hageman, R., Drummen, I., Stambaugh, K., Dupau, T., Herel, N., Derbanne, Q., Schiere, M., Shin, Y. & Kim, P.
2014. Structural fatigue loading predictions and comparisons with test data for a new class of US coast guard
cutters. The Ship Structure Committee Symposium: Vessel Safety & Longevity through Ship Structure Research,
1820 May 2014, Linthicum, USA.
Halkyard, J., Atluri, S. & Sirnivas, S. 2006. Truss spar vortex induced motions: Benchmarking of CFD and model
tests. The American Society of Mechanical Engineers (ASME) Conference Proceedings Ocean, Offshore and
Arctic Engineering (OMAE), 49 June 2006, Hamburg, Germany.
264
Halkyard, J., Sirnivas, S., Holmes, S., Constantinides, Y., Oakley, O.H. & Thiagarajan, K. 2005. Benchmarking of
truss spar vortex induced motions derived from CFD with experiments. The American Society of Mechanical
Engineers (ASME) Conference Proceedings Ocean, Offshore and Arctic Engineering (OMAE), 1217 June
2005, Halkidiki, Greece.
Hashimoto, H. & Le Touz, D. 2014. A coupled MPSFEM model for the analysis of violent flowsstructures
interaction. The International Workshop on Water Waves and Floating Bodies (IWWWFB), 30 March - 2 April
2014, Osaka, Japan.
He, G. & Kashiwagi, M. 2014. Numerical simulation of fluidstructure interaction by an adaptive Cartesiangrip
CID method. The International Workshop on Water Waves and Floating Bodies (IWWWFB), 30 March - 2 April
2014, Osaka, Japan.
Hendrikse, H., Kuiper, G.L. & Metrikine, A.V. 2011. Ice induced vibrations of flexible offshore structures: The
effect of load randomness, high ice velocities and higher structural modes. The 21st International Conference on
Port and Ocean Engineering under Arctic Conditions, POAC11, 1014 July 2011, Montreal, Canada.
Hendrikse, H. & Metrikine, A.V. 2013. The influence of friction at icestructure interaction on ice induced
vibration. The 22nd International Conference on Port and Ocean Engineering under Arctic Conditions,
POAC13, 913 June 2013, Espoo, Finland.
Hendrikse, H., Metrikine, A.V. & Evers, K.U. 2012. A method to measure the added mass and added damping in
dynamic icestructure interaction: Deciphering ice induced vibration, part 3. The 21st International Association
for HydroEnvironment Engineering and Research (IAHR) International Symposium on Ice, 1115 June 2012,
Dalian, China.
Hermundstad, O.A. & Moan, T. 2005. Numerical and experimental analysis of bow flare slamming on a RoRo
vessel in regular oblique waves. Journal of Marine Science and Technology, 10, 105122.
Hermundstad, O.A., Stansberg, C.T. & Hellan, O.Y. 2002. Numerical and experimental analysis of extreme
slamming loads on FPSO bows. The American Society of Mechanical Engineers (ASME) Conference
Proceedings Ocean, Offshore and Arctic Engineering (OMAE), 2328 June 2002, Oslo, Norway.
Hetmanczyk, S., Heinonen, J. & Strobel, M. 2011. Dynamic ice load model in overall simulation of offshore
wind turbines. International Offshore and Polar Engineering Conference (ISOPE), 1924 June 2011, Maui,
USA.
Hirdaris, S.E., Bai, W., Dessi, D., Ergin, A., Gu, X., Hermundstad, O.A., Huijsmans, R., Iijima, K., Nielsen,
U.D. & Parunov, J. 2014. Loads for use in the design of ships and offshore structures. Ocean Engineering, 78,
131174.
Hong, S.Y., Kim, B.W. & Nam, B.W. 2011. Experimental study on torsion springing and whipping of a large
container ship. International Offshore and Polar Engineering Conference (ISOPE), 1924 June 2011, Maui,
USA.
Hong, S.Y., Kim, K.H., Kim, B.W. & Kim, Y.S. 2014. Experimental study on the bow-flare slamming of a 10,000
TEU containership. International Offshore and Polar Engineering Conference (ISOPE), 1520 June 2014,
Busan, Korea.
Hu, A.K., Lin, Y. & Jiang, W. 2013. The wind tunnel experiment study of wind load on jack-up drilling unit. The
Engineering (OMAE), 914 June 2013, Nantes, France.
Huang, X., Liu, X., Sun, J., Zhang, Z. & Hua, H. 2014. Effect of the system imperfections on the dynamic response
of a high-static-low-dynamic stiffness vibration isolator. Nonlinear Dynamics, 76, 11571167.
Huang, X., Xu, S., Jiang, A. & Hua, H. 2012. Modeling and optimization of floating raft systems in
submarines under different objectives by using hybrid genetic algorithm. Journal of Vibration and Control 18,
268297.
Huang, Y., Ma, J. & Tian, Y. 2013a. Model tests of four-legged jacket platforms in ice: Part 1. model tests and
results. Cold Regions Science and Technology, 95, 7485.
Huang, Y., Yu, M. & Tian, Y. 2013b. Model tests of four-legged jacket platforms in ice: Part 2. analyses and
discussions. Cold Regions Science and Technology, 95, 86101.
Huijs, F., De Ridder, E.J. & Savenije, F. 2014. Comparison of model tests and coupled simulations for a
semi-submersible floating wind turbine. The American Society of Mechanical Engineers (ASME)
Conference Proceedings Ocean, Offshore and Arctic Engineering (OMAE), 813 June 2014, San Francisco,
USA.
Hwang, J.O., Chun, S.E., Joh, K.H., Cambos, P., De Lauzon, J., White, N., Kim, M.S., Park, J.B. & Lee, J.M. 2014.
Direct assessment of structural capacity against sloshing using dynamic nonlinear FE analysis. International
Offshore and Polar Engineering Conference (ISOPE), 1520 June 2014, Busan, Korea.
Hwang, S.Y., Lee, J.H. & Kim, S.C. 2012. Simplified impinging jet model for practical sloshing assessment of LNG
cargo containment. International Offshore and Polar Engineering Conference (ISOPE), 1722 June 2012,
Rhodes, Greece.
Hwang, S.Y., Lee, J.H. & Kim, S.C. 2014. Effect of gas bubble on the sloshing pressure simulated by a multiphase
flow of droplet and liquid. International Offshore and Polar Engineering Conference (ISOPE), 1520 June 2014,
Busan, Korea.
Hwang, S.Y., Lee, J.H., Kim, S.C., Nho, I.S., Jang, B.S. & Kim, Y. 2013a. Structural response analysis of LNG
CCS experiencing the sloshing impact determined by both convolution and fluid structure interaction method.
265
Hwang, S.Y., Lee, J.H., Nho, I.S., Kim, S.C. & Kim, Y. 2013b. Sloshing assessment procedure using structural
responses for membrane type LNG CCS. The International Symposium on Practical Design of Ships and Other
Floating Structures (PRADS), 2025 October 2013, Changwon, Korea.
Iijima, K. & Fujikubo, M. 2012. Impacts of waveinduced vibrations on ultimate strength and fatigue strength of
ships. The International Conference on Hydroelasticity in Marine Technology, 1921 September 2012, Tokyo,
Japan.
Iijima, K., Hermundstad, O.A., Zhu, S. & Moan, T. 2009. Symmetric and antisymmetric vibrations of a
hydroelastically scaled model. The International Conference on Hydroelasticity in Marine Technology, 810
September 2009, Southampton, UK.
Iijima, K., Kimura, K., Xu, W. & Fujikubo, M. 2011. Hydroelasto-plasticity approach to predicting the post-ultimate
strength behavior of a ships hull girder in waves. Journal of Marine Science and Technology, 16, 379389.
Iijima, K., Suzaki, Y. & Fujikubo, M. 2013. Scale model tests for the postultimate strength collapse behavior of a
ships hull girder under whip-ping loads. The International Conference on Marine Structures (MARSTRUCT),
2527 March 2013, Espoo, Finland.
International Association of Classification Societies (Iacs) 2013. Guidelines No 132 human element
recommendations for structural design of lighting, ventilation, vibration, noise, access and egress arrangements,
London, UK, IACS.
International Labour Organization 2006. Maritime labour convention 2006, Geneva, Switzerland, ILO.
International Maritime Organization 1982. Resolution A.468 (XII): Code on noise levels on board ship, London,
UK, IMO.
International Maritime Organization 2012a. Maritime safety committee resolution 337(91): Adoption of the code on
noise levels on board ships, London, UK, IMO.
International Maritime Organization 2012b. Maritime safety committee resolution 338(91): Adoption of amendments
to the international convention for the safety of life at sea, 1974, as amended, London, UK, IMO.
International Maritime Organization 2014. Marine environment protection committee Circ (833): Guidelines for
the reduction of underwater noise from commercial shipping to address adverse impacts on marine life, London,
UK, IMO.
International Organization for Standardization 1989. ISO 90521 Acoustics Determination of dynamic stiffness
Part 1: Materials used under floating floors in dwellings, Geneva, Switzerland, ISO.
International Organization for Standardization 2010a. ISO 3746 Acoustics Determination of sound power levels
and sound energy levels of noise sources using sound pressure Survey method using an enveloping
measurement surface over a reflecting plane, Geneva, Switzerland, ISO.
International Organization for Standardization 2010b. ISO 19906 Petroleum and natural gas industries Arctic
offshore structures, Geneva, Switzerland, ISO.
International Organization for Standardization 2012a. ISO 202834 Mechanical vibration Measurement of
vibration on ships Part 4: Measurement and evaluation of vibration of the ship propulsion machinery, Geneva,
Switzerland, ISO.
International Organization for Standardization 2012b. ISO/PAS 172081 Acoustics Quantities and procedures for
description and measurement of under-water sound from ships Part 1: General requirements for measurements
in deep water, Geneva, Switzerland, ISO.
International Organization for Standardization 2014. ISO/DIS 16554.3 Ship and marine technology Measurement
and reporting of underwater sound radiated from merchant ships deepwater measurement, Geneva,
Switzerland, ISO.
tibar, M.D., Unal, N.E. & Akyildiz, H. 2012. Numerical simulation of sloshing with SPH. International Offshore
Jacobi, G., Thomas, G., Davis, M.R. & Davidson, G. 2014. An insight into the slamming behaviour of large
highspeed catamarans through fullscale measurements. Journal of Marine Science and Technology 19,
1532.
Jameel, M., Ahmad, S., Islam, A.S. & Jumaat, M.Z. 2014. Wind induced nonlinear response of coupled spar
platform. The American Society of Mechanical Engineers (ASME) Conference Proceedings Ocean, Offshore
Jensen, J.J. 2010. Extreme value predictions using Monte Carlo simulations with artificially increased wave height.
The International Symposium on Practical Design of Ships and Other Floating Structures (PRADS), 1924
September Rio de Janeiro, Brazil.
Jensen, J.J. & Pedersen, T.P. 2009. Estimation of hull girder vertical bending moments including nonlinear and
flexibility effects using closed form expressions. Journal Engineering for the Marine Environment, 223, 337
390.
Jeon, J.H. & Joo, W.H. 2014. Prediction of propeller radiated noise by onboard measurement. The 2nd International
Conference and Exhibition on Underwater Acoustics, 2227 July 2014, Rhodes, Greece.
Jeong, H.K. & Yang, Y.S. 2013. Strength assessment of LNG CCS plate using failure criteria for composite
materials. The International Symposium on Practical Design of Ships and Other Floating Structures (PRADS),
2025 October 2013, Changwon, Korea.
Jeong, H.K. & Yang, Y.S. 2014. Strength assessment of LNG CCS plate using progressive damage modeling.
266
Jeong, S.M., Hwang, S.H., Nam, J.W. & Park, J.C. 2012. Numerical simulation of impact loads by sloshing in a 3D
rectangular tank using eulerian and lagrangian approaches. International Offshore and Polar Engineering
Conference (ISOPE), 1722 June 2012, Rhodes, Greece.
Jeong, W., Lee, B., Kwon, J. & Kim, M. 2013. A study on vibration characteristic of stiffened plates with fluid
coupling effect in a tank. The International Symposium on Practical Design of Ships and Other Floating
Structures (PRADS), 2025 October 2013, Changwon, Korea.
Jia, J. 2014. Investigations of a practical windinduced fatigue calculation based on nonlinear time domain dynamic
analysis and a full winddirectional scatter diagram. Ships and Offshore Structures, 9, 272296.
Jiang, L.S. & White, N. 2012. Nonlinear finite element dynamic collapse analyses of stiffened panels. The
International Conference on Hydroelasticity in Marine Technology, 1921 September 2012, Tokyo, Japan.
Jin, B.M., Kim, N.S. & Park, J.H. 2010. Stiffness estimation considering coupling of main engine platform and top
bracing. The Symposium on Korean Society of Ocean Engineers, 34 June 2010, Jeju, Korea.
Jing, L., Wang, Z., Shim, V.P.W. & Zhao, L. 2014. An experimental study of the dynamic response of cylindrical
sandwich shells with metallic foam cores subjected to blast loading. International Journal of Impact Engineering,
71, 6072.
Jussila, V. & Heinonen, J. 2012. Comparison of ice-induced vibration on a conical and a cylindrical offshore wind
turbine substructure. The 21st International Association for HydroEnvironment Engineering and Research
(IAHR) International Symposium on Ice, 1115 June 2012, Dalian, China.
Kahl, A., Fricke, W., Paetzold, H. & Von Selle, H. 2014. Whipping investigations based on large-scale
measurements and experimental fatigue testing. International Offshore and Polar Engineering Conference
(ISOPE), 1520 June 2014, Busan, Korea.
Kahl, A., Rathje, H., Rorup, J. & Schellin, T.E. 2013. Semiempirical assessment of longterm highfrequency hull
girder response of containerships an update. In: Soares, C.G. & Romanoff, J. (eds.) Analysis and Design of
Marine Structures. London, UK: Taylor & Francis Group.
Kang, Z., Zhang, X., Jiang, S. & Cheng, G. 2012. On topology optimization of damping layer in shell structures
under harmonic excitations. Structural and Multidisciplinary Optimization, 46, 5167.
Karimi, M.R., Brosset, L., Ghidaglia, J.M. & Kaminski, M.L. 2014. A study on conservatism of froude scaling for
sloshing model tests. International Offshore and Polar Engineering Conference (ISOPE), 1520 June 2014,
Busan, Korea.
Karimi, M.R., Kosinski, C. & Brosset, L. 2013. Comparison of sloshing model test results at scales 1: 10 and
1: 40. International Offshore and Polar Engineering Conference (ISOPE), 30 June5 July 2013, Anchorage,
USA.
Karna, T., Andersen, H., Gurtner, A., Metrikine, A., Sodhi, D., Loo, M., Kuiper, G., Gibson, R., Fenz, D.,
Muggeridge, K., Wallenburg, C., Wu, J.F. & Jefferies, M. 2013. Iceinduced vibration of offshore structures
Looking beyond ISO 19906. The 22nd International Conference on Port and Ocean Engineering under Arctic
Conditions, POAC13, 913 June 2013, Espoo, Finland.
Kashiwagi, M. & Hara, T. 2012. A method for ship hydroelasticity analysis by means of Rankine panel
method. The International Conference on Hydroelasticity in Marine Technology, 1921 September 2012, Tokyo,
Japan.
Kayal, B. & Berthon, C.F. 2013. Analytical approach to predict sloshing severity in LNG membrane tanks based on
optimized series of model tests. International Offshore and Polar Engineering Conference (ISOPE), 30 June - 5
July 2013, Anchorage, USA.
Kellett, P., Turan, O. & Incecik, A. 2013. A study of numerical ship underwater noise prediction. Ocean
Kerkeni, S., Dal Santo, X. & Metrikin, I. 2013. Dynamic positioning in ice: Comparison of control laws in open
water and ice. The American Society of Mechanical Engineers (ASME) Conference Proceedings Ocean,
Offshore and Arctic Engineering (OMAE), 914 June 2013, Nantes, France.
Kim, B.W., Kim, K.H., Kim, Y.S. & Hong, S.Y. 2014. Torsion moment conversion methods in model test with Ushape backbone. International Offshore and Polar Engineering Conference (ISOPE), 1520 June 2014, Busan,
Korea.
Kim, C.H. & Shin, Y.S. 2013. Numerical simulation of surface shield effects to waterblast wave. Ocean
Kim, D.H., Jeong, H.B., Choi, S.H., Jo, Y.C., Kim, D.E. & Jeong, T.S. 2014. Structural assessment under sloshing
impact for the IMO type B independent LNG tank. International Offshore and Polar Engineering Conference
(ISOPE), 1520 June 2014, Busan, Korea.
Kim, J.H., Kim, K.H., Lee, D.Y., Jung, B.H. & Kim, Y.H. 2013a. A fully coupled BEMFEM analysis on ship
structural hydroelasticity and experimental validation. The International Symposium on Practical Design of Ships
and Other Floating Structures (PRADS), 2025 October 2013, Changwon, Korea.
Kim, J.H. & Kim, Y.H. 2014. Numerical analysis on springing and whipping using fully-coupled FSI models.
Kim, J.H., Kim, Y.H., Kang, B.C. & Kim, Y.I. 2012. Ship springing analysis for very large container ship.
International Journal of Offshore and Polar Engineering, 22, 217224.
Kim, J.H., Kim, Y.H., Kim, Y.I. & Kim, K.H. 2013b. Study on ship structural hydroelasticity and fatigue
assessment in irregular seaways. The American Society of Mechanical Engineers (ASME) Conference
Proceedings Ocean, Offshore and Arctic Engineering (OMAE), 914 June 2013, Nantes, France.
267
Kim, J.M., Oh, M.H., Park, T.H., Moon, J.S., Sim, W.S. & Lee, M.S. 2013. Structural safety assessment of FLNG
cargo containment systems against sloshing loads. The Offshore Technology Conference (OTC), 69 May 2013,
Houston, USA.
Kim, K.H., Bang, J.S., Kim, J.H., Kim, Y.H., Kim, S.J. & Kim, Y.I. 2013. Fully coupled BEM-FEM analysis for
ship hydroelasticity in waves. Marine Structures, 33, 7199.
Kim, S.J. & Paik, J.K. 2013. Advanced method for ship structural design under slamming impact pressure loads.
Kim, S.R., Kim, H.S., Kim, J.S., Kim, B.K. & Lee, S.H. 2012. Sound insulation performance of prefabricated cabins
in cruise ships. Noise Control Engineering Journal, 60, 184191.
Kim, S.Y., Ahn, Y., Kim, K.H. & Kim, Y. 2013. Experimental studies on sloshing in a STX independence typeB
tank. International Offshore and Polar Engineering Conference (ISOPE), 30 June - 5 July 2013, Anchorage,
USA.
Kim, Y., Kim, S.Y., Ahn, Y. & Kim, K.H. 2013. Experimental analysis on sloshing loads for LNG cargo design.
Kim, Y.H. & Hermansky, G. 2014. Uncertainties in seakeeping analysis and related loads and response procedures.
Kim, Y.I., Kim, J.H. & Kim, Y.H. 2013. Whipping identification of a flexible ship using wavelet cross-correlation.
Kim, Y.I. & Park, S.G. 2014. Wet damping estimation of the scaled segmented hull model using the random
decrement technique. Ocean Engineering, 75, 7180.
Kimball, R., Goupee, A.J., Fowler, M.J., De Ridder, E.J. & Helder, J. 2014. Wind/wave basin verification of a
performancematched scale model wind turbine on a floating offshore wind turbine platform. The American
(OMAE), 813 June 2014, San Francisco, USA.
Kimmoun, O., Ratouis, A. & Brosset, L. 2012. Influence of a bubble curtain on the impact of waves on a
vertical wall. International Offshore and Polar Engineering Conference (ISOPE), 1722 June 2012, Rhodes,
Greece.
Kobayakawa, H., Kusumoto, H., Nagashima, T. & Neki, I. 2012. Hydroelastic response analysis using unsteady
time domain analysis of ship motions. The International Conference on Hydroelasticity in Marine Technology,
1921 September 2012, Tokyo, Japan.
Kong, X.S., Wu, W.G., Li, J., Chen, P. & Liu, F. 2014. Experimental and numerical investigation on a multi-layer
protective structure under the synergistic effect of blast and fragment loadings. International Journal of Impact
Koning, J. & Kapsenberg, G.K. 2012. Full scale container ship cross section loads first results. The International
Conference on Hydroelasticity in Marine Technology, 1921 September 2012, Tokyo, Japan.
Koo, B., Goupee, A.J., Lambrakos, K. & Lim, H.J. 2014. Model test data correlations with fully coupled
hull/mooring analysis for a floating wind turbine on a semi-submersible platform. The American Society of
13 June 2014, San Francisco, USA.
Koo, J.B., Jang, K.B., Suh, Y.S., Kim, Y.S., Kim, M.S., Yu, H. & Tai, J.S.C. 2011. Fatigue damage assessment
based on full scale measurement data for a large container carrier. International Offshore and Polar Engineering
Conference (ISOPE), 1924 June 2011, Maui, USA.
Korkut, E. & Atlar, M. 2012. An experimental investigation of the effect of foul release coating application on
performance, noise and cavitation characteristics of marine propellers. Ocean Engineering, 41, 112.
Korobkin, A.A. 2012. A model of body exit from water. . The International Workshop on Springing and Whipping
of Ships, 810 November 2012, Split, Croatia.
Korobkin, A.A. & Malenica, . 2005. Modified Logvinovich model for hydrodynamic loads on asymmetric
contours entering water. The International Workshop on Water Waves and Floating Bodies (IWWWFB), 29 May 1 June 2005, Longyearbyen, Norway.
Korobkin, A.A. & Malenica, . 2006. Local hydroelastic models for sloshing impacts. BV Technical note,
NT2912.
Kuhn, C., Bruns, B., Fischer, J., Gattermann, J. & Elmer, K.H. 2012. Development of a new underwater piling noise
mitigation system: Using hydro sound dampers (HSD). The American Society of Mechanical Engineers (ASME)
Brazil.
Kujala, P. & Arughadhoss, S. 2012. Statistical analysis of ice crushing pressures on a ships hull during hullice
interaction. Cold Regions Science and Technology, 70, 111.
Kundu, A. & Adhikari, S. 2014. Transient response of structural dynamic systems with parametric uncertainty.
Journal of Engineering Mechanics, 140, 315331.
Laakso, A., Romanoff, J., Remes, H. & Niemel, A. 2013. An analytical method for cabin deck fundamental
frequency, an update. In: Soares, C. & Romanoff, J. (eds.) Analysis and Design of Marine Structures. London,
UK: Taylor & Francis Group.
268
Lafeber, W., Bogaert, H. & Brosset, L. 2012a. Elementary loading processes (ELP) involved in breaking wave
impacts: Findings from the sloshel project. International Offshore and Polar Engineering Conference (ISOPE),
Lafeber, W., Brosset, L. & Bogaert, H. 2012b. Comparison of wave impact tests at large and full scale: Results from
the sloshel project. International Offshore and Polar Engineering Conference (ISOPE), 1722 June 2012,
Rhodes, Greece.
Lan, Z., Zhao, Y., Guo, J., Zhou, Z., Xing, J., Bai, G., Lu, H. & Li, H. 2011. The measurement and analysis of ice
induced vibration of JZ93W HPB oil platform in Bohai Sea. Engineering Science, 13, 8087.
Lavroff, J. 2009. The slamming and whipping vibratory response of a hydroelastic segmented catamaran model.
Ph.D. thesis, University of Tasmania.
Lavroff, J., Davis, M.R., Holloway, D.S. & Thomas, G. 2013. Wave slamming loads on wave-piercer catamarans
operating at high-speed determined by hydro-elastic segmented model experiments. Marine Structures, 33, 120
142.
Lee, A.K., Wang, G., Ternowchek, S. & Botten, S.F. 2014. Structural health monitoring on ships using acoustic
emission testing. The Ship Structure Committee Symposium: Vessel Safety & Longevity through Ship Structure
Lee, C.H., Kim, H.W. & Joo, W.H. 2014. Low frequency vibration control of the offshore fixed platforms by using
tuned liquid column damper. The American Society of Mechanical Engineers (ASME) Conference Proceedings
Lee, C.W., Kim, J.H., Lim, G.S., Lee, D.K. & Chung, T.S. 2010. Study on accuracy improvement of full ship
analysis considering dynamic characteristics of hydraulic topbracing. The Symposium on Korean Society of
Ocean Engineers, 34 June 2010, Jeju, Korea.
Lee, D.J., Shin, S.B. & Suhr, J. 2014. Numerical model for dynamic behavior of PUF in membrane type LNG cargo
containment system. International Offshore and Polar Engineering Conference (ISOPE), 1520 June 2014,
Busan, Korea.
Lee, J.H., Han, J.M., Park, H.G. & Seo, J.S. 2013. Improvements of model-test method for cavitation-induced
pressure fluctuation in marine propeller. Journal of Hydrodynamics, 25, 599605.
Lee, J.H., Park, H.G., Kim, J.H., Lee, K.J. & Seo, J.S. 2014. Reduction of propeller cavitation induced hull exciting
pressure by a reflected wave from air-bubble layer. Ocean Engineering, 77, 2332.
Lee, K.H., Lee, J., Kim, D., Kim, K. & Seong, W. 2014. Propeller sheet cavitation noise source modeling and
inversion. Journal of Sound and Vibration, 333, 13561368.
Lee, S.G., Kim, J.K., Nguyen, H.A. & Nam, J.H. 2012. Structural safety assessment of LNGC MARK III membrane
type CCS under sloshing impact loading. International Offshore and Polar Engineering Conference (ISOPE),
Lee, S.G., Lee, H.S., Choi, G.G. & Han, S. 2014. Shock response analysis of blast hardened bulkhead in part
chamber model of naval ship under internal blast. The AsianPacific Technical Exchange and Advisory Meeting
on Marine Structures (TEAM), 1316 October 2014, Istanbul, Turkey.
Lee, S.G. & Zhao, T. 2013. Structural safety assessment of LNGC CCS under iceberg collision using FSI analysis
technique. International Offshore and Polar Engineering Conference (ISOPE), 30 June - 5 July 2013,
Anchorage, USA.
Lee, S.K., Yu, K. & Tseng, R.K.C. 2012. CFD simulation of slamming loads on a containership fitted with energy
saving rudder fins. International Offshore and Polar Engineering Conference (ISOPE), 1722 June 2012,
Rhodes, Greece.
Lee, Y., White, N., Wang, Z., Hirdaris, S.E. & Zhang, S. 2011. Comparison of springing and whipping responses
of model tests with predicted nonlinear hydroelastic analyses. The 21st International Association for
HydroEnvironment Engineering and Research (IAHR) International Symposium on Ice, 1924 June 2011, Maui,
USA.
Lee, Y.W., White, N., Wang, Z., Tong, J., Xiao, Y. & Li, Q. 2014. Springing loads and fatigue assessment on large
container ships. International Offshore and Polar Engineering Conference (ISOPE), 1520 June 2014, Busan,
Korea.
Lee, Y.W., White, N., Wang, Z.H. & Park, J.B. 2012. Whipping responses and whipping effects on design bending
moments of a large container ship. The International Conference on Hydroelasticity in Marine Technology, 19
21 September 2012, Tokyo, Japan.
Lefevre, C., Constantinides, Y., Kim, J.W., Henneke, M., Gordon, R., Jang, H.C. & Wu, G. 2013. Guidelines for
CFD simulations of spar VIM. The American Society of Mechanical Engineers (ASME) Conference Proceedings
Lei, J. 2013. Optimal vibration control for uncertain nonlinear sampleddata systems with actuator and sensor
delays: Application to a vehicle suspension. Journal of Dynamic Systems, Measurement, and Control, 135,
021021.
Li, X., Wang, Z., Zhu, F., Wu, G. & Zhao, L. 2014. Response of aluminium corrugated sandwich panels
under air blast loadings: Experiment and numerical simulation. International Journal of Impact Engineering, 65,
7988.
Lim, S.J., Kong, C.D. & Park, H.B. 2013. A study on optimal design of filament winding composite tower for 2
MW class horizontal axis wind turbine system. International Journal of Composite Materials, 3, 1523.
Lin, L., Zhong, X. & Wu, Y. 2014. The drag, lift and Strouhal number of a circularsection cylinder with a
conical disturbance at subcritical Reynolds number. The American Society of Mechanical Engineers (ASME)
269
USA.
Lindberg, O., Bingham, H.B. & Engsig-Karup, A.P. 2012. A coupled finite difference and moving least squares
simulation of violent breaking wave impact. The American Society of Mechanical Engineers (ASME) Conference
Proceedings Ocean, Offshore and Arctic Engineering (OMAE), 16 July 2012, Rio de Janeiro, Brazil.
Lippert, S., Lippert, T. & Estorff, O.V. 2012. Prediction of underwater sound due to pile driving for offshore wind
farms a challenge for numerical simulation. The 41st International Congress and Exposition on Noise Control
Engineering, 1922 August 2012, New York.
Liu, J.H., Wu, Y.S., Wang, H.K. & Pan, J.Q. 2012. Application of the loading inherent subspace scaling method on
the whipping responses test of a surface ship to underwater explosions. The International Conference on
Hydroelasticity in Marine Technology, 1921 September 2012, Tokyo, Japan.
Longre, P., Geffroy-Grze, A.G., Lebl, B. & Dragon, A. 2013. Ship structure steel plate failure under nearfield air-blast loading: numerical simulations vs experiment. International Journal of Impact Engineering, 62,
8898.
Loysel, T., Chollet, S., Gervaise, E., Brosset, L. & De Seze, P.E. 2012. Results of the first sloshing model test
benchmark. International Offshore and Polar Engineering Conference (ISOPE), 1722 June 2012, Rhodes,
Greece.
Loysel, T., Gervaise, E., Moreau, S. & Brosset, L. 2013. Results of the 20122013 sloshing model test
benchmark. International Offshore and Polar Engineering Conference (ISOPE), 30 June - 5 July 2013,
Anchorage, USA.
Lugni, C., Bardazzi, A., Faltinsen, O.M. & Graziani, G. 2012. Hydroelastic slamming response in the evolution of a
flip-through event during shallow-liquid sloshing. Physics of Fluids, 26, 032108.
Lugni, C., Bardazzi, A., Faltinsen, O.M. & Graziani, G. 2013. Hydroelastic challenges for wave-impact phenomena
in sloshing flow. The American Society of Mechanical Engineers (ASME) Conference Proceedings Ocean,
Luppes, R., Van Der Heiden, H.J.L., Van Der Plas, P., Veldman, A.E.P. & Duz, B. 2013. Simulations of wave
impact and twophase flow with Comflow: Past and recent developments. International Offshore and Polar
Engineering Conference (ISOPE), 30 June - 5 July 2013, Anchorage, USA.
Lynch, J.P. & Loh, K.J. 2006. A summary review of wireless sensors and sensor networks for structural health
monitoring. Shock and Vibration Digest, 38, 91130.
Ma, L., Xin, J., Hu, Y. & Zheng, J. 2013. Ductile and brittle failure assessment of containment vessels subjected to
internal blast loading. International Journal of Impact Engineering, 52, 2836.
Ma, X., Jin, G. & Liu, Z. 2014. Active structural acoustic control of an elastic cylindrical shell coupled to a twostage vibration isolation system. International Journal of Mechanical Sciences, 79, 182194.
Maattanen, M., Loset, S., Metrikine, A., Evers, K.U., Hendrikse, H., Lonoy, C., Metrikin, I., Nord, T. &
Sukhorukov, S. 2012. Novel ice vibration testing in a largescale facility: Deciphering ice induced vibration, part
1. The 21st International Association for HydroEnvironment Engineering and Research (IAHR) International
Symposium on Ice, 1115 June 2012, Dalian, China.
Malenica, S. & Derbanne, Q. 2012. Hydroelastic issues in the design of ultra large container ships TULCS
project. The International Conference on Hydroelasticity in Marine Technology, 1921 September 2012, Tokyo,
Japan.
Malenica, S. & Diebold, L. 2013. Evaluation of sloshing impact loading and structural responses where we are?
The International Workshop on Ship Hydrodynamics, 2325 September 2013, Seoul, Korea.
Mao, W. & Rychlik, I. 2012. Estimation of extreme ship response. Journal of Ship Research, 56, 2334.
Marchesini, A. & Piana, E. 2012. Noise reduction on yachts. The International Congress on Sound and Vibration
(ICSV), 812 July 2012, Vilnius, Lithuania.
Mariani, R. & Dessi, D. 2012. Analysis of the global bending modes of a floating structure using the proper
orthogonal decomposition. Journal of Fluids and Structures, 28, 115134.
Maritime and Ocean Engineering Research Institute 2013. Wave induced loads on ships Joint industry project III
Interim report. MOERI Technical Report BSPIS7230103066. Daejeon, Korea: Maritime and Ocean
Engineering Institute.
Maron, A. 2012. Flexible model tests. The International Workshop on Springing and Whipping of Ships, 810
November 2012, Split, Croatia.
Mckenna, M.F., Ross, D., Wiggins, S.M. & Hildebrand, J.A. 2012. Underwater radiated noise from modern
commercial ships. Journal of the Acoustical Society of America, 131, 92103.
Mehl, B., Oppitz, J. & Schreier, S. 2013. Sensitivity study on the influence of the exciting motion in liquid sloshing
in a rectangular tank. International Offshore and Polar Engineering Conference (ISOPE), 30 June - 5 July 2013,
Anchorage, USA.
Mehl, B., Pttmann, A. & Schreier, S. 2014. Sensitivity study on the influence of the filling height on the liquid
sloshing behavior in a rectangular tank. International Offshore and Polar Engineering Conference (ISOPE), 15
20 June 2014, Busan, Korea.
Merz, S., Kessissoglou, N., Kinns, R. & Marburg, S. 2013. Passive and active control of the radiated sound power
from a submarine excited by propeller forces. Journal of Ship Research, 57, 5971.
Metrikine, A.V. 2011. The synchronization phenomenon in modern structural dynamics: Vortex, ice, and
pedestrianinduced vibrations of engineering structures. The 8th International Conference on Structural
Dynamics, European Dynamics (EURODYN), 46 July 2011, Leuven, Belgium.
270
Miao, S.H., Temarel, P. & Hirdaris, S.E. 2012. The antisymmetric dynamic behaviour of a modern containership in
regular waves. The International Conference on Hydroelasticity in Marine Technology, 1921 September 2012,
Tokyo, Japan.
Minnick, L., Bassler, C. & Percival, S. 2012. Analysis of the physics of bilge keel vortex generation. The American
Miyazaki, M.R., Tannuri, E.A. & De Oliveira, A.C. 2013. Minimum energy DP heading control: Critical analysis
and comparison of different strategies. The American Society of Mechanical Engineers (ASME) Conference
Moro, L. & Biot, M. 2013. Laboratory tests pave the way for the knowledge of dynamic response of resilient
mountings on board ships. The International Symposium on Practical Design of Ships and Other Floating
Structures (PRADS), 2025 October 2013, Changwon, Korea.
Moro, L., Biot, M., Mantini, N. & Pestelli, C. 2013. Solutions to improve accuracy in experimental measurement of
the dynamic response of resilient mountings for marine diesel engines. In: Soares, C. & Romanoff, J. (eds.)
Analysis and Design of Marine Structures. London, UK: Taylor & Francis Group.
Mutsuda, H., Baso, S. & Doi, Y. 2012. Numerical simulation on hydroelastic response of structure under impact
load from water using Eulerian scheme with Lagrangian particles. Journal of Shipping and Ocean Engineering, 2,
363371.
Nakamura, T., Kaneko, S., Inada, F., Kato, M., Ishihara, K., Nishihara, T. & Mureithi, N.W. 2014. Flow-induced
vibrations: Classifications and lessons from practical experiences, Amsterdam, Netherland, Elsevier.
Nandan, H., Younan, A.H. & Deng, L. 2011. Ice induced vibration implementation of Maattanen model and
development of design supplements. The 21st International Conference on Port and Ocean Engineering under
Arctic Conditions, POAC11, 1014 July 2011, Montreal, Canada.
Neto, S.F.S., Figueiredo, S.R., Reyes, M.C.T. & De Mesquita Ortiz, L. 2012. Numerical simulation of the added
mass of the fluid adjacent to the ship hull in vibration measured during seatrials in tanker ships to be converted
to offshore construction vessel. The American Society of Mechanical Engineers (ASME) Conference
Neugebauer, J., El Moctar, O. & Potthoff, R. 2014. Experimental and numerical investigation of single impacts
in a 2D tank. International Offshore and Polar Engineering Conference (ISOPE), 1520 June 2014, Busan,
Korea.
Newman, J.N. 1994. Wave effects on deformable bodies. Applied Ocean Research, 16, 4759.
Ng, D.J.T., Teng, Y.J., Magee, A., Aramanadka, S.B., Zukni, N.A., Malik, A.M.A., Ismail, N.H. & Ghani, M.P.A.
2014. Tandem riser VIV suppression fairing model test. The American Society of Mechanical Engineers (ASME)
USA.
Nho, I.S., Kim, S.C., Jang, B.S. & Lee, J.H. 2012. Parametric investigation on the simplified triangular impulse of
sloshing pressure and categorization of the structural response on the mark III LNG CCS. International Offshore
NichollsLee, R., Micklethwait, W., Walker, R. & Argall, R. 2014. Novel, practical and effective: A feasibility
study for a low motion, floating wind turbine platform. The American Society of Mechanical Engineers (ASME)
USA.
Nielsen, U.D., Jensen, J.J., Pedersen, P.T. & Ito, Y. 2011. Onboard monitoring of fatigue damage rates in the hull
girder. Marine Structures, 24, 182206.
Nihei, Y., Iijima, K., Murai, M. & Ikoma, T. 2014. A comparative study of motion performance of four different
FOWT designs in combined wind and wave loads. The American Society of Mechanical Engineers (ASME)
USA.
Nord, T.S. & Maattanen, M. 2012. Model analysis in icestructure interaction: Deciphering ice induced vibration,
part 2. The 21st International Association for HydroEnvironment Engineering and Research (IAHR)
International Symposium on Ice, 1115 June 2012, Dalian, China.
Nord, T.S., Maattanen, M.P. & Oiseth, O. 2013. Frequency domain force identification in icestructure interaction.
The 22nd International Conference on Port and Ocean Engineering under Arctic Conditions, POAC13, 913
June 2013, Espoo, Finland.
North Atlantic Treaty Organization Research and Technical Organization 2012. Application of fibre optic
instrumentation. NATO Research and Technology Organisation AGARDograph (RTO AG) 160 Flight Test
Instrumentation Series 22.
Nwankwo, E., Fallah, A.S., Langdon, G.S. & Louca, L.A. 2013. Inelastic deformation and failure of partially
strengthened profiled blast walls. Engineering Structures, 46, 671686.
Oakley, O.H. & Constantinides, Y. 2013. Prediction and screening of truss spar VIM with CFD. The American
(OMAE), 914 June 2013, Nantes, France.
Oberhagemann, J. & El Moctar, O. 2012. Numerical and experimental investigations of whipping and springing of
ship structures. International Journal of Offshore and Polar Engineering, 22, 108114.
271
Oberhagemann, J., Krmer, M., Cabos, C. & El Moctar, O. 2012a. A fluidstructure coupling method based on field
methods and a structural mode decomposition. The International Conference on Hydroelasticity in Marine
Technology, 1921 September 2012, Tokyo, Japan.
Oberhagemann, J., Ley, J. & El Moctar, B.O. 2012b. Prediction of ship response statistics in severe sea conditions
using RANS. The American Society of Mechanical Engineers (ASME) Conference Proceedings Ocean,
Offshore and Arctic Engineering (OMAE), 16 July 2012, Rio de Janeiro, Brazil.
Oberhagemann, J., Ley, J., Shigunov, V. & El Moctar, O. 2012c. Efficient approaches for ship response statistics
using RANS. International Offshore and Polar Engineering Conference (ISOPE), 1722 June 2012, Rhodes,
Greece.
Oberhagemann, J., Shigunov, V. & El Moctar, O. 2013. First order reliability analogies of nonlinear bending
moments in ships. The American Society of Mechanical Engineers (ASME) Conference Proceedings Ocean,
Obisesan, A.E. 2012. Uncertainty quantification and risk assessment of offshore structures. MS. thesis, University
of Aberdeen.
Ochi, K. 1958. Model experiments on ship strength and slamming in regular waves. The Society of Naval Architects
and Marine Engineers (SNAME), 66, 345383.
Ogawa, Y., Kitamura, O. & Toyoda, M. 2012. A study for the statistical characteristic of slamming induced
vibration of large container ship. The International Conference on Hydroelasticity in Marine Technology, 1921
September 2012, Tokyo, Japan.
Ogawa, Y. & Takagi, K. 2009. An evaluation of whipping vibration utilizing the displacement potential method. The
International Conference on Hydroelasticity in Marine Technology, 1921 September 2009, Southampton, UK.
Ogawa, Y. & Takagi, K. 2012. A study for the effect of operation on hydroelasticity of hull. The International
Conference on Hydroelasticity in Marine Technology, 1921 September 2012, Tokyo, Japan.
Paik, B.G., Kim, G.D., Kim, K.Y., Seol, H.S., Hyun, B.S., Lee, S.G. & Jung, Y.R. 2013. Investigation on the
performance characteristics of the flexible propellers. Ocean Engineering, 73, 139148.
Paik, J.K., Lee, S.E., Kim, B.J., Seo, J.K., Ha, Y.C. & Matsumoto, T. 2014. A new method for determining the
design sloshing loads for LNG FPSOs. The American Society of Mechanical Engineers (ASME) Conference
Proceedings Ocean, Offshore and Arctic Engineering (OMAE), 813 June 2014, San Francisco, USA.
Palmer, A. & Bjerkas, M. 2013. Synchronization and the transition from intermittent to lockedin iceinduced
vibration. The 22nd International Conference on Port and Ocean Engineering under Arctic Conditions, POAC13,
913 June 2013, Espoo, Finland.
Pang, F.Z., Song, H.B. & Miao, X.H. 2014. Research on vibration and sound radiation characteristics of ship
stiffened composite plate structure. The International Congress on Noise Control Engineering (INTERNOISE),
1619 November 2014, Melbourne, Australia.
Papaioannou, I., Gao, R., Rank, E. & Wang, C.M. 2013. Stochastic hydroelastic analysis of pontoon-type
very large floating structures considering directional wave spectrum. Probabilistic Engineering Mechanics, 33,
2637.
Papanikolaou, A., Mohammed, E.A. & Hirdaris, S.E. 2014. Stochastic uncertainty modelling for ship design loads
and operational guidance. Ocean Engineering, 86, 4757.
Papanikolaou, A. & Schellin, T. 1992. A three-dimensional panel method for motions and loads of ships with
forward speed. Ship Technology Research, 39, 145.
Park, H., Bernitsas, M.M. & Chang, C.C. 2013a. Map of passive turbulence control to flowinduced motions for a
circular cylinder at 30,000<Re<120,000: Sensitivity to zone covering. The American Society of Mechanical
2013, Nantes, France.
Park, H., Bernitsas, M.M. & Kim, E.S. 2013b. Selective surface roughness to suppress flowinduced motion of two
circular cylinders at 30,000<Re<120,000. The American Society of Mechanical Engineers (ASME) Conference
Pasquier, R. & Berthon, C.F. 2012. Model scale test vs. full scale measurement: Findings from the full scale
measurement of sloshing project. International Offshore and Polar Engineering Conference (ISOPE), 1722
June 2012, Rhodes, Greece.
Patraquim, R., Godinho, L., Tadeu, A. & AmadoMendes, P. 2011. Influence of the presence of lining materials in
the acoustic behaviour of perforated panel systems. The International Congress on Sound and Vibration (ICSV),
1014 July 2011, Rio de Janeiro, Brazil.
Peng, S., Temarel, P., Bennett, S.S., Wu, W., Liu, Z. & Wang, Y. 2014. Symmetric response of a hydroelastic scaled
container ship model in regular and irregular waves. The American Society of Mechanical Engineers (ASME)
USA.
Pereira, I. & Cordioli, J.A. 2011. Evaluation of the sound radiated from submerged flat structures using hybrid FE
SEABEM method. The International Congress on Sound and Vibration (ICSV), 1014 July 2011, Rio de
Janeiro, Brazil.
Peters, H., Kinns, R. & Kessissoglou, N. 2014. Effects of internal mass distribution and its isolation on the acoustic
characteristics of a submerged hull. Journal of Sound and Vibration, 333, 16841697.
Phelps, B. & Morris, B. 2013. Review of hull structural monitoring systems for navy ships [Online]. DTIC
Document. Available: http://dspace.dsto.defence.gov.au/dspace/handle/dsto/10246 [Accessed 17 Jan 2015].
272
Piro, D.J., Fu, T.C. & Maki, K.J. 2012. Joint high speed sealift (JHSS) segmented model test results. The American
Piro, D.J. & Maki, K.J. 2011. Hydroelastic wedge entry and exit. The International Conference on Fast Sea
Transportation (FAST), 2629 September 2011, Honolulu, USA.
Piro, D.J. & Maki, K.J. 2013. Hydroelastic analysis of bodies that enter and exit water. Journal of Fluids and
Popko, W. 2014. Comparison of fullscale and numerical model dynamic responses of Norstromsgrund lighthouse.
Qiu, W., Junior, J.S., Lee, D., Lie, H., Magarovskii, V., Mikami, T., Rousset, J.M., Sphaier, S., Tao, L. & Wang, X.
2014. Uncertainties related to predictions of loads and responses for ocean and offshore structures. Ocean
Qiu, Y., Wang, Q.B., Zhao, H.T., Chen, J.A. & Wang, Y.Y. 2013. Review on composite structural health
monitoring based on fiber Bragg grating sensing principle. Journal of Shanghai Jiaotong University (Science),
18, 129139.
Quinton, B.W.T., Daley, C.G. & Gagnon, R.E. 2012. Realistic moving ice loads and ship structural response.
Rafiee, A., Repalle, N. & Dias, F. 2013. Numerical simulations of 2D liquid impact benchmark problem using twophase compressible and incompressible methods. International Offshore and Polar Engineering Conference
(ISOPE), 30 June - 5 July 2013, Anchorage, USA.
Rahaman, M.M. & Akimoto, H. 2012. Analysis of the Mechanism of Slamming on the Bow Flare Region of a
Container Ship Using RaNS CFD Method. International Offshore and Polar Engineering Conference (ISOPE),
Rahaman, M.M., Akimoto, H. & Tanaka, K. 2012. Numerical simulation of hydrodynamic impact on ship sections
with bulb and flare in 2D and 3D conditions. The International Conference on Hydrodynamics (ICHD), 15
October 2012, St. Petersburg, Russia.
Razzak, S., Amaichan, J., Damion, J.P. & Sarraf, C. 2013. Dynamic pressure calibration. International Offshore and
Polar Engineering Conference (ISOPE), 30 June - 5 July 2013, Anchorage, USA.
Renaud, M., De Lorgeril, E., Boutillier, J.B. & Gerard, L. 2013. Fatigue and weather on ultra large containerships.
Resvanis, T.L., Rao, Z. & Vandiver, J.K. 2014. Effects of Strake Coverage and Marine Growth on Flexible Cylinder
VIV. The American Society of Mechanical Engineers (ASME) Conference Proceedings Ocean, Offshore and
Arctic Engineering (OMAE), 813 June 2014, San Francisco, USA.
Rezanejad, K. & Guedes Soares, C. 2013. Numerical study of a large floating oscillating water column device using
a 2D boundary element method. In: Guedes Soares, C.L.P., F (ed.) Developments in maritime transportation and
exploitation of sea resources. London, UK: Taylor & Francis Group, 951960.
Rijken, O. 2014. Examining the Effects of Scale, Mass Ratios and Column Shapes on the Vortex Induced Motion
Response of a Semisubmersible Through CFD Analyses. The American Society of Mechanical Engineers (ASME)
USA.
Riley, M., Smith, M., Van Aanhold, J.E. & Alin, N. 2012. Loading on a rigid target from close proximity
underwater explosions. Shock and Vibration, 19, 555571.
Robertson, A., Jonkman, J., Vorpahl, F., Popko, W. & Al, E. 2014. Offshore code comparison collaboration
continuation within IEA wind task 30: Phase II results regarding a floating semisubmersible wind system. The
Robertson, A.N., Jonkman, J.M., Goupee, A.J., Coulling, A.J., Prowell, I., Browning, J., Masciola, M.D. & Molta,
P. 2013. Summary of conclusions and recommendations drawn from the DeepCwind scaled floating offshore
wind system test campaign. The American Society of Mechanical Engineers (ASME) Conference Proceedings
Robinson, S.P., Theobald, P.D. & Lepper, P.A. 2012. Underwater noise generated from marine piling. The Meetings
on Acoustics 11th European Conference on Underwater Acoustics (ECUA), 2 - 6 July 2012, Edinburgh,
Scotland.
Rogers, L. & Stambaugh, K. 2014. Application of acoustic emission technology for health monitoring of ship
structures. The Ship Structure Committee Symposium: Vessel Safety & Longevity through Ship Structure
Rogers, L.M. & Carlton, J. 2010. The acoustic emission technique: Application to marine structures and machinery.
Lloyds Register Technology Day, Februrary 2010, London, UK. London, UK: Lloyds Register.
Roth, E.H., Schmidt, V., Hildebrand, J.A. & Wiggins, S.M. 2013. Underwater radiated noise levels of a research
icebreaker in the central Arctic Ocean. The Journal of Acoustic Society of America, 133, 19711980.
Rubanenco, I., Mirciu, I. & Domnisoru, L. 2013. On the nonlinear hydroelastic response in irregular head waves of
a structural optimized container ship. The International Conference on Marine Structures (MARSTRUCT), 2527
March 2013, Espoo, Finland.
273
Saelid, S., Jenssen, N.A. & Balchen, J.G. 1983. Design and analysis of a dynamic positioning system based on
Kalman filtering and optimal control. Institute of Electrical and Electronics Engineers (IEEE) Transactions on
Automatic Control, 28, 331339.
Saito, M., Masanobu, S., Taniguchi, T., Otsubo, K., Asanuma, T. & Maeda, K. 2012. Experimental evaluation of
VIM on MPSO in combined environmental conditions for waves and current. The American Society of
Salio, M.P. 2013. Numerical assessment of underwater noise radiated by a cruise ship. Ships and Offshore
Structures, DOI: 10.1080/17445302.2013.790271.
Sawamura, J. 2013. Numerical study on the ship distribution of ice pressure area and structure response for ship
maneuvering in level ice. International Offshore and Polar Engineering Conference (ISOPE), 30 June - 5 July
2013, Anchorage, USA.
Scheer, M. & Ritter, F. 2013. Underwater bowradiated noise characteristics of three types of ferries: Implications
for vesselwhale collisions in the Canary Island, Spain. The 27th Conference of the European Cetacean Society,
810 April 2013, Setubal, Portugal.
Schellin, T.E., Shigunov, V., Troesch, H.a.W., Kim, D.H. & Maki, K. 2013. Prediction of loads for ship structural
design. The Ship Structures Technical Track, American Society of Naval Engineers, 2122 February 2013,
Arlington, USA.
Schiffer, A. & Tagarielli, V.L. 2014a. The dynamic response of composite plates to underwater blast: Theoretical
and numerical modelling. International Journal of Impact Engineering, 70, 113.
Schiffer, A. & Tagarielli, V.L. 2014b. The one-dimensional response of a water-filled double hull to underwater
blast: Experiments and simulations. International Journal of Impact Engineering, 63, 177187.
Scolan, Y.M., Karimi, M.R., Dias, F., Ghidaglia, J.M. & Costes, J. 2014. Highly nonlinear wave in tank with small
density ratio. The International Workshop on Water Waves and Floating Bodies (IWWWFB), 30 March - 2 April
2014, Osaka, Japan.
Seigler, T.M. & Hoagg, J.B. 2013. Filtered dynamic inversion for vibration control of structures with uncertainty.
Journal of Dynamic Systems, Measurement and Control, 135, 041017.
Seiler, R. & Holbach, G. 2013a. Acoustic quality on board ships. The International Congress on Acoustics (ICA), 2
7 June 2013, Montreal, Canada.
Seiler, R. & Holbach, G. 2013b. Criteria for the determination of the acoustic quality of ships. The Italian Annual
Conference on Acoustics and German Annual Conference on Acoustics (AIADAGA), 1821 March 2013,
Merano, Italy.
Seng, S., Jensen, J.J. & Pedersen, P.T. 2012. Numerical prediction of slamming loads. Journal of Engineering for
the Maritime Environment, 226, 120134.
Senjanovi, I., Hadi, N. & Vladimir, N. 2012a. Improved methodology of ship hydroelastic analysis. The
International Conference on Hydroelasticity in Marine Technology, 1921 September 2012, Tokyo, Japan.
Senjanovi, I., Vladimir, N. & Cho, D.S. 2013. Simplified formulations of mass and geometric stiffness matrices in
vibration and stability analyses of thinwalled structures an update. In: Soares, C. & Romanoff, J. (eds.)
Analysis and Design of Marine Structures. London, UK: Taylor & Francis Group.
Senjanovi, I., Vladimir, N. & Malenica, . 2012b. Hydroelastic loading and response of ultra large container ships.
Senjanovi, I., Vladimir, N., Tomi, M., Hadi, N. & Malenica, . 2014a. Global hydroelastic analysis of ultra
large container ships by improved beam structural model. International Journal of Naval Architecture and Ocean
Senjanovi, I., Vladimir, N., Tomi, M., Hadi, N. & Malenica, . 2014b. Some aspects of structural modelling and
restoring stiffness in hydroelastic analysis of large container ships. Ships and Offshore Structures, 9, 199217.
Seok, H.I., Choi, S.H., Lee, D.K., Jeong, T.S. & Heo, J.H. 2011. A study on high pressure steam system noise
problem in a LNG carrier. The International Congress on Sound and Vibration (ICSV), 1014 July 2011, Rio de
Janeiro, Brazil.
Shadman, M. & Akbarpour, A. 2012a. Comparative study of utilizing a new type Vshaped tuned liquid column
damper and Ushaped tuned liquid column damper in floating wind turbines. The American Society of
Shadman, M. & Akbarpour, A. 2012b. Utilizing TLCD (tuned liquid column damper) in floating wind turbines. The
Engineering (OMAE), 16 July 2012, Rio de Janeiro, Brazil.
Shkhinek, K.N., Zhilenkov, A., Thomas, G. & Gebson, R. 2013. The vibration of fixed offshore structures under ice
action. The 22nd International Conference on Port and Ocean Engineering under Arctic Conditions, POAC13, 9
13 June 2013, Espoo, Finland.
Sieve, M.W., Kihl, D.P. & Ayyub, B.M. 2000. Fatigue design guidance for surface ships, Washington, USA,
Storming Media.
Sikora, J.P., Dinsenbacher, A. & Beach, J.E. 1983. A method for estimating lifetime loads and fatigue lives for
swath and conventional monohull ships. Naval Engineers Journal, 95, 6385.
SILENV. 2012. Deliverable 5.2 green label proposal. SILENV - Ships oriented innovative solutions to reduce noise
and vibrations, FP7 - Collaborative project No. 234181 [Online]. Available: http://www.silenv.eu [Accessed 17
Jan 2015].
274
Sohn, J.M., Kim, S.J., Kim, B.H. & Paik, J.K. 2013. Nonlinear structural consequence analysis of FPSO topside
blastwalls. Ocean Engineering, 60, 149162.
Song, Y., Wen, J., Yu, D., Liu, Y. & Wen, X. 2014. Reduction of vibration and noise radiation of an underwater
vehicle due to propeller forces using periodically layered isolators. Journal of Sound and Vibration, 333, 3031
3043.
Song, Y.K., Chang, K.A., Ryu, Y.G. & Kwon, S.H. 2013. Flow velocity and impact pressure in liquid sloshing. The
Southall, N.R., Lee, Y., Johnson, M.C., Hirdaris, S.E. & White, N.J. 2014. Towards a pragmatic method for
prediction of whipping: Wedge impact simulations using OpenFOAM. International Offshore and Polar
Engineering Conference (ISOPE), 1520 June 2014, Busan, Korea.
Souto-Iglesias, A., Botia Vera, E. & Bulian, G. 2011. Repeatability and two-dimensionality of model scale
sloshing impacts. International Offshore and Polar Engineering Conference (ISOPE), 1722 June 2011, Rhodes,
Greece.
Sovardi, C., Jaensch, S., Silva, C. & Polifke, W. 2014. Identification of sound sources in internal ducted flows: A
large eddy simulation-system identification approach. The International Congress on Sound and Vibration
(ICSV), 1317 July 2014, Beijing, China.
Stambaugh, K., Drummen, I., Cleary, C., Sheinberg, R. & Kaminski, M.L. 2014. Structural fatigue life
assessment and sustainment implications for a new class of US coast guard cutters. The Ship Structure
Committee Symposium: Vessel Safety & Longevity through Ship Structure Research, 1820 May 2014, Linthicum,
USA.
Storhaug, G. 2007. Experimental investigation of wave induced vibrations and their effect on the fatigue loading of
ships. Ph.D. thesis, Norwegian University of Science and Technology.
Storhaug, G. 2012. The effect of heading on springing and whipping induced fatigue damage measured on container
vessels. The International Conference on Hydroelasticity in Marine Technology, 1921 September 2012, Tokyo,
Japan.
Storhaug, G. 2014. Which sea states are dimensioning for container vessels when whipping is included? The
Storhaug, G., Choi, B.K., Moan, T. & Hermundstad, O.A. 2010a. Consequence of whipping and springing on
fatigue for a 8600TEU container vessel in different trades based on model tests. The International Symposium on
Practical Design of Ships and Other Floating Structures (PRADS), 1924 September 2010, Rio de Janeiro,
Brazil.
Storhaug, G. & Hareide, O.J. 2013. Assessment of hull monitoring measurements for a large blunt vessel. The
Storhaug, G., Malenica, S., Choi, B., Zhu, S. & Hermundstad, O. 2010b. Consequence of whipping and springing on
fatigue and extreme loading for a 13,000TEU container vessel based on model tests. The International
Symposium on Practical Design of Ships and Other Floating Structures (PRADS), 1924 September 2010, Rio de
Janeiro, Brazil.
Storhaug, G. & Moe, E. 2007. Measurements of wave induced vibrations onboard a large container vessel operating
in harsh environment. The International Symposium on Practical Design of Ships and Other Floating Structures
(PRADS), 15 October 2007, Houston, USA.
Storhaug, G., Oma, N., Blomberg, B. & Hirota, K. 2013. The effect of measured whipping and springing on LNG
vessels. The American Society of Mechanical Engineers (ASME) Conference Proceedings Ocean, Offshore and
Arctic Engineering (OMAE), 914 June 2013, Nantes, France.
Storhaug, G., Pettersen, T.A., Oma, N. & Blomberg, B. 2012. The effect of wave induced vibrations on fatigue
loading and the safety margin against collapse on two LNG vessels. The International Conference on
Stritzelberger, B., Abele, M. & Von Estorff, O. 2013. On the energy finite element method for the acoustic design of
ships. The International Congress on Acoustics (ICA), 27 June 2013, Montreal, Canada.
Swartz, R.A., Zimmerman, A.T., Lynch, J.P., Rosario, J., Brady, T., Salvino, L. & Law, K.H. 2012. Hybrid wireless
hull monitoring system for naval combat vessels. Structure and Infrastructure Engineering: Maintenance,
Management, LifeCycle Design and Performance, 8, 621638.
Takagi, K. & Ogawa, Y. 2007. Flow models of the flare slamming. The International Conference on Violent Flows
(VF), 2022 November 2007, Fukuoka, Japan.
Takaoka, Y., Murakami, A., Yoshida, T., Miyake, R., Yamamoto, N., Terai, K. & Toyoda, K. 2012. An
experimental study on the whipping response of large container carriers. The International Conference on
Takinac, A.C. & Taralp, T. 2013. Prediction and simulation of broadband propeller noise. Journal of Marine
Science and Technology, 21, 538544.
Tan, J.H.C., Magee, A., Kim, J.W., Teng, Y.J. & Zukni, N.A. 2013. CFD simulation for vortex induced motions of a
multi-column floating platform. The American Society of Mechanical Engineers (ASME) Conference Proceedings
275
Tan, J.H.C., Teng, Y.J., Magee, A., Ly, B.T.H. & Aramanadka, S.B. 2014. Vortex induced motion of TLP with
consideration of appurtenances. The American Society of Mechanical Engineers (ASME) Conference Proceedings
Tassin, A., Jacques, N., Alaoui, A.E.M., Nme, A. & Lebl, B. 2012. Hydrodynamic loads during water impact of
three-dimensional solids: Modelling and experiments. Journal of Fluids and Structures, 28, 211231.
Teng, X.Y., Han, J.S. & Peng, L. 2014. The application of BESO in vibration damping of ship plate. Key
Engineering Materials, 572, 185188.
Ternowchek, S. 2012. An early warning monitoring system for assessment of crack and corrosion activity in FPSO
vessels. The FPSO Research Forum, 711 May 2012, Oosterbeek, Netherlands.
The Federal Office of Bundeswehr Equipment, I.T.a.I.S.S.B. & Organisation., D.M. 2014. Shock resistance. BV
0230 / D50500599. Coblenz, Germany: BAAINBw & Hague, Netherlands: DMO.
Thomas, G., Matsubara, S., Davis, M., French, B., Lavroff, J. & Amin, W. 2012. Lessons learnt through the design,
construction and testing of a hydroelastic model for determining motions, loads and slamming behaviour in
severe sea states. The International Conference on Hydroelasticity in Marine Technology, 1921 September
2012, Tokyo, Japan.
Thomas, G., Winkler, S., Davis, M., Holloway, D., Matsubara, S., Lavroff, J. & French, B. 2011. Slam events of
high-speed catamarans in irregular waves. Journal of Marine Science and Technology, 16, 821.
Tian, Y. & Huang, Y. 2012. Model test of ice induced vibrations on fourleg jacket platforms. The 21st International
Association for HydroEnvironment Engineering and Research (IAHR) International Symposium on Ice, 1115
June 2012, Dalian, China.
Tiphine, E., Bigot, F., De-Lauzon, J., Sireta, F.X., Chung, Y.S. & Malenica, S. 2014. Comparisons of experimental
and numerical results for global hydroelastic response of container ship within the WILS III JIP. International
Offshore and Polar Engineering Conference (ISOPE), 1520 June 2014, Busan, Korea.
Tom, N., Seah, R. & Roddier, D. 2012. Analysis of roll motion due to asymmetric bilge keels. The American Society
of Mechanical Engineers (ASME) Conference Proceedings Ocean, Offshore and Arctic Engineering (OMAE),
16 July 2012, Rio de Janeiro, Brazil.
Toyoda, K., Matsumoto, T., Yamamoto, N. & Terai, K. 2012. Simplified fatigue assessment considering the
occurrence probability of hydroelastic response in actual sea state conditions. The International Conference on
Traverso, F. & Trucco, A. 2014. Analysis of underwater acoustic noise measured at the ship bow during
sea trial. The International Conference and Exhibition on Underwater Acoustics, 2227 July 2014, Rhodes,
Greece.
Tsouvalas, A. & Metrikine, A.V. 2013. A semi-analytical model for the prediction of underwater noise from
offshore pile driving. Journal of Sound and Vibration, 332, 32323257.
Tsouvalas, A. & Metrikine, A.V. 2014. A three-dimensional vibroacoustic model for the prediction of underwater
noise from offshore pile driving. Journal of Sound and Vibration, 333, 22832311.
Tuitman, J.T. 2008. TDWHIP theory manual, Burlington, USA, CRS Working Group.
Tuitman, J.T., Bosman, T.N. & Harmsen, E. 2013. Local structural response to seakeeping and slamming loads.
Marine Structures, 33, 214237.
Turan, O., Helvacioglu, I.H., Insel, M., Khalid, H. & Kurt, R.E. 2011. Crew noise exposure on board ships and
comparative study of applicable standards. Ships and Offshore Structures, 6, 323338.
Turbo, M.D. 2014. G60MEC9 vibration performance successful vibration measurements completed on sea
trial [Online]. Available: http://marine.man.eu/docs/librariesprovider6/technical-papers/successful-vibrationperformance-g60me-c9.pdf?sfvrsn=18 [Accessed 17 Jan 2015].
Utsunomiya, T., Sato, I., Yoshida, S., Ookubo, H. & Ishida, S. 2013. Dynamic response analysis of a floating
offshore wind turbine during severe typhoon event. The American Society of Mechanical Engineers
France.
Utsunomiya, T., Yoshida, S., Kiyoki, S., Sato, I. & Ishida, S. 2014. Dynamic response of a spar-type floating wind
turbine at power generation. The American Society of Mechanical Engineers (ASME) Conference Proceedings
Van Esch, B.P.M., Van Hooijdonk, J.J.A. & Bulten, N.W.H. 2013. Quantification of hydrodynamic forces due to
torsional and axial vibrations in ship propellers. The American Society of Mechanical Engineers (ASME)
Proceedings of Symposium on Fluid Machinery, 711 July 2013, Incline Village, USA.
Veen, D. & Gourlay, T. 2012. A combined strip theory and smoothed particle hydrodynamics approach for
estimating slamming loads on a ship in head seas. Ocean Engineering, 43, 6471.
Walters, A.P., Didoszak, J.M. & Kwon, Y.W. 2013. Explicit modeling of solid ocean floor in shallow underwater
explosions. Shock and Vibration, 20, 189197.
Wang, B., Shin, Y. & Wang, X. 2012. Reliability-based sloshing assessment of containment systems in LNGCs
and FLNGs. International Offshore and Polar Engineering Conference (ISOPE), 1722 June 2012, Rhodes,
Greece.
Wang, C.M., Gao, R.P., Koh, C.G. & Kitipornchai, S. 2012. Novel hybrid system for reducing hydroelastic response
of very large floating structures. The American Society of Mechanical Engineers (ASME) Conference
Wang, C.M. & Tay, Z.Y. 2011. Very large floating structures: Applications, research and development. Procedia
276
Wang, C.M., Watanabe, E. & Utsunomiya, T. 2008. Very large floating structures, London, UK, Taylor & Francis
Group.
Wang, G., Lee, M., Serratella, C., Botten, S., Ternowchek, S., Ozevin, D., Thibault, J. & Scott, R. 2010. Testing of
acoustic emission technology to detect cracks and corrosion in the marine environment. Journal of Ship
Production, 26, 106110.
Wang, H., Zhu, X., Cheng, Y.S. & Liu, J. 2014. Experimental and numerical investigation of ship
structure subjected to close-in underwater shock wave and following gas bubble pulse. Marine Structures, 39,
90117.
Wang, S. 2013. Damage detection in offshore platform structures from limited modal data. Applied Ocean Research,
41, 4856.
Wang, S. & Soares, C.G. 2013. Numerical study on the water impact of 3D bodies by an explicit finite element
method. Ocean Engineering, 78, 7388.
Wang, S.Q. & Li, N. 2013. Semi-active vibration control for offshore platforms based on LQG method. Journal of
Marine Science and Technology, 21, 562568.
Wang, S.Y. & Yue, Q.J. 2012. Vibration reduction of bucket foundation platform with fixed ice-breaking cone in
the Bohai Sea. Journal of Cold Regions Engineering, 26, 160168.
Wang, X., Hu, J., Gu, X., Ding, J. & Zhao, N. 2014. Wave loads investigation of a VLCC by experimental and
theoretical methods. The American Society of Mechanical Engineers (ASME) Conference Proceedings Ocean,
Offshore and Arctic Engineering (OMAE), 813 June 2014, San Francisco, USA.
Wang, Y., Liu, X., Wang, G. & Ren, B. 2014. Investigation of VIV suppression effect of the finned cylinder in
subcritical flow. The American Society of Mechanical Engineers (ASME) Conference Proceedings Ocean,
Offshore and Arctic Engineering (OMAE), 813 June 2014, San Francisco, USA.
Wang, Y. & Yue, Q. 2013. Physical mechanism of ice induced selfexcited vibration. The 22nd International
Conference on Port and Ocean Engineering under Arctic Conditions, POAC13, 913 June 2013, Espoo,
Finland.
Wang, Z.G. & Shen, J.P. 2014. Research on integrated design technique of active vibration isolating plant of marine
high-speed diesel for applying aboard. The International Congress on Sound and Vibration (ICSV), 1317 July
2014, Beijing, China.
Wei, Y. & Wang, Y. 2013. Unsteady hydrodynamics of blade forces and acoustic responses of a model
scaled submarine excited by propellers thrust and side-forces. Journal of Sound and Vibration, 332, 2038
2056.
Wei, Z., Ruan, S., Chen, X., Luo, S. & Yue, Q. 2014. An experimental investigation of liquid sloshing impact load
in a large-scaled prismatic tank. International Offshore and Polar Engineering Conference (ISOPE), 1520 June
2014, Busan, Korea.
Wei, Z., Yue, Q. & Ruan, S. 2012. An experimental investigation of liquid sloshing impact load in a rectangular
tank. International Offshore and Polar Engineering Conference (ISOPE), 1722 June 2012, Rhodes, Greece.
Wochner, M.S., Lee, K.M., Mcneese, A.R. & Wilson, P.S. 2014. Attenuation of low frequency underwater noise
using arrays of air-filled resonators. The International Congress on Noise Control Engineering (INTERNOISE),
1619 November 2014, Melbourne, Australia.
Wu, M. & Moan, T. 2005. Efficient calculation of wave-induced ship responses considering structural dynamic
effects. Applied Ocean Research, 27, 8196.
Wu, M.K., Lehn, E. & Moan, T. 2012. Design of segmented model for ship seakeeping tests with hydroelastic
effects. The International Conference on Hydroelasticity in Marine Technology, 1921 September 2012, Tokyo,
Japan.
Wu, M.K. & Stambaugh, K. 2013. Experimental and numerical study of hydroelastic responses in a highspeed
vessel. The International Conference on Fast Sea Transportation (FAST), 25 December 2013, Amsterdam,
Netherlands.
Wu, X., Ge, F. & Hong, Y. 2010. Effect of travelling wave on vortex-induced vibrations of submerged floating
tunnel tethers. Procedia Engineering, 4, 153160.
Xia, Z. & Tang, J. 2013. Characterization of dynamic response of structures with uncertainty by using gaussian
processes. Journal of Vibration and Acoustics, 135, 113.
Xiao, F., Chen, Y. & Hua, H. 2014. Comparative study of the shock resistance of rubber protective coatings
subjected to underwater explosion. Journal of Offshore Mechanics and Arctic Engineering 136, 112.
Xiong, J.J., Zheng, S.J., Hong, Y.P., Li, J., Wang, Y.L., Wang, W. & Tan, Q.L. 2013. Measurement of wireless
pressure sensors fabricated in high temperature co-fired ceramic MEMS technology. Journal of Zhejiang
University SCIENCE C, 14, 258263.
Xiong, Y.P. 2014. Novel approach for structural dynamic topology optimisation based on power flow mode theory
an update, Berlin, Germany, Springer International Publishing.
Xu, W., Iijima, K. & Fujikubo, M. 2012. Parametric dependencies of the post-ultimate strength behavior of a ships
hull girder in waves. Journal of Marine Science and Technology, 17, 203215.
Xu, W., Lv, Z. & Shuai, C. 2014. Theoretical and experimental analysis of high-frequency vibration isolation
for electric motor. The International Congress on Sound and Vibration (ICSV), 1317 July 2014, Beijing,
China.
Xue, M.A., Zheng, J., Yuan, X., Lin, P., Ma, Y. & Zhang, W. 2014. Numerical simulation of shallow water sloshing
characteristics in a rectangular tank. International Offshore and Polar Engineering Conference (ISOPE), 1520
June 2014, Busan, Korea.
277
Yan, Q. 2013. A new method for the analysis of ice intermittent crushing induced lock-in vibration. International
Yang, J., Kim, S., Park, J.S., Jung, B.H. & Lee, T. 2013. Numerical analysis for slamming impact loads and
dynamic structural responses of a containership. The International Symposium on Practical Design of Ships and
Other Floating Structures (PRADS), 2025 October 2013, Changwon, Korea.
Yang, J., Xiong, Y.P. & Xing, J.T. 2012. Dynamics and power flow behaviour of a nonlinear vibration isolation
system with a negative stiffness mechanism. Journal of Sound and Vibration, 332, 167183.
Yap, K.T. 2011. Level ice-vertical structure interaction: Steady-state self-excited vibration of structures. Ph.D.
thesis, National University of Singapore.
Yap, K.T. & Palmer, A.C. 2013. A model test on iceinduced vibrations: Structure response characteristics and
scaling of the lockin phenomenon. The 22nd International Conference on Port and Ocean Engineering under
Arctic Conditions, POAC13, 913 June 2013, Espoo, Finland.
Yoon, J.S., Cho, S.P., Jiwinangun, R.G. & Lee, P.S. 2014. Hydroelastic analysis of floating plates with multiple
hinge connections in regular waves. Marine Structures, 36, 6587.
Yoshikawa, T. & Maeda, M. 2013. Numerical simulation of structural response under bow flare slamming load. The
International Conference on Marine Structures (MARSTRUCT), 2527 March 2013, Espoo, Finland.
Young, A.C., Hettick, S., Dagher, H.J., Viselli, A.M. & Goupee, A.J. 2014. VolturnUS 1: 8scale FRP floating wind
turbine tower: Analysis, design, testing and performance. The American Society of Mechanical Engineers
(ASME) Conference Proceedings Ocean, Offshore and Arctic Engineering (OMAE), 813 June 2014, San
Francisco, USA.
Yucel, A. & Arpaci, A. 2013. Free and forced vibration analyses of ship structures using the finite element method.
Journal of Marine Science and Technology, 18, 324338.
Yue, Q. & Guo, F. 2012. Physical mechanism of ice-induced self-excited vibration. Journal of Engineering
Mechanics, 138, 784790.
Zdravkovich, M.M. 1981. Review and classification of various aerodynamic and hydrodynamic means for
suppressing vortex shedding. Journal of Wind Engineering and Industrial Aerodynamics, 7, 145189.
Zdravkovich, M.M. 1997. Flow around circular cylinders vol. 1: Fundamentals, Oxford, UK, Oxford Press.
Zdravkovich, M.M. 2003. Flow around circular cylinders: volume 2: Applications, Oxford, UK, Oxford Press.
Zhang, B.L. & Tang, G.Y. 2013. Active vibration H control of offshore steel jacket platforms using delayed
feedback. Journal of Sound and Vibration, 332, 56625677.
Zhang, D., Yue, Q., Li, H. & Huang, Y. 2012. The effects on facilities and crew members of the ice-resistant
platforms induced by ice vibrations in Bohai Sea. The American Society of Mechanical Engineers (ASME)
Brazil.
Zhang, G., Zhao, Y., Li, T. & Zhu, X. 2014. Propeller excitation of longitudinal vibration characteristics of marine
propulsion shafting system. Shock and Vibration, 2014.
Zhang, H., Yang, J., Peng, T. & Lu, H. 2012. Experimental investigation on added mass coefficient of a truss spar
subjected to vortex-induced motions. The American Society of Mechanical Engineers (ASME) Conference
Zhang, L., Chen, W. & Zheng, Z. 2010. Controlling parameter for wave types of long flexible cable undergoing
vortex-induced vibration. Procedia Engineering, 4, 161170.
Zhang, N. & Zong, Z. 2012. Hydro-elastic-plastic dynamic response of a ship hull girder subjected to an underwater
bubble. Marine Structures, 29, 177197.
Zhang, N., Zong, Z. & Zhang, W. 2014. Dynamic response of surface ship structure subjected to an underwater
explosion bubble. Marine Structures 35, 2644.
Zhang, Y. & Wan, D. 2012. Apply MPS method to simulate liquid sloshing in LNG tank. International Offshore
Zhang, Y., Yang, Y., Tang, Z. & Wan, D. 2014. Parallel MPS method for three-dimensional liquid sloshing.
Zhao, R., Faltinsen, O. & Aarsnes, J.V. 1996. Water entry of arbitrary two-dimensional sections with and without
flow separation. The Symposium on Naval Hydrodynamics, 2428 June 1996, Trondheim, Norway.
Zheng, M., Kim, Y., Kim, S.Y. & Ahn, Y. 2013. Study on the prediction of sloshing severity. International Offshore
and Polar Engineering Conference (ISOPE), 30 June - 5 July 2013, Anchorage, USA.
Zhou, L.B., Yang, T.J., Zhang, P., Pan, J.J. & Liu, Z.G. 2013. Tests for an active floating raft vibration isolation
system based on a flexible hull structure. Journal of Vibration and Shock, 32, 156161.
Zhu, G., Xiong, Y.P., Daley, S. & Shenoi, A. 2012. Dynamical analysis of a twostage nonlinear MRE vibration
isolation system. The International Congress on Sound and Vibration (ICSV), 812 July 2012, Vilnius, Lituania.
Zhu, S. & Moan, T. 2012. A comparative study of the influence of bow shape on hull girder vibrations through two
backbone models. The International Conference on Hydroelasticity in Marine Technology, 1921 September
2012, Tokyo, Japan.
Zhu, S. & Moan, T. 2013a. New insight into the wave-induced nonlinear vertical load effects of ultra-large container
ships based on experiments. Journal of Marine Science and Technology, 18, 87114.
Zhu, S. & Moan, T. 2013b. Nonlinear effects from wave-induced maximum vertical bending moment on a flexible
ultra-large containership model in severe head and oblique seas. Marine Structures, 35, 125.
Zhu, S., Wu, M.K. & Moan, T. 2011. Experimental investigation of hull girder vibrations of a flexible backbone
model in bending and torsion. Applied Ocean Research, 33, 252274.
278
Zhu, X., Xie, B., Yu, X. & Du, Q. 2014. Study on the sloshing load of FLNG. The American Society of Mechanical
2014, San Francisco, USA.
Ziemer, G. & Evers, K.U. 2014. Ice model tests with a compliant cylindrical structure to investigate ice-induced
vibrations. The American Society of Mechanical Engineers (ASME) Conference Proceedings Ocean, Offshore
Zong, Z., Zhao, Y. & Li, H. 2013. A numerical study of whole ship structural damage resulting from close-in under
water explosion shock. Marine Structures, 31, 2443.
Zou, J., Poll, P., Roddier, D., Tom, N. & Peiffer, A. 2013. VIM testing of a paired column semi submersible. The

710 SEPTEMBER 2015
CASCAIS, PORTUGAL
VOLUME 1
COMMITTEE III.1
ULTIMATE STRENGTH
COMMITTEE MANDATE
Concern for the ductile behaviour of ships and offshore structures and their structural components under
ultimate conditions. Attention shall be given to the influence of fabrication imperfections and in-service
damage and degradation on reserve strength. Uncertainties in strength models for design shall be
highlighted. Consideration shall be given to the practical application of methods.
COMMITTEE MEMBERS
Chairman:
T. Yoshikawa, Japan (Chair)

A. Bayatfar, Belgium
B. J. Kim, Korea
C. P. Chen, Taiwan
D. Wang, China
J. Boulares, USA
J. M. Gordo, Portugal
L. Josefson, Sweden
M. Smith, Canada
P. Kaeding, Germany
P. Jensen, Norway
R. Ojeda, Australia
S. Benson, UK
S. Vhanmane, India
S. Zhang, UK
X. Jiang, Netherlands
X. Qian, Singapore
KEYWORDS
Ultimate strength, buckling strength, yielding strength, nonlinear analysis, steel structures, aluminium
structures, composite structures, ship structures, offshore structures, initial imperfections, in-service
degradation, uncertainties, reliability, static/quasi-static loads.
280
ISSC committee III.1: ULTIMATE STRENGTH
CONTENTS
1.
INTRODUCTION ..................................................................................................................... 282
2.
FUNDAMENTALS .................................................................................................................. 283

2.1 Design for Ultimate Strength.......................................................................................... 283
2.2 General Characteristics of Ultimate Strength ................................................................ 283
3.
ASSESSMENT PROCEDURE FOR ULTIMATE STRENGTH ........................................... 284

3.1 Empirical and analytical methods .................................................................................. 284
3.1.1 Introduction........................................................................................................ 284
3.1.2 Hull structures .................................................................................................... 285
3.1.3 Residual Strength of Damage Hull Structures .................................................. 286
3.1.4 Plates and stiffened plates ................................................................................. 288
3.2 Numerical methods ......................................................................................................... 288
3.2.1 Introduction........................................................................................................ 288
3.2.2 Nonlinear FE method ........................................................................................ 289
3.2.3 Idealized structural unit method ........................................................................ 290
3.2.4 Conclusion ......................................................................................................... 290
3.3 Experimental methods .................................................................................................... 291
3.4 Reliability assessment ..................................................................................................... 292
3.5 Rules and regulations ...................................................................................................... 294
3.5.1 Harmonized common structural rules ............................................................... 294
3.5.2 Updates to offshore rules and guides ................................................................ 298
4.
ULTIMATE STRENGTH OF VARIOUS STRUCTURES .................................................... 299

4.1 Tubular members and joints ........................................................................................... 299
4.1.1 Tubular Members .............................................................................................. 299
4.1.2 Tubular Joints .................................................................................................... 300
4.2 Steel plate and stiffened plates ....................................................................................... 301
4.2.1 Introduction........................................................................................................ 301
4.2.2 Analytical Formulations for Ultimate Strength of Stiffened Panels ................ 302
4.2.3 Uniaxial Compression ....................................................................................... 302
4.2.4 Multiple Load Effects ........................................................................................ 303
4.2.5 Panels with openings, cut-outs or rupture damage ........................................... 304
4.2.6 Welding effects .................................................................................................. 304
4.2.7 In service degradation........................................................................................ 305
4.2.8 Experimental Testing......................................................................................... 305
4.2.9 Optimization ...................................................................................................... 306
4.2.10 Conclusions........................................................................................................ 306
4.3 Shells ............................................................................................................................... 306
4.4 Ship structures ................................................................................................................. 308
4.4.1 Progressive collapse methods............................................................................ 309
4.4.2 Damaged structures ........................................................................................... 310
4.4.3 Corrosion ........................................................................................................... 310
4.4.4 Complex ship structural components and complex loading............................. 310
4.4.5 Reviews and applications .................................................................................. 312
4.5 Offshore structures .......................................................................................................... 312
4.6 Composite structures ...................................................................................................... 314
4.6.1 Failure identification and material degradation models ................................... 315
4.6.2 Ultimate strength of composite stiffened panels and box girders .................... 316
4.6.3 Environmental effects ........................................................................................ 317
4.6.4 Compression after Impact ................................................................................. 317
4.7
281
Aluminum structures ...................................................................................................... 318

4.7.1 Introduction........................................................................................................ 318
4.7.2 Weld-induced effects ......................................................................................... 318
4.7.3 Formulation Development................................................................................. 320
4.7.4 Experimental Investigation ............................................................................... 320
4.7.5 Fiber-Reinforced Polymer Strengthened .......................................................... 321
4.7.6 Sandwich Panels ................................................................................................ 321
4.7.7 Hull Girder ......................................................................................................... 321
4.7.8 Summary and Recommendation for Future Works.......................................... 322
5.
BENCHMARK STUDY ........................................................................................................... 322

5.1 Small box girder.............................................................................................................. 322
5.1.1 Introduction........................................................................................................ 322
5.1.2 Model Parameters .............................................................................................. 323
5.1.3 Baseline Calculations ........................................................................................ 324
5.1.4 Comparison with Solid Element Mesh ............................................................. 327
5.1.5 Comparison with Smith method ....................................................................... 328
5.1.6 Effect of Imperfection Amplitude and Shape ................................................... 329
5.1.7 Effect of Material Model Parameters ................................................................ 331
5.1.8 Effect of Plating Thickness ............................................................................... 331
5.1.9 Summary/Conclusions....................................................................................... 332
5.2 Three hold model of hull girder...................................................................................... 332
5.2.1 Calculation cases ............................................................................................... 332
5.2.2 Calculation results ............................................................................................. 335
5.3 Summary and Recommendation for Future Works ....................................................... 338
6.
CONCLUSION AND RECOMMENDATION ....................................................................... 339
REFERENCES .................................................................................................................................. 340
282
1.
INTRODUCTION
Ultimate strength is a critical and fundamental assessment in the design of a ship or offshore structure.
The global ultimate strength of a structure is usually first assessed in the early phases of design. Evaluation
of the local ultimate strength of scantlings is also an important part of structural design and analysis.
Buckling and elasto-plastic collapse dominate the ultimate strength of slender members in compression
while yielding dominates the ultimate strength of members in tension.
This report concerns ductile behaviour of structural and components. Due to the improvements of
toughness of the material used in ships and offshore structures, brittle fracture is now a rare occurrence. For
plates greater than 50 mm in thickness, such as may be used in the upper deck plating of a container ship,
brittle fracture can developed as the result of fatigue crack propagation from an initial defect. As this can be
avoided through adequate inspection of defects, the limit state of brittle fracture is excluded from discussion
in this report.
It is now common to design ship and offshore structures to withstand buckling or yielding under the
design load. However, until the middle of the 19th century, the design criterion was the breaking strength of
the material. This is partly because brittle iron of low tensile strength was used for ship structures at that
time, and partly because buckling was not then well understood. It was after (Bryan, 1891) that panel
buckling was theoretically understood and calculated.
From the beginning of the 20th century, it had become common to consider the buckling strength as a
design criterion. Moving into the 21st century, this has evolved into the ultimate strength criteria.
The first attempt to evaluate the ultimate strength of ship structure was performed by Caldwell (1965). He
applied Rigid Plastic Mechanism Analysis to evaluate the ultimate hull girder strength. The influence of
buckling was considered by reducing the yield stress of the material in the buckled part.
The finite element method (FEM) was introduced in 1956 by Turner et al. (1956). At first FEM was used
only for the analysis of elastic behaviour of structures. To evaluate the ultimate strength of structural
members and systems, it is necessary to consider the influence of both buckling of structural members and
yielding of materials. Since the early 1970s, such analysis, so called elasto-plastic large deflection analysis,
has become possible to perform using FEM. However, computations were time consuming and it was two
decades before commercial codes were commonly used for such analysis.
As with previous ISSC reports, a literature survey is performed related to assessment procedures for
ultimate strength (see Chapter 3) and ultimate strength of various structures, such as tubular members, plates
and stiffened plates, shells, ship structures, offshore structures, composite structures, and aluminum
structures (see Chapter 4). It was from the 10th ISSC that benchmark calculation using different nonlinear
codes were completed by this committee (Oliveira, 1988), and benchmark calculations have been performed
in each committee since 1988.
For this report two benchmark calculations are performed (see Chapter 5). Both benchmarks show the
application of nonlinear finite element analysis to the ultimate strength problem. The first investigates a box
girder ultimate strength under four point bending loads and compares results with the experimental results
by Gordo et al. (2009). The second benchmark investigates the hull girder ultimate strength of a bulk carrier
considering the effect of initial imperfections and lateral loadings.
To complete a rational based design, it is usual to consider the ultimate strength as the strength standard
instead of yielding and buckling strength. Recent developments in design standards for the marine industry
have led to the Goal-Based New Ship Construction Standards (GBS) by International Maritime
Organization (IMO), and Common Structural Rules (CSR) by International Association of Classification
Societies (IACS). The GBS consists of five tiers as indicated in Figure 1. In Tier I, goals are specified for
design and construction of new ships. In Tier II, functional requirements are specified to achieve the goals.
Tier III, is verification of Tier IV, which is an existing framework of regulations, IMO conventions and
rules of recognized organization such as classification societies. CSR are closely related to GBS though Tier
IV. An important aspect of CSR is the requirement to evaluate ultimate hull girder strength as well as the
ultimate strength of plates and stiffened plates.
In the Common Structure Rules (CSR), which came into force in April 2006, the multi-step procedure
(Smiths method) is applied for calculating hull girder ultimate strength of bulk carriers (CSR-BC). In the
Harmonized Common Structure Rules (CSR-H), which combines CSR-BC(for bulk carriers) and CSR-T
(for double hull tankers) and comes into force in 2016, a double bottom factor, db will be introduced to take
account of the effect of lateral loading on the hull girder ultimate strength. Structural redundancy is also
taken into account in CSR-H according to SOLAS regulations.
With these developments ultimate strength assessment is now becoming a more important issue to ensure
the safety of ship structures. From this point of view the role of this committee continues to be very important.
283
Figure 1. Framework of IMO GBS (Goal Based Standard)
2.
FUNDAMENTALS
2.1
Design for Ultimate Strength
A limit state is defined as a condition that a structural member or an entire structure fails to perform its
designed function. There are four types of limit states: Ultimate limit states (ULS); serviceability limit states
(SLS); fatigue limit states (FLS) and accidental limit states (ALS). This report is concerned with ULS, with
a focus on the global hull girder strength of a ship. An equivalent treatment could be applied to an offshore
structure.
A ship hull girder, which is composed of stiffened plates, is subjected to a variety of forces including selfweight, cargo weight, buoyancy force and wave load. Under these loads, bending moment, torsional
moment, and shear force are exerted on the hull girder cross section. The strength of the hull girder to resist
these forces and moments is called longitudinal strength. This is the most fundamental and important
strength of ship structure. Ships must be designed to withstand the maximum expected loading, especially
bending moment, throughout its lifetime.
The following criterion is usually applied for hull girder ultimate strength assessment, using a partial
safety factor approach:
S M S + W MW MU / M
(1)
where, M S and M W are the maximum still water vertical bending moment and wave vertical bending
moment respectively, M U is ultimate moment capacity of the hull girder, and S , W and R are the
partial safety factors reflecting uncertainties and ensuring the overall required target safety margin
Recently, a probabilistic approach using the reliability index and AFORM method (advanced
first order second moment method) was attempted to apply for the assessment of ultimate strength:
=
D C
D2 C2
(2)
where, D and C are the mean value of demand (load) and capacity (strength) respectively, D
and C are the standard deviations of demand (load) and capacity (strength) respectively.
2.2
General Characteristics of Ultimate Strength
The ultimate strength is affected by many factors including:

1) Component scantlings (plate thickness, scantling of stiffener, supported span and space of
plate and stiffeners, etc.)
2) Material properties (elastic modulus, yield strength, stress- strain curve after yielding, etc)
3) Initial imperfections ( initial distortion, residual stress)
4) Load type (static or dynamic (ratio between duration and natural period of structures), etc.)
5) Additional loads (thermal load, lateral load in hull girder strength, etc.)
6) Age-related deteriorations (corrosion, fatigue crack)
7) Accidental issues (collision, grounding, and fire)
8) Human factors (mal- operation)
284
It can be said that among the above-mentioned factors, imperfection is one of the most influential
parameters on ultimate strength. The effect of imperfection, especially initial deflection, on the ultimate
strength is briefly explained as follows.
In the case of stiffened panel under compressive loading, greater deflection is produced when initial
deflection is present as an additional destabilizing moment is generated, and it reduces the ultimate strength.
As shown in the numerical results by Ueda et al. (1985), after buckling under the compressive loading,
especially in the case of elastic buckling, the deflection of buckling mode increases more while deflection of
the other mode decreases. Therefore, in discussing the effect of initial deflection on ultimate strength, it is
necessary to discuss not only the magnitude but also the mode shape of initial deflection, as both can
significantly influence ultimate strength.
The effect of initial deflection on buckling and ultimate strength is larger in shell structures that in
column and plate structures because of the difference in the post buckling behaviour. The load carrying
capacity of shell structure reduces greatly after buckling, and therefore the buckling strength can be usually
treated as the ultimate strength in shell structure. A constrained plate structure under in-plane compression
gradually increases after elastic buckling and reaches ultimate strength when it yields due to the bending
moment caused by lateral deflection. In this case, the ultimate strength is different from the buckling
strength.
The ultimate stage of hull girder bending under hogging moment occurs when the bottom reaches
ultimate strength in compression after the strength deck reaches the yield stress in tension. In this case, the
deck structure can maintain the yield stress after yielding.
Under the sagging moment, the hull girder can endure excessive bending moment after the strength deck
reaches ultimate strength by redistributing the stress and sharing the load with other members. It then
reaches ultimate strength when the stress of bottom structure achieves yield stress in tension. But in this
case, the deck structure cannot maintain its resistance after reaching ultimate strength in compression.
Accordingly, the hull girder reaches ultimate strength just after the time when the deck reaches ultimate
strength. Therefore it is important to take into account the post buckling strength of stiffened structure for
estimating accurately the hull girder ultimate strength under sagging condition.
The ultimate hull girder strength is reduced by the residual deformation from accidental loads such as
grounding and collision, and it is also affected by corrosion and fatigue cracks due to repeated loads. The
assessment of these effects on ultimate strength is essential.
In the case of composite structures, especially produced by fiber reinforced plastic, such as GFRP (Grass
Fiber Reinforced Plastic), CFRP (Carbon Fiber Reinforced Plastic), and AFRP (Aramid Fiber Reinforced
Plastic), the collapse behaviours differ from those of metallic structure, such as welded steel and aluminum
alloy. In tension, FRPs only show small plastic deformation after yielding. An exception is when the
composite material is composed of short length metal fiber and resin, in which sliding between fiber and
resin can take place. In compression, buckling of fibers can occur in addition to buckling of structural
member. Fiber buckling may greatly reduce the ultimate strength of the structural member in compression.
When the structure is damaged by lateral impact loading, delamination between fiber and resin takes place,
thus reducing the buckling strength of the fiber. Considerable research in this area in the aerospace field can
be applied to the design of ship structure using composite materials.
3.
ASSESSMENT PROCEDURE FOR ULTIMATE STRENGTH
3.1
Empirical and analytical methods
3.1.1
Introduction
Analytical methods for predicting hull girder ultimate strength began with Caldwell (1965), who
developed a direct assessment method for a simple rectangular stiffened hull section, taking into account
material yield and buckling. Direct assessment methods were further developed by Paik and Mansour
(1995) and others, but as these methods normally assume a stress distribution through the ship section,
they tend to be limited to simple structural geometries, and cannot account for reduced post-collapse
strength in compression. An alternative incremental iterative approach developed by Smith (1977) takes
account of the longitudinal elasto-plastic response of individual structural components, including their
post-buckling behaviour. In the years since these pioneering efforts, progressive improvements to both
direct assessment and incremental iterative methods have led to versions of both approaches being
adopted in the IACS rules.
Current research is focused on extension of these methods to biaxial bending and damaged structures; and
the development of empirical formula for the strength of damaged structure.
3.1.2
285
Hull structures
Paik et al. (2013) further extended the Paik-Mansour direct assessment method by improving the assumed
distribution of bending stresses at failure. Whereas the earlier Paik-Mansour formula assumed that only
the tension flange (i.e., the deck or hull bottom) achieves yield stress in tension, the Modified PaikMansour formula allows for a more extensive region of tensile yielding (see Figure 2). On the
compression side, the approach is the same as in the Paik-Mansour method, with the entire compression
flange and a portion of the side shells at the compressive ultimate strength, and the remaining parts of the
side shells in a state of elastic stress. The heights of the yielded area (hY) and the collapsed area (hC) in the
side shells are unknowns to be determined by satisfying force equilibrium in the section. An iterative
procedure is described for determining these two unknowns.
Figure 2. Assumed stress distributions used in the Modified Paik-Mansour method (Paik et al., 2013).
As with previous approaches based on aassumed stress distributions, the post-collapse behaviour of
structure is not accounted for, potentially leading to un-conservative ultimate strength predictions.
However, the results presented show that the method gives comparable ultimate strength predictions to
other calculation methods (Smiths method, ISFEM, and FEA) for a wide range of hull types. The
improvements to the original Paik-Mansour approach are shown to be most significant for double hull
structures in hogging, where large regions of yielding can develop in the upper side shell. In other cases,
the height of yield area (hY) remains small. Paik et al. (2012a) applied the modified method to the
assessment of double hull oil tankers with grounding damage and compared the results with those
obtained with FEA, ISFEM, and Smiths method.
Another important development has been the extension of Smiths method for progressive collapse to
compartment level failure modes. Benson (2011) and Benson et al. (2013d) describe an approach in
which load-shortening curves for compartment-length grillage panels are incorporated within the
framework of a Smiths method approach. The load-shortening behaviour of grillage panels is determined
using a semi-analytic method based on large deflection orthotropic plate theory. Benson (2011) describes
the semi-analytical method in detail, which involves computing the collapse load for all overall collapse
modes and comparing these with interframe collapse loads for plate, stiffener and beam-column modes to
determine the mode of failure. The method can be used to determine the load-shortening behaviour of a
grillage panel as well as the ultimate strength. The method is more efficient than nonlinear FEA of large
grillage panels, and can be used when sufficient regularity exists in the structure. Example applications of
box girders and aluminium hulls show that the compartment level collapse scheme can take into account
interframe as well as overall grillage collapse modes.
Figure 3 gives the moment-curvature relationships for a box girder designs predicted using the
compartment-level collapse analysis (ProColl), nonlinear FEM, and a semi-analytical interframe analysis
method similar to the standard Smiths method. The results are given for both a lightly framed (T1) and a
heavily framed (T2) design variants. The results indicate that the moment curvature responses and
ultimate strengths for lightly framed designs deviate widely from the interframe solution; while ProColl
predictions agree well with FEM results obtained using ABAQUS.
286
Reliability studies of hull girder ultimate strength often use Smiths method within the context of a
Monte Carlo simulation scheme (e.g. (Gaspar and Guedes Soares, 2013)). This work is covered more
fully in the Reliability section.
Lightly framed structure
Heavy framed structure.
(a) Moment-curvature response under vertical bending
(b) Deformed mesh plot
Figure 3. Calculation results of ultimate strength for box girder models (from (Benson et al., 2013d))
3.1.3
Residual Strength of Damage Hull Structures
An important step forward has been the development of Residual Strength vs Damage Index (R-D)
diagrams and their associated empirical formula for representing the residual bending strength of hull
girders with grounding damage. Paik et al. (2012a) developed empirical formulas for residual strength of
double hull oil tankers of the form,
Mu
= A2GDI 2 + AGDI
+ 1.0
(3)
1
M u0
where,
is the residual strength of the damage hull girder,
is the ultimate strength of the
undamaged hull girder, and the grounding damage index (GDI) is given by
GDI =
Aro
A
+ ri
Aoo
Aoi
(4)
where the first term is the ratio of the area lost by grounding damage (Aro) and the original area of the
outer bottom (Aoo), and the second term involves a similar area ratio for the inner bottom. The parameter
is a correction factor that weights the contribution of the inner bottom to the hull girder ultimate
strength.
The coefficients A1 and A2 are determined by fitting curves to ultimate strength calculations plotted on
an R-D diagram. By these empirical relations, the upper limits of GDI satisfying the IMO requirement
for residual strength of Mu/Mu0 0.9 are determined for the four tanker designs. The results, given in
Table 1, show that that the Panamax design is more damage tolerant to grounding than Suezmax,
Aframax and VLCC designs. The main factor influencing sensitivity to damage appears to be ship length,
and the upper limit of GDI is found to be linearly related to the overall length.
Table 1. Upper limits for the grounding damage index (GDI) for four sizes of double hull tanker (from
(Paik et al., 2012a)).
Hog
Sag
VLCC
0.2882
0.4225
Suezmax
0.2992
0.4344
Aframax
0.3376
0.4366
Panamax
0.3965
0.4363
287
Kim et al. (2013a) used a similar approach to develop R-D diagrams and empirical relations for four
container ship designs with grounding damage. The upper limits of GDI satisfying the IMO residual
strength requirement are given in Table 2 for four sizes of container ship. Kim et al. (2014) also
developed time-dependent R-D diagrams and associated empirical formula to describe the combined
effect of grounding damage and time-based corrosion wastage.
Table 2. Upper limits for the grounding damage index (GDI) for four sizes of
container ship (Kim et al., 2013a) .
Hog
Sag
3500 TEU
0.2882
0.4225
5000 TEU
0.2992
0.4344
7500 TEU
0.3376
0.4366
13,000 TEU
0.3965
0.4363
In general, asymmetric damage to a hull results in structural asymmetry as well as a biaxial loading
condition if flooding results in a non-zero heel angle. Some researchers have attempted to improve the
accuracy of Smiths method in this situation. Choung et al. (2012a) developed version of Smiths method
in which the position and rotation angle of the neutral axis plane is calculated and updated as bending
moment loads are applied. Their work differs from previous biaxial formulations of Smiths method in
that the moment plane (plane normal to the moment-producing shear forces acting on the hull) is defined
as an entity distinct from the neutral axis plane. In the general case of an asymmetric hull with a nonzero
heel angle, the neutral axis plane may be both displaced and rotated from the moment plane. By this
method, the vertical bending strength of a double hull tanker is evaluated using a Smiths method scheme
for grounding and collision damage scenarios, in which damage is modelled as loss of structure. It is
shown that the standard Smiths method approach (as recommended in the IACS CSR) can result in
overestimation of the ultimate strength, particularly when asymmetric damage to the hull girder and a
large heel angle are present.
Muis Alie et al. (2012) and Muis Alie (2013) derive a generalized expression for the incremental
moment-curvature relation of an asymmetric cross section under-biaxial bending. Their expression is
similar to those that have been previously developed, but use a derivation that is closer to the original
Smiths method approach developed by Smith C.S. They apply their formulation to two bulk carrier
and one oil tanker designs with different amounts of damage to the side shell, where damage is again
modelled as loss of structure. The moment-curvature relations are computed both with and without
rotation of the neutral axis for the case of an applied vertical bending moment. In their formulation,
neutral axis rotation arises from the growth of a horizontal component of curvature of the
asymmetrically damaged hull. Their results show that in some cases, a significant overestimation of the
vertical strength can arise by not considering the neutral axis rotation (horizontal curvature
component). Fujikubo et al. (2012a) extended this work by deriving simplified formulae for the
residual strength in terms of elastic properties of the section and the strength of a critical member. The
simplified expression is shown to compare well with direct calculation of the residual strength using
the HULLST program.
Two studies have treated residual hull girder strength for more specific damage cases than simple loss
of structure. Liu et al. (2012) proposed a combined elastic and rigid-plastic approach to determine
simplified load-shortening curves of double bottom structure subject to indentation damage. The ultimate
strength of damaged longitudinal stiffener-plate combination was determined to be the intersection point
of the load-deflection response for elastic compression and bending, and that of the rigid plastic
mechanism. The indentation damage for a longitudinal stiffener-plate combination was parameterized as a
vertical denting distance (h) and a tripping angle (see Figure 4).
The method is applied to determine the ultimate strength and resistance-strain relationship for a double
bottom structure with different amounts of indentation damage. Good comparisons between analytical
with purely numerical assessments were obtained, as shown in Figure 5 for the case of 150 mm
indentation damage of five stiffeners in an outer bottom structure.
Shi et al. (2012b) proposed a simplified formula for the strength under combined torsion and bending
of an open box girder with crack damage. The proposed formula employs a cross-sectional area reduction
factor to account for the loss in strength due to the presence of a vertically oriented crack in the side shell,
as well as a reduction factor to account for loss of stiffness from the large opening in the girder. The
proposed formula is applied to cases of pure bending, pure torsion, and combined bending and torsion.
In general, ultimate strengths predicted by this method are very conservative in comparison to FEM
calculations.
Intact
Damaged
Topflangelevel
bef
bf
hw
hwe
Resistanceforce(MN)
288
250
200
TotalDoublebottomFEM
OuterDoublebottomFEM
InnerDoublebottomFEM
TotalAnalytical
OuterAnalytical
InnerAnalytical
150
100
50
0
0.000
0.002
0.004
0.006
0.008
0.010
Meanstrain
b
Figure 4. Simplified representation of indentation damage (left) and comparison of load-shortening curves (right)
(from (Liu and Amdahl, 2012)).
3.1.4
Plates and stiffened plates
Researchers frequently use the results of a large number of FEA simulations to derive formulas describing
the strength as a function of one or more parameters. For example, Silva et al. (2013) used this approach
to describe the ultimate strength of rectangular plates with exponential functions of time, plate
slenderness, and the degree of degradation (DOD) from randomized patterns of corrosion pits. By this
approach, the expected (mean) value and standard deviation of ultimate strength is characterized. Jiang et
al. (2012b)used the results of numerical analysis of plates with partial depth pitting to derive empirical
formulas for the strength under uniaxial loading in terms of the plate slenderness ratio and the volume
loss due to corrosion. Jiang et al. (2012a) followed a similar approach to derive empirical limit state
relations for the strength of pitted mild steel plates under biaxial loading, also as a function of the plate
slenderness ratio and the volume loss due to corrosion.
Brubak et al. (2013) developed limit state criteria for rectangular stiffened plates under compressive
loading, where plate edges are either simply supported on four sides, or are simply supported on three
sides and with one free or partially restrained edge. The ultimate strength of the stiffened plates is
calculated using these criteria using a Rayleigh-Ritz method in conjunction with large-deflection theory
and an arc-length solution method. Examples of stiffened plates with different support and stiffener
arrangements are compared with FEM results. Yang et al. (2013) used a similar semi-analytical approach
in the context of small deflection theory to analyse the compressive ultimate strength of simply-supported
rectangular composite plates under uniaxial loading.
A different approach was taken by Cho et al. (2013), who adopted the generalized Merchant-Rankine
formula as the basis of an empirical representation for the ultimate strength of stiffened plates subject to
axial and transverse compression, shear loads and lateral pressure. Euler-column buckling, overall
grillage buckling, stiffener tripping, and plate buckling in compression and shear, are all accounted for
using elastic buckling stresses, each with an applied knockdown factor. The knockdown factors are
determined by regression, using numerical results in which each load type (axial compression, transverse
compression, and shear force) is applied separately. Predictions for combined axial and transverse loading
show good comparison with numerical results, except for thickest plates where the proposed formula
tends to underestimate ultimate strength. The stiffened plates used in the study were similar to those
described in the ISSC 2000 committee V1.2 report. These included flat bar, angle-bar and tee-bar
stiffened plates, with plate, column and stiffener tilt imperfections.
3.2
3.2.1
Numerical methods
Introduction
Since Caldwell (1965) firstly proposed a rational method considering both buckling and yielding effects,
there has been continuous effort worldwide with regard to the study of ultimate hull structure strength and
many researches have been carried out. The existing methods for calculating ultimate strength of hull
structures can be classified into three categories: direct methods (linear method, empirical formulas),
progressive collapse analysis (Smith method), and numerical methods such as idealized structural unit
method (ISUM) and nonlinear finite element method (nonlinear FEM). Currently, numerical research on
ultimate strength of hull structures focuses on the idealized structural unit method and nonlinear FE
approaches.
3.2.2
289
Nonlinear FE method
An early application of nonlinear FEM for the analysis of hull girders is presented by Kutt et al. (1985),
who uses the ABS nonlinear FEM program USAS to calculate the longitudinal strength of four ship hulls
including a passenger ship and a tanker. In recent years, with the development of computer technology,
many FE programs have been applied to ultimate strength analyses of hull structures, such as ABAQUS,
ANSYS, MARC, ADINA.
Typically, nonlinear FEM analyses of the progressive collapse utilize a static solver together with an
equilibrium convergence iterator using either the Riks arc length method or modified Newton-Raphson
method, (Cook et al., 2002). The static solver assumes that the time dependent mass and inertia effects are
small and thus can be neglected, (Hibbitt et al., 2010), which implies a quasi-static structural response. This
is usually valid if the loading frequency is less than a quarter of the lowest natural frequency of the structure.
S4R quad elements (4 node finite strain elements with reduced integration) are widely applied when
dealing with nonlinear problems using ABAQUS, (Shi and Wang, 2012a) (Shi and Wang, 2012b),
(AbuBakar and Dow, 2013), (Benson et al., 2013a), but other shell elements in ABAQUS are also used,
such as S8R5, (Shanmugam et al., 2014), S4, (Amlashi and Moan, 2008), and S9R5, (Kim and Yoo, 2008).
In analyses using ANSYS, the Shell 181 element is commonly used as it is well suited for large rotation
and large strain nonlinear applications.
Other FEM software such as ADINA and MARC are sometimes used for assessing the ultimate strength
of hull structures. Hu et al. (Hu and Jiang, 1998) carried out the collapse analysis of 12 full-scale stiffened
plates under axial and out-of-plane loads using the nonlinear FE program ADINA. Ferreira and used
ADINA to perform nonlinear FE simulations for the ultimate strength prediction of isotropic and orthotropic
plates under uniaxial compression and a four-node shell element MITC4 was applied. A series of collapse
analyses were performed by Tanaka et al. (2014) applying the program MSC.MARC to stiffened panels
subjected to longitudinal thrust by varying numbers, types and sizes of stiffeners.
Although the quasi-static response is commonly of interest for ultimate strength assessment, explicit
dynamic solvers, such as LS Dyna, are now also used to overcome convergence difficulties in large models.
Extending the model size for ultimate strength analyses to one hold or three hold models (to consider the
effect of lateral loading on the ultimate strength) increases the complexity of the numerical model so that
implicit static solvers often fail in convergence before reaching the post ultimate range of the structure,
leaving the analyst without a result. Explicit dynamic solvers do not face this convergence challenges, but
their results rely on controlling inertia and damping effects properly. Only if the inertia effect is reduced as
far as possible, e.g. by increasing the time step size or reducing the displacement increment, i.e. reducing the
loading velocity, a quasi-static solution can be expected. Otherwise the ultimate strength will be
overestimated by a dynamic response. The box-girder benchmark of this committee, as presented in section
5.1, shows that a well-controlled explicit dynamic solution gives comparable results to implicit static
solutions. The benchmark on a hull girder by this committee, given in section 5.2, clearly demonstrates the
advantage of the explicit solver: Only the explicit dynamic solver obtains results for all cases to be analyzed.
The implicit static solvers often do not convergence up to the final collapse. In cases where both implicit
and explicit solvers have given results, it is again shown that the results agree well.
AbuBakar et al. (2013) provide a general approach for calculating the ultimate strength of box girders
using a dynamic solver which was carried out in the ABAQUS explicit code. Benson et al. (2013a) compare
static with dynamic FEM solvers for the intact box girder which indicates that either approach is valid for
the purpose of progressive collapse analysis. LS Dyna is used for ultimate strength assessment considering
lateral loads by Toh et al. (2014). Yamada (2014) demonstrates the feasibility of using explicit dynamic
solvers to combine collision and collapse analyses.
Ships and marine components are examples of complex thin-walled structures that are composed of
various plate elements including stiffened and unstiffened plates characterized by different combinations of
geometry and loading conditions which are widely studied.
Hu et al. (Hu et al., 2004) analyzed the tensile and compressive residual ultimate strength of stiffened
panels and unstiffened plates considering the effects of fatigue cracks using the FE method. Paik (2007)
undertook series of ANSYS nonlinear FE analyses with varying the cutout size as well as plate
dimensions to investigate the ultimate strength characteristics of perforated steel plates under edge shear
loading. Alinia et al. (2007) (2008) investigate the influence of central cracks and edge cracks on the
residual strength and stiffness degradation of shear panels considering various geometrical and
mechanical characteristics of cracked panels using the FE program ANSYS. Paik (2008) investigated the
ultimate strength characteristics of perforated steel plates under combined biaxial compression and edge
shear loads with various plate dimensions. Paik (2009) examined the residual ultimate strength
290
characteristics of steel plates with longitudinal cracks under axial compressive actions by varying the
crack location, the crack size, the plate thickness and the plate aspect ratio. Kumar et al. (2009)
investigated ultimate load carrying capacity of stiffened plates with openings under axial load, and with
axial and out-of-plane loads using ANSYS. Considering the effects of a random corrosion thickness
distribution, Silva et al. (2013) investigated the ultimate strength of unstiffened rectangular steel plates
subjected to uniaxial compressive load. Shanmugam et al. (2014) investigated the ultimate load capacity
of stiffened plates subjected to combined action of in-plane load and lateral pressure using the computer
code ABAQUS.
To simplify the numerical models and structure strength problems, hull box girders are commonly used
by many authors to simulate large surface ships and double hull tankers.
Series of comparisons between results from nonlinear FEM analyses and tests of box girders are
presented by Qi et al. (2005) providing an integrated framework of finite element analysis which can be
used as benchmark of comparative study of ultimate hull girder strength. Amlashi et al. (2008) carried out
ultimate strength assessments of the hull girder of a bulk carrier under alternate hold loading conditions
based on the nonlinear FEA program ABAQUS. Kim et al. (2008) investigated the ultimate strengths of
steel rectangular box beams subjected to combined action of bending and torsion incorporating the effects of
residual stresses and initial imperfections using ABAQUS which proved in good agreement with those from
other researchers experimental investigations. The residual ultimate strength of cracked box girders was
studied as a function of crack sizes and crack locations subjected to torsional loading using same FEA
program by Shi et al. (2012a) and residual ultimate strength of open box girders with cracked damage under
combined loads was investigated by Shi et al. (2012b).
3.2.3
Idealized structural unit method
The idealized structural unit method (ISUM) can allow rapid assessment of large structural arrangements by
simplification into smaller constituent parts. The method has been widely recognized by researchers due to
large advantage in reducing modelling effort and computing time compared with nonlinear FEM.
Ueda et al. (1974) proposed the concept of idealized structural unit in which a large structure
member is regarded as one unit and this method was then applied to an analysis of offshore structures
by Ueda et al. (1983). Paik et al. (1996) developed a program of non-linear analysis of large plated
structure using ISUM which employed five types of the ISUM units (beam-column unit, unstiffened
plate unit, stiffened plate unit, hard unit and virtual unit). Paik et al. (2001) used the ALPS/HULL
software to investigate the ultimate strength characteristics of ship hulls with large hatch openings
under torsion based on the ISUM units. Fujikubo et al. (2002) proposed a new simplified model for
collapse analysis of stiffened plates in the framework of ISUM combining plate and beam-column
elements. Paik et al. (2003a) presented a summary of pertinent ISUM theory and its application to
nonlinear analysis of steel plated structures which illustrated the possible accuracy and versatility of the
ISUM method. ALPA/SCOL (ALPS/SCOL, 2006) applied the ISUM as a computer program for the
progressive structural crashworthiness analysis under collisions and grounding using three types of
ISUM elements. Paik et al. (Paik et al., 2008b) compared ANSYS FEA with ALPS/HULL to analyze
the ultimate vertical bending moment capacity of a double hull oil tanker structure which proved the
validation of ALPS/HULL method solutions. Paik et al. (Paik and Kim, 2008) modeled the rectangular
plate elements as ISUM plate elements to investigate the progressive collapse of box columns which
was verified by a comparison with ANSYS nonlinear FEM solutions. Underwood et al. (2012b) used
ISUM to investigate the collapse of stiffened steel panels and proposed a new ISUM for strength
assessment of damaged structural arrangements.
3.2.4
Conclusion
In maritime industry, the ultimate limit state is now applied as a basis of structural design and strength
assessment, (IACS, 2006b) (IACS, 2006a), (IMO, 2006) and (ISO, 2006), and it is now a mandatory task
for structural design of ships and offshore structures. The nonlinear finite element method (FEM) has
become a dominant computational approach for complex structural engineering problems with all modelling
herein considers both geometrical and material nonlinearities. However, taking into account modelling
effort and computing time when using FEM, ISUM can provide a preview of the ultimate collapse of ship
structures before considering the conduct of nonlinear FEM, and more valid ISUM elements need to be
developed. It seems that most of the collapse analyses neglect the time dependent mass and inertia effects
and imply quasi-static structural responses while ultimate dynamic responses of hull structures due to the
damage of grounding, collision and wave loads are still rarely studied.
3.3
291
Experimental methods
Experimental studies to assess the ultimate strength of whole ship structures are few. However, to
understand the trend and direction, some experimental studies which can be helpful are discussed in this
section. Generally, static ultimate strength experiments are controlled by applied displacement with very
slow velocity, such as 0.05mm/s applied on compressive stiffened plate which can be considered as static
or quasi-static loading regardless inertia effect, and static ultimate strength of stiffened plate can be
obtained once reacting load reaches maximum value, followed by a decline process of the reacting load.
Two series of longitudinally stiffened steel plates with and without transversal stiffeners subjected to
uniform axial in-plane load carried out to study the buckling and post-buckling up to final failure
(Ghavami and Khedmati, 2006).
Compressive tests on short, long and intermediate stiffened panels are conducted by Gordo et al. (2008a)
(2011) (2012) by changing distance between transverse frames. Every kind of stiffened plate with different
length has eight stiffened panels subjected to axial compression until collapse and beyond. The specimens
are three-bay stiffened panels with associated plate made of very high tensile steel S690. The use of this
very high strength steel led to the unconventional solution of using U stiffeners. Comparisons of this new
stiffener type and conventional bar stiffeners or L stiffeners are applied.
Load (kN)
1600
FB2B2F6A
1st cycle
2nd cycle
3rd cycle
Stiffener ; FB30x8
Frame ; L60x40x6
Frame space ; 200mm
1200
800
400
8
6
4
Displacement (mm)
10
Figure 5. Load-displacement curve for a sample panel (Xu and Soares, 2013a).
Tests of narrow and wide Stiffened Panels under axial compression until collapse and beyond are
presented to investigate the collapse behaviour and strength by Xu et al. (2013a) (2013b) which allows
analyzing the effect of the width on the strength of stiffened panels. The tests were made on panels with
two half bays plus one full bay in the longitudinal direction.
There are several ultimate strength experiments on plate or stiffened plate with cutout. Experimental
investigations are carried out up to collapse on eighteen stiffened steel plates having square cutout,
rectangular cutout and reinforcement around rectangular cutout (Alagusundaramoorthy et al., 1999).
Rectangular plates made by 1200A aluminum alloy with central inclined crack subjected to axial
compression are carried out (Paik, 2009). Residual ultimate strength of steel plates with longitudinal
cracks under axial compressive actions is examined (Seifi and Khoda-yari, 2011) . Tests of buckling and
post-buckling behaviours of steel plates having groove-shaped cutouts of various dimensions and angles
subjected to axial compression are investigated (Shariati, 2012).
An experimental study on stiffened plates subjected to combined action of in-plane load and lateral
pressure is investigated (Shanmugam et al., 2014). Lateral load carrying capacity of stiffened plate
drops with increase in axial load and vice-versa. It is found that plate slenderness ratio has significant
influence on the ultimate load capacity of stiffened plates subjected to both in-plane load and lateral
pressure.
There are several ultimate strength experiments on aluminum plate or stiffened plates. Axial
compression tests on longitudinally stiffened aluminum panels with open section (L-shaped) stiffeners
and closed section stiffeners were carried out. And two deformation modes were observed in the tests;
regular flexural buckling of the entire panel towards either the stiffener side or the plating side, and, for
the case of the panels with L-shaped stiffeners, collapse initiated by stiffener tripping (Aalberg et al.,
2001) . An experimental investigation of the torsional buckling strength of longitudinal stiffeners in
aluminum panels subjected to axial compression is conducted by Zha et al. (2003). Torsion buckling
strength of the stiffened panels is affected by the heat-affected zones and the reduction of ultimate
strength is up to 34%. Mechanical collapse testing on aluminum stiffened plate about fabrication related
292
initial imperfections is investigated by Paik et al. (2008a). And corresponding statistical database of
fabrication related initial imperfections on welded aluminum stiffened plate structures is also developed.
An experimental investigation was carried out to determine the ultimate strength of welded stiffened
aluminum panels subjected to in-plane compressive loads normal to the directions of the stiffeners by
Rnning et al. (2010). A total of 21 panel specimens with various aspect ratios and both open and closed
stiffener sections were tested. The panels failed by two different deformation modes; global flexural
buckling and local buckling of the plate elements between the stiffeners.
A series of experimental studies by Gordo and Guedes Soares feature a box girder structure under
vertical bending load. Pure bending test of a box girder with one span is conducted by Gordo et al.
(2004). Then pure bending test of box girder made by mild steel and high tensile steel are conducted
(Gordo & Guedes Soares, 2008b) (Gordo & Guedes Soares, 2009). Box girders with different spacing
between frames are also investigated (Gordo and Guedes Soares, 2014).
Experimental assessment of the ultimate strength of a box girder with different levels of corrosion is
conducted by Saad-Eldeen et al. (2011), (2013b). Corroded plate thicknesses have been measured in 212
points and a statistical analysis has been performed. The resulting corrosion wastage has been fitted by a
non-linear time variant degradation model. Pure bending test of a hull model with three longitudinal box
girder set on the deck is conducted by Wang et al. (2011).
Pure bending test of a container ship hull model under hogging is investigated to find possible failure
modes and ultimate strength, based on similar conversion relationship (Gui-Jie & De-Yu, 2012) . Pure
torque test of a ship-type hull girder with a large deck opening was investigated to find possible failure
mode and torsional ultimate strength by Sun et al. (2003).
Dynamic ultimate strength of ship or offshore structures, due to the time-dependence of the load, is
more complicated than static ultimate strength. In this condition, some additional issues have to be
considered, such as inertia effect, dynamic constitutive equation of the material. A fluid-solid impact
experiment was carried out by Li et al. (1993) for a plate falling on water from an over-water
slamming tower. Results showed that fluid-solid impact load has the millisecond order duration and a
half sine wave shape. According to the equivalent strain of a measuring point near the impacting end
of the plate, three critical criteria were used to define the critical loads of dynamic buckling, dynamic
plasticity and dynamic collapse. However, the position of the measuring point is somewhat arbitrary,
which will results in uncertainty. Fluid-solid impact on stiffened plate was not investigated in Lis
research. The elasticplastic dynamic post-buckling characteristics of simply supported columns under
axial solidfluid slamming are experimentally investigated by Cui et al. (2000). According to
observation of plastic collapse of the columns, a dynamic collapse criterion is defined. Experimental
investigations on the dynamic plastic impact buckling of slender beams were conducted by Kenny et
al. (2002). A 10 kg free-fall impact hammer applied an axial impulsive load to aluminum and cold
rolled steel beams with a fixed-slide bearing boundary condition. Cui did similar experiments and
numerical simulations, and presented similar conclusions Cu et al. (1999), Kiat et al. (2000) and
compared to Lis research. A series of dynamic collapse tests were carried out on steel plates under
axial compressive loads by Paik et al. (Paik and Thayamballi, 2003b), with loading speeds varying
from 0.05mm/s to 400mm/s. Experiments showed that the dynamic ultimate strength of steel plates
gets larger with the increase of the loading velocity. Experimental study was made for the dynamic
crushing of thin plates stiffened by stamping with 1-3 parallel V-grooves in a clamped-end condition
(So and Chen, 2007).
Currently structural reliability and uncertainty factors are seeing increased use in ship structures.
Experimental data that can be used as target values for calibration of structural models is therefore
necessary. Further experimental results on effect of ageing and imperfections in the analysis can be
helpful.
3.4
Reliability assessment
Despite significant recent advancements in ultimate strength evaluation procedures, use of reliability
methods to determine the level of uncertainty due to various parameters in the analysis is still ongoing, as
is evident from the harmonization of the classification society rules. However, more and more studies are
required in all the key design analysis of the ship structures so that uncertainty factors can be used in the
rule calibration and reliability analysis can be used in daily design practice.
Deco et al. (2012) proposed a reliability assessment frame work of ships under different operational
conditions. The reliability assessment was based on the flexural capacities associated with the ultimate
hull failure and the failure of the first stiffened panel within the selected ship section. This study was
extended to the development of a risk-informed decision tool for the optimal mission-oriented routing of
ships by Deco et al. (2013).
293
Saydam et al. (2013) analyzed an oil tanker hull girder by varying speeds, headings and sea states and
determined the reliability index for the intact and six damaged hull cases. The limit state function was
based on ultimate strength of hull girder section at midship. Also, the robustness index associated with
damage scenarios are presented using an optimization-based version of incremental curvature method. In
addition, aging effects on ship reliability are investigated.
An advanced reliability assessment of ship hull is performed by Zhu et al. (2013) using a Bayesian
updating method. They proposed an approach for reducing the uncertainty in the performance assessment
of ship structures by updating the wave-induced load effects with the data acquired from structural heath
monitoring (SHM). The proposed approach is illustrated with the Joint High Speed Sealift.
Zayed et al. (2013) considered the ultimate vertical bending moment capacity of the ship hull as a limit
state to assess the structural reliability of ship hulls incorporating the effects of corrosion degradation and
inspection and repair actions. The uncertainties in an inspection are accounted for by a probability of
detection model that introduces additional probabilistic events at the different inspection times. Ship
loading uncertainties are modelled based on the time ratio spent under each loading condition during the
ships service life.
An advanced study on the uncertainty analysis, Teixeira et al. (2013a) presented an approach to assess
the ultimate strength response of plates considering the random initial distortions, random material and
geometrical properties and random corrosion degradation. One important feature of this approach is that it
provides both the information on the contribution of each variable to the uncertainty of the response as
well as accurate predictions of the percentiles of the response in using general First-Order Reliability
Method (FORM) considerations.
On the reliability assessment of damaged ship hull girder structure, Prestileo et al. (2013) investigated
the scenario corresponding to bottom damage in the hull girder of an oil tanker. The accidental scenarios
were modeled considering the mutual relationships that arise in this case between the load and strength
stochastic variables involved. A wide range of specific bottom damage scenarios has been considered and
different Bayesian Network models (as shown in the Figure 6) have been developed to investigate each
one of them. The various damage cases are compared to each other and unconditioned to derive the
probability of failure extended to the ships life due to generic bottom damage. The vertical ultimate
bending moment has been computed for many damage cases, using a simplified progressive collapse
method.
Kim et al. (2013a) developed the residual ultimate longitudinal strength versus grounding damage
index diagram (RD diagram) for container ships. Each parameter (grounding damage location
(transverse) in the direction of the ships breadth, grounding damage height (vertical penetration extent),
grounding damage breadth (transverse extent) and rocks angle) affecting the structural damage are dealt
with as a random variable and are characterized probabilistically. In the present paper, the probability
density distributions for the grounding damage parameters defined by IMO are applied. Also, due to
computational cost limitations, only 50 damage scenarios have been selected. The grounding damage
index is defined to represent the damage severity for each of the selected scenarios. The proposed RD
diagram should be useful for defining acceptance damage criteria and making rapid salvage plans or
rescue schemes for containerships that have sustained a grounding accident.
Ibekwe et al. (2014) adopted Smiths progressive collapse approach to determine hull girder ultimate
capacity. They combined the ultimate strength with user-defined iterative-numerical framework to
determine the structural reliability of a frigate. Collision damage condition of the frigate is assessed under
pure vertical bending and combined vertical-horizontal bending. Randomness in the geometrical, material
and loading moment is considered and reliability analyses are performed by means of the curvature based
Adaptive Importance Sampling (AIS). The user-defined framework is based on the application of an
external numerical solver for the assessment of deterministic hull girder capacity and implicitly linked
within a probabilistic model for structural reliability analysis.
Nam et al. (2012) predicted the ultimate longitudinal strengths of hull girder of a VLCC (designed
according to the CSR-T) considering probabilistic damage extents due to collision and grounding
accidents based on IMO Guideline. The probability density functions of damage extents are expressed as
a function of non-dimensional damage variables. The accumulated probability levels of 10%, 30%, 50%,
and 70% are taken into account for the damage extent estimation. The ultimate strengths have been
calculated based on the progressive collapse method. Damage indices are provided for all heeling angles
due to any possible flooding of compartments from 0 degree to 180 degree with 15deg increment which
represent from sagging to hogging conditions, respectively. The analysis results reveal that minimum
damage indices show different values according to heeling angles and damage levels. A concept of
damage index is introduced in order to assess reduction of hull girder strength. In case of the collision,
the damage index is lowest at the moment plane MP=0 deg. and highest at MP = 120 deg. In case of the
grounding, the damage index is lowest at MP=180 deg. and highest at MP = 90 deg.
294

xpdLong
position
ypdTransverse
location
zidVertical
penetration
xidLongextent
yidTransverseextent
Belowinnerbottom(BIB)
5
DamageatB3p
xloc.
DamageatB3p
y loc.
8
FloodingB3p
9
Waveenvironment
10
MsStillwaterBM
MwwaveBM
11
MtTotalBM
12
MuUltimatestrength
oftheShip
13
HullGirderFailureIGrounding
14
15
Figure 6. Example of Bayesian Network model (Prestileo et al., 2013)
In the continued study, Nam et al. (2014) investigated the effect of average compressive strength of
stiffened panel on the hull girder strength and to assess the residual ultimate strength according to the
damage extents of a very large crude oil carrier considering damage extents due to collision and
grounding accidents. In order to determine extents of damage, two types of probabilistic approaches are
employed: deterministic approach based on regulations (ABS, DNV and MARPOL) and probabilistic
approach based on IMO probability density functions. Hull girder ultimate strength is calculated using
Smith method which is dependent on how much average compressive strength of stiffened panel is
accurate. A probability level of 80% in IMO; PDFs (Probability Density Function) is decided after
comparison of reduced section area among other damage extent models. They concluded that the level of
80% in IMO PDF show similar or slightly larger area reduction than DNV ship rule.
Dyanati et al. (2014) estimated the seismic reliability of a fixed offshore platform against collapse in
Persian Gulf area. With the base shear capacity and demand, seismic reliability analysis is performed
using first order reliability method (FORM) and seismic fragility curves are developed. The analysis
considers uncertainties in mechanical and hydrodynamic properties of members and foundation properties
and loading from topside with different lateral load patterns (linear, quad and first mode) for pushover
analysis. The capacity is evaluated through pushover analysis of the FEM at each iteration of FORM.
Importance analysis of the design parameters is also assessed. The results of reliability analysis indicate
that the seismic loading is not critical for the prototype structure as the lateral loads of wave and storm
dominate the design of the structural members when PGA (seismic intensity measures) is less than 0.5g
(that corresponds to much stronger earthquakes than rare intense earthquake in Persian Gulf, PGA=0.1g,
where the platform is located). Moreover, different load patterns of seismic pushover do not affect the
seismic fragilities in low probability of failures but do so noticeably when the probability of failure
becomes higher. Also, the first mode load pattern leads to more conservative reliability results and is
recommended to be used for the seismic pushover analysis of offshore platforms. Finally, importance
analysis of the design parameters in the reliability analysis indicate that steel yield stress is the most
important parameter that dominates the variability contribution to the limit state.
3.5
3.5.1
Rules and regulations

Harmonized common structural rules
(1) Introduction and history

The Harmonized Common Structural Rules (CSR-H) (IACS, 2014a) came to replace and combine the
Common Structural Rules (CSR) for Double Hull Oil Tankers (IACS, 2006b) and Bulk Carriers (IACS,
2006a) that were adopted by IACS Council in December 2005 and came into force in April 2006 in one set
of rules. The two sets of Rules were developed independently, and in an effort to remove variations and
295
achieve consistency, IACS made a commitment to Industry to harmonize the CSR. The harmonization
project for the two sets of Rules began in 2008.
Table 3. Damage index for collision condition (Nam et al., 2014).
Table 4. Damage index for grounding condition (Nam et al., 2014).
The first CSR-H made available on 1st July 2012, they were issued in draft form for public review and
comment during the first industry period from 1st July to 31st December 2012. This initial industry review
period was followed by a second industry review period in the spring of 2013 and a review by IACS
Societies Technical Committees in the autumn of 2013. The rules were then adopted on 18 December 2013
and issued on 1 January 2014.
These Rules enter into force on 1st July 2015 and supersede the Common Structural Rules for double hull
oil tankers, July 2012 and the Common Structural Rules for bulk carriers, July 2012.
CSR-H consists of two parts. Part one provides requirements common to both double hull oil tankers and
bulk carriers and Part Two provides additional requirements applied to either double hull oil tankers or bulk
carriers. CSR-H apply to the hull structures of single side skin bulk carriers having length L of 90 m or
above and to hull structures of double hull oil tankers having length of 150m or above.
CSR-H covers the buckling and ultimate strength of structural members such as stiffened and unstiffened
panels, corrugated bulkheads, struts, pillars, and brackets under part 1 Chapter 8. And also the CSR-H
covers hull girder ultimate strength and hull girder residual strength in a damaged condition for ships with
length greater than 150 m in under part 1 Chapter 5.
(2) CSR-H vs GBS

The IMO Global Based Standards (GBS) comes into force in 2016. Since the CSR was developed before
the release of the GBS, it was expected that some of the requirements would not be covered while the
CSR-H should be set out to achieve full compliance.
The original CSR complied with some of the GBS requirements when it came to considering the use of
the wave scatter diagram of the North Atlantic and the design life of 25 years. This has consequences on
several ship parameters; the long term period to be considered for the wave loads; the corrosion model to be
considered for this design life; and finally, the influence on the fatigue damage calculations. The areas of
296
noncompliance were defined by IACS through a gap analysis and were addressed in the CSR-H. These
areas included residual strength, structural redundancy and the wave induced hull girder vibration due to
whipping and springing for oil tankers.
For residual strength, IACS introduced in CSR-H the evaluation of the residual strength of the hull girder
of ships with length higher than 150m after collision or grounding as an effort to comply with the GBS
requirement of providing a reasonable residual strength after damage. Also the ultimate strength of the hull
girder is checked within the cargo area and in the machinery space for both collision and grounding
scenarios. This is done by separately considering that the assumed damage zone is removed from the hull
properties.
Structural redundancy was already applied to the side frames of single side skin bulk carriers. This
approach was implemented to avoid the single failure of a side frame leading to an overloading of its side
frame neighbours which could cause collapse and lead to the progressive collapse of adjacent frames. The
result of the IACS approach was the scantling reinforcement of the side frames for yielding and for
buckling. Subsequently IMO requested the same approach be extended to the areas of the cargo hold
structure subject to impact by grabs. Based on in-service experience of IACS members, a stiffened panel
typically does not collapse under a localized damage such as that which is defined. It is no longer a question
of grab impact on the structure generating deformation or cracks. The considered damage is relative to
normal sea-going operating conditions. Therefore in the IACS GBS submission IACS elected to
demonstrate that stiffened panels of different sizes made of elements with different scantlings but which are
of standard design in the shipping industry, have inherent structural redundancy to withstand localized
damages.
(3) Buckling and Ultimate strength Changes from CSR
CSR-H presented some modification to the CSR hull girder ultimate strength and buckling requirements.
Hull girder ultimate strength
For hull girder ultimate strength, two major modifications were introduced. The first one was about the
methods used for the appraisal of the ultimate bending moment and the second was to introduce a
coefficient to take into account the local bending of the double bottom.
In the CSR-OT the hull girder ultimate bending capacity is calculated by the single step method or by the
incremental-iterative method. The CSR-BC uses only the incremental-iterative method. As the single step
method was used only for oil tankers, IACS decided in the harmonization process to remove the single
step method and to keep only the incremental method.
For double bottom structure, where there is a large difference between external pressure acting on the
bottom plating and internal cargo pressure loading the inner bottom plating, e.g. alternate loading condition,
significant deformations and biaxial compressive stresses appear in the bottom plating in way of the midtank area and shear stresses load the ends of longitudinal girders in addition to hull girder stresses. These
additional biaxial and shear stresses reduce the overall hull girder ultimate capacity. To account for this
effect, a double bottom factor, db, was introduced to consider the decrease of hull girder ultimate capacity
by the above mentioned stresses corresponding to double bottom deformations. This effect obviously
appears when the tanker is in the hogging condition. Consequently, the db coefficient was fixed at 1.0 in the
sagging and at 1.1 in the hogging condition except for empty cargo holds in alternate loading condition of
BC-A bulk carrier. For this case, the db coefficient was fixed at 1.25.
Direct Strength Analysis
In CSR, the direct strength analysis through finite element analysis is required only for amidships structure.
After some concerns expressed about the validity of the approach and the strength of areas outside
amidships, IACS decided to increase the scope of the direct analysis in the CSR-H to include the entire
cargo hold region as well as the transition area forward and aft of the cargo region. As a result, loading
conditions outside amidships and for the foremost and aft most cargo holds were developed and the
boundary conditions, especially for the ends of the cargo area, were reconsidered and adjusted for this
purpose.
Also, the details to be checked by the screening method in the midship area for CSR-OT are now to be
checked in the full cargo area in CSR-H.
Buckling
The buckling control in CSR-H consists of three modules. The first module covers the slenderness
requirements of plates, longitudinal and transverse stiffeners, primary supporting members and end
brackets. The second module covers the prescriptive buckling requirements for members subjected to hull
297
girder stresses as plates and longitudinal stiffeners. The third module covers the buckling requirements for
FE analysis of plates, stiffened panels and other structures such as cross ties.
The slenderness criteria are based on CSR-OT and consist of maximum permissible slenderness
expressed as minimum thickness requirements, e.g. minimum thickness of plates, stiffener web or stiffener
flange. For plates which will be assessed for buckling based on actual stress level, this requirement is a high
slenderness requirement to avoid very thin plates resulting in very flexible structures with low stiffness. This
is visualized as Area C in Figure 7. For flanges of stiffeners and primary supporting members a low
slenderness requirement is applied so the flange is stocky and able to carry a stress very close to full yield
stress to avoid local buckling of the free edge of flange/face plate. This is visualized as Area A in Figure 7.
Area A
Area B
Area C
Buckling Stress / Yield Stress
1.0
Stock
members
Controlled by
specification of
minimum
proportions
and stiffnesses
Typical
stiffened
panels
Controlled by the application of
Either:
- Plate buckling
- Stiffener buckling
Slender
members
Controlled by
specification of
minimum
proportions
and stiffnesses
Slenderness
Figure 7 Concept of determination of scantling of stiffened panel in CSR-H.
The buckling capacity of plates and stiffeners is unified in a single toolbox applicable both for the
prescriptive buckling check and the FE buckling assessment. This toolbox, referred to as the Closed Form
Method (CFM), is a further development of the ultimate strength/buckling capacity method in CSR-BC and
the prescriptive method in CSR-OT. The ultimate capacity of plate and stiffeners are defined as the ultimate
limit state when the membrane Von Mises stress reaches specified minimum yield stress either in the plate
or at the top of the stiffener. The main updates of the Closed Form Method are:
Elastic buckling limit for stiffeners is included, as in the FE buckling method of the CSR-OT.
Torsional buckling of stiffeners has been replaced with a warping stress component in the lateral
buckling formulations for stiffeners.
For short wave plate buckling pattern (longitudinal plate buckling), the rotational support along the
long plate edge from the stiffener is taken into account.
Prescriptive Buckling Requirements

Buckling requirement for plates and stiffeners subjected to hull girder stresses are extended from the uniaxial buckling assessment in CSR-OT, where hull girder bending and hull girder shear are checked
separately. In CSR-H, the prescriptive buckling assessment is based on a combination of hull girder
bending, hull girder shear and local pressure similar to CSR-BC. Prescriptive buckling assessment shall
be carried out both for acceptance criteria AC-S (frequent loads, static loads) and AC-SD (extreme loads
in seagoing conditions, static + dynamic loads). This is also an increase in scope compared with CSR-OT
which has only prescriptive buckling requirement for acceptance criteria AC-SD.
FE buckling
For FE buckling the scope is the same as in CSR-OT, however the method has been changed from the
semi-analytical advanced buckling assessment method (PULS) to the Closed Form Method (CFM). The
elastic buckling limit for the webs of primary supporting structures in CSR-OT has been replaced with the
ultimate capacity based on plate panel without pull-in constraint at the plate edges, to avoid some of the
conservatism in the elastic buckling
(4) Consequence assessment
A consequence assessment (CA) was carried out by IACS in order to determine the effect of applying
CSR-H and presented in separate IACS individual Technical Background reports.
298
In the CA evaluation IACS Societies have used current and representative designs from major builders
in Asia. The designs assessed in this CA are compliant with the July 2010 Common Structural Rules.
For the consequence assessment the design has not been altered in any way. Strake size, seam
locations, material properties, stiffener spacing etc. have not been altered. The consequence
assessment is carried out using available software from each Classification Society and results were
cross checked.
The consequence assessment was conducted on VLCC, Suezmax, Aframax, Panamax and Handymax
oil tankers and also on Capsize, Babycape, Panamax and Handymax bulk carriers.
For most of the assessed tankers, the consequence assessment showed that CSR-H requirement for
buckling and ultimate strength generates a scantling increase for most of the ship structure.
For mid cargo tanks, the keel, sheer strake plating and non-watertight girders in double bottom and nonwatertight stringers in the double hull will see scantling increase due to minimum thickness increase. The
main deck, longitudinal bulkhead and inner hull longitudinal stiffeners will see increase in section modulus
due to more conservative buckling requirement. The FE buckling assessment also generated a scantling
increase in areas located at the hopper structure, inner hull upper part, longitudinal bulkhead upper part,
horizontal stringers in double hull and double bottom floors.
The keel in the fore-most and aft-most cargo tanks, sheer strake plating, non-watertight girders in double
bottoms, and non-watertight stringers in double hulls will see scantling increases for most of the tankers
assessed due to minimum thickness increase.
The fore part, aft part and machinery space will see scantling increase due to more conservative
requirements for plating and stiffeners.
FE buckling assessment in CSR-H shows that the offered scantlings for transverse bulkheads and
transverse web frames are not sufficient in some locations and will require reinforcement.
Also for most of the assessed bulk carriers, the consequence assessment showed that the CSR-H
requirement for buckling and ultimate strength generates a scantling increase for most of the ship structure.
For the mid cargo holds, the Rule requirements for longitudinal stiffeners are more conservative in
CSR-H (covering local pressure, prescriptive buckling) in comparison to CSR. Also some singe side shell
plating does not meet the prescriptive or FE buckling requirements of CSR-H. This leads to scantling
increase of various degrees.
The fore most and the aft most cargo hold will see increase in stiffener section modules due to more
conservative stiffener requirements.
The aft part and the machinery space will see increase in side shell thickness due to minimum thickness
requirement.
(5) Technical and critical opinions for further work
In the assessment of hull girder strength, the db coefficient was introduced in CSR-H. Based on the
CSR-H technical background (IACS, 2014b), the value of db was determined from the nonlinear FE
results. But the details of FEA, such as description of the solver used, model range, boundary conditions,
mesh size, and imperfection amplitude etc , were not shown. There are several papers that addressed
the effect of lateral loading for hull girder ultimate strength. The calculated results are different each
other, but the majority of these results show good agreement with CSR-H. But, further study is needed.
The scantlings according to CSR-H are somewhat more conservative than CSR. This is mainly due to
using the conservative side of both CSR rules when a theoretical back ground is not clear. Further study is
recommended in order to clarify the theoretical background, and if necessary, the rules should be
modified accordingly.
Some of technical background documents are not sufficient. It is recommended that the unknown and
unapparent items are clearly stated.
3.5.2
Updates to offshore rules and guides
No major changes have been made to the rules for buckling and ultimate strength for offshore structure.
Below are the few minor changes that have been implemented to the ABS (2014), LR (Lloyds, 2014)
and DNV (DNV-OS-J101, 2014) documents.
The ABS guide for Buckling and Ultimate Strength Assessment for Offshore Structures introduced an
additional check to its buckling control concept. In addition to checking the plate panels and the stiffened
panels for bucking and ultimate strength, it was introduced a check to the overall strength of the entire
stiffened panel to satisfy the required biaxial compression as satisfied in section 3-5.7 of the guide. The
proposed change addressed the buckling state limit check for large panels. The original requirement is from
ABS Rules for Building and Classing Steel Vessels.
299
The Lloyds Register Rules and Regulations for the Classification of Offshore Units, July 2014 have been
updated to implement similar buckling methods as in the CSR-H. For DNV, few updates related to buckling
have been implemented to DNV document DNV-OS-J101 Design of Offshore Wind Turbine Structures,
May 2014.
In Section 7 Design of Steel Structures [7.3.1.4], the requirement for using a material factor of
minimum of 1.2 for global buckling checks refers to IEC 61400-1 and is therefore changed to apply to
towers. An extension to also cover monopiles is natural and has been included. An exception is made that a
material factor of 1.1 can be used when formulas given in EN1993-1-6 are used for design.
In Sec.10 Foundation design, a new item [10.3.2.7] regarding permanent buckling and plastic
hinges in monopile foundations has been added.
4.
4.1
4.1.1
ULTIMATE STRENGTH OF VARIOUS STRUCTURES

Tubular members and joints
Tubular Members
Recent developments on the strength of tubular members focus on the various strengthening and
reinforcing schemes for tubular members subjected to different loading conditions. Many researchers
have examined two broad types of reinforcement on the strength of tubular members, i.e., the
reinforcement using the fiber reinforced polymers and the steel-concrete composite structures. In
addition, elliptical hollow sections, the design guides of which are still being developed, have emerged as
an increasingly popular structural solution due to its aesthetic properties. As the accidental loading poses
critical threats to the safety of such tubular members, recent research efforts have examined the resistance
of tubular members subjected to lateral impacts.
Reinforced Tubular Members
Concrete-filled steel tubes (CFST) have emerged recently as a popular structural member owing to
their significantly enhanced structural resistance with reasonably low cost. Goode et al. (2010) have
examined the buckling of slender composite concrete-filled steel columns through an experimental
program. The theoretical approach outlined in the Eurocode 4 presents a close estimation on the load
bearing capacity of the concrete-filled, circular and rectangular hollow section stubs and long columns.
Preloading on the concrete-filled hollow sections does not influence the load bearing capacities of the
columns.
Lu et al. (2010) reported an experimental investigation on six full scale double skin tubular columns filled
with self-consolidating concrete under fire tests. The double skin concrete-filled steel tubular columns
exhibit superior fire performance compared to the hollow tubular columns and concrete-filled tubular
columns. Kim et al. (2010) presented an experimental and numerical study on the load characteristics for
steel and concrete tubular members under jet fire in floating production storage and offloading (FPSO)
topsides, which can be used for the design of the passive fire protection. Uenaka et al. (2011) presented an
experimental study on the mechanical behaviour of concrete-filled double skin tubular circular deep beams
under three-point bending. Their study showed that the diameter ratio of the inner steel tube over the outer
steel tube imposes a strong effect on the observed failure modes of the specimens. Han et al. (2011) (2012)
reported some experimental investigations on the curved concrete filled steel tubular members under axial
compression, with the focus on the following parameters: the tube shapes, the initial deflection and the
slenderness ratio. The curved concrete filled steel tubes demonstrated a higher axial capacity than the
straight concrete-filled steel tubes. Both types of members are designed with the same amount of
eccentricity at the mid-height of the member. The eccentricity for the curved member, therefore, increases
from zero at the support to the maximum value at the mid-height, while the straight members have a
constant eccentricity along the length of the member. Dai et al. (2012) have proved the superior fire
resistance of concrete-filled circular hollow sections in comparison to elliptical, rectangular and square
hollow sections filled with concrete. Hou et al. (2013) developed the design equations to estimate the local
bearing resistance of the CFST members based on a comprehensive experimental study.
Sundarraja and Prabhu (2012) have shown, through an experimental study of thirty specimens, that the
external bonding of the carbon-fiber reinforced polymer (CFRP) on the CFST provides effective
confinement pressure and delays the local buckling of the steel tuber under axial compression. The same
authors have also demonstrated the significant enhancement on the flexural resistance of CFST members
strengthened by CFRP composites through experimental studies. Li et al. (2013b) presented a calibrated
numerical investigation on the high strength concrete filled square steel tubular columns with internal CFRP
circular tube under bi-axial eccentric loading. Their study covers the effect of the material parameters for
steel, concrete, CFRP, and the geometric slenderness ratios.
300
Elliptical Hollow Section
Elliptical hollow section (EHS) members have recently become a relatively new form of structural
elements. Chan et al. (2010) presented a state-of-the-art review on recent researches on the elliptical
hollow section members, including the elastic local-buckling and post-buckling, cross-section
classification, response in shear, member instabilities, connections and behaviour of concrete-filled EHSs.
Lam et al. (2010) have investigated experimentally the behaviour of axially loaded concrete-filled
stainless steel elliptical hollow sections, with the objective to develop a set of new equations to predict
axial capacity of the composite concrete-filled stainless steel elliptical hollow sections. Espinos et al.
(2011) have reported a numerical investigation on the fire behaviour of concrete-filled elliptical steel
columns. The fire resistance of the concrete-filled EHS depends on the slenderness of the column and the
load level. For a certain column length and load bearing capacity, the circular columns attain higher fire
resistance than the elliptical columns. Sheehan et al. (2012) examined the structural response of concretefilled elliptical steel hollow sections under eccentric compression. Uenaka (2014) reported an
experimental study on the capacity of concrete filled elliptical columns under compression and developed
a method to predict the axial loading capacity to account for the confinement effects on the in-filled
concrete. Dai et al. (2014) have subsequently confirmed from an extensive numerical investigation that
the design rules in Eurocode 4 for circular and rectangular CFST columns can be used to calculate the
axial buckling load of the elliptical CFST columns.
Impact Resistance
Accidental impact loads create a critical threat for the safety of the tubular members. Khedmati et al.
(2012) have demonstrated, through a numerical investigation, that the impact resistance of tubular
members decreases with increasing amounts of preloading. Their numerical study also confirms the
significant effect imposed by different boundary conditions on the ends of the tubular members on the
impact resistance. Wang et al. (2013) have investigated the impact response of concrete filled steel
tubular members with various levels of confinement effect. The confinement ratio imposes a strong effect
on the impact resistance of the CFST members. Han et al. (2014) developed a simplified model to
estimate flexural capacity of CFST members under impact loads, based on an experimental program and a
comprehensive numerical study. Wang et al. (2014) presented a comprehensive experimental and
numerical investigation on the impact of circular hollow members and double-skin circular specimens
(see Figure 8) filled with ultra-lightweight cement composite. They have demonstrated that the impact
process for hollow pipes and double-skin pipes consists of three phases, namely the vibration phase, the
stable phase and the unloading phase. The presence of the cement composite material in the annulus
between the inner and outer steel pipes restricts the local indentation of the outer steel pipe and enhances
significantly the impact resistance of the pipe structures.
Outer steel pipe

Cement composite
Inner steel pipe
Figure 8. Configuration of the double-skin circular sections (Wang et al., 2014).
4.1.2
Tubular Joints
Similar to the latest development in tubular members, various reinforcement schemes have emerged
recently for tubular joint due to the increasing demand by the industry to extend the operating life for
existing offshore platforms. Due to the increasing number of aging facilities in both onshore and offshore
infrastructure, the integrity assessment and residual strength examinations become increasingly important
to ensure the safe operation of these structures in their remaining life. Recent research efforts have also
covered the strength of complex tubular joints, e.g., the elliptical joints, multi-planar joints, for which the
existing design codes do not provide explicit strength equations. In addition, many researchers have
examined the strength of tubular joints under accidental loading, e.g., under fire and impact, to develop
the basis for engineering guidelines in such limit states.
Reinforced Tubular Joints
Lohning et al. (2013) presented a finite-element based design of grouted joints with shear keys for
offshore wind turbine. Their finite element analysis considers the complex interface between the grout
301
and steel materials, nonlinear material properties and geometric responses. Lesani et al. (2013b) have
reported a numerical investigation on the fiber-reinforced polymer (FRP) strengthened T-joints under
axial compression. The presence of the FRP delays the common failure modes, e.g., local bending of the
chord member, punching shear, plastic failure of the chord, in otherwise un-stiffened joints.
Residual Strength and Assessment
In addition, residual strength of the fatigue cracked tubular joints has become an increasingly important
concern for practicing engineers in assessing the aging structures. Qian (2013) have reported ductile
tearing assessment procedure for high-strength tubular X-joints subjected to in-plane bending. Their study
shows that the generic failure assessment curves do not provide a conservative assessment on the circular
hollow section X-joints, compared to the geometric and material dependent failure assessment curve
based on the option 3 failure assessment curve in BS7910 (2013). Qian et al. (2013a) has subsequently
proposed a set of specific failure assessment curves for circular hollow section X and K-joints under
brace axial loads. Meanwhile, Lie et al. (2014) have also presented a failure assessment diagram based
approach to predict the plastic collapse load for cracked circular hollow section T and Y joints. Qian et al.
(2013c) have further integrated the fracture representation in the load-deformation responses of tubular
joints in the global frame analysis. Qian et al. (2013b) (2014) reported a new lamellar splitting failure for
fatigue cracked tubular X-joints subjected to brace in-plane bending. The brittle splitting occurs at the
mid-thickness of the chord wall and creates a fracture surface almost perpendicular to the fatigue crack
surface.
Complex Tubular Joints
Zhao et al. (2011) have highlighted the recent developments in hollow section tubular joints, including the
elliptical hollow section joints, the thick-walled joints, etc. He et al. (2013) compared the static strength
of tubular Y-joints with and without partially thickened chord walls. They have proposed an equation to
estimate the strength of the Y-joints based on a numerical parametric study. Chiew et al. (2012) presented
an experimental and numerical study to examine the static strength of a complex tubular DKYY-joint.
Internal reinforced circular plates in the chord member become necessary to avoid excessive plastic
deformations in the chord. Lesani et al. (2013a) reported a detailed investigation on the un-stiffened T/Y
tubular joints under axial compression. They have examined the critical stress fields near the brace-tochord intersection of the T/Y joints and proposed subsequently a critical zone in the chord member near
the crown point.
Strength under Accidental Loading
Recent research has examined tubular joints under accidental loads, e.g., the impact loads, the blast and
fire conditions. Yu et al. (2011) reported an experimental study on the mechanical behaviour of tubular Tjoint under fire loads. The T-joint specimen experienced an impact load before fire loadings were applied.
Local buckling appeared to be the primary failure mechanism for T-joints under elevated temperatures.
Jin et al. (2012) have presented an experimental and numerical study on the post-fire behaviour of tubular
T-joints. The axial load level and the heating and cooling history impose significant effects on the
compressive stiffness of the T-joints. Xu et al. (2012) have subsequently developed an artificial neural
network approach to predict the ultimate bearing capacity of tubular T-joints under fire. The input
parameter required in their neural network approach includes the diameter ratio between the brace and the
chord, the chord radius to thickness ratio, the wall thickness ratio between the brace and the chord, and
the temperature. The neural network prediction shows reasonable agreement with the finite element
results. Tan et al. (2013) have also reported an experimental and numerical study on tubular T-joints
subjected to axial compression in fire conditions. The increase in temperature leads to the deterioration of
the material strength and therefore causes a more severe plastic deformation near the brace-to-chord
intersection under elevated temperatures. He et al. (2013) reported the experimental investigation on
circular hollow section K-joints subjected to brace axial loading under elevated temperatures. Local
plastic yielding near the brace-to-chord intersection appears to be the primary failure mechanism for the
K-joints studied. Qu et al. (2014) presented an experimental study on circular hollow section T-joints
subjected to a drop hammer impact loading with the initial velocity between 7 to 10 m/s. They have
proposed an approach to estimate the impact force based on the yield line theory.
4.2
4.2.1
Steel plate and stiffened plates

Introduction
Research efforts to further improve our understanding of the ultimate strength behaviour of stiffened
panels is essential for the further development of methods for local and global strength assessment of
ships and offshore structures. This is especially important when considering the continued adoption of
302
limit state practices for structural design. Stiffened plate panels form the fundamental building blocks of
ships and offshore structures. A stiffened panel usually fulfils multiple functions within a structural
arrangement, which may include:
forming an external or internal barrier to water or other liquid cargo;

supporting cargo or equipment loads;
resisting local loads such as hydrostatic pressure, sloshing or slamming;
contributing to the global strength of the platform under different loading conditions. A typical
example in a ship is a longitudinally stiffened panel which forms part of the continuous longitudinal
structure and resists longitudinal bending loads on the hull girder.
Proper treatment of all these functions, either in isolation or combination, is vital to a successful
structural design of a stiffened panel.
Stiffened plating is usually arranged to form fully continuous or semi continuous structure running
through the span of a thin plated structure. For a ship or ship shaped offshore platform, the continuous
structure contributes to the resistance of global wave-induced bending loads of the main hull girder.
Global bending of a hull girder imparts in plane axial loads on what are often relatively slender stiffened
panels. They must therefore be designed with regard to buckling and elasto-plastic collapse when the
dominant axial load acts in compression. This area of study continues to be important for improving code
based design formulas and direct strength analysis procedures.
4.2.2
Analytical Formulations for Ultimate Strength of Stiffened Panels
Analytical methods continue to be developed for predicting the critical buckling load or ultimate capacity of
a stiffened panel under in-plane compression. Whilst simple formulas may only predict an ultimate strength
value, more advanced methods attempt to describe the nonlinear load-shortening behaviour of the panel.
This may be useful for the direct strength assessment of the panel itself and is also essential information for
application in ultimate strength analyses of the hull girder using the Smith progressive collapse method
(known as the incremental-iterative method in IACS CSR) or Idealized Structural Unit Method (ISUM).
Further development of methods for predicting plate and panel load-shortening behaviour under different
loading conditions contribute to the enhancement of progressive collapse methods.
ISUM is an extendable approach where formulations for new types of elements (units) further improve its
capability for different structural configurations. An extension to ISUM formulations for modelling the
ultimate strength behaviour of very slender square plates and long plates under longitudinal axial load is
proposed by Lindemann et al. (2013) using higher-order components to describe the collapse mode of such
plates. Furthermore, a condensation procedure to eliminate the amplitudes of the panel deflection from the
degrees of freedom for the plate is introduced so that ISUM elements can be more easily implemented in
conventional finite element analysis frameworks, which handle only nodal degrees of freedom. The
procedure is validated by comparison with the original ISUM plate formulation and with nonlinear FEA.
Lindemann et al. (2013) investigate the influence of different components of the tangential stiffness matrix of
a combined ISUM plate / beam-column element model as originally proposed by Fujikubo et al. (2002). It is
highlighted that for the ISUM plate element the components derived by elastic large deflection analysis are
dominant while for the beam-column element considering the rotational stiffness of the plating is essential.
The Smith method is well-established for predicting the interframe collapse of a hull girder under
longitudinal bending, and is codified in IACS CSR for assessing the hull girder ultimate strength. An
extension to the method, requiring an adapted large deflection orthotropic plate method with the capability
to predict gross buckling of stiffened panels over multiple frame spaces, is proposed by Benson et al.
(2013d). The adapted orthotropic plate method is able to produce the complete load shortening curve for a
stiffened panel under uniaxial compression and accounts for the possibility of interframe and overall
buckling modes. The method is used in an extended Smith-type progressive collapse method which can
predict compartment level collapse modes of a lightweight hull girder.
4.2.3
Uniaxial Compression
IACS Common Structural Rules for Bulk Carriers (IACS, 2010a) and Tankers (IACS, 2010b), which are
soon to be superseded by the recently adopted harmonized rules (IACS, 2014a), provide design
formulations to determine the uniaxial buckling of the plating between stiffeners, column buckling of
stiffeners with attached plating and lateral-torsional buckling or tripping of stiffeners. CSR formulae are also
provided to determine the load-shortening curve for a stiffened panel for use in the incremental-iterative
progressive collapse method. These code formulations form the basis of several recent validation studies,
with differing levels of success in correlating between results.
303
Tanaka et al. (2014) examined the accuracy of the ultimate strength formulas in CSR for bulk carriers as
well as PULS, which is a nonlinear orthotropic plate method by DNV. Nonlinear FE analyses of stiffened
panels under uniaxial load are completed and show that both CSR and PULS give a good estimation of
ultimate strength. The CSR formulae are also compared to equivalent nonlinear finite element analyses by
Nam et al. (2013). This study showed some differences between methods and suggest some limitations in
the CSR formulae, particularly for flat bar and angle bar stiffened panels. However, only three panels are
investigated in total and so these findings cannot be taken as the general case.
Gaspar et al. (2011) applies structural reliability methods to assess the implicit safety levels of the
buckling strength requirements for longitudinal stiffened panels implemented in the IACS Common
Structural Rules (CSR) for double hull oil tankers. The study then uses this information to apply FORM
optimization of the scantlings for both plates and longitudinal stiffeners in order to reflect the minimum
strength required by the IACS formulations. The effect of corrosion is also considered for a sample of five
typical oil tankers.
The ultimate strength of an arbitrary dimensioned stiffened panel can be related directly to the plate and
column slenderness ratios. For example, the column collapse design curves published by Chalmers (1993)
present a simple and direct way to determine the ultimate strength of panels stiffened by UK Admiralty long
stalk tee bars. The curves are a function of plate slenderness, column slenderness and the ratio of stiffener
cross section area to the total panel cross section area. However, as shown by Yasuoka et al. (2012), the
ultimate strength can differ depending on the absolute size and number of stiffeners attached to the panel.
When stiffeners are small, and attached to thin panels, overall buckling dominates and the ultimate strength
differs depending on the number of stiffeners. When stiffeners are large and collapse is localized the
ultimate strength does not vary so much regardless of the number of stiffeners.
Choung et al. (2012b) evaluated the distribution of slenderness ratios for plates and stiffened panels using
data from 163 vessels including tankers, bulk carriers and container ships. The load shortening capacities of
the stiffened panels derived from the CSR formulae are compared to equivalent nonlinear finite element
analyses. Whilst ultimate strengths show reasonable correlation, the post ultimate behaviour shows
considerable differences. A new relative average strain energy parameter is introduced to better account for
post ultimate strength when using CSR formulations.
Several other papers present comparative studies of plates and stiffened panels with various geometries in
uniaxial compression. Linear eigenvalue and nonlinear buckling analyses of 27 stiffened panels are
presented by Ozdemir et al. (2013). El-Hanafi et al. (2013) present in-plane ultimate strength calculations
for Y stiffened panels designed for a double hull tanker. The ultimate strength of cylindrical curved panels is
investigated analytically, following the methodology outlined in Eurocode 3, and with nonlinear finite
element analysis by Tran et al. (2012). The behaviour of cylindrical curved panels is shown to depend on
curvature, slenderness and imperfections.
4.2.4
Multiple Load Effects
Secondary load components such as lateral pressure, biaxial load and shear may have a significant
influence on the longitudinal capacity of a stiffened panel. Several papers use nonlinear FE analysis
(NLFEA) to account for different combined loads.
Ergin et al. (2013) evaluate the effect of lateral pressure on the load shortening behaviour of 49 stiffened
panels. Example results for four panels are given together with non-dimensional ultimate strength values for
all 49 panels tested. The influence of lateral pressure on the average compressive strength of 189 panels
including flat bar, angle bar and tee bar stiffener geometries are compared to CSR formulations by Choung
et al. (2014). A new concept of the relative average compressive strain energy, instead of the ultimate
strength, is introduced in order to rationally compare the average compressive strength through a complete
compressive straining history. Similar to the authors earlier study (Choung et al., 2012b), whilst the ultimate
strengths predicted by CSR and NLFEA compare well, there are larger discrepancies found when using the
strain energy concept. Zhang et al. (2013) compare nonlinear FEM and CSR to predict the ultimate strength
of stiffened panels from an AFRAMAX tanker under combined compression and lateral pressure. This
study finds that CSR can over-estimate the panel ultimate strength. Xu et al. (2013b) examine the effect of
different boundary conditions and model extents for NLFEA of stiffened panels under biaxial in-plane load
and lateral pressure. The results show that a panel spanning two bays in the longitudinal direction
adequately represents a continuous stiffened panel.
The influence of shear on the compressive capacity of a plate or stiffened panel is studied by Fujikubo
et al. (2013). The modelling approach, using multi point constraints, is described in detail and an example
stiffened panel is analyzed. Shear of thin steel plates at non-uniform elevated temperatures is studied by
304
Salminen et al. (2014). A comparison of the buckling of stocky and slender plates under shear and bending
is completed by Alinia et al. (2012) using nonlinear finite element analysis and a theoretical p-Ritz energy
method. This shows inelastic buckling in the stocky plates at the ultimate strength. The interaction
relationship between shear and compression for a series of square plates are derived using nonlinear finite
element analysis by Syrigou et al. (2014). Two different in plane restraint boundary conditions on the edges
of the plate are investigated.
4.2.5
Panels with openings, cut-outs or rupture damage
Stiffened panels may be perforated deliberately by openings for running pipework, access hatches and
other design reasons. A panel may also be ruptured accidentally due to a gross damage event, which can be
approximated by removing structure. In both cases, the opening may be represented by a simple cut-out
from the otherwise intact panel. The reduction in the ultimate strength of a panel in this condition is the
subject of several papers.
The ultimate strength of perforated plates typically placed in ship structure for surveyor inspection, pipe
runs and weight reduction are investigated by Park et al. (P2013) using nonlinear FE analysis. Longitudinal
compression, transverse compression and in-plane shear were tested. Regression analysis was employed on
the results to develop empirical ultimate strength design formulations. The ultimate strength of stiffened
panels with rectangular openings are presented by Lee (2012), Yu et al. (2012) and Chang et al. (2014)
using nonlinear FE analysis approaches. Lee presents an empirical formula for estimating the reduction in
ultimate strength due to the opening shows good agreement with the numerical results. Elastic buckling of
skew plates loaded uniaxially and containing openings are presented by Tahmasebi et al. (2011). The
ultimate strength of a series of stiffened panels with cut outs to represent rupture damage are analyzed
parametrically by Benson et al. (2013e). The ultimate strength is shown to reduce as the cut out size is
increased and the results are presented in the form of column collapse curves for use in practical structural
design. The influence of the lengthwise and spanwise sizing of the cut-out is investigated. It is shown that
the strength is strongly influenced by the span of the opening, and is sometimes marginally influenced by
the opening length. Length is particularly important if the cut out removes.
Bayatfar et al. (2014) presented the results of a numerical study investigating the influence of cracks
(in terms of length and location) on the ultimate compressive strength characteristics of imperfect
unstiffened and stiffened plate elements, used in thin-walled structures such as ship hull girders. The
cracks were presumed to be through-thickness, having no contact between their faces and no propagation
was allowed. Flat-bar profile is the considered stiffener type in this study and materials are in the category
of high strength steel alloys. This study indicates that the length of cracks and especially its location can
significantly affect the ultimate strength characteristics of unstiffened and stiffened plate elements
subjected to axial compressive action.
4.2.6
Welding effects
The welding process has a significant influence on the fabrication factors associated with distortion and
residual stress of stiffened panels and thus on the ultimate strength. The welding process will create
residual deformations (distortions) in the panel, in particular out-of-plane deformations. These
imperfections are today included in numerical simulations of the ultimate strength, and often modeled as
a double sinusoidal variation in the plane of the panel.
To find the influence of the welding residual stress field, the shape of the residual stress field can often be
approximated (idealized) for a fillet welded stiffener to a panel if a single pass weld is performed. In the
longitudinal direction, a high tensile value in the weld region (often close to the yield stress value) balanced
by a lower compressive value away from the weld region.
Khan et al. (2011) studied axial compression of longitudinally stiffened plates, using such an idealized
longitudinal residual stress field and with the weld imperfections modelled with the double sinusoidal field.
It is seen that the decrease in the ultimate strength depends on the magnitude of the residual stress and the
slenderness of the plates, i.e. moderate thick plates are the most sensitive. Paik et al. (2011) studied a similar
panel geometry and found that the effect of residual stresses is larger for thicker plates. They also modelled
transverse residual stresses but found that they are less important prior to buckling for the axial load
situation.
Khedmati et al. (2013) studied curved stiffened panels in axial compression using the same approach, and
found that including the residual stress field will marginally reduce the ultimate strength.
Paik et al. (2012c) studied welding and subsequent buckling collapse of an aluminum stiffened plate.
Welding residual stresses (or rather strains) and deformations have been measured for a panel with three
bays. The shape of the welding residual stress field was found to be similar to the idealized stress field often
305
used. Experimentally determined buckling loads were compared to numerical results, and it is found that
even if agreement was good, it is essential to have a good model of the residual stress and deformation field
and the softening (yield stress) in the heat affected zone.
Tekgoz et al. (2013b) have studied the influence welding residual stresses in stiffened panels using both
the approach described above, with an idealized longitudinal residual stress field, and by modifying the
elastic perfectly plastic stress-strain curve used in the FE-simulations (as originally proposed by Faulkner).
They found this second approach to be fast and give good results.
Tekgoz et al. (2013a) studied a tee-stiffened plate, i.e. one stiffener fillet welded to a plane plate, and
show how the welding sequence for a stiffened plate influences the residual stresses and has a subsequent
effect on the ultimate strength of the plate. The welding residual deformations and residual stresses are
determined from thermo-mechanical FE analysis for three alternative weld sequences (in all cases the both
sides of the stiffener were welded simultaneously). It is shown that the welding sequence is the most
influential parameter which affects the lateral displacement of the stiffener, which then leads to more
ultimate load carrying capacity. The welding sequence also influences the displacement of the plate edges
which then governs the buckling shape of the plate. The welding residual stress field is used to determine a
modified stress-strain curve to be used for ultimate strength calculations.
Gannon et al. (2013) studied the influence of the welding residual stresses and deformation on the
strength of tee-stiffened panels under axial compression. Welding residual stresses and deformations were
obtained from a full thermo-mechanic FE analysis. Two weld passes were used for the stiffener, being
deposited sequentially with a cooling period in between. The simulations showed that the reduction in
ultimate strength varied between 5% to 18% for all models and decrease with increasing plate slenderness
ratio. In Gannon et al. (2012b) the effect of axial shakedown was studied for the same tee-stiffened panel.
Axial shakedown reduced the tensile welding stress peaks and the amplitudes of the welding deformations.
It was found that a partial relief of welding stresses would increase the ultimate strength somewhat, although
the amount of reduction depends on the failure mode as the residual stress reduction is not uniform. The
residual stresses are relaxed during the first cycle of a constant amplitude load history.
The influence of weld stiffness on the buckling strength of web-core steel sandwich plates is studied by
Jelovica et al. (2012). The study demonstrates that lower weld stiffness results in lower transverse shear
stiffness and also shifts mode intersections towards higher aspect ratios.
4.2.7
In service degradation
A parametric study of ship plates with non-uniform corrosion represented by a nonlinear random field
approach and then placed in axial compression are tested using nonlinear FE analysis by Teixeira et al.
(2013b). The results show that the complex corrosion process, which is spatially distributed using random
fields, produces an ultimate strength which is in most cases lower than the one obtained using a uniform
corrosion model.
The ultimate tensile strength of corroded plates, with the corrosion surface modelled using a random field
model, are tested using nonlinear FE analysis by Htun et al. (2013). The stochastic properties of the ultimate
tensile strengths of the corroded plate are then estimated using a Polynomial Chaos Expansion (PCE)
Method and also a Monte Carlo simulation. The study shows that PCE can be used to determine the
stochastic response with fewer samples. However, unlike the findings of Teixeira et al. (2013b) (which test
plates in compression rather than tension), the ultimate strength of the plates using random field corrosion
modelling are shown to have less strength reduction than when using a uniform corrosion model.
Jiang et al. (2012b) investigate the effects of corrosion pits on the ultimate capacity of mild steel
rectangular plates under uniaxial compression. The simulations show that the volume loss dominates the
degradation of the compressive capacity of pitted mild steel plates in addition to plate slenderness. An
empirical formula was proposed to predict the ultimate capacity of pitted plates under in plane compression.
The comparison between results from the FEM simulation and the formula showed a satisfactory fit.
A finite element procedure to determine the buckling behaviour of a cracked stiffened panel using a
mixed shell-solid mesh is proposed by Kimura et al. (2013). Shell element modelling is usually used in large
deflection elasto-plastic analysis of thin plated structures. However, shell elements are less suitable for
simulating the local behaviour of cracks or defects. Therefore a solid model is used in the local crack region
and attached to a shell model using rigid body elements.
4.2.8
Experimental Testing
The ultimate strength of a column segment typically found in square cross-section columns of new largesize semi-submersible platforms was carried out by Estefen et al. (2012). The modelling approach for
306
introducing imperfections was initially verified from small scale physical experiments on flat stiffened
panels which were repeated with nonlinear finite element analysis. A numerical model of the column
segment, including representative geometric imperfections, was then tested. The influence of imperfection
shape and amplitude is presented, comparing results to measured imperfections from a shipyard using a
laser tracker. The results show that initial distortions have a significant influence on the displacement to
failure but less influence on the axial load capacity.
4.2.9
Optimization
There is continued interest in the development and use of optimization methods to design plates and
stiffened panels for greater reliability, efficiency and economy. Such methods involve the optimization of
multiple geometric variables such as the material type, plate thickness, stiffener scantlings and the number
of stiffeners and frames for a given panel. Efficient optimization approaches employ analytical methods
which can quickly determine the strength of a panel and can therefore be employed in an optimization
algorithm.
Ma et al. (2013) apply two multi objective optimization methods, Pareto Simulated Annealing (PSA) and
Ulungu Multi-Objective Simulated Annealing (UMOSA), for a single panel optimization and then extend
the method to concurrently optimize a system of stiffened panels such as found in a ship structure.
Farkas et al. (2014) demonstrate the optimization of a square cellular stiffened panel (also known as a
sandwich plate) with an objective function of minimum fabrication cost, using an orthotropic plate method
to predict the strength and a systematic search algorithm to find the optima.
4.2.10 Conclusions
Research into the ultimate strength of stiffened panels for ship and offshore structures continues to
receive worthwhile attention. The parameters which affect the load shortening and strength of a stiffened
panel in compression are numerous and include the geometry, the stiffener type, the material properties,
the support conditions and the load components. Nonlinear FEM is used to conduct parametric studies to
determine the influence of certain parameters of interest. A particular focus over the last three years has
been on the ultimate strength formulations proposed in IACS CSR. Attempts at validating the formulas
with the results from equivalent nonlinear FEM are inconclusive, with some studies showing much better
correlation than others. This perhaps demonstrates the limitations of using a single rule formulation when
applied to a complex and multi-parametric problem.
Future work should continue to apply rigorous methods for setting up and analyzing nonlinear FEM of
stiffened panels with adequate consideration of the geometric and material imperfections which have a
significant effect on the resulting ultimate strength prediction. Further research into the effect of multiple
load combinations on stiffened panel ultimate strength is essential to enable a better treatment of complex
load patterns in progressive collapse type methodologies.
4.3
Shells
The effect of geometric imperfection on the buckling strength is larger in shell structures than in flat plate
structures. Figure 9 shows the schematic diagram which presents the post buckling behaviour of plate and
shell structures, in which the material assumed to behaves elastically.
Perfectplates
Imperfectplates
Load
Load
Perfectshells
Imperfectshells
Displacement
(a) Plate structure
Displacement
(b) Shell structure
Figure 9 Schematic diagram of post buckling behaviour
In the case of plate structures, the capacity does not reduce in the post buckling range although the
rigidity after buckling is small compared to before buckling. On shell structures, the capacity reaches its
307
peak value at buckling and reduces sharply after buckling. The dotted lines in Fig. 1 show the behaviour
of plate and shell with imperfection. A clear peak load at buckling is not apparent in plate structures,
whereas the peak load of imperfect shells is much reduced in comparison with perfect shells. These are
due to the difference of post buckling behaviour between plate and shell structures.
Koiter (1945) clarified the post buckling behaviour of shell structure. The Koiter formulation is based
on the analysis of the potential energy of the loaded structure in buckled equilibrium configurations.
Kotter initial postbuckling theory is a higher-order linearization that permits determination of slope and
the curvature of the secondary path at the bifurcation point. After that, there are so many researches
related to the effect of initial imperfection on the buckling strength of shells, such as Huchinson (1967)
and Almroth (1966).
Shells are used for areas of ships with high curvature. The main structures of submersible vehicles,
such as deep ocean research vehicles, deep sea rescue vehicles, and submarines, are constructed from
shell structure. And also shell structures are used for the spherical tanks and cylindrical tank skirts in
Moss type LNG carrier. In offshore floating platforms, critical parts are constructed from shell structures,
which are curved or cylindrical panels which may also include additional stiffening for example ring
stiffeners within a cylindrical shell. An example shell structure for an offshore structure is the column of a
semi-submersible. Shells are also used for support structures used in the offshore environment such as
piles, for offshore pipelines and for other submersible structures such as submarines.
As for the general buckling strength of ring-stiffened cylindrical shell under external pressure,
Tokugawa and Bryant proposed a split-rigidity method in which the formula for critical pressure
consists of a shell term and a stiffener term.
Pc =
EI (n2 1)
4
Et
+ e 2
2
R n2 + 2 / 2 1 n2 + 2
RRS Ls
)(
(5)
where, = R / L , L; total length of cylinder, LS; length between adjacent ring stiffeners, R; outer radius of
cylinder, RS; radius of neutral axis of ring-stiffener included effective breadth, E; elastic modulus, t;
thickness of shell, n; circumferential buckling wave number, Ie; bending rigidity of ring-stiffener with
effective breadth.
The main factor that determines the accuracy of eq. (6) is the effective breadth which is necessary for
estimating the rigidity of ring stiffeners. Formerly, 23 times of shell thickness is adopted to coincide to
the estimated collapse pressure to experimental results. These values are relatively smaller than actual
ones because the experimental collapse pressures are reduced by initial imperfections. Pulos et al. (Pulos
and Salerno, 1961) proposed the effective breadth of 1.56 Rt by considering the axi-symmetrical
deformation of shell, but this value gives a little higher value. Yoshikawa et al. (2006) (2008) proposed
the estimating procedure for effective breadth by considering both of axis-symmetrical and asymmetrical
deformation. Further studies and verifications in this field are required.
The other buckling modes of ring stiffened cylinder under external pressure are shell buckling and
torsional buckling of ring stiffener. As for the shell buckling, there are so many researches, such as
Windenburg et al. (Windenburg and Trilling, 1934) and Yamamoto (Yamamoto, 1965), are found. The
formulae for shell buckling strength have been proposed and the accuracy of formulae is confirmed by
experiments, although there still have existed some errors in the cases with large size of ring stiffeners. On
the other hand, there are few researches related to torsional buckling of ring stiffeners.
The buckling strength of ring stiffened cylinders under external pressure is investigated by Sawada et al.
(2012). Specifically, formulae to estimate the torsional buckling strength of the ring stiffener are derived and
validated with equivalent finite element analyses. The torsional buckling mode may occur if a large stiffener
height is used for the ring stiffening. The analytical formulae are derived by assuming the cylinder as an
equivalent flat panel and considering the stiffener web and attached plate as a spring. The method is adapted
to account for the reduction in bending stiffness of the plate under compressive stress and the different
spring stiffness of the plate and web. The resulting formulae correspond well to equivalent FEM analyses of
a series of cylinders considering a range of stiffener heights.
As mentioned above, researches on the simple formulae for estimating the elastic buckling strength of
ring stiffened cylinders under external pressure have been performed. To account for yielding, the equation
of Johnson-Ostenfeld andmethod can be adopted to estimate the elastic buckling strength, but there still
remains the problem of what kind of stress must be considered as a reference stress.
Some research has compared numerical calculations for the buckling strength with FE analysis to
experimental results.
308
An investigation into the use of nonlinear FE analysis for determining the collapse pressure of submarine
pressure hulls is completed by MacKay et al. (2013). Results from 47 experimental tests on small-scale ring
stiffened cylinders, with slenderness parameters equivalent to submarine pressure hulls, are compared to
numerical results using ANSYS. Consideration is given to the material model, the out of circularity,
artificially induced corrosion damage and boundary conditions. The accuracy of the numerical analysis is
shown to be within 10% of experimental results with a 95% confidence interval, which is a significant
improvement when compared to conventional design methods.
The effect of residual stresses on the buckling and ultimate strength of stiffened cylinders was
numerically investigated by Cerik et al. (2013) with an emphasis on shakedown which might occur during
the service, and the results of numerical calculation were compared with experimental results. They
showed that yield level tension stresses induced by welding have a significant effect on strength and
stiffness, and also that cyclic compressive loading does not cause any significant increase in ultimate
strength although the compressive stresses at shell midbay locations are reduced by external loading.
Offshore pipelines experience significant axial, bending and pressure stresses during installation and
operation on the seabed. Growing use of pipelines in deeper waters increases the bending deformation
and pressure loading on the pipe, which can cause buckling and collapse if not correctly accounted for.
Buckling modes include lateral (snaking) buckling, upheaval buckling and propagation buckling.
Gong et al. (2012) investigated how a buckle can propagate along a pipeline due to the combination of
high external pressure and bending deformation at one point in the pipe. Experimental tests are completed
in a hyperbaric chamber and are then compared to equivalent finite element analyses with good
correlation observed.
Propagation buckling is also investigated by Albermani et al. (2011). Quasi-static propagation buckling
tests in a hyperbaric chamber are completed and compared to a modified analytical solution and nonlinear
finite element analysis.
Faceted pipe geometry is proposed to increase the propagation buckling capacity of the pipeline. The
interaction between upheaval, lateral and propagation buckling is investigated numerically by Karampour
et al. (2013).
Yudo et al. (2012) investigated the mechanical behaviour of pipe under pure bending load considering
the oval deformation of the section. The study shows how nonlinear finite element analysis is required to
account properly for the oval deformation in the pipe, which has the effect of reducing the buckling
strength. The authors also investigate buckling of straight and curved pipes Yudo et al. (2014).
There are some investigations on how the production process used for pipes affects their buckling
strength.
Large scale four point bending tests on spirally welded tubes for combined wall piles were performed
by Van Es et al. (2013). Two tests are documented showing the performance of different steel grades and
wall thickness. The tests show the buckling behaviour of the tubes and provide some verification of
equivalent numerical analyses. Measurements of imperfection in the tubes were taken and compared to
the eventual buckling location. These show that dimples created during cold forming play an important
role in the location of the buckle.
Thick-walled steel pipes manufactured though the UOE process are used in deep-water pipeline
applications. The UOE manufacturing process steps includes crimping U-ing, O-ing, and expansion.
Chatzopoulou (2014) simulated the production process and also calculate the bending response of
externally-pressurized pipe. The analysis showed that the increase of the total expansion value leads to
minimization of the out-of-roundness of pipe, but at the same time it increases the hoop strain and
consequently it reduces the collapse pressure resistance of the pipe due to the influence of Bauschinger
effect. For combined pressure-bending loading condition, for relatively low external pressure levels, UOE
pipes exhibit higher bending deformation capacity compared to seamless pipes. On the contrary, as the
pressure level increases, the UOE pipes are less resistant to bending loads.
4.4
Ship structures
The research on ultimate strength of ships in the last years has been concentrated in few main topics.
Following the publication of the new Classification Societies Rules designated as Common Structural
Rules (CSR) concerning to ultimate strength, several works were dedicated to the comparison between
the CSR and different methods of evaluating the ultimate strength of ships as presented in previous ISSCCommittee III.1 reports.
In the present period, special attention has been dedicated to progressive collapse methods, FE analysis
and their applicability to intact ship integrity assessment, and degradation of the ultimate strength by
corrosion or local structural damage. Also, some work was done on the effects of interaction between
different types of loading and between vertical and horizontal bending.
4.4.1
309
Progressive collapse methods
Several researchers have been developing their own in house software for evaluating the behaviour of
the hull girder under applied bending moment, covering intact and damage conditions, combined
bending and corrosion, based on approximate progressive collapse methods. The original Smiths
method consists of a tangent or differential method and the secant or direct method. The direct method
is the one used by the CSR (Gordo and Guedes Soares, 1996) and may be represented by a simple
equation as follows:
M = i Pi Fi ()
(6)
where the position of the stiffened element is Pi, and the load in each element is Fi which depends on the
curvature and the position of the neutral axis at each curvature, which requires an iterative procedure to
balance the cross section. The above equations become the followings when horizontal and vertical
bending are made explicit:
MV = i ( yi yna)i Ai
(7)
MH = i (xi xna )i Ai
(8)
The secant method itself is not a progressive collapse method but the moment curvature relationship is
obtained by calculating successive points of the curve and an incremental curvature procedure facilitates
the convergence of the method at each step. With a linearization of the load shortening curves of the
elements, the method may be used for ultimate strength assessment (Paik et al., 2013) .
The tangent method (Smith, 1977) (Yao and Nikolov, 1992) is a real progressive method and its
potentialities are well and systematically described by Fujikubo et al. (F2012b). The governing
equation is:
M i +1 = M i + [ Di ] i
(9)
uses the slope of the load-shortening curves of the individual

The tangent stiffness matrix
structural elements at each curvature. This first order method allows the differential neutral axis, defined
as the axis of zero variation of strain due to an increment of curvature, to be determined, which is not
possible in the direct method.
Recently several in-house implementations of the method prescribed by IACS CSR have been
presented. Andric et al. (2014) compare some of them with their own software OCTOPUS and FEA for
five different hull girder structures. A coefficient of variation of between 4 and 10% was found for the
prediction of the different methods in the different types of structure. They concluded that differences
among calculated ultimate bending moments are much higher for double hull tankers and bulk carrier
than for container ship and single hull tanker.
A comparison of several methods to predict the longitudinal bending moment-curvature relationship
for a series of box beams representative of aluminum structures is made by Benson et al. (2013c),
accounting for compartment-level, gross panel buckling effects of the orthogonally stiffened structure.
The extended progressive collapse method is shown to compare favorably to the equivalent finite element
analysis when overall buckling modes dominate.
Chenfeng et al. (2013) proposed an approximate approach by combining the progressive collapse
mechanism and hulls beam theory. Similar work have been done by Li et al. (2013a).
A method based on the kinematic displacement approach is proposed by Tayyar (2012) where the
deflection of the hull is evaluated by the assembly of circular arcs by solving the differential equations of
the beam theory. The same method was used by Choung et al. (2013) in conjunction with the formulas of
the CSR.
One of the most important part of information required by progressive collapse analysis if the
definition of the load-shortening of stiffened elements. Gannon et al. (2012a) generated such curves
simulating by FE the three-dimensional distribution of welding-induced residual stress and distortion, and
applying it in Smiths method. Comparison with CSR and experimental data they concluded for a good
agreement of the procedure with experiments. The same methods were used by Gui-Jie et al. (2012) for
comparison with experimental data of a model of container ship hull.
The ultimate strength of composite ship hulls was studied by Misirlis et al. (2013) by developing a
progressive failure model based on the finite element method. The analysis of the study leads to the
310
conclusion that differences failure criteria and modes of collapse result in significant differences due to
interaction of local and global effects.
Makouei et al. (2014) developed software based on progressive collapse that uses as input the
behaviour of large stiffened panels under compression instead of the discretized stiffened elements. The
method is compared with the analytical incremental-iterative method proposed by the IACS CSR.
The effects of damage severity and location on the residual strength of a patrol vessel made with
aluminum are investigated Taylor et al. (2013) using ISFEM (Intelligent Supersize Finite Element
Method) and damage scenarios from DNV.
Toh et al. (2012) present a typical procedure of incremental-iterative method for the estimation of the
ultimate strength of damaged ships. The stress-strain relations (- curve) for damaged members are
obtained by FE analysis for tension and compression. Comparisons with the intact hull are used to
establish the reduction rates of hull girder ultimate strength by damage.
4.4.2
Damaged structures
Benson et al. (2013f) used the progressive collapse method to assess the strength of several box girder
models with different levels of ruptured damage applied and compare the results with FE analysis. The
effect of the transverse and longitudinal extent of the rupture is investigated by changing the area and
location of the damage.
Choung et al. (2012a) provide two convergence criteria to find translational and rotational locations of
the neutral axis and relates it to the heeling angle of ship considering different asymmetries due to
material, load and damage. It is concluded that the ultimate bending moment is correlated with the area
reduction due to damage. The method was extended with a probabilistic approach to define the extension
and location of damage applied to a VLCC (Choung et al., 2012a) (Nam and Choung, 2012).
Similarly an incremental formulation of the progressive collapse behaviour of hull girders with
damage subjected to longitudinal bending is presented by Zubair et al. (2012) and Fujikubo et al.
(2012b) based on the tangent method. Taking as examples bulk carriers and a tanker with collision
damage in the side structures, the movement of neutral axis and biaxial bending response were analyzed.
The effect of the side damage in the ultimate biaxial bending of the hull is shown for a different level of
side damage.
Yamada (2014) proposes a numerical methodologies using FEA in assessing residual ultimate hull
girder strength after collision simulation.
Makouei et al. (2015) also takes into account the influence of the rotation of the neutral axis on the
residual hull girder strength of asymmetrically damaged ships under longitudinal bending to obtain a set
of simple design formula to predict the residual strength of a damaged AFRAMAX tanker from its intact
strength.
The hull girder is modeled as a beam with asymmetric cross section subjected to biaxial bending. The
explicit expression of the rotation and translation of the neutral axis due to the buckling and yielding of
structural elements is given. The proposed formulation is applied to the residual strength analysis of bulk
carriers and a tanker having collision damages at the side structures. Particular focus is placed on the
influence of the rotation of neutral axial on the residual hull girder strength and the solution procedures to
obtain the residual strength including the case of biaxial bending.
Pollalis et al. (2013) present an investigation regarding the influence of modeling parameters in FE
codes on the ultimate strength of intact and damage hulls. The rotation of the neutral axis and the bi-axial
bending are studied to identify their effect on the reduction of the ultimate strength in comparison to the
intact hull.
A simple model for predicting the ultimate strength of cracked box beams under torsional loads is
proposed by Shi et al. (2012a) based on a FEA study considering different crack sizes and locations.
4.4.3
Corrosion
The effect of cracking and corrosion on ultimate strength of ships is analyzed by Gao et al. (2012).
NLFEM and ISUM are used on container ships structures and simple equations are derived to correlate
the ultimate strength with level of cracking and corrosion.
The effect of time-dependent corrosion wastage on the ultimate hull girder strength and geometric
properties of the cross section is studied by Kim et al. (2012a) (2012b) both for different sizes of double
hull oil tankers and container ships. The main objective of the study was to contribute to better definition
of the corrosion addition on rules.
4.4.4
Complex ship structural components and complex loading
The ultimate strength of a Capesize bulk carrier hull girder under combined global and local loads in the
hogging and AHL condition is performed by Shu et al. (2012) using FE software ABAQUS.
311
No.4
No.5
No.3
Deck
Hull girder N.A.
Cargo Pressure
Cargo Pressure
Figure 10. AHL condition on a typical bulk carrier (Shu and Moan, 2012)
Inner Bottom
Bottom
Sea Pressure
1.0
M/MU
0.8
0.6
Results from Nonlinear FEM analysis
Linear interaction curve determined by two points
0.4
0.2
0.0
0.0
0.2
0.6
0.4
0.8
1.0
P/PU
Figure 11. Interaction equation in AHL conditions and FE results. Ultimate pressure determined by FE
analysis.
The effect of modifying the geometry of the existing ship to satisfy the new regulations was evaluated
using design modification factor, which was determined to be 1.4. The research leads to the presentation
of a practical interaction equation between global hogging bending capacity and average external sea
pressure over the bottom in the form of:
M / M u + P / Pu = 1
(10)
Toh et al. (2014) performed similar study for a Panamax bulk carrier under AHL condition but
considering different levels of initial imperfections and found significant reductions on the ultimate
strength due to both effects of about 25% which compares with the 37% found in the previous study.
Pei et al. (2013) noted that the double bottom of an empty hold is subjected to both longitudinal thrust
due to hull girder bending in hogging and local bending caused by high pressure loads on bottom plating
and applied different methods to this case of alternate loading condition. A new method was presented
based on the Singularity Distribution Method and progressive collapse analysis by ISUM/FEM. The study
concluded that the ultimate hull girder strength is reduced by roughly 20% due to local bending from
secondary loads.
Some others (Kim et al., 2013b) pointed for a reduction of less than 10% due to the effect of the lateral
pressure on the ultimate vertical bending moment of a double-hull oil tanker.
Related to the same subject, Gordo (2012) performed a study on the behaviour and ultimate
compressive axial strength of a double-bottom subject to alternate lateral loading considering levels of
lateral pressure.
Estefen et al. (2012) analyzed the behaviour of a complex part of the column of a new generation of
semi-submersible platform up and beyond collapse. The buckling mechanism was related to the measured
initial geometric imperfections. Different imperfection shapes generate different sequences of collapse
deformations but the ultimate load does not change much, Figure 12. However a significant variation
occurred in the displacement at collapse depending on the number of half-waves of imperfections.
Underwood et al. (2012a) investigated the influence of a damage-hole on the ultimate collapse strength
of steel grillage arrangements. The influence of variations in geometrical arrangement of the structure is
investigated through the use of FE analysis and the size and dimension of damage is represented by a hole
312
on the structure. Results indicate that the damage aperture may influence the type of collapse depending
on the slenderness of the plate. The need to assess both interframe and overall collapse modes is pointed
out as necessary to ensure a good prediction of the ultimate strength of a damaged structure.
Figure 12. Different collapse deformations due to different shape of initial imperfections.
4.4.5
Reviews and applications
Zhang (2013) presents a review of buckling and ultimate strength assessment methods comparing them
with FE analysis results for existing oil tankers and bulk carriers.
Gordo et al. (2011) consolidated the research done by the Portuguese marine center (CENTEC) on the
ultimate strength of ship structures and components.
Geometrically and materially nonlinear finite element method and various incremental-iterative
methods based on Smiths approach are benchmarked against the results of experimental testing of
various stiffened box girders submitted to extreme pure bending by Kitarovic et al. (2013). The influence
of various relevant aspects of the employed methods is evaluated and discussed considering global and
local of structural response.
Li et al. (2013) analyzes the structural behaviour of a double skin bulk carriers by comparing FE results
from elastic direct analysis with the nonlinear analysis in order to evaluate the safety margin of the ship.
The identification of the main parameters that influence the ultimate strength of welded aluminum ship
structures and the validation of rapid assessment tools are dealt by Magoga et al. (2012), using ISFEM
(Intelligent Supersize Finite Element Method) available in ALPS/HULL module of the MAESTRO to
examine the ultimate hull-girder strength in vertical bending of a generic aluminum midship section. The
results are compared with the UltSAS (Ultimate Strength Assessment) program.
Nikolov (2012) studied the progressive collapse behaviour of box girder models by FE analysis
considering the effect of initial deflections and welding residual stresses. Unloading and reloading in the
pre-collapse range are carried out to simulate the welding stress relief in experiments, and concluded that
the ultimate strength was not affected by such alternate pre loading.
Saad-Eldeen et al. (2012) (2013a) (2013c) (2014) performed a series of simulations with FE to
compare with the experimental results of corroded box girders under pure bending. Different elastoplastic material models have been developed. Similar approach has been used by Xu et al. (2013a) to
evaluate the ultimate strength of a tanker considering that the stress-strain curves of panels obtained in
small scale tests are representative to the ones of stiffened panels of the compressive zone of a tanker hull.
The influence of different structural configurations of a river-sea container ship on the ultimate strength
is evaluated by Zhang et al. (2013) and the limit state in combined action of bending and torsion is
estimated by numerical and experimental evaluation (2013).
4.5
Offshore structures
Evaluation of the ultimate strength and history of collapse of structures in their intact and damaged
conditions is one of the key issues in advanced design techniques based on explicit safety evaluation and
collective optimization. The ultimate strength of an offshore structure is evaluated by using non-linear
finite analysis of the structural model, where gravity loading is applied as an initial load step, then the
ocean design loads for a chosen direction is incrementally applied until the ultimate strength of the
structure is reached (Ueda and Rashed, 1991). As an important attribute that affects the life expectancy,
requalification and life extension of the facility, the ultimate capacity of offshore structure can
significantly influence the reliability levels and operational costs.
Studies on ultimate strength of offshore structures are rather limited compared with considerable
number of documents on ships and ship-shaped structures. However, recent publications shed light on
313
effect of material property, aging factors (corrosion and fatigue), residual strength after collision with
supply vessels, and ultimate limit state based design and evaluation.
Arjomandi et al. (2012) investigated the response of sandwich pipe systems subject to pure bending
through a set of numerical parametric studies. The buckling response of the systems was evaluated by
using the linear eigenvalue buckling analysis method, while nonlinear analysis was used to calculate the
pre-buckling, buckling and post-buckling response. By employing nonlinear finite element analysis, the
effect of significant geometrical parameters, material properties and the intra-layer interaction mechanism
on the characteristic response of the systems was established. Ye et al. (2012) evaluated the ultimate
strength of a typical drilling semisubmersible platform. The progressive collapse of the object platform
was investigated using nonlinear finite element analysis with ANSYS. The analysis accounted for
material nonlinearities but excluded geometrical nonlinearities and imperfections. The simulation results
were compared with theoretical predictions by DnV rules and ABS rules and finally the corresponding
ultimate limit state formulations were developed. An et al. (2014) investigated the collapse behaviour of
sandwich pipe (SP) with strain hardening cementitious composites (SHCC) core under external pressure
by both experiment and modelling. The SHCC preparation and the SP fabrication process are described in
detail. Three SPs are tested using a hyperbaric chamber to obtain the very similar collapse pressure,
demonstrating the reliability of the experimental results. A quarter-ring model is adopted to perform the
FE analysis of structural behaviour of SPs under external pressure, where SHCC is modelled by a
concrete damaged plasticity model. The accuracy of the model is verified by the good agreement between
the numerical and experimental results.
Khedmate et al. (2012) calculated the ultimate strength of locally corroded tubular members under
axial compressive loads in order to provide an initial estimation. To carry out this, a parametric finite
element study was performed and the obtained results were examined against available experimental test.
Then, validated models were used to derive a semi-empirical formula as a function of some parameters
such as slenderness of the tubes and location of patch corrosion. It was found that location of corrosion
has great effect on reduction of ultimate strength. Also, the study on the effect of corrosion geometry
showed that tubes with different corrosion dimensions have different behaviours under compressive
loads. Zhang et al (2012) investigated the effect of corrosion damage on ultimate strength of semisubmersible platform deploying a 3000m deep sea semi-submersible platform, see Figure13.
Their research indicated that corrosion damage has a significant influence on the ultimate bearing
capacity of typical component and node on platform. The ultimate bearing capacity approximately
decreases linearly with the increase of service years considering corrosion damage. Strength of typical
component and node will be reduced and load will be enlarged under the combined effects of corrosion
damage and various kinds of loads. Hence, long-term and regular anticorrosion of platforms is of great
importance to improve the ultimate strength in the whole life cycle. Cerik et al. (2013) investigated the
ultimate strength of damaged stiffened cylinders (by collision impact) subjected to combined axial
compression and radial pressure, using the commercial FE code ABAQUS. After validating the numerical
modelling strategy, a series of FE analyses were carried out for both damaged ring-stiffened and stringerstiffened cylindrical shells, varying the extent of damage and the slenderness of the cylinders. This was
done to derive formulas for predicting the residual strength of the damaged stiffened cylinders which can
be useful for reliability-based studies for risk evaluation.
Figure 13. Stress distribution of typical component of semi-submersible platform (Zhang et al., 2012)
314
Nezamian et al. (2014) overviewed the ultimate strength assessments and their role in understanding the
structural system response and failure modes to extreme loads for requalification and life extension of an oil
field and for demonstrating fitness for purpose. An assessment of the structural integrity of thirteen
identified platforms under existing conditions was undertaken as these platforms are nearing the end of their
design life. This paper also describes an efficient method, the idealized structural system, to evaluate the
failure modes in ultimate strength of structures. Its application to offshore frame structures was presented
taking account of the nonlinear behaviour of members, joints, piling foundation and the structure as a whole.
Storheim et al. (2014) investigated the damage to offshore platforms subjected to ship collisions, see
Figure 14. The effect of the shipplatform interaction on the distribution of damage is studied by
modelling both structures using nonlinear shell finite elements. A supply vessel of 7500-ton displacement
with bulbous bow is modelled. A comprehensive numerical analysis program is conducted. It was
indicated that simplifying one body as rigid quickly leads to overly conservative and/or costly solutions,
and is in some cases non-conservative. Current design guidelines for ship collisions with offshore
structures need to be revised in view of significant increases in the supply vessel sizes and a wide variety
of bow configurations that can exert significant collision forces over small areas on platforms.
Impact force (MN)
50
40
30
20
10
2
Indentation (m)
Figure 14. Forcedeformation curves for a 5000-DWT supply vessel (Storheim and Amdahl, 2014)
Moreover, ultimate limit state based global structural analysis is addressed as well. Park et al. (2014)
developed a hybrid substructure for a 5MW offshore wind turbine in order to reduce the wave forces. The
fluid domain is divided into two regions to calculate the wave forces acting on the hybrid substructure
with multi-cylinder and the scattering wave in each fluid region is expressed by an Eigen-function
expansion method. The comparison between the mono pile and the hybrid substructure is made for wave
forces. Using the wave forces obtained from this study, the structural analysis of hybrid substructure is
carried out through ANSYS mechanical. It is found that the hybrid substructure system satisfy structural
safety in respect of ultimate limit state (ULS). Lee et al. (2014) introduced a procedure of global
structural analysis for a semi-submersible drill rig in accordance with DNV offshore rules. ULS of the
unit was analysed with LRFD method for operating and survival condition.
It can be noted that available documented publications on the ultimate strength of offshore structures is
rather limited compared with considerable numbers of publication on ship/ship shaped structures.
Offshore structure is a huge assembly of various components depending on the required function and
working conditions. Much efforts has been made on the ultimate strength at the level of specific
components, i.e., stiffened panels, tubular members or joints; and on the effect of various parameters, i.e.,
composite material, initial imperfection, aging factors (corrosion and fatigue), less efforts have been made
on the ultimate strength of the offshore structure globally and systematically. In this regard, systematic
assessment of ultimate strength of offshore structures deserves further efforts.
4.6
Traditionally the use of composite materials in the shipbuilding industry was limited to pleasure crafts
and small workboats. Nevertheless, in recent years there has been a reduction in the cost of composites
structures mainly due to improved design tools, new fabrication procedures and reduction in the cost of
high end materials. Furthermore, the need to achieve higher operational speeds and to reduce the
maintenance costs is currently leading shipyards and naval architects to use composites in the design of
larger vessels (Anyfantis and Tsouvalis, 2012). Mouritz et al. (2001) reports a steady increase in length of
full composite vessels in the naval industry particularly outlining patrol boats, mine countermeasure
vessels and corvettes and predicts that this trend could see warships between 120m-160m being built by
2020.
315
The increase in size of composite vessels has however uncovered new design and analysis challenges.
Chen et al. (2007) (2008) state that the analysis of a composite ship structure should not be restricted to
local structure, but that it should also include ultimate longitudinal strength assessment for ships over
60m. In fact, Chen et al. (2003) were the first authors to developed an analysis methodology for the
evaluation of the ultimate strength of a composite vessel. Following a Smiths type approach Chen et al.
calculated the ultimate strength of ship cross section using load shortening curves determined from
composite column theory. This approach was later modified (Chen and Guedes Soares, 2007) (Chen and
Guedes Soares, 2008) to include a progressive failure finite element model in the calculation of the
ultimate strength of the ship panels.
In a progressive failure model the ultimate strength of the composite structure is determined by coding
a suitable composite failure criterion within an incremental loading algorithm. Once failure is detected the
degradation model is used to reduce the properties of the laminate to account for the material damage and
to effectively weaken the panel. This iterative process, illustrated in Figure 15, is repeated until
equilibrium is no longer possible. At this point the structure is deemed to have completely failed.
Definition of initial state
Initial geometric imperfection

is defined
The initial pressure P0
is specified
Non-linear analysis to
establish equilibrium
No
Converged solution
obtained?
Yes
Predicted
final failure
load (Pult)
On-axial stress are

calculated
End
Are there failed

elements?
Re-establish
equilibrium by
restart analysis
Yes
Degrade
properties
of material
Increment load P=P0+P
Figure 15. Typical flow chart of a composite progressive failure algorithm (Yang et al., 2011)
A recent example of a progressive failure analysis is the work by Yang et al. (2011) on the ultimate
strength and reliability of laminated composite panels under axial compression. These authors used TsaiWu interactive failure criterion to identify the failure of the composite material combined with Engelstad
failure terms to establish if the failure was caused by matrix fracture, fibre breakage or shearing failure.
The material damage was represented using a limited, but constant, stiffness reduction specific for each
one of the three previously listed composite failure modes. Good agreement between experimental and
numerical predictions was obtained for thin and moderately thick plates however poor agreement was
obtained for thick plates.
4.6.1
Failure identification and material degradation models
The accuracy and efficiency of a progressive failure finite element simulation hinges on two crucial
components: an accurate composite failure criterion, used to identify the load that causes the initiation of
failure, and a degradation model, used to represent the material damage progression until the ultimate
failure.
Philippidis et al. (2013) introduced an anisotropic non-linear elastic constitutive model as well as
failure onset criteria and degradation strategies in order to calculate the ultimate strength of thin to
moderately thick composite structures under static loads. Puck and Schrmann, Lessard and Shokrieh as
well as combined Puck/Chang-Lessard failure criteria were used in three models to predict the inception
of failure in multi-directional (MD) laminates. Depending on the type of damage predicted by the
respective failure criteria, two different degradation strategies were adopted. In case of matrix cracking a
gradual degradation model was adopted whilst a total ply discount was used for fibre breakage. The
316
model accuracy was measured against simple experimental tensile coupon tests. Good agreement was
found in all cases.
In a similar line, Moncada et al. (2012) evaluated the capabilities of the of two micro-mechanics
theories, i.e. the generalized method of cells (GMC) and the high fidelity method of cells (HFGMC), for
the progressive failure prediction of polymer matrix composites and also evaluated the influence of four
different failure criteria (maximum strain, maximum stress, Tsai-Hill and Tsai-Wu) applied at the
microscopic level. The results indicate that the choice of failure criteria has a significant effect on
accuracy of the ultimate strength predictions with the maximum strain criterion showing the best
agreement with the experiments. The differences in results obtained by switching between the GMC and
HFGMC theories were small compared to those found among the four failure criteria.
The model allows the determination of the ultimate strength of laminates under complex tri-axial
loadings. The model uses the macroscopic (laminate level) loading to calculate the state of stress at a
microscopic (fibre-matrix level) level which allow to establish the failure and damage progression at the
mesoscopic level.
The model was successfully implemented to provide predictions of failure envelopes and non-linear
stressstrain curves of isotropic, unidirectional and mutli-directional laminates made of glass/epoxy and
carbon/epoxy materials and subjected to 3D loadings, including through-thickness stresses.
Chamis et al. (2013) also present a MS progressive failure model to provide theoretical predictions of
the ultimate strength of composite laminates. The critical damage events/indexes predictions tracked
trans-laminar and inter-laminar composite failures. Composite laminates, with and without a central hole,
were subjected to uniaxial tension and compression, bending and thermal load cases. Good agreement
was found in all cases. It is noted that although accurate, due to its inherit complexity, the MS method
proposed by both Chamis et al. and Carrere et al. are not yet suitable for design work.
It can be noted that all the investigations discussed so far have established the ultimate strength of the
composite structures using a mesa-scale (lamina level) damage progression model. Gan et al. (2011)
presented an investigation aimed at modelling the initiation and evolution of damage within a Glass Fibre
Reinforced Plastic (GFRP) by including real geometric variability. The investigation implemented 3D
textile modelling instead of classical laminate theory. Good agreement was observed between predicted
and actual stress-strain curve, however the final failure was poorly predicted.
All damage material models used in the investigations discussed so far fall into the elastic damage
category, limiting their application to laminates that exhibit and elastic-brittle behaviour. Nevertheless,
some thermoset and thermoplastic composites exhibit a plastic response under transverse and shear loads.
In order to establish the ultimate strength of composite structures with such characteristics, Chen et al.
(2011) developed a combined plastic damage model for composites which accounts for both the plasticity
effects and material properties degradation. The material model was verified by performing progressive
failure analysis of composite laminated unstiffened panels with a central hole. The model showed
sufficient accuracy in the prediction of failure loads.
4.6.2
Ultimate strength of composite stiffened panels and box girders
All the works discussed so far have focused on investigation of the ultimate strength of simple coupon
type samples or unstiffened laminated panels. Anyfantis et al. (2012) state that the ultimate strength of
composite stiffened panels has not been sufficiently studied. These authors proposed and validated a
progressive failure methodology for the calculation of the ultimate strength of longitudinally, blade
stiffened composite panels loaded in compression. The progressive failure methodology was incorporated
into the commercial FE code ANSYS. It can be appreciated from Figure 16 that good agreement was
achieved for test cases were the panels exhibit an abrupt failure mode. However it was found that the
proposed methodology was less effective in panels that exhibit mode switch prior to failure. For these
panels, it was recommended that inertial effects should be considered in the framework of a non-linear
dynamic analysis.
The review of literature shows that very little work has been done on simplified models for the
calculation of the ultimate strength of composite panels. Yang et al. (2013) have recently address this by
presenting a modifying Brubak et al. (2007) (2008) semi-analytical ultimate strength models for isotropic
stiffened models in an attempt to represent laminated composite panels. It is important to note that the
model is still limited to the analysis of unstiffened panels only. Thin panels simulations show reasonable
accuracy whilst thick panels show overly conservative results.
Recently, Sayyar et al. (2013) investigated the ultimate strength of pre-stressed composite box girders.
These authors state that experimental results on aircraft stiffened thin-sheet composite structures, as used
on aeroplane wings, indicate that the effective buckling strength is generally less than the tensile strength.
The reversible nature of the stress systems in thin walled wing structures thus leads to a design solution
governed by compression that leaves the superior tensile strength of the composite material under-
317
utilized. Experimental carbon composite box girder models pre-stressed using titanium rods were tested
to destruction using a cantilever bending load case. It was found that the pre-stressing of the box girder
increased the ultimate strength by 89% while adding only 15% to the weight of the structure.
700
600
Force (kN)
500
Experimental failure initiation
400
Experiment [6]
Simple nonlinear analysis
Failure analysis
300
200
100
0
-2250
-1500
-750
750
0
Strain - 22 (
1500
2250
Figure 16. Force versus axial strains measured at the web of the blade stiffeners (Anyfantis and Tsouvalis, 2012)
4.6.3
It is common knowledge that composite laminates are not affected by corrosion in the same way as metal
structures do. Nevertheless, recent work (Kant et al., 2010) suggests that carbon fibres under prolonged
exposure to sea water and elevated temperature have their mechanical properties, specifically elastic
modulus, reduced. Based on this evidence, Kant et al. (2011) reported on an ongoing investigation aimed
at establishing the effects of sea water on the ultimate tensile strength of carbon fibres using nano-tensile
testing. Based on limited experimental data available at the time of the publication of results, the authors
concluded that the fibres soaked with sea water did not show a clear trend of variation in measured
properties such as failure stress and modulus variation with strain as a function of exposure conditions,
and the potential degradations seems to occur for fibres with poor mechanical properties quantified by a
value corresponding to the storage modulus with respect to axial strain.
4.6.4
Compression after Impact
An area that has received attention in the last years is the ultimate Compression After Impact (CAI)
strength of composite structures.CAI strength is widely used as a damage tolerance threshold value for
composite structures in the aerospace industry but has potential applications in the shipbuilding industry
to the evaluation of the ultimate strength of composite structures damaged by ballistic impacts and
grounding.
Nettles and Jackson (2013) conducted an investigation into the CAI strength of out-of-autoclave
(OOA) processed laminates. OOA laminates are gaining much attention due to the elimination of needing
a costly autoclave large enough to hold the part to be cured. The CAI strength of two commercially
available OOA systems (IM7/8552 & IM7/MTM-45) was compared using a conventional autoclave
laminate (T40-800/5230) as a baseline. Three different levels of impact severity were investigated and
compared with undamaged control specimens as presented in Figure 17.
Compression strength (Mps)
750
650
IM7/8552 IM7/MTM45 T40800/5320
550
450
350
250
Undamaged
1.1J
2.3J
Impact severity level
4.2J
Figure 17. Compression strength versus impact severity (Nettles and Jackson, 2013)
318
Information on damage, size and morphology were assessed along with the CAI ultimate strength
values. The results of the investigation showed that the two OOA laminates tested have similar damage
tolerance characteristics to the autoclave processed laminate. This is important for the composite
shipbuilding industry where the large size of the structure may preclude the use of an Autoclave.
Lee et al. (2011) investigated the CAI strength of MD laminates using the Soutis-Fleck model, that
mathematically replaces micro-buckling/kinking damage with a through thickness line crack. The model
requires a damage area, reduced properties in the damage site and in-plane stress distribution near the
damage site. The damage site was model as an equivalent hole and as a soft inclusion. From the
comparison of the theoretical predictions and experimental measurements for CAI strength, it was found
that when fibre breakage occurs at certain plies, the equivalent hole model was more suitable for the
prediction that the soft penetration model. The differences between the theoretical and experimental
strength results were found to be less than 10%.
4.7
4.7.1
Aluminum structures
Introduction
Aluminium structures are used in a variety of marine applications such as hull and decks in high-speed
boats and catamarans and superstructures for ships. Other applications are box-girder bridges, walls and
floors in offshore modules and containers. The use of aluminium is primarily being driven by the demand
for reduced structural weight, increased payload, higher speed and reduced fuel consumption. Compared
to steel structures, the ultimate strength and collapse behaviour of aluminium structures are significantly
affected not only by initial imperfections (i.e. residual stresses and initial deformations), but also by the
material nonlinearity and the deterioration of mechanical strength in the heat-affected zones (HAZ)
caused by welding in assembly and fabrication. The level of the mechanical strength decrease in the HAZ
is dependent on alloy type, temper, welding process and welding parameters. This reduction can be
typically between 30%50% of the strength of the base material for 5xxx and 6xxx-series alloys, most
commonly considered for ship construction, joined by fusion welding (Sensharma et al., 2011). Such a
material strength decrease in HAZ may lead to a serious reduction of the aluminium structural strength.
The application of ultimate limit-state (ULS) for aluminium ships requires adapted ultimate strength
methods which take into account the influences of the alloy type, material softening in the HAZ, strain
hardening and loading type. However, the available literature is relatively limited over the last three years
when compared with the studies on steel structures.
4.7.2
Weld-induced effects
Extensive work was carried out by Sensharma et al. (2011) concerning the effect of welded properties on
aluminium structures. This study intended to provide a basis for modification to design standards by
analysing the impact of HAZ on the load-carrying capacity of aluminium stiffened panels under
compressive, tensile and bending loads. To that end, fine mesh finite element models were developed and
analysed for a range of tee bar-stiffened panels along with 5083-H116 and 6082-T6 aluminium alloys.
For each panel, three comparative FE models were made: the first consisting of all base metal throughout
the model with no separate HAZ regions, the second with all HAZ metal throughout the model, and the
third with HAZ regions adjacent to welded edges. Based on the analysis conducted, it appeared that there
is a significant advantage for modelling aluminium marine stiffened panels by explicitly considering the
HAZ extent, location, and strength rather than applying a uniform knock-down factor to the strength of
the HAZ material.
A parametric series of nonlinear finite element analyses of unstiffened plates with 5083-H116 and
6082-T6 material properties in uniaxial compression are presented by Benson et al. (2013b). The
study demonstrates the effect of the material model, imperfection parameters, residual stress and heat
affected zone on the plate load shortening curve. A series of plate load shortening curve datasets are
presented, accounting for three severity levels of imperfection and including effects of HAZ and
residual stress. These curves are proposed as suitable for direct implementation in a Smith type
progressive collapse method. As an example, two load shortening curve datasets are shown below.
These are representative of typical plates and use the base case material and geometric properties
(defined within the text). The aluminium alloy results use a HAZ ratio of 0.125 with HAZ located at
the four plate edges.
Paulo et al. (2013) carried out a set of finite element analyses (FEA) to reproduce the mechanical
behaviour of integrally stiffened aluminium panels under axial compression. The aim was to assess the
sensitivity to initial geometrical imperfections and material properties. Two different types of 6082-T6
panels used in high-speed catamaran ferries, equivalent to previous experimental tests, were taken into
consideration as FEM models (Figure 19).
319
1.0
0.8
0.8
0.6
0.6
0.6
0.4
0.2
0.0
0.2
0.4
0.6
0.8
1.0
Normalised Stress
1.0
0.8
Normalised Stress
Normalised Stress
(a) 5083-H116 aluminium alloy

1.0
0.4
0.2
0.0
0.2
0.4
0.6
Normalised Strain
0.4
0.6
0.8
1.0
0.8
1.0
3
0.4
0.2
0.0
0.2
Normalised Strain
Normalised Strain
(b) 6082-T6 aluminium alloy

Figure 18. Plate load shortening curves with slight (left), average (centre) and severe (right) imperfections
Rigid wall reference point
x, y, z, rx, rz = 0
z=0
(long. Supported model)
Rigid wall
Rigid wall
LOAD
z=0

y, z, rx = 0
Reference surface of
Shell elements
10
25
a) with L-shaped stiffeners

Rigid wall

x, y, z, rx, rz = 0
z=0
Xz plane symmetry
y, rx,rz = 0
Rigid wall

y, z, rx,rz = 0
20
8.5
LOAD
b) with trapezoidal stiffeners

Figure 19. Detailed description of FEM models for analysis
Reference surface of
Shell elements
320
Panels with L-shaped stiffeners were MIG welded whilst panels with trapezoidal closed-section were
joined using a friction stir welding (FSW) technique. In the numerical simulation models, particular
attention was given to the correct reproduction of the boundary conditions and the configuration of the
loading edges. As an overall conclusion, it was demonstrated that a reliable ultimate load prediction in the
presence of compressive loads should be preceded by a careful assessment of geometric imperfections
that are used in the initial model, with their influence more significant than that coming from the effective
stress vs. equivalent plastic strain curve in the HAZ of the studied panels.
A nonlinear FEA-based numerical study was done by Khedmati et al. (2014) in order to investigate the
effects of various geometrical imperfections types on the ultimate strength of aluminium stiffened plates
under different combined axial compression and lateral pressures. The finite element models taken into
consideration were triple-span panels stiffened by either extruded or non-extruded angle-bar profiles
(proposed by the Committee III.1 Ultimate Strength of ISSC2003). These models were made of
Aluminium alloy 6082-T6, by different arrangements of heat-affected zone (HAZ). The main outcomes of
this study confirmed the need for a subtle identification of the real shapes of the initial deformations,
especially in structures like catamarans and ferries that are designed to operate in severe sea conditions. In
a continuation study, Pedram et al. (2014) numerically investigated the effects of welding on the strength
of the aluminium stiffened plates under different combined uniaxial compression and lateral pressures. In
this study, the softening of the heat affected zone (HAZ) and the residual stress were taken into
consideration for all FE models. Based on the extensive finite-element investigations, different aspects of
the effect of welded induced initial imperfections on aluminium panels were outlined and some designoriented conclusions were drawn to which may make reference.
4.7.3
Formulation Development
Empirical formula and ANN model
Collette (2011) developed a series of rapid semi-analytical methods for predicting the compressive and
tensile response of aluminium plates and stiffened panels. These methods were designed to allow Smithtype progressive collapse approaches to be implemented for aluminium vessels. Unlike existing steel
ultimate strength methodologies, particular attention was paid to capturing aluminium-specific response
features, such as round material stress-strain curves and the weakening effect of fusion welds. In limited
comparison of the results to FEA studies and published experiments the methods generally performed
well, and represented a rapid aluminium-specific approach to ultimate strength calculations.
1
0.8
0.6
0.4
ANN model
Present empirical formula
0.2
0.2
0.4
0.6
FEM analysis
0.8
Figure 20. Comparison of the non-dimensional ultimate strength values obtained using the FEM versus those values
predicted using the present empirical formula and the ANN model
4.7.4
Experimental Investigation
An experimental and nonlinear FEA-based numerical study on buckling collapse of a fusion-welded

aluminium stiffened plate structure was carried out by Paik et al. (2012c). A set of aluminium stiffened
plate structures fabricated via gas metal arc welding in which the test structure was equivalent to a full
scale deck of an 80m long high speed vessel. The plate part of the test structure was made of 5383-H116
aluminium alloy, and extruded stiffeners were made of 5083-H112 aluminium alloy. A good agreement
was found between the nonlinear FEM computations and the experimental results. It was concluded that
the nonlinear FEM computations significantly depend on the structural modelling technique applied. In
particular, the welding-induced initial imperfections in terms of initial distortions, residual stresses, and
softening in the heat-affected zone need to be modelled as appropriate for the nonlinear FEA of welded
aluminium structures.
4.7.5
321
Fiber-Reinforced Polymer Strengthened
Zahurul Islam et al. (2011) presented an experimental investigation of fibre-reinforced polymer (FRP)
strengthened aluminium square and rectangular hollow sections experiencing web crippling failure due to
localised concentrated loads or reactions.
Figure 21. Test setup of End-Two-Flange loading condition
The work was mainly focused on the effects of different adhesive and FRP for the strengthening of 6061T6 heat-treated aluminium alloy tubular sections against web crippling (Figure 21).
The influence of different widths of FRP plate strengthening against web crippling subjected to endtwo-flange and an interior-two-flange loading condition was also investigated in this study. In general, the
test results revealed that the FRP strengthening significantly enhances the web buckling capacity of
aluminium tubular sections, especially for those sections with large value of web slenderness ratio. It was
also shown that the increases of FRP width do not provide much improvement in the strengthening of the
aluminium sections. Hence, the FRP strengthening against web crippling was found to be effective when
the FRP width is having the same width as the bearing length for aluminium tubular sections.
4.7.6
Sandwich Panels
There is significant and growing interest in using aluminium sandwich panels in the shipbuilding
industry, as an attractive and interesting solution to the increasing environmental demands. Crupi et al
(2013) made a comparison of static and impact response for aluminium foam and honeycomb sandwich
constructions. A combined experimental and theoretical approach was applied to investigate the static and
impact response. The obtained results confirmed that the capacity of energy dissipation under bending
loading is affected by the collapse mechanism, and also by the face-core bonding and the cell size for
foam and honeycomb panels, respectively.
4.7.7
Hull Girder
Benson et al. (2013d) completed nonlinear FE analysis to determine the ultimate strength response of
several example aluminium ship-type structures under vertical bending moment. Firstly a series of four
square box girders were analysed, with scantlings similar to equivalent large aluminium ships. A nominal
aluminium multihull midsection was also analysed. The FE models were extended over multiple frame
spaces to capture both interframe and overall grillage buckling effects. The FE analyses were compared to
results from an extended Smith-type progressive collapse method, which has the ability to predict hull
girder ultimate strength including both interframe and overall grillage collapse. Additional nonlinear FE
analyses of the flat panels making up the box girders are presented by Benson et al. (2013c). These are
compared to results from the adapted large deflection orthotropic plate method, which is the approach
used to develop the panel load shortening curve including overall buckling effects. The results from both
papers show relatively close agreement between analysis methods and demonstrated a significant effect
from overall buckling in reducing the ultimate strength of the section when compared to equivalent
interframe results.
The effect of damage severity and location on the residual strength of a generic aluminium hull
structure with scantlings typical of a patrol vessel are investigated by Taylor et al. (2013) using ISFEM.
Different damage scenarios as defined in classification guidance by DNV and Lloyds Register are
considered.
322
Magoga et al. (2014) presented ALPS/HULL ISFEM (intelligent supersize finite element method)
results of ultimate hull girder strength analysis under vertical and horizontal bending applied
independently. The aim was to quantify the degree to which weld-induced imperfections (i.e. initial plate
and stiffener deflection, plate and stiffener residual stress, HAZ breadth and HAZ yield stress softening)
reduce ultimate global bending load capacity of a patrol boat midship section and to determine if certain
imperfection parameters contribute more to the reduction than others. This investigation was based on a
one-bay hull girder ISFEM model containing only longitudinal stiffener fillet welds made by metal inert
gas welding process. All longitudinal stiffeners and girders included in the model are T-section profiles
made of aluminium alloys 6351-T6, 5083-H116 and 6082-T6. The obtained results of the analysis cases
illustrated that severe levels of HAZ-related imperfections (includes yield stress reduction, tensile residual
stress and HAZ breadth) and plate initial deflection are the most influential parameters in the reduction of
the ultimate strength of the aluminium patrol boat studied. The effect of plate and stiffener compressive
residual stress and stiffener column type initial deflection were found to be relatively small. The results
also showed that the ultimate strength of the patrol boat studied is more affected by imperfections under
sagging loads compared with hogging and horizontal bending loads.
4.7.8
Summary and Recommendation for Future Works
In summary, based on the literature review, it seems the research studies on the buckling/ultimate strength
of aluminium marine structures are rather limited. There is a growing continuation for using aluminium
alloy structures in marine applications. In particular, such a trend can be seen in the case of large highspeed crafts operating in deep ocean environments for which more rigorous ultimate strength baseddesign techniques and methodologies are required.
Following the studies made to date, attention should still be given to more numerically and
experimentally investigate the influences of alloy type, material nonlinearity, welding location along with
its parameters, HAZ characteristics and weld-induced imperfections on the ultimate strength of
aluminium plate and stiffened panels as well as ultimate global bending load capacity of aluminium
midship sections subjected to different types of loadings and its combinations. By such a contribution,
much more reliable buckling and ultimate strength formulations are developed which can, in turn, provide
a more suitable basis for modifications to aluminium design standards.
Other point of interest can be the study on the effects of some in-service damages like openings and
cracks which must be diagnosed accurately in security and safety considerations. Also, it would be
interesting to more investigate the structural behaviour of the structures consist of extruded integral
stiffened panels like deck and superstructure elements. Furthermore, it should be noted that due to
increasing environmental demands, more study on aluminium sandwich structures used in maritime
industries can be interesting research topic to carry out.
5.
5.1
5.1.1
BENCHMARK STUDY
Small box girder
Introduction
The purpose of this benchmark study is to demonstrate the level of uncertainty in an ultimate strength
calculation using NLFEM due to the effect of various physical parameters which are necessary for
nonlinear analysis. These parameters include the initial geometric imperfection definition, the material
plasticity model, the boundary conditions used to restrain the model, the mesh density, the element type
and the choice of finite element solver. The study also makes an attempt to quantify the magnitude of
uncertainty in the finite element solution which could be expected in the ultimate strength prediction of a
box girder using NLFEM. This is done by making a statistical evaluation of all the baseline results from
the benchmark study participants and producing a confidence band above and below an averaged
bending-moment curvature plot.
In ISSC Committee III.1 2012 it was concluded that results from ultimate strength analyses can still
be affected by large uncertainties that must at least be identified and then possibly estimated (Paik et al.,
2012b). The 2012 ISSC report divides the source of uncertainties into physical aspects (such as the
uncertainty in defining the material properties or imperfections) and modelling aspects (uncertainty in the
ability of the numerical model to adequately approximate the physics of a structural collapse).
Confidence in NLFEM for ultimate strength analysis comes from having a detailed understanding of
both these sources of uncertainty. There is not much information currently available to properly quantify
the expected uncertainty in the results from a particular ultimate strength assessment. Classification rules
do provide some guidance; for example IACS CSR uses a partial safety factor of 1.1 to account for both
modelling and physical uncertainties in an ultimate strength calculation. However, previous benchmark
studies including those by ISSC have shown larger differences between analyses. For example, results
323
from the hull girder vertical bending moment benchmark studies in ISSC 2012 showed differences in
ultimate strength predictions of over 30% for some of the ship hulls tested. This disparity indicates a
current need for research to better identify and quantify the influence of different modelling and physical
parameters on an ultimate strength analysis using NLFEM.
A common theme within many of the papers reviewed in this report is that the successful use of
NLFEM still relies on analysts having a high degree of expertise in constructing the finite element model,
sound engineering judgement to evaluate the uncertainty of the results and knowledge regarding the
limitations for a given finite element program. Analysts may try to gain confidence in their NLFEM
approach by comparing their results to a benchmark either from a previous numerical analysis or
physical experiment. For example, a well-used experimental test is the 1/3 scale frigate model tested by
the UK Admiralty Research Establishment (Dow, 1991). Previous benchmark studies by ISSC Ultimate
Strength committees have also provided excellent sources of numerical benchmarks for plates, stiffened
panels and hull girders. However, any comparison made by an analyst with a previous study can also be
clouded by the modelling and physical uncertainties within the original test results, which are often
poorly understood and documented. If it is unclear what the uncertainty is within an existing ultimate
strength result for a ship, it is difficult to judge the comparative success of a new result.
To produce a reliable result, ultimate strength analyses using NLFEM frequently need to quantify
physical uncertainties for inclusion. An analyst often makes recourse to a statistical average to
adequately represent the actual structure within the numerical model for example by specifying
average imperfection levels. The influence of any parameter can then be tested by running a parametric
study carefully varying the parameter within certain bounds and evaluating its effect on the
ultimate strength.
A common example which is often tested is the magnitude and shape of the geometric imperfections
within a plate or stiffener. It is normally essential to include some form of geometric imperfection within
the NLFEM mesh to enable the formation and nucleation of buckling, which often precipitates the
collapse of the structure and hence governs the ultimate strength. However, the exact shape of the
geometric imperfection for a given structure is rarely known and, as shown by numerous studies, the
shape and magnitude of the imperfection is known to affect the ultimate strength behaviour to some
extent. Therefore the analyst must make a proper consideration of the imperfection to ensure the resulting
ultimate strength prediction is reasonable.
Modelling uncertainties are much harder to estimate, specifically because the analyst will not normally
know what the real result should be and so the magnitude of the error in any calculation method is
impossible to realize. This is often true even when comparing with a physical experiment because the
modelling uncertainty of the test setup is often unknown and, especially for large-scale hull girder type
experiments, repeat tests are rare.
Efforts to improve the quantification of both physical and modelling uncertainties will lead to
improved confidence in ultimate strength analysis using NLFEM. This benchmark study hopes to
contribute progress in this regard.
5.1.2
Model Parameters
The benchmark study replicates physical experiments completed by Gordo et al. (2009) on three box
girders built from high tensile steel. These were tested to collapse under pure bending moment using a 4point bending rig. A schematic of the test rig is shown in Figure 22.
Each model has a cross section of 600 mm 800 mm (Figure 22). The length varies according to the
number of spans and the spacing between frames, and is 0.8 m, 0.9 m or 1.2 m (These are the lengths
considered within the simulations). All Box girders and stiffeners are modeled with 4 mm thickness. The
transverse frames are L stiffeners L50 28 6 made of mild steel. The longitudinal stiffeners are
continuous through the models.
150
150
150
100
200
150
400
600
100
400
Figure 22. Schematic of the test rig used for the experimental tests of the three box girders (left) and cross section of
box girders (right). All plate and stiffeners are 4mm thick. Stiffeners are 20mm height (Gordo and
Guedes Soares, 2009).
324
The girders have a relatively small and simple cross section, which enabled the initial specification for the
baseline studies attempted by all participants to be carefully defined. Typical meshes are of the order of
10,000 elements, which means solutions are reached in a relatively short time. The simple dimensions of
the box enables the geometric imperfection to be defined in the mesh using either Eigenvalue buckling
mode shapes or direct translation of the nodes. The specification is available separately for those who
wish to validate or practice their own NLFEM approach.
Baseline calculations for all three box girders were attempted by all participants. Sub studies to
investigate the parametric influence of different characteristics were then completed by different
participants using the H300 box girder.
5.1.3
Baseline Calculations
The baseline calculations were completed by all participants using an issued specification which
described the model geometry, the material yield strength and plasticity characteristics, initial
imperfection characteristics and required boundary conditions. The full specification is available
separately. A summary of the key characteristics is given here.
Geometric imperfection consists of a superposition of 3 different mode shapes: plate imperfection,
stiffener lateral imperfection and column imperfection. The imperfection shapes and amplitudes have
been chosen to represent an average level of imperfection typically found on ship structures as defined
in several previous studies (Smith et al., 1991). The mode shapes have been chosen to be relatively
conservative whilst retaining some realism. For example the plate imperfection shapes are assumed to
alternate direction, rather than form a hungry horse pattern. The plate imperfection is defined using a two
mode Fourier series in calculations.
w
w0 pl
x
mx y
= 0.8sin + 0.2sin
.sin
b
a
a
(11)
where w is the imperfection amplitude at a given coordinate position (x, y) from a plate corner. Each plate
has length a and breadth b. The maximum imperfection amplitude, wopl, is 0.4 mm for all plates.
The stiffener lateral-torsional imperfection describes the tilt of the stiffener from upright and is
defined using a single half wave between frames:
vs =
w
x
sin vos
hw a
(12)
where vs is the imperfection amplitude at a given coordinate position (x, w) from a stiffener-frame
intersection. The stiffener has a total web height hw of 20mm in this study. For the baseline studies the
maximum imperfection amplitude, vos, is specified as 0.015 hw = 1.5 mm for all stiffeners. The tilt
direction runs consistent with the corresponding plate imperfection (see Figure 22).
The column imperfection is a global mode shape and describes the distortion of the plate and
stiffeners as a single unit. The shape is applied to both plate and stiffener nodes using the
following equation:
w
x
(13)
vs = sin vos
hw a
where A is the length between frames (200,300 or 400 mm) and B is the breadth of the panel (600mm or
800 mm). The column imperfection is directed outward in the central bay of the box girder (see
Figure 22) with maximum magnitude, wc, of 1.5 mm.
Figure 22. Magnified column imperfection (left), plate imperfection and stiffener lateral Imperfection (right).
Magnification 20
325
A bilinear material model is used with E = 210 GPa, o = 732 MPa and a 1% post yield hardening
(engineering stress-strain curve). The stress-strain values for use in the FEM plasticity model are as
follows:
Table 5. Material Properties for HTS690 as used for numerical analyses.
enom
Yield
Ultimate
3.49E03
3.97E02
nom
(MPa)
7.32E+08
8.08E+08
e true
true (MPa)
eplastic
0.00348
0.038909
7.35E+08
8.40E+08
0.00E+00
3.49E02
The specification asked for the model extents to only include the test section as used in the original
experiments. A pure vertical bending moment is applied through rigid body constraints at each end of the
girder. Benchmark participants were free to set the exact boundary conditions appropriate to their own
modelling approach and the particular requirements of the software package used. A typical approach
uses a reference node to constrain all nodes at each end using a rigid body constraint. Incremental load or
rotation is then applied through the reference node. The resulting bending moment and rotation results
can be outputted directly from the reference node. Curvature is measured using the total relative rotation
between each end of the box girder for convenience and to ensure standardized results. The best method
to use for measurement of curvature is debatable to compare accurately with the experiment. A recent
paper by Gordo et al. (2014) discussing similar box girder tests to those used in this benchmark study
demonstrates the difference between measuring curvature from the ends of the model and measuring
curvature from adjacent frames to the nucleated buckling.
The rigid body boundary condition typically used in ANSYS, ABAQUS and other software packages
does not require the reference node to be placed on the neutral axis of the section. The reference node
position is irrelevant because it is used purely to control the rotation at the box girder end.
Participants were free to choose software package and solver type used, as shown in Table 6.
All participants were asked to conduct a mesh refinement study to determine an appropriate
characteristic element length which gives a converged prediction of the progressive collapse curve. An
example convergence study is shown in Figure 23. Participants typically found a characteristic length of
10mm to be sufficient. Some participants further refined the element length in the stiffeners to give an
increased number of elements through the depth. The mesh should have at least two elements in the web
height. A mesh with 1 element in the web height gives a conservative result.
Table 6. Summary of participants in benchmark study
Participant
1
2
3
4
5
6
7
8
9
9
10
FEM Package
Trident / VAST 9.1
ANSYS Workbench v11.0
ABAQUS v6.13
ANSYS APDL v14.5
ANSYS APDL v14.5
LS-Dyna v971d r 6.0.0
ABAQUS v6.13
NX-NASTRAN
ABAQUS v6.13
ANSYS APDL 14.5 and 15.0
Smith Method (HullColl )
Solver
Orthogonal Trajectory
Newton-Raphson
Arc Length (Riks)
Newton-Raphson
Newton Raphson
Dynamic-Explicit
Arc Length (Riks)
Newton-Raphson
Newton-Raphson
Arc Length
N/A
Element Type
Shell
Shell181
Shell S4R
Shell181
Shell181
Shell ELFORM2
Shell S4R
Shell
Shell S4R
Solid 95
N/A
Participants 1-9 used shell elements in the baseline calculations. Results for the three box girders are
presented in Figure 24 to Figure 26. A mean progressive collapse curve is also presented for each box
girder together with an upper and lower curve one standard deviation away from the mean. These are
provided to give some indication as to the statistical uncertainty of the numerical analyses.
The mean curve is formulated using a point by point averaging method. At each specified interval of
curvature (increment of 0.0002m-1), the resultant bending moment from each participants result is used to
calculate a mean BM. Because the raw data from the finite element analyses is not given with fixed
curvature increments, all the datasets were first made consistent using a standard curvature increment of
0.0002m-1 and applying linear interpolation. The standard deviation curves are also calculated on a point
by point averaging basis.
The differences between the numerical results are relatively minor. A comparison of the predicted
ultimate strengths between participants is shown in Table 7. It should be noted that the peak of the mean
326
1.2E+6
1.2E+6
1.0E+6
1.0E+6
8.0E+5
6.0E+5
H300 - Mesh Refinement - Participant 1
20mm element length
15mm element length
10mm element length
7mm element length
4.0E+5
2.0E+5
0.01
0.02
Curvature (1/m)
Bending Moment (Nm)
Bending Moment (Nm)
curves from the figures differ slightly from the mean ultimate strength reported in the table. This is
because the peak of each curve does not occur at the same curvature value. The point by point averaging
method will not distinguish the peak. Which is the most statistically significant average ultimate
strength from these results remains questionable.
Also, it should be noted that the experiment results from Gordo et al. (Gordo and Guedes Soares, 2009)
have been adjusted to fit the initial stiffness of the FEM models. The difference is due to difficulties in
accurately measuring rotation.
8.0E+5
6.0E+5
H300 - Mesh Refinement - Participant 5
20mm element length
15mm element length
10mm element length
7mm element length
4.0E+5
2.0E+5
0
0.03
0.01
0.02
Curvature (1/m)
0.03
1.6E+6
1.6E+6
1.4E+6
1.4E+6
Bending Moment (Nm)
Bending Moment (Nm)
Figure 23. Mesh refinement study by participant 1 (left) and participant 5 (right).
1.2E+6
1.0E+6
8.0E+5
6.0E+5
H200 Baseline Results

Exp.
Part.1
Part.3
Part.2
Part.4
Part.5
Part.6
Part.7
Part.9
4.0E+5
2.0E+5
0.01
0.02
Curvature (1/m)
1.2E+6
Max BM = 1.132MNm
1.0E+6
8.0E+5
6.0E+5
H200 Mean Line Numerical Result
4.0E+5
Exp.
Mean Numerical Result
Mean + st.dev
Mean - st.dev
2.0E+5
0
0.03
0.01
0.02
Curvature (1/m)
0.03
1.4E+6
1.4E+6
1.2E+6
1.2E+6
Bending Moment (Nm)
Bending Moment (Nm)
Figure 24. H200 bending moment curvature baseline results from all participants (left) and mean numerical results
(right)
1.0E+6
8.0E+5
6.0E+5
4.0E+5
2.0E+5
0.0
0.0

Exp.
Part.1
Part.2
Part.3
Part.4
Part.5
Part.6
Part.7
Part.9
Part.8
0.01
0.02
Curvature (1/m)
1.0E+6
Max BM = 0.955MNm
8.0E+5
6.0E+5
4.0E+5
Exp.
Mean + st.dev
Mean - st.dev
2.0E+5
0.03
0.0
0.0
0.01
0.02
0.03
Curvature (1/m)
Figure 25. H300 bending moment curvature baseline results from all participants using shell elements (left) and
mean numerical results (right).
327
1.2E+6
1.2E+6
1.0E+6
1.0E+6
Bending Moment (Nm)
Bending Moment (Nm)
8.0E+5
6.0E+5
4.0E+5
Exp.
Part.2
Part.4
Part.6
Part.8
2.0E+5
0.0
0.0
0.01
Part.1
Part.3
Part.5
Part.7
Part.9
0.02
Curvature (1/m)
Max BM = 0.817MNm
8.0E+5
6.0E+5
4.0E+5
Exp.
Mean + st.dev
Mean - st.dev
2.0E+5
0.0
0.03
0.0
0.01
0.02
0.03
Curvature (1/m)
Figure 26. H400 bending moment curvature baseline results from all participants (left) and mean numerical results
(right).
The most notable finding of the baseline study is that numerical results using shell elements predict a
significantly lower strength than the experimental test for all three girders. This suggests that one or more
parameters from the specification of the experiment differ from the reality. The additional parametric
studies investigating the influence of geometric imperfection, material model and plate thickness in the
sub sections below give some indication as to where differences between the NLFEM and experiment
may produce such disparity. Another possible reason for the large differences between NLFEM and
experiment, which is not tested here, is differences in the boundary conditions and load application. A
previous study by Benson et al. (2013a) showed that friction between the strips connecting the load to the
four point bending rig could inhibit buckling in the top flange of the box and increase the ultimate
strength capacity of the girder. However, it is not possible to verify whether this is indeed a cause of the
discrepancy between results.
The mean numerical results with standard deviation curves above and below show the approximate
bounds of a 95% confidence level between datasets, assuming a normal distribution about the mean curve
and a population sample of 8. Although the sample size is small this provides some clarity in judging the
correlation between different participants. At the curve peak the uncertainty is between 3.3% (H200) and
10.6% (H400). This is equivalent to the uncertainty suggested by the IACS CSR partial safety factor and
suggests that the safety factor is reasonable. It must be pointed out that this conclusion is only inferred
from a relatively small cross section box girder.
Table 7. Ultimate strength results of baseline box girder benchmark analyses.
Experiment
Participant 1
Participant 2
Participant 3
Participant 4
Participant 5
Participant 6
Participant 7
Participant 8
Participant 9
Participant 10 (Solid elements)
Participant 11 (Smith method)
Ult. Str. H200

(MN.m)
1.527
1.139
1.134
1.139
1.119
1.154
1.170
1.134
1.120
1.651
1.291
Ult. Str. H300

(MN.m)
1.269
0.958
0.947
0.944
0.956
0.969
1.015
0.961
1.011
0.958
1.175
Ult. Str. H400

(MN.m)
1.026
0.863
0.835
0.823
0.783
0.844
0.880
0.836
0.850
0.824
1.091
0.913
1.139
0.017
0.966
0.029
0.838
0.028
Ave. Ult. Str. (shell elements only)

St. Dev. (shell elements only)
5.1.4
Comparison with Solid Element Mesh
Participant 10 investigated the use of solid elements for the analysis. The cases H200 and H400 were
studied using meshes with an average element length of 30 40 mm, giving a total number of elements of
some 85000 (both tetrahedral and cubic). The resulting aspect ratio for elements in the plate parts is high,
328
but believed to have a minor influence on the results as quadratic elements were used. The current meshes
used 12 elements over the height on the webs of the stiffeners. Differences compared with the parallel
simulations using shell elements were the use of only one imperfection mode, here plate imperfection, the
use of a mild steel, with yield stress 250 MPa in the transverse frames (as indicated in the report from the
manufacturing of the box beams used for the experiments) and also the use of a trilinear stress-strain
curve for both materials. The simulations were performed using a prescribed rotation of one end section
or using the arc-length method, though in the latter cases it was found sometimes difficult to obtain
convergent results.
Figure 27 (left) shows the moment curvature relationship for the H200 Box girder. It is seen that the
maximum moment is overestimated using a bilinear stress strain curve, whereas a trilinear stress strain
curve will give a lower maximum bending moment. The material used for the transverse stiffeners is seen
to have a small effect. Figure 27 (right) shows the influence of material model for the H400 box girder
with transverse stiffeners modelled as mild steel. Uses of a more realistic trilinear material model will
also here give a lower maximum bending moment.
Compared to the use of shell elements, use of solid elements slightly overestimate the maximum
moment. It has not been possible to quantifiably explain this difference in results, or the difference seen in
elastic stiffness for the solid element meshes compared to the experimentally obtained elastic stiffness.
Note though that contrary to the shell element simulations, no adjustment of the elastic stiffness to match
the experimental results were made.
1.8E+6
1.2E+6
Bending Moment (Nm)
Bending Moment (Nm)
1.6E+6
1.4E+6
1.2E+6
1.0E+6
8.0E+5
H200 Solid Elements
6.0E+5
Experiment
4.0E+5
Trilinear MS transverse frames

Trilinear HTS transverse frames
2.0E+5
0.0
1.0E+6
8.0E+5
6.0E+5
H400 Solid Elements
4.0E+5
Experiment
Trilinear MS transverse frames
2.0E+5
Bilinear MS transverse frames
0.0
0.02
0.01
Curvature (1/m)
Bilinear MS transverse frames
0.0
0.03
0.0
0.02
0.01
Curvature (1/m)
0.03
Figure 27. Bending moment curvature results for participant 10 using solid elements, H200 (left) and H400 (right).
5.1.5
Comparison with Smith method
1.6E+6
1.4E+6
1.4E+6
1.2E+6
1.2E+6
1.0E+6
8.0E+5
6.0E+5
H200 Comparison with Smith Method
4.0E+5
Experiment
2.0E+5

(Shell Elements)
Smith Method
Bending Moment (Nm)
Bending Moment (Nm)
Participant 11 completed a Smith method progressive collapse calculation for each box girder using the
same geometry and material information as for the finite element experiments. Hard corners are assumed
at the corners of the box girder extending 60mm in both directions. The panel load shortening curves are
developed using the method described by Gordo et al. (1996).
1.0E+6
8.0E+5
6.0E+5
4.0E+5
2.0E+5
0.0
0.01
0.02
Curvature (1/m)
0.03
0.0
H300 Comparison with Smith Method
Experiment
(Shell Elements)
Smith Method
0.01
0.02
0.03
Curvature (1/m)
Figure 28. Bending moment-curvature results from Smith method. Comparison with experiment and mean curve
shell of finite element analysis. H200 (left) and H300 (right)
329
The resulting bending moment-curvature plots for are shown in Figure 28 and compared to the mean
numerical result from shell finite element analysis and the original experimental result. The results show
that the Smith method gives an optimistic prediction of ultimate strength in comparison to NLFEM for
these particular cases. The small scale of the box girders makes the Smith method highly sensitive to the
definition of the panel load shortening curve for the top flange of the box. It is also sensitive to the extent
of the hard corners. This sensitivity is more pronounced than found for a ship sized model. It suggests that
Smith method is less appropriate for use on very small cross section.
5.1.6
Effect of Imperfection Amplitude and Shape
1.0E+6
1.0E+6
9.0E+5
9.0E+5
8.0E+5
7.0E+5
6.0E+5
5.0E+5
4.0E+5
3.0E+5
2.0E+5
1.0E+5
0.0
0.0
H300 Plate lmp. Amplitude

Participant 1
wopl=0.0mm
wopl=0.4mm(baseline)
wopl=0.8mm
wopl=1.2mm
wopl=2.0mm
wopl=3.0mm
0.01
0.02
Curvature (1/m)
0.03
Bending Moment (Nm)
Bending Moment (Nm)
Several participants analyzed the effect of changing the parameters governing the geometric
imperfections. No information was available to describe the exact imperfections in the experimental
model, although it is noted in the experiment reports that the panels were relatively flat with minimal
level of imperfection amplitudes. Therefore all imperfection parameters used in this benchmark study are
nominal and based on statistical levels appropriate to ship type plating.
Participant 1 investigated the effect of parametrically changing the plate imperfection and column
imperfection independently. The other imperfection parameters are kept at baseline amplitudes in each
parametric study. The results are shown in Figure 29. This demonstrates insensitivity to both plate and
column imperfection amplitude.
8.0E+5
7.0E+5
6.0E+5
5.0E+5
H300 Column lmp. Amplitude

Participant 1
wopl=0.0mm
wopl=0.5mm
wopl=1.5mm(baseline)
wopl=1.0mm
wopl=3.0mm
wopl=5.0mm
4.0E+5
3.0E+5
2.0E+5
1.0E+5
0.0
0.0
0.01
0.02
Curvature (1/m)
0.03
Figure 29. Effect of Imperfection Amplitude (participant 1). Parametric change of plate imperfection (left) and
column imperfection (right)
Participant 4 also investigated the effect of changing the imperfection amplitude for the H300 box girder.
The study considered different amplitudes of the stiffener lateral-flexural imperfection. Plate imperfection
was kept at constant amplitude of 4mm. The initial deflection was applied in two directions, as shown in
Figure 30. Three values for amplitudes of this initial deflection have been used:
vos = 0.00025a = 0.075mm

vos = 0.0015a = 0.45mm
vos = 0.005a = 1.5mm (baseline imperfection)
where a is the distance between frames. In addition a case with no initial imperfection was completed.
All other parameters, such as the material model and mesh size, used baseline settings.
Figure 30. H300 with scaled initial imperfection, +ve up (left), ve down (right)
330
Figure 31 shows the bending momentcurvature plots for the variant models. With values of initial
deflection 0.00025a and 0.0015a almost no difference is detected between positive and negative initial
deflection models. However, when the maximum initial deflection value (0.005span) has been
considered, negative initial deflection model shows about 6% larger ultimate strength than positive initial
deflection model.
Considering the first two values for the initial deflection amplitude (0.00025a and 0.0015a) reduces
the ultimate strength 2.86% and 4.29% respectively. However, initial deflection increases it 2.72%
considering the maximum value (0.005a), all in positive case. In negative case, it reduces it 3.12% and
4.98% for 0.00025a and 0.0015a respectively and only 3.1% for 0.005a.
All considered values of initial deflection have not more than 5% influences on the ultimate strength.
This indicates that the box girder is relatively insensitive to the magnitude of imperfection applied.
1.0E+6
1.4E+6
8.0E+5
Bending Moment (Nm)
Bending Moment (Nm)
9.0E+5
7.0E+5
6.0E+5
5.0E+5
H300 Imperfection Amplitude

Participant 4
No Initial Deflection
neg. 0.00025span
neg. 0.0015span
neg. 0.005span
pos. 0.00025span
pos. 0.0015span
pos. 0.005span
4.0E+5
3.0E+5
2.0E+5
1.0E+5
1.2E+6
1.0E+6
8.0E+5
6.0E+5
4.0E+5
2.0E+5
0.0
0.0
0.01
0.02
Curvature (1/m)
0.0
0.03
H300 Imperfection Shape Participant 6

Experiment
Baseline
Case 3-1
Case 3-2
Case A-1
0.01
0.0
0.02
0.03
Curvature (1/m)
Figure 31. Effect of Imperfection Amplitude

(participant 4)
Figure 32. Effect of Imperfection Shape

(participant 6)
Participant 6 investigated the effect of changing each imperfection parameter whilst keeping the
remaining imperfections at the baseline level. The ultimate strength of the box girder for 9 cases is
summarized in Table 8. The study found the model relatively insensitive to all three imperfection
components. The largest effect was found on altering the column imperfection. With zero column
imperfection amplitude, the ultimate strength approached the zero imperfection value, which is about
7.7% higher than the baseline result. However, increasing the imperfection magnitude in any parameter
caused a relatively insignificant reduction on the ultimate strength and in some instances actually
increased it.
Participant 6 investigated the effect of different imperfection shapes and directions in the plating. The
plate imperfection lobes alternate direction in the baseline model. In case 3-1 and case 3-2 the plate
imperfection is changed to a hungry horse type pattern with all lobes directed inwards (3-1) or outward
(3-2). For Case A-1 the plate imperfection is modelled with two half waves along the length. The load
shortening curves (Figure 32) show minimal effect on the ultimate strength and the mode of collapse.
Table 8. Ultimate strength of H300 box girder with different imperfection amplitudes (Participant 6).
Plate
wopl [mm]
Baseline
CASE 2-1
CASE 2-2
CASE 2-3
CASE 2-4
CASE 2-5
CASE 2-6
CASE 2-7
CASE 2-8
CASE 2-9
Zero Imp.
0.4
LateralTorsional
vos [mm]
Column woc [mm]

Central
bay
End
bays
1.5
0
0.75
3.0
1.5
0.375
1.5
0.375
1.5
0.375
0
0.75
3.0
0
0
0.1875
0.8
0
LateralTorsional
0.4
Plate
0
0.2
0.8
1.5
Column
0.4
1.5
Ult.
Strength
(N.m)
1015800
1016300
1015300
1015800
1016400
1005900
1018000
1093500
1005500
1016700
1094400
Difference to
baseline (%)
N/A
0.05
-0.05
0.00
0.06
-0.97
0.22
7.65
-1.01
0.09
7.74
331
Based on these studies, we can conclude that including some imperfection in the model is important as
the resulting ultimate strength can be over predicted by up to 8%. The most important parameter is the
column imperfection. However, so long as some imperfection is present, the amplitude has a relatively
insignificant effect on the ultimate strength. The column imperfection is the most significant parameter
with influence on the ultimate strength behaviour.
5.1.7
Effect of Material Model Parameters
1.4E+6
1.4E+6
1.2E+6
1.2E+6
Bending Moment (Nm)
Bending Moment (Nm)
A parametric study to show the effect of different yield strengths was completed by participant 3 and
participant 5. The extent of the yield strength change was kept within realistic bounds of +-50MPa, which
could be the expected variation in actual material properties compared to the nominal yield of 732MPa.
Both studies (Figure 33) showed similar results, with the change in yield having only a moderate effect on
the overall ultimate strength and no effect on the mode of collapse. Participant 5 completed additional
parametric studies to compare the effect of different strain hardening models (Figure 34). Both studies
show a minor effect on ultimate strength and no effect on the mode of collapse.
1.0E+6
8.0E+5
6.0E+5
H300 Effect of Yield Stress

Participant 3
Experiment
= 732MPa(baseline)
= 690MPa
= 770MPa
4.0E+5
2.0E+5
0.0
0.01
0.0
0.02
1.0E+6
8.0E+5
6.0E+5
4.0E+5
2.0E+5
0.0
0.03
0.0
H300 Effect of Yield Stress

Participant 5
Experiment
= 732MPa(baseline)
= 695MPa
= 769MPa
0.01
0.02
0.03
Curvature (1/m)
Curvature (1/m)
1.4E+6
1.4E+6
1.2E+6
1.2E+6
1.0E+6
8.0E+5
6.0E+5
4.0E+5
2.0E+5
0.0
0.0
H300 Youngs Modulus Participant 5

Experime
nt
E=189GPa
E=200GPa
E=210GPa
E=221GPa
E=231GPa
0.01
0.02
Curvature (1/m)
0.03
Bending Moment (Nm)
Bending Moment (Nm)
Figure 33. Effect of Material Yield Strength Participant 3 (left) and Participant 5 (right)
1.0E+6
8.0E+5
6.0E+5
4.0E+5
2.0E+5
0.0
0.0
H300 Material Properties

Participant 6
Experiment
Baseline
Case B-1
Case B-2
0.01
0.02
0.03
Curvature (1/m)
Figure 34. Effect of Material Youngs Modulus (Participant 5) (left) and material properties (Participant 6) (right).
5.1.8
Effect of Plating Thickness
Participants 3, 5 and 11 investigated the effect of varying the plating thickness for the H300 box girder.
The load shortening curves from participants 3 and 5 are plotted in Figure 35. The results indicate that
changes in plate thickness have a significant effect on the ultimate strength of the box girder, but that the
mode of collapse remains the same. Results from all participants show a similar relationship between
ultimate strength and plate thickness (Figure 36). The load shortening curves also show that increasing
the plate thickness causes a noticeable increase in the stiffness of the box girder. This is as expected
because the radius of gyration increases with thicker plate. The results from participant 11, using the
332
Smith method, reflect the higher ultimate strength prediction found by using a simplified progressive
collapse method.
The relationship between ultimate strength and plate thickness is linear for the range of thicknesses
tested (3.8mm to 5.0mm), with a gradient of approximately 40%. For example, a 25% change in plate
thickness from 4mm to 5mm produces a 40% change in the ultimate strength. This means that, for small
scale cross sections, a relatively slight difference in plate thickness has a significant effect on the ultimate
strength.
Nominal plating thickness as stated by the steel supplier may differ from the actual measured thickness.
These experiments demonstrate the need for accurate measurements of plate thickness for a given cross
section to enable an accurate representation in the numerical model.
1.6E+6
1.2E+6
1.0E+6
8.0E+5
6.0E+5
4.0E+5
2.0E+5
0.0
0.0
H300 Plate Thickness Participant 3

Experiment
t=4.0mm(baseline)
t=3.8mm
t=4.2mm
t=4.4mm
t=5.0mm
0.01
0.02
0.03
Curvature (1/m)
Figure 35. Effect of changing the plating thickness

(participant 3).
5.1.9
Bending Moment (Nm)
Bending Moment (Nm)
1.4E+6
H300 Plate Thickness

Participant 3, 5 and 11
1.5E+6
1.4E+6
1.3E+6
Participant 3
Participant 5
Participant 11
1.2E+6
1.1E+6
1.0E+6
9.0E+5
8.0E+5
3.5
4.0
4.5
5.0
Plate Thickness (mm)
5.5
Figure 36. Relationship between plating thickness

and ultimate strength
Summary/Conclusions
The box girder benchmark study demonstrates that, given a relatively strict problem definition, a set of
experienced users of commercial nonlinear finite element analysis packages will produce consistent
results for a small scale ultimate strength problem. The range of ultimate strengths predicted within the
baseline calculations for the benchmark fall within the IACS CSR partial safety factor for strength
prediction uncertainty (m = 1.1). This provides good validation of the partial safety factor value, although
this conclusion is only made when considering a relatively small and simple box girder structure.
The benchmark also demonstrates clear variation between experimental results and equivalent baseline
numerical results. The baseline calculations were completed using nominal values for material and
geometric parameters. The variation suggests that one or more parameters from the specification of the
experiment differ from the reality. The boundary conditions in the experiment and numerical analyses
may also differ. To evaluate some of these potential differences, specific material and geometric
parameters were varied in the finite element models and the influence on the ultimate strength was
evaluated. These results show that the material stress-strain model, imperfection shape and imperfection
amplitude have only minor influence on the ultimate strength. Varying the plate thickness has a more
significant influence on the ultimate strength. The finite element analyses are also compared to a
simplified progressive collapse analysis using the Smith method. The Smith method predicts ultimate
strengths closer to the experimental results but still lower in magnitude. In this instance, the Smith
method is highly sensitive to the extent of the hard corner because of the small size of the cross section.
5.2
Three hold model of hull girder
The main purpose of this benchmark study using three hold model are to investigate the influence of
initial imperfections and lateral load on hull girder ultimate strength and also to clarify the differences of
calculation results provided by several researchers.
5.2.1
Calculation cases
As shown in Table 9, eight cases varying the initial imperfections and the lateral loads will be calculated.
The effect of initial imperfections and/or lateral loads on the hull girder ultimate strength will be
investigated. The mid-ship section of the modelled Capesize bulk carrier is shown in Figure 36. The
longitudinal stiffener space and the transverse frame space are also shown in Figure 36. The principal
333
particulars and the dimensions of longitudinal stiffeners are summarized in Tables 10 and 11,
respectively. The Youngs modulus (E) of longitudinal members in the mid-ship section is 206,000 [MPa]
in the elastic region, and the bi-linear relation between stress and strain is assumed with the strain
hardening rate of E/50. The yield stress is also shown in Figure 36.
In actual ships, initial geometrical imperfections and residual stresses caused by welding exist in plates
and stiffeners, and these initial imperfections will reduce the ultimate strength of both stiffened plate
structures and the entire hull girder.
Only geometrical imperfections are considered in the calculations of case_3, 4, 7 and 8 as shown in
Table 9. The local buckling mode given by Eq. (14) is considered in panels, and both the flexural
buckling mode and torsional buckling mode given by Eq. (15) are considered in stiffeners. The initial
imperfections are also given to flange so as to maintain right angles between flange and web. The
schematic diagram and an example of initial imperfections are shown in Figure 37 and 38, respectively.
m x
y
x
sin
+ B0 sin
a
b
a
x
x
= B0 sin
, v0 w = C0 sin
a
a
w0 p = A0 sin
(14)
w0 w
(15)
where:
A0 (= 0.12tp): Maximum deflection of the local buckling mode
B0 (= 0.0015a): Maximum deflection of the flexural buckling mode
C0 (= 0.0015a): Maximum deflection of the torsional buckling mode
(=
b
tp
Y
E
): Slenderness ratio
m (= a/b): Aspect ratio

tp: Panel thickness
a: Transverse frame space E: Youngs modulus
b: Longitudinal stiffener space Y: Yield stress
Table 9. Calculation cases
Condition
Case_1
Case_2
Case_3
Case_4
Case_5
Case_6
Case_7
Case_8
Hogging
Sagging
Hogging
Sagging
Hogging
Sagging
Hogging
Sagging
Table 10. Principal particulars.

Initial
imperfections
Lateral
loads
Without
Without
With
Without
With
With
Object
hold
Empty
Loaded
Empty
Loaded
Length [m]
Breadth [m]
Depth [m]
Draft [m]
Number of cargo holds
Hold length [m]
285.0
50.0
26.9
19.83
9
26.1
Table 11. Dimensions of longitudinal stiffeners.

No.*
Figure 36. Mid-ship section of Capesize

bulk carrier.
Dimensions
[mm]
Type
Yield stress
[MPa]
39027
Flat-bar
392
3339+10016
Tee-bar
352.8
2839+10014
Tee-bar
352.8
2839+10018
Tee-bar
352.8
3339+10017
Tee-bar
352.8
2839+10016
Tee-bar
352.8
18032.59.5
2839+10017
Bulb-bar
235.2
Tee-bar
352.8
3339+10018
Tee-bar
352.8
10
3339+10019
Tee-bar
352.8
11
3839+10017
Tee-bar
352.8
12
38310+10018
Tee-bar
352.8
13
38310+10021
Tee-bar
352.8
14
30027
Flat-bar
392
* Numbers correspond to those indicated in Figure 36.
334

a
a
a/2
tp
b/2
Longi.
b/2
a/2
Panel
Trans.
Stiffener
Figure 37. Schematic diagram of initial imperfections.

Trans.
Z
Y
Trans.
Trans.
(b) Longitudinal view
Deflection20
Trans.
Z
Y
(a) Birds-eye view
Z
X
(c) Transverse view

Figure 38. Example of initial imperfections (deflection 20).
This calculation is performed on the assumption that the bulk carrier has nine holds, with the odd
numbered holds heavily loaded with internal cargo under the alternate loading condition shown in Figure
39. Therefore, the double bottom structure is subjected to the global vertical bending moment as well as
local lateral pressure loads caused by external sea pressure on outer bottom plate and internal cargo
pressure on inner bottom plate in cargo holds. The internal cargo pressure and external sea pressure are
considered in the calculations of Case_5, 6, 7 and 8 as shown in Table 9.
In accordance with the requirements for the direct strength assessment in CSR-BC, the alternate loading
condition shown in Table 12 is considered, i.e., heavy cargo alternate loading condition (LC_10). Although
this is the only load case that must be considered for the direct strength assessment, the effect of lateral
loads on the hull girder ultimate strength in the sagging condition is also investigated in this calculation.
In the hogging condition, i.e., the calculations of Case_5 and 7 as shown in Table 9, the hull girder
ultimate strength in the empty hold is calculated by utilizing the No. 3 ~ No. 5 holds model as shown in
Figure 39. On the other hand, in the sagging condition, i.e., the calculations of Case_6 and 8 as shown in
Table 9, the hull girder ultimate strength in the cargo hold is calculated by utilizing the No. 4 ~ No. 6
holds model as shown in Figure 39.
The boundary conditions are shown in Figure 40. The symmetry conditions were applied to all
members in the center plane. A center point of the FE model was constrained X displacement. Rigid
bodies were attached at both ends of the FE model, and constrained Y and Z displacements and X and Z
rotations. Moreover, rigid bodies were also used for applying the global bending moment by the rotation
control (enforced rotation) on the FE model.
No.9
No.8
No.7
No.6
No.5
No.4
No.3
Cal.RegioninHog.
Cal.RegioninSag.
Figure 39. Alternate condition and calculation region.
No.2
No.1
335
Table 12. Calculating loading condition.

No.
Description
Draft
LC_10
Alternate load
(heavy cargo)
TS
Loading pattern
Aft Mid Fore
Load case
F2*
* F2: Following sea in hogging.

Rigid body
Uy=Uz=z=0
Rigidbody
(Uy =Uz = x =z =0)
Enforced
rotation
Enforcedrotation
(y)(y)
Enforced
rotation
Enforcedrotation
((
y) y)
Symmetry
condition
Symmetrycondition
(Uy =x =z =0))
(Uy=x=z=0
Center
point
Centerpoint
(Ux =0) )
(Ux=0
Figure 40. Boundary conditions.
5.2.2
Calculation results
Participants for this benchmark calculations and the mesh division of FE model are summarized in
Table 13.
The effect of initial imperfections
At first, participants investigated the effect of initial imperfections on the hull girder ultimate strength.
Figure 41 shows the relationships between the vertical bending moment and the curvature from all
participants by utilizing the 3 holds model. Figure 41 (a) and (b) show the results without and with
imperfections, respectively. Table 14 shows the peak moment obtained by FE analysis.
Participants 1 and 2, who used the static implicit FEA code, obtained the mostly linear momentcurvature relationships and failed to get the post-buckling behaviour of stiffened panel. Static analysis can
simulate the post buckling behaviour of simple stiffened panel by utilizing the enforced displacement
procedure or Riks method, and can get the ultimate strength of it. But, for the complicated structure, such as
actual hull girder structure which is composed with many stiffened panels, it is difficult to calculate the post
buckling behaviour of each buckled panel, because there are so many nodes with negative stiffness after
buckling.
Participant 2 also investigated the effect of modeling region on the hull girder ultimate strength. Figure
42 shows the relationships between the vertical bending moment and the curvature varying the modeling
region. The obtained peak moments are summarized in Table 15. From Table 15, the peak moment
obtained by using the 3 frames model was highest in all cases (Case_1-4). Furthermore, the results of the
3 hold models show the significantly low peak moment except for Case_1. In the case of small
calculation region, especially in hogging condition with imperfection, both ultimate strength and post
ultimate strength can be obtained, but the others cases seems to fail. The reason that the ultimate strength
can be obtained only for the 3 frames model with imperfection under hogging condition, is thought to be
follows.
In the small region model, the number of elements with negative stiffness by buckling is relatively
small, and the change of stiffness after buckling is smaller in an imperfect model than in a perfect
model, and the effect of buckling on ultimate strength is smaller in hogging condition than sagging
condition.
On the other hand, participants 3 and 4, who used the dynamic explicit FEM, can obtain not only the
ultimate strength but also the post ultimate behaviour. The calculation can be continued in stable at post
buckling and post ultimate region due to the inertia effect. The results of ultimate strength by both
participants show a small difference but same tendencies. Participant 4 also calculated with an implicit
procedure, and could obtain the ultimate strength only for Case_4 utilizing Riks method by trial-and-
336
errors changing the loading steps, but failed for another cases as shown in Figure 41 with the blue dotted
line. Overall, it seems that it is difficult to obtain the ultimate strength for actual ship hull girder by static
procedure even if using Riks method.
Table 13. Summary of participants in benchmark study.
FEA code
Solver
Plate*
Participant 1
Participant 2
Participant 3
Participant 4
Trident
ANSYS
LS-DYNA
Implicit
Implicit
Explicit
10 10
(10 30)**
Bottom part:4 12
Deck part :4 24
ANSYS
Explicit
Implicit
Bottom part: 612
Deck part: 6 25
Bottom part:10 29
Deck part :7 42
Web depth
Deck, 3; Bottom, 4
3 (6)**
Flange breadth
2 (4)**
*Plates bounded by longitudinal stiffeners and transverse web frames (breadth length).
**Figures between brackets indicate the number of elements only for the 3 frames model.
From results by participant 3, the effect of initial imperfections on the hull girder ultimate strength was
more remarkable under the sagging condition than under the hogging condition.
Under hogging moment, the deck part first reaches the yield stress in tension, and keeps its loading
capacity, and then the bottom part reaches its ultimate strength in compression. In this case, the yielding
load at the deck is not much affected by initial imperfections, and the reduction of ultimate strength due to
imperfection is small.
Under sagging condition, the deck part first reaches to its buckling strength, and reduces its loading
capacity after buckling. Redistribution of stress in the deck takes place, but the whole section reaches
ultimate strength because the reduction of capacity of buckled parts is large. The effect of imperfection on
ultimate strength in compression is larger than in tension, therefore the reduction of ultimate strength due
to imperfection is larger in sagging condition than in hogging condition.
The reduction of load carrying capacity after ultimate strength in sagging condition is larger than in
hogging condition.
The effect of lateral loads
Participant 3 investigated the effect of lateral loads on the hull girder ultimate strength. Figure 43 shows
the relationships between the vertical bending moment and the curvature of all cases. The calculation
results obtained by the Smiths method are also shown in Figure 43. The hull girder ultimate strength is
summarized in Table 16. Participant 1, 2 and 3 also tried to calculate the effect of lateral loads, and their
results are shown in Figure 44 and Table 17. But, many of the calculations failed to find ultimate strength
due to the same reason for case 1 to 4.
The effect of lateral loads on the hull girder ultimate strength was more remarkable under the hogging
condition than under the sagging condition. At the bottom plates at mid span of the empty hold under the
hogging condition, the compressive stress is introduced by the hogging moment, and the compressive stress
is also produced by the lateral pressure loads. Therefore, the buckling of bottom plates and its stiffeners
takes place earlier than the tensile yielding at deck plates. Under the hogging condition, it is found that the
calculation result by the Smiths method agree with the FEA results without considering the lateral loads,
and the FEA results including the effect of lateral loads show the safety value compared with the value of
MCSR-H as shown in Table 16. Under the sagging condition, however, differences between the calculation
result by the Smiths method and the FEA results are observed. This is because the initial imperfections
much affect the hull girder ultimate strength under the sagging condition, and the buckling mode of
imperfection in FEA seems to be safety side assumption compared to the actual imperfection in real ship
structure.
Table 14. Hull girder ultimate strength using 3 holds model (unit; GNm)
Case_1
Case_2
Case_3
Case_4
Participant 1
-17.57
-14.59
-15.74
-14.44/-14.54
Participant 2
-14.37
-10.26
-15.15
-11.04
Participant 3
-18.54
-17.02
-17.57
-12.88
Participant 4
-17.67
-16.24
-17.57
-12.88
337
Table 15. The difference of peak bending moment changing model region (unit; GN m).
(participant 2 with implicit static calculation)
3 frames
-14.82
-16.26
-17.94
-15.97
Case_1
Case_2
Case_3
Case_4
5 frames
-17.83
-14.37
7 frames
-17.58
-14.75
9 frames
-17.58
-13.15
1 hold
-13.03
-17.56
-14.92
3 holds
-14.73
-10.26
-17.49
-14.74
Table 16. Hull girder ultimate strength varying initial imperfections and lateral loads (unit; GN m).
(participant 3)
Initial imperfections
FEA
Without
-18.54
-13.68
-17.02
-17.00
Lateral loads
Without
With
Hogging
Without
Sagging
With
With
-17.57
-13.39
-12.88
-12.83
Smiths
method
(Mu)
CSR-H*
(MCSR-H)
-18.31
-13.32
-14.73
-13.39
*According to CSR-H, the reduction of hull girder ultimate strength due to lateral loads is considered as
follows: MCSR-H = Mu / (M DB), where, M = 1.1, DB = 1.25 (Hog.), 1.0 (Sag.),
M: Partial safety factor covering material, geometric, and strength prediction uncertainties,
DB: Partial safety factor covering the effect of double bottom bending
Participant
1
Participant
Participant 2
Participant
Participant
3
Participant
4
Participant
1
2
3
Participant 4
20.0
15.0
10.0
Hogging
Hogging
0.0
0.0
-5.0
-10.0
-15.0
-20.0
1.5 2.0 2.5

2.5
Participant
1
Participant
1
Participant
2
Participant
2
Participant 3
Participant
3
Participant
4
Participant
Static
implicit (Riks) 4
Static
implicit(Riks)
by participant
4
20.0
5.0
Sagging
Sagging
-25.0
-25.0
-2.5
-2.5 -2.0 -1.5 -1.0 -0.5 0.0
0.0 0.5 1.0
Curvature
[10-4-4m-1-1]
Curvature [ 10 m ]
25.0
25.0
Verticalbending
bending moment
moment [GN
m]
Vertical
[GNm]
Verticalbending
bending moment
moment [GN
m]
Vertical
[GNm]
25.0
25.0
15.0
10.0
Hogging
Hogging
by participant4
5.0
0.0
0.0
-5.0
-10.0
-15.0
-20.0
Sagging
Sagging
-25.0
-25.0
-2.5 -2.0 -1.5 -1.0 -0.5 0.0
0.0 0.5 1.0 1.5
-2.5
Curvature
[10-4-4m-1-1]
Curvature [ 10 m ]
(a) Case_1 (hog.) and Case_2 (sag.)
2.0 2.5
2.5
(b) Case_3 (hog.) and Case_4 (sag.)
Figure 41. Relationships between vertical bending moment and curvature using 3 holds model
20.0
15.0
3 frames
: 3frame
1 hold
: 1 hold
3 holds
Hogging
Hogging
: 3holds
10.0
5.0
0.0
0.0
-5.0
-10.0
-15.0
-20.0
Sagging
Sagging
-25.0
-25.0
-2.5 -2.0 -1.5 -1.0 -0.5 0.0
0.0 0.5 1.0
-2.5
Curvature
[10-4-4m-1-1]
Curvature [ 10 m ]
25.0
25.0
Verticalbending
bending moment
moment [GN
m]
Vertical
[GNm]
Verticalbending
bending moment
moment [GN
m]
Vertical
[GNm]
25.0
25.0
1.5 2.0 2.5

2.5
20.0
15.0
10.0
5.0
3 frames
: 3 frames
5 frames
: 5 frames
7 frames
9 frames
: 7 frames
1 hold
: 9 frames
3 holds
Hogging
Hogging
: 1 hold
: 3 holds
0.0
0.0
-5.0
-10.0
-15.0
-20.0
Sagging
Sagging
-25.0
-25.0
0.0 0.5 1.0 1.5
-2.5 -2.0 -1.5 -1.0 -0.5 0.0
-2.5
Curvature
[10-4-4m-1-1]
Curvature [ 10 m ]
2.5
2.0 2.5
Figure 42. Relationships between vertical bending moment and curvature varying modeling region
(participant 2)
338
Verticalbending
bending moment
moment [GN
m]
Vertical
[GNm]
25.0
25.0
: Case_1(Hog.)
Case_1
(Hog.)
: Case_2(Sag.)
Case_2
(Sag.)
Case_3
(Hog.)
: Case_3(Hog.)
Case_4
(Sag.)
: Case_4(Sag.)
Case_5
(Hog.)
: Case_5(Hog.)
Case_6
(Sag.)
: Case_6(Sag.)
Case_7 (Hog.)
:
Case_7(Hog.)
Case_8 (Sag.)
: Case_8(Sag.)
Smith's
method
20.0
15.0
10.0
5.0
Table 17: Hull girder ultimate strength

using 3 holds model (unit; GNm)
Hogging
Hogging
: Smiths method
0.0
0.0
Participant 1
-07.39
-14.54
-07.48
-14.37/-14.32
Case_5
Case_6
Case_7
Case_8
-5.0
Participant 2
-10.40
Participant 3
-13.68
-17.00
-13.39
-12.83
-10.0
-15.0
-20.0
Sagging
Sagging
-25.0
-25.0
-2.5 -2.0 -1.5 -1.0 -0.5 0.0
0.0 0.5 1.0
-2.5
Curvature
[10-4-4m-1-1]
Curvature [ 10 m ]
1.5 2.0 2.5

2.5
Figure 43. Relationships between vertical

bending moment and curvature varying initial
imperfections and lateral loads (participant 3)
20.0
15.0
Participant
1
: Participant
Participant 3
1
: Participant 3
Hogging
Hogging
10.0
5.0
0.0
0.0
-5.0
-10.0
-15.0
-20.0
Sagging
Sagging
-25.0
-25.0
-2.5
0.0 0.5 1.0
-2.5 -2.0 -1.5 -1.0 -0.5 0.0
-4 -1
-4 m-1 ]
Curvature
Curvature [[10
10 m
]
25.0
25.0
Verticalbending
bending moment
moment [GN
m]
Vertical
[GNm]
Verticalbending
bending moment
moment [GN
m]
Vertical
[GNm]
25.0
25.0
1.5 2.0 2.5

2.5
20.0
15.0
Participant
1
: Participant
Participant 2
: Participant
Participant
3
1
2
: Participant 3
Hogging
Hogging
10.0
5.0
0.0
0.0
-5.0
-10.0
-15.0
-20.0
Sagging
Sagging
-25.0
-25.0
-2.5
-2.5 -2.0 -1.5 -1.0 -0.5 0.0
0.0 0.5 1.0
Curvature
[10-4-4m-1-1]
Curvature [ 10 m ]
1.5 2.0 2.5

2.5
Figure 44. Relationships between vertical bending moment and curvature using 3 holds model,
5.3
Summary and Recommendation for Future Works
In the calculation of three hold model of a bulk carrier hull girder (section 5.2), the influence of initial
imperfections and lateral load on hull girder ultimate strength are investigated. The differences in
calculation results provided several researchers are clarified.
Unlike the box girder model in section 5.1, it is difficult to obtain the ultimate strength of a full scale hull
girder by using a static FE solver, because there are so many elements with negative stiffness after buckling.
However, by using time dependent dynamic analysis, it is easy to obtain the ultimate strength of large size of
model, such as three hold model.
As for the effect of initial imperfections on the hull girder ultimate strength, it was more remarkable
under the sagging condition than under the hogging condition. Under hogging moment, the deck part first
reaches yield stress in tension, and keeps its loading capacity, and then the bottom part reaches ultimate
strength in compression. Under sagging condition, the deck part first reaches buckling strength, and
reduces its loading capacity after buckling. The redistribution of stress at deck part takes place, but the
whole section reaches ultimate strength because the reduction of capacity of buckled parts is large. The
effect of imperfection on ultimate strength in compression is larger than in tension, therefore the
reduction of ultimate strength due to imperfection is larger in sagging condition than in hogging
condition.
As for the effect of lateral loads on the hull girder ultimate strength, it was more remarkable under the
hogging condition than under the sagging condition. Under the hogging condition, compressive stress is
introduced in the outer bottom plates by both the hull girder bending moment and the lateral pressure
loads. Therefore, the buckling of outer bottom plates and its stiffeners takes place earlier than the tensile
339
yielding at deck plates. But under the sagging condition, the ultimate strength mainly depends on the
buckling of deck, and therefore the effect of lateral pressure on the hull girder strength is small under
sagging moment.
The hull girder ultimate strength by FEM considering initial imperfection and lateral loads shows good
agreement with predicted value by CSR-H. The newly introduced partial safety factor covering the effect of
double bottom bending DB seems to be reasonable from the results of benchmark study.
6.
CONCLUSION AND RECOMMENDATION
This report describes the results of literature survey and benchmark calculations related to buckling and
ultimate strength of components and systems of ships and offshore structures, which have been conducted
during the last three years. The report consists of six chapters.
In chapter 1, the history of assessment of buckling and ultimate strength of ships and offshore
structures are briefly reviewed. Recent developments led by marine industry authorities on the buckling
and ultimate strength of ships and offshore structures, including GBS (Goal-based New ship Construction
Standard) in IMO, CSR (Common structural rules) by IACS and CSR-H (Harmonized Common
Structural Rule) by IACS are introduced.
In chapter 2, the two criterions for ultimate assessment, the partial safety approach and the probabilistic
approach, are briefly described, and the fundamentals and general characteristic of ultimate strength are
also discussed.
In chapter 3, the results of literature survey for assessment procedure for ultimate strength, such as 1)
Empirical and analytical methods, 2) Numerical methods, 3) Experimental methods, 4) Reliability
assessment, 5) Rules and regulations are shown.
For empirical and analytical methods, modifications and extensions of simplified progressive collapse
methods for predicting the ultimate strength of hull structures are investigated. The investigations of
empirical formula for ultimate residual strength of damaged hull due to collision and corrosion are also
performed.
For numerical methods, the numerical problem using static FEA in calculation of ultimate strength of
large size of model, such as a three hold actual ship model, is highlighted. The needs of development of
more valid ISUM elements, from the viewpoint of modeling effort and computing time, are also noted.
For experimental methods, experiments for stiffened panels, box girder models have been performed in
these three years, but the specimen sizes and scantlings are limited. More experiments are desired to be
performed and the detail information of experiments is expected to be opened for validation of analytical
and numerical calculations.
Reliability assessment of ultimate strength of various items, such as ships damaged by collision/
grounding and corrosion, are recently performed by many researchers. By accumulation of results of
reliability assessment of ultimate strength, it will be expected to be used more for confirming and
modifying the regulations.
For rules and regulations, the changes in CSR-H from CSR are mainly described. The scantlings
according to CSR-H are somewhat more conservative than CSR. This is mainly due to using the more
conservative formula in both CSR rules sets when a theoretical back ground is not clear. It is desired to
perform a further study in order to clarify the theoretical background, with modification of rules as
necessary. Some of the technical background documents are not sufficient. We understand that this is
unavoidable because not all been clarified yet. It is recommended that the unknown and unapparent items
be clearly stated in background documents.
In chapter 4, the results of literature survey for ultimate strength of various structures, such as 1)
Tubular members and joins, 2) Steel plate and stiffened plates, 3) Shells, 4) Ship structures, 5) Offshore
structure, 6) Composite structures and 7) Aluminum structures are described.
For tubular members and joins, many researchers have examined two type of reinforcement on the
strength of tubular members, i.e., using the fiber reinforce polymers and the steel- concrete composite
structure. As recent research efforts have examined the resistance of tubular member subjected lateral
impacts. The various reinforcement schemes for tubular joints have been investigated to extend the
operating life of existing offshore platform. Recent research efforts have also covered the strength of
complex tubular joints e.g. the elliptical joints, multi-planar joints, for which the existing design codes do
not provide explicit strength equations.
Moreover, many researchers have examined the strength of tubular joints under accidental loading to
develop the basis for engineering guidelines in such limit states.
For study of ultimate strength of steel plate and stiffened plates continues to be important for
improving code based design formulas and direct strength analysis procedures. The analytical formulation
for ultimate strength of stiffened panels have continued to study under compressive loading and multiple
340
loading, such with shear loading and lateral loading. And the behaviour of the panel with opening, rupture
damage and service degradation are also investigated by numerical procedures.
For shells, the ultimate strength can be calculated by numerical methods considering initial
imperfections and the effect of yielding, but the simple formulas for estimating the ultimate strength of
shells, such as stiffened shells under the external pressure, is desired to be investigated more.
For ship structures, the special attention have been dedicated to progressive collapse methods, FEA and
their applicability on intact ship integrity assessment, degradation by corrosion or reduction of the
ultimate strength of the ship girder by structural local damage. Also some work was done on the
interaction between different types of loading and their effect on the ultimate strength or interaction
between vertical and horizontal bending.
Available publications on the ultimate strength of offshore structures are rather limited compared with
considerable numbers of publication on ship/ ship-like structures. Offshore structures are huge assemblies
of various components depending on the required function and working conditions. Much effort has been
made on the ultimate strength at the level of specific components, i.e., stiffened panels, tubular members
or joints; and on the effect of various parameters, i.e., composite material, initial imperfection, aging
factors (corrosion and fatigue). Less effort has been made on the ultimate strength of the offshore
structure globally and systematically. In this regard, systematic assessment of ultimate strength of
offshore structures deserves further efforts.
The ultimate strength of composite structure is determined by coding a suitable composite failure
criterion within an incremental loading algorithm. Once failure is detected the degradation model is used
to reduce the properties of the laminate to account for the material damage and to effectively weaken the
panel. Recent research focused on failure identification and material degradation models. The very little
work has been done on simplified models for the calculation of ultimate strength of composite panels, and
an area that has received attention is the ultimate strength after impact.
For aluminum structures, the available literature is relatively limited over the last three years when
compared with the studies on steel structures. Compared to steel structures, the ultimate strength and
collapse behaviour of aluminium structures are significantly affected not only by initial imperfections, but
also by the material nonlinearity and the deterioration of mechanical strength in the heat-affected zones
(HAZ) caused by welding in assembly and fabrication. Therefore, weld-induced effects have been then
main focus of research on aluminium structures.
In chapter 5, the results of benchmark calculation for ultimate strength of small box girder model and
three hold model of bulk carrier are shown.
For the benchmark calculation of small box girder model, the discrepancy of results between
participants is in the range of uncertainly suggested by the IACS CSR partial safety factor. It should be
noted that the numerical results are significantly lower than the experimentally measured strength. There
is a possibility of difference of boundary condition between calculations and experiment, especially
friction effect at loading and supporting positions.
For the benchmark calculation of three hold model of bulk carrier, it is hard to obtain the ultimate strength
of hull girder by using the static FE solver, because there are so many elements with negative stiffness after
buckling. By using the time dependent dynamic analysis, it is easy to obtain the ultimate strength of large size
of model, such as three hold model. The hull girder ultimate strength by the dynamic explicit FEM considering
initial imperfection and lateral loads shows good agreement with predicted value by CSR-H. The newly
introduced partial safety factor covering the effect of double bottom bendingDB seems to be reasonable
from the results of benchmark study.
REFERENCES
Aalberg, A., Langseth, M. & Larsen, P. K. 2001. Stiffened aluminium panels subjected to axial compression. Thinwalled structures, 39, 861885.
Abs 2014. ABS guide for Buckling and Ultimate Strength Assessment for Offshore Structures. ABS.
Abubakar, A. & Dow, R. S. 2013. Simulation of grounding damage using the finite element method. Int J Solids
Struct., 50, 623636.
Alagusundaramoorthy, P., Sundaravadivelu, R. & Ganapathy, C. 1999. Experimental study on collapse load of
stiffened panels with cutouts. Journal of Constructional Steel Research, 52, 235251.
Albermani, F., Khalilpasha, H. & Karampour, H. 2011. Propagation buckling in deep sub-sea pipelines. Eng. Struct.,
33, 25472553.
Alinia, M. M. & Dibaie, A. 2012. Buckling and failure characteristics of slender web I-column girders under
interactive compression and shear. comp. Meth. Civil Eng., 3, 1534.
Alinia, M. M., Hosseinzadeh, S. a. A. & Habashi, H. R. 2007. Influence of central cracks on buckling and postbuckling behaviour of shear panels. Thin-Wall Struct., 45, 422431.
Alinia, M. M., Hosseinzadeh, S. a. A. & Habashi, H. R. 2008. Buckling and post-buckling strength of shear panels
degraded by near border cracks. J. Construct Steel Res., 64, 14831494.
341
Almroth, B. O. 1966. Influence of imperfections and edge restraint on the buckling of axially compressed cylinders.
NASA CR-432, Prepared under contract No. NAS 13778 by Lockeheed Missile and Space Company.
Alps/Scol 2006. A computer program for progressive structural crashworthiness under collisions and grounding.
Proteus Engineering S: MD, USA: Stevensville.
Amlashi, H. K. K. & Moan, T. 2008. Ultimate strength analysis of a bulk carrier hull girder under alternate hold
loading conditiona case study Part 1: nonlinear finite element modelling and ultimate hull girder capacity.
An, C., Duan, M. L., Filhoc, R. T. & Estefen, S. F. 2014. Collapse of sandwich pipes with PVA fiber reinforced
cementitious composites core under external pressure. Ocean Engineering, 82, 113.
Andric, J., Kitarovic, S. & Bicak, M. 2014. IACS incremental-iterative method in progressive collapse analysis of
various hull girder structures. Brodogradnja/Shipbuilding, 65, 6578.
Anyfantis, K. N. & Tsouvalis, N. G. 2012. Post buckling progressive failure analysis of composite laminated
stiffened panels. Applied Composite Materials, 19, 219236.
Arjomandi, K. & Taheri, F. 2012. Bending Capacity of Sandwich Pipes. Ocean Engineering, 48, 1731.
Bayatfar, A., Khedmati, M. R. & Rigo, P. 2014. Residual ultimate strength of cracked steel unstiffened and stiffened
plates under longitudinal compression. Thin-Walled Structures 84, 378392.
Benson, S. 2011. Progressive Collapse Assessment of Lightweight Ship Structures, Newcastle University.
Benson, S., Abubakar, A. & Dow, R. S. 2013a. A comparison of computational methods to predict the progressive
collapse behaviour of a damaged box girder. Eng Struct., 48, 266280.
Benson, S., Downes, J. & Dow, R. 2013b. Load shortening characteristics of marine grade aluminium alloy plates in
longitudinal compression. Thin-Walled Structures, 70, 1932.
Benson, S., Downes, J. & Dow, R. S. 2013c. A Comparison of Numerical Methods to Predict the Progressive
Collapse of Lightweight Aluminium Vessels. J. Ship Prod. Des., 29, 117126.
Benson, S., Downes, J. & Dow, R. S. 2013d. Compartment level progressive collapse analysis of lightweight ship
structures. Marine Structures, 31, 4462.
Benson, S., Leelachai, A. & Dow, R. S. 2013e. Ultimate strength of stiffened panels with circular cut-outs under inplane compression 12th International Conference on Fast Sea Transportation.
Benson, S., Syrigou, M. & Dow, R. S. 2013f. Longitudinal strength assessment of damaged box girders. Collision
and Grounding of Ships and Offshore Structures. Proceedings of the 6th International Conference on Collision
and Grounding of Ships and Offshore Structures.
Brubak, B., Andersen, H. & Hellesland, J. 2013. Ultimate strength prediction by semi-analytical analysis of
stiffened plates with various boundary conditions. Thin-Walled Structures, 62, 2836.
Brubak, L. & Hellesland, J. 2008. Strength criteria in semi-analytical, large deflection analysis of stiffened plates in
local and global bending. Thin-Walled Structures.
Brubak, L., Hellesland, J. & Steen, E. 2007. Semi-analytical buckling strength analysis of plates with arbitrary
stiffener arrangements. Journal of Constructional Steel Research, 63, 532543.
Bryan, G. H. 1891. On the stability of a plane plate under thrust in its own plane with application to the buckling of
side of a ship. London Math. Soc., 22, 5467.
Bs7910 2013. Guide to methods for assessing the acceptability of flaws in metallic structures. British Standards
Institute.
Caldwell, J. B. 1965. Ultimate longitudinal strength. Trans RINA, 107, 41130.
Cerik, B. C. & Cho, S. R. 2013. Residual strength of damaged stiffened cylinders under combined axial compression
and radial pressure. . The 12th Int. Symp. on Practical Design of Ships and Other Floating Structures
(PRADS13). Changwon City: Korea.
Chalmers, D. W. 1993. Design of Ships Structures, London.
Chamis, C. C., Abdi, F., Garg, M., Minnetyan, L., Baid, H., Huang, D., Housner, J. & Talagani, F. 2013.
Micromechanics-based progressive failure analysis prediction for WWFE-III composite coupon test cases.
Journal of Composite Materials, 47, 26952712.
Chan, T. M., Gardner, L. & Law, K. H. 2010. Structural design of elliptical hollow sections: a review. Proc. ICE
Struct. Build.
Chang-Li, Y. & Lee, J. 2014. Formulation of reduction rate for ultimate compressive strength of stiffened panel
induced by opening. China Ocean Eng., 28, 557568.
Chatzopoulou, G., Varelies, G.E. & Karamanos, S.A. 2014. Finite element analysis of UOE pipes under external
pressure and bending. In: Conference, P. O. T. I. O. a. P. E. (ed.).
Chen, J. F., Morozov, E. V. & Shankar, K. 2011. Plastic damage model for progressive failure analysis of composite
structures. ICCM International Conferences on Composite Materials.
Chen, N. Z. & Guedes Soares, C. 2007. Longitudinal Strength Analysis of Ship Hulls of Composite Materials under
Sagging Moments. Composite Structures, 77, 3644.
Chen, N. Z. & Guedes Soares, C. 2008. Ultimate longitudinal strength of ship hulls of composite materials. Journal
of Ship Research, 52, 184193.
Chen, N. Z., Sun, H. H. & Guedes Soares, C. 2003. Reliability analysis of a ship hull in composite material.
Composite Structures, 62, 5966.
Chenfeng, L., Huilong, R., Zhongqiu, Z., Xiaodong, Z., Ji, Z. & Donghao, X. 2013. An approximate calculation
approach for the ultimate strength of ship hull girder. OMAE 2013. San Francisco.
Chiew, S. P., Zhang, J. C., Shao, Y. B. & Qiu, Z. H. 2012. Experimental and Numerical Analysis of Complex
welded Tubular DKYY-Joints. Advances in Structural Engineeing, 15.
342
Cho, S. R., Kim, H. S., Doh, H. M. & Chon, Y. K. 2013. Ultimate strength formulation for stiffened plates subjected
to combined axial compression, transverse compression, shear force and lateral pressure loadings. Ships and
Offshore Structures, 8, 628637.
Choung, J., Nam, J. & Ha, T. 2012a. Assessment of residual ultimate strength of an asymmetrically damaged tanker
considering rotational and translational shifts of neutral axis plane. Marine Structures, 25, 7184.
Choung, J., Nam, J. M. & Ha, T.-B. 2012b. Slenderness ratio distribution and load-shortening behaviours of
stiffened panels. Mar. Struct., 26, 4257.
Choung, J., Nam, J. M., Tayyar, G. T., Yoon, S. W. & Lee, K. 2013. Assessment of Ultimate Longitudinal Strength
of a VLCC considering Kinematic Displacement Theory. Journal of the Society of Naval Architects of Korea, 50,
255261.
Choung, J., Park, J. B. & Song, C. Y. 2014. Lateral pressure effect on average compressive strength of stiffened
panels for in-service vessels. Ships Offshore Struct., 9, 110118.
Collette, M. D. 2011. Rapid analysis techniques for ultimate strength predictions of aluminium structures, Taylor &
Francis, London: UK.
Cook, R. D., Malkus, D. S., Plesha, M. E. & Witt, R. J. 2002. Concepts and applications of finite element analysis,
New York: Wiley.
Crupi, V., Epasto, G. & Guglielmino, E. 2013. Comparison of aluminium sandwiches for lightweight ship
structures: Honeycomb vs. foam. Marine Structures 30, 7496.
Cui, S., Cheong, H. K. & Hao, H. 2000. Experimental study of dynamic post-buckling characteristics of columns
under fluidsolid slamming. Engineering structures 22, 647656.
Cui, S., Kiat Cheong, H. & Hao, H. 1999. Experimental study of dynamic buckling of plates under fluidsolid
slamming. International Journal of Impact Engineering, 22, 675691.
Dai, X. H. & Lam, D. 2012. Shape effect on the behaviour of axially loaded concrete filled steel stub columns at
elevated temperature. J. Constr. Steel Res., 73, 117127.
Dai, X. H., Lam, D., Jamaluddin, N. & Ye, J. 2014. Numerical analysis of slender elliptical concrete filled columns
under axial compression. Thin Wall Struct., 77, 2635.
Deco, A. & Frangopol, D. M. 2013. Risk-informed optimal routing of ships considering different damage scenarios
and operational conditions. Reliability Engineering and System Safety, 119, 126140.
Deco, A., Frangopol, D. M. & Zhu, B. 2012. Reliability and redundancy assessment of ships under different
operational conditions. Ocean Engineering, 42, 457471.
Dnv-Os-J101 2014. Design of Offshore Wind Turbine Structures.
Dow, R. S. 1991. Testing and analysis of a 1/3 scale frigate model. Adv. Mar. Struct., 2, 749773.
Dyanati, M. & Huang, Q. 2014. Seismic Reliability of a fixed offshore platform against collapse. Proceedings of the
ASME 2014 33rd International Conference on Ocean, Offshore and Arctic Engineering. San Francisco:
California: USA.
El-Hanafi, A. S., Badran, S. F. & Leheta, H. W. 2013. Design of Y stiffened panels in double hull tanker under axial
compressive loads. Analysis and Design of Marine Structures, CRC Press, 291300.
Ergin, A. & Ozdemir, M. 2013. Ultimate Strength Analysis of Ship Structural Panels. TEAM 2013. Keelung:
Taiwan.
Espinos, A., Gardner, L., Romero, M. L. & Hospitaler, A. 2011. Fire behaviour of concrete-filled elliptical steel
columns. . Thin Wall Struct., 49, 239255.
Estefen, T. P. & Estefen, S. F. 2012. Buckling propagation failure in semi-submersible platform columns. Mar.
Struct., 28, 224.
Farkas, J. & Jrmai, K. 2014. Optimization of welded square cellular plates with two different kinds of stiffeners. J.
Constr. Steel Res., 101, 6165.
Fujikubo, M. & Kaeding, P. 2002. New simplified approach to collapse analysis of stiffened plates Mar Struct. 15,
15, 251283.
Fujikubo, M., Muis Alie, M. Z., Takemura, K., Iijima, K. & Oka, S. 2012a. Residual Hull Girder Strength of
Asymmetrically Damaged Ships. Journal of the Japan Society of Naval Architects and Ocean Engineers, 16,
131140.
Fujikubo, M., Uda, S., Tatsumi, A., Iijima, K., Ogawa, H. & Takami, T. 2013. Finite Element Modeling of a
Continuous Stiffened Panel under Combined Inplane Shear and Thrust. TEAM 2013. Keelung: Taiwan.
Fujikubo, M., Zubair, M. A., Takemura, K., Iijima, K. & Oka, S. 2012b. Residual Hull Girder Strength of
Asymmetrically Damaged Ships. Journal of the Japan Society of Naval Architects and Ocean Engineers, 16,
131140.
Gan, J. M., Bickerton, S. & Battley, M. 2011. Modelling the initiation and evolution of damage within GFRP by
including real geometric variability. ICCM International Conferences on Composite Materials.
Gannon, L., Liu, Y., Pegg, N. & Smith, M. J. 2012a. Effect of welding-induced residual stress and distortion on ship
hull girder ultimate strength. Marine Structures, 28, 2549.
Gannon, L., Liu, Y., Pegg, N. & Smith, M. J. 2013. Effect of three-dimensional welding-induced residual stress and
distortion fields on strength and behaviour of flat-bar stiffened panels. Ships Offshore Struct., 8, 565578.
Gannon, L. G., Pegg, N. G., Smith, M. J. & Liu, Y. 2012b. Effect of residual stress shakedown on stiffened plate
strength and behaviour. Ships Offshore Struct. , 8, 638652.
Gao, D. W., Shi, G. J. & Wang, D. Y. 2012. Residual ultimate strength of hull structures with crack and corrosion
damage. Engineering Failure Analysis, 25, 316328.
343
Gaspar, B. & Guedes Soares, C. 2013. Hull girder reliability using a Monte Carlo based simulation method.
Probabilistic Engineering Mechanics, 31, 6575.
Gaspar, B., Teixeira, A. P., Guedes Soares, C. & Wang, G. 2011. Assessment of IACS-CSR implicit safety levels
for buckling strength of stiffened panels for double hull tankers. Mar. Struct., 24, 478502.
Ghavami, K. & Khedmati, M. R. 2006. Numerical and experimental investigations on the compression behaviour of
stiffened plates. Journal of Constructional Steel Research, 62, 10871100.
Gong, S., Sun, B., Bao, S. & Bai, Y. 2012. Buckle propagation of offshore pipelines under external pressure
Mar. Struct., 29, 115130.
Goode, C. D., Kuranovas, A. & Kvedaras, A. K. 2010. Buckling of slender composite concrete-filled steel columns.
J. Civ. Eng. Manag., 16, 230237.
Gordo, J. M. 2012. Ultimate compressive strength of double-bottom under alternate lateral loading SOARES, C. G.
& SANTOS, N. A. (eds.) Engenharia e Tecnologia Martima, Lisboa: Salamandra.
Gordo, J. M. & Guedes Soares, C. 1996. Approximate method to evaluate the hull girder collapse strength Mar.
Struct., 9, 449470.
Gordo, J. M. & Guedes Soares, C. 2004. Experimental evaluation of the ultimate bending moment of a box girder.
Marine Systems & Ocean Technology, 1, 114.
Gordo, J. M. & Guedes Soares, C. 2008a. Compressive tests on short continuous panels. Marine Structures, 21,
113137.
Gordo, J. M. & Guedes Soares, C. 2008b. Experimental evaluation of the behaviour of a mild steel box girder under
bending moment. Ships and Offshore Structures, 3, 347358.
Gordo, J. M. & Guedes Soares, C. 2009. Tests on ultimate strength of hull box girders made of high tensile steel.
Marine Structures 22(4), 22, 770790.
Gordo, J. M. & Guedes Soares, C. 2011. Compressive tests on stiffened panels of intermediate slenderness. ThinWalled Structures, 49, 782794.
Gordo, J. M. & Guedes Soares, C. 2012. Compressive Tests on Long Continuous Stiffened Panels. Journal of
Offshore Mechanics and Arctic Engineering, 134, 0214030214038.
Gordo, J. M. & Guedes Soares, C. 2014. Experimental analysis of the effect of frame spacing variation on the
ultimate bending moment of box girders. Marine Structures, 37, 111134.
Gordo, J. M., Teixeira, A. P. & Guedes Soares, C. 2011. Ultimate strength of ship structures. Marine Technology
and Engineering.
Gui-Jie, S. & De-Yu, W. 2012. Ultimate strength model experiment regarding a container ships hull structures.
Ships and Offshore Structures, 7, 165184.
Han, L. H., He, S. H., Zheng, L. Q. & Tao, Z. 2012. Curved concrete filled steel tubular (CCFST) built-up members
under axial compression: experiments. J. Constr. Steel Res., 74, 6375.
Han, L. H., Hou, C. C., Zhao, X. L. & Rasmussen, K. J. R. 2014. Behaviour of high-strength concrete filled steel
tubes under transverse impact loading. J. Constr. Steel Res., 92, 2539.
Han, L. H., Zheng, L. Q., He, S. H. & Tao, Z. 2011. Tests on curved concrete filled steel tubular members subjected
to axial compression. J. Constr. Steel Res., 67, 965976.
He, S. B., Shao, Y. B., Zhang, H. Y., Yang, D. P. & Long, F. L. 2013. Experimental study on circular hollow section
, 34, 204216.
(CHS) tubular K-joints at elevated temperature. Eng. Fail. Anal.
Hibbitt, H., Karlsson, B. & Sorensen, P. 2010. Abaqus analysis users manual version 6, RI, USA: Dassault
Systemes Simulia Corp Providence.
Hou, C., Han, L. H. & Zhao, X. L. 2013. Concrete-filled circular steel tubes subjected to local bearing force:
Experiments. Journal of Constructional Steel Research, 83, 90104.
Htun, M. M. & Kawamura, Y. 2013. A Study on the Method to Estimate the Stochastic Property of the Strength of
Corroded Plate by Response Surface Method with Polynomial Chaos Expansion. In: Ceco (ed.) Proceedings of
the PRADS2013. Changwon City: Korea.
Hu, S. Z. & Jiang, L. 1998. A finite element simulation of the test procedure of stiffened panels. Mar Struct., 11,
7599.
Hu, Y., Cui, W. C. & Pedersen, P. T. 2004. Maintained ship hull girder ultimate strength reliability considering
corrosion and fatigue. Mar Struct., 17, 91123.
Huchinson, J. W. 1967. Imperfection sensibitivity of external pressurized spherical shells. JournalofApplied
34.
Mechanics
Iacs 2006a. Common Structural Rules for Bulk Carriers, International Association of Classification societies.
Iacs 2006b. Common Structural Rules for Double Hull Oil Tankers, International Association of Classification
Societies.
Iacs 2010a. Common Structural Rules for Bulk Carriers. International Association of Classification Societies.
Iacs 2010b. Common Structural Rules for Double Hull Oil Tankers. International Association of Classification
Societies.
Iacs 2014a. Common Structural Rules for Bulk Carriers and Oil Tankers, International Association of Classification
Societies.
Iacs 2014b. Technical Background Documents for CSR, International Association of Classification societies.
Ibekwe, A. U., Pu, Y., Ham, W. L. & Dow, R. S. 2014. Progressive collapse analysis and reliability of a damaged
hull girder. Proceedings of the ASME 2014 33rd International Conference on Ocean, Offshore and Arctic
Engineering. San Francisco, California: USA.
Imo. Goal-Based Standards. International Maritime Organization. November 2006 London.
344
Iso. Requirements for Their Ultimate Limit State Assessment. International Organization for Standardization. In: 2,
I. C.-S. a. M. T.-S. S.-P., ed., November 2006 Geneva.
Jelovica, J., Romanoff, J., Ehlers, S. & Varsta, P. 2012. Influence of weld stiffness on buckling strength of laserwelded web-core sandwich plates. J. Constr. Steel Res., 77, 1218.
Jiang, X. & Guedes Soares, C. 2012a. A closed form formula to predict the ultimate capacity of pitted mild steel
plate under biaxial compression. Thin-Walled Structures, 59, 2734.
Jiang, X. & Guedes Soares, C. 2012b. Ultimate capacity of rectangular plates with partial depth pits under uniaxial
loads Marine Structures, 26, 2741.
Jin, M., Zhao, J. C., Chang, J. & Zhang, D. X. 2012. Experimental and parametric study on the post-fire behaviour
of tubular T-joint. J. Constr. Steel Res. , 70, 93100.
Kant, M. & Penumadu, D. 2011. Sea water effects on ultimate tensile and fracture strength of carbon fibers with
nano-tensile testing. ICCM International Conferences on Composite Materials.
Kant, M., Siriruk, A. & Penumadu, D. 2010. Sea water and temperature effects on the mechanical properties of
single carbon fibers. 25th Technical Conference of the American Society for Composites and 14th US-Japan
Conference on Composite Materials 2010.
Karampour, H., Albermani, F. & Veidt, M. 2013. Buckle interaction in deep subsea pipelines. Thin-Walled Struct.,
72, 113120.
Kenny, S., Taheri, F. & Pegg, N. 2002. Experimental investigations on the dynamic plastic buckling of a slender
beam subject to axial impact. International journal of impact engineering, 27, 117.
Khan, I. & Zhang, S. 2011. Effects of welding-induced residual stress on ultimate strength of plates and stiffened
panels. Ships Offshore Struct., 6, 297309.
Khedmati, M. R. & Edalat, P. 2013. Ultimate Strength Characteristics of a Ships Deck Stiffened Plate Structure in
the Presence of Camber Parabolic Curvature. J. Offshore Mech. Arct. Eng., 135, 031601031601.
Khedmati, M. R. & Nazari, M. 2012. A numerical investigation into strength and deformation characteristics of
preloaded tubular members under lateral impact loads. Mar. Struct., 25, 3357.
Khedmati, M. R., Nazari, M. & Khalaj, A. F. 2012. A numerical investigation into ultimate strength and buckling
behaviour of locally corroded steel tubular members. The 31rd Int. Conf. on Ocean, Offshore and Arctic
Engineering (OMAE2012). Rio de Janeiro: Brazil: American Society of Mechanical Engineers (ASME).
Khedmati, M. R., Pedram, M. & Rigo, P. 2014. The effects of geometrical imperfections on the ultimate strength of
aluminium stiffened plates subject to combined uniaxial compression and lateral pressure. Ships and Offshore
Kiat Cheong, H., Hao, H. & Cui, S. 2000. Experimental investigation of dynamic post-buckling characteristics of
rectangular plates under fluid-solid slamming. Engineering Structures, 22, 947960.
Kim, B. J., Seo, J. K., Park, J. H., Jeong, J. S., Oh, B. K., Kim, S. H., Park, C. H. & Paik, J. K. 2010. Load
characteristics of steel and concrete tubular members under jet fire: an experimental and numerical study. Ocean
Eng., 37, 11591168.
Kim, D., Kim, B., Seo, J., Kim, H., Zhang, X. & Paik, J. 2014. Time-dependent residual ultimate longitudinal
strength - grounding damage index (R-D) diagram. Ocean Engineering, 76, 163171.
Kim, D. K., Park, D. K., Kim, H. B., Seo, J. K., Kim, B. J., Paik, J. K. & Kim, M. S. 2012a. The necessity of
applying the common corrosion addition rule to container ships in terms of ultimate longitudinal strength. Ocean
Kim, D. K., Park, D. K., Park, D. H., Kim, H. B., Kim, B. J., Seo, J. K. & Paik, J. K. 2012b. Effect of corrosion on
the ultimate strength of double hull oil tankers - Part II: Hull girders. Structural Engineering and Mechanics, 42,
531549.
Kim, D. K., Pedersen, P. T., Paik, J. K., Kim, H. B., Zhang, X. & Kim, M. S. 2013a. Safety guidelines of ultimate
hull girder strength for grounded container ships. Safety Science, 59, 4654.
Kim, K., Park, D. H., Kim, H. B., Kim, B. J., Seo, J. K. & Paik, J. K. 2013b. Lateral pressure effects on the
progressive hull collapse behaviour of a Suezmax-class tanker under vertical bending moments. Ocean
Kim, K. & Yoo, C. H. 2008. Ultimate strengths of steel rectangular box beams subjected to combined action of
bending and torsion. Eng Struct. , 30, 16771687.
Kimura, N., Tanaka, S. & Okazawa, S. Study of Buckling/Post-Bucklnig Behaviours of Plate Structures Using
Shell-Solid Mixed Analysis. TEAM 2013, 2013 Keelung: Taiwan.
Kitarovic, S., anic, V. & Andric, J. 2013. Progressive collapse analyses of stiffened box- girders submitted to pure
bending. Brodogradnja, 64, 437455.
Koiter, W. T. 1945. Over de Stabiliteit van het elastche Evenwichi, Dissertation, Delft, Holland, Tarnslation. On the
Stability of Elastic Equilibrium, NASA TTF-10833,1967,and AFFDL-TR-7025,1970, Sec.7.7.
Kumar, M. S., Alagusundaramoorthy, P. & Sundaravadivelu, R. 2009. Ultimate strength of stiffened plates with a
square opening under axial and out-of-plane loads. Eng Struct., 31, 25682579.
Kutt, L. M., Piaszczyk, C. M., Chen, Y. K. & Liu, D. 1985. Evaluation of the longitudinal ultimate strength of
various ship hull configurations. Trans Soc Naval Archit Mar Eng., 93, 3353.
Lam, D., Gardner, L. & Burdett, M. 2010. Behaviour of axially loaded concrete filled stainless steel elliptical stub
columns. Adv. Struct. Eng., 13, 493500. .
Lee, C. H., Jang, C. H., Jun, S. H. & Oh, Y. T. 2014. Global Structural Analysis for Semi-submersible Drill Rig. The
Twenty-fourth International Ocean and Polar Engineering Conference, . Busan: Korea.
345
Lee, J., Soutis, C. & Kong, C. 2011. Prediction of compression-after-impact (CAI) strength of CFRP laminated
composites. ICCM International Conferences on Composite Materials.
Lee, J. S. 2012. On the Ultimate Compressive Strength of Stiffened Panels with Rectangular Opening. Proceedings
of 5th PAAMES and AMEC2012. Taiwan.
Lesani, M., Bahaari, M. R. & Shokrieh, M. M. 2013a. Detail investigation on un-stiffened T/Y tubular joints
behaviour under axial compressive loads. J. Constr. Steel Res., 80, 9199.
Lesani, M., Bahaari, M. R. & Shokrieh, M. M. 2013b. Numerical investigation of FRP-strengthened tubular T-joints
under axial compressive loads. Comp. Struct., 100, 7178.
Li, C., Ren, H., Zhao, Z., Zhao, X., Zeng, J. & Xu, D. 2013a. An approximate calculation approach for ultimate
strength of ship hull girder. Proceedings of the International Conference on Offshore Mechanics and Arctic
Engineering - OMAE.2013.
Li, G. C., Yang, Z. J., Lang, Y. & Fang, C. 2013b. Behaviour of high strength concrete filled square steel tubular
columns with inner CFRP circular tube under bi-axial eccentric loading. Adv. Steel Constr., 9, 231246.
Li, S. Q., Zhang, Q. J. & Zheng, J. J. 1993. Experimental investigation on dynamic buckling and plastic collapse of
rectangular plates subjected to solid-fluid in-plane impact. Acta mechanica sinica, 25, 249258.
Lie, S. T., Li, T. & Shao, Y. B. 2014. Plastic collapse load prediction and failure assessment diagram analysis of
cracked circular hollow section T-joint and Y-joint. Fatigue Fract. Eng. Mat. Struct., 37, 314324.
Lindemann, T. & Kaeding, P. 2013. Investigations on the Influence of Particular Components of the Idealized
Structural Unit Method (ISUM) for Collapse Analysis of Plate Structures. Proceedings of the Twenty-Third
(2013) International Offshore and Polar Engineering. Anchorage: Alaska: USA.
Lindemann, T., Klostermann, L. & Kaeding, P. 2013. Approaches to the Improvement of the Applicability of the
Idealized Structural Unit Method for Collapse Analysis of Plate Structures. Proceedings of the Twenty-Third
(2013) International Offshore and Polar Engineering. Anchorage: Alaska: USA.
Liu, B. & Wu, W. 2013. Standardized nonlinear finite element analysis of the ultimate strength of bulk carriers.
Wuhan Ligong Daxue Xuebao (Jiaotong Kexue Yu Gongcheng Ban). Journal of Wuhan University of
Technology (Transportation Science and Engineering), 37, 716719.
Liu, Z. & Amdahl, J. 2012. Numerical and simplified analytical methods for analysis of the residual strength of ship
double bottom. Ocean Engineering, 52, 2234.
Lloyds 2014. Lloyds Registe Rules and Regulations for the Classification of Offshore Units, July 2014., Lloyds
Register.
Lohning, T. & Muuholm, U. 2013. Finite element-based design of grouted connections with shear keys for offshore
wind turbines. Struct. Eng., 23, 295302.
Lu, H., Han, L. H. & Zhao, X. L. 2010. Fire performance of self-consolidating concrete filled double skin steel
tubular columns: experiments. Fire Saf. J., 45, 106115.
Ma, M., Hughes, O. & Paik, J. K. 2013. Ultimate Strength based Stiffened Panel Design using Multi-Objective
Optimization Methods and Its Application to Ship Structures. Proceedings of the PRADS2013. CECO.
Changwon City: Korea.
Mackay, J. R. & Van Keulen, F. 2013. Partial safety factor approach to the design of submarine pressure hulls using
nonlinear finite element analysis. Finite Elem. Anal. Des., 65, 116.
Magoga, T. & Flockhart, C. 2012. Ultimate strength calculations of an aluminium patrol boat. International
Maritime Conference 2012, Pacific 2012. RINA, Royal Institution of Naval Architects.
Magoga, T. & Flockhart, C. 2014. Effect of weld-induced imperfections on the ultimate strength of an aluminium
patrol boat determined by ISFEM rapid assessment method. Ships and Offshore Structures, 9, 218235.
Makouei, S. H., Teixeira, A. P. & Guedes Soares, C. G. 2014. An approach to estimate the ship longitudinal strength
using numerical databases of stress-strain curves of stiffened panels Developments in Maritime Transportation
and Exploitation of Sea Resources. In: 2013, P. O. I. (ed.) 15th International Congress of the International
Maritime Association of the Mediterranea Coruna: Spain.
Makouei, S. H., Teixeira, A. P. & Guedes Soares, C. G. 2015. A study on the progressive collapse behaviour of a
damaged hull girder, Maritime Technology and Engineering. London, Taylor & Francis Group.
Misirlis, K., Downes, J. & Dow, R. S. 2013. Ultimate strength of composite hulls under vertical bending. FAST
2013 - 12th International Conference on Fast Sea Transportation.
Moncada, A. M., Chattopadhyay, A., Bednarcyk, B. A. & Arnold, S. M. 2012. Micromechanics-based progressive
failure analysis of composite laminates using different constituent failure theories. Journal of Reinforced Plastics
and Composites, 31, 14671487.
Mouritz, A. P., Gellert, E., Burchill, K. & Challis, K. 2001. Review of Advanced Composite Structures for Naval
Ships and Submarines. Composite Structures, 53, 2141.
Muis Alie, M. Z. 2012. Residual Longitudinal Strength Analysis of Ships Hull Girder with Damages. International
Society of Offshore and Polar Engineers (ISOPE). Proceedings of the Twenty-second (2012) International
Offshore and Polar Engineering Conference.
Muis Alie, M. Z. 2013. Residual hull girder strength of asymmetrically damaged ships. , Osaka University.
Nam, J. M. & Choung, J. 2012. Residual Longitudinal Strength of a VLCC Considering Probabilistic Damage
Extents. Proceedings of the Twenty-second (2012) International Offshore and Polar Engineering Conference.
Rhodes.
Nam, J. M., Choung, J., Park, S. Y. & Yoon, S. W. 2014. Progressive collapse analysis and reliability of a damaged
hull girder. Proceedings of the ASME 2014 33rd International Conference on Ocean, Offshore and Arctic
Engineering. San Francisco, California: USA
346
Nam, M. & Choung, J. 2013. Assessment of Average Compressive Strengths Effect of Stiffened Panels on Hull
Girder Ultimate Longitudinal Strengths. Proceedings of the PRADS2013. CECO. Changwon City: Korea.
Nettles, A. T. & Jackson, J. R. 2013. Compression after impact strength of out-of-autoclave processed laminates.
Journal of Reinforced Plastics and Composites, 32, 18871894.
Nezamian, A. & Nezamian, M. 2014. A Review of Failure Modes in Ultimate Strength Assessment for Requalification and Life Extension of an Oil Field. The Twenty-fourth International Ocean and Polar Engineering
Conference. Busan: Korea.
Nikolov, P. I. 2012. Box girder strength under pure bending: Comparison of experimental and numerical results.
Sustainable Maritime Transportation and Exploitation of Sea Resources. Proceedings of the 14th International
Congress of the International Maritime Association of the Mediterranean, IMAM 2011, 2012.
Oliveira, D. Report of Committee .1.Ductile Collapse. Proceedings of the 10th International ship and Offshore
Structures Congress, 1988 Lynby, Denmark, . 315404.
Ozdemir, M. & Ergin, A. 2013. Nonlinear buckling behaviour of stiffened ship panels, Taylor & Francis, Espoo:
Finland.
Paik, J., Kim, D. K., Park, D. H., Kim, H. B. & Kim, M. 2012a. A new method for assessing the safety of ships
.
damaged by grounding. Trans. RINA., Part A1, Intl J Maritime Eng., 154, 120.
Paik, J. K. 2007. Ultimate strength of perforated steel plates under edge shear loading. Thin-Wall Struct., 45, 301306.
Paik, J. K. 2008. Ultimate strength of perforated steel plates under combined biaxial compression and edge shear
loads. Thin-Wall Struct., 46, 207213.
Paik, J. K. 2009. Residual ultimate strength of steel plates with longitudinal cracks under axial compression
experiments. Ocean engineering, 35, 17751783.
Paik, J. K., A.K., T. & Che, J. S. 1996. Ultimate strength hulls under combined vertical bending, horizontal bending,
and shearing forces. Trans SNAME, 104, 3159.
Paik, J. K., A.K., T., Pedersen, P. T. & Y., P. 2001. Ultimate strength of ship hulls under torsion. Ocean Eng., 28,
10971133.
Paik, J. K., Amlashi, H., Boon, B., Branner, K., Caridis, P., Das, P. & Fujikubo, M. 2012b. Report of ISSC
Technical Committee III.1 - Ultimate Strength. 18th International Ship and Offshore Structures Congress.
Paik, J. K., Andrieu, C. & Cojeen, H. P. 2008a. Mechanical collapse testing on aluminum stiffened plate structures
for marine applications. Marine Technology, 45, 228240.
Paik, J. K. & Kim, B. J. 2008. Progressive collapse analysis of thin-walled box columns. Thin-Wall Struct., 46, 541
550.
Paik, J. K., Kim, B. J. & Seo, J. K. 2008b. Methods for ultimate limit state assessment of ships and ship-shaped
offshore structures: Part III hull girders. Ocean Eng., 35, 281286.
Paik, J. K., Kim, B. J., Sohn, J. M., Kim, S. H., Jeong, J. M. & Park, J. S. 2012c. On buckling collapse a fusionwelded aluminium stiffened plate structure: An experimental and numerical study. Journal offshore Mechanics
and Arctic Engineering, 134, 021402021409.
Paik, J. K., Kim, D. K., Park, D. H., Kim, H. B., Mansour, A. E. & Caldwell, J. B. 2013. Modified PaikMansour
formula for ultimate strength calculations of ship hulls. Ships and Offshore Structures, 8, 245260.
Paik, J. K. & Mansour, A. E. 1995. A simple formulation for predicting the ultimate strength of ships. J. Mar. Sci.
Tech., 1, 5262.
Paik, J. K. & Sohn, J. 2011. Effects of Welding Residual Stresses on High Tensile Steel Plate Ultimate Strength:
Nonlinear Finite Element Method Investigations. J. Offshore Mech. Arct. Eng., 134, 021401021401.
Paik, J. K. & Thayamballi, A. K. 2003a. A concise introduction to the idealized structural unit method for nonlinear
analysis of large plated structures and its application. Thin-Wall Struct., 41, 329355.
Paik, J. K. & Thayamballi, A. K. 2003b. An experimental investigation on the dynamic ultimate compressive
strength of ship plating. International Journal of Impact Engineering, 28, 803811.
Park, J., Chun, M., Jeon, S., Kim, B., Jang, K., Suh, Y., Kim, J. & Lee, J. 2013. Advanced Guideline to Predict the
Ultimate Strength of Perforated Plate with Edge Reinforcement. Proceedings of the PRADS2013. CECO.
Changwon City: Korea.
Park, M. S., Jeong, Y. J. & Y.J., Y. 2014. Structural analysis of a hybrid substructure with multi-cylinder for 5MW
offshore wind turbines. Proceedings of the ASME 2014 33rd International Conference on Ocean, Offshore and
Arctic Engineering. San Francisco, California: USA.
Paulo, R. M. F., Teixeira-Dias, F. & Valente, R. a. F. 2013. Numerical simulation of aluminium stiffened panels
subjected to axial compression: Sensitivity analyses to initial geometrical imperfections and material propertie.
Thin-Walled Structures, 62, 6574.
Pedram, M. & Khedmati, M. R. 2014. The effect of welding on the strength of aluminium stiffened plates subject to
combined uniaxial compression and lateral pressure. International Journal of Naval Architecture and Ocean
Pei, Z., Iijima, K., Fujjikubo, M., Tanaka, Y., Tanaka, S., Okazawa, S. & Yao, T. 2013. Collapse behaviour of a
bulk carrier under alternate heavy loading conditions. International Journal of Offshore and Polar Engineering,
23, 224231.
Philippidis, T. P. & Antoniou, A. E. 2013. A progressive damage FEA model for glass/epoxy shell structures.
Journal of Composite Materials, 47, 623637.
Pollalis, C. & Samuelides, M. S. 2013. Ultimate strength of damaged hulls. Collision and Grounding of Ships and
Offshore Structures. Proceedings of the 6th International Conference on Collision and Grounding of Ships and
Offshore Structures, ICCGS 2013.
347
Prestileo, A., Rizzuto, E., Teixeira, A. P. & Guedes Soares, C. 2013. Bottom damage scenarios for the hull girder
structural assessment. Marine Structures, 33, 3355.
Pulos, J. G. & Salerno, V. L. (eds.) 1961. Axisymmetric elastic deformation and stresses in a ring-stiffened perfect
circular cylindrical shell under external hydrostatic pressure, Office of Naval Research, Bethesda: MD.
Qi, E., Cui, W. & Wan, Z. 2005. Comparative study of ultimate hull girder strength of large double hull tankers.
Mar Struct., 18, 22749.
Qian, X. 2013. Failure assessment diagrams for circular hollow section X- and K-joints. Int. J. Press. Ves. Piping,
104, 4356.
Qian, X., Jitpairod, K., Marshall, P. W., Swaddiwudhipong, S., Ou, Z., Zhang, Y. & Pradana, M. R. 2014. Fatigue
and residual strength of concrete-filled tubular X-joints with full capacity welds. J. Constr Steel Res., 100, 2135.
Qian, X., Li, Y. & Ou, Z. 2013a. Ductile tearing assessment of high-strength steel X-joints under in-plane bending.
Eng. Fail. Anal., 28, 176191.
Qian, X., Ou, Z., Swaddiwudhipong, S. & Marshall, P. W. 2013b. Brittle failure caused by lamellar splitting in a
large-scale tubular joint with fatigue cracks. Mar. Struct., 34, 185204.
Qian, X., Zhang, Y. & Choo, Y. S. 2013c. A load-deformation formulation with fracture representation based on the
J-R curve for tubular joints. Eng. Fail. Anal., 33, 347366.
Qu, H., Huo, J., Xu, C. & Fu, F. 2014. Numerical studies on dynamic behaviour of tubular T-joint subjected to
impact loading. Int. J. Impact Eng., 67, 1226.
Rnning, L., Aalberg, A. & Kristian Larsen, P. 2010. An experimental study of ultimate compressive strength of
transversely stiffened aluminium panels. Thin-Walled Structures, 48, 357372.
Saad-Eldeen, S., Garbatov, Y. & Soares, C. G. 2011. Experimental assessment of the ultimate strength of a box
girder subjected to severe corrosion. Marine Structures, 24, 338357.
Saad-Eldeen, S., Garbatov, Y. & Soares, C. G. 2012. Influence of weld toe shape and material models on the
ultimate strength of a slightly corroded box girder. . In: Resources, S. M. T. a. E. O. S. (ed.) Proceedings of the
14th International Congress of the International Maritime Association of the Mediterranean, IMAM 2011, 2012.
Saad-Eldeen, S., Garbatov, Y. & Soares, C. G. 2013a. Effect of corrosion severity on the ultimate strength of a steel
box girder. Engineering Structures, 49, 560571.
Saad-Eldeen, S., Garbatov, Y. & Soares, C. G. 2013b. Experimental assessment of corroded steel box-girders
subjected to uniform bending. Ships and Offshore Structures, 8, 653662.
Saad-Eldeen, S., Garbatov, Y. & Soares, C. G. 2013c. Ultimate strength assessment of corroded box girders. Ocean
Saad-Eldeen, S., Garbatov, Y. & Soares, C. G. 2014. Strength assessment of a severely corroded box girder
subjected to bending moment. Journal of Constructional Steel Research, 92, 90102.
Salminen, M. & Heinisuo, M. 2014. Numerical analysis of thin steel plates loaded in shear at non-uniform elevated
temperatures. J. Constr. Steel Res., 97, 105113.
Sawada, Y., Izumi, N. & Yoshikawa, T. 2012. Torsional Buckling Strength of Ring Stiffener in External Pressurized
Cylinder. Proceedings of 5th PAAMES and AMEC2012. Taiwan.
Saydam, D. & Frangopol, D. M. 2013. Performance assessment of damaged ship hulls. Ocean Engineering, 68, 65
76.
Sayyar, M., Soroushian, P. & Weerasiri, R. R. 2013. Prestressing effects on the ultimate flexural strength of
composite box sections. Composite Structures, 106, 806814.
Seifi, R. & Khoda-Yari, N. 2011. Experimental and numerical studies on buckling of cracked thin-plates under full
and partial compression edge loading. Thin-Walled Structures, 49, 15041516.
Sensharma, P., Collette, M. D. & Harrington, J. 2011. Effect of welded properties on aluminum structures. Ship
Structure Committee Report 2011.
Shanmugam, N. E., Zhu, D. Q., Choo, Y. S. & Arockiaswamy, M. 2014. Experimental studies on stiffened plates
under in-plane load and lateral pressure. Thin-Wall Struct., 80, 2231.
Shariati, M. 2012. Experimental and numerical investigation of buckling in rectangular steel plates with grooveshaped cutouts. Journal of Zhejiang University-Science A; Applied Physics & Engineering, 13.
Sheehan, T., Dai, X. H., Chan, T. M. & Lam, D. 2012. Structural response of concrete filled elliptical steel hollow
sections under eccentric compression. Eng. Struct., 45, 314323.
Shi, G. J. & Wang, D. Y. 2012a. Residual ultimate strength of cracked box girders under torsional loading. Ocean
Eng., 43, 102112.
Shi, G. J. & Wang, D. Y. 2012b. Residual ultimate strength of open box girders with cracked damage. Ocean Eng.,
43, 90101.
Shu, Z. & Moan, T. 2012. Ultimate hull girder strength of a bulk carrier under combined global and local loads in
the hogging and alternate hold loading condition using nonlinear finite element analysis. Journal of Marine
Silva, J., Garbatov, Y. & Soares, C. G. 2013. Ultimate strength assessment of rectangular steel plates subjected to a
random localised corrosion degradation. Engineering Structures, 52, 295305.
Smith, C. S. 1977. Influence of local compressive failure on ultimate longitudinal strength of a ships hull. 3th Int.
Symposium on Practical Design in Shipbuilding. Tokyo.
Smith, C. S., Anderson, N., Chapman, J. C., Davidson, P. C. & Dowling, P. J. 1991. Strength of stiffened plating
under combined compression and lateral pressure. Trans. R. Inst. Nav. Archit., 131147.
So, H. & Chen, J. T. 2007. Experimental study of dynamic crushing of thin plates stiffened by stamping with Vgrooves. International Journal of Impact Engineering, 34, 13961412.
348
Storheim, M. & Amdahl, J. 2014. Design of offshore structures against accidental ship collisions. Marine Structures
37, 37, 135172.
Sun, H. H. & Guedes Soares, C. 2003. An experimental study of ultimate torsional strength of a ship-type hull girder
with a large deck opening. Marine Structures, 16, 5167.
Sundarraja, M. C. & Prabhu, G. G. 2012. Experimental study on CFST members strengthened by CFRP composites
under compression. J. Constr. Steel Res., 72, 7583.
Syrigou, M. S., Benson, S. & Dow, R. S. 2014. Strength of ship plating under combined shear and compression.
Proceedings of International Conference On Safety & Reliability of Ship, Offshore & Subsea Structures.
Glasgow: UK.
Tahmasebi Nejad, A. & Shanmugam, N. E. 2011. Elastic buckling of uniaxially loaded skew plates containing
openings. Thin-Walled Struct., 49, 12081216.
Tan, K. H., Fung, T. C. & Nguyen, M. P. 2013. Structural behaviour of CHS T-joints subjected to brace axial
compression in fire. J. Struct. Eng.-ASCE, 139, 7384.
Tanaka, S., Yanagihara, D., Yasuoka, A., Haradac, M., Okazawa, S., Fujikubo, M. & Yao, T. 2014. Evaluation of
ultimate strength of stiffened panels under longitudinal thrust. Mar Struct., 36, 2150.
Taylor, S. C., Magota, T. & Aksu, S. 2013. Residual strength assessment of an aluminium naval platform. In: Rina,
R. I. O. N. A. (ed.) International Conference on Damaged Ship II.
Tayyar, G. T. & Bayraktarkatal, E. 2012. A new approximate method to evaluate the ultimate strength of ship hull
girder. In: Resources, S. M. T. a. E. O. S. (ed.) Proceedings of the 14th International Congress of the
International Maritime Association of the Mediterranean. Genova: Italy.
Teixeira, A. P., L.D., I. & Soares, C. G. 2013a. Assessment of characteristic values of the ultimate strength of
corroded steel plates with initial imperfections. Engineering Structures 517527.
Teixeira, A. P., Soares, C. G. & Wang, G. 2013b. Probabilistic modelling of the ultimate strength of ship plates with
non-uniform corrosion. J. Mar. Sci. Technol., 18, 115132.
Tekgoz, M., Garbatov, Y. & Soares, C. G. 2013a. Advanced Guideline to Predict the Ultimate Strength of
Perforated Plate with Edge Reinforcement. Proceedings of the PRADS2013. CECO. Changwon City: Korea.
Tekgoz, M., Garbatov, Y. & Soares, C. G. 2013b. Finite element modelling of the ultimate strength of stiffened
plates with residual stresses. Analysis and Design of Marine Structures. Espoo: Finland: Taylor & Francis.
Toh, K., Maeda, M. & Yoshikawa, T. 2012. The effect of initial imperfection on the hull girder ultimate strength of
intact and damaged ships. Proceedings of the International Offshore and Polar Engineering Conference.
Toh, K., Maeda, M. & Yoshikawa, T. 2014. The Effect of Initial Imperfections and Lateral Loads on the Hull Girder
Ultimate Strength of Intact and Damaged Ships. 22th International Offshore and Polar Engineering Conference.
Tran, K. L., Davaine, L., Douthe, C. & Sab, K. 2012. Stability of curved panels under uniform axial compression. J.
Constr. Steel Res., 69, 3038.
Turner, M. J., Clough, R. W., Martin, H. C. & Topp, L. J. 1956. Stiffness and deflection analysis of complex
structure. J. Aeronautical Science, 23, 805823.
Ueda, Y. & Rashed, S. M. H. 1974. An ultimate transverse strength analysis of ship structure. J Soc Naval Arch
Japan, 13, 87104.
Ueda, Y. & Rashed, S. M. H. 1991. Modern Method of Ultimate Strength Analysis of Offshore Structure. Int J
Offshore and Polar Eng., ISOPE
Ueda, Y., Rashed, S. M. H., Nakacho, K. & Sasaki, H. 1983. Ultimate strength analysis of offshore structuresapplication of idealized structural unit method. J Kansai Soc Naval Architects Japan, 190, 13142.
Ueda, Y. & Yao, T. 1985. The influence of complex initial deflection modes on the behaviour and ultimate strength
of rectangular plates in compression. J. Construtional Steel Research, 5, 265302.
Uenaka, K. 2014. Experimental study on concrete filled elliptical/oval steel tubular stub columns under
compression. Thin Wall Struct., 78, 131137.
Uenaka, K. & Kitoh, H. 2011. Mechanical behaviour of concrete filled double skin tubular circular deep beams.
Thin Wall Struct., 49, 256263.
Underwood, J. M., Sobey, A. J., Blake, J. I. R. & Shenoi, R. A. 2012a. Determination of critical factors for the
damaged strength assessment of steel grillages. Proceedings of the International Offshore and Polar Engineering
Conference.
Underwood, J. M., Sobey, A. J., Blake, J. I. R. & Shenoi, R. A. 2012b. Ultimate collapse strength assessment of
damaged steel-plated structures. Eng Struct., 38, 110.
Van Es, S. H. J., Gresnigt, A. M., Kolstein, M. H. & Bijlaard, F. S. K. 2013. Local Buckling of Spirally Welded
Tubes. Twenty-Third Int. Offshore Polar Eng. Conf.
Wang, J. & Zang, S. 2011. Ultimate Strength Experimental Research for Longitudinal Box Girders Module Model.
Shipbuilding of China 2, 008.
Wang, R., Han, L. H. & C.C., H. 2013. Behaviour of concrete filled steel tubular (CFST) members under lateral
impact: experiment and FEA model. J. Constr. Steel Res., 80, 188201.
Wang, X., Qian, X., Liew, J. Y. R. & Zhang, M. H. 2014. Experimental behaviour of cement filled pipe-in-pipe
composite structures under transverse impact. Int. J. Impact Eng., 72, 116.
Windenburg, D. F. & Trilling, C. 1934. Collapse by instability of thin cylindrical shells under external pressure.
Trans, ASME., 56, 819825.
Wu, H., Wu, W., Gan, J. & Sun, H. 2013. Ultimate strength analysis of a river-sea ship under combined action of
torsion and bending. OMAE 2013.
349
Xu, J. X., Zhao, J. C., Song, Z. S. & Liu, M. L. 2012. Prediction of ultimate bearing capacity of tubular T-joint
under fire using artificial neural networks. Safety Sci., 50, 14951501.
Xu, M. C., Garbatov, Y. & Soares, C. G. 2013a. Ultimate strength assessment of a tanker hull based on
experimentally developed master curves. Journal of Marine Science and Application, 12, 127139.
Xu, M. C. & Soares, C. G. 2013a. Experimental Study on the Collapse Strength of Narrow Stiffened Panels. Journal
of Offshore Mechanics and Arctic Engineering, 135, 02140210.
Xu, M. C. & Soares, C. G. 2013b. Experimental study on the collapse strength of wide stiffened panels. Marine
Xu, M. C., Yanagihara, D., Fujikubo, M. & Guedes Soares, C. 2013b. Influence of boundary conditions on the
collapse behaviour of stiffened panels under combined loads. Mar. Struct., 34, 205225.
Yamada, Y. 2014. Numerical Study on the Residual Ultimate Strength of Hull Girder of a Bulk Carrier after ShipShip Ccollision. Proceedings of the ASME 2014 33rd International Conference on Ocean, Offshore and Arctic
Engineering, OMAE2014. San Francisco, California: USA.
Yamamoto, Y. 1965. General instability of a reinforced cylindrical shell under external pressure. Journal of the
society of naval architecture of Japan 113.
Yang, N., Das, P. K. & Yao, X. 2011. Ultimate strength and reliability assessment of laminated composite plates
under axial compression. Ships and Offshore Structures, 6, 105113.
Yang, Q. J., Hayman, B. & Osnes, H. 2013. Simplified buckling and ultimate strength analysis of composite plates
in compression. Composites Part B: Engineering, 54, 343352.
Yao, T. & Nikolov, P. I. 1992. Progressive collapse analysis of a ships hull girder under longitudinal bending (2nd
report). J. of Soc. Naval Arch. of Japan, 172, 437446.
Yasuoka, A., Tanaka, S., Yanagihara, D., Okazawa, S. & Yao, T. 2012. Collapse Behaviour and Ultimate Strength
of Stiffened Panels with Arbitrary Number of Stiffeners Subjected to Longitudinal Thrust. Presented at the The
Twenty-second International Offshore and Polar Engineering Conference. International Society of Offshore and
Polar Engineers.
Ye, Q., Jin, W. L., He, Y., Shi, Z. M. & Qu, Y. 2012. Assessment of ultimate strength of semi-submersible platform.
Journal of Ship Mechanics, 16, 277295.
Yoshikawa, T. & Yoshimura, K. 2006. General Buckling Strength of Ring-Stiffened Cylindrical Shells under
External Pressure and Effective Breadth of Ring-Stiffeners. Journal of The Japan Society of Naval Architects and
Engineers, 3, 253260.
Yoshikawa, T. & Yoshimura, K. 2008. General Buckling Strength of Ring-Stiffened Cylindrical Shells under
External Pressure and Effective Breadth of Ring-Stiffeners. Proceedings of OMAE2008, 27th International
Conference on Offshore Machanics and Arctic Engineering.
Yu, C. & Lee, J. S. 2012. Ultimate Compressive Strength of Unstiffened Plate with Rectangular Opening.
Proceedings of the 26th Asian-Pacific Technical Exchange and Advisory Meeting on Marine Structure. Kyushu
University, Fukuoka: Japan.
Yu, W. J., Zhao, J. C., Luo, H. X., Shi, J. Y. & Zhang, D. X. 2011. Experimental study on mechanical behaviour of
an impacted steel tubular T-joint in fire. J. Constr. Steel Res., 67, 13761385.
Yudo, H. & Yoshikawa, T. 2012. Mechanical Behaviour of Pipe under Pure Bending Load. . Proceedings of the
26th Asian-Pacific Technical Exchange and Advisory Meeting on Marine Structure. Kyushu University,
Fukuoka: Japan.
Yudo, H. & Yoshikawa, T. 2014. Buckling phenomenon for straight and curved pipe under pure bending. J. Mar.
Sci. Technol., 110.
Zahurul Islam, S. M. & Young, B. 2011. FRP strengthened alumnium tubular sections subjected to web cripplin.
Thin-Walled Structures, 49, 13921403.
Zareei, M. R., Khedmati, M. R. & Rigo, P. 2012. Application of artificial neural networks to the evaluation of the
ultimate strength of uniaxially compressed welded stiffened aluminium plates. Proceeding of the Institution of
Mechanical Engineering. Journal of Engineering for the Maritime Environment.
Zayed, A., Garbatov, Y. & Soares, C. G. 2013. Reliability of ship hulls subjected to corrosion and maintenance.
Structural Safety, 43, 111.
Zha, Y. & Moan, T. 2003. Experimental and numerical prediction of collapse of flatbar stiffeners in aluminum
panels. Journal of Structural Engineering, 129, 160168.
Zhang, J., Shen, Z., Ji, C. & Yin, Q. 2012. The influence research of ultimate strength to deepwater semisubmersible platforms structure under corrosion damage. Proceedings of the International Conference on
Offshore Mechanics and Arctic Engineering.
Zhang, S. 2013. Buckling and ultimate strength assessments of ship structures. International Conference on
Offshore Mechanics and Arctic Engineering - 2013. San Francisco.
Zhang, W., Luo, X. & Wu, W. 2013. Ultimate strength of river-sea container ship. Marstruct. Finland.
Zhao, X. L. & Tong, L. W. 2011. New development in steel tubular joints. Adv.Struct. Eng., 14, 699715.
Zhu, B. & Frangopol, D. M. 2013. Reliability assessment of ship structures using Bayesian updating. Engineering
Structure, 56, 18361847.
Zubair, M., Alie, M., Fujikubo, M., Iijima, K., Oka, S. & Takemura, K. 2012. Residual longitudinal strength analysis of
ships hull girder with damages. Proceedings of the International Offshore and Polar Engineering Conference.

710 SEPTEMBER 2015
CASCAIS, PORTUGAL
VOLUME 1
COMMITTEE III.2
FATIGUE AND FRACTURE

COMMITTEE MANDATE
Concern for crack initiation and growth under cyclic loading as well as unstable crack propagation and
tearing in ship and offshore structures. Due attention shall be paid to practical application and statistical
description of fracture control methods in design, fabrication and service. Consideration is to be given to the
suitability and uncertainty of physical models.
COMMITTEE MEMBERS
Chairman:
F. P. Brennan, UK
K. Branner, Denmark
J.H. den Besten, The Netherlands
A. M. Horn, Norway
B. Ki Choi, Korea
W. Mao, Sweden
T. Nakamura, Japan
E. Oterkus, UK
I. Paranhos Pasqualino, Brazil
G. Parmentier, France
C. Rizzo, Italy
J. Romanoff, Finland
J. Rrup, Germany
A. Samanta, India
A. Theodoulidis, Greece
F. Wang, China
G. Wang, Singapore
EXTERNAL CONTRIBUTOR
N. Chen, USA
KEYWORDS
Fatigue, Fracture Mechanics, Probabilistic Methods, Materials, Steel, High Strength Steel, Stainless Steel,
Aluminium, Inspection, Fitness for Service.
352
ISSC committee III.2: FATIGUE AND FRACTURE
CONTENTS
1.
INTRODUCTION ..................................................................................................................... 354
2.
FATIGUE LIFE-CYCLE DESIGN PHILOSOPHIES AND METHODOLOGIES ............. 354

2.1 Fatigue and fracture in marine structures ....................................................................... 354
2.2 Preliminary Design ......................................................................................................... 354
2.3 Detailed Design............................................................................................................... 354
2.4 Fabrication ...................................................................................................................... 355
2.5 In-service Maintenance................................................................................................... 355
2.5.1 Inspection techniques ........................................................................................ 355
2.5.2 Inspection planning............................................................................................ 355
2.6 Fatigue strength............................................................................................................... 355
2.6.1 S-N curves related to expected workmanship .................................................. 355
2.6.2 Crack propagation parameters........................................................................... 355
2.7 Fracture Strength............................................................................................................. 356
2.8 Fatigue Loads .................................................................................................................. 356
2.8.1 Wave loads......................................................................................................... 356
2.8.2 Loading unloading ............................................................................................. 356
2.8.3 Vibrations........................................................................................................... 356
2.9 Environmental effects ..................................................................................................... 356
2.9.1 In air ................................................................................................................... 357
2.9.2 Seawater ............................................................................................................. 357
2.9.3 Other aggressive environments ......................................................................... 357
2.9.4 Coating and coating life..................................................................................... 357
2.10 Fatigue, Fracture & Failure Criteria ............................................................................... 357
2.10.1 Failure definition ............................................................................................... 357
2.10.2 Uncertainties ...................................................................................................... 357
2.10.3 Safety Factors .................................................................................................... 358
3.
FACTORS INFLUENCING FATIGUE/FRACTURE ............................................................ 358

3.1 Resistance........................................................................................................................ 358
3.1.1 Thickness and size ............................................................................................. 358
3.1.2 Environment (corrosion) ................................................................................... 359
3.1.3 Temperature ....................................................................................................... 362
3.1.4 Residual Stress & Constraint, mean stress........................................................ 363
3.2 Materials.......................................................................................................................... 364
3.2.1 Metallic Alloys .................................................................................................. 364
3.2.2 Fatigue & fracture improvements through material changes, surface
treatment ............................................................................................................ 364
3.3 Loading ........................................................................................................................... 365
3.3.1 Stochastic Loading (load interaction effects (sequence))................................. 365
3.3.2 Cycle counting spectral, time-domain, stress ranges, means stress effect .... 365
3.3.3 Complex stresses ............................................................................................... 366
3.3.4 Recent developments in multiaxial fatigue criteria .......................................... 369
3.4 Structural Integrity/Life cycle management .................................................................. 373
3.4.1 Fabrication and repair ........................................................................................ 373
3.4.2 Inspection & Monitoring of structure and coatings .......................................... 374
3.4.3 Inspection and Maintenance .............................................................................. 376
3.5 Composites ...................................................................................................................... 377
353
4.
FATIGUE ASSESSMENT METHODS .................................................................................. 378

4.1 Overview ......................................................................................................................... 379
4.2 Fatigue damage models .................................................................................................. 381
4.2.1 Stress based concepts ........................................................................................ 381
4.2.2 Strain concepts ................................................................................................... 382
4.2.3 Notch-intensity factor, -integral and -energy density concepts........................ 382
4.2.4 Confidence and reliability ................................................................................. 383
4.3 Fracture mechanics models ............................................................................................ 385
4.3.1 Crack growth rate model ................................................................................... 389
4.3.2 Crack growth assessment .................................................................................. 390
4.3.3 Fracture mechanics based fatigue evaluation of ship structures ...................... 391
4.4 Rules, standards & guidance .......................................................................................... 392
4.4.1 Ship rules ........................................................................................................... 392
4.4.2 Design codes for Offshore Structures ............................................................... 394
4.4.3 IIW recommendation ......................................................................................... 395
4.4.4 ISO Standards .................................................................................................... 395
4.5 Acceptance criteria ......................................................................................................... 395
4.6 Measurement techniques ................................................................................................ 396
4.6.1 Crack growth and propagation .......................................................................... 396
4.6.2 Fatigue................................................................................................................ 397
4.6.3 Material properties ............................................................................................. 398
4.6.4 Fracture toughness ............................................................................................. 398
5.
BENCHMARKING STUDY .................................................................................................... 399

5.1 Problem Statement .......................................................................................................... 399
5.2 Analytical Methods ......................................................................................................... 400
5.3 Numerical analysis using FEM ...................................................................................... 402
5.4 Results ............................................................................................................................. 403
5.5 Discussion & Benchmarking Study Conclusions .......................................................... 404
6.
SUMMARY & CONCLUSIONS ............................................................................................. 404
REFERENCES .................................................................................................................................. 405
354
1.
INTRODUCTION
This report of the current committee describes the recent activities of the international ship and offshore
industry and the researchers that support it, with specific regard to current pertinent issues and trends
relating to Fatigue and Fracture. It is important to remember that the subject area is vast and developing
rapidly, and that this report should be seen not only in the context of the entire ISSC 2015 Proceedings
but also as a continuation of past ISSC reports.
In addition, the committee chose to focus on areas commensurate with the expertise of the committee
members to build on the vast knowledge base generated by previous Committee III.2 reports. This report in
addition to providing a critique of contemporary research and design methods applicable to fatigue and
fracture of ship and offshore structures undertook a detailed Benchmarking Study associated with very thin
walled plate structures, a topic gaining prominence as lower weight structures become more commonplace.
An important element of ISSC work is to provide an expert opinion on the subject matter reported and
readers will find such comments and recommendations interwoven into the report; the most prominent
however, are summarised in Section 6.
The report is structured so that Section 2 summarises from a designers perspective Fatigue Life-Cycle
Design Philosophies and Methodologies aimed to be a gentle introduction to the report and setting the
scene for the following two major review sections: Factors Influencing Fatigue/Fracture and Fatigue
Assessment Methods. The former focuses on the generic underpinning scientific literature focusing upon
subjects of particular relevance to current and emerging trends in structures operating in the marine
environment. The latter section reviews published work associated with assessment methods, both of
new design and build, but also for life-extension and reanalysis of fatigue life following unexpected
events.
The Benchmarking study illustrates the absolute necessity to perform analyses carefully and for rules,
design equations and interpretations to be clear and unambiguous as the fundamental logarithmic
relationship between stress and life inherently mean that life predictions are highly sensitive to local stress
and the accuracy to which stress is calculated is critical.
2.
FATIGUE LIFE-CYCLE DESIGN PHILOSOPHIES AND

METHODOLOGIES
Fatigue and fracture phenomena have been known about for a long time, illustrated, for example by the
breaks in the liberty ships during World War II.
2.1
Fatigue and fracture in marine structures
In current marine structures, fatigue, from local leakage to total structure brittle fracture and with the
increase of steels yielding stress, becomes more and more critical. Even if leakage consequences are
generally minor concerning immediate structural strength, pollution resulting from oil loss or other
aggressive cargoes may lead to an environmental disaster. Brittle fracture may also lead to a sudden loss
of the marine structure.
2.2
Preliminary Design
At the preliminary design stage, the objective of the shipyard is to make a quotation. At this stage the
issue is not to assess exhaustively the fatigue strength of the structure, as the details have not been
designed at this stage, but to assess that the global scantling is sufficient, and that the yard can safely
evaluate the structure building cost, before the order and in the next step the planned material. Generally
only the global hull girder fatigue strength is assessed at this stage, but the relevant primary structure may
also be assessed.
2.3
Detailed Design
At the detailed design stage, the structural area with high concentration factors and submitted to
significant load ranges is assessed versus fatigue strength. Depending on the case, when local structures
do not comply with fatigue strength requirements, the local design is modified. In principle, if the
assessment at the preliminary stage has been correctly carried out, target fatigue strength can easily be
reached by improving the local design details. Ship and marine structures are very complex and the
number of details with significant stress concentration factors is very large and therefore the fatigue
assessment of all details is not realistic. Shortlisting the locations where fatigue assessment is to be
carried out is one of the most critical tasks in marine structure fatigue assessment.
2.4
355
Fabrication
Fabrication will never carry out exactly what design has planned. There are misalignments, welding
defects, but also sometimes thicker sections used due to supply issues. Tolerance criteria and effects due
to workmanship are not always explicitly defined in the rules. In this regard, Jonsson et al. (2013) has
analysed the parameters relevant for fatigue among the parameters considered by inspection standards.
2.5
In-service Maintenance
Fatigue damage, on actual marine structures can neither be detected nor measured during initiation, but
during crack propagation the size of the flaw should be sufficient to be detected by non-destructive
techniques. The efficiency of maintenance depends on inspection efficiency and inspection planning
allied to inspection reliability.
2.5.1
Due to the complexity of marine structures, exhaustive close inspections are not always possible
everywhere.
Inspection techniques are chosen according to the cost, including delivery time constraints, and
efficiency/reliability. At the design stage, inspection areas are defined according to the probability
of failure and consequences of that failure. Efficiency in terms of ability of crack detection is
strongly dependent on the chosen Non Destructive Examination (NDE) approach. Cost of
inspection is also to be considered as resources are limited; cost includes not only the cost of the
inspection itself, but the time of surveyors, inspection devices and work required to make the
structure accessible, e.g. for passenger ships or navy ships, as well as the cost due to the
unavailability of the ship or marine structure during the inspection.
2.5.2
Inspection planning
From the above information at the design stage, it is necessary to assess that, according to the planned
inspection programme, a possible crack can be discovered and repaired before the occurrence of an
unacceptable event. It is, for example the case for LNG carriers with their constant monitoring of
potential gas leakage, where the designer has to demonstrate that no tank collapse will occur before a
harbour has been reached.
2.6
Fatigue strength
Fatigue strength may be either evaluated in terms of ability to resist damage accumulation or the ability to
resist crack propagation up to a maximum allowable crack size. Damage accumulation has no clear
physical meaning, but is very easy to apply, while crack propagation corresponds to a physical
phenomenon in principle well defined, but difficult to apply except for time history loading. In terms of
design, the most suitable model may differ from the physical phenomenon as the easiest model will be
used as long as the resulting failure prediction does not differ too much from the test results.
2.6.1
S-N curves related to expected workmanship
The S-N curve models using a Miner damage accumulation approach, corresponds predominantly to an
initiation model. This model is easily applied to simple details and reflects the welding quality effects
using the most applicable S-N curve for the welding process used, including when relevant, the post
welding process. The results are less convincing for complex details and design rules provide
correction factors, e.g. they propose a weighted combination of membrane and bending loads to fit test
results.
2.6.2
Crack propagation parameters
Crack propagation is, up to now, the only approach that allows estimating the residual life when a crack
has been discovered. Unfortunately more often than not, the residual life corresponding to a detectable
crack is too low to be economically acceptable as a design life. Alternative approaches have been
proposed defining an empirical initial crack size. Most of the life corresponds to propagation in short
crack conditions that are very complex to model and current approaches do not take into account this
complexity, meaning that this initial crack size determination is questionable as it is generally
recalculated retrospectively from S-N data. Final fatigue life estimation can be very sensitive to the
accuracy of the initial crack size.
356
2.7
Fracture Strength
Fracture strength is not up until now assessed for marine structures as material specifications are
generally supposed to fulfil the assessment of fracture strength. Due to the increase of high tensile steel
plate thickness in the largest container ships, fracture strength has become more and more critical.
Measures are required to stop any possible sudden crack propagation, either by careful structural
arrangement or by checking a materials ability to arrest crack propagation.
2.8
Fatigue Loads
Fatigue loads are highly complex and in order to simplify design rules, those that contribute the most are
considered neglecting the others. Wave loads are currently considered on all marine structures, whereas
other loadings may be considered, e.g. loading/unloading loads may in some cases have a significant
contribution.
2.8.1
Wave loads
Knowing the future actual route, the corresponding climatology, speed and sea-state distribution, relative
heading and encountering frequency can be deduced and then global and local loads applied to the marine
structure. At the design stage, generally the ship route is unknown except for specific ships such as fast
ferries with a dedicated area and the design climatology is necessarily arbitrary. For offshore structures
the site of operation is known and is an essential part of the design.
2.8.1.1
Definition of a fatigue load Standard
Taking into account the difficulty in anticipating the actual load applied to a ships structure, the rules aim
to define a standard load condition as a reference for checking fatigue strength.
2.8.1.1.1
Worst case e.g. North Atlantic
The first possible approach is to select a conservative navigation area. Generally IACS Recommendation
34 is the base of scantling for yielding ultimate strength and fatigue. The assumption for speed, e.g. 0.66
of maximum speed, remains questionable and should be considered case by case with attention to the type
of ship, which includes the propulsive power of the ship, mission profile, actual service at sea, and
likelihood of vibration (springing or slamming).
2.8.1.1.2
Specific worldwide, or local sea-state model
Loads resulting from the above approach generally provide very conservative results and do not cater for
existing ships that have not experienced fatigue issues. An alternative scatter diagram has been proposed
with a less severe wave height; the so-called Worldwide scatter diagram. Ferry ships, particularly fast
ferry ships, may be dedicated to a single route or a single route with a limited navigation, conditional on
the sea-state. Offshore structures are generally designed for a given site and a specific climatology
analysis is carried out.
2.8.1.1.3
Storms approach
When crack propagation is used with random loading it results in fatigue life prediction being dependent
on the order of cycles. It has been shown that organising the sequence of cycles in terms of storm and not
in terms of short-term does not change the mean value of fatigue life but increases significantly the
variance of fatigue life and consequently requires a larger safety margin.
2.8.2
Loading unloading
For a marine structure operating in a protected area and with frequent loading/unloading (e.g. FSO
(floating storage and offloading vessel) or Dredger), fatigue may be slightly influenced by the static
load variation or this static load may be the major source of fatigue.
2.8.3
Vibrations
The vibration effect due to slamming or springing is not commonly assessed, but for large ships such as
the recent largest container carriers or high speed craft, the vibration effect will have a significant effect
mainly due to its interaction with quasi-static loading.
2.9
References or base-line fatigue testing is carried out to establish S-N curves in dry air and in seawater
environments.
2.9.1
357
In air
The reference curve has been established in air and in principle, corresponding to a probability of failure
less than 2.5%. There is a change of slope conservatively assumed to correspond to 107 cycles. The
change of slope takes into account both the fatigue limit and the weakening due to damage of higher
number of cycles associated with random loading. It is generally assumed that fatigue in air is not
significantly influenced by loading frequency for specimens corresponding to typical details, the issue
being more in the control of loading.
2.9.2
Seawater
The actual phenomenon of fatigue strength weakening due to corrosion is not fully understood. It is not
easy to carry out representative testing of fatigue under corrosion as fatigue strength is very dependent on
test duration. Design rules have introduced empirical factors to ensure the design is conservative. As an
example, the IACS harmonized Common Structural Rules have chosen to modify the slope of the S-N
curve, keeping a single slope, disregarding the fatigue limit and dividing the lifetime by two for the higher
stress level. For offshore structure the factors are different.
2.9.3
Other aggressive environments
Sea water is not the most aggressive environment for metallic structures. Particularly for offshore
structures, oil with the hydrogen sulphide effects should also be considered. There is no general corrective
factor; it depends either on the corrosive element, the type of detail or material. Tests are to be carried out
and results treated with a suitable methodology, with respect e.g. to the test duration versus the structure
exposed time etc. It is generally difficult to define the slope of the S-N curve in such an aggressive
environment due to the sensitivity of the results to the test duration.
2.9.4
Coating and coating life
Coating is a way to protect the structure against corrosion. As long as the coating protection is efficient it
can be assumed that the fatigue strength corresponds to the fatigue strength in air; when the coating fails,
it is no longer efficient, then the fatigue strength is reduced and close to the fatigue strength in a corrosive
environment. Although some requirements are undertaken to improve coating life, the link between
coating life and assumptions of coating protection in the structural fatigue strength assessment remains
not fully understood. Moreover, attention should be paid to the link between improvement techniques and
coating protection as improvements become ineffective when the coating fails. According to the operating
conditions, it may be unrealistic to consider that an effective coating protection can be maintained. For
example in bulk carrier ship holds, due to the loading/unloading process, it is unrealistic to maintain
coating where ore is abrasive or scrapers are used to assist discharge.
2.10
Fatigue, Fracture & Failure Criteria
Fatigue is a progressive weakening process and acceptance criteria are neither obvious nor generic.
2.10.1
Failure definition
Fatigue criteria depends both on consequences of failure and on the accepted risk of failure. In the
offshore industry three different level of failure are defined: N1 where the stress range is measured by a
strain gauge decrease of 10% and considered as corresponding roughly to initiation; N2 corresponding to
a through thickness crack; and N3 corresponding to the end of the test due to too large displacements. For
simple design details there are small differences in life between these fatigue failure definitions, on the
other hand for complex details, N1 N2 and N3 may result in very different lives.
2.10.2
Uncertainties
For a given detail, i.e. welded or cut detail, for many reasons there is generally a large scatter in the S-N
curves. Most of the sources of uncertainty cannot be controlled either during design or the fabrication
process. Probability models related to these uncertainties have been widely discussed. From a practical
point of view actual fatigue strength under constant amplitude range is considered as a random parameter.
A common approach is that the uncertainty of the fatigue life follows a lognormal law but alternative
approaches exist, e.g. the Bastenaire approach where the uncertainty depends on the parameter of the S-N
curve. In addition to this uncertainty for constant stress range fatigue strength, there is also an uncertainty
for variable stress ranges as applied using the damage accumulation law, i.e. Miner law. Crack growth
retardation after overload is one of the phenomena contributing to this model uncertainty.
358
2.10.3
Safety Factors
To take into account the uncertainty in the design rules, safety factors are used. Safety factors may be
explicit by requiring that the total fatigue damage be less than 0.5 or even 0.1. They can also be implicit
defining a navigation condition as the worst possible condition or defining the design S-N curve as an SN curve with a probability less than that of a given proportion of the specimen of a sample which has a
fatigue life lower than the design curve. There needs to be a rational approach to define this safety factor
in order to adapt these factors to new materials or new fabrication processes. Probabilistic methods have
been developed to optimize this safety factor. For that reason, the consequences of fatigue failure are to
be evaluated, and an acceptable probability of failure clarified. Then a set of partial safety factors may be
defined to be applied in respect of design equations in order to reach the target safety level and to obtain a
failure probability that is as stable as possible versus the different possible designs.
3.
FACTORS INFLUENCING FATIGUE/FRACTURE
A large number of factors influence the fatigue life of component or structure as discussed in the previous
section. In the following subsections a comprehensive literature review of the main factors are critiqued
under the headings:
Resistance which address thickness and size effects, environmental impact, temperature effect and
residual end mean stresses;
Material which address fatigue and fracture improvement through microstructure and surface
treatments;
Loading which address different types of fatigue loading such as stochastic loading and multiaxial
stresses;
Structural Integrity which address fabrication, inspection, repair and different monitoring techniques.
3.1
3.1.1
Resistance
Thickness and size
Fatigue design methods for very thin deck plates, down to 3mm, have recently received considerable
interest due to increasing demands for energy efficiency and requirements for a more efficient use of
natural resources. The production of these panels is nowadays possible due to low-heat-input welding
techniques such as laser-; laser-hybrid (Fricke et al., 2013b) and friction stir welding (Azevedo et al.,
2014). Many classification societies consider the 5mm thickness as a threshold below which special
attention should be paid to the manufacturing and welding methods. Especially, the permissible tolerances
for axial and angular misalignment should not be exceeded. Recently considerable efforts have been made
to investigate the fatigue resistance of thin-welded deck structures based on nominal, structural and notch
stress methods; see for example the PhD theses of Molter (2013), Lilleme (2014) and Kder (2015).
Eggert et al. (2012) performed fatigue tests and their analysis on deck panels with thickness of 4mm and
concluded that in panels with these thicknesses the photogrammetric technique to obtain the distortion
shapes is important when the response is to be analysed, for example with non-linear FE-analyses. Panel
tests are quite expensive, especially if the development and optimization of the welding techniques are to
be carried out. Therefore, Fricke and Feltz (2013) and (Lilleme et al., 2012; Lilleme et al., 2013) have
been discussing the simplification of the testing to simple tensile specimens. It is clear that the highly
deformed panels are affected by the redistribution of stresses within the panels, as large deflection plate
theory would predict. This means that the test results on small-scale specimens e.g. see the excellent
review by Fricke et al. (2013a), are not directly transferrable to the design, but this redistribution of
stresses needs to be accounted for. (Lilleme et al., 2014a, b) showed how this redistribution happens
from dog-bone specimens to panels and further to passenger ship hull girders. Based on this research it
seems that on the large ships where the thin decks are typically located on moderately loaded decks,
linear hull-girder analysis with undeformed decks is sufficient. However, as we move towards panel
analysis the influence of initial deformations become more important and these should be modelled.
Geometrical non-linearity must also be considered as we analyse the small-scale specimens. When this is
done, it has been shown by Fricke et al. (2013a) that the nominal and structural stress approaches can be
used to predict the fatigue strength of the butt and cruciform joints, even with very low plate thickness. It
has been noted that the scatter, especially in the nominal stress approach, is larger than in thicker plates;
utilization of the structural stress approach improves the situation. Kder (2015) conducted fatigue tests
of 5 mm thickness butt-welded specimens as well as fillet-welded floating frame joints with 3.5mm plate
thickness. A laser scanner was used to measure the actual shape of the weld seams. It was found that the
fatigue strength of the fillet-welded floating frame joints is mainly influenced by the toe radius, undercut
359
depth and throat angle. Further research is needed on large-scale structures where the welding process of
secondary (web-frames, girders) and even primary structures (bulkheads) induces initial deformations to
the panels. In addition, the permitted deformation limits within standards need further investigation.
The thickness effect has been investigated already, decades ago, and is still considered in design
standards with the simplified formula f(t) = (tref /t)n. Where t is the relevant plate thickness, tref is a reference
thickness, usually 22 or 25 mm, and n is an empirical exponent. Different exponents should be considered
for different details/cases according to the current recommendations, such as those from the International
Institute of Welding, Hobbacher (2009). This factor reduced the used FAT-class for larger thicknesses, e.g.
down to 70% for a cruciform joint with 80 mm thickness. However, research demand still exists in respect
of the mechanism on the thickness effect and the influence for different structural details.
Doerk et al. (2012) compared fatigue tests for butt weld and longitudinal stiffener specimens with
thicknesses from 25 to 75 mm under axial load. For the butt welds a clear thickness influence could be
observed. The resulting thickness exponent of n = 0.21 complies well with the value given by Hobbacher
(2009). For the longitudinal stiffener specimens, where regardless of the main plate thickness a small
stiffener with t = 15 mm and 200 mm length was always attached, no distinct thickness influence could be
detected.
Yamamoto et al. (2012) investigated, using experiments, cruciform joints for various thicknesses from 12
to 80 mm under axial load. As a result, it was found that no thickness effect was observed in the case that
the thickness of the attached plate is constant irrespective of the thickness of the main plate. On the other
hand, in the case that the thickness of the attached plate increases in proportion to the thickness of the main
plate, a thickness effect of the exponent of 0.27 was estimated. Furthermore it was confirmed that the
thickness effect on fatigue strength is dominated by the stress concentration and stress gradient at the weld
toe which would change according to the form of the joint.
3.1.2
Environment (corrosion)
As recently assessed by Lieser and Xu (2010), the annual worldwide cost of maintenance due to
corrosion, repair and replacement is around 3%4% of GDP of an industrialized country. This fact
justifies the continued effort undertaken by the scientific community to understand this damaging
phenomenon which, if not detected in time, could lead to a catastrophic failure of a component or
structure. Corrosion is well known to reduce the fatigue life of structures. In recent years there has been
an increasing interest in using high-resolution microscopic techniques for observing the behaviour of the
material during fatigue tests. Li et al. (2012b) used a scanning electron microscope (SEM) to capture the
formation of a fatigue crack as a result of the coalescence of individual discontinuities (Figure 1) and the
surface exfoliation of a sample affected by pitting corrosion (Figure 2) in Aluminium. They concluded
that the corrosions pit is not only responsible for promoting the nucleation of fatigue cracks, but also for
influencing the crack propagation path, together with pre-existing flaws, grain size and pits distribution.
Figure 1. Fatigue crack path resulting from the coalescence of three individual discontinuities: a notch (A), a
corrosion pit (B) and a grain boundary (C).
As it is possible to see in Figure 2, the images obtained from SEM allow appreciating the growth in
size of the corrosion pits and their coalescence, resulting in the formation of a much larger void. Li et al.
(2012b) were also able to derive an empirical formulation for 6151-T6 aluminium alloy, which relates the
average pit depth ( D ) with the equivalent corrosion time ( Teq ):
D = 0.95 + 15.08 Teq
(1)
360
Figure 2. Time evolution of pitting corrosion: (a) 7 equivalent corrosion years, (b) 15 equivalent corrosion
years, and (c) 19 equivalent corrosion years.
As shown in the study of Lee et al. (2013), high-resolution digital images of specimens during fatigue
tests can be transformed into binary images, using an image segmentation method called Binarization.
Apart from the advantage of having images of a smaller size, the most important advantage is the possibility
of evaluating quite accurately the morphology of the corrosive phenomenon. In addition the authors showed
that the data could be used to create a 3D model of the specimen, which was modelled; von Mises stresses
were successfully extracted and predicted the onset of the crack observed in the specimen.
Problems become even more complex when considering structures such as oil & gas pipelines. It is
known that the corrosion rate of welded joints (WJ) is higher than for the base metal (BM). Moreover,
according to Slobodyan et al. (2009), electrochemical differences are expected between the heat-affected
zone (HAZ) and BM and lower mechanical properties are expected in the fusion zone (FZ) with respect to
the rest of the joint. These non-homogeneities are responsible for different corrosion rates. The current
approach to designing these structures is based on laboratory testing, engineering evaluations and safety
factors. To further complicate the matter, oil & gas pipelines are exposed to the fluids they transport.
According to Andreikiv and Hembara (2013), the hydrogen absorbed by the metal can double the corrosion
rate and they proposed a method to quantitatively evaluate the influence of hydrogenation on soil corrosion
rates of oil & gas pipelines.
Sahu et al. (2013) studied the effect of type I hot corrosion on the low-cycle fatigue life of a nickel base
super-alloy used in marine and offshore gas turbine engines. The phenomenon is caused by the high
operating temperatures and the exposure of the material to saline atmosphere, leading to the deposition of a
mix of fused salts. They used SEM to show the typical morphology of the phenomenon (Figure 3):
Figure 3. Typical morphology of type I hot corrosion: intact chromium oxide layer, exposed substrate and
corrosion product, (Sahu et al., 2013).
361
They concluded that the reduced low-cycle fatigue life induced by type I hot corrosion is due to the
oxidation of grain boundaries (Figure 4):
Figure 4. Grain boundaries oxidation due to type I hot corrosion, (Sahu et al., 2013).
According to Magdziarz et al. (2013), type I hot corrosion consists of two stages: (1) incubation period
with low corrosion rate and formation of the protective oxide layer and (2) accelerated corrosion attack
leading to the rupture of the protective oxide layer, consequent exposure of the substrate and formation of
corrosion products (Figure 4).
Corrosion has a critical influence on the ship structure because ballast tanks are fitted along the mid-ship.
Once the mid-structure is significantly corroded, it can cause a catastrophic failure as a result of the large
bending moment applied by dynamic sea loads. De Baere et al. (2013) indicated that the ships ballast tanks
are usually coated with epoxy and backed up by sacrificial zinc anodes in order to decrease the effect of
corrosion; however, this construction method has not been altered for many years. They proposed that
durable coating, combined with lifetime lasting aluminium anodes, can improve the ballast tank concept. It
was concluded that corrosion resistant steel also becomes attractive when the steel price becomes
competitive.
The microbiological corrosion behaviour of ship plate was studied by Li et al. (2012b) who proposed
that LaCl3-Zn epoxy coating offers better protection, since the resistance of LaCl3-Zn coating is much
better than Zn-epoxy alone. In this study, the images obtained, by using a scanning electron microscope,
from Zn-epoxy and LaCl3-Zn-epoxy coated specimens after 30 days exposure in sulphate-reducing
bacteria solution, are shown in Figure 5 (a) and (b) respectively. From Figure 5 (b), it can be seen that
the continuity of the organic rosin of LaCl3-Zn-epoxy coating is much better than that of the Zn-epoxy
coating presented in Figure 5 (a), and hence the structure was more successfully preserved by using the
LaCl3-Zn-epoxy coating.
There are two main limitations to the empirical analyses presented so far: (1) corrosion tests take
a long time to be executed and (2) the results of the analyses can be considered relative only to the
particular tested material and conditions. The first limitation is overcome by using accelerated
corrosion tests, whilst the second limitation, as well as the first, may be overcome thanks to
numerical models able to predict the materials degradation caused by the particular corrosive
phenomenon.
(a) Zn-epoxy coating
(b) LaCl3-Zn-epoxy coating
Figure 5. SEM of specimen after 30 days exposure in SRB solution (Li et al., 2012b).
362
3.1.3
Temperature
It is well known that temperature is one of the most influential environmental parameters that affect the
behaviour of materials. Certain damaging phenomena occur or become significant only when the
temperature is higher than a certain critical value (e.g. creep) or when the temperature is low enough (e.g.
embrittlement of the material).
According to the literature, generally the low-cycle fatigue resistance and even more generally, the
mechanical properties of metals, decrease at high temperatures due to phenomena such as creep, thermal
fatigue and oxidation. However, this is not always the case: the study conducted by Zhang et al. (2012b)
reported a material with higher low-cycle fatigue resistance at high temperatures. Their study was about the
assessment of the effect of temperature on the low-cycle fatigue resistance of a particular type of aluminium
alloy used for components in internal combustion engines. As shown in Figure 6, the material presents
higher low-cycle fatigue resistance at high temperatures for large total strain amplitudes:
Figure 6. Total strain fatigue life at 200C, 350C and 400C (Zhang et al., 2012b).
The SEM analysis of the fracture surface of samples tested at high temperature and large strain
amplitudes showed the presence of tear ridges, which are considered to impact positively on the lowcycle fatigue resistance of the material by lowering the local stresses and slowing down the crack
propagation. Yu et al. (2010) studied the high temperature creep and low-cycle fatigue resistance of a
nickel super-alloy and reported a similar behaviour.
Steam generators are another application where the material is required to operate at high temperatures.
For this reason, Choudhary et al. (2013) studied the deformation and fracture behaviour of a particular type
of ferritic steel used in steam generators under tensile loading. They reported SEM fractographs showing the
morphology of the fracture surface at different tested temperatures (Figure 7):
Figure 7. SEM fractographs: (a) 27C, (b) 300C and (c) 600C.
According to Figure 6, the necking of the fracture surface occurs only when the temperature is high
enough.
The Arctic presents a challenging environment both to humans and structures operating in the harsh
environment. Todays design codes and rules provide limited information and guidance on the
quantified requirements for steels and weldments at lower temperatures. NORSOK standards are only
applicable down to 14C and ISO 19906 Arctic Offshore Structure is not very specific when it comes
to laying down materials guidelines. The standard only states that material shall have adequate
363
toughness in order to be ductile at low temperatures, fatigue shall be given particular attention at cold
temperatures and in particular the validity of S-N curves with respect to low temperature applications
need attention. Hence for lower temperatures, it is up to the designer to show fit for purpose of the
selected material and some guidance can be found in (Hauge and stby, 2013; Horn et al., 2012b;
stby et al., 2013b).
Literature discussion on fatigue testing at low temperature is scarce; especially limited data are reported
on the fatigue behaviour of weldments. Low temperatures can have both a positive and negative effect on
fatigue resistance, depending on the specific load conditions. Evaluations of fatigue take into account both
the toughness and strength of the material. Alvaro et al. (2014) carried out a literature study and
recommended that S-N curves and fatigue threshold variation with temperature and weldments should be
studied further.
Crack growth testing was performed by Walters (2014) on base material grade S980 and S440 at room
temperature and 70C; he concluded that fatigue properties decrease with lower temperature until the
Fatigue Ductile-Brittle Transition, while below this level the fatigue increases again due to brittle material.
A recommended minimum toughness limit was set at T27J where a fatigue ductile to brittle transition occurs
below this value. The same recommendation is being discussed for being implemented by the new ISO
TC67 SC8 Arctic Operations Working Group on Arctic Materials.
The steel industry is currently improving their steel material in order to meet cold climate
requirements. Ichimiya et al. (2014) tested offshore grade steel YP420 achieving a CTOD>0.3 at 40C.
Akselsen and stby (2014) investigated different welding wires for structural steels under arctic conditions
and found that toughness tests showed a wide scatter, and coarse grained (CGHAZ) and the intercritically
coarse grained HAZ (ICCGHAZ) appeared to have low fracture toughness in the range of CTOD values of
0.05 to 0.15.
Also, the ship industry needs steel grades with low temperature toughness; Nakashima et al. (2014)s
work on YP390MPa steel shows J far above the requirement of 39 J toughness at 40C for large heat input
welding. Choung et al. (2014) investigated crashworthiness of a high strength steel grade FH32 for arctic
marine structures against ice impact. From their work consisting of both analytical and material testing,
they concluded that since the material strength properties increase (both the yield and tensile strength) at
lower temperatures, the effect of lower temperatures was beneficial and the impact provided less
deformation compared to a possible impact at room temperature.
For transportation and storage of LNG, such as the cargo tanks of LNG carriers, temperatures as low as 196C are common and in order to ensure the material is tough, nickel is used; the most common grade is
9% nickel. Kim et al. (2014) investigated the use of 7% nickel TMCP steel and found that the fatigue limits
for base metal, butt weld, and fillet weld increase by more than 30%, 33% and 15%, respectively, as
temperature decreases from room to cryogenic temperatures (20C to 163C). The crack growth results
were also compared with 9% nickel material and found to behave similarly.
3.1.4
Residual Stress & Constraint, mean stress
Due to high residual tensile stresses, the fatigue strength of WJs is often considered as independent of the
mean stress, Hobbacher (2009). On the other hand, theoretical considerations concerning stable crack
growth indicate a positive effect of mean compressive stresses on the fatigue life. These positive
expectations of the influence of compressive mean stresses are supported by operational experiences with
ships, e.g. full container vessels and open multipurpose vessels, exhibiting in still waters a pronounced
hogging bending moment. On such so-called hogging ships, fatigue cracks were found within the deck
area, which is loaded in the still water by tensile stress, but hardly at all in the compressively loaded
bottom area. Thus an increased effective material utilization exists and a more economical design for
structures which are continuously loaded by compressive mean stresses. The effect of the mean stress on
fatigue life can be interpreted for cracks which propagate under axial load (Mode I) as follows: The crack
undergoes partial closure during the loading cycle and the stress intensity at the crack tip is reduced due
to load carrying via contacting crack surfaces. Thus the effective stress intensity factor is regarded as
being proportional to the crack opening part during a load cycle. The effective portion of the stress
intensity factor, designated as effectiveness factor U, is often defined only as a function of the stress ratio
R. However, it is actually affected by further parameters such as residual stresses, external load, structural
configuration and crack geometry. Therefore, a general quantification of the mean stress effect is much
more difficult.
Yuen et al. (2013) reported a lack of commonality between different approaches to fatigue analysis and
design codes and standards when dealing with mean stress. Therefore, the objectives of their work were to
validate the models adopted in the various approaches, to seek to harmonize these approaches across the
364
codes and to develop an appropriate methodology for assessing the effects of mean stress. These objectives
were accomplished through a review of the available data on the effects of mean stress on fatigue strength
and they analysed the available fatigue data in order to develop an appropriate methodology for the
assessment of the mean stress effects in fatigue analysis for marine applications. Together with Polezhayeva
et al. (2013), they recommended using improvement factors on a fatigue strength of 1.2 for steel if the
applied stress is fully compressive and linear interpolation between 1 and 1.2 if the applied stress range is
between alternating or fully compressive. This applies to the loading patterns where the mean stress level is
not changing. Furthermore, Polezhayeva reported that residual stress relaxation due to the applied tensile
fatigue or static loading is beneficial if the subsequent fatigue loading produces stress ranges that are partly
or fully compressive.
3.2
3.2.1
Materials
Metallic Alloys
Although propeller design is mainly based on assumed quasi-static hydrodynamic loads (thrust, torque
and centrifugal load), propellers are subjected instead to cyclical stresses because of the non-uniform flow
behind the ships hull and because of the inclined flow due to shaft axis inclination. Indeed, fatigue failure
becomes a concern for high performance propellers, as propeller blades for common merchant ships are
relatively thick.
In fact, acting stresses are generally below the fatigue limit of the material. However, in the case of fast
ships and/or specific performance requirements (e.g. low radiated pressure and noise), blades become rather
thin and show high values of skew and rake. Moreover, detailed fatigue strength assessment procedures
would be beneficial to reduce the relatively high safety coefficients currently considered in rules and
guidelines, overcoming rather old-fashioned approaches.
Bertoglio et al., (2014); Bertoglio et al., (2015) analysed a high performance propeller blade and
proposed a procedure for the evaluation in time of the stress field acting on a blade during one revolution,
which is then considered for fatigue assessment. Comparing different computation approaches, assumption
and approximations were highlighted both as far as the loading conditions and the fatigue strength is
concerned.
3.2.2
Fatigue & fracture improvements through material changes, surface treatment
As discussed under section 3.1.3, low temperature affects the material properties and research is being
carried out in order to build more robust materials to prevent brittle fracture. One challenging issue for the
pipeline industry is the possibility of local brittle zones in longitudinal and girth welds, especially for low
temperature pipelines or pipelines subjected to high strain from reeling or environmental loading; these
local brittle areas are of particular concern. Moore and Nicholas (2013) investigated this effect on X70
grade pipe with a service temperature of 40C. HAZ toughness properties were measured and they
recommended that modifications were made to the steel making process to reduce the number of large
and elongated inclusions in the steel, and to ensure they were well distributed throughout the steel in
order to obtain adequate toughness in the HAZ. Tronskar and Andresen (2013) investigated the effect of
reeling and the study recommended that welding procedure qualification should involve CTOD in order
to demonstrate adequate toughness in the weld zone. They also found that it was beneficial to have steel
with a relatively low silicon content (< 0.2%wt) and vanadium as the main micro-alloying element (V
0.06%wt), with high Si (> 0.29%wt) and niobium content at 0.029%wt or greater, in order for the
material to be susceptible to formation of LBZs. To prevent/limit the possibility of a brittle crack to
initiate, or to prevent crack propagation in large container ships, Murakami et al. (2013) developed
YP460 Class steel plate (EH47 grade) with excellent brittle crack arrestability.
In order to investigate microscopic fractures in 420 MPa steel, acoustic emission (AE) was performed by
stby et al. (2013a). Based on this testing technique they could document the local arrest of propagating
microcracks in the HAZ, microcrack nucleation rate and the cumulative distribution of arrested microcrack
sizes, which are data of interest, in order to gain an understanding of micromechanical brittle fracture
models. A numerical model for dynamic crack propagation and arrest in steel plates based on the local
fracture stress criterion was developed by Aihara et al. (2014). Their model was compared with test data and
showed good alignment with the physical tests.
The makers of steel and weld consumables are developing steels and weld metal in order to meet the
demands for pipelines operating in more challenging environments such as e.g. ultra deep water, sour
service, low temperature or high strains from e.g. reeling, earthquakes, soil movements etc.
365
Harksen et al. (2013) investigated X-65 line pipe steel in the as-rolled condition (AR), quenched
condition (Q), and quenched & tempered condition (QT), in order to gain an understanding of the
mechanical material characteristics during the different processes. Nash (2013) presented the qualifications
of Line-Pipe for the Middle East to India Deepwater Pipeline project, with a water depth of 3500 m. The
effect of heat treatment on the hydrogen diffusion of process pipe steel was investigated by Han et al. (2013)
who found that post weld heat treatment degraded the pipe steel ability to resist hydrogen diffusion when
the micro structure had acicular ferrite compared to non-heat treated steel, this finding is opposite to other
microstructures such as ferrite/pearlite or bainite structures.
3.3
3.3.1
Loading
Stochastic Loading (load interaction effects (sequence))
The traditional approach to fatigue damage under varying loading is still widely used, based on the S-N
curve and the assumption of linear cumulative damage (Palmgren-Miners rule). Accordingly, the total
damage is expressed as the accumulated damage from each load cycle at different stress levels,
independent of the sequence.
Renaud et al. (2013) shows that the inclusion of flexible modes in the calculation of fatigue damage
accounts for a large share of the total fatigue damage by analysing measurements recorded onboard a 9,400
TEU class container vessel. Schaumann et al. (2012) investigated the effect of load sequence on the fatigue
performance of welded spatial tubular joints based on uniform material law and concluded it is important to
describe the load chronology and the maximum peaks precisely in order to take into account load sequence
effect in a practical way. Concerning the complex and randomness of fatigue loads acting on ship side-shell
structures, Li et al. (2013b) proposed a direct calculation procedure to compute the local stress of such
details using a so-called local stress factor instead of the conventional stress concentration factor. The
method was compared with full-scale measurements of a 4400TEU containership for the fatigue assessment
of ship side-shell structures with good agreement. Mao et al. (2015) proposed a modified beam theory using
coefficients regressed using bending moments computed by a 3D-panel hydrodynamic analysis and nominal
stressed calculated by the global FEM analysis. It is able to take into account fatigue damage of
containerships caused by horizontal bending and warping in a reliable and fast way with similar accuracy as
the direct FE analysis.
Fatigue crack propagation behaviour under the superimposed loading history containing whipping
response was studied by Matsuda and Gotoh (2013). Consideration of the high frequency component of the
loading history leads to overestimating the fatigue damage factor and shows that the traditional fatigue life
calculation method based on the S-N curve approach with the rainflow counting is inappropriate under
superimposed loading conditions.
Fricke and Paetzold (2012) carried out fatigue tests using simplified stress histories with superimposed
stress cycles of different frequencies and showed that most of the fatigue damage is caused by the waveinduced stress cycles enlarged by whipping stresses. They also proposed the simplified fatigue damage
assessment of wave induced stress cycles including whipping effects and this method has been adopted by
Germanischer Lloyd.
Gotoh et al. (2012) also performed fatigue tests with stress histories containing superimposed stresses and
proposed a procedure for extracting an effective loading sequence from random loading. They compared the
traditional S-N approach with RPG (re-tensile plastic zone generating) load criteria which provided better
results when considering the load sequence effects on fatigue life. Osawa et al. (2013) proposed carrying out
the fatigue tests of superimposed stress signals with a new testing machine powered by two exciting motors
and showed that a simplified counting based on the enlarged low frequency cycles yielded a good prediction
of the fatigue life.
3.3.2
Cycle countingspectral, time-domain, stress ranges, means stress effect
In order to apply Miners rule in the assessment of fatigue life, the complex stress time history is divided
into blocks of constant stress amplitude (or range). The rainflow counting method is still the most popular
technique in the analysis of fatigue data.
Several different assessment methods were assessed by Nielsen et al. (2011) and a comparison made of
the measured strains and results from the frequency domain fatigue analysis with those from a time domain
analysis based on the rainflow counting method using the Palmgren-Miner rule.
Fukasawa and Mukai (2014) investigated the effects of the hull-girder vibration on the fatigue strength of
a Post-Panamax container ship for short-term sea state. They showed that the fatigue damage factor was
reasonably evaluated by the rainflow counting method, even in the case where the smaller amplitude
366
component of the higher frequency vibration (flexible response) superimposed on the larger amplitude
component of the low frequency vibration.
Andersen and Jensen (2013) investigated four spectral methods in order to assess whether spectral
methods for fatigue damage estimation are applicable in onboard systems for monitoring the fatigue damage
of a 9,400 TEU container ship and compared them with the result of rainflow counting. They showed that
the agreement between the spectral method and rainflow counting is generally good and the narrow-band
approximation seems to yield a fast and fair estimation of the fatigue damage.
Huang (2013) proposed a new method, which is equivalent to Lows method to estimate the fatigue
damage of combined high and low frequency load effects and this method also identified the reduced effects
of the rainflow counting.
3.3.3
Complex stresses
Ship and offshore structures are often subjected to complex stress fields due to their unique structural
configuration and also due to various wave encounters from the sea. However, the fatigue tests conducted
in the laboratory have a simple uniaxial stress field. Conversion of the complex stress field into an
equivalent uniaxial state is still a challenging task for the researchers working in the area of fatigue. The
following literature highlights the complex stress field as an important factor governing the fatigue life.
Yu et al. (2011) carried out strain controlled multiaxial fatigue experiments on extruded magnesium alloy
cylindrical specimens. The experiments included fully reversed tension-compression (path a), cyclic torsion
(path b), proportional axial torsion (path c), and 90o out-of-phase axial-torsion (path d), as shown in Figure 8.
Figure 8. Load paths used in the experiment.

For the same equivalent strain amplitude, fatigue life under proportional loading was higher than in the case
of non-proportional loading (Figure 9).
Figure 9. Fatigue lives obtained for various load paths.
367
It is observed that the FatemiSocie criterion predicts the fatigue life well in the low cyclic fatigue regime
for all the loading paths investigated, but does not do so for the high cycle fatigue regime. The modified
Smith Watson and Topper (SWT) criterion combining the normal and shear components of the stresses and
strains on material planes can predict the fatigue lives of the material satisfactorily. Both models (FatemiSocie and SWT) fail to predict fatigue life and early cracking behaviour for the loading conditions under
cyclic torsion with a high compressive axial stress.
Rozumek and Marciniak (2012) investigated, by the use of experiments, the crack growth in
rectangular cross sections subjected to out-of-phase bending and torsion loadings. The tests were
conducted in a high cycle fatigue regime for in phase and 90 out-of-phase conditions and
it was observed that the fatigue life of the specimen is less when the loadings are applied out-ofphase.
Wu et al. (2012) conducted strain controlled multiaxial low-cycle fatigue experiments on tubular
specimens made up of Ti-6AI-4V for both proportional and non-proportional load cases. The strain paths
used were push pull (PP), reversed torsion (TP), 90o out-of-phase circular strain path (CP) and 90o out-ofphase elliptical strain path (EP). The number of cycles to failure obtained from the experiments are
correlated with the von Mises equivalent strain range and principal strain range (Figure 10). Use of both of
these classical methods resulted in scatter of fatigue data which indicates the inaccuracy of the classical
methods for predicting fatigue life in non-proportional load cases.
Figure 10. Correlation of number of cycles to failure with the von Mises equivalent strain.
Shamsaei et al. (2010b) investigated the fatigue behaviour of pure titanium and titanium alloy cylindrical
specimens under proportional and non-proportional loading conditions. It has been observed that cyclic
stress strain data were correlated by von Mises equivalent criterion. This indicates that there is no
additional hardening effect observed for this material under non-proportional cyclic loading. However,
non-proportional loading resulted in significantly shorter lives and fatigue data could not be correlated
by the von Mises criterion which indicates that the non-proportional loading shortens the fatigue life
though the non-proportional hardening is not observed. The authors explained that the non-proportional
additional hardening is not the only parameter which provides differences in fatigue life in nonproportional cases.
Zhang et al. (2012b) carried out experiments on solid cylindrical aluminium alloy bar specimens to
investigate the effect of the stress amplitude ratio (ratio of amplitude of shear stress and normal stress),
mean shear stress and phase angle on axial-torsion high cycle fatigue failure of an aluminium alloy under a
certain equivalent stress. Macro- and micro-analysis of the specimen fracture appearance were conducted to
obtain the fracture characteristics. It was concluded that with increasing stress amplitude ratio, fatigue life
reduced gradually under proportional loading and the reduction is more significant under non-proportional
loading. This indicated that in tension-torsion fatigue, the effect of the stress amplitude ratio is more severe
than for uniaxial fatigue.
Zerres and Vormwald (2013) conducted experiments on thin-walled tubes made of fine grained steel
S460N and aluminium alloy AlMg4.5Mn under proportional and non-proportional tension and torsion
loading under stress controlled conditions. For both cases, compared to proportional loading conditions the
368
non-proportional load case provided longer crack growth life compared to the proportional load case
(Figure 11).
Figure 11. Crack growth lives of steel specimens.

Thvenet et al. (2013) conducted experiments (deformation controlled) on tubular Welded Joints
subjected to multiaxial fatigue stresses. The fatigue assessment approaches according to the International
Institute of Welding (IIW), Det Norske Veritas (DNV) and American Bureau of Shipping (ABS) were
applied for this particular case of tubular joint subjected to multiaxial fluctuating stress. It was observed
by the authors that under the loading, the application of these kinds of recommendation does not lead
systematically to conservative predictions (Figure 12). Dang vans multiaxial fatigue criterion has been
recommended by the authors as a suitable alternative. The authors also highlighted that the approaches
mentioned above do not mention the critical point to be assessed when a combination of different load
components occurs.
Figure 12. Numerical and Experimental Fatigue Lives.

Shamsaei and Fatemi (2014) investigated small crack growth behaviour of 1045 and 1050 medium carbon
steels, 304L stainless steel, and Inconel 718 under multiaxial states of stress. Experimental observations
from solid and thin-walled tubular round specimens under various combined stresses including in-phase
(IP) and out-of-phase (OP), tension-torsion and tension-tension, with or without mean stresses, were used
369
to characterise small crack growth behaviour. The effects of friction induced closure, material ductility,
stress gradient, strain amplitude, non-proportionality of loading, and mean stress on small fatigue crack
growth behaviour were also discussed. The ReddyFatemi (RF) effective strain-based intensity factor was
used to correlate crack growth rate data for various materials and loading conditions presented. The
difference in fatigue crack growth behaviour in proportional and non-proportional loading for 1050 N
steel is shown in Figure 13.
Figure 13. Effect of non-proportionality in crack growth behavior (1050 N Steel).
3.3.4
Recent developments in multiaxial fatigue criteria
In the ISSC 2012 report, multiaxial fatigue was discussed and also included a comprehensive bench mark
study of a bulkhead, where the weld seam was subjected to parallel, normal and shear stresses. Different
codes were assessed and the results showed a great divergence between the results, which gives clear
evidence of the fact that fatigue problems in the case of multiaxial stresses need further research
supported by experimental work, which was also pointed out by the official discusser of the report.
Arajo et al. (2011) proposed a new method of finding the shear stress amplitude on the critical plane for
both synchronous and asynchronous loadings. The authors used the Maximum Rectangular Hull (MRH)
method for finding the shear stress amplitude in the critical plane. It has been observed that the use of this
MRH method instead of Minimum Circumscribed Circle (MCC) method improves the fatigue life
prediction for the data evaluated.
Fatemi and Shamsaei (2011) presented an overview of some important issues in multiaxial fatigue and
life estimation. The authors focused on damage mechanisms and damage quantification parameters, material
constitutive response and non-proportional hardening, cycle counting and damage accumulation in variable
amplitude loading, and mixed-mode crack growth. It is highlighted that capturing the correct damage
mechanism is essential to develop a proper damage quantification parameter. Also the authors emphasized
that critical plane damage models with both stress and strain terms are more appropriate since they can
reflect the materials constitutive response under non-proportional loading.
It has been emphasized that classical yield criteria, such as the von Mises distortion energy criterion, are
commonly extended to multiaxial fatigue life estimations, although these criteria may work for in-phase or
proportional loading, they often under-estimate the typically observed shorter lives for out-of-phase or nonproportional loading. The shorter life under non-proportional loading is often attributed to non-proportional
hardening. However, materials without this hardening also typically exhibit shorter life under such a
loading. Classical criteria, such as von Mises equivalent strain, cannot represent this behaviour and should
not be used for non-proportional loading.
Overall, it is concluded by the authors that due to the challenging nature of the multiaxial fatigue
problems, much additional research and development work is still needed for accurate and reliable
multiaxial fatigue life estimation.
Li et al. (2011) proposed a new multiaxial fatigue criteria based on the critical plane concept. The nonproportional factor proposed by Chen et al. (1996 ) has been used in this method to evaluate and account for
the degree of multiaxiality. It was observed that the fatigue life was reduced when the additional nonproportional cyclic hardening increased.
370
Matsubara and Nishio (2013) proposed a new criterion to estimate fatigue limit under multiaxial high
cycle loading. The proposed criterion consists of two parameters, one for crack initiation and the other for
initial crack growth. When compared with the available experimental results, the results are within 10% of
error, which is better than other criteria used for comparison.
Marciniak et al. (2013) compared the fracture plane position gained from experimental tests of specimens
under multiaxial loading and theoretical ones from calculations according to variance and damage
accumulation methods. The authors concluded that the damage accumulation method predicts the fatigue
life better than the variance method. However, for the pseudo-random loading, the variance method is better
than the damage accumulation method.
Ince and Glinka (2014) proposed two different forms of an original multiaxial fatigue damage
parameter related to the maximum fatigue damage plane for performing fatigue life prediction under
various loading conditions loadings. Both the damage parameters are correlated to sets of experimental
data published in the literature to verify the prediction accuracy of the damage parameters. The damage
parameter in the form of the GSA, when applied to the uniaxial loading, provides very good correlations
with four sets of experimental mean stress fatigue data for Incoloy 901 super alloy, ASTM A723 steel,
7075-T561 aluminium alloy and 1045 HRC 55 steel. In the case of multiaxial loadings, both the GSE and
GSA parameters are found to correlate well with fatigue data of 1045 steel and Inconel 718 tubular
specimens under proportional and non-proportional loadings. In addition, the damage parameters show
reasonably acceptable correlations with experimental fatigue data of SAE 1045 steel notched shafts
subjected to proportional and non-proportional loadings.
Factors affecting multiaxial fatigue performanceMaterial and Additional hardening due to nonproportional load path
The difference in fatigue life is attributed to the type of material subjected to cyclic loading. This can be
explained from a metallurgical point of view. In materials having low stacking fault energy, only planar
slip systems evolve under a proportional load. However, during a non-proportional load the principal
stress plane rotates and causes plastic strain at different slip mechanisms. This cross slip activation due to
plastic strain can result in a significant increase of hardening of the material when compared to a
proportional load cycle.
However, for materials with high stacking fault energy, slip is easier and material slip occurs both under
proportional and non-proportional load cases. In these materials the additional hardening in a
non-proportional load cycle is rarely observed. To illustrate this, aluminium material has a higher stacking
fault energy hence, this material shows virtually the same stress strain curve independent of the
proportionality of the loading path and no significant additional hardening is observed in this material.
Literature that relates the material as a parameter affecting the fatigue life in non-proportional load case
has been reported by several researchers. Shamsaei and Fatemi (2009) who studied the effect of material
hardness on multiaxial fatigue behaviour based on experiments conducted on tubular specimens made from a
medium-carbon steel with three hardness levels obtained from normalizing (low hardness), quenching and
tempering (medium hardness) and induction (high hardness). The authors concluded from the experiments
they conducted that the induction hardened tubular specimen lives were 8 and 25 times longer than the
normalized tubular specimen in the high cycle fatigue regime, whereas, in the low-cycle fatigue regime, the
induction hardened tubular specimen fatigue lives are approximately three times lower than the normalized
tubular specimen. Based on these results the authors concluded that the ductile material sustains more in the
low-cycle fatigue regime, whereas, in the high cycle fatigue regime, the higher hardness improves the fatigue
life. Based on these experiments, it is understood that the fatigue life of the material in the multiaxial stress
state depends on the hardness of the material as well as the type of loading (proportional or non-proportional).
Itoh and Yang (2011) studied the material dependence of multiaxial low-cycle fatigue to develop a
suitable strain parameter for life estimation under non-proportional loading. Strain controlled multiaxial
low-cycle fatigue tests under proportional and non-proportional strain paths were carried out using hollow
cylindrical specimens of several materials. Test materials employed were 12 metallic materials of which
crystal structures (CS) are of a face-centred cubic (FCC) structure and body-centred cubic (BCC) structure.
It is concluded that reduction in fatigue life has a close relationship with additional hardening under nonproportional loading, and depends on the crystalline structure of the tested materials. The authors proposed a
material dependent equation to calculate the equivalent strain range which takes into account the additional
hardening effect. It has been observed that by using this equivalent strain range parameter proposed by the
authors, predicted and experimental fatigue lives were in good agreement (Figure 14).
371
Figure 14. Predicted and experimental fatigue lives of various materials of BCC and FCC category using
proposed equivalent strain parameter.
Shamsaei and Fatemi (2010) investigated the effect of micro structure and hardness on non-proportional
cyclic hardening of metallic materials. Tubular specimens made from 1050 steel in normalized, quenched
and tempered, and induction hardened conditions, as well as 304L stainless steel were used to study the
effect of microstructure on multiaxial cyclic deformation. Multiaxial data generated in this study as well
as multiaxial deformation data of several materials from literature suggest non-proportional cyclic
hardening can be related to uniaxial cyclic hardening. Non-proportional hardening coefficients predicted
from a proposed equation based on this observation were found to be in good agreement with the
experimental values in this study and from the literature. It was concluded by the authors that a nonproportional hardening coefficient cannot be predicted if only based on monotonic properties, as nonproportional cyclic hardening also depends on the material slip system induced under cyclic loading. The
non-proportional cyclic hardening coefficient was found to strongly correlate with the uniaxial cyclic
hardening coefficient. Similar trends were observed for both coefficients with a change in total plastic
strain. With the proposed non-proportional hardening coefficients, the authors have shown that the
predicted and experimental equivalent stress values are in good agreement (Figure 15).
Figure 15. Predicted and experimental equivalent stress values for various materials.
Additional hardening due to non-proportional load path
In general, additional hardening is observed in the low-cycle fatigue behaviour where the stress exceeds
the yield stress of the material. When the stress exceeds the yield stress in case of monotonic loading
there is strain hardening due to locking of the grain after the initial plastic flow. In the case of the nonproportional cyclic loading it is observed that the hardening is either more or less than the case of
proportional cyclic load case/ monotonic load case. Several slip systems are activated in non-proportional
loading and block each other, which finally results in additional hardening. This additional hardening is
generally noted in the literature as additional hardening due to non-proportional load case.
Literature that relates the additional hardening as a parameter controlling the fatigue life in multiaxial
fatigue has been reported by several researchers.
Doring et al. (2006) carried out experiments and numerical simulations using a new plasticity model to
compute the additional hardening in the non-proportional load case. The experiments were conducted by
altering 100 proportional load cycles and 100 non-proportional load cycles with 90o out-of-phase in the
372
non-proportional case. It is observed that during the non-proportional loading the stress value rose
significantly, which is known as the additional hardening due to non-proportional load cycle (Figure 16).
Figure 16. Alternating cycles from proportional and non-proportional loading to observe additional
hardening behaviour in non-proportional load case.
Shamsaei et al. (2010a) observed from experiments that the fatigue life of 1050 steel with no nonproportional cyclic hardening was found to be more sensitive to non-proportional loading than for 304L
stainless steel with significant non-proportional cyclic hardening. Therefore, additional cyclic hardening
due to the non-proportionality of loading is not the only factor causing shorter fatigue lives under nonproportional loading.
Noban et al. (2012) studied the load path sensitivity in the fatigue life estimation of a material
(30CrNiMo8HH). Three proportional and seven non-proportional load paths were studied by the authors
using strain controlled experiments. In addition, the authors used two different cyclic plasticity models
(Mroz and Chaboche) for predicting the cyclic stress strain behaviour. The authors concluded that for
multiaxial load paths the Chaboche model produces results in good agreement with the experiment, whereas
for uniaxial load paths the Mroz model has been observed to be closer to the experiment.
It is observed that the examined material did not show any significant additional hardening due to nonproportional loading. However, the material showed significant reduction in fatigue life compared to the
proportional loading case. The cause of lower lives in non-proportional loading is not necessarily the
additional hardening, rather it depends on the sensitivity of material to out-of-phase loading of which
additional hardening may be one part.
Based on the cyclic stress strain behaviour of the material for various loading paths, a material dependent
non-proportionality modification factor has been proposed to account for the severe path dependency of the
material. The fatigue life assessment with this modification factor is observed to be more accurate with the
non-proportionality factor for this material (Figure 17).
Figure 17. Predicted fatigue life vs. experiments, before (a) and after (b) applying the loading path effect
on maximum equivalent strain.
373
Anes et al. (2013) explained that in shear stress space, the use of a constant factor to convert the axial
stress is not appropriate. The authors further explained that the stress scale factor (ssf) on axial stress is
dependent on the stress amplitude level and also based on the shape of the loading path. The authors have
conducted a series of multiaxial fatigue tests on a high strength steel to determine the multiaxial fatigue
strength under proportional and non-proportional loading conditions. An ssf function was mapped based
on the experimental results using as arguments, the axial stress amplitude and the stress amplitude ratio,
which have proved to be sensitive to the loading path nature. Using the ssf surface, an equivalent shear
stress was defined which was used for the fatigue life calculation. Results indicate that the ssf is not a
constant value and it is strongly dependent on stress amplitude level and loading path shape. The authors
concluded that the equivalent stress was successfully applied to proportional and non-proportional
loading paths with satisfactory results.
Fremy et al. (2013) studied the load path effect on mixed mode conditions experimentally on stainless
steel. The authors observed that the load path has a significant effect on the rate of crack growth (Figure
18), though the crack path is not significantly different. It is observed from scanning the electron
microscope in a non-proportional mixed mode condition that a complex system of slip bands is formed
at the crack tip and is used to promote the crack growth. It is concluded that, in non-proportional mixed
mode loading conditions, criteria based solely on the maximum, minimum and mean values of the stress
intensity factors, cannot provide accurate predictions of the crack growth rate or of the crack path. The
entire load path should be considered so as to predict accurately fatigue crack growth in nonproportional mixed mode conditions.
Figure 18. Crack lengths vs. Number of cycles for various cases.
Itoh et al. (2013) proposed a method of evaluating the principal stress and strain ranges and the mean
stress and strain, and also proposed a method of calculating the non-proportional factor which expresses
the severity of non-proportional loading in 3-dimentional stress and strain space. This study also
examines the material constant used in the strain parameter proposed by the authors for life estimation
under non-proportional multiaxial LCF and presents a simple method to re-evaluate material constants in
relation to the material constants obtained in a static tensile test. For various materials with different nonproportional loading, the authors have shown that the predicted and measured fatigue lives are in good
agreement.
3.4
3.4.1
Structural Integrity/Life cycle management

Fabrication and repair
Fatigue improvement methods were discussed in the ISSC 2012 Fatigue and Fracture report and more
details regarding the different methods can be found in the ISSC 2015 V.3 report. Hence, only the most
applicable research within this topic is listed below.
High frequency mechanical impact (HFMI) technology has undergone a significant amount of research
with the aim of improving the fatigue classes of Welded Joints in high strength steels (HSSs) with yield
strength up to 950MPa (Yildirim, 2013; Marquis and Barsoum, 2013; Yildirim and Marquis, 2014). The
benefits of the technique are small tools, low induced noise levels and compressive residual stresses
374
produced at the weld. Recommended design curves are proposed for nominal, structural and notch stress
approaches when the load ratio is R=0.1; see Figure 19. These are collected and presented in Marquis et al.
(2013) and Haagensen and Maddox (2013). (Kirkhope et al., 1999a, b) and Yildirim and Marquis (2014)
state that future research should be directed towards large-scale structures, variable amplitude loading and
other load ratios. The quality control issues have been investigated by Yekta et al. (2013); the varying load
ratio by Mikkola et al. (2013) and the residual stress states by Suominen et al. (2013). The IIW Commission
XIII design proposal includes proposals for the material strength, high load ratio and variable amplitude
loading. The presented investigations show the potential of this method to improve the fatigue strength,
which would enable weight reduction or longer design lives for marine structures when high strength steels
are considered.
Figure 19. Proposed maximum possible fatigue class improvement in % for A) nominal stress method, B)
structural stress method (non-load-carrying), C) structural stress method (load-carrying) and D)
all joints using the notch stress method (Yildirim, 2013).
The long term effects of the different improvement methods are not well documented. Haagensen et al.
(2014) studied the long term effect of different extension methods applied to the floater Vestlefrik B in
the North Sea during 1999-2000. From their study they concluded that toe burr grinding and hammer
peening showed an improvement of 4 to 6 in fatigue improvement and only one crack was observed in the
treated area during inspection 10 years later, however this crack was due to a lack of penetration not
discovered during the first inspection.
3.4.2
Inspection & Monitoring of structure and coatings
Marine structures are exposed to dynamic, continuous loadings coming from the harsh marine
environment. This environment can not only cause damage to the structure due to extreme loadings in the
short term or cyclic loadings in the long term, but also engenders corrosion. In order to guarantee
reliability and integrity of marine structures, an application of structural health monitoring (SHM) is
required. The utilization of SHM in the marine industry allows increasing both human and environmental
safety in conjunction with reduction in maintenance costs. Sielski (2012) indicated that SHM should have
the following main components: (1) accurate information about a structures failure modes, fatigue
characteristics and material corrosion characteristics, (2) damage detection capability in real-time and (3)
prediction capability of the forthcoming condition of structural members. In this study it was proposed
that SHM can be improved, if the following characteristics are achieved: (1) develop more accurate
fatigue and fracture models, (2) improve the model in terms of corrosion detection and (3) install sensors
for detecting damage continuously.
375
One of the most common failure modes of ship structures occurs from fatigue crack initiation and
propagation. In the case of a ship being under dynamic sea loading, fatigue cracks are easily initiated since
the structural components are assembled by using a welding process. According to stiffened plate fatigue
experiments, the combination of lamb wave propagation and power spectral density methods is effective
enough to detect fatigue crack initiation and monitor crack propagation. In addition to the development of a
fatigue crack model, the selection of an appropriate sensor is crucial to collect strain data accurately.
Murawski et al. (2012) suggested that Fibre Bragg Grating (FBG) sensors should be used in marine
structures to increase the precision of obtained strain data during the SHM process. In this study, a damage
detection experiment was conducted on an offshore platform leg model by using FBG sensors. The location
of the FBG sensors on the offshore platform leg model can be seen in Figure 20. The mechanical strain
changes due to the dynamic loading are illustrated in Figure 21. The scheme of the SHM system for real
marine structures is also discussed in order to calculate strain-stress absolute values with main axis
directions by arranging three FBG sensors as a strain rosette.
The conventional non-destructive methods for crack detection consist of strain monitoring, ultrasonic
testing, radiographic testing and magnetic testing. The strain monitoring process can be applied to marine
structures by allocating strain-gauges on various locations of the structure. However, this process can
require a large number of strain-gauges depending on the complexity of the structure. Ultrasonic sound
waves can be applied to the structure in order to carry out the ultrasonic monitoring. These waves can
become discontinuous when they reach a crack in the structure, and hence the crack inspection can be done
by using this process. In contrast, this test system is not adequate for the in-service screening of fatigue
cracks. The radiographic testing system generally relies on X-rays and Gamma-rays which can penetrate the
structure and allow identification of the crack. Nevertheless, it is impractical for them to be actually applied
to marine structures, such as offshore platforms, due to the size of the structure. Although the magnetic
testing method is very simple to use, it requires human interpretation to notice the crack on the structure
surface. Hence, it is not beneficial to employ this method in real-time SHM applications (van der Horst
et al., 2013).
On the other hand, van der Horst et al. (2013) proposed a technique based on wireless monitoring as an
application of SHM for marine structures. They indicated that the implementation of networks of wireless
sensors can effectively enable the detection of fatigue cracks on a structure, even though it has some
disadvantages such as a lack of robust connectivity, inadequate data rates, etc. It was concluded that the
conventional methods for detecting cracks and recent techniques for monitoring crack propagation can be
combined in order to increase the sensitivity and efficiency of the SHM.
Figure 20. Laboratory experiments: a) the offshore platform leg model b) sensor with steel carrier,
c) sensors without steel carriers.
376
Figure 21. Mechanical strains before and after yielding of undamaged model.
3.4.3
Inspection and Maintenance
The structural strength of ships is a key issue that affects the safety of crew, economic costs and pollution
of the environment. The required structural management and safety of ships can be achieved by
performing appropriate inspections at the right intervals and repairing the defects that are identified. Chao
et al. (2013) indicated that a vision system based on the laser measurement system (LMS), which is a new
sensor to accurately calculate distances of the target in an arc section, can be used for bulk ship
inspection. Their study concluded that ship loaders can achieve the bulk cargo loading operation
automatically based on the inspection data calculated by using LMS scanners and the architecture of this
operation is illustrated in Figure 22.
Figure 22. The Architecture of the Automatic Ship Loader.

The importance of using autonomous underwater vehicles has recently increased in the inspection of
ships hulls and marine structures after the more challenging application of robotics has emerged. Hover
377
et al. (2012) have built and applied navigation algorithms that can control the hovering autonomous
underwater vehicle (HUAV) in order to achieve full imaging coverage of hull structures in high
resolution, and with better simultaneous localization and mapping processes.
Todays structural reliability applications have had a great impact on the determination of inspection and
maintenance planning. The assessment of structural strength by using probabilistic techniques is one of the
most popular structural reliability applications. The target of this method is to calculate the failure
probability of the system by evaluating a reliability index for the assumed scenarios. Cmara and Cyrino
(2012) developed a statistical model of hull structure containing the effects of fatigue and corrosion. Based
on the developed model, time-dependent reliability of the hull structure is calculated by using Monte Carlo
Simulation. Figure 23 illustrates the probability of failure change versus operation time for their considered
model. It was concluded that maintenance/inspection time, which should be done around 7.5 years of
operation for the considered model, can be arranged by using target reliability.
A powerful fast integration technique based on the first order reliability methods was adopted by
(Zayed et al., 2013a, b) to calculate the structural reliability of ship hulls. Guo et al. (2012) used the Latin
Hypercube Sampling (LHS) method with Monte Carlo Simulation in order to determine the failure
probability level for corroded aging tankers. The probability distribution of each variable is divided into
non-overlapping segments and a value for each variable can randomly be generated from each segment in
the LHS procedure, so that the total variety of the distribution is sampled more evenly and steadily. This
study concluded that LHS is mainly more accurate than conventional Monte Carlo direct sampling during
the simulation. Structural Health Monitoring (SHM) data obtained from sensors was used by Zhu
and Frangopol (2013) to improve the accuracy and redundancy of reliability assessments of ship crosssections.
Figure 23. Failure probability in relation to operation time.

Li et al. (2012a) introduced a hull structure information integration model that can be used in mobile
devices in order to record the data during the inspection process, and therefore transferring these data is
much easier than using conventional ways. Lee et al. (2013) suggested a radio-frequency identification
device (RFID, which can be applied to cloud technology, for an effective maintenance/inspection
operation. It was proposed that RFID can be used to identify correct data and save these data accurately
via a cloud system, thus if RFID is used during the inspection process, it will improve the efficiency of
maintenance operations.
3.5
Composites
Coronado et al. (2012) studied the influence of temperature on the static and fatigue resistance of a
carbon-fibre epoxy composite, which is a material commonly used in aerospace and naval applications
specifically because of its low weight. According to the authors, whilst fatigue in metals can be studied in
terms of crack nucleation and propagation, fatigue in composites may result in a more complex
phenomenon, where multiple damage mechanisms nucleate, grow and interact (e.g. fibre breakage, fibrematrix debonding, delamination and matrix cracking). The fracture mechanics of these materials resulting
from static loading are currently the subject of extensive research. On the otherhand, few studies have
focused on dynamic loadings and the influence of temperature. Coronado et al. (2012) used the data
reduction method called the Modified Beam Theory (MBT) method to calculate the interlaminar fracture
toughness in mode I resulting from their experiments. As seen from Figure 24, the fracture energy
necessary to initiate delamination increases with the increment of temperature and decreases with the
increment of the number of fatigue cycles.
378
Figure 24. Delamination fatigue curves at different temperatures.

Vina et al. (2011), conducted a similar study on a different composite material: glass fibre reinforced
polypropylene. According to the authors, the tensile strength of the material decreases with the increment
of temperature, as well as its fatigue life (Figure 25).
Figure 25. Results of fatigue tests (Vina et al., 2011).

An interesting finding of this study is represented by the non-linear shape of the fatigue fitting curve
corresponding to a temperature of 100C. This change in behaviour of the material was explained by the
fact that the material is preparing to change from its viscoelastic temperature zone to its viscous
temperature zone.
4.
FATIGUE ASSESSMENT METHODS
Generally speaking, ship- and offshore structures can be classified as stiffened (curved) panel-, truss-, or
frame assemblies consisting of shells, plates, profiles, tubular members and castings. Arc-welding, the
commonly applied joining technique, introduces notches at the weld toe and, depending on the weld
penetration level, notches at the weld root as well. Since notches have been identified as fatigue sensitive
locations and ship and offshore structures typically contain a O(103) [m] weld seam, particular attention is
paid to (arc-)welded joints and V-shaped notches. Different assessment methods have been developed
over time and will be discussed.
Fatigue assessment requires defining the Fatigue Loads, Strength and Damage accumulating rule.
IACS (2014) defines the calculation of fatigue loading in a direct calculation (spectral approach) and
describes the assumption of service conditions used to define the Equivalent Design Wave simplified
approach which is used in rules to check fatigue criteria by analytical formulae.
379
Derbanne et al. (2011) analysed the effect of different probability levels of rule-based loads on the fatigue
damage. The sensitivity of the Weibull shape parameter on long-term distribution has been studied. It has
been observed that while the shape parameter is unity, the 10-2 probability level of rule-based loads has less
sensitivity to the final fatigue damage. Based on this study it has been concluded that the Bureau Veritas
(BV) rule for fatigue assessment will be updated from the existing 10-5 probability level to the 10-2
probability level.
4.1
Overview
As in the report of the ISSC 2012 TC III.2 only a brief overview of newly developed fatigue
assessment methods was presented, it was decided this time to pay more attention to this subject.
Actually, several approaches and methodologies have been and continue to be proposed in the literature
dealing with a wide range of cases of ships and offshore structures and more in general with welded
structures. It seemed interesting, therefore, to propose an updated overview offering a categorization of
approaches based on the type of parameter believed to govern the fatigue phenomena, but having also
in mind their complexity and application limits that were well described by Figure 1 of the ISSC 2012
TC III.2 report.
Noticeable advancements in fatigue assessment approaches of welded joints using local approaches, have
appeared in recent years since the publication of the well-known book on the subject by Radaj et al. (2006).
These same authors recently published a review of recent developments in local concepts of fatigue
assessment of welded joints that can be seen as an appendix to the original book (Radaj et al., 2009b).
Fricke (2014) outlines the most recent developments and future challenges in fatigue strength assessment of
welded joints, pointing out the links among different approaches.
A new book by Radaj and Vormwald (2013) reviews and explains advanced methods of brittle fracture
and fatigue assessment, looking at them even more locally than before as they move from the generalized
Neuber (1968) concept of fictitious notch rounding (FNR) to explain methods based on the stress intensity
factor (SIF) and on the energy density concepts, finally touching upon the crack propagation approach based
on both linear elastic and elasto-plastic fracture mechanics principles.
Traditional approaches to fatigue assessment are based on simplified structural analyses and fatigue tests
of standard structural details. The well-known nominal stress approach is still widely applied and provides
consistent results in several practical cases, even if the trend is more and more towards local approaches.
Local stress approaches were introduced decades ago for tubular joints of offshore structures in the hot
spot stress fashion and their applications to plated structures is nowadays mainly based on finite element
(FE) models of relatively large parts of the structure, refined according to well established regulations,
limiting their application to specific cases. These methods are now referred to as structural or geometric
stress approaches.
The effective notch stress approach has recently been regulated by Fricke et al. (2008) and is nowadays
commonly applied by the marine industry, even if not actually included in classification societies rules for
ship construction. Indeed, recent rules, such as the IACS Common Structural Rules, apply the structural
stress approach in as far as the notch stress is obtained by a stress concentration factor, rather than FE
analysis simulating the weld seam effect, while the effective notch stress approach is more common in the
offshore field.
Local approaches should be able to avoid long lasting tests of specific structural details as required by
the nominal stress approach. However, results are not always conservative and sometimes are largely
over-conservative, depending on the loading and geometry of the structural detail and applied
approaches. It is noted that often fatigue assessments carried out by different local approaches, even
when based on the same principle, are not consistent. It appears that not only the method (e.g. structural
stress can be evaluated using different techniques such as surface stresses extrapolation, throughthickness stress linearization, stress at critical distances from hot spot, etc.) but also the FE analysis has
an influence on the final result.
Indeed, the trend towards local approaches to assess the fatigue strength of structural details is still in
progress and methodologies for fatigue strength assessment of welded structures are becoming more and
more local while new ones are continuing to appear in the literature. Since computational tools are
continuously being advanced, fatigue assessment approaches proposed several decades ago have now
become applicable in everyday engineering practice.
In general, test results in terms of S-N curves do not clearly distinguish between crack initiation and
complete failure; rather a parameter that is believed to govern the phenomenon is considered and
associated with the fatigue strength of the structure, i.e. to the cycles to failure, without distinguishing
380
between different propagation phases (Schijve, 2012, 2014). However, the crack propagation approach
is still felt to be not suitable or sufficiently reliable for basic design, rather it is commonly applied for
failure analysis and detailed assessments of in-service structures, especially in the oil and gas industry.
Fatigue strength assessment approaches are categorised below on the basis of the parameter believed
to govern the fatigue strength behaviour. Figure 26 proposes an improvement to the well-known
overview of fatigue assessment approaches given by Radaj et al. (2006) in order to provide a clear
summary of first principles on which the approaches are based. Actually, the parameter believed to
govern fatigue failure is shared by several different variants within the approach category; each
approach has specific application limits. The idea is, therefore, to provide a guideline for design
engineers and at the same time to present an overview of the fatigue assessment approaches recently
proposed in the literature.
Stress field
intensity
K1,W,
NSIF,SED,
PSM,local
energy,
Figure 26. Overview of fatigue assessment approaches, adapted from Radaj et al. (2006).
In summary, several local approaches have been developed in the last 20-30 years in addition to the
nominal stress approach, which is based on the stress in a component resolved using general theories and
disregarding all local stress raising effects of the welded joint. They are generally categorized as:
Structural stress approaches, based on the structural (or geometric) stress at the hot spot that includes all
stress raising effects of a structural detail, excluding all stress concentrations due to the local weld seam
profile itself;
Notch stress approaches, based on the effective notch stress, i.e. the total stress in the notch, obtained
assuming linear-elastic material behaviour, generally replacing in the calculations the actual weld seam
profile by an effective one to account for the scatter of weld shape irregularities as well as for the plastic
material behaviour of the notch root;
Notch strain approaches, based on the notch strain instead of notch stress and considering the elastoplastic behaviour of the material occurring in the notch, thus overcoming at the root the problem of the
unrealistically high stress values obtained by the notch stress approach but not removing the need for
rather complex calculations to obtain the stress and strain at the notch root;
Crack propagation approaches, evaluating the crack growth up to failure, according to the fracture
mechanics model, i.e. the fatigue strength is related to the stress intensity factor range by the wellknown Paris-Erdogan law, provided that an initial crack is assumed to exist in the welded detail and is
propagated by each applied loading cycle in a stable way.
The fatigue phenomena in welded structures are far more complex than the existing descriptions and are
controlled by what occurs in the area surrounding the weld notch, therefore involving many other
geometrical and material parameters not included in the current approaches. Indeed, every approach has its
own features and applicability limits while a unified approach seems rather far from current practice.
Among the others, there is a new category of promising fatigue assessment approaches based on the
concept of notch stress intensity, explicitly accounting for the singularity of the stress field at the notch tip
without a fictitious description of it (see section 4.2.3). Such approaches can be regarded as in between the
notch stress or notch strain based approaches and the rather more complex crack propagation approaches.
Hence, they can provide the link for a unified approach to fatigue assessment.
4.2
381
Fatigue damage models
The following advancements have been developed in recent years:
4.2.1
Stress based concepts
Effective notch stress concept: Notches typically introduce stress concentrations reducing the fatigue
strength and fatigue life time, but not to the extent of the linear elastic notch stress increase Kt .
The effective stress concentration often turns out to be smaller. Adopting a microstructural support
hypothesis, the influence of the notch stress gradient is included by averaging the notch stress along the
(presumed) crack path over a material-characteristic micro-structural length *. The real weld toe and
weld root notch radius is artificially enlarged by employing a fictitious component f = s* to obtain
an effective one e = + f and the corresponding notch stress of the original geometry (Radaj et al.,
2009a; Fricke, 2013); an intact geometry parameter. However, a real weld seam is somehow curved and
anything but simple, e.g. is widely scattered meaning the statistical weld volume effect has to be
incorporated. Using the effective notch radius e = (, ) + s*, the effective notch stress range
becomes: e = Kf = Kt(e). The statistical component, however, is often left in the fatigue
resistance data scatter. The support factor s depends predominantly on the loading mode (uni-axial, ,
multi-axial), level of constraint (plane stress; plane strain), notch angle (2 = (5/4, 2)) for respectively
idealized weld toe- and weld root notches) and notch shape (blunt hyperbolic, root hole or blunt circular
for weld toe notches; elliptic, key-hole or U-hole for weld root notches) and last but not least the adopted
strength criterion (e.g. von Mises). Plateau values are in the range (1, 10) Radaj and Vormwald (2013)
and (Radaj et al., 2013).
The microstructural length * is typically obtained in an implicit way. Using fatigue test data, Se-N curve
parameters can be estimated. Assuming the data correlation is at maximum for the correct *, its most likely
value can be found in (Zhang et al., 2012a; Zhang, 2012). Although * is a material characteristic parameter,
the HAZ and weld material, the effects of respectively the weld toe and weld root notch are ignored. In an
engineering approach, the most likely e value can be obtained directly as well, meaning average (r, f)
contributions are involved.
For engineering applications, one fatigue resistance curve related to one reference radius r = e = 1 [mm]
has been defined (for both steel and aluminium weld toe and weld root notches) because of the
simplifications (concerning notch angle, elasto-plasticity, notch acuity, etc.,) with respect to the original
concept. The reference radius r = 1 [mm] is restricted to plate thickness tp 5 [mm] because of crosssectional weakening (weld root notches). Strengthening of weld toe notches is ignored. In both cases
structural stress corrections should be applied. Concerning notch shape, root notches are obviously critical
and classified as conservative (e.g. key-hole) and non-conservative (U-hole) as noticed in a Round Robin
(Fricke et al., 2013a), although the main criterion should be that the adopted shape, as used to obtain the
fatigue resistance curve, and the one used for the fatigue assessment, are in agreement. Note that only the
absolute notch acuity has been taken into account. It is proposed to be replaced by a relative one (Schijve,
2012), althoughat least for weld toe notchesthe plate thickness seems a better solution than a weld leg
length.
The case tp < 5 [mm], r = 0.05 [mm] has been adopted based on a completely different hypothesis, i.e.
the relationship between the stress intensity factor and notch stress as well as crack tip blunting. At the same
time, it is a compromise with respect to FE modelling and the calculation of a reasonable local stress
component (Sonsino et al., 2012).
Battelle structural stress concept: Assuming the fatigue life time of welded joints is predominantly related
to (micro-) crack growth because of the welding process induced flaws, the stress intensity factor KI has
been adopted to obtain a damage tolerant equivalent stress parameter, i.e. a through-thickness criterion
(Selvakumar and Hong, 2013).
The through-thickness weld notch stress distribution has been bi-linearised, and is related to a
relatively mesh-insensitive linear far field stress distribution (s, rs) and has been used to approximate
the crack tip wake field, i.e. to obtain a blend stress intensity factor weight function YnYf including the
notch and far field contribution. However, the bi-linearisation induced transition at 0.1tp is valid for low
weld load carrying levels only; for increasing ones the notch affected region should be increased. The
stress intensity KI however shows singular behaviour for a 0 and cannot be explained as a higher
order effect.
The notch and far field related stress intensities have been translated into a two-stage crack growth
model, a modified Paris equation. Notch plasticity has been assumed, i.e. the (anomalous) crack growth
behaviour is non-monotonic. Integration yields an equivalent stress parameter. The adopted initial defect
382
size is based on crack growth integral convergencea result of the assumed notch plasticity. The crack
growth integral I(rs) has been approximated using a polynomial formulation. It is applicable to monotonic
weld notch stress distributions only (rs > 0) and is based on a single edge crack formulation. For this reason,
only half the plate thickness has been considered to obtain the structural stress for WJs showing symmetry
with respect to half the plate thickness as adoptedalthough significantly weld load carryingin a Round
Robin (Fricke et al., 2013a). ASME implemented the BSS concept for fatigue design. Its structural strain
equivalent has recently been proposed (Dong et al., 2014).
4.2.2
Strain concepts
Most marine and offshore structures are designed such that the nominal strains remain elastic; however,
stress concentrations often cause plastic strains to develop in the vicinity of notches. The strain-based
methods, i.e. Coffin-Manson equation and its modified forms accounting for mean stress effects (i.e. the
Morrow, Manson-Halford and Smith-Watson-Topper equations) are often used to predict the crack
initiation at notches (Chen and Wang, 2012; Wang et al., 2014).
Chen and Wang (2012) proposed an approach to determine the minimum allowable size of a crackstopping hole, in which the Morrow, Manson-Halford and Smith-Watson-Topper equations are used for
low-cycle fatigue analysis. Gladskyi and Fatemi (2013) investigated the notch effects on the axial and
torsion fatigue behaviours of low carbon steel and found that the Coffin-Manson equation, with the use
of the effective strain obtained from Neubers rule, results in poor correlation of the fatigue life data of
smooth and notched specimens under axial and torsion loadings. Karakas (2013) investigated three types
of welded joints with regard to their fatigue strength based on the Smith-Watson-Topper equation,
accounting for the mean stress effects and the reference notch radius concept. His study showed that the
mean stress effects on fatigue strength were described properly by calculating the elastic-plastic
deformations in weld toes and roots and by evaluating them using the Smith-Watson-Topper equation.
Gates and Fatemi (2014) investigated the effects of notches on multiaxial fatigue behaviour of thinwalled tubular 2024-T3 aluminium specimens with a circular transverse hole. Constant amplitude fully
reversed axial, torsion, and in-phase and 90 out-of-phase axial-torsion tests were performed under load
control. The results showed that the local strain approach using Neubers rule with the fatigue notch
factor Kf was able to correlate axial and torsion data for both smooth and notched specimens in the lowcycle fatigue regime. The use of the elastic stress concentration factor Kt in Neubers rule resulted in
overly conservative life predictions.
The strain-based methods were also applied for probabilistic fatigue analysis. Based on the Morrow
equation, Lorn and Svensson (2012) performed a second moment reliability evaluation and Monte Carlo
simulations for a welded structure to investigate the variability of the fatigue life. Zhu et al. (2013) applied
the Smith-Watson-Topper equation and Bayesian framework to perform a probabilistic low-cycle fatigue
life prediction of aircraft turbine disk alloys and the results showed the prediction of fatigue life agrees well
with the experimental data. Correia et al. (2013) proposed a local unified probabilistic approach to model
both crack initiation and crack propagation of a notched rectangular plate with the use of the Coffin-Manson
and Smith-Watson-Topper equations. The results showed that there is a good agreement between
predictions and experimental data.
The strain-based methods have also been accepted by industry standards as useful methods for predicting
the fatigue life of a notched component. The American Society for Testing and Materials (ASTM) and the
Society of Automotive Engineers (SAE) have published procedures and practices for performing straincontrolled tests and predicting fatigue lives based on the test data.
Although strain-based methods have been widely applied to predict the crack initiation of a notched
component in the low-cycle fatigue regime, there are still some weaknesses in these methods:
Some effects considered in strain-based methods, i.e. mean stress effects and notch size effects, are
still empirical in nature;
Further studies still need to be conducted to improve the strain-based methods for predicting the crack
initiation of a notched component under variable amplitude loading;
The constants used in strain-based methods depend on the condition of the specimens tested. There is
still no defined way to account for differences in surface finish, plating, surface treatment, etc.
4.2.3
Notch-intensity factor, -integral and -energy density concepts
A few methods linking fracture mechanics concepts with more traditional stress based approaches were
recently revived, moving forward from the generalised Neuber concept of Fictitious Notch Rounding
(Radaj et al., 2013; Zappalorto and Lazzarin, 2014; Radaj and Vormwald, 2013) and looking at
383
parameters describing the singular stress field at the notch tip; the Notch Stress Intensity Factor (N-SIF)
concept, first introduced by Williams in 1952, is in fact the basis for various fatigue assessment
approaches ranging from the N-SIF approach itself, to the Strain Energy and Strain Energy Density
(SED) concepts, to the Peak Stress Method (PSM).
In some cases such approaches are able to explicitly account for the very local plastic behaviour of the
material and may consider either sharp or blunt notches, despite in the latter cases analytical formulae are by
far the more complex. The bridge linking a novel version of the effective notch stress approach with the NSIF concepts has been recently explained by Zhang et al. (2012a).
The literature review presented in Rizzo (2011) reports the fundamentals of the most recently proposed
N-SIF based approaches in view of their application in the shipping industry.
A research group in the University of Padua has mostly developed such methods in recent years and one
of the most promising features for the marine industry is the possibility of automatically implementing them
in finite element sub-modelling techniques, capable to transfer appropriate boundary conditions from large
and coarse shell element models to smaller and refined solid element models.
N-SIF based methods promise to overcome the need for extensive and dedicated fatigue experiments and
of validations based on component failures and in-service experience, even if it should be admitted that
developments are still necessary in two directions: on the one side Whler curves for such approaches are
still to be fully developed and/or validated for their application in the ship and offshore field, on the other
side no guidelines exist for the implementation of novel, N-SIF based, fatigue assessment approaches, rather
only research papers were presented in the literature dealing with simplified and mostly two-dimensional
geometries.
The N-SIF based approaches are relatively young and therefore their application is not widespread. Very
few references were found dealing with ship and offshore structures despite the methods appearing to be
quite promising and few independent applications have been carried out other than those by the researchers
most involved in the development of such approaches.
The most suitable approach evolving from the N-SIF method, and applicable to ship and offshore
structures, is that based on the strain energy density (SED) in a volume. Feltz et al. (2010) and Fischer
et al. (2010) compared the SED results with experimental data of lap joints and cover plates; the crack
propagation approach was applied as well. Two-dimensional finite element models were used and some
preliminary analyses were carried out showing general agreement with experimental data but also some
discrepancies because of the three-dimensional stress fields. Fischer et al. (2011a) analysed the finite
element evaluation of SED and highlighted some differences depending on the element type and shape.
Fischer et al. (2011b) presented the application to a very typical yet still challenging case, the hopper
knuckle, building a solid elements sub-model surrounded by a larger shell element model, hence
considering a complex three-dimensional stress field. Overly-conservative results were generally found.
Later another method based on the N-SIF concept was proposed allowing relatively straightforward
fatigue assessments of welded joints if well calibrated in advance: the Peak Stress Method (PSM) actually
largely simplifies the estimate of N-SIF when using particular elements and meshes. Applications to ship
structures were found in (Fischer et al., 2013; Fischer et al., 2015; Cosso et al., 2014).
The above mentioned approaches seem to overcome the difficulties introduced by other local fatigue
assessment approaches, including several different effects not considered by other methods. For example,
observing that an energy parameter is in principle a function of all loading modes, particularly the SED
criterion can be successfully used to estimate fatigue strength of welded details even if subjected to
multiaxial fatigue loading. However, N-SIF based approaches are not yet recognised as a procedure for
fatigue life assessment in the various industrial fields nor were they agreed to be included in any regulation
or standard.
4.2.4
Confidence and reliability
The determination of S-N curves is associated with large scatter (DNV GP RP 0005, 2014). Parameters in
S-N curves are often regressed from fatigue tests under constant amplitude loads. While stress responses
in ship and offshore structures are often variable amplitude random loads, in particular for larger and fast
container ships, fatigue contribution due to high frequency loads has attracted a great deal of attention.
Agerskov (2000) tested welded steel plate specimens with longitudinal attachments under variable
amplitude loads using three typical offshore load spectra: broad band, narrow band , and a modified
Pierson-Moscowitz wave spectrum with broad band character, see Figure 27. A significant difference was
indicated by the test series between constant amplitude and variable amplitude fatigue test results. It
concluded that for broad-banded random loads, the Miner fatigue accumulation value of 1/31/2 should
be used corresponding to fatigue failure.
384
Figure 27. Fatigue test results under constant amplitude loads and variable amplitude loads of different
spectra.
Osawa et al. (2013) developed fatigue tests using superimposed vibrations with a special testing machine
powered by two exciters. The preliminary test results illustrated that using the simple envelope counting
method can give a good prediction of the fatigue life. In order to simulate the low-frequent wave loading
and higher frequent whipping stresses, Fricke and Paetzold (2014) created the stress signals by using two
superimposed sinusoidal functions with different frequencies for the fatigue tests. Full-scale
measurements of stress histories on a Panamax container ship, containing typical whipping events, were
applied for the fatigue test of the specimen with double-sided transverse stiffeners. Based on the fatigue
tests, a simple method was proposed to take into account the whipping signals by multiplying a factor of
the maximum increase of wave-induced stress ranges due to whipping by the cumulative damage caused
by wave-induced stress ranges.
Due to the time, expense and difficulties associated with fatigue testing many researches have devoted
attention to the development of methods to consider the S-N parameter scatter, using a limited number of
tests or currently available test results for practical fatigue estimations. Ronold and Lotsberg (2012) showed
how to estimate characteristic S-N curves with confidence when the amount of data is limited. The
confidence level in estimating the characteristic value of cycle numbers to failure from limited data was
established to be able to maintain the same safety level under an infinite number of fatigue tests. The results
provided a new way to optimize fatigue design and may form a useful corrigendum to existing
recommendations for given confidence levels for the estimation of S-N curves.
Based on existing fatigue test data, some methods were developed to consider S-N scatter for more
reliable fatigue assessment. Johannesson et al. (2005) proposed using the equivalent stress for the regression
of S-N parameters by the maximum likelihood method. It was then applied to estimate fatigue damage
caused by random loads. The features and usefulness of this method were demonstrated from fatigue test
data by Agerskov (2000). In addition to the scatter of S-N parameters from the data, Mao et al. (2010)
employed the Cornell safety index to account for the uncertainty from encountered sea environments for
ship structures. The procedure was validated by a full-scale measurement of a container ship.
Some statistical techniques were also used to investigate the fatigue test database in order to find a
probabilistic way to describe the relation between stress cycles and fatigue failure. Recent fatigue tests have
explored the gigacycle fatigue properties of materials. The results indicated the introduction of
modifications to the conventional statistical fatigue life models characterized by a single fatigue failure
mechanism and a fatigue limit, because specimens may fail even if the applied stress value is smaller than
the conventional fatigue limit. Paolino et al. (2013) presented a unified fatigue model using random
variables to describe (knee points) transition stresses between two failure mechanisms. This model is able to
account for different failure mechanisms and the possible presence of the fatigue limit. Angelo and
Nussbaumer (2013) also pointed out that disregarding run-out data and arbitrarily fixing the fatigue limit at
107 cycles is unrealistic. A random fatigue limit model may be used to overcome this aspect, but only for the
test results with an existence of fatigue limit. While for S-N test results without a fatigue limit, this paper
proposed an alternative four-parameter fatigue limit asymptote model based on the hypothesis that S-N
curves have an oblique asymptote in the high cycle fatigue region (N>107). In the model, the parameters of
median curve and stress range-dependent standard deviation represent the randomness of the fatigue life.
385
By making use of the Database on fatigue strength of Metallic Materials published by the Society of
Materials Science Japan, Mukoyama et al. (2014) identified the correlations between the regression
parameters of S-N parameters and static mechanical properties (tensile strength) of material. Based on the
correlation, a statistical estimation method of the S-N curve was developed for iron and structural steels in
terms of their tensile strength. Pedersen et al. (2013) also collected a fatigue test database from public
literature to investigate the thickness effect for the fatigue assessment of transverse butt joints. The test data
were taken to compare with the thickness correction according to the IIW recommendations and most other
codes/guidelines; it was concluded that these codes/guidelines are very conservative for fatigue estimation
of butt joints.
4.3
Fracture mechanics models
The trend in the ship and offshore building industry is to construct ever bigger structures by means of
their size and as a result, strength of material must be satisfied under several loading scenarios. Apart
from strength, toughness of material has become an important issue for big structures and both must have
acceptable limits. Unfortunately, the attainment of both of these properties is not a simple issue and there
are several studies on this subject. Ritchie (2011) indicated strength-toughness relationships for several
engineering materials (Figure 28) including new state-of-the-art materials, i.e. metallic glasses, metallicglass composites, as well as future trends of materials (indicated as a white arrow in Figure 28). In this
study, it is concluded that both of these properties are related to several length scales, i.e. nano to macro
scale, and one should consider the effect of different length scales on material behaviour.
Figure 28. Conflicts of Strength versus Toughness (Ritchie, 2011).

Constructing ever bigger ship structures leads to the application of thicker steel plates in the shipbuilding
industry. Several attempts have been undertaken in order to gain both strength and toughness of
shipbuilding materials, especially for thick steel plates. In January 2013, the International Association of
Classification Societies (IACS, 2013) released a new rule, the so-called Requirement for Use of
Extremely Thick Steel Plates. In this documentation, several requirements and recommendations are
highlighted for the use of extremely thick steel plates, i.e. between 50 mm and 100 mm, concerning brittle
fracture toughness and brittle crack arrestabiliy.
There are several parameters describing the fracture toughness of materials, such as energy release rate
(G), stress intensity factor (K), J integral, crack tip opening displacement (CTOD) and crack tip opening
angle (CTOA). These parameters are related to material behaviour which can be linear elastic, non-linear
elastic or elastic-plastic. While measuring and evaluating these parameters, proper test methods must be
adopted. A detailed research reviewing these issues from past to state-of-the-art developments has been
undertaken by Zhu and Joyce (2012). Their review is based on the American Society for Testing and
Materials (ASTM) standards. Usually fracture tests give too conservative results when compared with actual
386
structures. For this reason, Bayley and Aucoin (2013) examined the fracture behaviour of welded single
edge notched tension specimens using constraints and loading conditions encountered by ship structures in
service. They examined large scale samples of hull structural material. They developed a crack mouth
opening displacement (CMOD)CTOD transfer function utilizing the Finite Element Method (FE)
simulations of these samples and this led to the determination of fracture toughness values. So, fracture
toughness values of non-standardised tests specimens can also be determined using the same procedure. On
the other hand, while measuring fracture toughness values of thick steel plates, large test specimens and a
testing machine, with significantly large expenditures, are necessary. This issue was addressed by Yajima et
al. (2011) who used a small sized centre-notched tension test specimen using steel plate thickness as the
specimen width. They accomplished evaluating the fracture toughness of 70mm thick steel plate using this
specimen.
The relationship between the microstructure of weld joints and fracture toughness was investigated by
Leng et al. (2012). They performed several experiments including the CTOD test, metallographic analysis
and fracture surface analysis on high strength, low alloy steel S335G10+N, which is suitable for marine
structures. Welded joints are subjected to different thermal cycles and experience crystallization and solid
transformation, which lead to heterogeneity. Samples from weld position (WP), fusion line (FL) and
fusion line plus 2mm (FL+2) are used in experiments. In their study, average grain sizes of samples were
measured and compared with CTOD values that are listed in Tables 1 and 2. It was observed that average
grain size distribution is in agreement with CTOD values. The larger the CTOD, the smaller the average
grain size will be.
Table 1. CTOD Values.
Table 2. Average grain sizes.
In fracture mechanics, crack arrest toughness (Ka) has also become an important parameter to determine
the failure characteristics of materials, along with crack initiation toughness (Kc). After crack initiation
takes place, it is important to arrest its propagation in order to prevent failure or fracture of the related
structure. In the fatigue study field, crack arrest toughness and/or crack retardation effects are also
investigated for failure characteristics of structures under several variations of loads, i.e. sea loadings. An
et al. (2014) investigated crack arrest fracture toughness (KIa) characteristics of thick steel base material
and its heat-input weld for 50 mm thick high strength shipbuilding steel and concluded that KIa is a linear
function of temperature. In their experiments, while base material (BM) satisfied IACS (2013) rule, this
was not the case for the weld material. In addition, they obtained a straight crack propagation path along
the fusion line (FL) of weld and that is why investigation into the material properties of the path, i.e. grain
size, hardness, Charpy impact energy, was carried out; they found that localised degradation of materials
strength and toughness in the impact notch region of the experimental setup can be the cause of
disparities. Note that it is also important to obtain satisfactory fracture toughness values for weld joints of
ship structures. Moon et al. (2013) carried out several experiments to obtain CTOD values of API 2W
Gr.50 steel at welds and their heat affected zone (HAZ) using two types of welding: submerged arc
welding (SAW) and flux cored arc welding (FCAW). Apart from the utilisation of crack initiation and
arrest toughness, crack tip opening displacement (CTOD) is a commonly used parameter for evaluating
fracture toughness of structures in fracture mechanics. It can be related to the J integral using the plastic
constraint factor, m (Equation 2).
J = m y
(2)
where is the CTOD value. From several experiments using two types of welding Moon et al. (2013)
concluded that a plastic constraint factor based on the ASTM standard, E1820, does not include the effect
of weld process and it is too conservative. They also carried out Charpy impact and hardness tests of
387
SAW and FCAW welding processes and compared them with each other. They concluded that CTOD
values have similar tendencies to impact energy and hardness results.
A comprehensive research study has been made by the Japan Ship Technology Research Association
(JSTRA) to prevent failure in large container ships and the results are published by Sumi et al. (2013). They
mainly focused on the initiation of brittle crack, fatigue crack growth under several sea loading conditions
and investigated crack arrest designs in the deck structure of large container ships for a thickness range of 50
75mm steel plates. It is emphasized that allowable initial defects in the weld joints can grow in critical
sizes and may cause brittle fracture. They also referred to guidelines published by ClassNK on Brittle Crack
Arrest Design for countermeasures of brittle crack propagation. It has also been shown that the most reliable
method to prevent crack propagation along welded joints is a weld line shift, as shown in Figure 29.
Figure 29. Concept of weld line shift (Sumi et al., 2013).

A more detailed research on fatigue crack growth has been undertaken by Sumi (2014), considering the
same locations and thickness ranges of container ships. In this study, fatigue crack growth was simulated
using two types of model, 1) crack growth model based on K RP and simple crack growth model, the socalled Paris-Elber law, and 2) a random sequence of clustered loading, the so-called storm model. Fatigue
lives of K RP criterion were found to be 23 times longer than the simple method, because retardation
and acceleration effects cannot be considered by the latter model which does not properly take into
account the increase in plastic wake (Figure 30). Crack growth model based on K RP is basically defined
by Equation 3; please refer to Toyosada et al. (2004) for a more detailed discussion.
da / dN = C ( K RP )
Figure 30. Formation of plastic wake during fatigue crack growth (Sumi, 2014).
(3)
388
In the same paper by Sumi (2014), the effect of slamming-induced whipping stress on fatigue crack
growth is also investigated since the longer marine structures mean more slamming-induced vibratory
stresses considering their flexible bodies. It is concluded that M-series specimens may be used in order to
estimate fatigue crack propagation life under whipping stress. But still further research is needed in this
area.
Qumner et al. (2013b) made a fatigue crack evaluation of stiffened plate longitudinal elements
located at the bottom and deck regions of ships where critical bending stresses occur. First, critical crack
lengths were evaluated for extreme bending moments which represent the fracture resistance of members
based on actual CTOD values and these critical lengths correspond to the fatigue life of each component
for which fatigue crack propagation becomes unstable. Another fatigue crack evaluation study for
longitudinal elements of ship structures has been undertaken by (Mao, 2014; Mao and Ringsberg, 2013).
They considered an effective method for fatigue crack propagation which is based on the narrow band
spectral fatigue method and is validated by full scale measurements of a 2800 TEU container ship. In
general, ships are subjected to several cyclic loads during their service life and it is hard to expect an
occurrence frequency of these loads. Additionally, other uncertainties such as residual stresses from
manufacturing, corrosion, weld defects etc. can also cause difficulties in fatigue analysis. So it is highly
probable that fatigue cracks may occur earlier than expected. Due to the different approaches that are
applied for fatigue strength assessment of ship structures, Fricke et al. (2012) performed several tests
using two types of structure, i.e. web frame corners as well as the intersection between longitudinal and
transverse web frames, under several loading conditions and all available techniques were applied to
models as far as possible. They investigated both large and small scale specimens experimentally and
numerically. In their analysis, fabrication-related conditions, welding induced pre-deformations and
residual stresses were all included. At the end, they summarized the strength behaviours of the chosen
complex welded structures.
Mechanical properties of materials such as fracture toughness, strength, fatigue life can be improved with
several techniques. In this respect, Rubio-Gonzlez et al. (2011) studied duplex stainless steels which are
used in different fields including the ship industry. These steels have high strength and excellent fracture
toughness as well as high corrosion resistance. They investigated the relatively new laser shock processing
(LSP) technique and its effect on mechanical properties of these steels. It was already known that the LSP
technique, especially, increases fatigue crack initiation life and reduces the fatigue growth rate of materials.
Their work confirmed that the application of LSP on 2205 duplex steel improved fatigue properties and no
effect has been found on micro-hardness and microstructure of material but an improvement in fracture
toughness was reported. Kim et al. (2012a) studied the mechanical properties of adhesive joints which are
used to join stainless steel sheets. Stainless steel sheets have been used at a cryogenic temperature of
150C for their relatively high mechanical and low coefficient of thermal expansion (CTE) properties in
the containment system of liquefied natural gas (LNG) ships. However, adhesives of steel sheets become
quite brittle at that temperature and need to be reinforced to improve fracture toughness values. In this study
the film-type epoxy adhesive was reinforced with randomly oriented aramid fibre mats. As fracture
toughness values are influenced by mechanical properties as well as thermal residual stresses, both
properties are investigated. It was found that aramid fibres could reduce the thermal residual stresses
between the stainless steel and adhesive layer. The optimum volume fraction of the aramid fibre mats was
found to be 16.3% from the double cantilever beam (DCB) tests of the adhesive joint. It was concluded that
the adhesive reinforced with aramid fibre mat has a higher load-carrying capacity and fracture toughness at
a cryogenic temperature of 150C than those of other choices, such as polyester or glass fibre mats. An
LNG ships containment system is mainly composed of dual barriers and insulation board. The insulation
board must be reliable against leakage of LNG and have a high thermal insulation performance. The
reliability of insulation board, considering sufficient fracture toughness at a cryogenic temperature of
163C, was studied by Yu et al. (2013). Conventionally, the polyurethane foam (PUF) reinforced with
glass fibre (RPUF) is used for insulation board to increase fracture toughness. This implementation not only
increases costs but also decreases thermal insulation characteristics. As a result the volume fraction of glass
fibre is restricted to less than 0.5% and sufficient fracture toughness is uncertain. This uncertainty can
possibly lead to crack propagation in insulation board and induce leakage of LNG. Cracks usually initiate
and propagate as a result of local tensile stress concentrations due to the temperature change. In this study,
the crack resistance of insulation board increased with the reinforcement of glass fibre polymer composite in
selective areas which have high stress concentrations. Mainly glass fibre reinforcement sustains high
thermal stresses by its high stiffness property. The tensile failure strengths of both a conventional RPUF
system and glass composite reinforcement were measured and the safety factor of the latter is calculated as
389
42% higher than the former by Equation 4 below:

(4)
4.3.1
Crack growth rate model
A very important engineering method used to predict fatigue crack growth is still based on Paris law. A
number of modifications to the Paris law have been proposed in recent years to account for different
fatigue phenomena such as load sequence effect, load ratio effect, specimen size effect, environmental
effect etc., especially load sequence effect. Ship structures are subjected to complex variable amplitude
loading (VAL) stress-time histories during their service lives. As a result, many modifications to crack
growth rate models are focused on a particular loading mode, such as the retardation effect resulting from
the applied peak tensile overload cycles.
Iranpour and Taheri, 2013, 2014 validated the effectiveness of some of the commonly used analytical
crack growth models for estimating the fatigue life of the material under the VAL stress-time history using
the Willenborg retardation approach, combined with the Walker fatigue model, which has proved to be nonconservative, thus undesirable for design purposes. The differences seen in the prediction of the fatigue life
by the use of the widely used Willenborg retardation model, would provide an incentive to researchers to
further investigate the influence of the loading effect. Chen et al. (2012) proposed an improved crack growth
rate model expressed by Equations (5)(6) to consider the effect of variable amplitude loading by modifying
the crack closure level based on the concept of partial crack closure due to crack-tip plasticity, as shown in
Figure 31, which can explain the phenomena of retardation due to overload and the tiny acceleration due to
underload.
m
da = A( K max K op K effth )
(5)
n
dN
K
1 max
K C
aOL + rOL ryi rUL
aOL ai aOL + rOL ryi rUL

where =
ai
ai > aOL + rOL ryi rUL

1
(6)
rUL
ai
aOL
rpi
rOL
Figure 31. Schematic representation of modifying plastic zone sizes to account for the effect of underload
following overload.
Abdullah et al. (2012) tested the applicability of a unified two-parameter fatigue crack growth model for
high-low sequence loading conditions and identified some uncertainty parameters to evaluate the
prediction accuracy of the model. Dai et al. (2013) proposed a plasticity-corrected stress intensity factor
to describe the effects of overload on the fatigue crack growth in ductile materials. From the use of this
new crack driving force parameter, a theoretical correlation of the Paris law with crack tip plastic zone is
established, which can describe fairly well the effects of the crack length and different loading variables,
including the applied stress intensity factor range, load ratio, overload extent and overload number, on the
overload retardation. Liu et al. (2013) discussed a new formula under single overload and evaluated the
residual life under single overload by adopting a dynamic coefficient mechanics model. Silitonga et al.
(2014) pursued a numerical method based on a cohesive zone model aimed at simulating fatigue crack
growth as well as crack growth retardation due to an overload. Zhan et al. (2014) proposed a new
approximate model for predicting the crack growth rate which can condense the stable crack propagation
stage curves for non-zero stress ratios to that for R = 0 in a narrow band. By using the new approximate
390
model, the obtained material constants corresponding to R = 0 can be further and widely used for
different loading amplitude ratios. All the above mentioned new modified models in recently published
literatures are verified by experimental data to some extent.
However, almost all existing fatigue prediction models using fracture mechanics have a reversal-based
approach (or cycle-based formulation). However, for most engineering structures, load history is usually
very complex and far from being regularly cycled. The conversion from random load history to cyclic load
sequence (such as the cycle counting technique) introduces additional complexities and uncertainties in
fatigue damage modelling. And the cycle-based approach makes it impossible to continue reducing the time
scale for more fundamental investigation since the smallest time scale is one cycle, as shown in Figure 32
(Lu and Liu, 2010). It is also observed that crack growth is not uniformly distributed within a loading cycle
and only happens during a small portion of the loading path. Multiple mechanisms exist within one cyclic
loading, which is not able to be captured using the classical cycle-based approaches (Zhang and Liu, 2012).
The recently developed small time scale fatigue crack growth model can be expressed as follows:
da (t ) =
ctg
2
d = Cd (t )
(7)
, is CTOD, and is the crack tip opening angle (CTOA).

where, C =
The crack length at any arbitrary time can be calculated via direct time integration as,
ctg
2
t + t
where, is redefined as = 3
8E y
1 da
Ca dt
t + t
d
2
1C 2 dt
(8)
so as to include the hardening effect.
Figure 32. Schematic illustration of the small time scale crack growth model: (a) difference in time scales of
the small time scale model and the cycle-based approach; (b) Geometric relationship between
crack growth and CTOD (Lu and Liu, 2010; Zhang and Liu, 2012).
4.3.2
Crack growth assessment
Crack path prediction is one of the bases in the remaining life assessment of structures when considering
crack growth, crack arrest and the design of critical details. In the past, the crack growth simulation has
usually been based on a predefined path. In the last five decades, there have been substantial advances in
the understanding and prediction both of the macroscopic aspects of crack paths, largely through
developments in fracture mechanics, and in the application of modern computers and microscopes.
The papers presented at the International Conference on Crack Paths (CP 2009) show that understanding
of crack paths is increasing (Pook, 2013). Judt and Ricoeur (2013) present numerical methods used for
predicting crack paths in technical structures, based on the theory of linear elastic fracture mechanics. A
precise FE-mesh technique is required, as shown in Figure 33. Several crack extension criteria have been
proposed in the past decades to describe crack paths under mixed mode loading conditions, such as
maximum principal stress criterion and the minimum extension of the core plastic zone criterion. The
crack path is mainly affected by loading conditions, structural details, residual stress distributions and
material microstructural features, such as unavoidable inhomogeneities when the conditions of crack
initiation and crack growth under fatigue loading have to be evaluated (Brighenti et al., 2012; Sumi and
Okawa, 2013). So, in complex conditions, the propagation path was unknown, the accuracy of simulation
was decreased and the application of fracture simulation was limited (Chen and Chen, 2012). However,
with the development of computational techniques, some researchers developed algorithms to simulate
three-dimensional crack growth more precisely under non-proportional mixed-mode loading based on
linear elastic fracture mechanics, e.g. Yang and Vormwald (2013). In recent years, (Sumi and Okawa,
391
2013; Sumi, 2013; Sumi et al., 2013) continuously introduced the computational simulation methods
developed for multiple fatigue cracks propagating in a three-dimensional stiffened panel structure used in
ships, by taking into account the effects of weld, complicated stress distributions at three-dimensional
structural joints, and structural redundancy.
Figure 33. FE-mesh of the grown curved crack shown in a deformed configuration.
Multiple crack identification plays an important role in the vibration-based crack identification of structures.
Complex engineering structures with multiple cracks are difficult to simulate accurately. A traditional crack
detection method of a single crack is difficult to be used in a multiple crack diagnosis (Li et al., 2013a).
Some research efforts in recent years have focused on numerical technique developments on multiple crack
initiation and propagation. Kim et al. (2012b) proposed an extended finite element method to be applied to
predict lifetime under constant amplitude cyclic loadings of fatigue tests on several multiple site damage
specimens. The numerical results were compared to testing data by using Forman and NASGROW fatigue
equations. Hollnder et al. (2012) developed an advanced numerical algorithm to analyse the multiple crack
growth at each crack simulation step automatically in combination with a numerical crack path simulation.
The additional crack initiated during the crack growth process is considered in a fully automatic analysis.
Oliveira and Leonel (2013) applied a boundary element method to model multiple crack propagation in twodimensional domains. An interactive scheme is proposed in order to predict the crack growth path using
maximum circumferential stress criteria and the crack length increment by using a displacement correlation
technique at each load step. The scheme proposed by them is proved to be accurate enough to simulate
localisation and coalescence phenomena. Li et al. (2013a) presented a three-step-meshing method for the
multiple cracks identification in structures. The multiple cracks quantitative diagnosis process can be
successful with an accurate and reasonably finite element structural model whose calculated values of
natural frequency agree well with the measured values. Judt and Ricoeur (2013) present numerical methods
used for predicting crack paths in technical structures based on the theory of linear elastic fracture
mechanics in combination with a smart re-meshing algorithm. Shi et al. (2013) used an extended meshless
method, based on partition of unity to simulate multiple cracks, which has a higher computational efficiency
and precision compared with the modified intrinsic enriched meshless method. Similar work on meshfree
methods to simulate multiple crack growth and coalescence, has been undertaken by Barbier and Petrinic
(2013). Pang and Chew (2014) presented an algorithm for multiple surface cracks originating from weld
toes. All these efforts are compared with experimental findings to show their accuracy and efficiency, which
can provide the basis for further application.
4.3.3
Fracture mechanics based fatigue evaluation of ship structures
Present ship building rules follow the S-N curve approach to evaluate the fatigue damage at identified
locations while the use of a fracture mechanics approach has yet to receive due attention. Feng et al.
(2012) calculated the crack growth using Paris law on the basis of simulating the fatigue stress
392
history of ship structures and calculating the stress intensity factors in random sea states. Doshi
and Vhanmane (2013) have done some work aimed at exploring the prospect of fatigue evaluation of
ship structure detail, such as the connection of the longitudinal stiffener with the transverse
web frame, using the fracture mechanics technique. The study is performed on weld details at the midship of an oil tanker vessel. Paris law, combined with model parameters reported by others, are used.
Randomness in the crack size, crack growth model parameters and loads has been considered and the
results are thus presented within a probabilistic framework using Monte Carlo simulations. It is
observed that the fracture mechanics based fatigue life predictions are in good agreement and
conservative compared to the predictions using the S-N curves, as given in the common structural
rules. Bardetsky (2013) analysed the progressive structural failure caused by cracks emanating from
the damaged area by using a procedure for predicting the crack propagation under sea wave loading
and based on Paris law combined with a finite element model to calculate the stress intensity factor.
The procedure for the estimation of the crack propagation is implemented for a typical modern
170,000 DWT bulk carrier in full load condition. Qumner et al. (2013a) investigate the fracture
failure of ship longitudinal members, constituted of a stiffener and its attached plate, in which a
crack has propagated by fatigue from either the stiffener flange or fillet weld. The study employs the
failure assessment diagram methodology to assess the conditions of failure at the crack tip, which
enables determining critical fatigue crack lengths corresponding to the maximum longitudinal stresses
derived from extreme loads. Zilakos et al. (2013) established a common base for evaluating
difference crack arrest technologies following the guidelines proposed by the CSR based on the
Paris law.
To sum up the recent research work, fracture mechanics based methods have attracted attention but the
footsteps stop in the Paris law. More progress has been made on the crack growth theory which requires
more research work for application in the fatigue life assessment of ship structures.
4.4
4.4.1
Rules, standards & guidance

Ship rules
Harmonized Common Structural Rules (CSR-H)

Some of the highlights of CSR-H in comparison with CSR-OT and CSR-BC are as follows:
1. Loads: CSR-H uses the Equivalent Design Wave (EDW) load approach, which is similar to CSR-BC;
however, CSR-OT uses the envelope load approach. Oblique Sea (OS) load cases are included in
CSR-H but are not considered in CSR-BC and CSR-OT.
2. Probability level: In both CSR-BC & CSR-OT, the probability level of 10-4 is used for fatigue load
calculation; however, in CSR-H the 10-2 probability level is used. This change in level of probability
is because loads at this probability level have maximum contribution to damage, and fatigue life
shows less sensitivity with the Weibull shape parameter.
3. Details to be assessed using very fine mesh analysis: In the case of CSR-OT, only the hopper knuckle
connection is considered for fatigue assessment by using very fine mesh analysis, but in CSR-H some
more details are considered (e.g. connection of lower stool to inner bottom plating, etc.)
4. Fatigue assessment by screening: This method is newly introduced in CSR-H; in this method fine
mesh analysis with 50 mm mesh size at given locations is carried out and the obtained stress is
multiplied by a stress magnification factor (SMF), to obtain the hot spot stress. This stress is
used to determine fatigue life. This type of fatigue screening is not available in both CSR-OT &
CSR-BC.
5. Reference stress for hot spot stress calculation: CSR-H uses principal stress within +/ 45 as well as
outside +/ 45 with respect to normal to the weld. Stress obtained outside +/ 45 is multiplied with
factor 0.9, to take into account the effect of the S-N curve. The maximum stress range out of the two
is chosen for fatigue life calculation. CSR-OT uses stress normal to the weld and CSR-BC uses
principal stress within +/ 45 with respect to normal to the weld.
6. Calculation of stress range: In CSR-H, stress range and mean stress, for a particular loading
condition are calculated using individual stress components for load cases i1 and i2. However in
CSR-BC, the stress range and mean stress are calculated using principal stress in individual load
cases i1 and i2. CSR-OT uses stress combination factors for the calculation of stress range.
7. Fatigue assessment at stiffener end connections: The nominal stress approach is used in CSR-OT, and
the notch stress approach in CSR-BC. In CSR-H, the hot spot stress approach is used for fatigue
assessment at stiffener end connections.
393
8. Stress due to relative displacement: In CSR-BC and CSR-OT, two separate approaches are used to
calculate stress due to relative displacement. These individual approaches are kept as they are for bulk
carriers and oil tankers, in CSR-H.
9. Special procedure for web stiffened cruciform joints: In CSR-H, a special procedure has been
introduced (considering bending effect) for hot spot stress calculations for web stiffened cruciform
joints. In CSR-BC, for the assessment of stress in a cruciform connection, stress correction factor is
used, which is based on the angle between intersecting plates. CSR-OT uses stress normal to the weld.
10. Corrosive environment consideration: CSR-BC and CSR-OT use factors fcoat and fSN respectively to
take into account the effect of corrosion. However CSR-H uses two different S-N curves to calculate
individual damages (in air, in corrosive environment), and finally added to obtain fatigue life.
11. Thickness correction: CSR-BC uses thickness correction factor fthick which is a function of thickness
of the member. In CSR-OT, thickness effect is included in the type of S-N curve used. CSR-H uses
factor fthick, for the correction of thickness effect; this factor is a function of thickness of member and
the attachment types.
Fatigue assessment of ship structuresDNV guideline

This Classification Note is intended to give a general background for the rule requirements for fatigue
control of ship structures, and to provide detailed recommendations for such control.
In the DNV GP RP 0005 (2014) version the requirements are aligned with DNV Ship rules and IACS
CSR-H rules for specific details. In addition improved formulations and less conservative approaches for
details with limited damage experience have been included.
Fatigue assessment of ship structuresABS guideline

ABS (2012) is an update of ABS (2009) which has introduced a new guideline to provide supplementary
requirements for the application of higher-strength, thick steel plates, i.e. greater than 51 mm, within the
structure of large container carriers. The requirements given in the guideline are based on extensive experience
with the design, construction and in-service performance of large and ultra large container carriers.
When the hull girder strength is designed to the rules minimum requirements, the accompanying effects
of higher-strength thick steel plates are largely associated with higher stress levels and reduced fatigue and
fracture strength characteristics. In the upper flange of the hull structure, wave-induced fatigue damage of
the thick plated weld connections is absolutely critical. As a countermeasure, the fatigue behaviour of these
weld connections is to be extensively evaluated to avoid crack initiation. Satisfactory fatigue and fracture
characteristics are to be attained from improvements in structural design measures, steel materials, welding
consumables, welding procedures and post-weld enhancements.
In this guideline, special attention is provided for the butt welds and hatch corners of upper flanges for the
high strength thick steel plates of H40 or H47 grade steel. In order to minimise the fatigue failures of thick
plates in the container ship, the rules recommend the construction procedures, weld treatment methods and
fatigue design procedures specific to the thick plates. The guideline also incorporates the effect of springing
and whipping on the fatigue strength of upper flange of hull structure. ABS (2012) pays special attention to
fracture avoidance, particularly by using inserted plates as crack arrestors.
Fatigue assessment of ship structuresLR rule

As part of the Lloyd's Register (LR) ShipRight design and construction procedures, LR has developed a
fatigue design assessment procedure, ShipRight FDA ICE (2010), to assess the fatigue damage of ship
structure induced by ice loads for ships navigating in ice covered regions.
The ShipRight FDA ICE assessment procedure examines ship-ice interaction loads, ice-load impact
frequency, ice-load distribution, structural responses and the fatigue behaviour of hull structures in cold
temperatures, including associated fatigue responses. The fatigue-response assessment is determined for
different winter conditions and ice thicknesses on typical routes for winter trade.
Zhang et al. (2011) presented the salient features of the Ship Right FDA ICE guideline. It is explained by
the authors that compared to non-ice classed ships, ice classed ships have increased scantlings to strengthen
the hull structure against ice loads; it is normally expected that ice classed ships would have fewer cracks
and fractures due to fatigue. The data analysis, however, shows the opposite, which implies that cracking of
the hull structure caused by cyclic ice loading during ships navigating in ice regions is an important
phenomenon that warrants investigation and assessment in the structural design of the ships.
394
Fatigue assessment of ship structuresGL rule

GL (2012) introduced guideline for the fatigue analysis of racing yacht keels. The method of calculating
dynamic stress amplitudes and fatigue calculations at keel connections are described in the rule. The rule
is applicable for fatigue analysis of Yachts with a canting keel as well as those with a fixed keel.
GL (2013) introduced a guideline to assess the high frequency hull girder response of container ships.
This guideline was established to aid the designer in assessing the effects of hull girder fatigue loads caused
by the high frequency response of the ships structure. It comprises a semi-empirical approach based on the
results from the measurement campaigns as well as on collected damage data from the fleet of container
ships, currently classed by Germanischer Lloyd. The method used reflects all kinds of hull girder vibrations,
including whipping and springing. These high frequency loads are transferred to any considered container
ship as a function of averaged rule slamming pressures which are considered to comprise parameters
important for high frequency ship response.
In GL (2014) the thickness effect for fatigue strength was updated. Now as well as the geometrical notch
effect, the requirement comprises a consideration of the allowable axial misalignment. Finally, for butt
welds with standard quality (FAT80) a moderate thickness exponent of only 0.1 is required.
Fatigue assessment of ship structuresBV rule

BV NR 583 (2012) is dedicated to long term loading calculation when springing or whipping is prone to
occur. Fatigue is concerned both for linear quasi-static and dynamic fatigue damage, and non-linear
fatigue assessment including whipping. Quasi-static and linear dynamic assessment may be carried out by
a spectral analysis, when non-linear fatigue needs a time domain analysis.
BV NR 483 (2011) is dedicated to Naval ship design. Fatigue rules are very similar to rules applicable to
merchant ships but due to specific operating conditions, naval ship loads are different, e.g. longitudinal
bending moment depends on the characteristic Froude number, and may be corrected by a factor taking into
account dynamic loads due to bow flare impact.
IACS URS 3
These unified requirements deal with the use of extremely thick steel plates, mainly on container ships. It
is intended to define either structural arrangements or material properties leading to stopping a sudden
break propagating, starting from an undetected defect. It is applicable for thick plates such as container
ships decks and for higher tensile steels.
4.4.2
Design codes for Offshore Structures
DNV (Fatigue Design of Offshore Steel Structures)

DNV GP RP 0005 (2014) deals with the fatigue design of offshore structures. Fatigue assessment can be
carried out based on the S-N curve approach or fracture mechanics approach. In the present version of the
code, the stress concentration factors for tubular butt weld connections due to eccentricities (due to the
change in thickness or misalignments) have been updated. Stress concentration factors have also been
updated for members of equal thickness.
Recommendations for applicable S-N curves at the inside of tubular joints and S-N curves for piles have
been added. In addition a section on thickness effect for butt welds and cruciform joints has been added
and the thickness exponent for S-N class C and C1 has been modified.
DNV RP-C206 (Fatigue Methodology of Offshore Ships)

DNV RP-C206 (2012) has moved design fatigue factors and details to be assessed to DNV OS-C102
(2012). Criteria related to initial fatigue assessment are updated (stress concentration factor kg).
ABS (Fatigue Assessment of Offshore Structures)

ABS (2013) (fatigue) has updated requirements related to the benefits due to grinding, in the case of the
tubular intersection connections. The use of grinding is allowed, and the requirements for the same are
specified. A benefit factor of 2 is also applicable in the case of ultrasonic/hammer peening providing
suitable documentation is provided.
BV NR593 Ship Conversion into Surface Offshore Units and Redeployment of Surface Offshore
Units
BV NR 593 (2012) is dedicated to the assessment of converted marine structures. Fatigue assessment
involves both fatigue due to operation before conversion and expected fatigue after conversion.
395
Evaluation of fatigue due to operation before conversion may take advantage of the knowledge of actual
loading conditions. Within the standard, repaired details are considered as having a damage
corresponding to their actual life.
BV NR578 Rules for the classification of Tension Leg Platforms (TLPs)

BV NR 578 (2012) is dedicated to the assessment of TLP fatigue requirements involving both hull and
Tendon Leg Systems (TLSs). Depending on the linearity of loading, spectral analysis is proposed, or time
domain analysis is required. When appropriate, deterministic analysis is accepted. Fracture mechanics are
also considered as an alternative approach.
BS 7910
BS 7910 (2013) is intended to justify the ability of structures to resist loading, by considering different
failure modes such as fatigue but mainly concentrating on crack propagation and unstable cracks. BSI
7910 is currently used to assess, on offshore structures, the ability to postpone repairs when cracks are
discovered during inspection. It has been updated from BSI:2005, adding annexes on NDT and corrosion.
It has been drastically modified for the analysis of unstable fracture mechanics using FITNET and R6
results.
The fracture strength reference has been shifted from critical CTOD to critical Stress Intensity Factor
( ). The effect of the residual stress contribution to the stress intensity factor has been reanalysed, and
total stress intensity factor now involves three components: one dealing with tension stresses, another
dealing with bending, the last one corresponding to a residual stress, stress intensity factor based on a design
residual stress distribution. For a crack length greater than the thickness of the structure, it results in a less
conservative approach than in the 2005 revision. A procedure has been developed to avoid an overconservatism related to shallow crack effect (constraint effect).
4.4.3
IIW recommendation
Document IIW (2013c) XIII-2460-13 is an IIW document following Document XIII-2151-r4/XV-1254-r4

dealing with the fatigue of welded and cut structures or components. In particular, this document provides
procedures to check fatigue life according to nominal stress, structural stress or effective notch stress
analysis. In addition, with this last document, a crack propagation procedure is developed, and a
background document has been drafted for traceability reasons.
Document IIW (2013a) XIII-2301r3-13 is an incomplete draft to Guideline Weld quality; in relation to
fatigue strength it reanalyses the current welding quality standards in order to clarify the criteria dealing
with fatigue performances. The quality criteria are ranked according to their sensitivity to fatigue life with
respect to the weld design. Considerations are provided for the inspection methods and the consequences on
fatigue strength confidence.
Document IIW XIII-2380r3-11/XV-1383r3-11 provides a comprehensive description of methods to
analyse fatigue strength in the weld bead commonly called root crack fatigue. Up to now this analysis is
not explicitly considered in ship rules, but implicitly through the requirements of minimum weld throat.
With the higher strength characteristic of modern steels, the relevance of these requirements have to be
examined.
4.4.4
ISO Standards
ISO/TR 14345:2012 (2012) is guidance on the best practices for fatigue testing under constant or variable
amplitude loading, including specimen selection, e.g. measurement of radius at weld toe, monitoring of
testing itself and data treatment of testing results to obtain a fatigue strength including data sheet. This
ISO document has been developed as Document XIII-2140-6 within the IIW framework.
ISO 12110-1:2013 (2013) and ISO 12110-2:2013 (2013) deal with variable amplitude load testing. The
documents in addition to describing the overall test parameters focus on methods for cycle counting and
signal reconstruction.
Although different counting methods are enumerated, the second part of ISO 12110 focuses on rain
flow counting and provides numerical examples allowing assessment of the software using the rain flow
method.
4.5
Acceptance criteria
Traditionally, the acceptance criteria for the high-cycle fatigue strength assessment of a structural
component are established based on the design fatigue life and the applied safety factor. If the calculated
fatigue life exceeds the design life divided by the applied safety factor, the fatigue strength of the
396
component is considered to meet the design requirements. Otherwise, the component should be
redesigned to satisfy the requirements.
As for the assessment of fracture resistance, the so-called Failure Assessment Diagram (FAD) (API RP
579-1/ASME FFS-1, 2007; BS 7910, 2013) is normally performed based on the principles of fracture
mechanics. Typically, the vertical axis of the FAD represents a ratio of the applied conditions to the
conditions required to cause fracture and the ratio of the applied load to that required to cause plastic
collapse is plotted on the horizontal axis. Fracture mechanics analysis for a crack provides either the coordinates of an assessment point or a locus of points. These positions are compared with a defined
assessment line to determine the acceptability of the crack.
Recently, the acceptance criteria were also developed for fatigue and fracture assessment under extreme
environmental loading. This development was triggered by the unexpected large number of cracks observed
in many platforms in the Gulf of Mexico after hurricanes passed. API RP 2T (2010) recommended a socalled single event fatigue analysis, in which components that are susceptible to low-cycle/high-stress
fatigue should also be analysed to assess fatigue damage during a rare/extreme event that may be of
extended duration, such as a 48-hour rise and fall of a 100-year storm. Analyses have found that fatigue
damage during such a rare/extreme event may be comparable or even higher than high cycle fatigue
accumulation over the service life.
Chen et al. (2012) developed acceptance criteria to determine the minimum allowable size of a crackstopping hole. The limiting condition of this stopping hole is that no new crack would initiate from it when
the structure is subject to a storm with a return period of one year or longer. The time to initiate a new crack
at a crack-stopping hole was predicted based on the low-cycle fatigue analysis. A safety factor of 3 was
selected such that the predicted crack re-initiation time would be 3 times longer than the time needed for the
cracked structure to sail to a repair facility.
4.6
4.6.1
Measurement techniques
Crack growth and propagation
Conventional experimental techniques, such as strain gauge, photo elasticity and more interferometry,
have been used to determine stress intensity factors (SIFs) during recent decades. Although these
techniques have greatly improved the accuracy of SIF measurement, they also have some limitations. For
example, the angular and radial locations of strain gauges are complicated to determine when measuring
the SIF near the crack tip region. Sarangi et al. (2013) conducted experiments to determine the optimal
strain gauge locations, and results showed that highly accurate values of SIFs can be obtained if the strain
gauges are placed at the optimal locations. Chakraborty and Murthy (2014) proposed a new single strain
gauge technique for accurate determination of mode I SIF in orthotropic composite materials. Their
research also demonstrated the existence of reasonable strain gauge locations.
In the last few years, digital image correlation (DIC) techniques have become increasingly popular due
to their simplicity and effectiveness. Pataky et al. (2012) developed an anisotropic least-squares
regression algorithm using displacements from DIC to find the effective SIFs, including KI, KII and Tstress. They studied the effects of anisotropy during mixed mode fatigue crack growth in single crystal
316L stainless steel, and determined crack tip plastic zones using an anisotropic yield criterion. Strains in
the plastic zone obtained from DIC showed a dependence on the crystallography and load ratio. Crack tip
slip irreversibility was also measured and showed an increasing trend with increasing crack length. In
parallel, Zhang and He (2012) investigated the determination of the mixed-mode SIFs using the DIC
method. Experiments were performed on an edge fatigue cracked aluminium specimen and the full-field
displacements around the crack-tip region of the test sample were calculated using DIC, and then the SIFs
associated with unavoidable rigid-body displacement motion were calculated from the experimental data.
A coarse-fine searching method was developed to determine the crack tip location. This proposed
technique is validated by comparing it with the theoretical results of SIF, which show DIC to be a
practical and effective tool for full-field deformation and SIF measurement.
For small fatigue cracks (physical size of the order of 500 m), Lachambre et al. (2014) utilized X-ray
tomography to monitor the initiation and growth of a 3-D fatigue crack in a nodular cast iron sample, and
SIF values along the curved crack front were extracted from displacement fields by digital volume
correlation (DVC) technique. The obtained results of SIF were in agreement with analytical calculations and
showed variations: larger values of SIF were found close to the sample surface, which was attributed to a
higher level of closure compared to the bulk of the specimen. However, the presented method is limited by
the relatively large size of the nodules.
397
In addition, Shlyannikov and Zakharov (2014) investigated the fatigue crack growth behaviour in the
elastic-plastic stress fields; the distributions of the governing parameter of elastic-plastic stress fields, the Infactor, have been determined in various specimens. Biaxial and mixed mode fatigue crack propagation tests
have also been carried out on cruciform and compact tension shear specimens, and discrepancies in fatigue
crack growth rate have been observed in various multiaxial loading conditions. The experimental results
provided the opportunity to explore the suggestion that multiaxial fatigue crack propagation may be
governed more strongly by the plastic SIF than the magnitude of the elastic SIFs alone. Both the In-factor
and T-stress are used as the main parameters of the elastic-plastic SIF, which is sensitive to the in-plane and
out-of-plane constraint effects.
4.6.2
Fatigue
By using DIC and electron backscatter diffraction (EBSD) techniques, Carroll et al. (2013) presented the
sub-grain level full field plastic strain accumulation, near a growing fatigue crack, measurement results
for the first time. In these experiments, inhomogeneities were observed in the strain field behind the crack
tip in both macro- and micro-scales which suggests the necessity for using crystal plasticity models to
capture an accurate fatigue behaviour.
By following a different approach, Rethore et al. (2012) performed in situ fatigue experiments using a
tomograph. In order to extract displacement fields and crack geometry from images, digital volume
correlation was utilized. As part of post-processing, crack front positions and stress intensity factor ranges
were obtained which were used to extract local Paris law coefficients.
Temperature change can also be used as a good fatigue damage indicator. Wang et al. (2010)
developed a Quantitative Thermographic Method to assess the residual fatigue life and perform non-linear
stress measurement. The technique allows obtaining several fatigue parameters including fatigue limit, SN curve and residual fatigue life. Furthermore, in-depth analysis of stress concentration zones was done
by using lock-in thermography. Moreover, Amiri and Khonsari (2012) proposed a technique for the
estimation of remaining fatigue life (RFL) by using the slope of the temperature rise as a function of time.
By using the thermography technique, their results showed that as the materials age, the slope of the
temperature increases.
For very high cycle regime conditions, Stanzl-Tschegg (2014) investigated the characterization of the
fatigue properties of materials and components at a very high cycle fatigue regime. It was concluded that
ultrasonic machines are the most suitable for characterization in these cases. The main advantages of
ultrasonic machines are time and energy saving, no need for water cooling equipment, low maintenance
costs and low noise. As an example, Heinz et al. (2013) performed ultrasonic fatigue tests for the very
high cycle fatigue regime up to 1010 cycles. Moreover, they used 3-Dimensional laser scanning
vibrometry to investigate the oscillation behaviour of the specimens in different fatigue states and to
perform non-contact strain measurements during ultrasonic fatigue loading. Furthermore, Mller and
Sander (2013) also used an ultrasonic fatigue testing system for a very high cycle regime to investigate
the effect of variable amplitude loading. The potential drop technique was also utilized to quantify crack
growth in conjunction with an optical microscope for surface crack measurement.
There are three different available techniques to measure slip during fretting fatigue experiments; clip
gauge, Wittkowsky and DIC. Clip gauge technique characteristically captures a significant amount of elastic
deformation of the surrounding material since it measures the slip remotely. This problem is eliminated in
the Wittkowsky technique. However, this approach suffers from calibration inaccuracies. The DIC
technique developed by De Pauw et al. (2014) allows measuring global slip closer to the contact region
while eliminating the effect of elastic deformations. Moreover, Meriaux et al. (2010) developed a dualactuator fatigue set-up which is capable of separating the fatigue and the fretting load. The test set-up
contains a potential drop technique device in order to accurately detect crack nucleation and propagation.
The technique allows detection of cracks as small as 50 m in size.
The analysis of experimental data is of paramount importance. Full-scale component fatigue tests are
complex and several effects should be accounted for in order to apply such results to fatigue assessment
procedures, for example: scale effects, fabrication imperfections and defects, environmental conditions,
loading conditions, etc. Notable examples are provided in (Kim et al., 2010; Rother and Rudolph, 2011;
Fricke et al., 2012; Li et al., 2013b; Fricke and Feltz, 2013). The Benchmarking study reported in this
document is also an example illustrating the complexities of component testing and interpretation of
results.
Validation of assessment approaches with experimental data need particular care and often data available
in literature are not complete and sufficient for the intended purpose. Often, component testing carried out in
398
past years does not distinguish clearly the crack initiation and/or the failure criteria making it rather difficult
to use such data for comparison of newly proposed methods.
4.6.3
Material properties
The principal material properties required for fatigue and fracture mechanics analysis include modulus of
elasticity, E, yield strength, y, tensile strength, u, constants C and m in Paris law, fracture toughness data
(KIc, J values, and crack tip opening displacement, CTOD), n, cyclic strain hardening exponent, b,
fatigue exponent, c, fatigue ductility exponent, K, cyclic strength coefficient, 'f , fatigue strength
coefficient, 'f , fatigue ductility coefficient, etc.
E, y, and u can be determined from the tests in terms of standards, e.g. (BS EN ISO 6892-1 Metallic
materials, 2009; BS EN ISO 6892-2 Metallic materials, 2011; ASTM Standard E8 / E8M13a, 2013;
ASTM Standard E111-04, 2010), etc., y is to be taken as either the lower yield or the 0.2% proof strength
depending on the material type. The constants C and m in Paris law may be determined from the tests
according to standards, e.g., ASTM Standard ASTM E64713a (2013), BS ISO 12108 Metallic materials
(2012), etc. The constants C and m also depend on the material and the applied conditions, including
environment and cyclic frequency. KIc, J values, and CTOD can be determined from the tests in accordance
with standards, e.g. ASME E399, BS EN ISO 15653 Metallic materials (2010), etc. n, b, c, K, 'f , and 'f
can be determined from the tests provided in standards, e.g. ASTM Standard E606 / E606M12 (2012), BS
7270 Metallic materials (2006), ISO 12106 (2003), etc.
4.6.4
Fracture toughness
Fracture toughness is considered to be one of the key input material properties for fatigue life assessment
based on fracture mechanics. In recent years, there have been some new developments on this topic.
Many researchers studied the fracture toughness under different temperatures and for different
components. For example, Ganesh et al. (2014) studied the initial fracture toughness (J0.2) and critical
crack tip opening displacement fracture toughness values of laser rapid manufactured structures. The
fracture toughness is usually obtained by using laboratory size specimens that meet certain requirements.
This represents a major challenge due to the pronounced effect of the specimen size on the mechanical
and fracture properties. Fatigue pre-cracking of the specimens, in order to obtain sharp crack, is one of the
standard procedure conditions to determine the plane strain fracture toughness, which is a costly and time
consuming process. Then some researchers tried to find the relationship between the fracture toughness
value and other material or structural properties. Pettin et al. (2014) presented a virtual testing tool with a
fracture toughness determination module to estimate fracture toughness vs. specimen thickness. It is
realized from transforming the simple tensile tests of the material. Mourad et al. (2013) investigated the
effect of specimen notch radius r and specimen size (thickness B and ligament b) on the apparent fracture
toughness KIC,app and verified the equations for predicting the apparent toughness by comparing with tests,
as shown in Figure 34. Two models have also been found in this research work to reasonably predict the
apparent fracture toughness KIC,app values, one of which can be expressed by:
(a)
(b)
Figure 34 . (a) The apparent critical fracture toughness KIC,app as a function of the ligament length b and
compared with experimental results (JIc vs. b for q = 0.05 mm); (b) Predicted KIC, app
compared with the experimental data.
K l,app = K lc,app + ( Pmax / B W ) e
399
( rp / )
f (a / W )
(9)
where B is the specimen thickness, W is the specimen width, Pmax is the maximum load and KlC,app and m
are two constants, f(a/W) is the geometry factor, rp is the plastic zone size. The two constants KIC,app and
m can be determined by testing two specimens with different .
5.
BENCHMARKING STUDY
Increasing needs to have higher payload-to-structural weight ratio have increased the need to study
lightweight, structural solutions in ships. This is especially important in ships where structural weight
is a significant part of the displacement, e.g. passenger ships. The previous committee Horn et al.
(2012a) carried out a benchmark investigation on the bulkhead structures of passenger ships and
focused on the fatigue damage assessment under multi-axial stresses. Another way to reduce weight is
to decrease the plate thickness of the passenger decks within the superstructure. However, this requires
design around the minimum plate thickness requirement of 5mm, set typically by the classification
societies. Thus, the starting point of this Benchmarking study is the discussion of the previous
committee work on plate thickness effect (Horn et al., 2012a). The issue was raised by the official
discusser, Bacicchi, who stated that recent experimental work has shown an increasing trend of fatigue
strength for decreasing plate thickness in the thickness range of t = 5mm to t = 20mm. The question
was commented upon by Professor Fricke who stated that at thicknesses around 5mm, the fatigue
strength actually shows a decreasing trend based on recent experimental findings. Since 2012, fatigue
tests have been carried out in the range from 3mm to 6mm (Fricke et al., 2013a; Fricke and Feltz, 2013;
Lilleme et al., 2012; Remes and Fricke, 2014) in the large EU-project BESST (Breakthrough in
European Ship and Shipbuilding Technologies) involving all the major European shipyards. Their
experimental evidence has been provided to complement this discussion, and is described below to
identify the limitations for the Benchmarking study.
Typically welding processes are developed by butt- and T-joint welding experiments to optimize the
welding parameters and weld geometry, and mechanical properties. A dog-bone specimen is commonly
used in fatigue tests during this process; see Figure 35. Lilleme et al. (2012) showed that both the
shape and magnitude of the welding-induced initial imperfections have a large impact on the fatigue
strength of the thin and slender butt-welded specimens. When the shape and geometrical non-linearity
were taken into account, the test specimen straightening was properly modelled. Their investigation
showed that in this case the fatigue strength defined, based on notch or structural stress, corresponds to
that of the thicker plates. They also showed that the relationship between structural and notch stress is
constant; thus the emphasis should be put on the structural stress and how it develops during
straightening. Fricke et al. (2013a) extended these findings for larger experimental sets of both butt and
fillet welds. They also noted that the slope issue needs further investigations as the normally used slope
of m = 3 results in non-conservative fatigue strength in comparison to design curves. The slope of m =
5 gave a better agreement between the test results and existing design curve. A slope of 5is also
recommended by Sonsino et al. (2010) for thin plates below t = 5mm. Remes and Fricke (2014), using
structural stress assessment and different stress extrapolation strategies showed that there is no
beneficial plate thickness effect if the butt and T-joints of 34 mm and 68 mm specimens are
compared. However, they also showed that the weld quality and especially the undercuts have a
significant impact on the strength of thin-plates, especially in the high-cycle domain. They further
showed that accounting for undercuts and carrying out the stress extrapolation through the thickness
and using geometrically non-linear analysis would result in smaller scatter in the predicted fatigue life
and proposed that the influence of undercuts be taken into account in FEM modelling emphasising the
role of specimen shape.
Based on these observations, this Committee decided to focus on the assessment of structural stresses in
thin slender specimens. Altogether, seven parties using ten different approaches contributed to the
Benchmarking study.
5.1
Problem Statement
A dog-bone specimen was selected for this Benchmarking study. The specimen represents the average
geometry of 10 specimens produced by arc welding (Lilleme et al., 2012). The specimen is curved and
the geometry is shown in Figure 35. It has a length of L = 250 mm, width varies from B = 50 mm at the
ends to B = 20 mm at the middle. The radius of R = 160 mm defines the dog bone shape where the curved
shape is also given in terms of coordinates, weld details and load and boundary conditions.
400

H=1.87mm
t=3.11mm
W=11.39mm
=23.14o
=28.27 o
=22.66o
e=-0.03mm
=24.87o
G=-0.21o
W=6.91mm
H=1.03mm
50, -1. 61
Load and support
-50, -1.63
Load and support
-30, -1.65
G=L
30,
-1. 64
-70, -1.68
70, -1. 67
-10, -1.76
-85,
-1.76
L=170mm
-6.02, -1.76
-3.81, -1.77
10, -1.76
85, -1.76
6.02, -1.80
3.81, -1.82
Figure 35. The dog-bone specimen.

This kind of specimen is typically assessed using stress-based fatigue assessment methods. The peak
stress at the weld notch, notch; see Equation 9, is needed for fatigue assessment. This can be obtained
using a notch stress approach where the shape of the specimen and the weld is modelled explicitly.
However, the approach suffers from considerable modelling efforts, thus structural stress, HS, and
nominal stress, nom, can also be used as input for the fatigue assessment. In the case of the structural
stress method, the notch stress can be approximated using calculated structural stress, HS, and notch
factor, Kw. In the case of the nominal stress method, the same is obtained by replacing, structural stress
with the product of nominal stress, nom, by the stress magnification, km; i.e.,
(10)
An objective of the Benchmarking Study is to understand the range of predicted km values when
different analysis approaches are used. Special emphasis is put on the influence of geometrical nonlinearity.
The specimen is exposed to tensile loading with a load ratio of R = 0 and a nominal stress level of nom =
30200MPa. Both ends of the specimen are free to rotate around the y-axis, i.e., the secondary bending
effect due to clamping is neglected. The distance between the load introduction points is 170 mm while the
entire length is 250 mm. During the fatigue test, the specimen straightens. The material is steel with E =
206GPa, y = 355MPa and = 0.3.
5.2
Analytical Methods
The stress concentration factors were obtained according to the following codes.
DNV (2003) Classification Note No. 30.7 Fatigue Assessment of Ship Structures February 2003,
taking into account the effects of axial and angular misalignment, multiplying these and using the
initial, undeformed shape to define km,
RINA Rules (2014) for the classification of ships, 2014 edition, taking into account the effects of
axial misalignment only and using the initial, undeformed shape to define km,
IIW (2013b) Fatigue Recommendations IIW-doc. XIII-2460-13/XV-1440-13 (rev. April 2013)
taking into account the effects of axial and angular misalignment and using the deformed shape to
define km.
It is worth noting that IIW recommendations account for the reduction of angular misalignment due to
acting membrane stress. RINA Rules are not specifically suited for the assessment of butt joints of tensioned
plates and even less of very thin plates rather the focus is on ship structural details.

DNV
RINA
IIW
401
402
Figure 36. Assumptions of the various analytical approaches by DNV, RINA and IIW.
5.3
Numerical analysis using FEM
As indicated by Fricke et al. (2013a) and Remes and Fricke (2014), the weld shape and straightening are
important factors when defining the km-factor in thin-plates, and therefore, they need to be modelled
properly. An accurate scheme is to use fine 3D-solid elements to model the weld and specimen shape
based on geometrical measurements see Figure 37 (A). However, this scheme is very time consuming and
is seldom used in practice. Often, simplifications are made in modelling of geometry.
As determining the straightening km value is complex, simplifications can be done by: excluding the dogbone or curved shape and using plane strain or plane stress assumption (Figure 37 (B)); excluding the weld
and considering only the curved specimen shape (Figure 37 (C)) or simplifying the weld shape and using
shell elements (Figure 37 (D)).
A)
B)
C)
D)
Figure 37. Various FE-modelling schemes. A) Full 3D-solid, B) 2D-plane strain or plane stress, C)
Simplified 3D-solid and D) Shell element mesh.
403
The analyses utilised here employed various approaches to extract km values from the solid element
results. Basically, the total stress concentration factor, K, was derived with FEA. The fixed Kw values were
obtained from the rules. The km values were then calculated according to Eq. (9). The results are given in
Italic font in Table 3. The kw values vary from 1.46 to 1.50 (Anthes et al., 1993; DNV, 2003). Similarly, in
three cases, the km factor is obtained by linear surface interpolation at x=0.4t and x=1.0t or x=0.5t and
x=1.5t. This with the total stress concentration factor, K, gave in turn the Kw factor.
Table 3. Calculated notch stress values and fatigue parameters for different nominal stress levels. The kwvalues given in italic font are taken from rules. The element length is denoted by Le and plate
thickness by t.
5.4
Analysis
Code
Mesh type
Geometry
Analytical
DNV CN30.7 -
Analytical
IIW, XIII2460-13/XV1440-13
Analytical
RINA Rules
2014 Pt. B,
Sect.4 Para
3.3.
FEM
DNV CN30.7 A) 8-node Hex,

Fully modeled
L e=t/4
& Sesam
FEM
Ansys
A) 8-node Hex,
Half modeled
L e=0.01mm,
radius 0.05mm
FEM
Abaqus
B) 8-node plane Transverse section

strain,
only
L e =0.07mm
FEM
DNV CN30.7 D) 4-node Quad Fully modeled,

& Sesam
L e =t/2
t weld=2t
FEM
Ansys
C) 20-node
Hex, L e =t
Plate strip, weld

excluded
FEM
Ansys
A) 20-node
Hex, L e =t/5
Fully modeled
FEM
Abaqus
B) 8-node plane Transverse section

stress,
only
L e=0.01mm
Straight plates,
separately angular
and axial
misalignment
Straight plates,
separately angular
and axial
misalignment
Straight plates,
axial
misalignment
Kw-values
Km -values
Non-linear
Kw-calculation Nominal Linear Non-linear Linear
Stress
30
1,46
Rules
1,41
100
200
1,26
30
1,03
1,17
Rules
1,60
100
1,13
200
30
2,33
Rules
1,03
100
200
30
1,13
Rules
1,50
100
200
30
FEM
1,15
100
200
30
1,64
1,10
FEM
1,44
1,30
100
1,62
1,14
200
1,62
1,15
30
Rules
1,10
1,50
100
200
30
1,16
Anthes et al.
1,11
1,46
100
1,20
(1993)
200
1,20
30
1,08
1,09
Not computed
100
1,07
200
30
1.63
1,19
FEM
100
1.64
1,15
200
1.65
1,13
Results
As Table 3 indicates, the linear 3D-solid element models without any geometrical modifications gives km
= 1.091.15 using the surface extrapolation; this means a deviation of 5.5%. The difference is due to the
modelling of the weld toe radius, the smaller radius causing a higher km. When only plate strip is
modelled (without modelling of the weld), the km value falls to this range with km=1.11. When Kw is taken
from the rules and km is derived from the total stress concentration value using Eq. (9) the km = 1.101.13.
So the values are again within the range. The 2D-plane strain model gives km=1.30; the difference is due
to plane strain assumption. From the comparison between the linear 3D-solid results (Le = t/4) and shell
element results, with equivalent weld modelled with twice the base plate thickness, the results indicate
that the km values are, 1.10 and 1.13, respectively. Analytical results by IIW and RINA give the same, unconservative results of km=1.03 while DNV gives a much more conservative result of km=1.41. It should
be noted that different fatigue analysis methods use very different assumptions, and a direct comparison
of km may not be relevant. It should also be noted that the way the clamp length is determined has an
effect. For example, when s changes from 210mm to 170mm, see Figure 35, the calculated km reduces
from 1.41 to 1.34.
When the analysis is carried out with geometrical non-linearity, the km value decreases from km,30MPa=1.19
to km, 200MPa=1.13. Thus, the straightening causes 5.3% decrease in non-linear range km value when nominal
stress is increased from 30MPa to 200MPa. In the case that the curved shape is not modelled, the weld is not
404
considered to be in the middle of the specimen and/or weld toe extrapolation is used for km while failure
occurs on the root, the km values can have an unphysical increasing trend; see values km,30MPa = 1.16 to km,
200MPa = 1.20 and km,30MPa = 1.10 to km, 200MPa = 1.15. The analytical equation of IIW also predicts a decreasing
trend; however, the decrease is stronger than in FE-analyses. It should be noted that this equation does not
account for a curved shape.
5.5
Discussion & Benchmarking Study Conclusions
The comparison of the FE-results shows that the assumption made in modelling can cause relatively
moderate deviation of the predicted stress state from that observed in the specimen. Therefore, special
attention should be put on the accuracy of the modelling of the geometry, discretization and analysis type.
Even for a single, simple dog-bone geometry, the different stress assessment methods show significant
scatter. It should also be noted that this specimen had relatively moderate deformations and in practice
higher amplitudes and different shapes can occur within specimen series, causing even higher scatter.
The highest and lowest km values observed were 1.41 and 1.03, respectively. Both were observed in
analytical methods. None of the analytical methods were able to model the effects of curved shape properly
thus, there is a need to update the design equations to account for the shapes observed in thin specimens. In
the case of solid elements in 3D or 2D, the problem becomes far more complicated, depending on the
method by which the km value is being determined. In some cases the km value is determined from the total
stress concentration factor using rule value for weld factor, Kw. The values used in this investigation ranged
from Kw=1.46 (Anthes et al., 1993) to Kw=1.5 given for example by the DNV (2003) rules (CN30.7).
However, as noted by IIW, Maddox and Fricke (2008) both the structural hot spot stress and the effective
notch stress at a weld toe need to be checked. A minimum value of Kw=1.6 is recommended for welds with
high weld toe radius for design purposes. The non-linear analyses based on plane stress and plane strain
values indicate that the Kw value is 1.62-1.65. Linear analysis gives lower Kw=1.44 and thus higher km=1.30.
Thus, it is recommended to follow the IIW proposal.
In practical analysis of ship structures, the curvature occurs in two directions. This means that the stresses
are being redistributed within the panels, and that the stiffeners carry more loading (Eggert et al., 2012;
Fricke and Feltz, 2013; Lilleme et al., 2013, 2014b). As Eggert et al. (2012) pointed out, the consequence is
that the fatigue damage can also occur along the stiffeners that are parallel to loading direction. Lilleme
et al. (2013) in turn stated that the curvature in two directions reduces the influence of straightening
considerably. This can lead to false conclusions on the dog-bone results on a panel level as there are
compensating effects; see also Fricke and Feltz (2013). When decks are considerably distorted, the
redistribution of stresses also happens between various decks of the superstructure. The study did not
consider variation of the specimen geometry. As Fricke et al. (2013a) and Remes and Fricke (2014) show,
this causes additional scatter, which needs to be incorporated into the analysis of test results. Considering
the fact that the slope of the S-N curve has been shown to vary between m=3 and m=5, this means that the
sensitivity of modelling the specimens is of utmost important. Considering these facts, one recommendation
is that large-scale tests should be carried out where the surrounding structure is included to consider
properly the stress redistribution, initial imperfections and residual stresses. This would allow conclusions
on what kind of impact these shapes have on the panel and, further, to the entire hull.
6.
SUMMARY & CONCLUSIONS
This report is the latest update in the long series of ISSC Fatigue & Fracture Technical Committee reports
and as highlighted earlier is not meant to be a stand-alone report rather than to build on the accumulated
knowledge reported previously. However, we have structured the report in a manner to make it accessible
to the relatively new researcher and designer. We therefore included Section 2 which summarised
approaches and methodologies to ship and offshore structure design, setting out some basic terminology
and characteristics of most structures in the marine environment e.g. damage tolerance which may not
necessarily be familiar to those accustomed to other application areas. Readers will have appreciated the
emphasis on uncertainty and reliability,y including inspection and maintenance through the life-cycle.
Marine structures are subject to stochastic loading and corrosion processes meaning that an absolute
determination of remaining life is always impossible but offshore and marine engineers use
probabilistic models coupled with safety factors to ensure acceptable confidence in calculated levels of
structural integrity.
Section 3 considered factors influencing fatigue and fracture categorised under Resistance, Materials,
Loading and Complex Stresses including multiaxial aspects. The section then developed the concepts of
Structural Integrity and Life-cycle management addressing fabrication and repair and Inspection and
Maintenance. It was seen that significant activity has taken place during the review period, probably the
405
greatest emphasis has been the urgent need for more component testing results to better validate models.
Section 4 presented a brief overview of the recent developments in fatigue and fracture assessment methods,
as well as an update of rules, standards & guidelines. Measurement techniques were also reviewed and the
latest experimental results available concerning fatigue, crack growth resistance and material properties
were presented and reviewed.
The main conclusion from benchmark study is that as we move to very thin decks the work by the
committee on ultimate strength analysis of ships starts to approach the work of this committee. The main
reason is that the analysis requires large deflection plate theory, but instead of looking at the average loadend-shortening behaviour, we have to look at details of the stress distribution. This is theory. In practise the
shape is far more complex that the theory is capable to model properly. Thus, we need full-scale data in
terms of actual measured shapes and residual stresses of the thin deck structures. Such data should be
exposed to new benchmarking exercise so that we can extend this to practical application. The Committees
benchmarking study reveals moderate deviations in predicted stress state as a result of modelling techniques
and assumptions.
Overall the report highlights a highly active subject area which has developed significantly to address
growing trends in the industry for weight saving and cost efficiencies during fabrication and maintenance.
These drivers demand more of materials and our understanding of them in order to model accurately
progressive through-life failure mechanisms.
REFERENCES
Abdullah, N. N., Hafezi, M. H. & Abdullah, S. 2012. High-Low Sequence Loading Effect on the Crack Growth Rate
using UniGrow Model. In: Oechsner, A., Da Silva, L. F. M. & Altenbach, H. (eds.) Materials with Complex
Behaviour II. Springer Berlin Heidelberg, 647660.
ABS 2009. Application of Higher-Strength Hull Structural Thick Steel Plates in Container Carriers. American
Bureau of Shipping.
ABS 2012. Application of Higher-Strength Hull Structural Thick Steel Plates in Container Carriers. American
Bureau of Shipping.
ABS 2013. Fatigue assessment of offshore structures. American Bureau of Shipping.
Agerskov, H. 2000. Fatigue in steel structures under random loading. Journal of Constructional Steel Research, 53,
283305.
Aihara, S., Shibanuma, K. & Namegawa, T. 2014. Numerical Simulation of Brittle Crack Propagation and Arrest in
Steels Considering Shear-Lip Formation. ISOPE, Volume 4
Akselsen, O. M. & stby, . 2014. Evaluation of Welding Consumables for Application down to 60C. ISOPE,
Volume 4.
Alvaro, A., Akselsen, O. M., Ren, X. & Kane, A. Fundamental Aspects of Fatigue of Steel in Arctic Applications.
Proceedings of the Twenty-fourth International Ocean and Polar Engineering Conference, June 2014 Busan,
Korea. 1520.
Amiri, M. & Khonsari, M. M. 2012. Nondestructive estimation of remaining fatigue life: Thermography technique.
Journal of Failure Analysis and Prevention, 12, 683688.
An, G. B., Woo, W. & Park, J.-U. 2014. Brittle crack-arrest fracture toughness in a high heat-input thick steel weld.
International Journal of Fracture, 185, 179185.
Andersen, I. M. & Jensen, J. J. Hull girder fatigue damage estimations of a large container vessel by spectral
analysis. PRADS, 2013. 557565.
Andreikiv, O. E. & Hembara, O. V. 2013. Influence of Soil Corrosion and Transported Products on the Service Life
of Welded Joints of Oil and Gas Pipelines. Materials Science, 49, 5258.
Anes, V., Reis, L., Li, B., Fonte, M. & De Freitas, M. 2013. New approach for analysis of complex multiaxial
loading paths. International Journal of Fatigue, 62, 2133.
Angelo, L. D. & Nussbaumer, A. 2013. Reconsideration of standard S-N curves for welded components using
maximum likelihood based fatigue models with and without fatigue limit. International Institute of Welding,
WG1163.
Anthes, R. J., Kttgen, V. B. & Seeger, T. 1993. Stress Concentration Factors for Butt and Cruciform Joint.
Schweien und Schneiden, 45, 685688.
API RP 2t 2010. Recommended practice for planning, designing, and constructing tension leg platforms. The
American Petroleum Institute.
API RP 5791/ASME FFS-1 2007. Fitness-for-service. The American Society of Mechanical Engineers and the
American Petroleum.
Arajo, J. A., Dantas, A. P., Castro, F. C., Mamiya, E. N. & Ferreira, J. L. A. 2011. On the characterization of the
critical plane with a simple and fast alternative measure of the shear stress amplitude in multiaxial fatigue.
International Journal of Fatigue, 33, 10921100.
ASTM Standard ASTM E64713a 2013. Standard test method for measurement of fatigue crack growth rates. West
Conshohocken, USA: ASTM International, DOI: 10.1520/E0647.
ASTM Standard E8 / E8m13a 2013. Standard Test Methods for Tension Testing of Metallic Materials. West
Conshohocken, USA: ASTM International, DOI: 10.1520/E0008_E0008M.
406
ASTM Standard E11104 2010. Standard test method for Young's modulus, tangent modulus, and chord modulus.
DOI: 10.1520/E011104R10.
ASTM Standard E606 / E606m12 2012. Standard test method for strain-controlled fatigue testing. West
Conshohocken, USA: ASTM International, DOI: 10.1520/E0606_E0606M-12.
Azevedo, J., Infante, V., Quintino, L. & Dos Santos, J. 2014. Fatigue Behavior of Friction Stir Welded Steel Joints.
Advanced Materials Research, 891892, 14881493.
Barbier, E. & Petrinic, N. 2013. Multiple Crack Growth and Coalescence in Meshfree Methods with Adistance
Function-Based Enriched Kernel. Key Engineering Materials, 560, 3760.
Bardetsky, A. 2013. Fracture mechanics approach to assess the progressive structural failure of a damaged ship.
Collision and Grounding of Ships and Offshore Structures, 77.
Bayley, C. & Aucoin, N. 2013. Fracture testing of welded single edge notch tensile specimens. Engineering
Fracture Mechanics, 102, 257270.
Bertoglio, C., Gaggero, S., Rizzo, C. M. & Viviani, M. Fatigue strength assessment of propellers by means of
weakly coupled CFD and FEM analyses. OMAE, Proc.s of the 33rd Int. Conf. on Ocean, Offshore and Arctic
Engineering, June 2014 San Francisco: CA. 24655.
Bertoglio, C., Gaggero, S., Rizzo, C. M., Viviani, M., Vaccaro, C. & Conti, F. Improvements in fatigue strength
assessment of marine propellers. MARSTRUCT, Proc.s of the 5th Int. Conf. on Marine Structures, March 2427
2015 Southampton: UK.
Brighenti, R., Carpinteri, A., Spagnoli, A. & Scorza, D. 2012. Crack path dependence on inhomogeneities of
material microstructure. Frattura e Integrita Strutturale, 20.
BS 7270 Metallic Materials 2006. Constant amplitude strain controlled axial fatigue. Method of test. London:
British Standard Institute.
BS 7910 2013. Guide to methods for assessing the acceptability of flaws in metallic structures. London: British
Standard Institute.
BS EN ISO 68921 Metallic Materials 2009. Tensile testing. Method of test at ambient temperature. London:
BS EN ISO 68922 Metallic Materials 2011. Tensile testing. Method of test at elevated temperature. London:
BS EN ISO 15653 Metallic Materials 2010. Method of test for the determination of quasistatic fracture toughness of
welds London: British Standard Institute.
BS ISO 12108 Metallic Materials 2012. Fatigue testing. Fatigue crack growth method. London: British Standard
Institute.
BV NR 483 2011. RNS_PartB Rules for the Classification of Naval Ships. Bureau Veritas.
BV NR 578 2012. Rules for the Classification of Tension Leg Platforms. Bureau Veritas.
BV NR 583 2012. Whipping and Springing Assessment. Bureau Veritas.
BV NR 593 2012. Ship Conversion into Surface Offshore Units and REdeployment of Surface Offshore Units.
Bureau Veritas.
Cmara, M. C. & Cyrino, J. C. R. Structural Reliability Applications in Design and Maintenance Planning of Ships
Subjected to Fatigue and Corrosion. ASME 2012 31st International Conference on Ocean, Offshore and Arctic
Engineering, 2012. American Society of Mechanical Engineers, 503514.
Carroll, J. D., Abuzaid, W., Lambros, J. & Sehitoglu, H. 2013. High resolution digital image correlation
measurements of strain accumulation in fatigue crack growth. International Journal of Fatigue, 57, 140150.
Chakraborty, D. & Murthy, K. S. R. K. 2014. A new single strain gage technique for the accurate determination of
mode I stress intensity factor in orthotropic composite materials. Engineering Fracture Mechanics, 124125,
142154.
Chao, M., Haiwei, L., Ning, Z. & Yang, S. 2013. A Ship Cargo Hold Inspection Approach Using Laser Vision
Systems. TELKOMNIKA Indones. J. Electr. Eng, 11, 330337.
Chen, F., Wang, F. & Cui, W. 2012. Fatigue life prediction of engineering structures subjected to variable amplitude
loading using the improved crack growth rate model. Fatigue & Fracture of Engineering Materials & Structures,
35, 278290.
Chen, G. H. & Chen, L. J. 2012. Solution-dependent path crack propagation simulation and its application. Pressure
Vessel Technology, 29, 3640, (in Chinese).
Chen, N. Z. & Wang, G. Low-cycle fatigue and a rational method for determining the size of a crack-stopping hole.
Proceedings of the First International Conference on Performance-based and Life-cycle Structural Engineering,
2012 Hong Kong, China. 18951903.
Choudhary, B. K., Samuel, E. I., Sainath, G., Christopher, J. & Mathew, M. D. 2013. Influence of Temperature and
Strain Rate on Tensile Deformation and Fracture Behaviour of P92 Ferritic Steel. Metallurgical and Materials
Transactions A, 44, 49794992.
Choung, J., Nam, W., Noh, M.-H., Lee, J.-Y. & Shim, S. 2014. Impact Bending Tests and Simulations of an High
Strength Steel FH32 for Arctic Marine Structures. ISOPE.
Coronado, P., Argelles, A., Via, J., Molln, V. & Via, I. 2012. Influence of Temperature on a Carbon-Fibre
Epoxy Composite Subjected to Static and Fatigue Loading Under Mode-I Delamination. International Journal of
Solids and Structures, 49, 29342940.
Correia, J. a. F. O., De Jesus, A. M. P. & Fernndez-Canteli, A. 2013. Local unified probabilistic model for fatigue
crack initiation and propagation: application to a notched geometry. Engineering Structures, 52, 394407.
407
Cosso, G. L., Rizzo, C. M. & Servetto, C. 2014. Fitness for service assessment of defected welded structural details
by experimental evaluation of the fatigue resistance S-N curve. IIW doc. XIII-2536-14. Paris: International
Institute of Welding.
Dai, P., Li, S. & Li, Z. 2013. The effects of overload on the fatigue crack growth in ductile materials predicted by
plasticity-corrected stress intensity factor. Engineering Fracture Mechanics, 111, 2637.
De Baere, K., Verstraelen, H., Rigo, P., Van Passel, S., Lenaerts, S. & Potters, G. 2013. Study on alternative
approaches to corrosion protection of ballast tanks using an economic model. Mar. Struct., 32, 117.
De Pauw, J., De Waele, W., Hojjati-Talemi, R. & De Baets, P. 2014. On the use of digital image correlation for slip
measurement during coupon scale fretting fatigue experiments. International Journal of Solids and Structures,
51, 30583066.
Derbanne, F. R., De Hauteclocque, G. & Chen, X. B. Evaluation of Rule-Based Fatigue Design Loads Associated at
a New Probability Level. Proceedings of the Twenty-first International Offshore and Polar Engineering
Conference, 2011 Maui, Hawaii, USA. 1924.
DNV 2003. Classification Notes, No. 30. Fatigue assessment of ship structures. 7.
DNV Gp Rp 0005 2014. Fatigue Design of Offshore Steel Structures. Det Norske Veritas.
DNV Os-C102 2012. Structural Design of Offshore Ships. Det Norske Veritas.
DNV Rp-C206 2012. Fatigue Methodology of Offshore Ships. Det Norske Veritas.
Doerk, O., Fricke, W. & Von Selle, H. Validation of Different Fatigue Assessment Approaches for Thick Plate
Structures Made of High Tensile Strength Steel YP47. Proceedings of the Twenty-second International Offshore
and Polar Engineering Conference, 2012 Rhodes, Greece.
Dong, P., Pei, X., Xing, S. & Kim, M. H. 2014. A structural strain method for low-cycle fatigue evaluation of
welded components. Int. J. of Pressure Vessels and Piping, 119, 3951.
Doring, R., Hoffmeyer, J., Seeger, T. & Vormwald, M. 2006. Short fatigue crack growth under nonproportional
multiaxial elasticplastic strains. International Journal of Fatigue, 28, 972982.
Doshi, K. & Vhanmane, S. 2013. Probabilistic fracture mechanics based fatigue evaluation of ship structural details.
Eggert, L., Fricke, W. & Paetzhold, H. 2012. Fatigue strength of thin-plated block joints with typical shipbuilding
imperfections. Welding in the World, 56.
Fatemi, A. & Shamsaei, N. 2011. Multiaxial fatigue: An overview and some approximation models for life
estimation. International Journal of Fatigue, 33, 948958.
Feltz, O., Fischer, C. & Fricke, W. Fatigue assessment of weld toe and root cracks with the notch stress intensity
factor and crack propagation approach. Proc. 11th Int. Symp. on Practical Design of Ships and Other Floating
Structures (PRADS), 2010 Rio de Janeiro, Brazil.
Feng, G., Cao, J., Ren, H. & Li, H. The Fatigue Crack Propagation of the Ship Structures in Random Sea States.
ASME, 31st International Conference on Ocean, Offshore and Arctic Engineering, July 2012. American Society
of Mechanical Engineers, 3944.
Fischer, C., Dster, A. & Fricke, W. Different finite element refinement strategies for the computation of the strain
energy density in a welded joint. Proc. MARSTRUCT 2011 Congress, 2011a Hamburg, Germany.
Fischer, C., Feltz, O., Fricke, W. & Lazzarin, P. 2010. Application of the Notch stress intensity and crack
propagation approaches to weld toe and root fatigue. IIW-Doc. XIII-2337-10/XV-1354-10. Paris: International
Institute of Welding.
Fischer, C., Fricke, W., Meneghetti, G. & Rizzo, C. M. Fatigue strength assessment of HP stiffener joints with filletwelded attachments using the peak stress method. In: Brinkmann, B. & Wriggers, P., eds. Proc.s of the fifth
Conference on Computational Methods in Marine Engineering (MARINE 2013), 2013 Germany. 660669.
Fischer, C., Rizzo, C. M. & Fricke, W. 2011b. Fatigue assessment of hopper knuckles according to N-SIF based
approaches. In: Rizzuto & Guedes Soares, C. (eds.) Sustainable Maritime Transportation and Exploitation of Sea
Resources. London: Taylor & Francis Group.
Fischer, C., Rizzo, C. M. & Fricke, W. Fatigue Assessment of Joints at Bulb Profiles by Local Approaches.
MARSTRUCT Proc.s of the 5th Int. Conf. on Marine Structures, March 2427 2015 Southampton. UKFitnessfor-service, (2007) The American Society of Mechanical Engineers and the American Petroleum API RP 579
1/ASME FFS-1.
Fremy, F., Pommier, S., Poncelet, M., Raka, B., Galenne, E., Courtin, S. & Roux, J.-C. L. 2013. Load path effect on
fatigue crack propagation in I+II+III mixed mode conditions - Part 1: Experimental investigations. International
Journal of Fatigue.
Fricke, W. 2013. IIW guideline for the assessment of weld root fatigue. Welding in the World, 57, 753791.
Fricke, W. 2014. Recent developments and future challenges in fatigue strength assessment of welded joints. Proc
IMechE Part C : J Mechanical Engineering Science, DOI : 10.1177/0954406214550015.
Fricke, W., Bollero, A., Chirica, I., Garbatov, Y., Jancart, F., Kahl, A., Remes, H., Rizzo, C. M., Von Selle, H.,
Urban, A. & Wei, L. 2008. Round Robin Study on Structural Hot-Spot and Effective Notch Stress Analysis. Ship
and Offshore Structures 3, 335345.
Fricke, W., Codda, M., Feltz, O., Garbatov, Y., Remes, H., Risso, G., Rizzo, C. M. & Romanoff, J. 2013a. Round
robin study on local stress and fatigue assessment of lap joints and doubler plates. Ship and Offshore Structures,
8, 621627.
Fricke, W. & Feltz, O. 2013. Consideration of influence factors between small-scale specimens and large
components on the fatigue strength of thin-plated block joints in shipbuilding. Fatigue & Fracture of Engineering
Materials & Structures, 36, 12231231.
408
Fricke, W. & Paetzold, H. 2012. Experimental Investigation of the Effect of Whipping Stresses on the Fatigue Life
of Ships. Proc. Intl. Marine Design Conf. IMDC2012, 3, 3 10.
Fricke, W. & Paetzold, H. 2014. Effect of whipping stresses on the fatigue damage of ship structures. Weld World,
58, 261268.
Fricke, W., Remes, H., Feltz, O., Lilleme, I., Tchuindjang, D., Reinert, T., Nevierov, A., Sichermann, W.,
Brinkmann, M., Kontkanen, T., Bohlmann, B. & Molter, L. Fatigue strength of laser-welded thin plate ship
structures based on nominal and structural hot-spot stress approach. Analysis and Design of Marine Structures Proceedings of the 4th International Conference on Marine Structures, MARSTRUCT, 2013b. 249254.
Fricke, W., Von Lilienfeld-Toal, A. & Paetzold, H. 2012. Fatigue strength investigations of welded details of
stiffened plate structures in steel ships. International Journal of Fatigue, 34, 1726.
Fukasawa, T. & Mukai, K. 2014. On the effects of hull-girder vibration upon fatigue strength of a Post-Panamax
container ship disaggregated by short-term sea state. Int. J. Nav. Archit. Ocean Eng, 6, 431441.
Ganesh, P., Kaul, R., Sasikala, G., Kumar, H., Venugopal, S., Tiwari, P. & Kukreja, L. M. 2014. Fatigue Crack
Propagation and Fracture Toughness of Laser Rapid Manufactured Structures of AISI 316L Stainless Steel.
Metallography, Microstructure, and Analysis, 3, 3645.
Gates, N. & Fatemi, A. 2014. Notched fatigue behavior and stress analysis under multiaxial states of stress.
International Journal of Fatigue, 67, 214, http://dx.doi.org/10.1016/j.ijfatigue.
GL 2012. Guidelines for the Structural Design of Racing Yachts Greater than 24m. Hamburg: Germanischer Lloyd.
GL 2013. Guidelines to Assess High-Frequency Hull Girder Response of Containerships. Hamburg: Germanischer
Lloyd.
GL 2014. Rules for Classification and Construction, I Ship Technology, Part 1 Sea-Going Ships, Chapter 1Hull
Structures. Hamburg: Germanischer Lloyd.
Gladskyi, M. & Fatemi, A. 2013. Notched fatigue behavior including load sequence effects under axial and torsional
loadings. International Journal of Fatigue, 55, 4353.
Gotoh, K., Matsuda, K. & Kitamura, O. Numerical Simulation of Fatigue Crack Propagation under Superposed
Loading Histories with Two Different Frequencies. Proc. Hydroelasticity in Marine Technology, 2012. 287297.
Guo, J., Wang, G., Perakis, A. N. & Ivanov, L. 2012. A study on reliability-based inspection planning Application
to deck plate thickness measurement of aging tankers. Mar. Struct., 25, 85106.
Haagensen, P. & Maddox, S. 2013. IIW recommendations on methods for improving the fatigue lives of welded
joints. Cambridge: Woodhead Publishing Ltd. International Institute of Welding, Paris.
Haagensen, P. J., Larsen, J. E. & Vrdal, O. T. 2014. Long Term Effectiveness of Life Extension Methodologies
Applied to Offshore Structures. Procedia Materials Science, 3, 21872194.
Han, D. K., Jung, H. G., Kim, K. Y. & Kim, S. J. 2013. Effect of Heat Treatment on Hydrogen Diffusion and
Hydrogen Induced Cracking Behavior of Process Pipe Steel in Sour Environment. ISOPE, 4.
Harksen, S., Lischewski, I., Schneider, A., Schmidt, T. & Mahn, D. 2013. A Modern Methodology of Materials
Development for High Toughness Seamless Line Pipe Products for Offshore Applications. ISOPE, 4.
Hauge, M. & stby, E. Fracture Integrity of Arctic StructuresHow to Determine Characteristic Fracture Toughness
for Engineering Critical Assessment. The Twenty-third International Offshore and Polar Engineering Conference,
June 2013. International Society of Offshore and Polar Engineers.
Heinz, S., Balle, F., Wagner, G. & Eifler, D. 2013. Analysis of fatigue properties and failure mechanisms of
Ti6Al4V in the very high cycle fatigue regime using ultrasonic technology and 3D laser scanning vibrometry.
Ultrasonics, 53, 14331440.
Hobbacher, A. 2009. Recommendations for Fatigue Design of Welded Joints and Components. IIW doc, Welding
Research Council Bulletin 520, 182307.
Hollnder, D., Wnsche, M., Henkel, S. & Theilig, H. 2012. Numerical Simulation of Multiple Fatigue Crack
Growth with Additional Crack Initiation. Key Engineering Materials, 488, 444447.
Horn, A., Aihara, S., Andersen, M., Biot, M., Bohlmann, B., Van Der Cammen, J., Choi, B. K., Garbatov, Y.,
Mishra, B., Qian, X., Remes, H., Ringsberg, J., Samanta, A., Wang, D. & Zhang, S. Committee III.2 Fatigue and
Fracture. Proceedings of the 18th International Ships and Offshore Structures Congress, September 913 2012a
Rostock, Germany.
Horn, A. M., stby, E., Hauge, M. & Aubert, J. M. Robust material qualification for Arctic applications. Proc. 22nd
Intl Ocean & Polar Eng Conf, 2012b Greece. ISOPE, volume 4, available at www.isope.org.
Hover, F. S., Eustice, R. M., Kim, A., Englot, B., Johannsson, H., Kaess, M. & Leonard, J. J. 2012. Advanced
perception, navigation and planning for autonomous in-water ship hull inspection. Int. J. Rob. Res, 31, 1445
1464.
Huang, W. 2013. The Rain-Flow Estimate of Fatigue Damage of Combined High and Low Frequency Load Effects.
ASME, 32nd International Conference on Ocean, Offshore and Arctic Engineering (OMAE2013). American
Society of Mechanical Engineers.
IACS 2013. Requirements for Use of Extremely Thick Steel Plates.
IACS 2014. Technical Background for Equivalent Design Waves (EDW) for Fatigue Loads.
IIW 2013a. Draft for Development: IIW guidelines on weld quality in relationship to fatigue strength, IIW-doc.
XIII-2301r313.
IIW 2013b. Fatigue Recommendations, IIW-doc. XIII-246013/XV-144013 rev. (April 2013).
IIW 2013c. Recommendation for Fatigue Design of Welded Joints and Components, IIW-doc. -246013.
Ince, A. & Glinka, G. 2014. A generalized fatigue damage parameter for multiaxial fatigue life prediction under
proportional and non-proportional loadings. Int. J. of Fatigue, 62, 3441.
409
Iranpour, M. & Taheri, F. 2013. Influence of the Peak Tensile Overload Cycles on the Fatigue Crack Growth of
Aluminum Alloy Under Spectrum Loading. Journal of Materials Engineering and Performance, 22, 34903499.
Iranpour, M. & Taheri, F. 2014. Analytical and computational investigation into the influence of the compressive
stress cycles on crack growth under variable amplitude loading using CTOD. Fatigue & Fracture of Engineering
Materials & Structures, 37, 645658.
ISO 12106 2003. Metallic materials- Fatigue testing. Strain control axial fatigue method. International Organization
for Standardization.
ISO 121101:2013 2013. Metallic materials -- Fatigue testing -- Variable amplitude fatigue testing -- General
principles, test method and reporting requirements. International Organization for Standardization.
ISO 121102:2013 2013. Metallic materials --Fatigue testing -- Variable amplitude fatigue testing -- Cycle counting
and related data reduction methods. International Organization for Standardization.
ISO/TR 14345:2012 2012. Fatigue --Fatigue testing of welded components --Guidance. International Organization
for Standardization.
Itoh, T., Sakane, M. & Ohsuga, K. 2013. Multiaxial low cycle fatigue life under non-proportional loading.
International Journal of Pressure Vessels and Piping, 110, 5056.
Itoh, T. & Yang, T. 2011. Material dependence of multiaxial low cycle fatigue lives under non-proportional loading.
Johannesson, P., Svensson, T. & Mar, J. D. 2005. Fatigue life prediction based on variable amplitude tests
methodology. International Journal of Fatigue, 27, 954965.
Jonsson, B., Dobmann, G., Hobbacher, A., Kassner, M. & Marquis, G. 2013. IIW-XIII-251013 IIW Guidelines on
weld quality in relationship to fatigue strength.
Judt, P. & Ricoeur, A. 2013. Efficient simulation of crack growth in multiple-crack systems considering internal
boundaries and interfaces. In ICF13.
Karakas, O. 2013. Consideration of mean-stress effects on fatigue life of welded magnesium joints by the
application of the Smith-Watson-Topper and reference radius concepts. International Journal of Fatigue, 49, 1
17.
Kim, J. G., Hwang, Y. J., Yoon, S. H. & Lee, D. G. 2012a. Improvement of the fracture toughness of adhesively
bonded stainless steel joints with aramid fibers at cryogenic temperatures. Composite Structures, 94, 29822989.
Kim, J. H., Chau Dinh, T., Zi, G. & Kong, J. S. 2012b. The effect of compression stresses, stress level and stress
order on fatigue crack growth of multiple site damage. Fatigue & Fracture of Engineering Materials &
Kim, M. H., Kang, S. W., Kim, J. H., Kim, K. W., Kang, J. K. & Heo, J. H. 2010. An experimental study on the
fatigue strength assessment of longi-web connections in ship structures using structural stress. International
Journal of Fatigue, 32, 318329.
Kim, Y. W., Lee, J. M., Kim, M. H., Noh, B. J., Sung, H. J., Ando, R. & Matsumoto, T. An Experimental Study for
Fatigue Performance of 7% Nickel Steels for Type B LNG Carriers. ASME 2014 33rd International Conference
on Ocean, Offshore and Arctic Engineering, 2014. American Society of Mechanical Engineers.
Kirkhope, K. J., Bell, R., Caron, L., Basu, R. I. & Ma, K.-T. 1999a. Weld detail fatigue life improvement
techniques. Part 1: review. Marine Structures, 12, 447474.
Kirkhope, K. J., Bell, R., Caron, L., Basu, R. I. & Ma, K.-T. 1999b. Weld detail fatigue life improvement
techniques. Part 2: application to ship structures. Marine Structures, 12, 477496.
Kder, T. 2015. On the Fatigue Strength of Thin-Plated Ship Structures. Ph.D. Thesis, University of Southern
Denmark.
Lachambre, J., Rthor, J., Weck, A. & Buffiere, J. Y. 2014. Extraction of stress intensity factors for 3D small
fatigue cracks using digital volume correlation and X-ray tomography. International Journal of Fatigue, 71, 3
10.
Lee, J.-M., Lee, K.-H. & Kim, D.-S. 2013. Cloud-Based RF-Inspection for Ship Maintenance. Int. J. Distrib. Sens.
Networks, 2013, 18.
Leng, S. W., Miao, Z. M., Qiu, F. X., Niu, L. N. & Miao, T. 2012. Analysis of the Relationship between CTOD
Toughness and Micromechanism of Marine Steel Weld Joints. Applied Mechanics of Materials, 117119, 1867
1873.
Li, B., Chen, X. & He, Z. 2013a. Three-steps-meshing based multiple crack identification for structures and its
experimental studies. Chinese Journal of Mechanical Engineering, 26, 400405.
Li, J., Zhang, Z. P., Sun, Q. & Li, C. W. 2011. Multiaxial fatigue life prediction for various metallic materials based
on the critical plane approach. International Journal of Fatigue, 33, 90101.
Li, K., Chen, M. & Lin, Y. 2012a. Research on marine structure inspection support system on mobile device
platform. J. Shanghai Jiaotong Univ, 17, 7075.
Li, X. D., Wang, X. S., Ren, H. H., Chen, Y. L. & Mu, Z. T. 2012b. Effect of Prior Corrosion State on the Fatigue
Small Cracking Behavior of 6151T6 Aluminum Alloy. Corrosion Science, 55, 2633.
Li, Z., Ringsberg, J. W. & Storhaug, G. 2013b. Time-domain fatigue assessment of ship side-shell structures.
Lieser, M. J. & Xu, J. 2010. Composites and the Future of Society: Preventing a Legacy of Costly Corrosion with
Modern
Materials.
[Accessed
26th
March
2014].
Available
from:
http://www.ocvreinforcements.com/pdf/library/CSB_Corrosion_White_Paper_Sept_10_10_English.pdf.
Lilleme, I. 2014. Influence of Initial Distortion on the Fatigue Assessment of Thin Ship Deck Structure. Ph.D.
Thesis, Aalto University.
410
Lilleme, I., Lammi, H., Molter, L. & Remes, H. 2012. Fatigue strength of welded butt joints in thin and slender
specimens. International Journal of Fatigue, 44, 98106.
Lilleme, I., Remes, H. & Romanoff, J. 2013. Influence of initial distortion on the structural stress in 3 mm thick
stiffened panels. Thin-Walled Structures, 72, 121127.
Lilleme, I., Remes, H. & Romanoff, J. Fatigue assessment of large thin-walled structures with initial distortions.
Proceedings of the 11th International Fatigue Congress, March 27 2014a Melbourne, Australia. Presented on
Advanced Materials Research, 123129.
Lilleme, I., Remes, H. & Romanoff, J. 2014b. Influence of initial distortion of 3 mm thin superstructure decks on
hull girder response for fatigue assessment. Marine Structures, 37, 203218.
Liu, Z. F., Gu, L. X. & Xu, Z. Y. 2013. Fatigue Crack Propagation Prediction under Single Overload Variable
Loading. Applied Mechanics and Materials, 275, 215219.
Lorn, S. & Svensson, T. 2012. Second moment reliability evaluation vs. Monte Carlo simulations for weld fatigue
strength. Quality and Reliability Engineering International, 28, 887896.
Lu, Z. Z. & Liu, Y. M. 2010. Small time scale fatigue crack growth analysis. International Journal of Fatigue, 32,
13061321.
Maddox, S. J. & Fricke, W. 2008. Guideline for the Fatigue Assessment by Notch Stress Analysis for Welded
Structures, IIW-Doc. XIII-2240r108/XV-1289r108.
Mao, W. 2014. Development of a Spectral Method and a Statistical Wave Model for Crack Propagation Prediction
in Ship Structures. Journal of ship research, 58, 106116.
Mao, W., Li, Z., Ogeman, V. & Ringsberg, J. W. 2015. A Regression and Beam Theory based Approach for Fatigue
Assessment of Containership Structures including Bending and Torsion Contributions. Marine structures, in
press.
Mao, W. & Ringsberg, J. W. Analysis of Fatigue Crack Initiation and Propagation in Ship Structures. 2013
Proceedings of the Thirteenth International Conference on Fracture (ICF13). 10.
Mao, W., Rychlik, I. & Storhaug, G. 2010. Safety index of fatigue failure for ship structure details. Journal of Ship
Research, 54, 197208.
Marciniak, Z., Rozumek, D. & Macha, E. 2013. Verification of fatigue critical plane position according to variance
and damage accumulation methods under multiaxial loading. International Journal of Fatigue, 58, 8493.
Marquis, G. & Barsoum, Z. 2013. A guideline for fatigue strength improvement of high strength steel welded
structures using high frequency mechanical impact treatment. Procedia Engineering, 66, 98107.
Marquis, G., Mikkola, E., Yildirim, H. C. & Barsoum, Z. 2013. Fatigue Strength Improvement of Steel Structures
by HFMI: Proposed Fatigue Assessment Guidelines. Welding in the World, 57, 803822.
Matsubara, G. & Nishio, K. 2013. Multiaxial high-cycle fatigue criterion considering crack initiation and nonpropagation. International Journal of Fatigue, 47, 222231.
Matsuda, K. & Gotoh, K. 2013. Numerical Simulation of Fatigue Crack Propagation Under Simulated Whipping
Loading Arising in Hull Structures. OMAE201310985.
Meriaux, J., Fouvry, S., Kubiak, K. J. & Deyber, S. 2010. Characterization of crack nucleation in TA6V under
fretting-fatigue loading using potential drop technique. International Journal of Fatigue, 32, 16581668.
Mikkola, E., Dore, M. & Khurshid, M. 2013. Fatigue strength of HFMI treated structures under high R-ratio and
variable amplitude loading. Procedia Engineering, 66, 161170.
Molter, L. 2013. Fatigue strength of laser hybrid weld seams in ship construction. Ph.D. Thesis, University of
Southern Denmark.
Moon, D. H., Lee, J. S., Lee, J. M. & Kim, M. H. 2013. Estimation of Constraint Factor on the Relationship
Between J Integral and CTOD for Offshore Structural Steel Weldments. Materials Technology; Ocean Space
Utilization, 3, V003T03A023.
Moore, P. & Nicholas, J. The Effect of Inclusions on the Fracture Toughness of Local Brittle Zones in the HAZ of
Girth Welded Line Pipe. OMAE, 2013. 10729.
Mourad, A. H. I., El-Domiaty, A. & Chao, Y. J. 2013. Fracture toughness prediction of low alloy steel as a function
of specimen notch root radius and size constraints. Engineering Fracture Mechanics, 103, 7993.
Mukoyama, K., Hanaki, K., Okada, K., Sakaida, A., Sugeta, A., Nishikawa, I., Ueno, A. & Sakai, T. 2014. A Study
on Estimation of S-N curves for Structural Steels Based on Their Static Mechanical Properties. Advanced
Materials Research, 891892, 16391644.
Mller, T. & Sander, M. 2013. On the use of ultrasonic fatigue testing technique variable amplitude loadings and
crack growth monitoring. Ultrasonics, 53, 14171424.
Murakami, Y., Takeuchi, Y. & Kazukuni Hase, K. 2013. Development of YP460 Class Steel Plate with Excellent
Brittle Crack Arrestability for Large Container Ships. ISOPE, Volume 4.
Murawski, L., Ostachowicz, W., Opoka, S., Mieloszyk, M. & Majewska, K. 2012. Practical Application of
Monitoring System Based on Optical Sensors for Marine Constructions. Key Eng. Materials, 518, 261270.
Nakashima, K., Hase, K., Endo, S., Eto, T., Fukai, H. & Shiomi, H. Development of YP390MPa Steel Plate for
Shipbuilding with Superior Low Temperature Toughness for Large Heat Input Welding. The Twenty-fourth
International Ocean and Polar Engineering Conference, August 2014. International Society of Offshore and Polar
Engineers.
Nash, I. 2013. The Production and Testing of MEIDP Line-Pipe for 3500m Application. ISOPE Volume 4.
Neuber, H. 1968. ber die Bercksichtigung der Spannungskonzentration bei Festigkeitsberechnungen.
Konstruktion, 20, 245251.
411
Nielsen, U. D., Jensen, J., Pedersen, J., Preben, T. & Ito, Y. 2011. Onboard monitoring of fatigue damage rates in
the hull girder. Marine Structures, 24, 182206.
Noban, M., Jahed, H., Ibrahim, E. & Ince, A. 2012. Load path sensitivity and fatigue life estimation of
30CrNiMo8HH. International Journal of Fatigue, 37, 123133.
Oliveira, H. L. & Leonel, E. D. 2013. Dual BEM Formulation Applied to Analysis of Multiple Crack Propagation.
Key Engineering Materials, 560, 99106.
Osawa, N., Nakamura, T., Yamamoto, N. & Sawamura, J. Development of a new fatigue testing machine for high
frequency fatigue damage assessment. ASME 2013 32nd International Conference on Ocean, Offshore and Arctic
Engineering (OMAE), 2013. American Society of Mechanical Engineers.
stby, E., Akselsen, O. M. & Kristensen, T. A. Monitoring of Cleavage Microcrack Arrest in Weld Thermal
Simulated Microstructures by Means of Acoustic Emission. Proceedings of the Twenty-third International
Offshore and Polar Engineering Conference, 2013a Anchorage, Alaska.
stby, E., Horn, A. M., Hauge, M. & Akselsen, O. M. Fracture Mechanics Design Criteria for Low Temperature
Applications of Steel Weldments. Proceedings of the twenty-third International Offshore and Polar Engineering
Conference, 2013b Anchorage, Alaska.
Pang, J. H. L. & Chew, Y. X. 2014. Fatigue Crack Growth and Coalescence Algorithm Starting from Multiple
Surface Cracks. Advanced Materials Research, 891, 10031008.
Paolino, D. S., Chiandussi, G. & Rosetto, M. 2013. A unified statistical model for S-N fatigue curves: probabilistic
definition. Fatigue and Fracture of Engineering Materials & Structures, 36, 187201.
Pataky, G. J., Sangid, M. D., Sehitoglu, H., Hamilton, R. F., Maier, H. J. & Sofronis, P. 2012. Full field
measurements of anisotropic stress intensity factor ranges in fatigue. Engineering Fracture Mechanics, 94, 13
28.
Pedersen, M. M., Andersen, J. G. & lafsson, . M. 2013. Investigation of the thickness effect for butt welded
joints. IIW, WG1, 15412.
Pettin, M., Farahmand, B., Berto, F. & Abdi, F. 2014. Virtual testing for fracture toughness, fatigue crack growth
and fatigue life data estimation of metallic components. Key Engineering Materials, 577, 177180.
Polezhayeva, H., Maddox, S. J., Howarth, D. & Robinson, A. Fatigue Testing to Identify Effect of Mean Stress on
Fatigue Strength of Welded Joints in Ship Structures. The Twenty-third International Offshore and Polar
Engineering Conference., 2013. International Society of Offshore and Polar Engineers.
Pook, L. P. 2013. Fifty years of crack path research. In CP2009.
Qumner, Y., Huang, C.-H. & Lee, C.-F. 2013a. Assessment of Critical Fatigue Crack Length Considering the
Fracture Failure of Ship Longitudinal Members. The 27th Asian-Pacific Technical Exchange and Advisory
Meeting on Marine Structures (TEAM2013) Keelung, Taiwan.
Qumner, Y., Huang, C.-H. & Lee, C.-F. 2013b. Fracture Resistance of Ship Longitudinal Members Including
Fatigue Crack. Structures, Safety and Reliability, 2a, V02AT02A026.
Radaj, D., Lazzarin, P. & Berto, F. 2009a. Fatigue assessment of welded joints under slit-parallel loading based on
strain energy density or notch rounding. Int. J. Fatigue, 31, 14901504.
Radaj, D., Lazzarin, P. & Berto, F. 2013. Generalised Neuber concept of fictitious notch rounding. Int. J. of Fatigue,
51, 105115.
Radaj, D., Sonsino, C. M. & Fricke, W. 2006. Fatigue assessment of welded joints by local approaches Cambridge,
Woodhead Publishing Ltd.
Radaj, D., Sonsino, C. M. & Fricke, W. 2009b. Recent developments in local concepts of fatigue assessment of
welded joints. International Journal of Fatigue, 31, 211.
Radaj, D. & Vormwald, M. 2013. Advanced Methods of Fatigue Assessment, Springer.
Remes, H. & Fricke, W. 2014. Influencing factors on fatigue strength of welded thin plates based on structural stress
assessment. Welding in the World, 58, 915923.
Renaud, M., De Lorgeril, E., Boutillier, J. B. & Gerad, L. Fatigue and weather on ultra large containerships.
Proceedings of the PRADS2013, 2013. 384394.
Rethore, J., Limodin, N., Buffiere, J.-Y., Roux, S. & Hild, F. 2012. Three-dimensional analysis of fatigue crack
propagation using X-Ray tomography, digital volume correlation and extended finite element simulations.
Procedia IUTAM, 4, 151158.
RINA Rules 2014. Pt.B, Ch.7, Sect.4, Para 3.3 and Tables 7 and 9.
Ritchie, R. O. 2011. The conflicts between strength and toughness. Nature Materials, 10, 817822.
Rizzo, C. M. 2011. Application of advanced notch stress approaches to assess fatigue strength of ship structural
details: literature review, Report 655. Germany: Schriftenreihe Schiffbau, Technische Universitt.
Ronold, K. O. & Lotsberg, I. 2012. On the estimation of characteristic S-N curves with confidence. Marine
Rother, K. & Rudolph, J. 2011. Fatigue assessment of welded structures: practical aspects for stress analysis and
fatigue assessment. Fatigue & Fracture of Engineering Materials & Structures, 34, 177204.
Rozumek, D. & Marciniak, Z. 2012. The investigation of crack growth in specimens with rectangular cross-sections
under out-of-phase bending and torsional loading. International Journal of Fatigue, 39, 8187.
Rubio-Gonzlez, C., Felix-Martinez, C., Gomez-Rosas, G., Ocaa, J. L., Morales, M. & Porro, J. A. 2011. Effect of
laser shock processing on fatigue crack growth of duplex stainless steel. Material Science and Engineering A,
528, 914919.
Sahu, J. K., Gupta, R. K., Swaminathan, J., Paulose, N. & Mannan, S. L. 2013. Influence of Hot Corrosion on Low
Cycle Fatigue Behaviour of Nickel Base Superalloy SU 263. International Journal of Fatigue, 51, 6873.
412
Sarangi, H., Murthy, K. S. R. K. & Chakraborty, D. 2013. Experimental verification of optimal strain gage locations
for the accurate determination of mode I stress intensity factors. Engineering Fracture Mechanics, 110, 189200.
Schaumann, P., Raba, A. & Lochte-Holtgreven, S. Load Sequence Effects in the Fatigue Design of Welded Spatial
Tubular Joints in Jackets. The Twenty-second International Offshore and Polar Engineering Conference, 2012.
International Society of Offshore and Polar Engineers.
Schijve, J. 2012. Fatigue predictions of welded joints and the effective notch stress concept. Int. J. of Fatigue, 45,
3138.
Schijve, J. 2014. The significance of fatigue crack initiation for predictions of the fatigue limit of specimens and
structures. International Journal of Fatigue, 61, 3945.
Selvakumar, P. & Hong, J. K. 2013. Robust mesh insensitive structural stress method for fatigue analysis of welded
structures. Procedia Engineering, 55, 374379.
Shamsaei, N. & Fatemi, A. 2009. Effect of hardness on multiaxial fatigue behaviour and some simple
approximations for steels. Fatigue & Fracture of Engineering Materials & Structures, 32, 631646.
Shamsaei, N. & Fatemi, A. 2010. Effect of microstructure and hardness on non-proportional cyclic hardening
coefficient and predictions. Materials Science and Engineering: A, 527, 30153024.
Shamsaei, N. & Fatemi, A. 2014. Small fatigue crack growth under multiaxial stresses. International Journal of
Fatigue, 58, 126135.
Shamsaei, N., Fatemi, A. & Socie, D. F. 2010a. Multiaxial cyclic deformation and non-proportional hardening
employing discriminating load paths. International Journal of Plasticity, 26, 16801701.
Shamsaei, N., Gladskyi, M., Panasovskyi, K., Shukaev, S. & Fatemi, A. 2010b. Multiaxial fatigue of titanium
including step loading and load path alteration and sequence effects. International Journal of Fatigue, 32, 1862
1874.
Shi, J., Ma, W. & Li, N. 2013. Extended meshless method based on partition of unity for solving multiple crack
problems. Meccanica, 48, 22632270.
Shlyannikov, V. N. & Zakharov, A. P. 2014. Multiaxial crack growth rate under variable T-stress. Engineering
Fracture Mechanics, 123, 8699.
Sielski, R. A. 2012. Ship Structural Health Monitoring Research at the Office of Naval Research. JOM: The Journal
of The Minerals, Metals & Materials Society, 64, 823827.
Silitonga, S., Maljaars, J., Soetens, F. & Snijder, H. H. 2014. Numerical Simulation of Fatigue Crack Growth Rate
and Crack Retardation due to an Overload Using a Cohesive Zone Model. Advanced Materials Research, 891,
777783.
Slobodyan, Z. V., Kyryliv, V. I., Mahlatyuk, L. A. & Kupovych, R. B. 2009. Influence of the Inhibitor Treatment on
the Corrrosion Resistance of Different Zones of Welded Joints. Materials Science, 45, 136139.
Sonsino, C. M., Bruder, T. & Baumgartner, J. 2010. S-N lines for welded thin joints suggested slopes and FAT
values for applying the notch stress concept with various reference radii. Welding in the World, 54, 375392.
Sonsino, C. M., Fricke, W., De Bruyne, F., Hoppe, A., Ahmadi, A. & Zhang, G. 2012. Notch stress concepts for the
fatigue assessment of welded joints background and applications. Int. J. of Fatigue, 34, 216.
Stanzl-Tschegg, S. 2014. Very high cycle fatigue measurement techniques. International Journal of Fatigue, 60, 2
17.
Sumi, Y. 2013. Computational crack path prediction for ship structural details. In FCP2003.
Sumi, Y. 2014. Fatigue crack propagation in marine structures under seaway loading. International Journal of
Fatigue, 58, 218224.
Sumi, Y. & Okawa, T. 2013. Crack Paths from Weld Details in Three-dimensional Plate Structures. In CP2006.
Sumi, Y., Yajima, H., Toyosada, M., Yoshikawa, T., Aihara, S., Gotoh, K., Ogawa, Y., Matsumoto, T., Hirota, K.,
to the deck structure of large container ships. Journal of Marine Science and Technology, 18, 497514.
Suominen, L., Khurshid, M. & Parantainen, J. 2013. Residual stresses in welded components following post-weld
treatment methods. Procedia Engineering, 66, 181191.
Thvenet, D., Ghanameh, M. F. & Zeghloul, A. 2013. Fatigue strength assessment of tubular welded joints by an
alternative structural stress approach. International Journal of Fatigue, 51, 7482.
Toyosada, M., Gotoh, K. & Niwa, T. 2004. Fatigue crack propagation for a through thickness crack: a crack
propagation law considering cyclic plasticity near the crack tip. International Journal of Fatigue, 26, 983992.
Tronskar, J. P. & Andresen, V. Avoiding Local Brittle Zones in Offshore Pipeline Girth Welds for Reeling
Installation Involving Large Plastic Strain. ASME 2013 32nd International Conference on Ocean, Offshore and
Arctic Engineering (OMAE), June 2013. American Society of Mechanical Engineers, 11094.
Van Der Horst, M. P., Kaminski, M. L. & Puik, E. Methods for Sensing and Monitoring Fatigue Cracks and Their
Applicability for Marine Structures. Proceedings of the 23rd International Offshore and Polar Engineering
Conference (ISOPE), June 2013. 455462.
Vina, J., Argelles, A. & Canteli, A. F. 2011. Influence of Temperature on the Fatigue Behaviour of Glass Fibre
Reinforced Polypropylene. Strain, 47, 222226.
Walters, C. L. 2014. The effect of low temperatures on the fatigue of high-strength structural grade steels. 20th
European Conference on Fracture (ECF20). Trondheim, Norway.
Wang, G. R., Chen, L. Y., Zhao, M., Li, R. & Huang, B. S. 2014. The research on failure analysis of fluid cylinder
and fatigue life prediction. Engineering Failure Analysis, 40, 4857.
Wang, X. G., Crupi, V., Guo, X. L. & Zhao, Y. G. 2010. Quantitative thermographic methodology for fatigue
assessment and stress measurement. International Journal of Fatigue, 32, 19701976.
413
Wu, M., Itoh, T., Shimizu, Y., Nakamura, H. & Takanashi, M. 2012. Low cycle fatigue life of Ti6Al4V alloy
under non-proportional loading. International Journal of Fatigue, 44, 1420.
Yajima, H., Watanabe, E., Jia, Z., Yoshimoto, K., Ishikawa, T. & Funatsu, Y. 2011. Study on Fracture Toughness
Evaluation by centre-notched small size specimen for heavy-thick steel plates. Welding World, 55, 7883.
Yamamoto, N., Mouri, M., Okada, T. & Mori, T. Analytical and Experimental Study on the Thickness Effect to
Fatigue Strength (1st ReportResults of Fundamental Specimens). Proceedings of the Twenty-second
International Offshore and Polar Engineering Conference, 2012 Rhodes, Greece.
Yang, Y. & Vormwald, M. 2013. Three-dimensional Fatigue Crack Growth Path Simulation under Nonproportional
Mixed-mode Loading. In ICF13.
Yekta, R. T., Ghahremani, K. & Walbridge, S. 2013. Effect of quality control parameter variations on the fatigue
performance of ultrasonic impact treated welds. International Journal of Fatigue, 55, 245256.
Yildirim, H. C. 2013. Design aspects of high strength steel welded structures improved by high frequency
mechanical impact (HFMI) treatment. Ph.D. Thesis, Aalto University.
Yildirim, H. C. & Marquis, G. 2014. Fatigue design of axially-loaded high frequency treated welds by effective
notch stress method. Materials and Design, 58, 543550.
Yu, J., Sun, X., Jin, T., Zhao, N., Guan, H. & Hu, Z. 2010. High Temperature Creep and Low-Cycle Fatigue of a
Nickel-Base Superalloy. Materials Science and Engineering: A, 527, 23792389.
Yu, Y. H., Kim, B. G. & Lee, D. G. 2013. Cryogenic reliability of the sandwich insulation board for LNG ship.
Composite Structures, 95, 547556.
Yuen, B. K., Koko, T. S., Polezhayeva, H. & Jiang, L. 2013. Mean Stress Assessment in Fatigue Analysis and
Design, Ship Structure Committee, Report No. 466. Canada: Martec Limited.
Zappalorto, M. & Lazzarin, P. 2014. Some remarks on the Neuber rule applied to a control volume surrounding
sharp and blunt notch tips. Fatigue & Fracture of Engineering Materials & Structures, 37, 349358.
Zayed, A., Garbatov, Y. & Guedes Soares, C. 2013a. Reliability of ship hulls subjected to corrosion and
maintenance. Structural Safety, 43, 111.
Zayed, A., Garbatov, Y. & Guedes Soares, C. 2013b. Time variant reliability assessment of ship structures with fast
integration techniques. Probabilistic Eng. Mech, 32, 93102.
Zerres, P. & Vormwald, M. 2013. Review of fatigue crack growth under non-proportional mixed-mode loading.
Zhan, W., Lu, N. & Zhang, C. 2014. A new approximate model for the R-ratio effect on fatigue crack growth rate.
Engineering Fracture Mechanics, 119, 8596.
Zhang, G. 2012. Method of effective stress for fatigue: Part IA general theory. International Journal of Fatigue,
37, 1723.
Zhang, G., Sonsino, C. M. & Sundermeier, R. 2012a. Method of effective stress for fatigue: part II applications to
V-notches and seam welds. Int. J. of Fatigue, 37, 2440.
Zhang, Q., Zuo, Z. & Liu, J. 2012b. High-Temperature Low-Cycle Fatigue Behavior of a Cast Al-12Si-CuNiMg
Alloy. Fatigue & Fracture of Engineering Materials & Structures, 36, 623630.
Zhang, R. & He, L. 2012. Measurement of mixed-mode stress intensity factors using digital image correlation
method. Optics and Lasers in Engineering, 50, 10011007.
Zhang, S., Bridges, R. & Tong, J. Fatigue Design Assessment of Ship Structures Induced by Ice Loading - An
introduction to the ShipRight FDA ICE Procedure. Proceedings of the Twenty-first International Offshore and
Polar Engineering Conference, 2011 Maui, Hawaii, USA. 1924.
Zhang, W. & Liu, Y. 2012. Investigation of incremental fatigue crack growth mechanisms using in situ SEM testing.
Zhu, B. & Frangopol, D. M. 2013. Incorporation of structural health monitoring data on load effects in the reliability
and redundancy assessment of ship cross-sections using Bayesian updating. Structural Health Monitoring, 12,
377392.
Zhu, S. P., Huang, H. Z., Smith, R., Ontiveros, V., He, L. P. & Modarres, M. 2013. Bayesian framework for
probabilistic low cycle fatigue life prediction and uncertainty modeling of aircraft turbine disk alloys.
Probabilistic Engineering Mechanics, 34, 114122.
Zhu, X.-K. & Joyce, J. A. 2012. Review of fracture toughness (G, K, J, CTOD, CTOA) testing and standardization.
Engineering Fracture Mechanics, 85, 146.
Zilakos, I. K., Karatzas, V. A., Chatzidouros, E. V., Papazoglou, V. J. & Tsouvalis, N. G. 2013. A FE based
numerical tool for crack assessment in ship structures employing the CSR loading scheme. Analysis and Design
of Marine Structures, 281.

710 SEPTEMBER 2015
CASCAIS, PORTUGAL
VOLUME 1
COMMITTEE IV.1
DESIGN PRINCIPLES AND CRITERIA

COMMITTEE MANDATE
Concern for the general concept of goal orientated design, for the quantification of general sustainability
aspects in economic, human and environmental terms and for the development of appropriate procedures for
rational life-cycle design of marine structures. Special attention shall be given to the issue of GBS as
presently implemented within IMO. Possible differences between the current regulatory framework for ship
structures and the design requirements developed for offshore and other marine industries shall be
considered.
COMMITTEE MEMBERS
Chairman:
E. Rizzuto, Italy (Chair)

J. Downes, UK
M. Radon, Germany
G. Egorov, Ukraine
Y. Kawamura, Japan
S. ONeil, The Netherlands
R. Skjong, Norway
A. Teixeira, Portugal
KEYWORDS
Ship design principles, goal orientated design, design for sustainability, rational design criteria,
internalization of costs, societal cost benefit analysis.
416
ISSC committee IV.1: DESIGN PRINCIPLES AND CRITERIA
CONTENTS
1.
INTRODUCTION................................................................................................................................... 418
1.1
General concept of sustainability oriented design ...................................................................... 418
1.2
Goal oriented normative framework ........................................................................................... 418
1.3
Procedures for the impact analysis of regulations ...................................................................... 419
2.
QUANTIFICATION OF SUSTAINABILITY ASPECTS ................................................................... 419

2.1
Economic aspects ........................................................................................................................ 419
2.2
Human aspects ............................................................................................................................. 420
2.3
GCAF and NCAF indicators for loss of life ............................................................................... 420
2.3.1
Life Quality Index ......................................................................................................... 421
2.3.2
DALY and QALY indicators ....................................................................................... 422
2.4
Environmental aspects................................................................................................................. 423
2.4.1
Cost of averting a tonne of oil spilt (CATS) ................................................................ 423
2.4.2
CO2 emissions costs ...................................................................................................... 427
2.4.3
Other emissions costs.................................................................................................... 428
3.
DEPRECIATION RATES IN DECISION MAKING .......................................................................... 430

3.1
Pure time preferences .................................................................................................................. 431
3.2
Precautionary approach vs standard economic theory ............................................................... 431
3.3
Integrated Assessment Models ................................................................................................... 432
3.4
Tails of the probability distributions ........................................................................................... 434
3.5
Role of the discounting rate ........................................................................................................ 434
3.6
Conclusion (depreciation rates) .................................................................................................. 437
4.
EXAMPLES RELATED TO SUSTAINABILITY ORIENTED DESIGN ......................................... 437

4.1
Probability based design.............................................................................................................. 437
4.2
Lifecycle Design.......................................................................................................................... 439
4.3
Lifecycle Design Considering Future Climate Change.............................................................. 441
5.
REGULATORY FRAMEWORK FOR MARINE STRUCTURES..................................................... 443

5.1
Development of Goal Based Standards at IMO ......................................................................... 444
5.1.1
IACS Harmonized Common Structural Rules for Bulk Carriers and
Tankers .......................................................................................................................... 444
5.1.2
Goal Based Standards/Safety Level Approach (GBS/SLA) at IMO........................... 446
5.2
Regulatory actions implemented at IMO targeting environmental protection .......................... 447
5.2.1
Energy Efficiency Design Index (EEDI) ..................................................................... 447
5.2.2
NOx SOx control ............................................................................................................ 447
5.2.3
Emission Control Areas ................................................................................................ 447
5.2.4
MARPOL Annex V Prevention of Pollution by Garbage from Ships ........................ 448
5.2.5
IMO Ship Recycling (The Hong Kong Convention) ................................................... 448
5.2.6
Pre-normative investigations at IMO in the field of noise radiation
into water ....................................................................................................................... 449
5.3
Other (non IMO) regulatory actions in the field of ships ........................................................... 449
5.3.1
Developments in the Naval Ship Code ........................................................................ 449
5.3.2
Inland vessels ................................................................................................................ 450
5.3.3
EU Directive on Safety of Offshore Oil and Gas Operations ...................................... 451
5.4
Comments on the recent developments in the normative framework ....................................... 451
6.
STUDIES FOCUSSING ON ENVIRONMENTAL IMPACT ............................................................. 451

6.1
Studies on Green House Gas Emissions ..................................................................................... 451
6.2
Studies on countermeasures to limit emissions .......................................................................... 452
6.2.1
Slow steaming ............................................................................................................... 452
6.2.2
Scale effects and propulsive improvements ................................................................. 452
6.2.3
Discussions of the EEDI concept ................................................................................. 452
6.2.4
Studies on control of NOx and SOx emissions ............................................................. 453

6.2.5
6.2.6
7.
417
Emissions Trading Schemes ......................................................................................... 453

Alternative Fuels ........................................................................................................... 453
CONCLUSIONS ..................................................................................................................................... 453
REFERENCES ................................................................................................................................................ 454
418
1.
INTRODUCTION
1.1
General concept of sustainability oriented design
The present report, as the previous ISSC reports of Committee IV.1, follows the same general definition
of sustainability that was given by the Brundtland Commission of the United Nations: economic
development which meets the needs of the present generation without compromising the ability of future
generations to meet their own needs (UN-WECD, 1987).
A sustainable development of the maritime transport involves therefore a detailed consideration of all
the negative implications such transport mode has for the human society (social costs) and a proper
assessment of compliance to the general sustainability target, based on a cost/benefit analysis.
As stated by Korzhenevych et al. (2014), transportation contributes significantly to economic growth
and enables a global market. Transport modes, however, also produce negative effects. Shipping traffic,
in particular, contributes to air and water pollution and shipping accidents to losses in terms of human
lives, economical losses and ecological damages. These effects give rise to costs that can be expressed in
monetary terms and affect in various ways the Society: health costs caused by air pollution (due to NOx,
SOx, PM, ), lives lost and loss of biodiversity in traffic accidents, costs related to the world scale
climate impact of shipping, etc.
These societal costs are referred to as external costs, that sum up to those directly borne by the
transport first and second parties (private or internal costs, such as: wear, tear and energy cost due the
ship operation, own time costs, transport fares and transport taxes and charges (port fees, pilot, insurance,
etc.). The same kind of classification applies to benefits: first and second parties take advantage of most
of them, even though society also experiences a gain.
The external costs of transportation are generally not paid by transport actors and hence not taken into
account when they take a decision about a transport activity.
Internalisation of external costs means making such effects part of the decision making process of
transport first actors. This can be done directly through regulation, i.e. issuing specific requirements in
terms of operational and/or design control measures, or indirectly, providing suitable incentives to
transporters with market-based instruments (e.g. taxes, charges, emission trading, etc.). Combinations of
these basic types are also possible.
In other terms, the role of the Regulator is to assess if and under which conditions a transport activity is
acceptable from a Societal viewpoint and to establish a Regulatory Framework able to control the societal
losses and redistribute their costs in a balanced way, in particular on the first actors of the transportation
process (internalisation of the external costs). In a technical analysis of sustainability, it is important that
all external costs are included. All kinds of costs incurred during the lifetime of the product must be
assessed. Additionally, some social costs are incurred also long after the product lifetime has expired, as
is the case for the CO2 emitted during the lifetime of the product. It is also important to include external
costs of all raw materials that goes into the product, like e.g. the steel used to construct a ship. In a
comparison of sustainability of e.g. different transport mode, it is also important to properly account for
the differences in tax regimes.
The key point of this concept is a proper evaluation of the external costs, (also called social or implied
costs) that are now not directly perceived as such by the actors.
This means, in a first place, the identification of all implications (negative and, in case, positive) of the
transportation process and a quantification on a monetary scale.
The results of this type of investigations that has been carried out by different research groups will be
reviewed and commented in chapter 2 with reference to the shipping sector.
1.2
Goal oriented normative framework
Another issue for the Regulator when establishing a regulatory framework is the way the various
requirements are organized internally. Such requirements can regard subjects of different degrees of detail
and targets of different levels of generality. A trend in the definition of the worldwide shipping regulatory
framework at the international Maritime Organisation (IMO) in the last decade has been to organize the
targets of the regulatory action in hierarchical order (from an upper level, the most general, to the most
detailed ones, corresponding to the provisions on specific items of the ship). The procedure has a
qualified characteristics in the internal coherence of the various levels, which is to be checked by codified
procedures. This verification process in particular is applied to check the correspondence of the lower
level requirements (Classification Societies Rules and Standards) to the upper level targets (goals and
functional requirements).
419
This approach has been named in the maritime field Goal Based and the first applications have been
reviewed in the past reports of this committee, while a few recent ones will be considered in chapter 5.
Without recalling in details these applications, it is here noted that different results can be obtained
depending on the way the requirements at the higher levels are formulated and on the way the coherence
between levels is assessed.
1.3
Procedures for the impact analysis of regulations
The study of the impact of normative requirements to assess their acceptability from a societal point of
view has been carried out by different Regulatory Bodies under different terms and in different ways,
generally focussing on specific aspects of the impact.
At OECD level, these investigations have been termed Regulatory Impact Assessment (RIA), which,
also from a semantic viewpoint, does not set boundaries to the evaluation of the regulation impact (neither
in the type, nor in the domain of the impact). RIA has been conceived as a key instrument for improving
regulatory quality and good governance by ensuring more coherent and transparent policies and for
making regulation more effective and efficient. By analysing potential consequences of proposed
regulations, and by comparing different options, RIA is a methodological framework and an
administrative procedure for better-informed policy-making. If properly conducted, the impact
assessment should systematically assess the impact likely to arise from government regulation and
communicate this information to the decision-makers (Jacob et al., 2011).
The generic statement above reported about the necessity in a RIA of a systematic assessment of all
types of impacts arising from the regulation to be evaluated has found an increasingly broader
interpretation with time at OECD. Following again (Jacob et al., 2011) and related bibliography
Originally, RIA focused on identifying the direct economic costs and benefits of different regulatory
alternatives on a wide range of actors, see (Kirkpatrick and Parker, 2007), (Radaelli and De Francesco,
2010). But in recent years RIA has experienced a high degree of diversification of approaches regarding
orientation, ambition, institutionalisation and transparency of the procedures. Nowadays, in several
countries, RIA requires the assessment of all types of possible impacts. It varies among countries,
however, to what extent these assessments comprise a consideration of environmental issues.
In other contexts than OECD, the impact assessment has been focussed specifically on environmental
aspects, (Life Cycle Assessment (LCA) of a manufact or an activity), but in this case the analysis is
limited to regular emissions: see, in particular for ships, e.g.: (Mountaneas et al., 2015).
On the other hand, as mention above, at IMO impact evaluations concentrated so far on safety,
following the FSA procedure and focussing on low probability accidental consequences for humans.
A possibility, which has not been yet enforced in the maritime field and is here proposed for future
applications, is to set the sustainability of the ship transport system as top goal for the normative
framework of shipping. This implies (in a Goal Based-structured perspective) to verify the functional
requirements at the second level in terms of a regulatory impact assessment based on a sustainability
evaluation, accounting for all internal and external costs. This is very well in line with the general trend
above recalled towards a generalisation of the impact concept in a regulatory context.
It is here noted that, from this viewpoint, the term sustainability assumes an holistic meaning (wider
than in other contexts). Its definition expands in space, in time (to intra-generational periods) and in the
probability domain (from regular to very rare events), in order to include all possible types of impacts.
In this respect, the influence of the emissions of a ship on the global climate changes on a planetary
scale, as well as the life cycle impact of the ship on future generations and the expected values of the
damages caused to humans, property and bio-diversity by rare accidental events are all possible examples
of aspects to be included in the impact assessment.
2.
QUANTIFICATION OF SUSTAINABILITY ASPECTS
In this chapter, measurement and assessment criteria regarding to various aspects included in the
sustainability concept are presented. In particular several approaches developed for internalizing costs
associated to assets, loss of human life, human injures, oil spills and environmental pollution are
reviewed. This allows the inclusion of sustainability-related costs into cost benefit/effectiveness analyses
and therefore in a rational decision-making process.
2.1
Economic aspects
In the context of a cost benefit analysis carried out from a societal viewpoint, the cost of a human activity
(e.g. a transport mode) represents the effort the society is called to pay in order to implement that activity.
Such cost should be balanced by a societal gain.
420
In a broad sense, the cost includes all types of negative effects the activity implies. The economic
burden of an activity is part of this picture, perhaps the most evident and studied one, at least for the part
internal to first parties of the process. Expenses for designing, building, running, maintaining, repairing
and dismantling a ship are all examples of resources devoted to a maritime transport mode. Some of the
expenses may be deterministically defined, other are affected by uncertainties (e.g. fuel consumption, in
dependence of weather conditions) or need to treat in a probabilistic way (accidents).
Also in case of the economic items of the balance, however, it is important to include all aspects in the
analysis. With reference again to Korzhenevych et al. (2014) it is here recalled that the societal costs for a
generic transportation mode should include on the long term also the building costs for re-building and or
adapting to new requirements the infrastructures needed for the particular transportation mode. This in
addition to the maintenance costs for the existing infrastructure that are definitely included in the external
costs. In the specific case of shipping, this item relates f.i. to the construction of new ports, in addition to
the dredging, maintenance, and service costs related to the normal functioning of ports
2.2
Human aspects
In this section it is covered how is it possible to account for the aspects related to the safety (life, health
and well being) of humans in the context of (normative) decision making.Indicators (indexes) and criteria
for the assessment will be reviewed.
2.3
GCAF and NCAF indicators for loss of life
Indicators based on a cost-effectiveness ratio related to the safety of human life are available in the
maritime industry since some time (Norway, 2000) and have been formally accepted by including them
into the consolidated FSA guidelines (IMO, 2007) and later into their revised version (IMO, 2013b). The
implication is that internalizing costs related to safety, injuries and ill health in any decision model
becomes possible.
Indicators used to express the cost effectiveness of risk control measures are the Gross and Net Cost of
Averting a Fatality (GCAF/NCAF), as described in Appendix 7 of the FSA Guidelines (IMO, 2013b).
Definition
The Gross Cost of Averting a Fatality (GCAF) is defined in terms of ratio of marginal (additional) cost
C of the risk control option divided by the reduction in risk to personnel in terms of the fatalities averted
R; i.e.:
GCAF =
C
R
(1)
The Net Cost of Averting a Fatality (NCAF) index is defined by subtracting possible economic benefits
B of the risk control option (e.g benefits deriving from the reduction in economic lossess and
environmental damages that is expected by implementing the safety measure) from the cost of the risk
control option (RCO):
NCAF =
C B
R
(2)
According to appendix 7 of the revised FSA guidelines (IMO, 2013b), either the two indices can in
principle be used. However, it is recommended to firstly consider GCAF and if the cost effectiveness of
an RCO satisfies the inherent acceptance criterion (see below), then NCAF may be also considered. The
reason is that NCAF may be misused in some cases for pushing certain RCOs, by including more
categories of economic benefits when considering preferred RCOs.
NCAF and GCAF can always be adopted to compare different RCOs from the viewpoint of their costeffectiveness.
Assessment (acceptance criteria)
For both the GCAF and the NCAF, appropriate cost-effectiveness criteria can be specified in terms of
quantitative reference values for the maximum cost of averting a fatality. In other words, limit
GCAF/NCAF values can be set, below which safety measures are to be considered cost-effective in an
absolute way.
A criterion defined in terms of a limit value of GCAF/NCAF of US$ 3 million was suggested by
(Norway, 2000) and this was the value provided by IMO (2007) for illustrative purposes (Table 1).
According to IMO (2007) and later to IMO (2013b), specific values selected as appropriate should be
explicitly defined and used in any FSA study.
421
Table 1. Cost-effectiveness criteria for NCAF and GCAF indicators ((IMO, 2007), (IMO, 2013b)).
Criterion covering risk of fatality,

injuries and ill health
Criterion covering only
risk of fatality
Criterion covering risk of
injuries and ill health
NCAF [US $]
GCAF [US $]
3.0 10
3.0 10 6
1.5 106
1.5 106
1.5 106
1.5 106
These criteria are not static and, on the contrary, should be updated every year according to the average
risk free rate of return (see the discussion of rates in chapt. 3 below) or by calibrating the limit value on
the basis of other indicators and inherent acceptance criteria (e.g the Life Quality Index criterion, see
below).
The NCAF is a price which society should be willing to pay for saving lives according to its ethical
principles and which society can afford, i.e. for safety-relevant regulations in cost-benefit calculations.
Independently from the way its is found, the NCAF, however, should never be taken as the value of
human life (Rackwitz, 2002b).
2.3.1
Life Quality Index
Definition
The Life Quality Index (LQI: see (UNDP, 1990), (Lind, 2002)) is a social indicator reflecting the length
of a quality life in a given society.
It is constructed from two aggregated indicators, i.e. the life expectancy at birth (e) and the gross
domestic product pro capite (g),
LQI = g w e (1-w )
(3)
where the exponent w is the proportion of life spent in economic activity (w=1/8 in developed countries).
Acceptance criterion
Using the definition in Eq. 4, a Life Quality Index criterion ((Nathwani et al., 1997), (Skjong and Ronold,
1998), (Skjong and Ronold, 2002), (Rackwitz, 2002a), (Rackwitz, 2002b)) may be defined: according to
such criterion, safety interventions are regarded as justifiable as long as they contribute positively to LQI.
The improved safety by implementation of a safety measure is expressed through a positive change e in
the life expectancy, e. The cost of implementing the measure is expressed through a change, g in the
gross domestic product, g. The Life Quality Index Criterion implies that the measure is implemented
when
g w
e
>
e
g 1 w
(4)
which results from differentiation of the expression that defines the LQI and requiring a positive
increment in the index: LQI>0
The use of this criterion for decision-making on a particular normative option implies that one is
willing to take on an increased risk, expressed in terms of a negative e, if the associated gain or
compensation g is large enough. Optimality is achieved when the inequality of the Life Quality Index
Criterion is turned into equality.
Assuming that saving a human life corresponds to saving, in years, half the life expectancy of an
individual: e=e/2, the limit acceptable value for the net cost of averting a fatality, (NCAF)acc, can then be
calculated as:
( NCAF )acc
g e 1 w
4 w
(5)
The acceptable NCAF value of Eq. (6) can be updated from global statistics, available at the websites
of OECD, CIA World Factbook, International Monetary Fund (IMF) and World Bank (WB). The OECD
member countries are typically adopted to calculate average values of NCAF as they represent about 95%
of the global GDP and presumably a similar share of the maritime transport. When updating the NCAF
criterion all parameters can be revised (i.e. g, e w).
422
From these, the g (Gross National Product pro capite) is the one with larger variability, while the
percentage of the lifetime spent in economic activity (w) is typically not easily obtained.
Figure 1 shows the variation on the average NCAF for the OECD countries from 1995 to 2011. Table 2
shows that the total average values of OECD countries in the period 2002-2011 is approximately
4.63 106 US$ , which corresponds to an increase of 50% compared to the average of the period 19952005 ( 3.07 10 6 US$ ).The main changes are due to the fact that the number of OECD countries has
increased, the pro capite Gross Domestic Product has increased, life expectancy at birth has increased and
less time is spent in economic activity. In addition the US$ has decreased its value against most other
currencies.
Table 2. Time variation of the average NCAF value of the OCDE member countries ( 10 6 US$ ).
Year
1995
2000
2005
2009
2012
1995-2005
1995-2011
2002-2011
NCAF
2.74
2.70
4.31
5.05
5.76
3.07
3.85
4.63
7.0
6.0
NCAF
5.0
4.0
3.0
2.0
1.0
0.0
1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014
Year
Figure 1. Variation of the average NCAF for the OCDE countries from 1995 to 2011
(Data source: OCDE statistics).
2.3.2 DALY and QALY indicators

For quantifying health effects and injuries, the FSA Guidelines advocate the use of the DALY (Disability
Adjusted Life Years)/QALY (Quality Adjusted Life Years) concept, which is promoted by the World
Health Organization (WHO).
Definitions
The disability-adjusted life year (DALY) is a measure of overall disease burden, expressed as the number
of years lost due to ill-health, disability or early death.
The DALY was first conceptualised by Murray and Lopez in work carried out with the World Health
Organization and the World Bank known as the global burden of disease study (Murray and Lopez,
1996).
The DALY is becoming increasingly common in the field of public health and health impact
assessment (HIA). It extends the concept of potential years of life lost due to premature death...to
include equivalent years of healthy life lost by virtue of being in states of poor health or disability. In
so doing, mortality and morbidity are combined into a single, common metric.
Traditionally, health liabilities were expressed using one measure: (expected or average number of)
Years of Life Lost (YLL). This measure does not take the impact of disability into account, which can
be expressed by: Years Lived with Disability (YLD). DALYs are calculated by taking the sum of these
two components. In a formula:
DALY = YLL + YLD
(6)
The DALY relies on an acceptance that the most appropriate measure of the effects of chronic illness is
time, both time lost due to premature death and time spent disabled by disease. One DALY, therefore, is
equal to one year of healthy life lost.
423
The quality-adjusted life year or quality-adjusted life-year (QALY) is a measure of disease burden,
including both the quality and the quantity of life lived. The QALY was originally developed as a
measure of health effectiveness for cost-effectiveness analysis, a method intended to aid decision-makers
charged with allocating scarce resources across competing health-care programs (Fanshel and Bush,
1970; Torrance et al., 1972; Weinstein and Stason, 1977).
The QALY is based on the number of years of life that would be added by the intervention. Each year in
perfect health is assigned the value of 1.0 down to a value of 0.0 for being dead. If the extra years would not
be lived in full health, for example if the patient would lose a limb, or be blind or have to use a wheelchair,
then the extra life-years are given a value between 0 and 1 to account for this.
Acceptance criteria
A possible way to develop an acceptance criterion based on the DALY concept, i.e. to define an acceptable
cost for averting a loss (or obtaining the gain) of 1 year of healthy life, is to derive such criterion from the
NCAF criterion (see sect. 0 above). This can be obtained by assuming that one prevented fatality implies 35
(i.e. e/2 (where e is the life expectancy) Quality Adjusted Life Years gained. i.e.:
NCAFacc / 2
(7)
e
which, assuming for the acceptable NCAF a value of 3106 US$, provides an acceptable cost per DALY
gained of 42000 US$.
Although the DALY criterion may have some limitations and more research on this is required, the
criterion may be sufficient in prioritizing risk control options. It is surprising how the value of
US $ 42 000 is close to other criteria derived by different techniques such as the one used for decisions
in the health care sector, where e.g. Gafni (1999) refer to a DALY of $35,000. Also, the average value for
life saving interventions in the US in (Tengs et al., 1995) is also $42.000 per life-year (Skjong, 2009).
In any case both the CAF and DALY/QALY Criteria can be regulary updated based on the LQI
formulas above. One of the lhe latest updates was carried out in (Spouge and Skjong, 2014).
( Cost for averting a loss of 1 year of healthy life )acc
2.4
Environmental aspects
In this section, indicators for quantifying some of the environmental consequences of a normative
decision will be discussed.
2.4.1
Cost of averting a tonne of oil spilt (CATS)
A cost effectiveness criterion related to accidental oil spills of tankers has been suggested by Vanem et al.
(2008) based on the work performed under the EU project SAFEDOR. Vanem et al. (2008) have proposed
an evaluation criteria based on cost effectiveness considerations, i.e. the cost of averting a tonne of oil spilt
(CATS). The rationale behind the approach suggested is in line with cost effectiveness criteria normally
employed in Formal Safety Assessment (FSA) studies on safety issues, i.e. the GCAF/NCAF (Gross/Net
Cost of Averting a Fatality). Based on a review of available oil spill statistics and a generic, global average
cost per tonne of oil spilt, Vanem et al. (2008) have formulated a criterion in terms of CATS, suggesting that
options with a CATS value less than F $40,000 should be implemented. An exact value for the assurance
factor F was not established, but it was indicated that it should take a value between 1 and some upper limit
of 3. This work has also compared the proposed criteria with previous actual decisions related to the OPA
90 regulations. Overall, it was found that the proposed methodology is in general agreement with previous
decisions and that the suite of OPA regulations corresponds to a CATS value of approximately USD 63,000
showing that the proposed CATS criteria are appropriate and the overall OPA 90 regulations are sensible
and associated with a reasonable degree of cost effectiveness. This would correspond to an assurance factor
F of 1.5 for the global criterion, and would also be in agreement with previous decisions. However, Vanem
et al. (2008) recommended that further studies should be carried out in order to obtain more accurate
estimates of oil spill costs, but until better statistics become available, the CATS value of $60,000 (1.5
$40,000) could be adopted.
A debate led by Greece has also started at the 56th MEPC (Greece, 2007), focusing on the formulation
and the value of CATS, alerting for the extreme variability of the per tonne cost (clean-up and damage) of
oil spills worldwide. Furthermore, the document stated that the use of a single criterion focused on the
volume of oil spilt was not sufficient, given that, from the multiple variables that influence this cost,
volume is not the most relevant. MEPC 56 had therefore agreed to establish a correspondence group,
under the coordination of Greece to carry out an in-depth analysis of the proposed environmental risk
424
evaluation criteria for the purpose of the Formal Safety Assessment (FSA) before inclusion of such
criteria in the IMO FSA Guidelines.
At MEPC 60 (March 2010), after considerable debate, the majority of the members of the Working
Group on Environmental Risk Evaluation Criteria within the context of Formal Safety Assessment
expressed its preference for a non-linear approach instead of a constant CATS threshold. To this effect,
MEPC 60 agreed that in order to arrive at the recommended CATS criterion, the following should be
considered, among other things (IMO, 2010), MEPC 60/WP.11):
(1) Member governments or interested organizations having their own additional data attempt to verify,
and adjust as necessary, the said regression formula by incorporating their additional (chosen) data in
the analysis. In this connection, MEPC 60 agreed to invite the interested stakeholders to submit their
data for each cost component and the results of their analysis for consideration.
(2) Following a more reliable establishment of the cost curve, a proposed CATS formula, to be used in
the cost-effectiveness step of FSA can be established by introducing a margin or factor value (socalled assurance factor), still to be agreed, representing societys willingness to prevent an accident
rather than to simply neutralize its consequences.
(3) MEPC 60 invited member governments and interested organizations to use the non-linear cost
function in FSA studies with a view to gain experience with its application and provide information to
the IMO which may help to improve the proposed functions.
In the meantime several studies had been performed to quantify the total cost of an oil spill using
different regression approaches and different sources of data, as reviewed by Kontovas et al. (2010).
Shahriari and Frost (2008) proposed a mathematical method for estimating clean up costs based on
regression analysis of 80 incidents occurred during the period 19672002. The models parameters
included the quantity of oil spilled, the oil density, the distance to shore, cloudiness and level of readiness
based on ITOPF estimations on how well different world regions cope with oil spills. Liu et al. (2009)
proposed a log linear relation between the total cost and the spill size based on a combination of
simulating and estimation methods.
Yamada (2009) performed a regression analysis of the quantity of oil spilled and the total cost of the
spill using data from the Annual Report from the IOPCF (International Oil Pollution Compensation
Fund). The study of Yamada (2009) was officially submitted by Japan to IMO MEPC recommending a
volume based approach to quantify the total cost of oil spills (Japan, 2008), MEPC 58/17/1) given by the
regression formula:
C = f (V ) = 38,735V 0.66
(7)
where C is the total spill cost in US % and V the spill size in tonnes.
Psarros et al. (2011) have analysed cost data for 185 oil spill incidents combining the IOPCF data with
spill data derived from the accident database developed during the EU Research SAFECO II and
proposed a value of $80,000/tonne, as a CATS criterion. This value was achieved using an assurance
factor of 1.5 and the mean total cost value of $54,721/tonne (i.e. E(C/V), where C is the total spill cost
and V the spill size in tonnes). Psarros et al. (2011) have also obtained a regression model to estimate the
total unit cost of an oil spill as a function of the amount of oil spilled that is very similar to that of
document MEPC 59/INF. 21 submitted by Norway (2009) to MEPC 59 in July 2009 as
C = f (V ) =
60,515
= 60,515 V 0.647
V 0.353
(8)
The mean total cost value of $54,721/tonne and the corresponding CATS value of $80,000 have been
criticised by Psaraftis (2011), as the $54,721/tonne do not correspond to the average of the ratio spill
cost/spill volume (C/V) but to 93th percentile of the distribution of the ratio C/V. According to Psaraftis
(2011), the $54,721/tonne value can lead to erroneous results, particularly when extrapolated and applied
to all spill sizes. Therefore, a better option would be to use the non-linear function that relates spill cost
(C) and spill volume (V) instead of linearizing it or even approximating it as, C= $54,721 V.
Kontovas et al. (2010) have reviewed some previous studies and have also performed a regression
analysis of oil spill cost data provided by the IOPCF. The dataset used to estimate the total cost of oil
spills consisted of 83 spills in the period 1979-2006 obtained after removing incomplete entries and
outliers. The minimum volume was 0.1 tonnes and the maximum was 84,000 tonnes. The average spill
425
was 4854.29 tonnes but the median was just 140 tonnes, which clearly show that most claims came from
relatively small spills. In fact, only 11 spills were above 5000 tonnes and, thus, one should be very careful
when using the regression formulas to extrapolate the cost of large spills. Using this dataset Kontovas
et al. (2010) have proposed the following non-linear function of total spill cost versus spill weight that has
been also submitted by Greece (2009) as MEPC 60/17, annex 2:
C = f (V ) = 51,432 V 0.728
(9)
The Working Group on Environmental Risk Evaluation Criteria at MEPC 60 noted that the research
work independently conducted by Japan (2008) (MEPC 58/17/1), Norway (2009) (MEPC 59/INF.21) and
Greece (2009) (MEPC 60/17, annex 2) summarised in Table 3 led to very similar non-linear regression
functions of total spill cost versus spill weight. Among these non-linear regressions models the one
proposed by Greece was considered as more conservative, resulting in larger cost figures per tonne of oil
spilled, except for spills less than 4 tonnes as illustrated in Figure 2 and was proposed as a basis for
further analysis (IMO, 2010), MEPC 60/WP.11).
Table 3. Non-linear functions of total spill costs (obtained by regression) (IMO, 2010).
Total Spill Cost (C) in US $
Data Source
Reference
versus spill weight (V) in tonnes
0.66
(Japan, 2008) (MEPC 58/17/1)
38,735 V
IOPCF
IOPCF, SAFECO II
(Norway, 2009) (MEPC 59/INF.21
60,515 V 0.647
EU project
(Greece, 2009) (MEPC 60/17, annex 2) 51,432 V 0.728
IOPCF
Total unit spill cost [US %/tonne]
80000
70000
Japan (2008)
(MEPC 58/17/1)
60000
Norway (2009) (MEPC 59/INF.21
50000
Greece (2009) (MEPC 60/17, annex 2)
40000
IMO (2013) MEPC 62/INF.24
30000
20000
10000
0
1
10
100
1000
10000
100000
Spill weight [Tonne]
Figure 2. Total unit spill cost [US$/tonne] versus spill weight for V1 tonne (IMO, 2010).
In July 2011 at the 62th MEPC (IMO, 2011c), MEPC 62/WP.13), the work on CATS was completed and
a spill sized dependent formulation was agreed and included in the annex 7 of the revised IMO FSA
Guidelines (IMO, 2013b). First, a consolidated oil spill database, developed jointly by Germany, Japan,
Greece and the United States and based on IOPCF data, US Data and Norwegian data, was endorsed.
Second, MEPC 62 agreed that the following volume-based total spill cost regression formulae derived
from the consolidated oil spill database could be appropriate to be used in environmental FSA studies
(Table 4).
Table 4. Volume-based total spill cost regression formulae (IMO, 2013b).
Dataset
C = f(V) = Total Spill Cost (TSC) [in 2009

US dollars]
0.5893
All spills
67,275 V
V > 0.1 tonnes
42,301 V 0.7233
Reference
MEPC 62/INF.24
MEPC 62/18
426
Figure 3 shows the data of the consolidated oil spill database in terms of specific costs per tonne spilled
(figure 5 of document MEPC 62/INF.24). Further information with respect to the basis of
the database can be found in document MEPC 62/INF.24. It should be acknowledged that the
consolidated oil spill database has limitations and possible deficiencies. These are described in
document MEPC 62/INF.24 and may also involve incomplete or missing data on costs or other
information.
Figure 3. Oil spill costs as a function of spill size. All specific oil spill cost data in 2009 US$ (spill cost per tonne).
Source: document MEPC 62/INF.24.
It is interesting to note (and quite unexpected) that even though data from the USA and Norway were
added (and these are quite expensive spills), the new total spill cost functions obtained are well below the
ones proposed by Greece, based on IOPCF data alone, and by Norway, based on IOPCF and data from
the SAFECO EU project, for spills larger than 10 tonne, as shown in Figure 2.
Although the regression formulae of Table 4 derived from the consolidated database were suggested in
the revised FSA guidelines (IMO, 2013b), the submitter of the FSA can amend this database with new oil
spill data and is free to use other formulae, as long as these are well documented. For example, if a FSA is
considering only small spills, the submitter may filter the data and perform his or her own regression
analysis, however, this amendment should be properly documented.
It is recommended that the FSA analyst use the following formula to estimate the societal oil spill costs
(SC) used in the analysis:
SC threshold = F assurance * F uncertainty * f(V)
(10)
where FAssurance is assurance factor (allowing for societys willingness to pay to averting accidents),
FUncertainty is the uncertainty factor (allowing for uncertainties in the cost information from occurred spill
accidents, and f(V) is the volume-dependent total cost function representing the fact that the cost per unit
oil spilled decreases with the spill size in US$ per tonne oil spilled.
According to IMO (2013b), the values of both assurance and uncertainty factors should be well
documented. In addition, if value of FAssurance and FUncertainty other than 1.0 are used, a cost-effective
analysis using FAssurance= 1.0 and FUncertainty = 1.0 should be included in the FSA results, for reference. In
order to consider the large scatter, the FSA analyst may perform a regression to determine a function f(V)
that covers a percentile different than 50 % and document it in the report.
Psaraftis (2012) identified some open issues on this topic that are related to the considerable options
that the amended FSA guidelines provide to the FSA analyst when performing an environmental FSA.
These are mainly related to the selection of both assurance and uncertainty factors and the possibility to
perform a regression analysis not at 50 % (as is standard in all regressions), but at a level different from
(and likely higher than) 50 %. For instance, there has been no agreement on the specific value for the
assurance factor, even though there is a clear belief by some IMO delegations that this factor should be
well above 1.0. On the other hand, some other delegations have suggested that this factor should not be
427
the object of an FSA study but should be left for policy makers to decide. As things stand, the value of
this factor is open, and FSA allows any value, so long as it can be well documented.
Also, it is not clear how the uncertainty factor can be calculated with a reasonable degree of
confidence. In fact, as spill claims are typically inflated, the uncertainty factor can even be less than 1.0,
but some IMO delegations want to use an uncertainty factor much higher than 1.0 (Psaraftis, 2012). As
things stand, any uncertainty factor can be used in an environmental FSA, as long as it is well
documented.
It is still unclear if the present formulations will be corresponding to the willingness to pay for averting
oil pollution accidents that are implicit in the present MARPOL. An agreement at IMO on how to
internalize the societal cost of oil spill in the regulatory process, is anyway important for the design for
sustainability approach (Skjong, 2009). This may be compared to the decision processes in EU on
offshore regulations. The EU Impact Assessment (EU, 2011) is largely a worst case scenario analysis
estimating a return period for a Macondo type accident in Europe.
2.4.2 CO2 emissions costs

A Cost of Averting a Tonne of CO2-eq Heating (CATCH) criterion was proposed by (Skjong, 2009). The
methodology proposed is in line with the analysis carried out by Intergovernmental Panel on Climate
Change (IPCC) presented in the Fourth Assessment Report, in the contribution from Working Group III
(IPCC, 2007a). The report contains estimates of the risk reduction at different carbon price levels, both
based on top-down and bottom-up studies and for two different scenarios, reproduced here in Table 5 and
Table 6 (Skjong, 2009).
Eide et al. (2009) applied the methodology for assessing the cost-effectiveness of technical and
operational measures for reducing CO2 emissions from shipping, through the development of the
evaluation parameter called the CATCH and a decision criterion for the implementation of measures,
against which the evaluation parameter should be evaluated. The decision parameter for emission
reduction CATCH has been established using the same approach adopted in the development of the
decision parameter, NCAF (Net Cost of Averting a Fatality), already included in the FSA guidelines
(IMO, 2007, IMO, 2013b), and the similar parameter for assessing measures for oil spill reduction, CATS
(Cost of Averting a Tonne of oil Spill): see Vanem et al. (2008); Skjong et al. (2005); (Skjong, 2009).
Table 5. Global economic mitigation potential in 2030 estimated from bottom-up studies.
5-7
Reduction Relative
to SPES A1B
(68 GtCO2 eq/yr) (%)
7-10
20
9-17
14-25
19-35
50
13-26
20-38
27-52
100
16-31
23-46
32-63
Carbon Price Economic Potential

(US$/tCO2eq)
(GtCO2 eq/yr)
Reduction Relative
to SPES B2
10-14
Table 6. Global economic mitigation potential in 2030 estimated from top-down studies.
Carbon Price
(US$/tCO2eq)
Economic Potential
(GtCO2 eq/yr)
20
50
100
9-18
14-23
17-26
Reduction Relative
to SPES A1 B
13-27
21-34
24-38
Reduction Relative
to SPES B2
18-37
29-47
35-53
The economic potential for emission reduction estimates is surprisingly consistent at all carbon price
levels, (Skjong, 2009). Two scenarios are defined as follows: the A1 storyline and scenario family
describes a future world of very rapid economic growth, global population that peaks in mid-century and
declines thereafter, and the rapid introduction of new and more efficient technologies. Major underlying
themes are convergence among regions, capacity building and increased cultural and social interactions,
with a substantial reduction in regional differences in pro capite income. The A1 scenario family develops
into three groups that describe alternative directions of technological change in the energy system. The
three A1 groups are distinguished by their technological emphasis: fossil intensive (A1FI), non-fossil
energy sources (A1T), or a balance across all sources (A1B) (where balanced is defined as not relying too
428
heavily on one particular energy source, on the assumption that similar improvement rates apply to all
energy supply and end use technologies).
The B2 storyline and scenario family describes a world in which the emphasis is on local solutions to
economic, social and environmental sustainability. It is a world with continuously increasing global
population, at a rate lower than A2, intermediate levels of economic development, and less rapid and
more diverse technological change than in the B1 and A1 storylines. While the scenario is also oriented
towards environmental protection and social equity, it focuses on local and regional levels.
Assuming that the politically expressed wish to reduce the emission by 80%, compared to the current
level B2 scenario at 2030, and ignoring the uncertainties, this indicates that all measures that can avert a
tonne of CO2 eq emission for less than or about $50 should be implemented now or in the near future.
This is higher than the current price in e.g. the EU market and demonstrates that the current market based
instrument fail to internalize the societal costs. At IMO this way of deciding to implement RCOs would
be consistent with current decision making processes and FSA, e.g.
(CATCH)acc = acceptable Cost of Averting a Tonne CO2 eq Heating effect = $50
(11)
It may obviously also be argued that due to the uncertainty in the estimates, and the long term irreversible
effect of climate change, a safety factor should be introduced too. For EU and others, a reference to the
precautionary approach would be of relevance, as this is also representing an agreed policy.
Using CATCH < 50 USD/tonne as an acceptance criterion ensures that shipping contributes to the
reduction in GHG emissions in a way that is cost-effective and complies with the (IPCC, 2007a)
recommendations. On the other hand, depending on the cost of reducing emissions caused by shipping,
this criterion cannot be used to achieve a specific amount of reduction. Again, this is in agreement with
both the IPCC way of thinking and the Kyoto protocol (UNFCCC, 1997), the only difference between
these two being that the trading mechanisms established under the Kyoto protocol assume that the
CATCH is established by market mechanisms and the trading of CO2 quotas, whereas the IPCC uses a
specific concentration target and analyses the cost of reducing the CO2 level to that amount.
The suggested CATCH <50$/T CO2-eq has been used as a decision criterion for investment in GHG
emission reduction measures for shipping. Eide et al. (2009) have analysed a number of specific technical
and operational measures for reducing CO2 emissions for selected ships showing that several measures
are cost effective according to the proposed criterion. Assuming independence between the measures, the
cost effective measures (not including speed reduction) considered by Eide et al. (2009) add up to an
emission reduction in the order of 30% for the bulk carrier, and 40% for the container vessel.
The CATCH concept has been also applied by Longva et al. (2010a) for setting the target Energy
Efficiency Design Index (EEDI), but could be used in various ways, including introducing a fuel levy.
The approach quantifies the cost of averting greenhouse gas emission, which indicate the costs that needs
to be internalized in the decision processes, by regulators and individuals to achieve sustainability.
2.4.3 Other emissions costs

Generally, there are two methods of estimating external costs of transport: bottom-up and top-down
approaches (Essen et al., 2007) (Friedrich and Bickel, 2001). The bottom-up approach begins at the
micro-level, in which basic elements are initially specified in detail and then connected to form a
complete system. Hence, this approach is more precise with a potential for differentiation and is more
suited to deriving marginal external costs. By contrast, a top-down approach operates at the macro-level,
e.g., external costs of a country can be calculated and divided by the national transport volume, leading to
average external costs. This method is easier to manipulate, but fails to incorporate specific details
(Bickel and Friedrich, 2005). The above approaches have been widely used in many studies for evaluating
air pollution effects, including ExternE (Bickel and Friedrich, 2005), UNITE (Tervonen et al., 2002) and
(Nash, 2003), CAF (Holland et al., 2005) , HEATCO (Bickel et al., 2006), RECODIT (Black et al.,
2003), and GRACE (Nash et al., 2008), as reviewed by Jiang et al. (2014) in the context of the analysis of
the costs and benefits of reduction measures for the shipping industry to comply with the forthcoming
sulphur emission regulations.
Although the estimation of external costs has to consider several uncertainties, there is a wide consensus
on the best approach. The state-of-the-art is the damage cost approach or the dose-response method
(bottom-up approach). However, for assessing external costs of carbon emission, the avoidance cost
approach (which evaluates the cost of achieving a given amount of emissions reduction) is typically used
(Korzhenevych et al., 2014). The dose-response method focuses on the quantification of the explicit impact
that the emissions have on human health, environment, economic activity, etc. Efforts undertaken in the
429
last 20 years to develop standardised approaches involve a detailed analysis of the long chain of events
preceding the final impact on the exposed population. The EU funded series of projects ExternE (Bickel
and Friedrich, 2005) formalised this solution under the title Impact Pathway Approach (IPA) (Figure 4).
IPA starts with from the quantification of the pollutant emissions to the determination of impacts and
subsequently to the quantification of economic damage in monetary terms often based on valuation studies
assessing e.g. the willingness to pay for reduced health risks, (Korzhenevych et al., 2014).
Figure 4. The Impact Pathway Approach (IPA) (EEA, 2011).
Recently, Korzhenevych et al. (2014) provided a comprehensive overview of the approaches for
estimating external costs of transport, including air pollution cost from shipping, and recommend a set of
methods and default values for use when conceiving and implementing transport pricing policy and
schemes. This handbook on the external cost estimation updates the 2008 Handbook (Maibach et al.,
2008) with new developments and data generated from research projects and scientific papers in Europe.
In particular for air pollution costs, Korzhenevych et al. (2014) report more recent damage costs from
the NEEDS project (Preiss et al., 2008). In addition to covering all major pollutants and all Member
States, the values provided in NEEDS have several features that are especially relevant for the purpose of
policy application. First, they cover all European sea territories (very relevant for correctly calculating the
external costs of maritime transport). Second, they cover not only health effects (that surely correspond to
over 90% of the total external effect), but also quantify the side effects of emitted NOx and SO2 on
materials (e.g. buildings), biodiversity, and crops. Table 7 reports cost values for all major pollutants
calculated for all European sea regions using the EcoSense model (Preiss and Klotz, 2007) in the NEEDS
project, updated to the price level of 2010.
Table 7. Damage costs of main pollutants in sea areas, in per tonne (2010).
Sea region
NMVOC
NOx
PM2.5
SO2
Baltic Sea
1100
4700
13800
5250
Black Sea
500
4200
22550
7950
Mediterranean Sea
750
1850
18500
6700
North Sea
2100
5950
25800
7600
Remaining North-East Atlantic
700
2250
5550
2900
Source: NEEDS (Preiss et al. 2008), updated to year 2010 using EU nominal pro
capite GDP (PPP) figures. All values are rounded.
Table 8 and Table 9 report unit cost values for maritime transport. The types of vessels and the
corresponding emission factors are taken from (den Boer et al., 2011). It is important to note that some
important vessel categories are not included (e.g. Ro-Ro, container ships due to a lack of comprehensive
430
data in a consistent format. The damage costs presented below are differentiated by sea area according to
damage costs reported in Table 7.
Table 8. Marginal air pollution costs (2010) for maritime transport (average load), EU average,
per 1000 tkm.
Marginal air pollution cost, per 1000 tkm

Average
Remaining
Type of ship
load,
Baltic Black Mediterranea North
North-East
tonnes Sea
Sea
n Sea
Sea
Atlantic
Crude oil tanker 0-10 kt 1761
4.94
5.22
3.02
6.70
2.37
Crude oil tanker 10-60 kt 18413 1.45
1.55
0.91
1.99
0.70
Crude oil tanker 80-120 kt 49633 0.95
1.01
0.59
1.29
0.45
Products tanker 0-5 kt
810
6.71
7.07
4.09
9.09
3.22
3150
4.36
4.59
2.65
5.91
2.09
General Cargo 0-5 kt
1527
2.57
2.73
1.59
3.49
1.23
4174
2.90
3.08
1.81
3.94
1.39
Bulk carrier (feeder)
1440
4.71
5.01
2.93
6.41
2.26
Bulk c. (handysize)
14300 1.39
1.48
0.87
1.89
0.67
Bulk c (handymax)
24750 1.01
1.08
0.63
1.38
0.48
Source: own calculations using emission factors from (den Boer et al., 2011). Damage cost
factors (non-urban) from Table 7.
Table 9. Marginal air pollution costs (2010) for maritime transport (average load), EU average, per ship-km.
Type of ship
Average
load,
tonnes
Crude oil tanker 0-10 kt

Bulk carrier (feeder)
Bulk carrier (handysize)
Bulk carrier (handymax)
1761
18413
49633
810
3150
1527
4174
1440
14300
24750
Marginal air pollution cost, per ship-km

Baltic
Sea
Black
Sea
Mediterranean North
Sea
Sea
8.70
26.78
46.93
5.43
13.73
3.92
12.09
6.78
19.86
25.03
9.19
28.60
50.03
5.72
14.45
4.16
12.87
7.21
21.18
26.72
5.33
16.83
29.38
3.31
8.35
2.43
7.55
4.22
12.44
15.71
11.81
36.55
63.98
7.37
18.63
5.33
16.46
9.23
27.06
34.14
Remaining
North-East
Atlantic
4.17
12.83
22.49
2.61
6.57
1.88
5.79
3.25
9.52
12.00
Source: own calculations using emission factors from (den Boer et al., 2011). Damage cost factors
(non-urban) from Table 7.
3.
DEPRECIATION RATES IN DECISION MAKING
In chapter 2 above a review has been presented of different items that should be taken into account for
internalisation of the costs of transportation in general and of maritime transportation in particular.
Costs and benefits, however, occur at different positions along the time axis. This raises the key
question of how to compare these elements i.e. how to account for the dependence on time of the utility
function.
The concept of time-dependence of the utility function in decision making was introduced by
Samuelson (1937) when proposing the discounted-utility (DU) model. Despite Samuelsons clear
reservations about the normative and descriptive validity of the model he had proposed, the DU was
accepted instantly, not only as a valid normative standard for public policies (e.g., in cost-benefit and
cost-effectiveness analyses), but as a descriptively accurate representation of actual behaviour.
A key assumption of the DU model is that all of the disparate motives underlying inter temporal choice
can be condensed into the discount rate.
It is important, however, to distinguish among the varied considerations that underlie inter temporal
choices and to distinguish time discounting from time preference. The first broadly is to encompass
any reason for caring less about a future consequence, including factors that diminish the expected utility
431
generated by a future consequence, such as uncertainty or changing preferences. The second term refers,
more specifically, to the preference for consumption now over delayed consumption. Shane et al. (2002)
demonstrates that many of the studies to reveal time preferences are actually masked by other reasons for
discounting.
3.1
Pure time preferences
Pure time preferences reflect individuals preference for consumption now, rather than later, and are a
central element in economic models used in policy recommendations related to sustainability and climate
change.
Shane et al. (2002) carried out a rather detailed review of the background for discounting pure time
preferences. The paper demonstrates that economists working on this item have been predominantly
focused on peoples perception of time preferences (by using questionnaires), or by studying revealed
preferences in actual decisions.
This approach is therefore not aimed at identifying what would be the rationally and ethically correct
depreciation rate to be used in studies looking into normative decisions that have impacts over long
periods of time, like CO2 emissions or nuclear waste storage, but rather takes a picture of people
perception on the subject.
The DU model, on the other hand, explicitly accounts for time discount only assuming that the overall
value of a sequence of outcomes is equal to the discounted sum of the utilities in each period. The
distribution of utility across time makes no difference beyond that of discounting. It rules out any kind of
preference for patterns of utility over time, e.g. a preference for a flat utility profile over a fluctuating
utility profile with the same discounted utility.
This is clearly in conflict with the precautionary approach (see next paragraph), which assumes that the
next generations have equal or better opportunities than the current.
The studies show similar biases as risk perception studies. For example:
Gains are discounted more than losses (This is exactly as in risk perception studies: risk of loss is
systematically overrated compared to gains)
Small amounts are discounted more than large amounts;
In choices over sequences of outcomes, improving sequences are often preferred to declining
sequences though positive time preference dictates the opposite
In choices over sequences, people seem to prefer spreading consumption over time in a way that
diminishing marginal utility alone cannot explain.
The studies seems to demonstrate that respondents ignore the existence of a capital market
The last point indicates that much of the input used to derive a time preference is just a perception bias.
When people are asked to choose between $100 today and $110 in a year, this can rationally be answered
by considering if the capital market could return more or less than 10%.
3.2
Precautionary approach vs standard economic theory
A classical cost-benefit analysis, in its calculation of the expected present value, will attribute little
importance to a possible future disaster with major impacts. This is because the product of quite high
costs and a low probability could still be a relatively small number. In addition this number will also have
to be discounted. In particular for high discount rates the future thus tends to vanish from the calculations.
This is a reason for scepticism about the treatment of such events by economics, and the development
of policy rules that focus more explicitly on uncertainty, and on avoiding at any cost irreversible
phenomena and potential disasters.
The precautionary approach is a much referenced environmental policy principle cited in international
regulations and policy documents. The most commonly used definition is from the Rio Declaration on
Environment and Development from 1992: In order to protect the environment, the precautionary
approach shall be widely applied by States according to their capabilities. Where there are threats of
serious or irreversible damage, lack of full scientific certainty shall not be used as a reason for postponing
cost-effective measures to prevent environmental degradation.
By including the last part and referring to only cost-effective measures to prevent environmental
degradation, the definition becomes actually unclear, as cost-effectiveness normally must be evaluated
using economic theory. In any case, the most important element in the precautionary principle is that it
shifts the burden of proof in cases where irreversible or serious damage may occur. In case of foreseen
432
un-acceptable consequences to the environment, an action is to be put in place. Cost-effectiveness is only

used to prioritize the different abatement options.
Another difference from standard economic theory is that the environment shall be given the benefit of
the doubt in case of uncertainty.
This is quite contrary to the principle in economic theory where the uncertainty, i.e. the lack of
knowledge and the inherent possibility of reducing uncertainty by gaining more knowledge, is treated as a
(positive) option. This led the committee behind the recommendations formulated in (Norway, 2012) to
conclude When faced with irreversible effects, one should, if the project can be postponed and such
postponement enables new information of relevance to decision making to be gathered, take into account
the (quasi-) option value associated with a wait-and-see alternative. A positive net present value based on
a cost-benefit analysis will not necessarily imply that it is profitable to implement the project
immediately. The expected net present value of implementing the project immediately must also exceed
the option value associated with the wait-and-see alternative.
It may therefore be noted that uncertainties from the viewpoint of economic theory play prejudicially
against a corrective action, while the precautionary approach suggests the opposite.
It may be questioned if some of the assumptions and ideas used in standard economic theory are part of
the problem when dealing with global heating and other environmental effects, rather than a solution.
3.3
Integrated Assessment Models
As mentioned, a subject for debate on inter-generational utility functions is represented by long term
climate changes. Economists often use Integrated Assessment Models (IAM) to analyse this challenge.
Such models combine climatology and economics, and claim to be able to estimate both the costs
associated with global emission-abatement measures and the global welfare effects of the greenhouse gas
emission levels. In principle, IAM models are used to analyse what is an optimal global level of ambition
for climate policy, or to analyse climate policy requirements for reaching a specific target, for example
the agreed two-degree target, see (Randalls, 2010). Many of the IAM models are based on standard
growth theory, where society invests in capital, education and technology, thus refraining from
consumption at present for the purpose of expanding future consumption opportunities.
It is assumed that society will seek to optimise its consumption over time, based on its current and future
consumption needs. Professor William Nordhaus (Yale University) has developed what is generally
believed to be the best IAM model; the so-called DICE model (Dynamic Integrated model of Climate and
Economy: (Nordhaus, 2007). The model expands the traditional growth theory approach. By investing in
natural capital today, future reductions in consumption opportunities as the result of climate-induced
deterioration are limited. In the DICE model, Nordhaus (2007) studies the trade-off between traditional
production investments and climate investments. His findings show that it is profitable to make relatively
large investments in traditional activities in an early phase, which also include improved technologies and
additional intellectual capital, and only some investments in designated climate measures.
He concludes that this can be achieved by introducing general, harmonised CO2 taxes at relatively low
rates in this early phase. In the longer run, the environmental costs are expected to increase as greenhouse
gases accumulate and global warming increases. It will then become profitable to shift investments to
much more aggressive, emission reduction measures. This can be achieved through a steep increase in the
tax rates in the intermediate and long run. This gradual approach to the climate challenge has been termed
the climate policy ramp
Many IAM calculations lead to conclusion that actions may be postponed, with a moderate emission
reduction effort in the near future. This conclusion depends on debatable choices of assumptions made,
including the choice of discount rate.
Whilst Nordhaus (2007) applies a capital-market-based discount rate (5.5%) in his calculations, Stern
(2007) uses a much lower discount rate in his calculations (1.4%). This contributes to his arguing in
favour of starting early with relatively large, global emission cuts, in sharp contrast to Nordhaus (2007)
and many other IAM estimations.
The key point is represented by the comparison between the expenses due to the implementation of
greenhouse gas cutting actions (possibly starting in the near future) and the benefits corresponding to
avoiding serious damages in a medium term future: a different way of discounting affects the results of
such comparison.
Ingham and Ulph (2003) also criticize the fact that the IAM-based computations ignore the possibility
that there may exist tipping points in nature, beyond which additional heating may trigger dramatic and
irreversible processes (criteria usually referred to in the definition of the precautionary approach).
433
Since the main consideration is to safeguard against a potential climate-induced disaster with a low
probability, it is better to base analysis of the climate issue on a pure risk approach than to base it on a
comparison of costs and benefits, as is the approach adopted by most IAM models.
It is also worth noting, (Skjong, 2009), that (IPCC, 2007b) used a totally different approach: the cost
effectiveness of a series of measures was calculated and it was found that all measures that cost less than
$50 for preventing a ton of CO2 equivalent heating effect need to be implemented to reach the 2 degree
target. (IPCC, 2007b) therefore defined a limit value for the CATCH of $50, that was in agreement with a
2C target, which many environmental scientist believe is a tipping point.
The dismal theorem presented by Weitzman (2009) has generated another debate. The theorem
criticises cost-benefit analysis (CBA) of the climate issue, as included in IAM. The basis for the
criticism is that we do not know the probability of very serious consequences from global heating. For
example, the lack of knowledge relating to rapid global heating and large-scale feedback effects on
production and consumption may be non-negligible. Weitzman (2009) also observes that potentially
catastrophic feedback effects from increased CO2 concentrations in the atmosphere are currently omitted
from most IAM models. He also criticises the damage functions used in IAM analyses of the climate
issue. The standard CBA damage function is quite simplistic. It reduces the welfare equivalent of
production in the event of a mean global temperature increase T by a multiplier expressed as (Weitzman,
2011):
M(T)=T2/(1+T2)
(12)
Rather than the catastrophic scenarios described in other literature, such analyses find a relatively
modest effect on world production from large temperature increases. Furthermore, other effects like
biodiversity and health effects are ignored in many models.
Nordhaus (2011) accepts the dismal theorem to be important because it may assist in determining
when extreme outcomes are of relevance. He argues that the theorem is valid only under special
conditions: it assumes strong risk aversion in society, a very fat tail for uncertain variables and,
moreover, that society is unable to learn and act in a timely manner. According to Nordhaus, the dismal
theorem assumes that the marginal utility of consumption grows infinitely large when consumption
approaches zero, as in the case of a disaster. This implies that society will have an unlimited willingness
to pay to avert such a scenario, even considering the low probability. If this assumption is true, the
expected value of the welfare loss will not be infinite, and the premises underpinning the dismal theorem
will be false. That leads us, according to Nordhaus, back to standard cost-benefit analysis, as used by the
many IAM models.
The key question asked by Nordhaus is whether the international community does in fact have an
infinite willingness to pay for avoiding a very low, but non-negligible, probability that the basis for
human existence will be wiped out. He notes, as an example, that the probability of an asteroid hitting the
Earth is about 10-8 per annum. If the dismal theorem was valid, we would, according to Nordhaus, be
willing to pay an unlimited amount for a tiny reduction in this probability. He notes that society does not,
generally speaking, behave as if infinite negative utility is associated with catastrophic outcomes at the
limit. Besides, Nordhaus maintains that the nature of climate change (unlike asteroid disaster) is such as
to give us time to learn, and to postpone the large-scale emission reductions until more effective
technologies have been developed. This is obviously contrary to facts: there are programs in place to
deflect asteroid impacts, but the technology is far from ready. However, in the next 100 years there might
be technology in place, and the probability of the event in the next 100 years is in the order of 10-6. This is
contrary to climate change, where much of the technology is ready for use, and the probability of severe
consequence in the next 100 years is almost 1, not as severe as a big asteroid impact, but still with severe
implications to our welfare.
Pindyck (2011) notes, as did Nordhaus, that Weitzman assumes a utility function exhibiting special
characteristics, especially with regard to societys risk aversion. Pindyck believes that marginal utility
may be very high when consumption approaches zero, but not that it approaches infinity. If we introduce
an upper cap, thus implying that marginal utility approaches a finite level, the willingness to pay will also
be finite, according to Pindyck. He agrees with Weitzman that it is reasonable to assume that the
probability distributions are fat-tailed (see next chapter). He does not deny the possibility of an extreme
climate-generated outcome, and notes that sufficiently fat tails justify swift action without any complex
analysis, but he believes that steep emission reductions may also be justified even if assuming thin tails.
The last must be correct, see next paragraph.
434
Weitzman, (2011) argues that the existence of other potential disasters does not eliminate the special
cause for concern occasioned by climate change. He could also have chosen different specifications for
societys utility function than those adopted in his original analysis. The key point according to Weitzman
is that potentially fat tails should make economists less confident about cost-benefit calculations within
this area!
3.4
Tails of the probability distributions
Economists often refer to fat & thin tail probability distributions when discussing climate change. The
normal distribution, which is much used in economic theory, features a thin tail. A thin-tailed
probability distribution implies that the probability decreases exponentially or faster when moving
away from the expected value. Hence, there is a very low probability of outcomes that are far from the
expected value.
However, many phenomena in economics and elsewhere may have a probability distribution that
deviates from this. One example may be stock market fluctuations. On 19 October 1987, the US stock
market slumped by 23 percent. Nordhaus (2011) observed that if the stock market had adhered to a
normal distribution, we would have observed a 5-percent change in prices only once every 14,000 years.
However, historical stock market data show that major fluctuations happen much more frequently than
would be indicated by a normal distribution. This is indicative of a probability distribution with fat tails.
The reasons are better explained and understood by physicists/mathematicians working in economics by
use of Agent Based Models, (OECD, 2011), (Turner, 2011).
Classical economists observe that extreme outcomes will not be very improbable under fat-tail
assumptions. The problem with this debate amongst economists when debating climate change, is that the
predicted consequences are actually around the expected value of the prediction, so the debate about the
tails is rather irrelevant and brings the debate out of focus.
3.5
Role of the discounting rate
As above recalled, the discount rate is a very critical parameter in cost-benefit and cost-effectiveness
analysis when costs and benefits are different in their distribution over time, especially when they occur
over a long time period. For example in the debate about climate change the horizon may be hundreds
or even thousands of years. As an illustration: 1/1.01100=0.369 and 1/1.05100=0.0029, a factor difference
of 125.A very high discount rate implies that the far future (both costs and benefits) has no impact on a
decision now, and will therefore often be in conflict with a precautionary approach. The best
illustration of the extreme importance is illustrated by the Stern-Nordhaus debate. It is now well
understood that it was a low discount rate that drove the policy conclusions of the Stern report, as
opposed to previous cost-benefit analyses of global heating. Sterns cost-benefit analysis of global
heating assumed a real discount rate of 1.4% and concluded there was a case for immediate action.
Stern used a rather unique argument for setting the depreciation rate for true time preferences
exceedingly low. He argued that the only reason for this depreciation was the likelihood that
humankind went extinct and was not present to enjoy future consumption. Stern interprets the
depreciation factor e-t as the probability that the world exists at that future time, because this is the
probability of survival that would apply if the destruction of the world was the first event in a Poisson
process with parameter . Nordhaus assumed a 5.5% discount rate, and favoured a modest carbon price,
increasing over time. The different assumptions of discount rates could explain the whole difference in
conclusions. This was verified by Nordhaus, when he ran his DICE model using Sterns discount rates,
he got similar results to Stern (Nordhaus, 2007).
There are essentially two ways of deriving discount rates:
Based on comparing to opportunity costs, and use of the Capital Value theory
Based on the Ramsey equation
The logic of comparing with opportunity costs is based on the idea that cost-benefit analysis normally
uses willingness to pay studies to measure benefits. The opportunity cost is the value that the alternative
would have produced. Identifying the costs and benefits of a change involves comparing the result of the
proposed change to outcomes without the change (often referred to as the zero option, or the business as
usual). Analysts typically observe the value people place on something by observing their willingness to
pay, for example by observing market behaviour. So, logically a decision should be made in favour of the
change if it is better than without change. This is reflected in the opportunity cost. Obviously, for issues
435
like climate change, the question may be said to relate to the observable market behaviour or the stated
willingness to pay.
In finance, the capital value theory is used to determine a theoretically appropriate required rate of
return of an asset, if that asset is to be added to an already well-diversified portfolio, given that assets
non-diversifiable risk. The model takes into account the assets sensitivity to non-diversifiable risk (also
known as systematic risk or market risk), often represented by the quantity beta () in the financial
industry, as well as the expected return of the market and the expected return of a theoretical risk-free
asset. The market reward-to-risk ratio is effectively the market risk premium
E(Ri) = Rf + i (E(Rm) Rf)
(13)
where E(Ri) is the expected return on the assets, E(Rm) is the expected return in the market, Rf is the risk
free return and i = Cov (Ri, Rm)/Var (Rm)
The Social Time Preference Rate, derived by the Ramsey approach, represented by r, is the sum of
these two components, i.e.
r=+g
(14)
The Social Time Preference Rate has two components:
The rate at which individuals discount future consumption over present consumption, on the
assumption that no change in pro capite consumption is expected, represented by
An additional element, if pro capite consumption is expected to grow over time, reflecting the fact
that these circumstances imply future consumption will be plentiful relative to the current position
and thus have lower marginal utility. This effect is represented by the product of the annual growth in
pro capite consumption (g) and the elasticity of marginal utility of consumption () with respect to
utility, which is defined as the percentage fall in the marginal utility when consumption increases by
one per cent.
Each element of Social Time Preference Rate is examined in turn below.
Estimates of : This comprises two elements:
Catastrophe risk premium (); and

Pure time preference ().
The first component, catastrophe risk, is the likelihood that there will be some event so devastating that
all returns from policies, programmes or projects are eliminated, or at least radically and unpredictably
altered. Examples are natural disasters, major wars etc. Newbery (1992) estimates as 1.0, Kula (1987)
as 1.2, Pearce and Ulph (1995) as 1.2, OXERA (2002) as 1.1 currently and 1 in the near future. It should
be noted that in risk analysis, the risk component is explicitly quantified. A risk premium should therefore
not be included in the discounting, as this would account for the same effect twice. Alternatively a risk
premium may be calculated from the risk model (Skjong and Lereim, 1988).
The second component, pure time preference, reflects individuals preference for consumption now,
rather than later, assuming an unchanging level of consumption pro capite over time. In (Scott, 1977)
and (Scott, 1989) is estimated as 0.5. Other literature suggests it lies between 0.0 and 0.5. However, if
zero, this implies pure time preference does not exist, which is not regarded as plausible by most
economists.
The UK Green Book (UK, 2011) concludes that the evidence suggests that these two components
indicate a value for of around 1.5 per cent a year for the near future. Scott (1977) derives a central
estimate value of 1.5 from past long-term returns on savings. A later estimate in (Scott, 1989) updated
this estimate to 1.3. However, this was based on United States, as well as UK, evidence. OXERA (2002)
estimates to lie between 1.0 and 1.6.
Estimates of : According to (UK, 2011) the available evidence suggests the elasticity of the marginal
utility of consumption () is around 1. Pearce and Ulph (2005) estimate a range from 0.7 to 1.5 with 1.0
being considered defensible. Cowell and Gardiner (1999) Cowell and Gardiner (1999) estimate as being
just below or just above 1.0; OXERA (2002) estimate a range from 0.8 to 1.1.This implies that a marginal
increment in consumption to a generation that has twice the consumption of the current generation will
reduce the utility by half.
436

Table 10. Discount Rates Derived by Ramsey Formula.
Reference
Stern (2007)
Quiggin(2006)
Cline (1993)
Garnaut (2008)
Harrison (2010)
Nordhaus (2007)
Weitzman (2007)
Arrow (2007)
Dasgupta (2006)
Gollier (2006)
UK (2011)
0.1
0
0
0
0-3
1.5
2
0
0
0
1.5
1
1
1.5
1-2
0.2-4
2
2
2-3
2-4
2-4
1
g
1.3
1.5
1
1.3
1.2-2.1
2
2
1-2
1-2
1.3
2
r
1.4
1.5
1.5
1.3-2.6
0.24-11.4
5.5
6
2-6
2-8
2.6-5.2
3.5
Harrison (2010) also gives an overview of discount rates used by governments and governmental
agencies. This is reproduced in Table 11, with some updates since 2010.
Table 11. Discount rates in national guidelines.
Country
Philippines
India
Pakistan
Developing Banks
New Zeeland
Canada
China
South Africa
USA
USA
European Union
Italy
The Netherlands
France
UK (*)
Norway(*)
Germany
Agency
World Bank
Asia Development Bank
Inter-American Development Bank
European Bank for
Reconstruction and Development
African Development Bank
Treasury and Ministry of finance
Treasury Board
Office of Management and Budget

EPA
European Commission
Central Guidance to Regional
Authorities
Ministry of Finance
Commissariat General du Plan
HM Treasury
Norway (2012)
Federal Finance ministry
Rate
15%
12%
12%
10-12%
10-12%
10%
10%
10%
8% (10% 82-2008)
8% (10%, 76-2007)
8%
8%
7% (10%, < 1992)
2-3% (Test 7%)
5% (6%, 2001-2006)
5%
4% (risk free)
4% (8%, 85-2005)
3.5% (see Table )
4% (See Table)
3% (4%, 99-2004)
(*) Updated as compared to Harrison (2010)

Estimates of g: Maddison (2001) shows growth pro capite in UK to be 2.1 per cent over the period 1950
to 1998. Surveying the evidence, the UK (2013) Treasury paper Trend Growth: Recent Developments
and Prospects also suggest a figure of 2.1 per cent for output growth to be reasonable. This estimate
removes the impact of net migration. In UK (2011) the annual rate of g is therefore put at 2 per cent per
year.
The calculated Social Time Preference Rate by UK (2011) is therefore:1.5% + 1 2% = 3.5%.
These and other results from use of Ramseys formula may be found in Table 10.
Where the appraisal of a proposal depends materially upon the discounting of effects in the very long
term the UK (2011) view is that a lower discount rate for the longer term (beyond 30 years) should be
used. The main rationale for declining long-term discount rates results from uncertainty about the future.
This uncertainty can be shown to cause declining discount rates over time, Weitzman (2001) and Gollier
(2002). In light of this evidence, UK (2011) recommends that for costs and benefits accruing more than
30 years into the future, appraisers use the schedule of discount rates provided in Table 12.
437
Table 12. Long Term Discount Rates (UK).
Years
Discount
rate
0-30
31-75
76-125
126-200
201-300
301-
3.5%
3.0%
2.5%
2.0%
1,5%
1%
The recommendation in Norway (2012) is slightly simpler, but also suggests a declining rate, see
Table 13.
Table 13. Long Term Discount Rates (Norway).
Years
Risk free Rate
Risk Premium
Risk Adjusted Rate
0-40
2.5
1.5
4
40-75
2
1
3
75-
2
0
2
From 1967 to 1999, Norway used an approach where the discount rate was based on the Ramsey
model. Based on a report in 1967, in which it was assumed that = 1 percent, = 3 and g = 3 percent,
the discount rate was put at 10 percent in a government circular from 1975. In a circular from 1978 the
discount rate was changed to 7 percent. When revising the cost-benefit analysis framework in 1998, the
assumption of a small, open economy was held to be reasonable for Norway. Therefore, the risk-free
interest rate and risk premiums were deemed to be increasingly determined by international markets.
Hence, estimates of the relevant discount rate were based on the Capital Asset Pricing Model. The
Ministry of Finance guides from 2000 and 2005 discuss, in more detail, how the model may be adapted
to provide an expression for a reasonable discount rate for use in cost-benefit analysis. A circular in
1999 stipulates a risk-free real rate of 3.5% should be assumed in cost-benefit analysis.
3.6
Conclusion (depreciation rates)
It is observed that there are many diverging views on depreciation rates to be used in cost-benefit and cost
effectiveness studies. The assumptions have very large consequences on policy decisions where the effect
of the decision has long term effects, like the greenhouse gas emission, and CO2 emissions in particular,
where emission now have effects over periods of thousands of years.
However, it is observed that there is some tendency that the depreciation rates are now typically
recommended at around 3% in developed economies. In some of the fast growing economies the discount
rates are much higher, which can be easily justified by the Capital Value Model. The challenge with
global polices for greenhouse gas emission reduction can be understood in this context, because at these
high rates, few abatement technologies are cost effective, and policy recommendations from the models
will tend to be wait-and-see rather than a precautionary approach.
It should also be observed that the approach used in IPCC(2007) is less dependent on these assumption,
as it makes a direct link between the politically agreed 2C target and the cost effectiveness of available
abatement technologies.
4.
EXAMPLES RELATED TO SUSTAINABILITY ORIENTED DESIGN
As described in Chapter 1, general aspects such as economic, human and environmental terms should be
considered for sustainabe design. In this chapter, recent works related to risk-based design are firstly
reviewed, which are mainly related with the human aspects (fatality) and are applied to regulatory
process. Later, lifecycle design examples including evaluation of environmental aspect and adaptation to
climate change are considered.
4.1
Probability based design
In this section, recent applications of probability based design are reviewed: these applications, however,
do not cover the whole spectrum of consequences included in the concept of sustainability, but
concentrate on single aspects.
Vassalos (2012) reported the recent history of the Design for Safety (Risk-Based Design) concept and
presented implementation examples, highlighting challenges opportunities of the concept. An applications
to contemporary regulatory development, the IMO Passenger Ship Safety, is reported. Modern safety
expectations are expressed as a set of specific safety goals and objectives, addressing design (prevention),
operation (mitigation) and decision making in emergency situations with an overarching safety goal,
corresponding to no loss of human life due to ship related accidents.
438
The concept of alternative and/or equivalent design, more recently summarised in (IMO, 2013a), is
exemplified with an application to fire safety regulation, in which designs not strictly complying with the
existing prescriptive fire safety regulations can be accepted, provided that such designs can be shown to
be at least as safe as the design made in accordance with the conventional rules by using the risk-based
approach. In the risk analysis, example of evacuation simulations by which the number of fatalities in the
fire scenarios is computed, and the dynamic flooding/survivability simulation is used to evaluate the
consequence of the flooding scenarios.
In (Vassalos, 2012), the concept of the Safety Management System (SMS) is discussed, too. SMS is a
lifecycle process, starting at the concept design stage and continuing throughout the life of the vessel. In
this process, safety must be continuously monitored and reviewed to ensure developments/changes in the
design, and its operations are reflected in the way safety is managed. As an example of life-cycle safety
management system, the development of iStand is described, a Decision Support System (DSS) installed
on-board with general features such as: real time sensors and hardware integration, vulnerability log,
critical assessment, corrective action search etc.
In the short report by Andrews et al. (2012), the risk-based design approach is seen as a classical
application of the holistic/multi-objective ship design concept, where in addition to safety issues
represented by a variety of hazards, economic and environmental impact aspects are jointly optimized for
best design solutions.
Boulougouris and Papanikolaou (2013) reported the application of risk-based design to combat ships.
They used risk-based assessment methods to the evaluation of the survivability of surface combatants
after damage. In this case, damage stability criteria are coupled to a probabilistic model of the damage
due to weapon impact. For the estimation of the probability of survival after damage in waves, two
methods are applied: the quasi-static approach, in which semi-empirical deterministic criteria are used
and a full dynamic flooding/capsizing approach in which a time domain flooding and/or capsizing
simulation code is used. Also, the use of the flooding simulation by the pressure-correction method
implemented in a commercial code is reported. The introduced method is applied to the assessment of the
damaged stability of a generic frigate operating in specified seaway conditions, and typical results of this
type of study are presented.
Given the ever increased amount of historical records with improved accessibility, a situation of
flood with data, but difficult to harvest is becoming a common situation, when trying to derive
statistics from accident/incident data. Cai and Konovessis (2012) elaborates on a method for integrating
the effort devoted to marine accident investigation into a risk-based design framework. In this concept,
basic ship design parameters and operational parameters which affect the safety performances are
firstly investigated. Then the historical accidental events are transformed into the accident database by
using these parameters. Next, a risk model is built, using a Bayesian Belief Network (BBN) based on
the database. The parameters in the risk model are defined by using learning techniques such as
Bayesian Structure Learning or Bayesian Parameters Learning. The model is then used to identify a
risk-based optimal design solution. The concept is demonstrated through an investigation on fire safety
onboard passenger ships. Even though the demonstration shown in this paper is very simple, the
proposed concept is very interesting in that they proposed the development of accident database
system, subsequent model training, and application in risk analysis, for the implementation of riskbased ship design.
Even though a risk-based approach has been applied a few times in the rule-making process or for
novel ship designs, there are difficulties and challenges in applying RBD concepts to conventional ship
designs. Lee et al. (2012) discuss how to meet these challenges. In their paper, the existing concepts and
methodologies for risk-based design of ships and marine systems are reviewed. Next, difficulties in
applying these concepts in a practical design environment are discussed, together with possible ways of
overcoming such difficulties. The potentiality of the approach are shown in an application to the pilot
design of a new concept container vessel and of CO2 carriers. In their conclusions the authors note that
the standardization of Risk-based ship design is necessary to establish the RBD concept in practical
ship design.
Recently, applications of risk-based design methods have been proposed and developed in various EU
funded research projects. NMRI (National Maritime Research Institute) in Japan also carried out a
research on development of fatality risk evaluation method as a cooperative research with the Japan
Shipbuilding Research Association, and now continue to carry out researches on more advanced riskbased design methods. Kaneko et al. (2012) reported about present status and future plan of the research
project related to further enhancement of risk-based design in NMRI covering the items below.
439
Accident scenarios and estimation of their frequencies

In this part of the research, improvement of HAZID method, which is the first step of risk assessment for
identification of possible hazards relevant to the considered ship, is discussed for more suitable
development of accident scenarios. In particular, typical form of the HAZID sheet and the HAZID
database as well as the use of historical accident data is discussed, by which efficient hazard identification
and development of accident scenarios might be possible. Moreover, the use of the field measurement
data such as AIS data for estimation of frequency of initiating events (collision situation) is reported with
some examples of ship trajectories.
Consequence analysis method
This part of the investigation focuses on flooding and fire as main causes of serious accidents with
fatalities and, accordingly, on simulation programs of fire escalation, flooding (in the time domain) and
evacuation for consequences analysis in risk evaluation. In the paper, use of software for fire propagation
and evacuation simulations is reviewed and discussed with some results. Plans for the development of
CFD simulations of HNS (Hazardous and Noxious Substances) and oil spillage out through breaches in a
tanker hull are also reported.
Data conversion programs
The paper also discusses about the necessity of data conversion programs, for example from internal data
of design tools into input format for programs carrying out consequence analysis.
Validation of the approval procedures
When a ship is designed by prescriptive methods (rule-based design), authorities such as Classification
Societies can easily approve the design based on the existing rules. On the other hand, in the approval
process of a ship by risk-based design, accident scenarios, frequencies, probabilities and consequences
should be validated by the designer and by the approval body. In addition, a verification of the simulation
and computation programs used in risk evaluation should be carried out. In the paper it was commented
that approval authorities carry out this activity directly and that open source programs are desirable for
the verification of simulations. Another practical example of risk-based design, is reported in (Choung
et al., 2012) where they carried out a specific risk-based design using Structural Reliability Analysis
(SRA). The structural safety assessment of a longitudinal stiffener of the outer shell of a double hull
tanker is carried out. A first yielding limit state function in bending is defined for the local supporting
member subjected to lateral pressure and hull girder longitudinal stress. A probabilistic model of design
variables is set, based on the literature review. Then, the probability of failure is computed using a
commercial reliability program for both full-load and normal ballast conditions, at different heading
angles. Based on the results of SRA, a Design Safety Level (DSL) is evaluated as a combination of the
computed probabilities and compared to a Target Safety Level (TSL) defined according to (Denmark
et al., 2004a), assuming no serious failure consequences and a ductile failure with no reserve capacity
as failure development category. Finally, the assessment results for design modification cases (different
stiffener dimension) are reported to show the validity of the proposed procedure.
It is here noted that a more updated source for the definition of target probabilities of failure is
provided in (IACS, 2006b)
4.2
Lifecycle Design
For a sustainable design of ship and marine structures, it is important to evaluate also the environmental
impact in the lifecycle of the structures themselves. (Aspen et al., 2012) reported current status,
opportunities and barriers for integrating the lifecycle analysis in ship design processes. Life cycle
assessment (LCA) and life cycle costing (LCC) concepts are reviewed in this paper. The authors account
for the adverse environmental effects from the entire lifecycle of products and processes considering raw
material extraction, processing, product manufacture, transportation, operation, maintenance, repair and
scrapping. In the paper, the application of the concept to vessel designs is discussed, and different LCA
software programs which can be used for the purpose are reviewed. Next, in order to demonstrate the
applicability of the concept, a case study of two longline fishing vessels is reported: a conventional vessel
is compared to a newer version where innovative solutions (e.g. the moonpool technology) are employed.
Results were related to the functional unit of 1 kg of fish captured and brought to shore for further
440
processing. The analysis uses four environmental indicators from the CML (Institute of Environmental
Sciences at Leiden University) Baseline characterization method:
Acidification potential (AP) (Unit: Kg SO2 Eq)

Global warming potential (GWP 100a) (Unit: Kg CO2 Eq)
Photochemical oxidation potential (POCP) (Unit: Kg Ethylene Eq)
Stratospheric ozone depletion potential (ODP) (Unit: Kg CFC-11 Eq)
In addition, weights are added in the various inventory flows based on the willingness to pay to avoid the
various impacts from the lifecycle of the product, so that it becomes possible to produce a unique
indicator that shows environmental impacts as damage costs. As a result, a difference in eco-efficiency
shows the advantage of the new vessel. The paper further discusses the relationship between the
framework of environmental regulation and the LCA (environmental impact indicators). Difficulties in
introducing the LCA and LCC into marine systems design are discussed, related to the definition of the
functional unit, system boundaries and data collection.
Environmental performance is now becoming an important subject in the maritime industries.
However, ship designers may not be able to find the appropriate tools to evaluate the environmental
performances during the lifecycle of the ships and marine structures. Even though tools are found, an
holistic approach is still missing. Fet et al. (2012) and Fet et al. (2013) report about existing
environmental assessment tools and introduce systems engineering as an approach to holistic life cycle
design. Figure 5 shows the environmental performance improvement tools classified by the scope of
temporal concern and of environmental concern.
Figure 5. A Classification of methods and tools for environmental performance improvement [Fet et al, 2012].
The inner smallest area is related to production processes, for which CP (Cleaner Production, in the
narrow sense), EAc (Environmental Accounting) and I/O (Input-Output Analysis) are appropriate tools.
The next area is related to products and their life cycles: here appropriate tools are LCA (Life Cycle
Assessment), LCC (Life Cycle Costing), EEA (Eco-Efficiency Assessment), CFP (Carbon Footprint of
Products), WFP (Water Footprint of Products), EPD (Environmental Product Declaration) and CP. Area 3
represents the company lifetime: EMS (Environmental Management System), EA (Environmental
Auditing), EPE (Environmental Performance Evaluation), EEA, GHG management are relevant. To
achieve a sustainable development in a long-term perspective, policy programs and international
regulations are necessary, which are illustrated with the outermost circle in the figure. The paper also
describes the advantage to introduce System Engineering (SE) and shows how environment impact
assessment tools can be used in the SE process. The authors conclude that the SE has proved to be a
systematic tool for a methodology for Life Cycle Management (LCM), and can be used as the framework
for design for environment.
In conventional ship design, the ship structure is usually designed to minimize the overall weight of the
vessel, meeting constraints such as strength requirement introduced by regulations and standards. This
approach, however, does not take into account the maintenance and repair costs incurred by the
shipowner in the lifecycle of the vessel. (Temple and Collete, 2013) assessed year-by-year costs of a
naval vessel over her service life to determine a lifecycle maintenance cost value for the ship. They use a
lifetime structural model that captures the effect of degradation due to corrosion and fatigue. For fatigue
model, a probabilistic formulation based on a classical S-N approach is used, by which the probability
441
that the fatigue crack is found is estimated and the corresponding fatigue costs can be calculated
considering the repair cost. For corrosion damage, a time-dependent wastage model is used to estimate
the thickness reduction in each year of the vessels service life. Corrosion costs due to plates replacement
because of local corrosion and due to degradation in the overall global strength are considered. The model
includes also scheduled maintenance, unplanned-maintenance and costs associated with dry-docking to
carry out repairs. An optimization using MOGA (Multi-Objective Genetic Algorithm) is adopted to
minimize both the structural volume and the lifetime maintenance cost for the mid-ship section of a
nominal DTMB-51 hull form. The scantlings of the panels in different functional locations are used as
design variables of the optimization. By comparing the optimum solutions on a typical Pareto frontier, the
trade-offs between structural weight and lifecycle maintenance cost are investigated. For example, as the
design become lighter, the fatigue costs increase from near zero to a significant part of the lifetime
maintenance costs for the vessel. The conclusion is that, by incorporating the lifetime maintenance costs
into the initial evaluation of the structural design, it is possible to optimise the performance over the
whole operational life of the vessel.
Based on the concept of holistic risk evaluation, already outlined in preceding issues of the report
of committee IV.1 (Design Principle and Criteria) of ISSC (Brunner et al., 2012) (Moore et al.,
2009), a lifecycle structural optimization of the mid-ship section of a double hull tanker was carried
out in (Kawamura et al., 2013). They defined the optimization problem as a maximization of the
life-cycle benefit (LCB) as shown below.
LCB = ROPE LCC
(15)
where ROPE is the lifecycle revenue and LCC the lifecycle cost, evaluated as:
tlife
LCC = (COPEi + CINSi ) + CM + CRISK + CI

i =1
(16)
where COPEi is the annual operational cost for i-th ship-year, CINSi is the insurance cost per year, CM
is the maintenance cost, and CI is the construction cost. It is noted that the lifecycle risk of the ship
(CRISK) is evaluated as a part of the objective function (LCB) in this optimization. Within the
lifecycle risk, the components due to CO2 emissions, to the oil-outflow risk and to the risk of
structural failure are considered. CO2 emissions are evaluated by life cycle assessment (LCA), while
possible oil-outflow consequences are computed according to the MARPOL convention. The risk of
failure is evaluated by structural reliability analysis (SRA) of the longitudinal strength of the ship,
adopted to compute the probability of failure. Based on these risk models, various single objective
optimization problems are solved by genetic algorithm, maximising the LCB considering (a) only
the risk of oil outflow, (b) only the impact of CO2 emissions, (c) only the risk of failure. Finally
LCB has been maximised considering all the contributions (holistic risk). The optimal solutions of
these separate problems are compared to discuss the influence of each objective function and to
show the meaning of including an holistic perspective in the design.
4.3
Lifecycle Design Considering Future Climate Change
Despite the efforts part of the world is making to limit this effect, according to the Intergovernmental
Panel of Climate Change (IPCC), the globe is definitely experiencing a climate change. One of the
implications is that frequencies and intensities of extreme weather events are likely to increase. For an
effective lifecycle design of ships and marine structures, it is crucial to estimate future trends in the metocean conditions. A possible increase in significant wave height (SWH), f.i., would in fact increase waveinduced structural loads to a higher level than in the present situation.
Bitner-Gregersen et al., (2013) published a very valuable monograph about climate change, in
particular wave climate and its impact on ship and offshore structural design. They reviewed IPCC
reports and recent papers related to climate change, and summarize the prediction of future changes
in storminess, wind and waves. Their conclusions about the trends of wave climate are as follows.
There is a consensus about an increase in significant wave height from the middle of the 20th century to
the early 21st century in the northern hemisphere winter, at high latitudes in the North Atlantic and the
North Pacific, while there has been a decrease at souther latitudes of the northern hemisphere. The
increase of the 99-percentile significant wave height (SWH) has been observed to be up to 0.5% per year.
However, if records are extended back to late 19th century, the picture changes, as studies show that
442
storminess and wave heights in the late 19th/early 20th centuries were about the same as near the end of the
20th century. Thus, it is unclear if the increase observed during the last 45 decades is caused by
anthropogenic climate changes or is just a manifestation of long-term natural variability. Regional
increases in wind speeds and wave heights are observed, too, more pronounced in extreme values than in
the average ones. The increase in the 20-years return period SWH is generally in the range of 0.5-1.0 m in
the North and Norwegian Seas, immediately west of the British Isles, off the northwest of Africa, around
30N from the east coast of the United States to 50W and in the Pacific between 25N and 40N and
from the west coast of the United States to 170W. Increases up to 18% for the 99th percentile SWH have
been reported for the southern North Sea. There are indications that the increase in extreme wave heights
may reach more than 10% above present day extremes in some areas. Projections are however influenced
by: choice of climate model, emission scenario and downscaling method for waves. The uncertainties
affecting the estimated increase is of the same order of the estimates itself, so they should not be ignored
when impacts of climate change on design and operation of ship and offshore structures are considered.
The authors propose to deal with climate changes as shown in Figure 6, in which changes in metocean conditions and relevant uncertainties are integrated as a part of the risk based design.
Figure 6. Position of climate changes in a Risk-based approach (Bitner-Gregersen et al., 2013).
Finally, based on the analysis of wave climate change, they investigated the potential impacts of
climate change on a tanker design by using structural reliability analysis (SRA). A limit state
function for hull girder collapse is formulated as follows:
g (X ) = M U X R (M WV X st X nl + M SW X sw )
(17)
where MSW is the random still water bending moment, MWV is the random wave bending moment, Mu
is the random ultimate capacity while XR, Xst, & Xnl and Xsw represent model uncertainty factors for
the capacity, wave-induced bending moment and still water bending moment, respectively.
Randomness of MWV is evaluated by analysing structural response (RAO) using hydrodynamic
analysis. Short- and long-term response analyses were carried out to get annual extreme value
distribution. When evaluating short-term response, the increase of the extreme SWH (0.5, 1.0 and
2.0 m) corresponding to wave climate change is considered in this study. The results of SRA are
compared with those of the base case without increase of SWH. It is concluded that observed and
projected changes in climate will have large effects on tanker design practice. The presented results
show that, in order to maintain the safety level, the steel weight of the deck for net scantlings should
be increased by 58% if the extreme SWH increases by 1 m.
The authors also conclude that, in case of climate changes leading to more extreme weather, rules
for tankers would need revisions in order to maintain the same structural reliability level. This could
be done either by revising the IACS formula for the characteristic wave bending moment or by
increasing the partial safety factor for the wave bending moment. Alternatively, one could consider
443
adopting direct calculation of the characteristic wave bending moment and applying the
environmental model inclusive of climate changes in the calculation.
Vanem and Bitner-Gregersen (2012) presented a Bayesian hierarchical space-time stochastic model in
order to predict future trends of the significant wave height. They used the database of the ERA-40
project, including data of significant wave height sampled every 6th hour from 1958 to 2002 in the North
Atlantic Ocean. By such data, the stochastic time-dependent model of the significant wave heights is
developed, including long-term temporal trend, seasonal components, conditionally dependent term of the
nearest neighbour locations, etc. Bayesian inference and MCMC (Markov Chain Monte Carlo) techniques
are used to identify the parameters in the model. It is noted that not only the original data, but also the
log-transformed data are applied to construct the stochastic model. From the model, an increasing trend in
the significant wave height is observed in the recent 50 years. It is possible to extrapolate these
oservations into the future to predict future trends, even though the predicted results might be uncertain. It
is reported that an increase in the monthly maximum significant wave height of about 1.6m in 100 years
is predicted by using the proposed model with the original data. An increase of 10-15% in 100 years is
expected by using the log-transformed data.
Further, the relation between such results and the calculation of future environmental loads and
responses is discussed in the paper. The long-term trends obtained by the prediction are incorporated into
the joint distribution of significant wave height and wave period based on the ERAinterim data. Concretely,
it is assumed that the joint distribution ( f H S ,T (h, t ) ) can be constructed by the product of a 3-parameter
Weibul distribution of significant wave height ( f H S (h ) ) and a log-normal conditional distribution of
wave period f T |H S (t ) as shown in Eq. 18 below
f H S ,T (h, t ) = f H S (h ) f T |H S (t )
(18)
To predict the shape of the joint distribution function in the future, the location parameter and the scale
parameter of the 3-parameter Weibul distribution are modified using the predicted trend contribution of
significant wave height (in this paper ct = 1.6m and ct = 0.39m is used as an example), while the
log-normal distribution is not assumed to change due to the climatic trends.
Figure 7 shows the environmental contour lines of H S and TZ for the North Atlantic location with
original and modified data. As expected, the modification of the distribution of significant wave height
moves the contours to the right. Further, it is observed that the contours are narrowed and the maximum
1-, 10- and 25- return H S and related TZ are increased.
Finally, the 25-years stress amplitude and response period for a considered oil tanker have been
calculated in the paper incorporating the effect of the climatic trend. A significant increase in the 25-year
stress amplitude (7%) is shown, with an increase in the zero-crossing period of 2%. The authors conclude
that even though the presented results are used for illustration purposes only and any evaluation would be
unavoidably affected by uncertainties, the effect of climatic changes in the wave climate are not
negligible and might have impact on load and response calculation of floating structures.
Figure 7 Projections of the impact of climate changes on contour lines for sea state characteristics prediction
(Vanem and Bitner-Gregersen, 2012)
5.
REGULATORY FRAMEWORK FOR MARINE STRUCTURES
Some of the recent implementations of the regulatory framework at IMO will be recalled in the following.
The latest developments of goal based standards will be covered first, while later a short review of the
444
ongoing evolution of requirements targeting particular environmental aspects will be presented. Finally, a
few notes regarding the normative framework of other ship activities will be reported.
The latest IMO strategic plan for the period 2014 to 2019 reiterates that the mission of the IMO is to
enhance not only the safety and security of shipping but also the sustainability and environmental impact
of shipping: The mission of the International Maritime Organization (IMO), as a United Nations
specialised agency, is to promote safe, secure, environmentally sound, efficient, and sustainable shipping
through cooperation. This will be accomplished by adopting the highest practicable standards of maritime
safety and security, efficiency of navigation and prevention and control of pollution from ships, as well as
through consideration of the related legal matters and effective implementation of the IMOs instruments,
with a view to their universal and uniform application (IMO, 2014d)
It is clear that the IMO intends to continue along the path set over previous periods identifying those
effects from shipping that adversely affect the environment whilst ensuring that shipping is able to
develop further sustainable practices.
5.1
Development of Goal Based Standards at IMO
IMO, 2011e listed, in the High Level Action Plan and Priorities for the 2012-2013 biennium, the High
Level Action (HLA) 10.0.1 Further develop measures to apply goal-based standards for maritime safety
and environmental protection including the two planned outputs:
10.0.1.1 Implementation of goal-based new ship construction standards for tankers and bulk carriers
10.0.1.2 Development of goal-based ship construction standards for all types of ships, including safety,
security and protection of the marine environment
This re-proposed the situation already analysed in the previous term of this committee, characterised by
two types of initiatives at IMO in the field of Ship Construction Standards, both related to the term Goal
Based Standards (GBS) and in line with the concept it represents. GBS is in fact a term that relates to how
IMO wants to structure and justify the maritime regulatory system in general. The reference is to a
framework that reports with a top-down approach overall goals, objectives, general and specific
requirements organised in a hierarchical, coherent and transparent way.
While the GB way of organising the framework is specifically devoted to enforce the qualities of
transparency and coherence, it does not imply a specific choice on how the various objectives
(particularly those at upper levels) are set and, consequently, how the coeherence between the various
levels should be enforced (and checked).
IMO, 2011d is the most updated IMO document containing GBS Guidelines. In the title such
Guidelines are defined as generic. Quoting from paragraph 2: It should be noted that these Guidelines
are generic and where they use phrases such as required level of safety, this does not imply any
preference for a specific technical approach.
This explain why two different targets have been sought in the last years at IMO under the same term
of GBS:
the development of Construction Standards for Bulkers and Tankers (GBS/B&T), first separately for
the two class of ships and, more recently, for the two classes together. In this case, the formulation of
goals and their verification are carried out with a deterministic approach
the development of a procedure for the formulation of Construction Standards for ships based on risk
evaluation (GBS/SLA: safety level approach). In this case, the formulation of goals and and their
verification is carried out with probabilistic methods
In the following part of the chapter, the recent developments of IMO activity in these two directions
will be briefly summarised
5.1.1
IACS Harmonized Common Structural Rules for Bulk Carriers and Tankers
The IACS Harmonized Common Structural Rules for bulk carriers and oil tankers (CSR-H) (IACS,
2014a) are a declination of the goal based standards at the level of tier IV. The Harmonised Structural
Rules are now in the audit process at IMO: accordingly, satisfaction of the goals and especially of
functional requirements is being verified.
The IACS CSR-H rules (IACS, 2014a) were adopted on December 18th, 2013. They will enter into
force on 1st July 2015 replacing the IACS Common Structural Rules for double hull oil tankers, (IACS,
2012b) and the IACS Common Structural Rules for bulk carriers, (IACS, 2012a).
The IACS CSR-H Rules are divided in 2 parts. Part one contains common requirements to both vessel
types (oil tankers and bulk carries). Part two contains all requirements that are ship type specific.
445
In order to understand the approach used in the CSR-H rules, IACS published the technical
background information (IACS, 2014b). The document is based on the same main structure as the CSR-H
rules.
A consequence assessment (CA) has been carried out by IACS members to assess the impact of the
new CSR-H rules on the design of vessels. (IACS, 2014c) summarises the analyses and results of the
CA. The CA was carried out on the following oil tanker types: VLCC, Suezmax, Aframax, Panamax,
Handymax and different bulk carrier types: Capesize, Babycape, Panamax, Handymax. The design of
the sample vessels has not been altered. Same materials and structure element arrangement were used.
The CA analysis was carried out for the two ship types (bulkers and tankers) respectively against the
common structural rules for bulk carriers (CSR-BC) and the common structural rules for oil tankers
(CSR-OT).
The analyses covered the local scantling requirements (yielding, buckling and simplified fatigue
assessment), hull girder ultimate residual strength and direct strength assessment (yielding, buckling and
fatigue assessment). The general result of the CA is that, for both ship types in all sizes of vessels, in most
areas of the hull structure the scantlings according to CSR-OT and CSR-BC do not satisfy the
requirements of the new CSR-H rules. Vessels dimensioned according to the CSR-H will have increased
scantlings and weight.
The CSR-H rules were developed in the main frame of the goals based standards (GBS). To demonstrate
that top-level goals and objectives have been met, the Rules were developed using a hierarchical
framework as shown in Figure 8. The framework of the Rules represents a top-down approach that
provides transparency and ensures that structural requirements developed reflect the overall objectives
(IACS, 2014d).
The issues of the Rule framework are addressed as follows:
The Objectives state the clear and unambiguous goals of the Rules with respect to safety and
performance aspects. These objectives provide the basis for deriving the detailed structural acceptance
criteria.
The Systematic Review identifies and evaluates the hazards due to operational and environmental
influences and the likely consequences of these on the structure of a ship, in order that these can be
addressed in the Rules and thereby minimised. The consequences include those that have an effect on the
safety of life, the environment and property (ship and cargo). The Systematic Review also identifies
whether some of the risks or hazards to the structure are controlled by other standards or regulations (e.g.
MARPOL).
Figure 8 Rules Framework
The General Assumptions specify aspects that are beyond the scope of the Rules, but affect the
application and effectiveness of the rules. Typically, these include references to other international
regulations and industry standards, e.g. SOLAS and MARPOL. The General Assumptions also includes
information on the shared responsibilities of Societies, builders and owners.
446
The Design Basis specifies the premises that the Design Principles of the Rules are based on, in terms
of design parameters and the assumptions about the ship operation.
The Design Principles define the fundamental principles used for the structural requirements in the
Rules with respect to loads, structural capacity and assessment criteria. The principles are based on sound
engineering practise to provide suitable safeguards against the hazards identified by the systematic
review.
The Application of the Design Principles describes how the Design Principles and methods are
applied and what criteria are used to demonstrate that the structure meets the Objectives. It includes
definition of load and capacity models along with the corresponding acceptance criteria.
(IACS, 2014d) resumes reports describing in detail how the CSR-H complies with the GBS
framework.
In (Horn et al., 2013) the philosophy of the goal based ship construction standards (GBS) and their
implementation in the CSR-H are summarized. In the introduction a general description of the GBS and
the declination into 15 functional requirements is described.
Additionally comes a description of the application of GBS in CSR-H, the overview of the CSR-H
covering:
o
o
o
o
o
o
o
o
o
o
Wave loads
Equivalent design wave concept
Fatigue loads
Corrosion
Hull girder ultimate strength
Direct strength analysis
Buckling (including FE analyses for buckling)
Fatigue
Welding
Ship in operation
There were mentioned the functional aspects not directly covered by the CSR-H.
In (Wenquing et al., 2013) a study of the impact of the CSR-H on the design of tankers is described.
The authors conducted a study based on following ship types: 320K VLCC, 163K Suezmax, 115K
Aframax, 76K Panamax, 48K MR tankers. A consequence is resumed to the increase of steel weight. In
general a weight increase for all type of vessels investigated was found. The analysis is split to show
separately the influence of: increased scantlings, minimum thickness requirements, application of load
combination in Static +Dynamic conditions for oil cargo tank, the FE analysis covering a larger area of
the hull, acceptance and yielding criteria, ultimate buckling capacity of the stiffeners. The lengthening of
the design and approval phase was also mentioned.
(Shijian et al., 2013) summarise a comparative study of a corrugated bulkhead of a large product
tanker. The study is based on an Aframax 115k DWT vessel. The study contains a rule comparison that is
followed by a local scantling analysis. As a next step a direct coarse mesh strength analysis and local fine
mesh analysis are discussed. In conclusion a little weight increase has been found for a design according
to CSR-H rules. The factors influencing the results are discussed in the conclusions.
5.1.2 Goal Based Standards/Safety Level Approach (GBS/SLA) at IMO

This work started with submissions by different delegations (Germany, 2004) (Denmark et al., 2004b)
and (Denmark et al., 2004c). These documents made clear references to extending GBS beyond just
structural issues, including all aspects of safety and environmental protections and made clear references
to the use of risk assessment (FSA) to justify regulations. For structural issues, (Denmark et al., 2004b)
and (Denmark et al., 2004c) refer to Structural Reliability Analysis (SRA).
Over the following years, there were a large number of submissions regarding the GBS/SLA
(Safety Level Approach) and IACS even developed an alternative Tier II for the GBS/SLA (IACS,
2006a) and an example of use of SRA according the ALARP principle (IACS, 2006b). These and all
other papers on GBS/SLA were not really discussed at IMO, because all available time was spent on
the GBS/B&T.
IMO, 2011d is the most updated IMO document containing GBS Guidelines. It is worth noting that the
Guidelines are generic, and do not contain a preference for SLA. Actually, it states in paragraph 2 It
should be noted that these Guidelines are generic and where they use phrases such as required level of
447
safety, this does not imply any preference for a specific technical approach. As might be understood,
this indicates that there is not much agreement at IMO on the way forward.
The GBS/SLA work is now progressed in the Maritime Safety Committee (MSC) in a work group on
Goal-based Standards/Safety Level Approach (GBS/SLA) and Formal Safety Assessment (FSA). It is
now at least clear that only GBS/SLA is discussed as the way ahead (IMO, 2014b), but there are still
discussions on the need to quantify risks.
In the GBS debate over the years there has not been any explicit discussion about sustainability,
particularly with the global meaning adopted in the present report. The term is actually not present in the
many submitted documents. However, (IMO, 2011e) lists, as above mentioned, safety, security and
protection of the marine environment as targets for the action.
The way ahead may be foreseen. The first issue to be dealt with is probably that the FSA Guidelines
needs to develop risk criteria also for environmental issues. Currently, this is mentioned in (IMO, 2014b),
but the current FSA Guidelines only contain criteria related to oil spill, not to the global heating issues or
to the releases to air and sea. By including such criteria, the FSA Guidelines would effectively be changed
to general Regulatory Impact Assessment Guidelines. Some of the criteria required, like the external cost
of premature deaths due to emissions to air, is already in the FSA Guidelines (QALY Criteria), but this is
currently not transferred to criteria needed in IMO policy decisions. IMO typically sets the system border
around the ship and would require the criteria as an external cost per ton of air pollutant (e.g. NOx, SOx
and PM) or Cost of Averting a Ton of CO2 equivalent Heating effect (CATCH).
In conclusion it might be stated that there is currently no connection made at IMO between GBS and
sustainability, but that it is possible to foresee a development in that direction.
5.2
Regulatory actions implemented at IMO targeting environmental protection
A good overview of recent regulatory and legal developments in the maritime field is given by the United
Nations Conference on Trade and Development (UNCTAD) in the review of maritime transport (UN,
2013).
In the following, recent examples of the evolution of the IMO regulatory framework in the direction of
the limitation of environmental impact are covered
5.2.1 Energy Efficiency Design Index (EEDI)

Amendments to IMO MARPOL Annex VI regulations for the prevention of air pollution to ships entered
into force on January 1st 2013 and introduced the Energy Efficiency Design Index (EEDI) for all new
ships and the Ship Efficiency Management Plan (SEEMP) for all ships (IMO, 2011f). As known, the aim
of this requirement is to limit the installed power onboard ships and to push an efficient use of it, in view
of a reduction in emissions The subject was already reported in (Brunner et al., 2012). Recently, concerns
about the effect of these requirements on the manouvrrability of ships in bad weather conditions were
taken into account in (IMO, 2014a). In Chapter 6 below a part of the continuing discussion in literature on
the theme is recalled.
5.2.2 NOxSOx control

Reductions in the emission of nitrogen oxides (NOX) have been decided by IMO over recent years and
inherent requirements are entering into force, as already pointed out in (Brunner et al., 2012). Following
MEPC 66, recently amendments have been adopted to MARPOL Annex VI, regulation 13, on Nitrogen
Oxides (NOX), concerning the date for the implementation of Tier III standards within emission control
areas (ECAs, see 0 below and http://www.imo.org/MediaCentre/MeetingSummaries/MEPC/Pages/
MEPC66.aspx).
Sulphur oxide (SOX) emission limits were in effect from 1st January 2012, with MARPOL Annex VI
establishing SOX thresholds for marine bunker fuels (UN, 2013) and these thresholds will continue to
reduce over time to 0.5% by 2020, subject to a feasibility review in 2018. The EU has mandated that this
0.5% limit will enter force in European waters by 2020.
5.2.3 Emission Control Areas

Emission control areas were created under MARPOL Annex VI Regulations for the Prevention of Air
Pollution from Ships and defined as Special Areas. Such Special Areas are also possible under
MARPOL Annex I Prevention of Pollution by Oil, Annex II Control of Pollution by Noxious Liquid
Substances, Annex IV Prevention of Pollution by Sewage from Ships, and Annex V Prevention of
Pollution by Garbage.
448
Figure 9. Existing and Possible Future ECAs (DNV, 2011).
Special Areas are defined as certain sea areas in which for technical reasons relating to their
oceanographical and ecological condition and to their sea traffic, the adoption of special mandatory
methods for the prevention of sea pollution is required. Under the Convention, these special areas are
provided with a higher level of protection than other areas of the sea(International Maritime
Organisation, 2014e).
New Emission Control Areas covering SOX, NOX, and PM emissions have come into effect for North
American (1 August 2012) and United States Caribbean Sea (1 January 2014) areas. These add to the
Baltic Sea and North Sea ECAs already in force. Additional potential areas that can become ECAs are
highlighted in Figure 9. The considered ECA in North Norway will not be implemented. This decision
was based on a CBA along the lines described above.
5.2.4 MARPOL Annex V Prevention of Pollution by Garbage from Ships

On January 1st 2013 entered into force the revised MARPOL Annex V, adopted in July 2011 by MEPC
62 (IMO, 2011g). In March 2012, MEPC 63 adopted also guidelines for the implementation of the Annex
(IMO, 2012d) and for the development of garbage management plans (IMO, 2012c).
The view taken by the revised Annex V is that garbage can be just as dangerous to the marine
environment as classical pollutants like oil and chemicals. The annex now generally prohibits the
discharge of all garbage into the sea, with a few exceptions, listed in specific regulations (n. 4, 5, and 6 of
the Annex). The Annex binds Governments to ensure adequate reception facilities at ports and terminals
for the reception of garbage.
Special sea areas in a similar way as for the control of emissions are defined according to their
oceanographic and ecological conditions and/or to the particular traffic characteristics where the adoption
of special mandatory methods for the prevention of marine pollution by garbage is required.
5.2.5
IMO Ship Recycling (The Hong Kong Convention)
The regulatory activity at IMO regarding recycling of ships after reaching the end of their operational lives
started at the 44th MEPC session in March 2000 and is not yet finalised. As already mentioned in (Brunner
et al., 2012), a major step was the adoption of the Hong Kong International Convention for the Safe and
Environmentally Sound Recycling of Ships (IMO, 2009). This convention aims at ensuring that ships which
are being recycled do not pose any unnecessary risks to human health, safety and to the environment.
The convention is not yet in force, still requiring the agreement of 15 states representing 40% of
merchant shipping gross tonnage to ratify the convention. In addition the combined annual recycling
volume of these states during the preceding 10 years must constitute not less than 3% of the combined
merchant shipping tonnage.
A number of application guidelines have been adopted in the meanwhile by MEPC 62, 63, and 64
(IMO, 2011b, IMO, 2011a, IMO, 2012a, IMO, 2012b, IMO, 2012f, IMO, 2012e).
When entering into force, the convention will introduce regulations for the: design, construction,
operation and preparation of ships in order to facilitate the safe and environmentally sound recycling of
ships without compromising their safety and operational efficiency. Among other provisions, a Green
Passport for ships was established, containing an inventory of all materials used in the construction of a
449
ship that are potentially hazardous to human health or to the environment.The Convention covers, too, a
proper operation of ship recycling facilities and the establishment of a system of certification and
reporting requirements enforcing an appropriate application of the provisions for ship recycling.
5.2.6
Pre-normative investigations at IMO in the field of noise radiation into water
In recent years, IMO dealt also with the problem of noise radiation into water by ships. in particular
MEPC 58 approved the inclusion of a new high priority item in the work programme of the Committee,
on noise from commercial shipping and its adverse impact on marine life.
A Correspondence Group was established to identify and address ways to minimize the underwater
noise radiation from commercial shipping operations. The concern was about the short and long-term
negative consequences on marine life, especially marine mammals.
An outcome of that work are the later approved Guidelines for the reduction of underwater noise from
commercial shipping to address adverse impacts on marine life (IMO, 2014c)
These are non-mandatory guidelines intend to provide general advice to designers, shipbuilders and
ship operators about reduction of underwater noise.
It is here noted that the the control of underwater noise radiated by ships has been and is being
investigated by EU funded research projects see (SILENV, 2012) (AQUO, 2015) (SONIC, 2015). It is
foreseeable that all the above activities will produce in a not far future design and/or operational
requirements aimed at reducing this type of impact. Other (non IMO) regulatory actions in the field of ships
5.3
Other (non IMO) regulatory actions in the field of ships
5.3.1
Developments in the Naval Ship Code
Naval platforms are large and complex engineering systems that are typically a product of a long and
lengthy development process. They are also generally a unique engineering challenge with considerable
limits on the information available at the time that key decisions are made. Whilst the development of
goal based standards discussed in section 5 of this report has involved the commercial marine industry,
the naval world has also embraced similar concepts and ideas. The development of the Naval Ship Code
(NSC) ANEP 77 (NATO, 2014) clearly shows goal based ideas being comprehensively implemented
supported by the International Naval Safety Association, whose vision is that the Naval Ship Code
becomes established as a cost-effective goal based standard for naval ship safety benchmarked against
statute, and accepted by the global naval community and intergovernmental bodies(INSA, 2014).
The code is intended to produce an equivalent level of safety to corresponding merchant codes whilst
the role of naval ships is such that the safety of the naval ship and embarked personnel may be a
secondary consideration (NATO, 2014). The code covers systems such as structure, stability, and fire
safety but excludes military specific components. As discussed by (Nappi and Collette, 2013), ANEP 77
does not prescribe specific regulations but rather it is a code for the selection of suitable regulations. In
other words, the higher levels and goals are set by the responsible administrations in line with the code
and then suitable regulations are developed to achieve these overall goals.
The code was discussed in detail by ISSC Specialist Committee V5 (Ashe et al., 2009) and is not further
reviewed in detail here. Recently it has been suggested that the code should be split into two segments with
the upper part of the pyramid being the responsibility of the Naval Flag Administrations in setting the
goals, and the other segment covering the prescriptive requirements of level 4 via the implementation of
relevant Naval Standards or Class Society regulations (Fredriksen, 2013) as shown in Figure 10.
Figure 10. Potential future direction of Naval Code (Fredriksen, 2013).
450
This splitting of the overall goals from the specific choice of prescriptive requirements enables the standards
to be maintained for applicability to a wide range of naval ship types and allows the choice of the best suited
standard in each individual design case (Fredriksen, 2013). It introduces the principle of choosing the best
suited prescriptive requirements and provides the flexibility to reduce the requirement to commit to single or
narrow set of technical solutions. Current ongoing work by INSA is to improve the consistency of the part 1
safety goals by clarifying the goals and the principles behind them themselves in further detail. Framework
and guidance for use of the Naval Ship Code certification are also being developed.
5.3.2 Inland vessels

An overall picture of the dimension of inland water transport (IWT) in Europe can be derived by the
recent survey by the Central Commission for the Navigation of the Rhine (CCNR, 2014). There, about
8000 dry good vessels and 1,650 tankers are mentioned, with various sizes.
For numbers related to Russia, reference is made to (Egorov, 2014), mentioning for dry cargo about
1200 motor vessels and 4900 barges, for tankers 700 motor vessels and 900 barges and, finally, about
5800 pusher boats. An analysis of the state and age of this fleet is reported in the same paper.
For more than six decades, UNECE Inland Transport Committee (ITC) has promoted the develop of
international inland water transport at European level. A recent document exploiting the characteristics
and advantages of this transport mode is the White Paper on Efficient and Sustainable Inland Water
Transport in Europe (UNECE, 2011b)
UNECE provided a platform for intergovernmental cooperation to facilitate IWT. An important target
of UNECE is the development of a unified normative framework at European level.
This is particularly needed as the IWT suffers, even at a European level, from a historical
infrastructural, institutional, legal and also technical fragmentation, due to the presence of local
institutions (River Commissions: Danube DC, Rhine CCNR, Mosel MC, Sava SC) providing
requirements for their specific river basins on various overlapping subjects.
The Recommendations on Harmonized Europe-Wide Technical Requirements for Inland Navigation
Vessels (UNECE, 2011a) establish a Pan-European regime of technical requirements for inland
navigation vessels engaged in international transport of goods and passengers. They are a result of
Governments efforts, aimed at unifying the divergent regulations in force within different
intergovernmental organizations and within individual UNECE member countries. These unified
regulations have been brought in line with relevant European Unions legislation and are to facilitate
vessels engaged in international transport by inland waterway (see http://www.unece.org).
The Recommendations contain, in particular, strict regulations on limitation of air and water pollution
by vessels and on abatement of the noise radiated as well as internationally agreed standards for minimum
manning requirements and working and rest hours of crews.
Operational requirements are also contained in the European Code for Inland Waterways-CEVNI
(UNECE, 2009)
Another relevant document is the European Agreement concerning the International Carriage of
Dangerous Goods by Inland Waterways (ADN), issued by the Central Commission for the Navigation of
the Rhine (CCNR, 2015).
The issues of pollution prevention and waste management are addressed by several UNECE and River
Commission instruments, such as special resolutions and relevant provisions of the technical prescriptions
for vessels.
For example, on November 1st 2009, the convention on the Collection, deposit and reception of waste
produced during navigation on the Rhine and Inland Waterways (CDNI) entered into force (CDNI,
2009). Upgrades have been issued in the following years.
The CDNI has been adopted by six countries (Luxembourg, Switzerland, the Netherlands, Belgium,
Germany and France) and is applicable over the entire length of the river Rhine and on all inland
waterways in Germany, the Netherlands and Belgium, on the international part of the river Moselle in
Luxembourg and France. In order to assure the application on the entire geography of the Convention, the
respective rules have been transposed into the national law of each Member State so as to become
applicable in the conventional network.
The same subject are covered in Res. 21 Prevention of Pollution of inland Waterways by vessels
(UNECE, 2007) and CEVNI (UNECE, 2009) and following updates, including cooperation with various
River Commissions (in particular Sava and Danube)
In 2010 the Russian Federation approved new technical regulations for the safety of Russian IWT
objects. These requirements regulate the presence of double bottom and double sides for tankers and
vessels transporting dangerous cargoes.
451
5.3.3 EU Directive on Safety of Offshore Oil and Gas Operations

It is here just mentioned that attention to environment in the Off shore field is shown by the European
Union with the adoption of the Directive 2013/30/EU on Safety of Offshore Oil and Gas Operations (EU,
2013) which intends to enhance the safety standards on platforms across European waters whilst
minimising potential damage to the environment and the livelihoods of coastal communities.
5.4
Comments on the recent developments in the normative framework
From the review presented in this chapter, some general trends in the development of normative
frameworks for the design and operation of ships can be detected.
On one hand, the need for a simplification and systemisation of the general lay out of the Norms is
quite diffused even though it could be implemented to a different extent in the different fields. For
commercial shipping, the international nature of the activity and the presence of a world wide normative
organisation like IMO facilitates the process, while for naval vessels and for inland vessels, this is
objectively more difficult due to the particular interests and local characteristics of the activities.
The trend for enlarging the spectrum of implications of shipping activities to include the impact on
environment in its various forms is also easily identifiable, with concerns that are oriented both to global
effects on planetary scale (GHG emissions) and to more local effects (chemical pollution, waste
treatment, noise). For inland water vessels, the local effects show a higher priority, giving the higher
proximity with inhabited areas and their specific features.
At IMO level, too, in addition to general requirements set world wide, specific ones are set in special
areas.
Requirements at all levels are however still formulated in a prescriptive way, i.e. they force the
achievement of specific results on single aspects of the design. In this respect, the present state of the art
of the regulatory framework for shipping is not in line with the approach defined in 1,2 and 3 of design
for sustainability. As always when dealing with prescriptive requirements formulated independently from
each others, the efforts needed to meet the requirements themselves are not globally optimised in a costbenefit sense. In other words, this type of formulation prevents the possibility of finding the optimal
global solutions that could be sought by a rational and holistic approach to the design for sustainability.
6.
STUDIES FOCUSSING ON ENVIRONMENTAL IMPACT
The present chapter aims at providing a few references about the recent developments of the scientific
and technical framework in which the concept of design for sustainability finds its roots. The focus is on
ship design, even though many issues have much more general implications.
6.1
Studies on Green House Gas Emissions
When dealing with the global impact on planetary scale of anthropogenic emissions, very comprehensive
and authoritative references are represented by the Assessment Reports by the Intergovernamental Panel
on Climate Changes, in particular the latest one (5th one: AR5) with the synthesis report (IPCC, 2014c)
and the detailed reports by the three working groups (IPCC, 2013, IPCC, 2014a, IPCC, 2014b).
The AR5 provides an update of knowledge on scientific, technical and socio-economic aspects of
climate change. The study analyses the observed changes in the climate, with causes, effects and impacts.
It identifies past and recent drivers of climate change linking them to human activities.
The Report concentrates, even more than in previous editions, on the future evolution, identifying
possible scenarios for discussing risks, adaptation and in particular mitigation measures to contain the
changes.
Among many other findings, a confirmation of an overall possible objective of a limitation of the
earths temperature rise to 2C, as previously set (IPCC, 2007b) and discussed, see e.g.(den Elzen and
Meinshausen, 2006) and a further deepening of the twofold global effect (cooling, but also heating) due to
a control of some air pollutants typical of ship emission: (SO2 and NOx: see (IPCC, 2013), chapt. 8). As
regards CO2 emissions from ships, the Third IMO Greenhouse Gas Study (GHG) 2014 (Smith et al.,
2014b) has updated the outcomes of the previous edition (Buhaug et al., 2009) with predicted emissions
from shipping of 3.1% of annual global CO2 emissions in 2007 and these are expected to increase by
2050. The work develops a multi-year inventory and future scenarios for the period 2007 2050.
This has lead toward significant efforts being proposed, and some implemented, as to how to quantify
and reduce the potential impact of shipping and related activities.
Investigations in this area have been carried out by the UK Low Carbon Shipping consortium (Smith
et al., 2014a) and the Clean North Sea Shipping consortium (CNSS, 2014).
452
Fitzgerald et al., 2011 assessed the methodologies for calculating greenhouse gas emissions from
international maritime transport through a case study of the imports and exports from New Zealand.
This investigation in particular highlighted the potential for differing results when considering
emissions calculated from fuel bunkered for international transport and emissions computed on the
basis of the actual movement of goods in and out of New Zealand. International policy implications are
discussed.
Murphy et al., 2013 discussed that there are numerous initiatives for the assessment of the
environmental performance of ships and highlighted that the two most widely used are able to address
more than one pollutant of interest. They went on to consider the effectiveness of current and possible
future forms of environmental indexing schemes.
6.2
Studies on countermeasures to limit emissions
6.2.1 Slow steaming

Psaraftis and Kontovas, 2010 investigated the implications of various maritime emission reduction
policies for maritime logistics exploring the trade-offs between environmental benefits associated with
slow steaming and fleet size and the economic attributes such as in-transit inventory holdings.
(Claudepierre et al., 2012) discussed the potential use of ultra-slow ships for CO2 reduction. This work
considered the advantages and disadvantages of slow steaming, in particular in crowded shipping lanes,
adverse weather or areas where there may be security concerns, concentrating on determining if available
propulsion power, when slow steaming, enabled the safe operation of this ship. This concept was also
considered by (Smith et al., 2014a) who highlighted the potential need for more ships to be constructed in
order to meet supply chain demand and the emissions from the manufacturing processes may outweigh
any potential benefits.
Lindstad et al., 2011 considered the potential reductions in greenhouse gases and cost by shipping at
lower speeds for a selection of shipping classes showing that there is a potential for substantial reductions
in direct CO2 emissions from ships.
6.2.2 Scale effects and propulsive improvements

Lindstad et al., 2012 investigated the importance of economies of scale for reducing the levels of
greenhouse gas emission from ships through the consideration of the effect of increasing vessel size. This
study suggests that replacing the current fleet with larger vessels would result in a reduction in CO2
emissions.
Molland et al., 2014 reviewed the practical improvements in the propulsive efficiency of future ships.
This work investigated the potential design changes and placed them into the context of economic and
environmental efficiency.
6.2.3 Discussions of the EEDI concept

The intention of the EEDI is to reduce the installed power on ships, however there is concern as to the
unintended consequences of this change potentially in the safe operation of ships. This was discussed in
the previous report of this committee (Brunner et al., 2012).
Verhulst et al., 2013 discussed the applicability of EEDI values to small cargo ships concluding that
these ship types will require correction factors to the EEDI formulae in order to account for the effects of
different operational profiles, class notations, and installed loading gear found on such vessels.
Pundt and Kruger, 2013 reported the development of a methodology to verify the speed of a ship for
the EEDI.
Egorov and Kolesnik, 2012 show that Russian new generation river-sea vessels satisfy EEDI
requirements.
Longva et al., 2010b reported on the development of a required energy efficiency design index level
based on a cost-effectiveness criterion. This work considered how to set the required targets for the
EEDI based on a decision criterion of 50 USD per tonne CO2-eq averted. 11 emmission reduction
measures were analysed within this study for a representative ship type with results indicating the
potential for a reduction in the required index level of 2530% in a cost effective manner. It is
suggested that as the EEDI only consideres CO2 emission reductions, there is potential for adverse
impacts on other emissions, discharges, safety levels for crew and ship which should be addressed with
a total impact analysis.
453
6.2.4 Studies on control of NOx and SOx emissions

CNSS, 2014 reports that any potential delay in the ongoing change over to tier 1 and 2 engines will have a
significant impact on the contribution from shipping related activities to global levels of NO2.
Dinwoodie et al., 2013 used a Delhi study to investigate the perceptions of maritime specialists of the
potential changes in maritime oil fright flows to 2050. This study suggests that the effects of local
sourcing, new artic seaways, and reduced fossil fuel usage will result in lower oil freight but that the
introduction of Emission Control Areas and piracy may result in longer haul lengths as shipping re-routes
to reduce the impact of these restrictions.
A similar study investigating the Dry Bulk shipping flows to 2050 was undertaken by (Dinwoodie
et al., 2014). This study reported that an expected increased raw material shipping might be offset by
reduced haul lengths and that an expected rise in demand for biofuels combined with a reduction in fossil
fuels will lead to system based approach to future planning.
6.2.5 Emissions Trading Schemes

Within the European Union (EU), greenhouse gas emissions have been a topic of significant debate for some
time. For the aviation sector, the EU has begun to implement an emissions trading scheme (ETS) which aims
to limit or cap the total amount of certain greenhouse gases which can be emitted within European airspace.
This has proven highly controversial and is subject of much international political debate.
A similar scheme has been proposed for the shipping sector within European waters which is the
subject of considerable debate. (Smith et al., 2014a) provide an outline of the parallels and differences
between such ETS schemes for the aviation and shipping sectors.
6.2.6 Alternative Fuels

A review of future ship powering options was undertaken by the UK Royal Academy of Engineering
(2013). This report was prepared by a working group of approximately 45 leading industrialists and
academics considered the potential benefits of a range of propulsion options with the aim to reduce the
impact on the environment whilst maintaining operational efficiency. Whilst other propulsion options
such as LNG, fuel cells, biofuels, and hydrogen are all potentially adoptable, a key conclusion from the
report is that the adoption of any alternative propulsion options will be dependent upon the prices of fuels,
the impact of present and future legislation, and the potential introduction of carbon taxes. For many of
these alternative options, significant research and development challenges remain both in the actual
technology development but also in the infrastructure and supply chain required to actually provide them.
Bengtsson, Fridell, and Andersson (2012) investigated the life cycle performance of two alternative
pathways to biofuels in the shipping industry by considering the use of bio-diesel and liquefied biogas.
Within the context of this life-cycle analysis, it is reported that the use of biogas resulted in potentially
better environmental performances that the use of biodiesel, due to a lower overall environmental impact,
but that the biodiesel had more infrastructure in place. The study also reported that, although the use of
biofuels would have an effect on reducing the global warming impact from shipping, this would be at the
expense of greater potential impact from other methods of reduction in environmental impacts, and that
the biomass itself may be more cost effectively used in other industries.
7.
CONCLUSIONS
In the first three chapters of this report, the general bases of the concept of design for sustainability have
been outlined. Such concept seeks a rational approach to decision making in a normative context by
pursuing an holistic approach in considering the implications of the regulatory action. This means to
account for effects ranging in the probability domain from certain to very rare events, in the time domain
from contemporaneity to far future and spanning in different fields (preservation of assets, human health
and environment).
Basic tools for probabilistic methods of analysis are available since many years and their adoption in
the development of the normative framework at IMO is slowly gaining momentum, as recalled in 5. In
general the use of these methods is growing in the technical and scientific literature in the field of ship
construction, as partially reported in 4 and 6.
The discussion about depreciation rates in decision making reported in 3 relates mainly to long term
global effects of GHG emissions and is not specific for shipbuilding: concepts and procedures can (and
have to) be adopted in line with other fields.
More room seems to be available for quantifying better those aspects of environmental impact that are
specific of the maritime sector and may differ significantly in terms of emissions, of effects (or both) in
respect to other transport modes and/or human activities.
454
As remarked in 5, the normative framework in shipbuilding is still nowadays dominated by

prescriptive requirements, whose formulations not necessarily represent the most balanced design
solution in respect to sustainability. What is felt to be important for defining rational design criteria for
ships, in line with the sustainability target, is to give a proper weight to all the local aspects of the
environmental impact: on the marine as well as the atmospheric part of the ambient. This includes
chemical emissions, but also f.i. noise radiation in water (as it is done in air for other transportation means
and inland vessels). In the atmosphere, specific aspects could include those effects (e.g. PM emissions)
which, with the exception of a few sensitive areas, might have a different impact as compared to
analogous emissions on land and should therefore be valued accordingly.
Operations like the one carried out in (Korzhenevych et al., 2014) of pricing external costs of
transport in general could be carried out in more details for the shipping sector in order to get a more
precise picture of the actual societal impact of maritime activities (and of the subdivision into the various
components). This would make possible to take more rational normative and design decisions on where
allocating resources to enforce sustainability in this sector.
REFERENCES
Andrews, D., Papanikolaou, A. & Singer, D. J. 2012. Design for X. 11th International Maritime Conference on
Design for Safety. Glasgow.
Aquo. 2015. FP7 Project AQUO: Achieve QUieter Oceans by shipping noise footprint reduction [Online].
Available: http://www.aquo.eu/.
Ashe, G., Cheng, F., Kaeding, P., Kaneko, H., Dow, R., Broekhuijsen, J., Pegg, N., Fredriksen, A., De Francisco, F. V.,
Leguen, J. F., Hess, P., Gruenitz, L., Jeon, W., Kaneko, H., Silva, S. & Sheinburg, R. 2009. Naval Ship Design. 17th
International Ship and Offshore Structures Congress ISSC 2009. Seoul, Korea: Seoul National University.
Aspen, D. M., Fet, A. M. & Johansem, B. A. 2012. Life Cycle System Analysis Methods in Ship Design A
Review of Current Status, Opportunities and Challenges. 11th International Maritime Conference on Design for
Safety. Glasgow.
Bickel, P. & Friedrich, R. 2005. ExternE. Externalities of energy: Methodology 2005 Update. Luxembourg:
European Commission.
Bickel, P., Friedrich, R., Burgess, A., Fagiani, P., Hunt, A., Jong, G. D., Laird, J., Lieb, C., Lindberg, G., Mackie,
P., Navrud, S., Odgaard, T., Ricci, A., Shires, J. & Tavasszy, L. 2006. Proposal for Harmonised Guidelines.
Deliverable 5 of HEATCO (Developing Harmonised European Approaches for Transport Costing and Project
Assessment) project. Stuttgart: Institut fr Energiewissenschaft und Rationelle Energieanwendung.
Bitner-Gregersen, E. M., Eide, L. I., Hrte, T. & Skjong, R. 2013. Ship and offshore structure design in climate
change perspective, Springer.
Black, I., Seton, R., Ricci, A. & Enei, R. 2003. Actions to promote intermodal transport. Final report of RECORDIT
(Real cost reduction of door-to-door intermodal transport) project.
Boulougouris, E. & Papanikolaou, A. 2013. Risk-based design of naval combatants. Ocean Engineering, 65, 4961.
Brunner, E., Birmingham, R. W., Byklum, E., Chen, Y., Cheng, Y. F., Dasgupta, J., Egorov, G., Juhl, J., Kang, B.
S., Karr, D., Kawamura, Y., Klanac, A., Oneil, S., Rizzuto, E., Teixeira, A. & Yoshida, K. 2012. Design
Principles and Criteria. In: Fricke, W. & Bronsart, R. (eds.) 18th International Ship and Offshore Structures
Congress ISSC 2012. Rostock, Germany: Schiffbautechnische Gesellschaft e. V., 435506.
Buhaug, ., Corbett, J. J., Endresen, ., Eyring, V., Faber, J., Hanayama, S., Lee, D. S., Lee, D., Lindstad, H.,
Mjelde, A., Plsson, C., Wanquing, W., Winebrake, J. J. & Yoshida, K. 2009. Second IMO Greenhouse Gas
Study. International Maritime Organization, London
Cai, W. & Konovessis, D. A Methodology of Marine Accident Investigation for Risk-Based Ship Design. 11th
International Maritime Conference on Design for Safety, June 2012 2012 Glasgow. 263282.
Ccnr 2014. Inland Navigation in Europe - Market observation 2014.
Ccnr 2015. European Agreement concerning the International Carriage of Dangerous Goods by Inland Waterways
(ADN).
Cdni 2009. Convention on the collection, deposit and reception of waste produced during navigation on the Rhine
and Inland Waters CDNI
Choung, C. H., Kim, H. K., Park, J.-H. & Ha, T. B. Study on the risk-based methodology for ship structural design.
11th International Maritime Conference on Design for Safety, 2012 Glasgow. 497506.
Claudepierre, M., Klanac, A. & Allenstrom, B. 2012. ULYSSES - the ultra slow ship of the future. Green Ship
Technology Conference. Copenhagen, Denmark.
Cnss. 2014. Clean North Sea Shipping Project Final Report: Key Findings and Recommendations. [Online].
Available: Available from: http://cnss.no/wp-content/uploads/2014/03/CNSS_Final_Report_web.pdf.
Cowell, F. A. & Gardiner, K. 1999. Welfare Weights. Economics Research Paper 20. STICERD London School of
Economics.
Den Boer, E., Otten, M. & Van Essen, H. 2011. STREAM international freight 2011: Comparison of various
transport modes on an EU scale with the STREAM database. Commissioned by: Dutch Ministry of Infrastructure
and the Environment ed.: Delft, CE Delft, July 2011.
455
Den Elzen, M. & Meinshausen, M. 2006. Meeting the EU 2 C climate target: global and regional emission
implications. Climate Policy, 6, 545564.
Denmark, Iceland, Norway & Islands, F. 2004a. GOAL-BASED NEW SHIP CONSTRUCTION STANDARDS
Issues for consideration
Denmark, Iceland, Norway & Islands, F. 2004b. GOAL-BASED NEW SHIP CONSTRUCTION STANDARDS,
Issues for consideration.
Denmark, Iceland, Norway & Islands, F. 2004c. GOAL-BASED NEW SHIP CONSTRUCTION STANDARDS,
Issues for consideration.
Dinwoodie, J., Landamore, M. & Rigot-Muller, P. 2014. Dry bulk shipping flows to 2050: Delphi perceptions of
early career specialists. Technological Forecasting and Social Change, 88, 6475.
Dinwoodie, J., Tuck, S. & Rigot-Mller, P. 2013. Maritime oil freight flows to 2050: Delphi perceptions of
maritime specialists. Energy Policy, 63, 553561.
Eea 2011. Revealing the costs of air pollution from industrial facilities in Europe. European Environment Agency
(EEA).
Egorov, G. V. 2014. Condition and Perspectives of Russian River Shipbuilding. Morskoe Strakhovanie Journal,
pp. 6680.
Egorov, G. V. & Kolesnik, D. V. 2012. Energy Efficiency of River-Sea Vessels of New Generation. Morskoy
Vestnik Journal, pp. 97103.
Eide, M. S., Endresen, ., Skjong, R., Longva, T. & Alvik, S. 2009. Cost-effectiveness assessment of CO2 reducing
measures in shipping. Journal of Maritime Policy and Management. Taylor and Francis Group.
Essen, H. P. V., Boon, B. H., Maibach, M. & Schreyer, C. 2007. Methodologies for External Cost Estimates and
Internalisation Scenarios. Discussion paper for the workshop on internalisation on March 15, 2007. Delft: CE Delft.
Eu 2011. IMPACT ASSESSMENT, Accompanying the document: Proposal for a Regulation of the European
Parliament and of the Council on safety of offshore oil and gas prospection, exploration and production activities.
COM(2011) 688 final, SEC(2011) 1293 final, SEC(2011) 1294 final.
Eu 2013. Safety of offshore oil and gas operations In: Parliament, E. (ed.) Directive 2013/30/EU
Fanshel, S. & Bush, J. W. 1970. A health status index and its application to health services outcomes. Oper Res.
Fet, A. M., Aspen, D. M., Ellingsen, H. & Skaar, C. Sustainable Design in View of Life Cycle Management. 11th
International Maritime Conference On Design For Safety, June 2012, 2012 Glasgow. 395408.
Fet, A. M., D.M., A. & Ellingsen, H. 2013. Systems Engineering as a Holistic Approach to Life Cycle Designs.
Fitzgerald, W. B., Howitt, O. J. & Smith, I. J. 2011. Greenhouse gas emissions from the international maritime
transport of New Zealands imports and exports. Energy Policy, 39, 15211531.
Fredriksen, A. 2013. Future use of NSC. Presentation to INSA Action 12AGM-24:. Oslo, Norway: DNV.
Friedrich, R. & Bickel, P. 2001. Environmental External Costs of Transport. Berlin: Springer.
Gafni, A. 1999. Economic Evaluation of Health Care Interventions: An Economists Perspective. ACP Journal Club
Germany 2004. GOAL BASED NEW SHIP CONSTRUCTION STANDARDS, Guiding strategy
Greece 2007. FORMAL SAFETY ASSESSMENT, Environmental Risk Evaluation Criteria. MEPC 56/18/1. IMO.
Greece 2009. FORMAL SAFETY ASSESSMENT, Report of the Correspondence Group on Environmental Risk
Evaluation Criteria (Submitted by Greece, on behalf of the Coordinator of the Correspondence Group). MEPC
60/17. IMO.
Holland, M., Pye, S., Watkiss, P., Droste-Franke, B. & Bickel, P. 2005. Damages per tonne of PM2.5, NH3, SO2,
NOx and VOCs of EU25 Member State (excluding Cyprus) and surrounding seas. Service contract for carrying
out cost-benefit analysis of air quality related issues, in particular in the Clean Air for Europe (CAFE)
programme. Oxon. UK.: AEA Technology Environment.
Horn, G. E., Arima, T., Baumans, P., Boe, A. & Ocakli, H. 2013. IACS summary of the IMO GBS and the
Harmonised Structural Rules. Tanker Structure Cooperative Forum TSFC.
Iacs 2006a. GOAL-BASED NEW SHIP CONSTRUCTION STANDARDS, Draft Tier II Safety Level Approach.
Iacs 2006b. GOAL-BASED NEW SHIP CONSTRUCTION STANDARDS, Linkage between FSA and GBS.
Iacs 2012a. Common Structural Rules for bulk carriers.
Iacs 2012b. Common Structural Rules for double hull oil tankers.
Iacs 2014a. Common structural rules for bulk carriers and oil tankers.
Iacs 2014b. Common structural rules for bulk carriers and oil tankers, Technical background rule reference.
Iacs 2014c. Consequence Assesment (CA) Summary report.
Iacs 2014d. Harmonised CSR, Rule General Principles and Validation.
Imo 2007. Consolidated text of the guidelines for formal safety assessment (FSA) for use in the IMO rule-making
process (MSC/Circ.1023MEPC/Circ.392). MSC 83/INF.2. IMO.
Imo 2009. Hong Kong International Convention for the safe and environmentally sound recycling of ships, 2009.
Imo 2010. FORMAL SAFETY ASSESSMENT, Report of the Working Group on Environmental Risk Evaluation
Criteria within the context of Formal Safety Assessment. MEPC 60/WP.11. 24 March 2010: IMO.
Imo 2011a. 2011 Guidelines for the development of the inventory of hazardous materials
Imo 2011b. 2011 Guidelines for the development of the ship recycling plan.
Imo 2011c. FORMAL SAFETY ASSESSMENT, Report of the Working Group on Environmental Risk Evaluation
Criteria within the context of Formal Safety Assessment. MEPC 62/WP.13.
Imo 2011d. Generic Guidelines for developing IMO Goal-Based Standards.
Imo 2011e. High-Level Action Plan of the Organisation and Priorities for the 20122013 Biennium.
456
Imo 2011f. Inclusion of regulations on energy efficiency for ships in MARPOL Annex VI.
Imo 2011g. Revised MARPOL Annex V.
Imo 2012a. 2012 Guidelines for safe and environmentally sound ship recycling.
Imo 2012b. 2012 Guidelines for the authorization of ship recycling facilities.
Imo 2012c. 2012 Guidelines for the Development of Garbage Management Plans.
Imo 2012d. 2012 Guidelines for the Implementation of MARPOL Annex V.
Imo 2012e. 2012 Guidelines for the inspection of ships under the Hong Kong Convention.
Imo 2012f. 2012 Guidelines for the survey and certification of ships under the Hong Kong Convention.
Imo 2013a. Guidelines for the Approval of Alternatives and Equivalents as provided for in various IMO
instruments.
Imo 2013b. Revised guidelines for formal safety assessment (FSA) for use in the imo rule-making process. MSCMPEC.2/Circ. 12.
Imo 2014a. 2013 Interim guidelines for determining minimum propulsion power to maintain the manoeuvrability of
ships in adverse conditions.
Imo 2014b. GOAL-BASED NEW SHIP CONSTRUCTION STANDARDS, Report of the GBS/FSA Working
Group (part 1).
Imo 2014c. Guidelines for the reduction of underwater noise from commercial shipping to address adverse impacts
on marine life.
Imo 2014d. Strategic plan for the Organization (for the six year period 2014 to 2019).
Ingham, A. & Ulph, A. 2003. Uncertainty, Irreversibility, Precaution and the Social Cost of Carbon. Tyndall Centre
for Climate Change Research.
Insa. 2014. Available: http://www.navalshipcode.org.
Ipcc 2007a. Climate Change 2007: Synthesis Report. An Assessment of the Intergovernmental Panel on Climate Change.
Ipcc 2007b. Fourth Assessment Report of the Intergovernmental Panel on Climate Change.
Ipcc 2013. Climate Change 2013 - The Physical Science Basis.
Ipcc 2014a. Climate Change 2014 - Impacts, Adaptation, and Vulnerability.
Ipcc 2014b. Climate Change 2014 - Mitigation of Climate Change.
Ipcc 2014c. Climate Change 2014 - Synthesis Report.
Jacob, K., Weiland, S., Ferretti, J., Wascher, D. & Chodorowska, D. 2011. Integrating the environment in regulatory
impact assessments. OECD.
Japan 2008. FORMAL SAFETY ASSESSMENT, Relevant information in relation to the Draft Environmental Risk
Evaluation Criteria. MEPC 58/17/1.
Jiang, L., Kronbak, J. & Christensen, L. P. 2014. The costs and benefits of sulphur reduction measures: Sulphur
scrubbers versus marine gas oil. Transportation Research Part D: Transport and Environment.
Kaneko, F., Oka, H., Asami, M., Ohnishi, S., Hirao, Y., Kawagoe, Y., Itoh, H., Yuzui, T., Kudo, J. & Mitomo, N.
Research on Advanced Risk Based Design. 11th International Maritime Conference on Design for Safety, June
2012 2012 Glasgow. 517530.
Kawamura, Y., Ohba, Y. & Y., K. Lifecycle Structural Optimization of Mid-Ship of Double Hull Tanker Based on
Holistic Risk Evaluation. In: Guedes Soares, C. R., J. , ed. Marstruct 2013, Espoo, Finland, March 2527, 2013),
Pp. 555563., March 2527, 2013 2013 Espoo, Finland. Crc Press/Balkema, 555563.
Kirkpatrick, C. H. & Parker, D. 2007. Regulatory impact assessment: towards better regulation?, Edward Elgar
Publishing.
Kontovas, C. A., Psaraftis, H. N. & Ventikos, N. P. 2010. An empirical analysis of IOPCF oil spill cost data. Marine
Pollution Bulletin.
Korzhenevych, A., Dehnen, N., Brcker, J., Holtkamp, M., Meier, H., Gibson, G., Varma, A. & Cox, V. 2014.
Update of the Handbook on External Costs of Transport. London: Ricardo AEA.
Lee, J., Park, G. J. & Kang, H. J. Risk-based Approaches in Design for Ship Safety. 11th International Maritime
Conference on Design for Safety, Lee, J., Park, G.J., Kang, H.J. (2012) Risk-based Approaches in Design for
Ship Safety. Proc. IMDC 2012 (11th International Maritime Conference on Design for Safety), Vol.2, Glasgow,
June 2012, 6374 2012 Glasgow. 6374.
Lind, N. 2002. Social and economic criteria of acceptable risk. Reliability Engineering and Systems Safety. Elsevier.
Lindstad, H., Asbjrnslett, B. E. & Strmman, A. H. 2011. Reductions in greenhouse gas emissions and cost by
shipping at lower speeds. Energy Policy, 39, 34563464.
Lindstad, H., Asbjrnslett, B. E. & Strmman, A. H. 2012. The importance of economies of scale for reductions in
greenhouse gas emissions from shipping. Energy Policy, 46, 386398.
Liu, W., Wirtz, K. W., Kannen, A. & Kraft, A. 2009. Willingness to pay among households to prevent coastal
resources from polluting by oil spills: a pilot survey. Marine Pollution Bulletin.
Longva, T., Eide, M. S. & Skjong, R. 2010a. Determining a Required Energy Efficiency Design Index Level for
New Ships Based on a Cost-Effectiveness Criterion. Maritime Policy & Management.
Longva, T., Eide, M. S. & Skjong, R. 2010b. Determining a required energy efficiency design index level for new
ships based on a cost-effectiveness criterion. Maritime Policy & Management, 37, 129143.
Maibach, M., Schreyer, C., Sutter, D., Van Essen, H. P., Boon, B. H., Smokers, R., Schroten, A., Doll, C.,
Pawlowska, B. & M., B. 2008. Handbook on Estimation of External Cost in the Transport Sector. Internalisation
Measures and Policies for All external Cost of Transport (IMPACT). The Netherlands: Delft, CE.
457
Molland, A., Turnock, S., Hudson, D. & Utama, I. 2014. Reducing ship emissions: a review of potential practical
improvements in the propulsive efficiency of future ships. Transactions of Royal Institution of Naval Architects
Part A, 156, 175188.
Moore, W. H., Arai, M., Besse, P., Birmingham, R., E., B., Chen, Y. Q., Dasgupta, J., Friis-Hansen, P., Boonstra,
H., Hovem, L., Kujala, P., Mcgregor, J., Rizzuto, E., Teixeira, A., Zanic, V. & Yoshida, K. 2009. Design
Principles and Criteria. 17th International Ship and Offshore Structures Congress ISSC 2009. Seoul, Korea:
Seoul Nationa University.
Mountaneas, A., Georgopoulou, C., Dimopoulos, G. & Kakalis, N. M. P. 2015. A model for the life cycle analysis
of ships: Environmental impact during construction, operation and recycling. In: Santos, G. S. (ed.). Taylor &
Francis Group, London.
Murphy, A. J., Landamore, M. & Pazouki, K. 2013. Modelling Ship Emission Factors and Emission Indicies. Low
Carbon Shipping Conference
Murray, C. J. L. & Lopez, A. D. 1996. The Global Burden of Disease: A Comprehensive Assessment of Mortality
and Disability from Diseases, Injuries, and Risk Factors in 1990 and Projected to 2020. . Cambridge, MA: .
Global Burden of Disease and Injury Series. Cambridge, MA: Harvard University Press.
Nappi, N. S. & Collette, M. Structural Design of Naval Vessels: Recent Developments and Emerging Challenges.
American Society of Naval Engineers Day 2013. , 2013.
Nash, C. 2003. UNITE (Unification of Accounts and Marginal Costs for Transport Efficiency). Final Report for
Publication. University of Leeds, Leeds: ITS.
Nash, C., Matthews, B., Link, H., Bonsall, P. W., Lindberg, G., Voorde, E. V. D., Ricci, A., Enei, R. & Proost, S.
2008. Final Activity Report. GRACE (Generalisation of Research on Accounts and Cost Estimation). University
of Leeds, Leeds: ITS.
Nathwani, J. S., Lind, N. C. & Pandey, M. D. 1997. Affordable safety by choice: the life quality method. Institute
for Risk Research, (Canada): University of Waterloo.
Nato 2014. ANEP 77 Naval Ship Code E, Version 1 ed.
Nordhaus, W. D. 2007. A Review of the Stern Review on the Economics of Climate Change. Journal of
Economic Literature, 686702.
Nordhaus, W. D. 2011. The economics of tail events with an application to climate change. Review of
Environmental Economics and Policy, 5, 240257.
Norway 2000. Formal Safety Assessment - Decision Parameters Including Risk Acceptance Criteria.
Norway 2009. FORMAL SAFETY ASSESSMENT, Analysis of Various Environmental Risk Evaluation Criteria.
MEPC 59/INF.21. IMO.
Norway 2012. Cost-benefit analysis. Official Norwegian Reports, NOU.
Oecd 2011. Future Global Shocks IMPROVING RISK GOVERNANCE. Reviews of Risk Management Policies. OECD.
Oxera 2002. A Social Time Preference Rate for Use in Long-Term Discounting. A report for ODPM, DFT and
DEFRA. ODPM, DFT and DEFRA.
Pearce, D. & Ulph, D. 2005. A Social Discount Rate For The United Kingdom. CSERGE Working Paper School of
Environmental Studies University of East Anglia Norwich.
Pindyck, R. S. 2011. Fat tails, thin tails, and climate change policy. Review of Environmental Economics and Policy,
5, 258274.
Preiss, P., Friedrich, R. & Klotz, V. 2008. Report on the procedure and data to generate averaged/aggregated data.
Deliverable n 1.1 - RS 3a of the NEEDS project. NEEDS: New Energy Externalities Developments for Sustainability.
Preiss, P. & Klotz, V. 2007. Description of updated and extended draft tools for the detailed site-dependent
assessment of external costs. Technical Paper No. 7.4 - RS 1b of NEEDS Project.
Psaraftis, H. N. 2011. Letter to the Editor. Marine Pollution Bulletin.
Psaraftis, H. N. 2012. Formal Safety Assessment: an updated review. Journal of Marine Science and Technology.
Springer.
Psaraftis, H. N. & Kontovas, C. A. 2010. Balancing the economic and environmental performance of maritime
transportation. Transportation Research Part D: Transport and Environment, 15, 458462.
Psarros, G., Skjong, R. & Vanem, E. 2011. Risk acceptance criterion for tanker oil spill risk reduction measures.
Marine Pollution Bulletin.
Pundt, L. & Kruger, S. Development of a Software to Conduct and Analyse Sea Trails with respect to Speed
Verification for the EEDI. 12th International Symposium on the Practical Design of Ships and Other Floating
Structures, 2013 Changwon, Korea. pp. 861868.
Rackwitz, R. 2002a. Optimization and risk acceptability based on the Life Quality Index. Structural Safety.
Rackwitz, R. 2002b. A rational risk acceptability criterion based on the life quality index. Invited keynote lecture.
International Conference on Offshore Mechanics and Arctic Engineering (OMAE2002). Oslo, Norway.
Radaelli, C. & De Francesco, F. 2010. Regulatory Impact Assessment. In: Baldwin, R., Cave, M. & Lodge, M.
(eds.) The Oxford handbook of regulation. Oxford University Press.
Randalls, S. 2010. History of the 2C climate target. Wiley Interdisciplinary Reviews: Climate Change, 1, 598605.
Samuelson, P. 1937. A Note on Measurement of Utility. Rev. Econ. Stud. , 4, 155161.
Scott, M. F. G. 1977. The Test Rate of Discount and Changes in Base Level Income in the United Kingdom. The
Economic Journal, 23.
Scott, M. F. G. 1989. A New View of Economic Growth, Oxford University Press.
Shahriari, M. & Frost, A. 2008. Oil spill cleanup cost estimation developing a mathematical model for marine
environment. Process Safety and Environmental Protection.
458
Shane, F., Loewenstein, G. & Odonoghue, T. 2002. Time Discounting and Time Preference: A Critical Review.
Journal of Economic Literature, 40, 351401.
Shijian, C., Yiqian, L. & Weiqiang, Q. 2013. Comparison analysis between CSR-OT and CSR-H for corrugated
bulkhead of large product tanker. Tanker Structure Cooperative Forum TSCF.
Silenv. 2012. FP7 Project SILENV: Ships oriented Innovative soLutions to rEduce Noise and Vibrations [Online].
Available: www.silenv.eu.
Skjong, R. 2009. Regulatory framework. Risk-based ship design. Springer, 97151.
Skjong, R. & Ronold, K. O. 1998. Societal indicators and risk acceptance. Proc. of the Int. Conf. on Offshore
Mechanics and Arctic Engineering (OMAE1998). Lisbon, Portugal.
Skjong, R. & Ronold, K. O. 2002. So much for safety. International Conference on Offshore Mechanics and Arctic
Engineering (OMAE2002). Oslo, Norway.
Skjong, R., Vanem, E. & Endresen, R. 2005. Risk Evaluation Criteria. SAFEDOR: Design, Operation and
Regulation for Safety.
Smith, T., Day, S., Bucknall, R., Mangan, J., Dinwoodie, J., Landamore, M. & Turan, O. 2014a. Low Carbon
Shipping: A Systems Approach. UK.
Smith, T., Okeeffe, E., Aldous, L., Parker, S., Raucci, C., Traut, M., Corbett, J. J., Winebrake, J. J., Jalkanen, J.-P.,
Johansson, L., Anderson, B., Agrawal, A., Ettinger, S., Ng, S., Hanayama, S., Faber, J., Nelissen, D., T Hoen, M.,
Lee, D., Chesworth, S. & Pandey, A. 2014b. Third IMO GHG Study 2014. International Maritime Organisation.
Sonic. 2015. FP7 Project SONIC: Suppression of underwater Noise Induced by Cavitation [Online]. Available:
http://www.sonic-project.eu/.
Spouge, J. & Skjong, R. 2014. Risk Level and Acceptance Criteria for Passenger Ships. First interim report, part 2.
Stern, N. 2007. The economics of climate change: the Stern review, cambridge University press.
Temple, D. & Collete, M. Optimization of Structural Design to Minimize Lifetime Maintenance Cost of Naval
Vessels. In: Romanoff, G. G. S. J., ed. Marstruct 2013, March 2527 2013 2013 Espoo, Finland. Crc
Press/Balkema, 525532.
Tengs, T. O., Adams, M. E., Pliskin, J. S., Safran, D. G., Siegel, J. E., Weinstein, M. C. & Graham, J. D. 1995. FiveHundred Life-Saving Interventions and Their Cost-Effectiveness. Risk Analysis.
Tervonen, J., Hmekoski, K., Otterstrm, T., Peter, A., Bickel, P. & Schmid, S. A. 2002. UNITE (Unification of
Accounts and Marginal Costs for Transport Efficiency). Deliverable 11 Appendix: Marginal Cost Case Study 9c
Nordic Maritime Shipping. University of Leeds, Leeds: ITS.
Torrance, G. W., Thomas, W. H. & Sackett, D. L. 1972. A utility maximization model for evaluation of health care
programs. Health Serv Res.
Turner, S. 2011. Systemic financial risk: agent based models to understand the leverage cycle on national scales and
its consequences. OECD/IFP Project on Future Global Shocks. OECD.
Uk 2011. THE GREEN BOOK, Appraisal and Evaluation in Central Government.
Un-Wecd 1987. Our Common Future. United Nation World Commission on Environment and Development, - the
Brundtland Commission.
Un 2013. Review of Maritime Transport. United Nations Conference on Trade and Development United Nations.
Undp 1990. Human Development Report. United Nations Development Programme. New York, USA: United Nations.
Unece 2007. PREVENTION OF POLLUTION OF INLAND WATERWAYS BY VESSELS
Resolution No. 21 First revised edition. Inland Transport Committee.
Unece 2009. CEVNI European Code for Inland Waterways (Rev.4).
Unece 2011a. Recommendations on Harmonized Europe-Wide Technical Requirements for Inland Navigation
Vessels - Resolution 61.
Unece 2011b. White Paper on Efficient and Sustainable Inland Water Transport in Europe. Inland Tranport
Committee - WP on Inland water Transport.
Unfccc 1997. Kyoto Protocol to the United Nations Framework Convention on Climate Change. FCCC/CP
/1997/L.7/Add.1. UNITED NATIONS FRAMEWORK CONVENTION ON CLIMATE CHANGE (UNFCCC).
Vanem, E. & Bitner-Gregersen, E. M. 2012. Stochastic modelling of long-term trends in the wave climate and its
potential impact on ship structural loads. Applied Ocean Research, 37, 235248.
Vanem, E., Endresen, . & Skjong, R. 2008. Cost effectiveness criteria for marine oil spill preventive measures.
Reliability Engineering and System Safety.
Vassalos, D. 2012. Design for Safety: Risk-Based Design Life-cycle Risk Management. . 11th International
Maritime Conference on Design for Safety. Glasgow.
Verhulst, M., Duursema, W. & Nieuwenhuis, J. J. EEDI Values of Small General Cargo Ships. 12th International
Symposium on the Practical Design of Ships and Other Floating Structures, 2013 Changwon, Korea. pp. 658665.
Weinstein, M. C. & Stason, W. B. 1977. Foundations of cost-effectiveness analysis for health and medical practices.
N Engl J Med.
Weitzman, M. L. 2009. On modeling and interpreting the economics of catastrophic climate change. The Review of
Economics and Statistics, 91, 119.
Weitzman, M. L. 2011. Fat-tailed uncertainty in the economics of catastrophic climate change. Review of
Environmental Economics and Policy, 5, 275292.
Wenquing, Z., Yinhua, L. & Jun, G. 2013. Impact of Harmonized CSR on Tankers. Tanker Structure Cooperative
Forum TSFC
Yamada, Y. 2009. The cost of oil spills from tankers in relation to weight of spilled oil. Marine Technology

710 SEPTEMBER 2015
CASCAIS, PORTUGAL
VOLUME 1
COMMITTEE IV.2
DESIGN METHODS
COMMITTEE MANDATE
Concern for the synthesis of the overall design process for marine structures, and its integration with
production, maintenance and repair. Particular attention shall be given to the roles and requirements of
computer-based design and production, and to the utilization of information technology.
COMMITTEE MEMBERS
Chairman:
M. Collette, USA
R. Bronsart, Germany
Y. Chen, China
S. O. Erikstad, Norway
P. Georgiev, Bulgaria
V. Giuglea, Romania
H. K. Jeong, Korea
I. Lazakis, UK
L. Moro, Italy
Z. Sekulski, Poland
M. Sicchiero, Italy
M. Toyoda, Japan
M. Ventura, Portugal
V. Zanic, Croatia
KEYWORDS
Design Methodology, Product Lifecycle Model, Optimization, Classification Society, Lifecycle Structural
Management, CAD, Design Software.
460
ISSC committee IV.2: DESIGN METHODS
CONTENTS
1.
INTRODUCTION.................................................................................................................................... 461
2.
DESIGN METHODOLOGY................................................................................................................... 461

2.1 Developments in procedural aspects of ship design methodology ............................................. 462
2.2 Developments in Design-for-X and risk-based design ............................................................ 462
2.3 Developments in ship form-function mapping, tradespace searches .......................................... 465
2.4 Handling uncertainty in future operating context ........................................................................ 466
3.
DESIGN TOOLS ..................................................................................................................................... 467

3.1 Introduction .................................................................................................................................. 467
3.2 Development of design tools........................................................................................................ 467
3.3 Tools for lifecycle cost modeling and lifecycle assessment ....................................................... 469
3.4 Links between design tools and production and operational phases .......................................... 469
3.5 Developments in integrated naval architecture packages ........................................................... 471
4.
OPTIMIZATION DEVELOPMENTS.................................................................................................... 472

4.1 Introduction to Design Support Systems (DeSS) ........................................................................ 472
4.2 Parallel processing and hardware developments ......................................................................... 475
4.3 Developments in structural optimization algorithms (optimization solvers-) .......................... 477
4.4 Surrogate modeling and variable fidelity approaches (surrogate solvers) ............................. 482
4.4.1 Surrogate modeling in design and optimization ........................................................... 483
4.4.2 Surrogate modeling in risk and safety analyses ............................................................ 484
4.5 Optimization for production (design quality modulesPRODUCTION) ......................................... 484
4.6 Optimization for lifecycle costing (design quality modulesLCC) ............................................ 486
5.
CLASSIFICATION SOCIETY SOFTWARE REVIEW ....................................................................... 487

5.1 Background, motivation, and aim ................................................................................................ 487
5.2 Tool analysis ................................................................................................................................. 488
5.2.1 Overall functionality ...................................................................................................... 488
5.2.2 Evaluation Criteria ......................................................................................................... 488
5.3 Classification societies tools details............................................................................................. 490
5.3.1 American Bureau of Shipping (ABS)www.eagle.org ................................................ 490
5.3.2 Bureau Veritas (BV)www.bureauveritas.com ............................................................ 491
5.3.3 China Classification Society (CCS)www.ccs.org.cn .................................................. 491
5.3.4 Croatian Register of Shipping (CRS)www.crs.hr ....................................................... 492
5.3.5 DNVGL ........................................................................................................................ 493
5.3.6 Korean Register of Shipping (KR)www.krs.co.kr ...................................................... 495
5.3.7 Nippon Kaiji Kyokai (ClassNK)www.classnk.com ................................................... 496
5.3.8 Polish Register of Shipping (PRS)www.prs.pl ........................................................... 497
5.3.9 Registro Italiano Navale (RINA)www.rina.org .......................................................... 498
5.4 Conclusions and future challenges............................................................................................... 498
6.
STRUCTURAL LIFECYCLE MANAGEMENT .................................................................................. 499

6.1 Introduction .................................................................................................................................. 499
6.2 Tool development ......................................................................................................................... 500
6.3 Data interchange and standards.................................................................................................... 502
6.4 Integration with repair .................................................................................................................. 503
6.5 Integration with structural health monitoring systems ................................................................ 504
6.6 Summary of the lifecycle structural management systems ......................................................... 506
7.
OBSTACLES, CHALLENGES, AND FUTURE DEVELOPMENTS................................................. 506
8.
CONCLUSION ........................................................................................................................................ 508
ACKNOWLEDGMENTS................................................................................................................................ 509
REFERENCES ................................................................................................................................................. 509

1.
461
INTRODUCTION
Structural design methods continued to increase in both scope and fidelity over the time period covered
by the 19th ISSC. The committees work revealed progress along themes established by previous ISSC
IV.2 Committees. Advances addressing lifecycle modeling, lifecycle data management, risk-based design,
and optimization continue to appear in both design methodology and design tools. In the 19th ISSC, the
Committee elected to revisit a survey of classification society software packages first conducted during
the 14th ISSC in the year 2000. Classification societies continue to play a central role in the development
and maintenance of ship structures from design inception through disposal. Thus, their software offerings
are frequently deeply involved in the design and operational support of ship structures. In the intervening
15 years since the first survey of these tools, finite element analysis (FEA), spectral fatigue assessment,
and lifecycle data management have all grown significantly in prominence. Not surprisingly, these
capabilities are now woven into the software packages of most classification societies. Chapter Five of
this report provides an overview and comparison of ten societies offerings.
The preceding IV.2 Committee chose to organize their report around the concept of lifecycle
management (LCM), with a strong focus on data management. The current Committee continued this
investigation with an update on developments in this area in Chapter Six and with additional, relevant
discussions about design tools in Chapter Three and a classification society review in Chapter Five. The
current committee also chose to expand the focus beyond data management to include design approaches
for lifecycle performance, structural monitoring approaches, and integration with repair. These are all
topics where the committee saw exciting, but generally preliminary, initial studies. The Committee
foresees rapid future growth in these topics.
Probabilistic modeling, risk-based design, and the inclusion of uncertainties in design continue to
receive attention as well. Sustained efforts to include more realistic safety scenarios and sources of risk in
early-stage and detailed structural design are apparent throughout the literature. Such modeling now
routinely goes beyond safety and risk topics, with scenario-based design and epoch-era analysis
techniques also continue to be explored by the community. Chapter Two presents a comprehensive
review of recent advances and literature in these areas.
Design methods are undoubtedly influenced by the available design tools. Perhaps no tool is more
discussed than optimization. The Committees review reveals that larger and larger optimization problems
are becoming computationally practical. Building off the overall optimization strategy first presented in
the last ISSC, the Committee continued its discussion of the role of optimization in design and the types
of advances in optimizers, computational hardware, surrogate modeling, and links to lifecycle cost
optimization and production optimization in Chapter Four. Design tools, including links to lifecycle
costing and analysis, are reviewed in Chapter Three.
While the Committees report primarily focuses on the impressive progress made during the last three
years, important gaps still remain. These have been identified throughout the Committees report and then
summarized in Chapter Seven. Struggles around data transfer, intellectual property (IP) rights,
management of monitoring data, and providing realistic lifecycle modeling parameters during design are
all retarding wider implementation of LCM approaches. The continued growth in the number, size, speed,
and fidelity of numerical modeling and optimization approaches is also resulting in more pressure to
develop solid data links, tool integration, and design methodology. Finally, while both risk and
uncertainty remain topics of intense interest, relatively few risk or uncertainty approaches have been
codified into rules, regulations, and design methodologies to date. This gap between what is desired for
applied design and what has been demonstrated at a proof-of-concept level appears to be more difficult to
bridge than first anticipated.
2.
DESIGN METHODOLOGY
What actually comprises design methodology? Based on existing research literature in ship structural
design, it is apparent that the term design methodology is used to denote several disparate aspects of the
complete ship design process. One aspect is the procedural aspect of design, in which a design
methodology prescribes a structured sequence of activities, starting with an initial statement of needs and
ending with a design specification. An alternative interpretation of design methodology is related to the
predominant performance goal that is the focus of the process, often captured in the term Design-for-X
(DfX). Here, the X normally represents a specific goal such as operability, environment, safety, or
production. A third aspect of design methodology relates to various strategies for handling the two-way
mapping between the form space and the function space related to design, that is, identifying basic
decision support methods that bring a designer from a set of needs and requirements all the way to a final
design description. In the remaining sections of this chapter, each of these aspects will be discussed in
more detail.
462
2.1

Developments in procedural aspects of ship design methodology
Perhaps the most well-known, classical design model is the design spiral, first presented by Evans
(Evans, 1959). Though often used as a process description, the design spiral is often regarded more
valuable as an abstract model, capturing the sequential, iterative nature of the design process. The spiral
approach assumes a relatively smooth process of balancing out conflicting requirements. This
idealization of the design process is challenged in recent studies of naval ship design processes,
pointing to the fact that the design search space contains important regions of cliffs and plateaus in the
functional relationships between core capabilities, size, and cost (Andrews et al., 2012b). Further, the
design spiral is generally criticized for locking the designer into his first assumptions. As a
consequence, the System-Based Design methods (Levander, 2006) have emphasized the development
of a functional structure transforming the customer needs and high-level requirements into a relatively
detailed definition of specific functional requirements. In this way, the specification of form elements is
postponed until a fairly well balanced solution can be proposed. This follows the basic principles of the
German VDI-model (Pahl et al., 2007) and has been applied to many ship types, predominantly cruise
vessels, ferries, and most recently, offshore service vessels (Erikstad and Levander, 2012). More recent
additions include the building-block design proposed by UCL (Andrews, 2003a) and methods adopted
from the systems engineering domain, such as the Responsive Systems Comparison method (Ross and
Rhodes, 2008).
Other process models put more emphasis on the creative and innovative elements, especially in the
early design stages. There has been a discussion whether stringent methodologies and advanced tools will
set new constraints, rather than enhance innovation, (Koelman, 2013) on lines generation. Also, the
general availability of 3-D prototyping in the early stages is likely to influence the choice of design
process. These models might, on the one hand, lock the designer into first assumptions, while on the other
hand, provide a platform for a visual, fast-feedback design sketching environment (Alonso et al., 2013,
Koelman, 2013) in which the designers experience and preference for design style can be expressed
(Pawling and Andrews, 2011). By connecting parametrically defined building blocks in the 3-D model
with their corresponding functions in a system-based model, changes in customer needs and functional
requirements can be seamlessly forwarded to updates in one or several design templates (Erikstad and
Levander, 2012). This can be further combined with decision support systems for space configuration,
layouts, and arrangements, using both spatial (Oers, 2011) and non-spatial models as well as topological
models (Gillespie et al., 2012, Parker and Singer, 2013).
2.2
Developments in Design-for-X and risk-based design
A summary of the most recent developments related to the design for specific performance aspects, also
known as Design-for-X (DfX), is given by Andrews et al. (Andrews et al., 2012a). From a structural
design point-of-view, the most relevant DfX aspects are design-for-production and design-for-safety.
Both of these will be handled separately later in this report. Furthermore, design-for-environment and
design-for-energy-efficiency have received considerable attention the last few years. In Hirdaris and
Cheng (2012), a number of alternative technology options are reviewed, and DNV looks into different
options for CO2 reductions (Hoffmann et al., 2012).
In general, DfXs concept puts the emphasis on the performance achievement, and, at least in principle,
has no specific requirements towards the specific design solution or the design process to be followed.
Thus, DfX is closely related to the current trend towards goal-based design methodologies in general
(Vassalos, 2012), and risk-based design in particular (Lee et al., 2010, Boulougouris and Papanikolau,
2013), endorsed by recent International Maritime Organization (IMO) regulations.
IMO and the merchant maritime industry have made advances in proposing safety standards for merchant
ships by adopting pro-active safety measures for future rules and regulations in the frame of a holistic
approach to ships safety. IMO and International Association of Classification Societies (IACS) decided to
move from prescriptive concepts to probabilistic assessment methods and goal-based standards (GBS).
Therefore, a risk-based approach and its assessment methods have become important design tools, especially
in the shipbuilding industry, by facilitating the accomplishment of safety objectives in a cost effective
manner. A risk-based approach is generally recognized as a proactive and rational method for handling the
safety and environmental issues as well as enabling accelerated innovations and the introduction of new
technologies in ship design. As some on-going research efforts continue in this regard, the risk-based ship
design approach is applied to new concept container vessels and CO2 carriers, and from this, standardization
framework for risk-based ship design is proposed, as shown in Figure 1 (Lee et al., 2012a).
463
Figure 1. Proposed standardization framework for risk-based ship design (Lee et al., 2012a).
Currently, there are challenges, especially for the maritime industry, to cross the bridge from rules-based
to risk-based design. This means that, although there are near enough scientific and technological
research findings available for risk-based design to be fully implemented in the maritime industry,
industrialization of processes, design tools, techniques, data, regulations, and approval still require careful
attention. This implies that the implementation of the risk-based design in ship design requires a lot of
time and efforts. Additionally, perhaps most importantly, the actual need of a risk-based ship design
approach must be accepted by all stakeholders in order to facilitate the transition from prescriptive to
goal-setting regulations. A practical example of the application of a risk-based design approach in the
design of a cruise ship is reported in Vassalos (2012). Here, risk-based design is understood as the
integration of risk assessment in the ship design and decision-making processes, as shown in Figure 2.
Figure 2. A high level framework for risk-based design (Vassalos, 2012).
Risk is the product of the frequency of an event and the associated consequences. Risk-based design
approach is an improved alternative to the traditional design process, as it integrates safety as additional
design objective. Thus, the designer has to justify that the risk of any feasible design is less than or equal
to the specified acceptable risk. Risk categories, such as system failure, human life, environment, or
property, can be varied, and these are treated separately. The total risk is obtained by the sum of partial
risks from different damage categories such as explosion, fire, collision, and grounding. Each partial risk
can be computed with the aid of risk models like Fault Tree Analysis (FTA), Event Tree Analysis (ETA),
or Bayesian networks.
Risk-based design concept is applied to the assessment of the damaged stability of a generic navy
vessel. In this study, numerical simulation tools are implemented in modern risk-based assessment
methods for assessing risk after flooding. Survivability of the vessel is calculated using a mathematical
464
risk model, and as a result, an attained survivability index can be calculated and used as a reference for
setting the acceptable risk level (Boulougouris and Papanikolau, 2013). Risk-based approach is also used
to generate the risk matrix of an LNG cargo handling operation. A hazard analysis of the LNG cargo
handling devices, such as valves, the piping system, and the control system, is performed prior to risk
analysis. Using the results, maintenance, and operational recommendations for reducing probability of
failures of LNG cargo handling devices are presented (Roldan and de Souza, 2012).
The risk-based design project RISPECT (Hifi et al., 2012) provides a methodology that combine a
detailed analysis of long-term experience from a large number of ships with reliability or risk-based
methods to work out useful and justifiable risk-based inspection plans. The goals of the project are better
inspections, the detection and repair of more important defects, fewer pollution incidents, and the savings of
lives. The RISPECT inspection planning tool is intended to be used by the shipping industry. It aims to
combine the traditional experience-based, class rule driven inspection method with first principles,
reliability based, and inspection methods. By developing this combination it aims to demonstrate an
improved decision making method for safe, cost-effective structural inspection and repair of existing ships.
The use of advanced computational tools allows the quantification of the risk level of a particular
design and its exhaustive comparison with alternative designs. In this aspect, risk is no longer a constraint
but a measure of safety performance and design objectives, which can be used in an optimization
procedure. This can become optimal ship design for minimum risk, having rational efficiency and
economy computational performance. For example, the lifecycle structural optimization of the midship
section of a double hull tanker, considering multiple risks such as oil outflow, CO2 emissions, and failure,
is performed (Kawamura et al., 2013). In this case, the lifecycle benefit is used as the objective function
of the optimization, in which the various risks are evaluated as a part of the lifecycle cost.
There are research outcomes from the risk-based approach which contribute to the implementation of
risk-based design methodology in ship design. From them, the safety assessment process, based on
previous design process, supporting tools, and standardized framework, will help to establish risk-based
design concept and advance the safety level of the new types of ships. Such benefits can be obtained for
environmental protection as well as promoting the development of safety level approach based goal-based
standard from IMO. Related to this, one of the research outcomes is the introduction of a time-based ship
safety assessment model (SSAM) and SSAM-based safety assessment framework as a supplement to use
the IMO formal safety assessment (FSA) and the gains of SAFEDOR (Kang et al., 2012). From IMO
FSA, it is possible to obtain a balance between the various controversial, technical, and operational issues.
From the SAFEDOR project, the meaning of a safe ship design has been clarified. The SSAM describes a
ships mission-based functional hierarchy including not only functions, systems, and subsystems but also
their environmental context, organizational and management infrastructure, personal subsystems, and
technical/engineering systems on a time basis. Safety assessment framework can be used for safety
assessments and for selecting the best risk control options for a novel ship design if designers have an
objective and concretized generic model.
Reliability-based approaches to the marine industry are active to support and provide one of the
fundamental elements of the risk-based ship design capability.
In case of subsea applications, the reliability assessment of subsea pipelines is conducted to prepare
their risk assessment procedures and related decision making processes (Gazis, 2012). Also, the
reliability-based load and resistance factor design (LRFD) method-based composite pressure vessel
design procedure for subsea blowout prevention is studied. Target reliability is set to establish LRFDbased composite pressure vessel code. It determines the probability of failure of the vessel subjected to
external hydrostatic pressure (Cai et al., 2011).
For naval applications, the probabilistic approach is used for naval ship structures to build integrated
lifecycle maintenance, monitoring and managing under the uncertainty of multiple and conflicting
objectives (Frangopol et al., 2011). Efforts are also made to reduce the uncertainty in the performance
assessment of ship structures by updating the wave-induced load effects with the information acquired
from structural health monitoring. The reliability analysis is performed with respect to the midship cross
section of the Joint High Speed Sealift (Zhu and Frangopol, 2013, Zhu and Collette, 2013)
For ship structural material and member cases, the probabilistic approach is used for composite
laminated structures, aiming to find optimal use of them in non-high speed light craft (Sanchez et al.,
2011). A reliability approach is used to conduct the ultimate strength assessment of aging steel plates
subjected to compressive load and randomly non-uniformly distributed corrosion wastage. In this study,
the reliability of a series system composed of two time variant events, corrosion deterioration and
ultimate strength assessment of a rectangular plate is studied. It was assumed that the correlation between
them was unknown (Silva et al., 2012). The Bayesian network and a reliability approach are utilized, and
as a result, a new reliability purposed dynamic algorithm for the Bayesian network is presented to deal
with the fatigue cracking problems of the stiffened panels (Zhu and Collette, 2013).
465
For offshore applications, a rational reliability assessment procedure for hull girder ultimate strength
assessment of ship-shaped FPSOs is investigated. Four ship-shaped FPSOs are considered as the case
study, and the effect of the return period of VWBM, environment severity factor, and corrosion effects on
hull girder reliability index are investigated (Chen et al., 2013b). Also, a methodology for fatigue
reliability-based design optimization with respect to bending stiffener, the key structural member of
umbilical/flexible risers to protect the upper connection against damage from over-bending, is introduced
(Yang and Wang, 2012).
Implementation of HiP-HOPS in the reliability and availability analysis of ship machinery systems,
such as propulsion and electric distribution systems, are studied (Ruede et al., 2012). Here, HiP-HOPS is
a compositional, model-based safety analysis technique. Use of HiP-HOPS permits the rapid evaluation
of multiple design variants and design modifications, especially when compared to the development and
analysis of fault trees separate from the graphical system model.
2.3
Developments in ship form-function mapping, tradespace searches
Fundamentally, the concept of design can be defined as a mapping from the function space that defines
needs and requirements, to a form space containing the description of the final design (Coyne et al.,
1990). The function-to-form mapping is a fundamental aspect of design and is sometimes contrasted to
engineering analysis, which is fundamentally aimed at the opposite process of deriving the function
(performance) from a given form (description).
To support this mapping, from needs and requirements to preferable solutions in the design space, there
are several competing strategies. One is optimization, where heuristic and nature-inspired methods in
particular, such as genetic algorithms, ant-colony optimization, and simulated annealing, have received
much attention (Collette, 2014). Optimization methods will be discussed later in this report.
An alternative approach is set-based search strategies (McKenney et al., 2011, McKenney et al., 2012).
Set-based design has been combined with a rule-based configuration for arrangement optimization in the
conceptual stage for both naval and offshore vessels (Oers et al., 2007, Oers, 2011). Designing ships
inside-out, the importance of functional development (Keane, 2012).
Since the failure of artificial intelligence to live up to its expectations in the 1980s and 1990s, we have
recently seen renewed interest in using knowledge-based systems in the design process. In the work of Cui
and Wang (2013), experiences of design experts, design rules, and successful previous designs are stored in
a knowledge base. Other contributions are intelligent arrangement design for naval vessels (Daniels et al.,
2011), knowledge-based systems used in structural design (Yang and Wang, 2012), and a prediction model
based on neural networks, which is trained with data of optimized structural arrangements and is capable of
estimating the structural properties for different hull geometries (Chaves et al., 2014).
From a design perspective, the form-to-function mapping is basically about analysis methods that allow
for a fast and efficient, yet accurate, evaluation of specific design alternatives as part of an overall design
process. Throughout the last decade, these analysis methods have developed considerably, where highfidelity, first principles tools, such as computational fluid dynamics (CFD) and advanced finite element
analysis (FEA), have moved from the research institutions to become an intrinsic part of everyday design
processes. Fundamentally, there is a tradeoff between the fidelity level of the model and resource
expenditure. This corresponds to a more fundamental tradeoff between breadth and depth in the search for
design solutions and is thus an intrinsic part of the design process (Zheng et al., 2013). From a normative
perspective, one should generally start with lower fidelity model in the conceptual stage (go broad) and then
migrate towards higher fidelity models in the detailed design stages (go deep). However, maintaining a
rational approach to this tradeoff throughout the design process is often difficult since we have few explicit,
quantitative measures of these tools and model properties. The wide adoption of CFD and FEA tools has
contributed to a tendency towards too high fidelity models in the early stages, thus missing potentially
preferred solutions in other parts of the design space. Empirical prediction methods that have proved their
usefulness for decades, such as Holtrop and Mennen resistance prediction methods or Schneekluths weight
approximation methods, have not been updated for several decades (Koelman, 2013).
To alleviate this situation, there are two main strategies available. The first strategy is to shift the
Pareto frontier that represents this tradeoff outwards, for instance, by allowing the more efficient,
seamless integration of high fidelity tools into CAD software (Yu et al., 2012, Cui and Wang, 2013,
Roberts et al., 2013); or by allowing more efficient searches through the design space, simply by updating
key empirical methods; or by applying surrogate modeling methods (Collette, 2014, Prebeg et al., 2014),
including response surfaces (Pajunen and Heinonen, 2013, Chaves et al., 2014), and experimental design
methods (Zheng et al., 2013).
An alternative strategy is to support a better-informed balance between model fidelity and resource
expenditure by associating available tools and methods with relevant metrics. One example of this is the
466
hydrodynamic optimization of catamarans regarding the choice between high fidelity (URANS) and low
fidelity (potential flow) methods.
A somewhat different approach to using simulation for analyzing system performance is to use discrete
event models that capture the complex operation of a ship. These have been used for both deriving a more
detailed and realistic operational profile for the vessel (as opposed to idealized design cases), as well as
for aggregating lifecycle performance measures such as energy efficiency, operability, and safety. This
will typically call for an extensive modeling of the operating context of the vessel such as metocean, fleet
logistics, and ice. This approach has been applied to the design of LNG transport (Erikstad and Ehlers,
2014), arctic transport (Bergstrm et al., 2014), the design of ships in rouge sea states (Bitner-Gregersen
and Toffoli, 2014), and for ship transport energy efficiency (Coraddu et al., 2014). This reflects a general
extension of system boundaries, analyzing ship performance as a part of larger operation or a fleet/logistic
chain (Andrews, 2003b, Hagen and Grimstad, 2010, Ulstein and Brett, 2012).
2.4
Handling uncertainty in future operating context
When designing ocean engineering systems, such as ships, platforms, and transportation services, there is
a considerable degree of risk and uncertainty related to key aspects of the systems future operating
context. The total size of the market is highly dependent on external exogenous factors, such as GDP
growth rates and oil prices, affecting both the size and structure of the market as well as freight and day
rates. Future environmental regulations, in particular, the location of emission control areas, will be
important for the selection of fuels as well as other emission controls; and rapid technology development
may render design solutions with non-optimal configurations non-competitive after only a few years.
Thus, the design of a new vessel needs to strike a balance between optimizing the vessel for the (likely)
first contract or trade entered, while investing in additional functionality and performance capabilities to
meet future requirements and changes in its operating context. This could be increased fuel prices, stricter
environmental regulations, or new and more cost-efficient technology in more northern and deeper
waters. To further add complexity, these additional capabilities might either be made part of the vessel at
the design time, or they may be provided as design options, to be executed depending on information that
will only be available in the future. Thus, there has been a need for developing efficient methods to be
used for the conceptual design of innovative ship solutions, able to deliver sustained value to stakeholders
over time in a complex, uncertain, and changing lifecycle.
Though shipping markets have always been volatile, the traditional approach in marine design has been
to assume a fixed set of requirements to which the final design must comply. To the extent that alternative
scenarios for the future operating context have been considered relevant, the most common response has
been to design towards the most likely scenario and subsequently analyze the sensitivity of the final
solution towards variations in key operating parameters.
This is not to say that flexibility is not being valued in the maritime sector. We have seen many examples
of versatile vessels able to serve multiple cargoes, services, and markets. OBO carriers, car carriers with
hoistable decks, and offshore vessels with multi mission capabilities are typical examples. To some extent,
vessels and systems are designed with additional capabilities that go beyond the immediate requirements.
This implies hedging towards changes in the operating context that may render the vessel less competitive in
the future while at the same time, providing a level of specialization that can be a liability in a second-hand
market that puts a premium on standard solutions. Also, options are routinely provided as parts of
specifications and build contracts, setting a predetermined price on changes that may be exercised in the
future. However, the quantitative valuation of flexibility and options is a challenge, both with respect to
determining the optimal capability level, the price of options, and for conveying the operating value of these
additional capabilities and options to prospective customers. Thus, the approach taken today can best be
described as predominantly a gut feeling approach by key decision makers.
Methods for quantitative risk management was first developed in the financial sector, used to design
client portfolios with a predetermined risk profile. This was further developed into a theory for pricing
financial options, using the Black-Scholes formula to determine the time-dependent option price (Black
and Scholes, 1973). The financial market approach to handling uncertainty and risk was later adapted and
applied to engineering systems using real options analysis. Real options made it possible to put a value on
future opportunities to change or expand the capabilities of a design, switching between markets, or
retrofitting components and technology. In the shipping sector, real options analysis has been used, for
instance, for the valuation of combination carriers (Sdal et al., 2008), the analysis of options in
shipbuilding contracts (Hegh, 1998), and naval ship design and acquisition (Gregor, 2003).
An alternative path for design decision-making under uncertainty is stochastic optimization. This
approach extends a more traditional deterministic optimization by explicitly modeling alternative future
scenarios and corresponding probability distributions, and then explicitly taking into consideration
opportunities to change the design or project at later stages (two-stage/multi-stage models). Marine
467
technology applications include the design of emission controls for ships (Balland et al., 2013), uncertain
environmental policy (Niese and Singer, 2013), and vehicle routing problems (Fagerholt et al., 2010).
Further, the proposed research area will share important characteristics with the emerging field of RiskBased Design (Papanikolaou and Soares, 2009) in taking a probabilistic approach towards modeling
future events and scenarios (Wagner and Bronsart, 2012).
A broader engineering systems perspective on designing for flexibility is provided in the book
Flexibility in Engineering Design (de Neufville and Scholtes, 2011). Here, a combined theoretical and
practical framework is provided for identifying, analyzing, and implementing flexibility in a broad range
of large-scale engineering systems. Other work has integrated design for flexibility and real options
evaluation as part of a more complete systems design process (Ross et al., 2008, Beesemyer et al., 2011),
which has further been adapted to ship design (Gaspar et al., 2012).
3.
DESIGN TOOLS
3.1
Introduction
The number of available computer design and analysis tools available to the naval architect has expanded
since the preceding ISSC Congress assembled in 2012. Several trends impact the development of such
tools. There is a persistent desire to increase the fidelity of modeling throughout the design process while
simultaneously trying to push such high-fidelity modeling into early stages of design. At the same time,
recognition of the value of the digital models created by such design tools is growing. Efforts continue to
leverage this value by making data from such models available in new contexts, including in the shipyard
and throughout the vessels service life. Finally, there is a struggle with fragmentation of such tools as
their number and focus expand. The combination of these factors has driven both feature addition to
monolithic design software packages and classification society toolsets as well as long-term investment in
more distributed architectures. In turn, this has continued to support investments in data exchange formats
for use during and after the vessels design is completed. Implementation of concurrent engineering
approaches has made design data available on the shop floor via lightweight viewers and augmented
reality devices. These developments are reviewed in this chapter.
3.2
Development of design tools
During the time since the last ISSC Congress, the overall state of computer-aided ship design (CASD) in
the marine industry has been reviewed by several authors. A recent review has been presented on the
underlying numerical techniques used in hydrodynamics, structures, and production simulation (McNatt
et al., 2013). More applied reviews have also been published, including one focused on the engineering
simulation tool used primarily within the GL Group (Bertram and Peric, 2013) and a review of the
MAESTRO code and history of computerized ship structural design (McNatt et al., 2013). A special issue
of the journal CAD was also produced, highlighting in-part developments in structural design and
outfit/production planning via design tools (Kim et al., 2012). As shown in these works, continued
development of high-fidelity simulation and smooth data exchange between different software packages
and different phases of the vessels lifecycle remain key focus areas for the marine industry. The
continued use of increasingly detailed simulation in structural design, with extensions into collision, fire,
and blast simulations, is expected in the future. A key challenge in industrial projects is in ensuring that
an appropriate level of modeling is used, supported by necessary experts in conducting such modeling.
The widespread adoption of advanced simulation methods in design has also resulted in a number of
shortcomings being identified to date. Notably, continued unhappiness with the industry-standard
NURBS modeling of hull surfaces remains (Koelman, 2013, McNatt et al., 2013), owing to complications
ranging from handling complex geometry to mathematical limitation in the NURBS formulations that
make automatic data processing difficult. A method of handling this complex surface information for
automated finite element meshing was recently presented (Wang et al., 2014), highlighting both the
power and complexity of NURBS. Despite these misgivings and the development of other approaches,
such as subdivision surfaces and T-splines, no clear alternative has been established. Several authors have
also noted in general terms the challenges of the increasing number of often-fragmented design
approaches (Ulstein and Brett, 2012, Koelman, 2013) and the growing concern that the increased focus on
analysis is producing minimal returns in the performance of the design (Ulstein and Brett, 2012). This
echoes the concerns raised in the conclusion of Bertram and Peric (2013), where the case for a careful
blend of appropriate modeling strategies and expert users is noted.
The past three years have seen extensive development in linking multiple design tools together to
conduct a more comprehensive structural evaluation of new ships and offshore structures. These
approaches have involved both distributed concepts, which included the linking of independent programs
into a larger framework, and the monolithic program concept, where development of a monolithic
program that handles several linked calculations for a specific vessel or platform type was attempted.
468
Both approaches involve tradeoffs. In the distributed approach, middleware is needed to link existing
codes together and handle data translation, and often times some type of data processing such as
interpolation to transfer CFD loads to finite element models. Such middleware takes energy to create and
maintain, and deep linking into the data structure of the individual programs is often not possible.
However, developing monolithic tools often requires re-creating existing analysis and simulation
capabilities, which can be cost-prohibitive. Thus, it is not surprising that the reported new monolithic
tools appeared primarily for newer areas, such as wind turbines (Thomassen et al., 2012), where longstanding, existing analysis packages may not exist.
Several ongoing efforts at creating the middleware for distributed approaches have been updated in the
past three years. The Lloyds Register DIME (Data Interface Management System) is one proposed
middleware approach that was reviewed in detail in the 2012 Committee IV.2 report. This system allows
translation between complex CAD/CAM ship product models, and it meshes and simplifies the geometry
required for analysis calculations through a mix of semantic and geometrical processing routines. Recent
extensions to new naval ship design tools in the LR/MARTECH suite have been reported (Pegg et al.,
2012, Roberts et al., 2013) along with additional information about DIME. An online middleware
solution for small and medium Korean shipyards was also presented (Shin et al., 2012). In this case, a
web-based system linked together databases of previous designs, hull form modifications, calculation
tools, and design information management. STEP AP215 output is planned for this system in the future.
Nupas-Cadmatic is working on a building block approach capable of linking many different codes in a
product data model that is focused on both re-use of previous designs and support for a shipyard company
with multiple shipyards around the world (de Jonge and Kramer, 2014). Nupas-Cadmatic notes that they
explicitly support the distributed tool vision in this architecture.
An area where such linkage has been particularly active is linking the output of either 3-D potential
flow codes or CFD codes to finite element analysis. This presents a challenge as the mesh domain in the
fluid is either usually substantially larger (potential flow) or smaller (viscous flow) than a global FE
model of vessel, and thus, an interpolative framework is necessary in addition to raw data translation.
Recent advances in this area from both the Maestro code (McNatt et al., 2013, Zhao et al., 2013) and
within the GL group for the next generation of large, flexible containerships have recently been presented
(Payer and Schellin, 2013) as well as a similar shipyard-based approach for fatigue (Won et al., 2013). In
general, these approaches are built into marine-specific structural tools, although middleware translators
can be used to couple to generic, finite element packages.
Bespoke and one-off middleware solutions also appear frequently in publications and have been shown
to provide encouraging results for complex design studies. These approaches typically build off of
published program APIs or standard file structures such as XML files. Such published interfaces make the
creation of custom middleware more cost-effective. Zanic demonstrated a decision support framework by
linking CAD/CAM code NAPA Steel with the analysis code Maestro and the optimization code Octopus
for the design of small, open-topped general cargo ships (Zanic et al., 2013c). Danese and Aasen (2013)
presented an approach to tracking weight data across distributed design, purchasing, and other enterprise
activities with ShipWeight, noting that a distributive approach is often necessary, as all-in-one software
for naval-specific CAE applications are lacking. At a more detailed level, Ruy et al. (2012) developed a
Python-based automated system to request cut-outs in structural members and run distributed systems
using Tribon M3s API. In this approach, the outfitting engineers can request specific cut-out locations
and topologies. These requests are tracked and forwarded to the hull structure group who can approve the
location. Then, the cut-out is automatically added to the ship product model. Similarly, Rox and Astrup
(2013) reported on an XML-based approach to link the initial steel design in the German design package
E4 to rule-approval in DNVs Nauticus Hull. Liu and Jin (2012) reported on linking Tribon to other CAE
applications via Tribons XML file. Such industry-specific custom solutions clearly continue to be an
important component of the ship design information flow. All the solutions mentioned here make
extensive use of standard technologies for interfacing (e.g. Python API scripts, SQL databases, XML),
which indicate the value of including standard data storage routines, even in custom software.
Feature extraction from CAD/CAM databases for analysis remains a topic of interest for the industry,
especially for costing approaches. Recently, Caprace (2011) proposed a novel, bottom-up, feature-based
costing approach for vessels that relies on automatic extraction and identification of structural features, such
as cut-outs, weld lines, and bevels, from CAD/CAM data. With the expanding use of such modelling in early
stage design, it is hoped that such an approach can provide more realistic costing earlier on. Critically, if such
a system can be automated or moved to near-real-time, cost feedback can be given to the designer to reduce
the cost of proposed structures during the design stage. Similar work was also reported in Japan (Goldghate
and Yamaguchi, 2012), with a system for tracking design throughout the shipyard design process.
Structural fatigue assessment is a complex design task, especially when the direct-calculation spectral
method is used. In this case, stress RAOs for all the fatigue-critical structures must be determined for
469
numerous speed, heading, and loading combinations via linked seakeeping and FEA models. Then, fatigue
calculations must be made over a wide range of operating conditions. While basic guidance on this
calculation and some software support have been available for over a decade now, usage has not been
widespread. Primarily, the complexity and difficulty of the calculation, even with existing tools, have been
noted as the reason for this situation (Elsen, 2013, Hunter et al., 2013, Won et al., 2013). Not surprisingly,
efficient spectral fatigue design tool development remains a subject of interest. Recently, Won et al. (2013)
reported on a shipyard-specific tool for conducting spectral fatigue assessment of proposed vessels that
intentionally provides limited options and flexibility, compared to existing classification society packages,
to reduce both analysis time and the chance for analyst error. In addition to automating the load-to-FEmodel transfer process as noted above (McNatt et al., 2013, Payer and Schellin, 2013, Zhao et al., 2013),
this code automates much of the hot-spot stress calculation procedure for numerous structural details, and it
contains as few user-settable options as practical to ensure a fast, consistent application of the yards fatigue
life methodology. The MAESTRO suite has also been extended to support fatigue calculations, with a
detailed overview of the new calculation procedure and example presented at COMPIT (Hunter et al.,
2013). GL also reported on a re-engineering of their spectral fatigue toolchain that removes friction points
and difficult operations to reduce the time required for analysis (Elsen, 2013). Another approach attempted
is to simplify the method, on which Chen et al. reported. Relying instead on a simplified procedure from
DNV (Chen et al., 2012), this development dispenses with the fine-mesh FEA component of the analysis.
Notably, all of these publications do not focus on increased fatigue assessment accuracy or fidelity as their
primary contribution, but instead promote ease-of-use.
While much work of the past three years has focused on code linking and automation, development of
stand-alone structural design tools continues. Primarily, these are simple tools designed either for rapid
feedback during early-stage design or for helping with specific structural tasks where data re-use and
integration is less critical. Addressing early stage design, Stenius et al. (2011) reported a simple, rulebased approach for developing minimum weight designs for high-speed craft. This was used to compare a
number of steel, aluminum, and composite concepts during the early design stage and to quantify the
weight impact of structural configuration and material choices. Yang and Wang (2012) reported on the
development of a knowledge-based engineering (KBE) system to perform efficient design of deck
structures. Making production decisions easier and more informed, Dong et al. (2012) proposed a simple,
math-based producibility evaluation approach for titanium structures. BVB Caf is a more comprehensive
stand-alone tool. While it has links to several external solvers for 2-D CAD drawings and FEA, it
primarily attempts to provide a more efficient geometry-to-structure CAD solution. Further developments
with this code have recently been highlighted in COMPIT 2014 (Zagkas and Bralic, 2014). Given the
increased focus on distributed solutions, it seems clear that stand-alone tools can grow in concert with
more integrated design approaches by following the lead of programs such as BVB Caf.
3.3
Tools for lifecycle cost modeling and lifecycle assessment
Reports of successful tools for lifecycle cost (LCC) modeling and lifecycle assessment (LCA) of
environmental impacts remain sparse. This is at odds with the increased level of interest being generated
around each of these topics. A further discussion of LCC and LCA approaches proposed during the past
three years is contained in Chapter 6 of this report. ABB reported on a large effort to bring design prediction
to onboard decision support tools and return the link to the owners office via fleet analysis (Ignatius et al.,
2013). Aspen et al. (2012) reviewed current lifecycle assessment tools available to the marine industry and
carried out a case study demonstration on a longline fishing vessel. This study concluded that much work
remains to be done to move LCA into a practical method for vessels under construction, including better
system boundary standards, industry-specific assessment databases, and tighter integration into ship design
software. Overall, while both LCC and LCA continue to gain traction, there is currently a gap between the
capability of easily-available tools and the desire to address these topics.
3.4
Links between design tools and production and operational phases
The enormous value represented by digital vessel design models and databases created during ship design
and construction continues to attract attention for data re-use over life. Much progress has been made for
data re-use within the constructing shipyard (e.g. Morais et al. (2014) presents recent progress with
ShipConstructor), but struggles remain to make this data available over the vessels entire lifespan. One
major issue in sharing such data is the fact that the intellectual property associated with the majority of
this data is normally owned by the shipyard who does not want it to fall into the hands of competing
yards. Further complicating matters, the building yard typically has the shortest temporal interest in the
vessel, especially when compared to the vessels owners, flag states, and classification societies. Thus,
470
while this section does report notable progress over the last three years in this area, there is still no
industry-wide agreed approach for handling ship design data over the vessels lifecycle.
The ability to view 3-D product data over the entire vessels lifecycle continues to be a fairly basic
requirement of any design-model use. Several proposals have been developed for lightweight viewers,
typically independent of the underlying CAD/CAM/CAE system that stores the full design model. Notable
progress has been made with Siemens JT lightweight open viewing format. This format was codified in
ISO Standard 14306:2012 in December 2012. While its applicability far exceeds the marine field alone, as
of 2014, several major marine players served in the corporate program group for JT including Electric Boat,
Hyundai Heavy Industries, and ThyssenKrupp Marine Systems. Marine progress with JT was reported in
several venues, including RINAs Naval Architect (Malay, 2012), and JT was adopted for an NSRP
visualization pilot for the common parts catalog in the United States (Electric Boat, 2013). A rival approach
was presented by ShipConstructor, whose Autodesk compatibility allows direct tie-in with the Navisworks
suite, which is widely used in other industries (Larkins et al., 2013a, Larkins et al., 2013b). In marinespecific software, Nupas-Cadmatic reported on efforts to use 3-D models directly on the shop floor to
replace 2-D paper drawings via a variety of viewers (Cadmatic, 2012). While vendors are clearly leading
much of these efforts to date, some vendor-neutral efforts have also been attempted including a Dutch effort
to create a vendor-free, simple, 3-D visualization package (Asmara et al., 2012). The use of such lightweight
models for production interference checks has also been proposed (Suh et al., 2013), indicating that some
calculations that occur down-stream from engineering may also migrate to such platforms in the future.
Augmented reality (AR) approaches, where the visualization of data from the ship product model is
linked to viewing the actual structure, are rapidly growing extensions of these visualization approaches. At
the simplest level, an early study reported on using embedded RFID tags in the structure and equipment
itself to help locate the user in the 3-D model and associated design data (Lee et al., 2012b). More recently,
Halata et al. (2014) presented a detailed account of the design and prototype of an AR approach for
outfitting tasks such as pipe installation. Based upon extensive observation, information needs were
investigated and divided into three categories. A prototype AR device using an Android tablet was
implemented and used in a production environment. Post-use survey results were gathered with largely
positive feedback. Matsuo and Shiraishi (2014) reported on two AR demonstrations. The first demonstration
integrated on-piece markers, laser scanners, and an AR tablet to track progress in shaping curved hull plates.
In this approach, the current shape of a complex curved plate is compared to the desired final shape, and the
differences are displayed to the shop floor craftsman to help plan the next forming operation. In the second
application, an AR approach is used to plan final pipe connecting runs. Here, the details of the final pipe run
are not in the ship product model. Instead, the AR device examines the location that the pipe must be run
though and determines how the pipe must be shaped. Future work will link the as-built pipe information
back into the shipyards drawing system. Helle et al. (2014) presented some initial results of a 2-year AR
study in Finland using a Windows 8 tablet/phone AR device that located itself by using both visual cues and
RFID tags on the vessel. Helle et al. explored two sample use cases with this system. The first was
documenting vessel construction by adding virtual notes to a ship product model from the device during
inspections. The second was requesting new structural cut-outs complete outfitting whose routing
information was not complete when the structural steel was initially fabricated. All three AR approaches
combined a variety of off-the-shelf AR software packages and existing mobile devices, and they are
targeting similar, but distinct, steps in the vessel construction process.
In addition to viewing, more direct model exchange and keeping digital design models with the vessel
throughout the vessels life continues to be a goal of many owners and regulators. However, as mentioned
above, there are serious intellectual property considerations associated with such data sharing. Thomson and
Renard (2013) provide an overview of these issues, and highlight some proposed responses such as keeping
some data ashore, separate from the ship, or allowing access to more data as the vessel design ages and the IP
associated becomes less valuable. Naval applications are unique in this regard, as they offer an application
where a single entity often owns or has rights to access all associated design IP and is the vessels sole owner
and operator. In this context, implementing through-life data solutions is much simpler, with only technical
hurdles to overcome. The US Navys Achieving Service Life Program (Anderson et al., 2013) recently
demonstrated the use of a single FEA model throughout the vessels life, including aging-related updates for
corrosion and fatigue life calculations. This paper indicated that such integration is indeed valuable.
While IP issues remain difficult, open standards for such data exchange continue to gain traction to bring
this reality to commercial shipping. The ShipDEX extension to the S1000D standard for technical
information is widely established for marine equipment manuals. The OpenHCM hull condition and
thickness database is seeing more widespread use, including integration into major vendor packages, like
Aveva and Sener, and classification societies such as LR, BV, and DNV GL. The integration with BV was
471
recently profiled (Renard, 2012), and OpenHCM export has been added to the BVB Caf software (Zagkas
and Bralic, 2014). A key advantage of applications such as OpenHCM is that many of the IP-sensitive
production details of the vessel can be removed because the full CAD/CAM model is normally simplified
before exported into such a format (Thomson and Renard, 2013). To push this concept, Thomson and
Renard (2013) advocate the creation of a Model Repository File, or MRF, standard similar to the Ship
Construction File standard to store more detailed digital models with access control built into the standard.
In addition to open standards, many vendors have released through-life extensions to their design-stage
CAD/CAM/CAE packages. Most of these extensions attempt to offer a hub for through-life information that
can interface both with the vendors own CAD/CAM/CAE packages and with other software or data
structures that each ship owner may have. Complete all-in-one solutions seem to be less in favor, though
some of the bigger, more general software packages attempt to offer such a feature set. PTCs CADDS 5
software integrates with a variety of PTCs through-life portfolio, including PTCs Optegra for viewing and
integrated PLC data from CADDS 5 and other programs, and PTCs Windchill which is a more robust PLM
solution for CADDS 5 and other data streams. The FORAN ship CAE database was reportedly integrated
with Windchill (Penas, 2013) so that it could be used through-life. Likewise, SMART 3D, Nupas-Cadmatic,
and AVEVA have their own enterprise packages capable of tracking digital design data throughout the
vessels lifecycle. Siemens recently highlighted the difference between marine-specific PLM models and
more general PLM systems and proposed a new way forward to receive the benefits of each type of PLM
system (Bresler et al., 2013) following the general trend discussed in this paragraph.
3.5
Developments in integrated naval architecture packages
The development of monolithic naval architecture packages continues, with new features and capabilities
added to many of the existing packages used around the world. By enforcing tight data model coupling,
such packages can offer easy transferability between different structural and production tasks as well as
other naval architecture disciplines such as hydrodynamics and stability. A description of the major
packages was presented at the last ISSC Congress in 2012 and will not be repeated here; however, the
major packages have been summarized in Table 1. Specialized software for classification society analysis,
approval of structural designs, and links to the rest of the product lifecycle from classification society
databases have been selected for special review in Chapter 5 of this committees report.
Table 1. The number of officially reported plaque cases in the world.
Name
Autoship
AVEVA Marine
BVB Caf
CADDS 5
Design for Sea/
Catia
FORAN
Maxsurf
NAPA
Nupas-Cadmatic
ShipConstructor
SMART 3D
Structural Modeling Capabilities

Structural member definition via SQL/AutoCAD
approach, nesting, lofting.
Integrated structural design and detailed design/production
package, with other naval architecture features as well.
Features easy definition of structure on top of imported
geometry, basic calculations, and extensive FEM
import/export.
PTCs discrete manufacturing CAD/CAM platform.
Focused on detailed steel and outfitting design. Includes
lifecycle links via PTCs Optegra.
Specialized marine tools built into more general
CAD/manufacturing software. Supports initial design,
FEM analysis, and links to lifecycle applications.
Comprehensive naval architecture package. Supports
initial vessel design and structural detailing. Links to ship
production and PLM management
Supports initial structural design and built-in FEM
analysis with beam/plate tool Multiframe. Links to other
disciplines such as seakeeping, resistance, and stability.
Basic through detailed design and production, including
pre/post FEM processing capability.
Marine-focused modelling suite with extensive tools for
steel definition and production. Includes LCM links via its
eShare module, which can also share design data from
other packages.
3-D detailed design CAD/CAM model for structure and
outfitting built on top of AutoCAD and SQL.
Intergraphs latest iteration of what was originally
Intelliship, then SmartMarine 3D. Supports detailed steel
design and manufacturing, strong links to process and
outfitting/piping from co-focus on plant design.
Website (as of Fall 2014)

cadcam.autoship.com
www.aveva.com/marine
www.bvbcafe.com
www.ptc.com/product/cadds-5/
www.3ds.com/productsservices/catia/welcome/
http://www.senermar.es/
www.bentley.com/enUS/Products/Maxsurf/Marine+
Vessel+Analysis+and+Design.htm
www.napa.fi
www.nupas-cadmatic.com
www.ssi-corporate.com
www.intergraph.com/
ppm/default.aspx
472
4.
OPTIMIZATION DEVELOPMENTS
4.1
Introduction to Design Support Systems (DeSS)
Modern optimization methods developed in areas such as naval architecture, aerospace, and mechanical
engineering, are capable of validating innovative vehicle concepts as well as generating competitive
designs for standard ship types. The report of TC IV.2 at the ISSC 2012 contains a section on the design
requirements, mathematical models of required fidelity for design phases (concept, preliminary, and
detail), basic taxonomy, and applicable optimization methods and formulations, including safety as design
objective and other topics. The methods presented in 2012 help designers to achieve significant savings
for the shipyard and the ship owner; increase deadweight; decrease the price and weight of construction
steel; decrease production costs; improve lead times; increase safety (and robustness); increase savings
regarding lifecycle cost (LCC) among other beneficial results.
In the last three years, we have seen further progress in methodology and hardware:
Large scale practical problems: Inclusion of the designers talent and heuristic knowledge is shifting
the focus from the simple optimization problem to the design support (DeS) paradigm, where the
optimization module is a utility, since the real quality of the design process is not based on the
inclusion of all possible or available complex calculations. Rather, it is based on the interactive
designers exclusion of all unnecessary considerations. DeSS enables the designer to concentrate on
the relevant design attributes used in his/her educated (supported) key decisions regarding the design
characteristics.
Design Analysis: Development of IMO GBS and related IACS Rules (CSR) as well as the emergence
of new fast general methods for reliability analysis based on the dimension reduction method (DRM).
Design Synthesis: Emergence of techniques of surrogate modeling, problem decomposition, and
coordination enable the designer to handle large scale structural models (i.e. full ship models). For
optimal performance, the development of synthesis models still needs to be closely coordinated with
analysis models.
Hardware developments: Computing power increases by increasing the number of CPUs or CPUs
cores. Novel frameworks can enable distribution of parallel computation sequences to both standard
CPU cores and massive parallel GPU cores (e.g. NVIDIA Tesla Kepler K20).
Building off our report in 2012, this chapter starts with an introduction to the overall design procedure
for large scale design problems. The growth in problem size implies application of the novel hardware
developments (see Section 4.2) that are paving the way for novel design synthesis approaches in
formulation of practical mathematical models. Those approaches will enable the application of more
complex rules in optimization, including the following:
IACS Harmonized Rules (H-CSR) for double hull tankers and bulk-carriers requiring 3-hold models
for each hold to be approved, resulting in large scale multi-hold design optimization problems.
Development of practical mathematical models that use safety measures as design objectives (not only
as constraints), leading to socially responsible design practices.
This chapter is further concerned with developments of practical optimization methods (Section 4.3)
applicable in design offices, solving manipulated multi-criteria design problems (Section 4.4) in an
acceptable time-frame and applying it to the design tasks:
Optimization for production cost (Section 4.5), important for the yard decision makers.
Optimization for LCC (Section 4.6), important to owner and operator.
In the remainder of this introduction, the overall design procedure, embedded into the interactive
design environment with optimization utility, will be briefly discussed. The design problem definition for
efficient DeS (decision support), is presented in Table 2. This framework is intended to be a generic
description of a structural optimization process. As computational power grows via improved computer
hardware and parallelized codes, it becomes increasingly important to tie such optimization approaches
into a formal decision support system for the designers. The overall problem is composed of three main
steps: design problem identification, design problem formulation, and design problem solution. These
steps descend in a hierarchy from high-level problem description to the details of computing a range of
potential solutions to the problem. In brief, they may be summarized as following:
(1) Design problem identification (DPI) includes the identification of mathematical models (i.e. criteria
function sets and their supports and domains). Efficient taxonomy is instrumental for precise and
473
concise, yet comprehensive, definition of the decision blocks (see Zanic et al. (2013c)). For the
example of commercial vessels (Oil Tankers and Bulk Carriers), critical constraint functions values
are taken from the set of design loadcases using CSR.
(2) Design problem formulation (DPF) implies IT-formulation of requirements defined in DPI. It also
includes problem manipulation to develop the most efficient algorithm. The efficient solutions for
large scale practical problems are best obtained if the mathematical design problem is formulated
together with the associated solvable mathematical model (and the corresponding IT-modules to
support the stakeholders in their design-related decision making). It combines an efficient design
analysis model (AM) for response and evaluation assessment, and the synthesis model (SM) for
subjective and objective decision support using optimization solvers as utilities (see Table 3).
(3) Design problem solution (DPS) requires the integration of the DPF developed or selected modules
embedded into the interactive environment for the inclusion of the designers subjectivity and
heuristic knowledge to provide adequate support for the decision making (DM) process.
In the DeSS interactive environment, DPS needs control modules (sequencers) and a set of GUI
utilities.
Table 2. Design problem definition, including basic taxonomy (Zanic et al., 2013c).
DESIGN PROBLEM IDENTIFICATION - DPI (CONCEPTUAL LEVEL)
Determination of design objectives set o* and related measures of robustness/sensitivity set

rx*, for selected set of design load conditions/cases LC and design descriptors d ,
Selection of design variables nv-tuple x, among the set of design descriptors d, with
corresponding structural (often represented as finite) element Groups (xeG).
Selection of design criteria functions c (constraints g>0, attributes a) with corresponding
function supports ceG-s, geG-s, etc. Note that design objective functions o could be obtained
from attribute functions a, for given direction of improvement (min, max).
DESIGN PROBLEM FORMULATION - DPF (ALGORITHMIC LEVEL)
Selection of two basic mathematical models and corresponding IT-modules:

Design analysis model (AM) for technical (response, feasibility criteria) and economical evaluations (cost criteria,
risk), see Table 3bc.
Design synthesis model (SM) for objective and subjective decision making, see Table 3a
SM formulation (given AM) includes:

Design problem manipulation into equivalent but mathematically more convenient form. Selection of solution
strategies (e.g. solvers based on optimization techniques , surrogates ) for the manipulated problem (Section 4.3).
Development of method for the final subjective selection of preferred design(s) among generated design variants in
modules, using modules for sensitivity /uncertainty/ robustness analysis.
DESIGN PROBLEM SOLUTION - DPS (PROCEDURAL LEVEL)

Application of the design procedure with practical implementation of selected AM inbuilt into the SM interactive
decision-making shell, see Tables 4 and 5.
AM/SM utilities involved should enable the efficient synthesis (optimization and sensitivity modules, databases,
graphics, etc.) possibly based on parallel processing due to the required workload and time available to the given
design phase, see Table 3.
The design problem formulation must ultimately lead to a single solution, denoted preferred solution,
being selected for construction. In this step, two design support approaches are possible, which are
denoted as Pareto Supported Decision-Making (PSD) and Decision Support Problem (DSP) to signify
their basic differences. PSD enables the stakeholders educated selection of the preferred design based on
the sequentially generated geographical map of the Pareto frontier in the design space X (spanned by nv
design variables, used for form description, see Chapter 2 of this report) and attribute space Y (spanned
by na design attributes or functionalities, measured by KPIs). DSP generates preferred designs along the
designers route in those spaces using knowledge engineering. Of these approaches, PSD is the more
commonly used system to date. Zanic et al. (2013b) provides a comprehensive overview of the various
Operational Research methods that can be used for tackling this problem, building on the work in the
2012 ISSC report of this committee.
474
While evaluation mapping is a direct procedure (getting criteria function values for given design
variables), the inverse design mapping (getting values of design variables for the aspired levels of
objective functions), implied in the design process, is entangled with many mathematical problems. They
lead to different methods in operations research tailored to the characteristics of objective and constraint
functions of the problem at hand with selected optimization solvers. Based on the design problem
definition (Table 2), a sequence of standard optimization problem formulations can be developed by
integrating IT-modules defined in Table 3. Corresponding IT-modules, embedded into the interactive
design environment (see Table 4) are enabling the design process evolution and the development of
efficient designs.
Table 3. Criteria implementationIT modules (Pradillon et al., 2012).
(2a) SYNTHESIS MODULES (SM)

= set of modules for the interactive definition of the synthesis (optimization) problem: selection of variables x
(subset of d) and criteria fns a and g via module , problem decomposition and coordination;
= set of synthesis modules (GUI) in synthesis model (SM) for optimization (using preference data Pu, Pv for
subjective definition of utility and value fns u and v), designer interaction with the design process, filtering of
designs and visualization of X, Y, and metric M, L sets/spaces. Output: normalized m, l values
= set of optimization solvers (e.g.: Multi Objective Seq. Linear Programming- (MOSLP), Fractional Factorial
Experiments (FFE), Multi Objective Particle Swarm Optimization (MOPSO), Multi Objective Genetic
Algorithms (MOGA), Evolution Strategy-Adaptive Monte Carlo (ES-AMC) , etc.) for generating Pareto
frontier {xk, yk}ND by filtering designs in feasible subspaces X U Y based on objectives o.
= surrogate solvers (e.g. Response surfaces (RS), Kriging, Radial Basis Functions (RBF), etc.) Input: set (d, z)k for
fns ci, Output: quality measures qi for selected data in data container Z = d X (variates) zLC (load effects).
(2b) RBDO MODULES

= reliability/robustness modules; subset of AM containing modules based on fns rx ( = np-tuple of probabilistically
based c functions, e.g.: rx REL: g(Z) pfail; (set of failure probabilities/freqs of unwanted event e); rx ROB:
ai(Z) robustness measure (e.g.Taguchs SNR); rxRISK: ( g(Z)T * CF(Z)) pfail * cf (e.g. risk of losing a ship).
Output: values of probability (Pfailure), robustness measure, risk ($)
(2c) ANALYTICAL MODULES (AM)
= design quality modules; subset of AM containing criteria functions/mappings ai. Output: y (attribute values)
= adequacy modules; subset of modules in the analysis model (AM) containing safety constraint
functions/mappings gi>0 (e.g. class rules). Ouptut Igi(pass, fail), normalized g-values [-1,1].
= response modules; subset of AM, containing modules for FEM procedures ri. Output: zLC (load effects).
= environment/economy modules (loads, costs, etc.); subset of modules in AM with data generators E:
d= {dpressuresLC, daccelerationsLC, dmasses, dcosts} = E(LC) d
= structural (physical) modules; subset of modules / modelers in AM/SM; data generators F, e.g. NAPA, MG, INDAT:
d= {dtopology, dgeometry, dmaterial, dscantlings} = F(d0) d ; d0= dPROTOTYPE
Table 4. Integration of Design Support SystemDeSS (Prebeg et al., 2014).

A8
8
ID SHELL / MODULE
( ())
B7
((,))
C6
((,))
RBDO
(,)
(,) RBD
D
5
E4
F3
G2
H1
(),
( (), ) - surr.
(, )
( (, ) - surr.
()
(innovat, exper.,
prototype)
Back to A II
SHELL DESCRIPTION
Interactive shell (GUI) containing control of modules in embedded shells. Used for:
A I : DP (re)definition via structural and load descriptors dand d (see Table 3)
A II: subjective design selection in metric space using PSD techniques and AHP, HAW,
ELECTRE solvers, see Table 6b
Shell for MODM or MADM adaptive generation of non-dominated (Pareto) designs based on
MOGA, MOPSO, ES, MOSLP solvers, see Table 3a
Shell for reliability criteria based adaptive generation of designs using FASTREL and CALREL
solvers (Zanic et al. 2014, Piric et al 2012), see Table 3b
Shell for definition of objectives and adaptation of criteria status (objectives and constraints) for
their evaluation using criteria libraries in MAESTRO and OCTOPUS solvers or CSR library in
MAESTRO-XML solver, see Table 3c
Set of programs for quality, feasibility calculation (direct or via surrogate solvers ) using Class
or IACS CSR modules in OCTOPUS EPAN, MAESTRO ULSAP, ALPS/HULL, LUSA
Shell for calculation (direct or via surrogate solvers) of response fields (stresses, displacements)
using MAESTRO, OCTOPUS and NASTRAN solvers, see Table 3c
Shell for definition/ input of design loads in TD and FD (ships RAO) and mission profile using
solvers HYDROSTAR or MAESTRO WAVE, see Table 3c
Shell (environment) for structural modeling using data generators MAESTRO, NAPA STEEL,
FEMAP, USCS Ship Explorer, see Table 3c
Problem redefinition, MCDM decision making in PSD procedure
475
The resulting design support systems (DeSS) include objective (optimization solver) and subjective DM
tools as parts of an iterative and interactive design process. They will endow the stakeholders with direct
involvement in the design process and will support their educated decisions by sophisticated techniques
for subjective decision making. The mathematical model for design synthesis, as used in DeSS, is usually
formulated using Multi-criteria Decision Making (MCDM) techniques. They are divided into two basic
groups: Multi-Attribute DM (MADM) and Multi-Objective DM (MODM). MADM is based on the
selection among generated and evaluated designs that are based on a-posteriori defined preferences, while
MODM generates designs for the selected set of designers preferences using optimization techniques.
The structure of DeSS, given in Table 4, can be presented as a cascade of embedded shells (with IDs
A-H in the babushka style) of mathematical models or IT modules (direct and surrogate, see Section
4.6) for design analysis and synthesis (see Table 3). For integration tradeoffs, see also Section 4.3.
In the first row of Table 4 (denoted A or 8), the symbol ( ()) denotes that interactive shell
contains controls of design definition shell used to define criteria functions c via module . It is design
synthesis approach. AM are often integrated using inverse approach (in order, numbered 15 in Column
1 of Table 4). SM is mostly integrated in the AC direction for control of running AM. For example,
papers (Zanic et al., 2010, Prebeg, 2014 #297) present the structural DeSS rationale for the reliability
based design optimization (RBDO) in the concept design phase (CDP), a phase when the most farreaching decisions are made, regarding the attributes measuring the ships safety and costs. These papers
also demonstrate that using novel reliability based design (RBD) technique, the Pareto supported
decision-making (PSD), with safety as one of the objectives, can replace the old paradigm where safety is
only a constraint and thus promote a socially responsible design approach
While the concepts presented here are meant as a generic description of coupling of optimization and
decision support, the DeSS system described has been directly implemented in the
MAESTRO/OCTOPUS software. The initial DeSS system presented in Table 4 and the first row of Table
5 is based on the parallel processing-based RBDO procedure (Zanic et al., 1993) called OCTOPUSLegacy. The new RBDO version is capable of working in the MAESTRO/OCTOPUS design
environment (Prebeg et al., 2012) by using the very fast reliability calculation module FASTREL (Piric
et al., 2013). Examples of the practical integration styles for modules and shells for MAESTRO and
OCTOPUS DeSS are presented in Table 5 along with test models and recent references. Regarding novel
MAESTRO optimization modules and applications have also been described in recent conference
papers: Ma et al. (2013b), Ma et al. (2014a), Hughes et al. (2014), Freimuth and Ma (2014), and Ma et al.
(2013a), as well as a novel, very efficient loading utility described in a paper by Ma et al. (2014b).
Table 5. Integration of MAESTRO/OCTOPUS modules into DeSS.
Software / Test-model /
(Reference)
(1) OCTOPUS-Legacy
SWATH Patria
(Zanic et al, IOS 1993)
(2) OCTOPUS DeMak
Test BOX / SSC-398
(Zanic et al, OMAE 2014)
(3) MAESTRO-Legacy
LNG /EU FP6 MPROVE
(Zanic et al, JEME2013)
(4)
MAESTRO/OCTOPUS
RRM open ship
(Zanic et al, PRADS
2013)
(5) MAESTRO/OCTOPUS
IACS CSR BC
(Zanic et al, IMAM, 2015)
4.2
Shell level
Integration IDs
A or 8
B or 7
A H RBDO
DeView ES-AMC
Parallel proc.
ver.1
XPC k X
40 processors.
A H RBDO
DeView MOSLP
DRMSIMULATOR
SP /US
E or 4
F or 3
G or 2 H or 1
Faulkner
SEA
MADM NLN
2D FEM KEEPING INDAT
FEM
FD
C or 6 D or 5
FOR
M
DRM MODM SSC-398 ANALIT
17
OPTDAT SLIP2
MIN
WGT
OCTOPUS
A E1
MAESTRO
1 42
DeView
MIN
WGT
MIN
WGT
OCTOPUS
AE
MAESTRO
13
DeView
SLP/FD
SLP/FD
Hughes
INDAT INDAT
3D FEM
MG
MG
Hughes2
3D FEM
Deflection1
MG
MG
CSR
LOADS
MG
CSR
-ADEQ
LUSA
MULTIMODEL
3x3HOL
D
3DFEM
Parallel processing and hardware developments
The time span covered by this committees report has seen rapid progress in both computational hardware
for optimization problems and the usage of such resources for structural problems. Historically,
computational hardware advances have been focused on allowing a new generation of modeling codes to
become tractable. Any benefit for optimization has tended to be a secondary concern. Additionally, within
476
the marine technical community, hydrodynamic simulation has typically been viewed as requiring highperformance computing (HPC) resources, while structural design has not. For example, in a recent
review, McNatt et al. (2013) only associate HPC resources with the loads side of the structural design and
optimization loop but not the structural response. However, HPC resources are beginning to appear in
structural optimization literature. Experience of the members of the IV.2 committee indicates that 1,000
to 2,000 CPU hour structural optimization runs are easily within reach of most researchers using HPCs to
parallelize out over tens or hundreds of cores. Such resources do result in the ability to tackle harder and
more complex problems such as trade-space definition. To date, most work in this domain has focused on
steps B-G defined in Table 4 above.
Cluster-style HPC resources still dominate the HPC literature to date. The long-term growth and
dramatic reduction of cost of cluster-style computers, where hundreds or thousands of standard off-theshelf CPUs are aggregated together have resulted in many opportunities for parallelization. Access to
such machines is now commonplace at any major research university. As of 2014, the University of
Michigan cluster Flux provides approximately 16,000 CPU cores for on-campus researchers. Similarly,
the N8 cluster available to the University of Newcastle Upon Tyne has approximately 5,300 CPU cores,
and the University of Zagreb has the 800-core cluster Isabella in-house. Typical optimization jobs can
span tens or hundreds of these cores. Most clusters are configured with a Linux-based operating system
and a variety of proprietary compilers and high-performance mathematics libraries. Linking the cores
together is typically done through high-speed networking, with both custom-design protocols like
Infiniband and more standard protocol, such as Gigabit Ethernet, implemented. Fully taking advantage of
such resources typically requires manually modifying the code of each application to include multiple
threads via threading libraries, such as MPI, and then compiling and linking with the system compilers
and libraries. Thus, this approach is difficult to implement for closed-source applications unless the
vendor can provide binary-level compatibility. Additionally, applications that have evolved around
graphical user interfaces must be modified to run in a command or batch environment, as most HPC
resources do not provide any display forwarding from the compute cores.
The growth of heuristic optimization algorithms, such as genetic algorithms or particle swarm
algorithms, in which a population of candidate points needs to be evaluated frequently, is perfectly
matched towards this type of cluster. Because each individual to be evaluated from the population is
typically independent of all other individuals, this type of problem parallelizes very easily across multiple
CPU cores. There is typically minimal need to communicate among individuals during evaluation, so data
sharing and transfer speeds tend to be less critical, although data must be able to flow back to the master
optimization process quickly. In practice, such optimization approaches have proven practical, using
hundreds of cores joined only via standard gigabit networking. Another advantage of this approach is that
the MPI and thread code can be confined to the optimizer with the objective function relatively
untouched. In fact, if the per-call time of the objective function is long enough, it is possible to use
operating system-based file writing to move data in and out, greatly reducing the need for efficient
parallel threading in the codebase. This also allows existing, closed-source applications (e.g. ANSYS,
Abaqus) to interact efficiently with an optimizer over tens or hundreds of cores, provided sufficient
software licenses exist to run the required instances of these codes.
A more recent development has been the introduction of computation on graphical processing units
(GPUs). Driven by the need to provide renderings for computer games and other 3-D applications,
graphics cards now feature anywhere from tens to thousands of limited, calculation-focused
computational cores. By running computationally-intensive code on these cores, instead of on the main
CPU, it is possible to further accelerate many calculations. As most graphics cards are small, it is
possible to have a mini-cluster on each users desktop. Additionally, such functionality can be
accessed via conventional GUI-type applications and development environments like MATLAB and
Maple which have been compiled with GPU support. The downside to GPU processing is that custom
code is needed to access and run on the GPU themselves, much like (but incompatible with) the MPI
code needed for cluster computing. Thus, programs typically need to be re-written, not just recompiled, to work with GPUs. While this re-writing can focus only on the performance-critical areas of
the code, this still represents a time-consuming undertaking and requires full access to the relevant
source code. An alternative approach proposed by Prebeg et al. (2014) is to use surrogate models, built
in a GPU-aware framework to stand in for optimization components that it would be inefficient to recompile. However, simulations that are needed to build up the surrogate models must still be executed
outside of the GPU, or even in a parallel computing setup if the original applications are not compatible
with the available parallel computing hardware.
Cloud computing has also emerged during the last three years as a significant approach for managing
optimization and high-performance computing resources. In the cloud approach, the vendor bundles the
software and necessary licenses, computational hardware, and data storage into a single package that
477
can be rented or leased to the designer. For example, Altairs Hyperworks and the related topology
optimization can be run remotely on Altairs servers, or Altair will build a custom hardware and
software package that can be physically installed in a clients office. This abstraction of much of the
details of applying high-performance computing to design allows design offices to treat highperformance computing as sort of utility cost that can be scaled up or down as business demands.
Additionally, end designers can access the increase computational power through similar GUI-driven
applications they are used to using. While it frees the business from the overhead of developing,
running, and training staff on a bespoke HPC solution, most cloud solutions are standardized and lack
the full flexibility of a bespoke solution.
It is evident that high-performance parallel computing infrastructure has become easily available, not
just to research institutions but even to small design offices. Even more, cloud computing enables usage
of those resources on demand, which significantly reduces overall computing cost to users that do not
have constant requirements for HPC computing. However, it seems that those capabilities are not
adequately utilized in the ship structural design community and that the main reason for this is a low
level of adoption of ship structural design software to parallel execution on currently available high
performance parallel computing platforms. The reason for this is most likely due to the fact that in the
small ship structure community, ship structural design tools are developed by ship structural analysis
and design experts, not by numerical mathematicians or computer science experts (e.g. MAESTRO,
LBR5). At the same time some general purpose structural analysis tools (e.g. Abaqus) are well adopted
for almost all HPC platforms, but those tools are not widely used in the ship structures community,
especially not for purpose of dimensioning the structural components according to classification society
rules (e.g. Stage 2 models in IACS CSR for Bulkers and Tankers). Another possible reason is that the
actual performances of those tools were good enough for applications on real life structures, even in
early 1980 (Hughes, 1988). Thus, the identification of a necessity of a code adoption to parallel
computing came fairly late. Currently, this need is primarily driven by the desire to be able to generate
a Pareto frontier for the entire global ship structural design, not just for substructures. Additionally, the
ability to include reliability-base calculation (e.g. structural safety) in the constraints or objectives of
this global problem is desired. Constraints, or even objectives, include reliability-based calculation (e.g.
structural safety). Therefore, one of the largest challenges in the following period is to adopt the tailormade ship structural analysis and design software for parallel execution on currently available high
performance parallel computing platforms.
4.3
Developments in structural optimization algorithms (optimization solvers-)
Currently, manufacturers of ships and offshore structures are operating under difficult conditions. Facing
strong competition, they are forced to meet the expectations of their demanding customers. Such
circumstances require ship and offshore structure construction enterprises to focus on assuring quality,
satisfying their customers wishes, and continuously improving of their businesses. The ability to design
and manufacture high quality structuresbuilt as various one-off items or in short serieswhile
simultaneously fulfilling the economic manufacture requirements are factors that determine whether the
enterprise will survive in the market, hold its market position, and keep developing.
Given the above, it is normally impossible to assess variants of proposed solutions using a single criterion
(which allows for a very clear interpretation, such as the cost, and is sufficiently, accurately definable in
early design stages). Therefore, the conditions under which the decision is to be taken can be improved by
entering multiple criteria into the decision-making model. This initially entails the need to design and
manufacture structures characterized by cost-effective exploitation as well as cost-effective construction, or
simply the need for ship or offshore structure optimization in respect of cost-effective exploitation and costeffective construction. More and more attention is given to the requirement of cost-effective and
environment-friendly utilization of such structures after their lifecycle has ended, namely cost-effective
utilization. Therefore, both in the design stage and during the manufacture, the most advantageous decisions
possible that guarantee goal attainment are expected. One of the main measures allowing for such goals to
be attained is optimization. Favorable conditions for this have been provided by the development of ship
and offshore structure optimization methods and techniques that have been observed for many years.
Unfortunately, the progress in their practical utilization has been much slower. Many of them were
discussed during previous congresses. Because structure optimization is mainly concerned with complex
computational procedures, in order for such optimization to be run, it is necessary to formulate suitable
mathematical models for the structures subjected to optimization and to apply algorithms that are effective
in solving defined problems and that render solutions at acceptable costs, namely cost-effective optimization.
In most cases the multi-objective optimization problems in ship and offshore structures design are
solved in two steps:
478
(1) Determining a set of compromises/trade-offs (Pareto optimum),

(2) Selection of the preferred solutions/variants/candidates from the set of compromises. Such an approach
corresponds with the Pareto Supported Decision-Making (PSD) strategy discussed in the introduction
to this chapter.
To date, mainly deterministic assessment criteria have been used for the purposes of determining the
set of compromises. All such criteria are treated as equally important and taken into account together with
subjectively determined weighting factors representing value differences among such criteria for the
decision maker. When choosing the best solution, or the most preferred compromise in the set of
compromises, selection is done with the use of an additional (typically new) criterion. In papers published
within the time-frame concerned (since 2011), their authors present a single set of compromises, a set of
non-dominated solutions that constitutes a solution of the multi-objective optimization problem they
formulated. In their works, they do not discuss the issues of selecting the preferred compromise or set of
compromise solutions in the set of non-dominated solutions. Neither do they discuss the quality of the set
of non-dominated solutions obtained. However, it is known that when applying optimization algorithms
using random solution generation procedures (e.g. genetic algorithms) and/or random procedures for
selecting generated solutions for subsequent iterations (e.g. genetic algorithms, simulated annealing),
different sets of non-dominated solutions are obtained when any change is made to the variables
controlling the course of the optimization calculations. In such cases, one should therefore produce more
than single set of compromises and then all the sets should be analyzed and evaluated using statistical
methods as well. The rationale for the production of a single set of compromises can be high cost and
long calculation times as well as high complexity of the optimization object which are ships and off-shore
structures. In most cases may be satisfactory to obtain one set of compromises only; however the lack of
rigorous convergence criteria for this class of optimizers represents a continued challenge.
Since 2011, no definitively new approach has been proposed to formulating and/or solving ship and
offshore structure optimization problems. Existing computational algorithms and computer codes have
been developed. Computational procedures have been expanded and tested in subsequent cases. The
authors have improved the existing algorithms to increase the effectiveness of the computational
procedures and/or to build user-friendly computing environments (e.g. a graphical user interface). The
existing development environments (e.g. LBR-5, OCTOPUS) are being improved, or new ones are being
built, which integrate numerous computational tools including ones for structure optimization and
analysis for the purposes of assessing the limitations formulated. The results of reach and application
works are presented and reviewed in condensate form in Table 6.
Table 6. Summary of research and application of optimization for ships and offshore structures.
Authors
Optimization Method
Problem solved, main results and comments
Deterministic algorithms for single-objective optimization (1 publication)
Bayatfar et al.
SIMPLEX method
The single-objective optimization of deck structure.
(2013)
Deterministic algorithms multi-objective optimization (1 publication)

Motta et al.
(2011)
Convex linearization and a dual

approach
The hull structures of a 60-m mega yacht multi-objective

optimization regarding the cost and weight of the structure.
They showed that the optimization analysis application in
the preliminary design stage leads us to important gains in
terms of cost and weight with respect to the initial
scantlings. The advantages of a structural optimization are
particularly useful if they can be applied during the first
stages of the project.
Randomized (e.g. evolutionary etc.) algorithms for single-objective optimization (4 publications)
Devine and
Single-objective
genetic A single-objective optimization T-Craft midship section
Collette (2013)
algorithms (GA) and the with respect to production cost. A single-objective
Bayesian optimization algorithm optimization was conducted to minimize production cost
(BOA)
while maintaining sufficient longitudinal strength
requirements as recommended by the Guide to High Speed
Craft as set by the American Bureau of Shipping.
Kawamura et al. Single-objective
genetic The midship section of a double hull tanker VLCC
(2013)
algorithm (GA)
structural optimization with respect to construction cost and
Life Cycle Benefit (LCB). The authors present results of
lifecycle structural optimization of the midship section of a

Authors
Optimization Method
Fu et al. (2012)
Adaptive Simulated Annealing

(ASA)
Ji and Wang
(2013)
Pedersen et al.
(2015)
Multi-Island Genetic Algorithm

(MIGA)
Particle Swarm Optimization
(PSO) algorithm
Sekulski
(2011a, 2011b,
2011c, 2013,
2014, 2015)
Multi-objective
genetic
algorithm (GA) with combined
fitness function
Sobey et al.
(2013)
Genetic algorithm (GA)
479

double hull tanker considering multiple risks about not only
the risk of failure but also the environmental risks and the
economic risks.
Kitamura et al.
Genetic algorithm (GA)
The weight minimal optimization problem of the bottom
(2011b)
structure of a bulk carrier.
Sun and Wang
Support vector machine (SVM) A statistical learning theory applied in SVM was proposed,
(2012)
and genetic algorithms (GA)
which specializes in studying the situation with a small
number of samples. This theory could help to solve the
structure selection problem and local minimum problem
existing in traditional learning machines. As an example,
they presented the optimization of the midship of a very
large crude carrier (VLCC) ship according to the direct
strength assessment method in common structural rules
(CSR) with respect to structural mass. In this paper, a hybrid
process of modeling and optimization, which integrates a
SVM and GA, was introduced to reduce the high time cost
in structural optimization of ships.
Randomized (e.g. evolutionary etc.) algorithms for multi-objective optimization (10 publications)
Cui et al. (2012) Multi-objective particle swarm The multi-objective optimization of a bulk carrier structure
optimization and the multi- with respect to weight of the structural materials and
objective genetic algorithm cumulative fatigue damage. The results obtained indicate a
NSGAII
(Non-Dominated possibility to make weight savings compared to the original
Sorting Genetic Algorithm-II)
design. The fatigue indexes were also reduced, which means
the structure can last longer under the same operational
conditions.
Cui and Wang
Multi-Island Genetic Algorithm Optimization of a container ship cargo tanks midship
(MIGA)
section with respect to weight of structure. They presented
(2013)
an application of knowledge-based engineering in container
ship cargo tank structural design and optimization. Ship
design is such a complicated multi-discipline task that
knowledge-based engineering can assist in design and
optimization. In this design process for new ship structures,
the relevant knowledge is automatically distracted from
knowledge base and executed together with the knowledge
reasoning technique.
Ehlers and
Particle Swarm Optimization The optimization of stiffened panel from a case study vessel
Kujala (2013)
(PSO) algorithm
with respect to lowest cost per mass (C/M) ratio, which
reduces both cost and mass [(C+M)/(C/M))]. A PSO
algorithm, coupled to nonlinear FE-simulations and a
production cost assessment, was utilized in order to
optimize the design alternatives in compliance with the
uniform and non-uniform ice loading.
A multi-objective optimization of a container ship structure

considering a tee-stiffened plate structure with respect to
structural mass in the static analysis and maximum
acceleration of structure in the dynamic analysis. A new
improved collaborative optimization (CO) model, one of
multidisciplinary design optimization (MDO), is used.
A multi-objective optimization for the ultimate strength of a
ship hull in multiple load cases.
Identify the conceptual design alternatives for ice-classed
vessels. The midship structural arrangement of an LNG
tanker and of a general cargo ship were optimized with
respect to structural weight and cost (M, C/M).
The multi-objective optimization of the hull structure of a
passenger and vehicle ferry. Figure 3 shows condensed data
for calculations and the results obtained: a set of solutions
for the assessments of which are non-dominated in a Pareto
sense.
The composite boat hulls optimization with respect to
structural mass and cost. This paper examines the way in
which rapid assessment of stiffened boat structures
(composite structure) can be performed for the concept
480
Authors
Optimization Method
Temple and
Collette (2013a)
Multi-objective
genetic
algorithm
NSGAII
(NonDominated Sorting Genetic
Algorithm-II)
Genetic algorithms (GAs)
Vasconcellos
et al. (2015)

design stage.
The multi-objective optimization of a naval vessel structure
with respect to structural weight and lifecycle maintenance
costs. Fatigue and corrosion damage over the course of its
service life are taken into consideration.
The midship section design of a double-hull tanker
optimization with respect to power, deck wetness,
slamming, steel weight, and structural safety. Mode Frontier
software was used to integrate the mathematical model and
find the best solution.
As it appears from the review, the majority of the works are concerned with the use of randomized
algorithms to single- and multi-objective optimization. Only two teams reported results related to a
deterministic single-objective optimization. In all papers concerned with a randomized single-objective
optimization, genetic algorithms were used. As it comes to works focusing on multi-objective
optimization, most authors report the use of genetic algorithms. Two teams reported that they had used a
particle swarm optimization algorithm. One team reported that they had used a simulated annealing
algorithm.
There are additional works related to the development of the frameworks for ship structural
optimization and analysis. Frangopol et al. (2011) presented a general framework for the probabilistic
analysis of ship structures in terms of reliability, redundancy, fatigue, material deterioration, damage
detection, monitoring, and inspection optimization. Kitamura et al. (2011a) presented a structural
optimization system using the Finite Element Method (FEM) for the initial design stage of a ship. A
general bulk carrier was selected as the object for the optimization. A numerical example showed that the
proposed method makes it possible to optimize the shape of the ships bottom structure. McNatt et al.
(2013) used MAESTRO, a practical CAE tool for ship structural design and optimization. The
achievement of improved performance, including several aspects related to improved performance of the
ships structure, is fundamental to the evolution of ship classes and size for different types of cargo and
service. Examples of structural performance contribution to overall ship engineering and economic
performance include the following:
Higher performance structures in terms of reduced weight with higher degrees of safety and reliability
Lower fabrication costs
Better economic performance in terms of lower contribution to lightship weight and hence, larger
payload fractions
Reduced structural maintenance costs over the lifecycle
Recognition of social responsibility in terms of environmental protection, collision/damage tolerance,
reduced risk of failure, etc.
Recent decades have also seen increasing pressure by ship owners and operators for ships that deliver
high returns on investment. Zanic (2013) presented the application of the OCTOPUS DeView 5D
(OCTOPUSDesigner 2012) system to the optimization of the ships structure in respect to its weight. The
basic concepts and methods for multi-criteria synthesis of complex thin-walled ship structures in concept
and preliminary design are presented. Multi-criteria decision making (MCDM) techniques are used. They
are divided into two basic groups, multi-attribute decision making (MADM) and multi-objective decision
making (MODM). MADM is based on the selection among generated and evaluated designs from a
posteriori defined preferences, while MODM is based on optimization techniques to generate designs for
a selected set of designers preferences. Zanic et al. (2013a) presented the application of the
MAESTRO/OCTOPUS design support system to the multi-objective optimization of a Ro-Pax ship
structure. They used several optimization algorithms: Monte Carlo sampling, sequential adaptive
generation, multi-objective genetic algorithms (MOGA), evolution strategies, simulated annealing, and
multi-objective particle swarm optimization. The design support methodology (i.e. techniques and
procedures) for multi-criteria synthesis of large thin-walled ship structures in concept and preliminary
design is presented. Papanikolaou et al. (2011) describe the essential features of a modern, integrated
approach to ship design, such an integrated design software platform, and demonstrate its implementation
in practice: multi-objective optimization of oil tanker hull structure in reference of Freight Rates (RFR),
Oil Outflow Index (OOI), Energy Efficiency Design Index (EEDI) and maximum speed for given main
engine margins as integrated design and multi-objective optimization.
481
Yang and Wang (2012) proposed a methodology for fatigue reliability-based design optimization
(RBDO) for the design of a bending stiffener. RBDO is more meaningful, which is concerned with the
probabilistic constraints evaluated through the reliability analysis.
There are various criteria that can be used for assessment of the effects of optimization in designing
ship and offshore structures. An optimally-shaped structure can be compared to a design made by an
experienced designer. However, certain difficulties are to be expected here. Thus, for certain typical
simple structures, the optimization effects amount to a few percentage points, whereas for more complex
and untypical structures, such effects may amount to a dozen or so percent.
Because running an optimization process would extend the design stages duration and increase its
cost, the conclusion is that optimization is only meaningful in the case of manufacturing series of
structures or their elements in which even a slight percentage unit profit will yield large-scale global
savings and in the case of untypical costly structures, for which substantial unit savings will thus be
secured.
Figure 3. Data and results of evolutionary multi-objective optimization of ship hull structure (Sekulski, 2012).
There are various randomized multi-objective optimization methods reported in this section among
which multi-objective evolutionary algorithms have received significant attention (7 publications).
These techniques usually generate a finite set of solutions to approximate the Pareto frontier of a
multi-objective optimization problem (e.g. Fig. 3). However, obtaining the best possible set to
represent the entire Pareto frontier is not always a trivial (or even an objectively-defined) task. From
these reason performance assessment and comparison study of such techniques should gained much
attention. One obvious way to compare the algorithms is to simply visualize the final sets of
solutions and rely on intuitive judgments to decide on superiority of one technique to another.
However intuitive and visual assessment is not a reliable tool for comparison of different multiobjective optimization techniques. Because of that a theoretical as well as practical developments
regarding to a performance metrics that can be used for a comparison study of different algorithms is
needed. These performance metrics generally should be capable to assign an absolute or relative
value to determine whether it is a good performance of the algorithm. These metrics should be
useful for the effective and quantitative measure all desired aspects of quality in compromise
solution sets obtained by different algorithms as well as performance of optimization algorithms
without redundancy. Proposed metrics should allow for helps the decision-maker a quantitative and
objective comparison of the performance of the different multi-objective optimization algorithms
applied/developed for ship and offshore structures.
482
4.4

Surrogate modeling and variable fidelity approaches (surrogate solvers)
Surrogate modeling refers to the replacement of a computationally-intensive prediction model with a

simpler surrogate, typically built from strategic samples taken from the computationally-intensive model.
The methodology of surrogate (also referred to as approximation or metamodeling) modeling evolved
from classical Design of Experiments (DoE) techniques for conducting expensive physical experiments
into Design and Analysis of Computer Experiments (DACE). The main and particularly important
differences between the methodologies are the nonexistence of random errors for deterministic computer
experiments and the adequacy of the models, which is determined by systematic bias.
The surrogate modeling (metamodeling) and its role in support of engineering design optimization is
illustrated in Figure 4.
Figure 4. The role of metamodeling in support the engineering design optimization (Wang and Shan, 2006).
A recent survey (Viana et al., 2014) of more than 200 scientific papers examines application of surrogate
modelling in aerospace engineering multidisciplinary optimization community in the past 25 years. The
authors opinion is that the curse of dimensionality still exists as problems have just gotten larger and
that new metamodeling techniques are needed to handle the peculiarities of high-dimensional problems.
They believe that global sensitivity analysis will play an important role in alleviating the curse of
dimensionality, and many are investigating approaches to reduce the computational expense associated
with high-dimensional problems. In addition they specify problems with computational complexity,
numerical noise and challenges of handling mixed discrete/continuous variables. The validation of
metamodels and the underlying model is as critical as before. A related wide open field for research is the
incorporation of the error of the metamodel and the underlying model itself into the problem formulation.
One of the conclusions of ISSC2012 concerning the surrogate modeling was that the Kriging models
and other surrogate modeling techniques have been explored in limited depth at the present time but clear
recommendations on their use cannot yet be made, and more research on these topics is required
(2012). In 2015, there are many reported results from comparative studies of several metamodeling
techniques, which are applicable for different engineering fields. The metrics used in this comparison are
as follows:
Accuracy: the capability of predicting the system response over the design space of interest
Robustness: the capability of achieving higher accuracy for different problem types and sample sizes
Efficiency: the computational effort required for constructing the surrogate model and for predicting
the response for a set of new points by surrogate models
Transparency: the capability of illustrating explicit relationships between input variables and
responses
Conceptual simplicity: the ease of implementation
The common opinion according to Prebeg et al. (2014) from the different tests problems with low or
high level of nonlinearity and small (2, 3) or large ( >10) number of input variables concerning the
accuracy of accepted modeling techniques is summarized in Table 7. Three of the most popular modeling
techniques are response surface (RS), Kriging (KG), and radial basis functions (RBF). The accuracy of
483
every model must be validated by comparing the model and underlying prediction method on samples
distinct from those used for fitting the model.
Table 7. Preferred techniques for different problems
concerning accuracy.
Nonlinearity
Low
High
Number of input variables

Small
Big
RS
KG
RBF
RBF
The survey of the reports from last three years points out two main directions in development of the
surrogate modeling: automatic design and optimization, and risk and safety analysis.
4.4.1
Surrogate modeling in design and optimization
The main objective of the research of Prebeg et al. (2014) is to improve the design process of complex
thin-walled ship structures through an application of surrogate modeling. The methods considered are
integrated into an interactive computing environment for multi-criteria design in order to enable effective
and efficient interaction between the designer and the design problem within a given design timeframe.
Surrogate modeling was applied to the approximation of structural responses in order to reduce the
number of FEM calculations inside of the optimization loop. Applicability of the proposed approach and
suitability of different surrogate modeling methods is extensively tested on a simple barge example.
The cost of surrogate modeling is influenced by the number of control factors and the number of
computer experiments (i.e. simulations) using FEM. One possible solution to reduce the total number of
control factors is to treat beam descriptors (height and thickness of web and breath and thickness of
flange) as one composite descriptor using its e.g. moment of inertia or section modulus as a control value.
In this way, it is possible to reduce the number of control factors in the barge example from 35 to 11,
eight of which are beam composite control factors.
Another important factor that influences the computational complexity of the structural response
surrogate approach is the number of surrogate models that must be created. For example, in the case of
the RoPax ship design, about 7000 surrogate models for each load case are necessary. One possible
reduction is to create a surrogate model of the mean value of certain structural response (e.g. x) on all
adequacy elements (patches) in some substructure.
Due to the complexity of the design process in the real, practical applications, and in order to enable a
flexible execution, the DeMakOCTOPUS Designer environment is used. An important property of the
proposed design sequences is the possibility of parallelization which can take advantage of todays multicore and multi-processor workstations.
A hybrid between the well-established strength Pareto evolutionary algorithm (SPEA2) and some
commonly used surrogate models is presented by Kunakote and Bureerat (2013). Several surrogate
models, including quadratic function, radial basis function, neural network, and Kriging models, are
employed in combination with SPEA2 using real codes. The various hybrid optimization strategies are
implemented on eight simultaneous shape and sizing design problems of structures, and they take into
account the structural weight, lateral bucking, natural frequency, and stress. Structural analysis is carried
out by using a FEA. For a design problem with mass and natural frequency as objective functions and
stress and buckling as constraints, SPEA using RBF is the best method. Nevertheless, for all considered
design problems, the overall top performer is SPEA2 using the quadratic regression; SPEA2 using the
Kriging model is a close second-best.
Surrogate models are used by Chen et al. (2013a) for reliability-based robust design. The proposed
Surrogate Based Particle Swarm Optimization (SBPSO) algorithm combines the surrogate modeling
technique and particle swarm optimization. The algorithm and its efficiency are displayed through
numerical examples of a composite pressure vessel comprised of an aluminium liner and
T-300/Epoxy composites. The responses are analysed by using FEM (performed by ANSYS
software). An optimization problem for maximizing the performance factor is formulated by
choosing the winding orientation of the helical plies in the cylindrical portion, the thickness of metal
liner, and the drop off region size as the design variables. Numerical examples show that the optimal
results of the proposed model can satisfy certain reliability requirements and are robustness to
fluctuations of the design variables.
An automated optimization procedure based on successive response surface method is presented by
Pajunen and Heinonen (2013). The approach is applied to the weight optimization of a stiffened plate
used in marine structures. In the design space, the surrogate model is spanned sequentially into an
optimally restricted subspace that converges towards at least a local optimum. Both the objective function
484
and all constraint functions are modeled using the linear response surface method, enabling the use of a
robust and efficient simplex algorithm for the optimizations. The approach links SolidWorks and ANSYS
software, and optimization is made in a MATLAB environment. Being a practical optimization problem,
to generate a discrete variable optimum solution from the continuous problem solution, a simple binary
integer optimization is adopted. The binary integer optimization problem is solved using the branch-andbound method.
4.4.2
Surrogate modeling in risk and safety analyses
Polynomial Chaos Expansion (PCE) is a special kind of response surface for random variables by which
the stochastic properties of the responses can be easily obtained by using the orthogonal properties of the
polynomial. This kind of response is used by Htun and Kawamura (2013) to study the stochastic
properties of the ultimate tensile. First, the hypothetical corroded surfaces, which represent the thickness
reduction of corrosion, are generated by a random field model (Karhunen-Loeve Expansion Method).
Second, the random characteristics of the minimum cross sectional area of the generated plates with
random field corrosion are estimated by PCE method as well as Monte Carlo simulation to evaluate the
availability of PCE to estimate stochastic properties of strength of the corroded plate. By using PCE, the
number of samples can be reduced significantly to get the most accurate results for time consuming
analysis. Third, the ultimate tensile strength of the plates with random field of corrosion is obtained by
non-linear three dimensional FEA. Based on FEA results, stochastic properties of the ultimate tensile
strength of the plates with random field corrosion are obtained.
The surrogate modeling penetrates the rapidly growing industry of offshore wind turbines. Taflanidis
et al. (2013) discussed the quantification and evaluation of risk and development of automated risk
assessment tools, focusing on applications to offshore wind turbines under extreme environmental
conditions. The framework is based on a probabilistic characterization of the uncertainty in the models for
the excitation, the turbine, and its performance. Risk is then quantified as the expected value of some risk
consequence measure over the probability distributions considered for the uncertain model parameters.
Stochastic simulation is proposed for the risk assessment, corresponding to the evaluation of some
associated probabilistic integral quantifying risk, as it allows for the adoption of comprehensive
computational models for describing the dynamic turbine behavior. For improvement of the
computational efficiency, a surrogate modeling approach is introduced based on moving least squares
response surface approximations. The framework for risk quantification and assessment was also
extended to an efficient, sampling-based sensitivity analysis. Such analysis aims to identify which of the
critical uncertain model parameters are contributing the most to overall risk. The efficiency of the
surrogate model was further exploited to develop a standalone risk assessment tool that can facilitate an
automated implementation of the proposed risk quantification and assessment framework.
The surrogate modeling is very often combined with Monte Carlo simulations. Such an approach is
used by Georgiev (2011) for safety analyses of still water loads for bulk carriers. The goal was to study
the influence of deviations from cargo loading plan on net load of double bottom and still water bending
moments. For example, a Handymax BC-A type ship is used. The metamodels approximate the work of
the installed, on-board, mandatory loading instrument and give the relation between the distribution of
cargo and the trim and still water bending moments in controlled sections. The Monte Carlo simulation
uses the fitted metamodels to obtain a rich set of statistical data that permit preparing event tree analysis
for possible overloading of double bottom and the evaluation of likelihood for exceeding of permissible
bending moments.
4.5
Optimization for production (design quality modulesPRODUCTION)
An important subset of the overall optimization problem is that focused on optimization for production.
For many commercial shipyards, minimizing the production cost of the vessel is critical to winning
competitive contracts, and hence this topic has received extensive attention. The University of Wisconsin
(UW) and the Marinette Marine Corporation (MMC) developed a shipyard orientation program course
targeted to newcomers and non-shipbuilders in order to teach a basic understanding of shipyard layouts
and functions through a 3-D virtual shipyard layout modeled on Navisworks Freedom. According to the
University, the model represents various items seen in a shipyard and can even be customized to represent
a specific yard. Through this development, UW became the first institution in North America to offer fast
track, distance-delivered courses in a full certificate program in modern shipbuilding design that could
provide an asset in effectively training production personnel Lundquist (2012).
Lodding et al. (2011) presented Virtual Reality (VR) as a tool to aid in the adherence of budgets and
due dates in the shipbuilding industry by developing accurate planning and fast reactions to unplanned
events such as late deliveries or assembly parts. In particular, their article addresses that there is no
concept for an automatic session preparation according to new assembly and no software-based support
485
for finding and verifying a new assembly sequence using VR. Thus, an assembly planning prototype was
developed for collecting model data and dynamic meta data, which allows for fast and automatic model
manipulation and planning information. This reduces time and the occurrence of errors during the
preparation and execution session. A prototype of the VR model is currently under evaluation in
cooperation with a German shipyard.
Production management based on simulations rejects decision-making based on experience and
intuition and values the establishment of improvement methods based on quantitative and concrete data.
Lee et al. (2014) applied a simulation using software tools, such as DELMIA, to the work plan as part of
the production planning in shipyards in order to provide optimal decision-making through models
reflecting virtual shipyard facilities and spaces. The simulation environment is based on discrete event
simulation and is designed to enable the determination of production output, capacity, process flow
congestion, bottleneck identification, and material requirements at the work-stage level.
Nienhuis (2012) initiated the development of an automatic sequence generation system, capable of
generating the interference-free assembly sequences that identify the relationships among outfitting
activities and estimate the reasonable mounting time. The system contains three fundamentally automated
processes: assembly knowledge representation, data acquisition and preparation, and sequence generation.
The goal of this model is to generate a reasonable assembly sequence of non-structural components in the
outfitting processes, under the assumption that the various resources, such as drawings, material, and
equipment, required to perform the activities are always available. The research only takes the
geometrical aspects of components into account. The output of this research project is not the
development of a scheduling tool that commits all kinds of resources among a variety of possible tasks,
but the determination of the technical dependencies among different assembly tasks that helps to optimize
both the planning and production processes. Thus, the onsite coordination and installation effort can be
reduced, and the level of advance outfitting practices maybe increased, which leads to the reduction of
cost and lead-time.
Dong et al. (2013) present a two stage queuing model for shipbuilding outfitting process. Their article
tackles the problem of determining when the outfitting work for each block should be processed during
ship construction. Stages of ship production are simplified into two with the first stage representing the
general assembly process associated with block construction and the second stage representing the grand
block construction. Therefore, a model is formulated as a queuing system to provide information on how
to optimally distribute the outfitting work at each stage in the shipbuilding process. A closed formed
equation for system cycle time is applied by using Kingmans equation from queuing theory and
calculates the optimal percentage of total outfitting work processed at each stage can be calculated given
any scenario.
Kajiwara (2013) tries to identify the important wastages in a shipyard that can increase production time
and cost and to design the production system of a shipyard based on lean philosophy. Simulation
modeling is used to evaluate the performance of the production system, and Kanban is a pull-type
scheduling method that can be used to simulate the pull production system of a shipyard for the
optimization analysis. The reduction of waiting times for various block assembly strategies are obtained,
and as a result, the overflows of the sub-blocks and blocks in stockyards are notably reduced. Kanban is
used as the supplementary tools that can contain the information of the products and production system,
and it can help to pull the blocks in production lines and to maintain the amount of production.
Tokola et al. (2013) study the scheduling of block erection in shipbuilding and formulates a
mathematical model to minimize the time between the start and finish of the erection schedule when
different lifting and joining times are considered by taking structural stability and no-skipped-blocks as
constraints. Although the mathematical model used is simplified and should be further examined by
varying constraints, the solutions presented in the article give insight into how the design and production
parameters generally affect the production time, which could be an important factor in the early design of
the ship.
In the 10th International Conference on Computer and IT Applications in the Maritime Industries,
Caprace (2011) presented a ship block erection process using discrete event production simulation and
optimization. The purpose of this study is to examine how various computer-based analyses and
simulation techniques can be used to improve the efficiency of the block sequence definition process, as
the process is usually lengthy and subject to errors, leading to a non-optimal block sequence. The study
also discusses the reasons why shipyards do not use simulation tools alongside the various advantages of
simulation in terms of layout and production planning. The case study focused on the erection of a Suez
Max tanker, which initially begins with selecting blocks one by one to be erected, and at each stage of the
selection process, a list of potential neighbor blocks are provided, which satisfy the technical conditions
in order to proceed with the block selection sequence process. The sequences are then entered into the
DES software (QUEST) which evaluates the lead time of the production process. Once this is executed,
486
optimization software (ModeFRONTIER) is applied to optimize lead time by modifying input variables
and analyzing the outputs, which are defined as objective functions or constraints. Although the initial
results of this study reached no convergences for reasons explained by the authors, it is important to
highlight that the model generates only the feasible block sequences, thus avoiding non-feasible design
processes, and the analysis concluded that the block erection rule by layer takes approximately 6% more
time to complete compared to erection by slice or pyramidal propagation. Production simulation tools are
useful regarding the possibilities of gains in the process of production and cost reduction.
4.6
Optimization for lifecycle costing (design quality modulesLCC)
To effectively support future ship design, maintenance, and service-life extension decisions, Collette
(2011) proposed that it is necessary to extend the existing semi empirical, component-based structural
design rules based primarily on safety concerns for a system performance model for ship structures. This
system-based approach extends the existing rule-based approach by formally stating performance
requirements for the structure based on the vessel systems that the structure supports. By setting
performance targets, it is possible to assign key performance parameters (KPP) and key system attributes
(KSA) to the structural system, and to allow the overall vessel design synthesis work to evaluate different
structural concepts and weight budgets against achieved vessel performance attributes. Potential
difficulties in making this transition are discussed- such as combining structural tools capable of
producing the required metrics with associated measures of uncertaintythat are critical for combining
structural with other design metrics in an absolute frame such as cost effectiveness. Moreover, difficulties
in following traditional system engineering partitioning and decomposition approaches with structural
systems are observed.
Bharadwaj and Wintle (2011) present an innovative and a pragmatic application of risk-based
principles for planning inspection to the ship hull structure. The approach described in this paper uses
semi-quantitative measures (FMEA, RPN) and the ECI planning is informed by reports from CI and
engineering judgment/expert opinion, as well as operating data from a relevant sample of ships from a
generic database such as the one within the RISPECT project. This paper extends principles in asset
integrity management and the risk model is separated into two parts. The first part is a technique for
assigning priority to structures within the ship hull based on certain measures of risk. The second part is
used for optimizing inspection actions, given the risk-based order of priority established in the first part.
The application of this approach has the potential to reduce the operation and maintenance costs of ships
and increase their reliability.
Moreover, the American Society of Naval Engineers published a research article (SNAME, 2011)
aiming at identifying what actions can be taken to control LCC of a surface ship and developing a
potential list of obstacles in the design, construction, testing, and in-service support. Total LCC is
addressed by proposing actions such as reducing cost through weight constraints (cost estimates not
accurate, operations and support costs not updated as design matures), manning, and automation
(inadequate manpower evident from reduced readiness, deferred maintenance etc., deficiencies in
hardware and software). The article also compares surface ships to the submarine community, which
places considerable priority to maintainability, a major factor in altering the LCC.
Apostolidis et al. (2012) conducted an investigation of the relationship between dry-docking cost and
its significant determinants for a major ship repair yard. Based on 414 cases of ship repairs over a time
span of four years at major ship repair yards, by using a GMM model, size, age, and dry-dock days are
depicted as the main determining factors behind operating costs. The research also suggests that future
studies can focus on investigating the determined parameters on each dry-docking individual element and
on examining dry-docking cost geography, comparing similar dry docking projects and elements in
different regions.
In the 12th International Conference on Computer and IT Applications in the Maritime Industries,
Thomson and Renard (2013) explained the role of shipyards as producers of lifecycle maintenance
models. One of the greatest benefits of shipyards using 3-D models is that the models can potentially play
a greater role in the lifecycle of their products. They can also gather feedback on the through-life
performance of their vessels, which enables optimization opportunities (e.g. steel thickness frame
spacing). A lifecycle model could be the first step towards the extension of shipyards in the aspect of
providing lifecycle services such as management of technical databases, calculations, preparing and
implementing dry-docks and repairs, retrofitting, and training. The article also describes current lifecycle
management systems, such as AVEVA NET, which offer gateways for many software systems and 3-D
design tools by enabling the extraction and visualization of intelligent documents and datasets and has
been used in the oil and gas industry.

5.
CLASSIFICATION SOCIETY SOFTWARE REVIEW
5.1
Background, motivation, and aim
487
Traditionally, classification-society provided rules and simplified, easy-to-use formulas are applied to
determine the scantling dimensions of a ships structure. These formulas still remain useful in the early
stage of the design process, but the increase in ship complexity and research in first principal methods has
become a driver for class societies and designers to utilize a more rational-based design. While class
societies still publish and develop rule-based formulas, such rule-based approaches now exist with
guidelines for structural assessment analysis methods that must be applied. Such guidelines specify how
to investigate the structure, and what acceptance criteria must be used. Unlike rule-based approaches, the
guidelines themselves do not directly specify the final structural dimensions. 3-D finite element (FE)
models are extensively used for global and detailed analysis, and thanks to software and hardware
updating, model definition (decreasing of mesh size) has been improved with the additional possibility to
increase the number of details subject to fine mesh model assessment. Also, the approach to load-cases
definitions has changed. In particular for fatigue analysis, the support of software for long-term
hydrodynamic loads calculation has been introduced.
Advanced methods used for structural assessment offer to shipyards the opportunity to use a more
efficient design recognized and requested by owners. Often, this added value is certified by a
classification society with a specific class notation assigned to new buildings and is monitored throughout
the ships life.
General purpose software and tools are not completely able to support the design process without
complex and expensive customization to facilitate quick modeling, loading (interaction between
hydrodynamic and structural software), and post-processing the results. However, classification society
tools are inherently marine-specific. For these reasons, classification societies tools are evolving to
support higher-fidelity analysis and more complex design procedures. Availability of software supporting
the design and approval process is advertised as an important service and as a clear index of competence
and efficiency. Recent acquisition of the marine software firm NAPA by ClassNK may be considered as
an evident interest in expanding and improving the range of services offered to shipyards and shipowners, including a portfolio of software modeling tools.
The increase of information supporting the approval process has also encouraged classification
societies to develop applications for data storage and through-life monitoring. The aim is to create a
platform that serves shipyards, subcontractors, and surveyors so that ship owners may efficiently interact
not only during the approval process, construction phases, but also the rest of the ships lifecycle.
Advances in technology offer the possibility to maintain in databases product data models representing all
details of the vessels structural history throughout its service life. Not only drawings, support
documentation, notes, remarks, and certification would be stored, but also, for particular purposes, models
that may be used during ship life for ship monitoring and maintenance for re-fitting assessment and
accident support. These applications are developed to guarantee the availability of information but also,
thanks to appropriate access control, to ensure intellectual properties, which is very important considering
the number of different parties involved in ship design.
Classification society tools were last subject to evaluation by the ISSC Technical Committee IV.2 in
2000. Criteria of analysis defined in that report still remain as good guidelines for the evaluation of tools, but
advances in information technology, combined with continued rules development, keep alive the Technical
Committees interest in this particular topic with the intention to also extend the attention to application
supporting information storage and sharing during the ship lifecycle.
The aim of this chapter is not to present a benchmark study but to give evidence of how classification
societies are extending their offers in terms of software development by covering not only the design stage
but also the entire ship life. In Figure 5, a workflow is presented to map typical main activities, parties
involved, and tools used as support with evidence of their interactions.
488
Figure 5. Typical flow of structural information through classification society tools over a ships lifecycle.
5.2
Tool analysis
5.2.1
Overall functionality
The software tools provided by the classification societies, which are evaluated in this section, can be
divided in two main categories: tools for the assessment process of the ship structure and tools for the
Project Lifecycle Management (PLM).
Both the main categories, in turn, can be broken into sub-categories. As for the software tools for
scantling assessment of ship structures, all categories have been presented previously in the appendix of
the ISSC2000 IV.2Design Methods:
Hypertext-based rule search and presentation, which allow the user to quickly find the relevant rules
within the full rule text by means of built-in search facilities and cross-references in the form of
hyperlinks.
Calculation of scantlings of structural configurations according to rule requirements. These tools,
provided by some classification societies, are designed for the assessment of plates and stiffeners of
ship structures according to their rules.
2-D analysis of the hull cross section that allows the user to calculate the cross sectional data, once
the user has input the arrangement and dimensions of every structural part making up the ship cross
section. Also, some tools are equipped for the assessment of local structure according to the rules.
3-D modeling of parts or of the whole ship structure and direct calculation methods. The user can
create a 3-D FE model of ship structures and carry out direct FE calculation for structural
verification.
Data exchange between design and analysis tools.
The Project Lifecycle Management tools can be classified into two sub-categories:
Database for management of plan approval process information.
Database for management of ship performances as well as of the performances of the entire fleet.
5.2.2
Evaluation Criteria
The evaluation of the software is carried out in two different levels of analysis. The first level is common
to both categories of the software. First, the software tool under evaluation is associated with a
subcategory in order to identify which is the general functionality of the software. Then, the users to
whom the software tool is addressed are identified (i.e. the designer, the shipyard, classification societies
plan approval or surveyor, owner) along with the phase of the design process in which the software tool
has been designed to be used (i.e. pre-contractual, preliminary design, functional design, detail design,
approval process, monitoring during the construction, warranty and ship life monitoring or ship
489
decommissioning). Other information gathered in this phase of the software tool evaluation is the
application field of the software tool, that is, whether the tool has been designed to be used for all types of
ships or for a precise category of ships.
After the first level analysis of the software, a second analysis is carried out, for which two sets of
questions have been defined: one set related to software tools for scantling assessment, the other related to
PLM tools. The evaluation of a software tool is performed, answering each question individually.
Questions related to the evaluation of software tools for the assessment of ship structures are listed
below:
What is the functionality of the software tool?
Software tools for the scantling of the ship structures can be used for different types of analysis or they
can be designed for global or local analysis of structures. For this reason, further questions going into
more detail are below:
What are the components of the ship structure that are covered by the software?
Do restrictions apply that limit the usage to special types of elements, e.g. longitudinally oriented
or to regions like 0.4L mid ship?
Which assessment procedures are supported? Minimum scantling according to rules; stresses and
displacements based on direct calculations; limit state analysis: first principle design; vibration
analysis (eigenvalue and eigenvector, frequency response); fatigue design assessment.
Does the system support 2-D or 3-D ship structure models to be applied for the assessment?
What is the input mode of the software? All tools require input data, either 2-D or 3-D geometry
models for carrying out the analysis. In this part of the software evaluation all the input modes are
taken into account for each tool. Moreover, the compatibility with other commercial software typically
used in the maritime industry has been taken into account, which is described in more in detail below:
Does the tool allow a manual input of geometry models and data? If so, is a parametric and/or
associative modeling approach supported? How is symmetry handled?
Does the tool assist with the input of recommended scantlings from rule calculation software?
Does the tool allow the importing of a geometric model from CAD software?
Is a possibility function available to import FE models from other FE software?
Which types of loads can be taken into account in the analysis? How can the user manage different
load cases or combinations thereof in the pre-process phase?
As for the FE software tools, does the classification society provide an FE solver, or does the society
have the direct calculation be carried out using an external solver (e.g. Nastran or Ansys)?
What are the FE capabilities offered by the tool? Does the FE software allow carrying out special
structural analysis (e.g. Smiths method collapse routines, simplified ultimate strength codes)?
What are the capabilities to export raw data or complete models? As for the FE software, on the one
hand, the software sometimes allows the export of the FE model in a special format so that the
database can be loaded in other software, or the FE model can be solved by different solvers. On the
other hand, the other structural tools sometimes allow to create automatic reports or to export the data
and the results in different formats.
With respect to the PML tools, the second level evaluation is carried out, individually answering each
of the following questions:
What is the information taken into account in the database (e.g. drawings, certificates, missing
information for the approval of the project, FE models)?
Does the data management allow all shipbuilding parties (i.e. owner, classification society, shipyard)
to interact (e.g. status of remarks, drawing submission)?
What is the software architecture (i.e. is the software web based)? Is a particular database structure or
capability offered for models?
Does the software integrate with the class societies through-life inspection, corrosion/gauging data
collection, and other through-life tasks? Does the software have the ability to update the design stage
geometry/FE model with gauging data to build an as corroded model of the vessel in service? Are
any lifetime fatigue management tools offered? Are critical areas identified in the software to help
surveyors? Are these data shared between similar ships?
Does the software allow the user to automatically create reports, or to export and exchange data with
other PLM software?
Are special functions built in to support emergency incidents (e.g. grounding, fire, collision)?
490
5.3
Classification societies tools details
5.3.1
American Bureau of Shipping (ABS)www.eagle.org
The American Bureau of Shipping (ABS) has developed a series of tools for the strength analysis of ship
structures. The ABS Eagle SAFEHULL helps designers, shipyards, and engineers of the classification
society carry out the scantling calculation according to the rule requirements. The software allows a 2-D
analysis of hull cross sections as well as a 3-D modeling of parts of the ship or of the entire ship structure.
It can be used in the analysis of bulk carriers, containerships, and tankers that are not built in accordance
to the Common Structural Rules. To help the engineers in the structural scantling of bulk carriers and
tankers that are designed in accordance with the Common Structural Rules, ABS provides the ABS Eagle
CSR. This software allows the user to perform the same analyses that can be carried out using the ABS
Eagle SAFEHULL, but the results are checked according to the Common Structural Rules. The software
is used in several design phases, from the pre-contractual phase until the detail design and approval
process. The parts of the ship that can be analyzed are the hull cross sections (in 2-D analysis) and the
three cargo hold block centered around midship (in 3-D analysis). In the pre-processing phase of the FE
model, SAFEHULL is able to import the FE 3-D model, as well as the 3-D geometric model, that is
created in other software. It is also able to identify tanks in the 3-D model, and the user can apply loads
and external pressures, which mimic full and light drafts, to the FE model. The FE analysis is solved by
Nastran and SAFEHULL post-processes, and the results can be displayed in tabular or colored contour
formats.
Two modules have been developed for evaluating an LNG vessels design and an LGC vessels design:
SAFEHULLLNG and LGC ISE, respectively. Both modules allow the user to carry out the same analyses
that are available in SAFEHULL, but the scantling of ship structures is done according to the
classification society rules for LNG and LCG vessels. ABS assigns the SAFEHULL (SH) notation to oil
or fuel oil carriers, bulk or ore carriers, combination carriers, and container carriers that are designed
according to Part 5C, Chapters 1, 3, and 5 of the ABS Rules for Building and Classing Steel Vessels,
respectively. Also, the SH notation may be assigned to membrane tank LNG carriers designed in
accordance with Part 5C, Chapter 12 of the ABS Rules for Building and Classing Steel Vessels. The SH
notation applies to container vessels over 130 m; bulk carriers, oil carriers, and membrane tank LNG
carriers over 150 m; and liquefied gas carriers with independent tanks over 90 m in length. The
requirements of these portions of the rules are collectively referred to as the SAFEHULL Criteria.
ABS provides the ABS Eagle DLA/SFA Analysis System which integrates the dynamic load
approach and the spectral fatigue analysis in a software package that includes the major analysis
components. The software is provided to shipyards, designers, and engineers of the classification societies
and can only be used for ship-shaped hull forms. A seakeeping analysis of the ship can be carried out
through the application of 3-D potential theory based on hydrodynamic analysis in frequency or time
domain. A probability analysis of wave and ship response statistics follows the seakeeping analysis. The
DLA/SFA tool provides several modules for the assessment of loads for liquid, bulk and container
cargoes: the buckling and yielding strength assessment and a full spectral-based fatigue analysis. The FE
analysis is carried out in Nastran, and the results can be post-processed within the tool environment. ABS
assigns the SAFEHULL Dynamic Loading Approach class notation to vessels that have been evaluated
using an enhanced structural analysis procedure and criteria for calculating and evaluating the behavior of
hull structures under dynamic loading conditions. Such vessels must be built in accordance with plans
approved on the basis of the results of such analysis, in addition to full compliance with the other
requirements of the rules. Spectral Fatigue Analysis (SFA) notation is assigned to vessels on which
Spectral Fatigue Analysis is performed in accordance with an acceptable procedure and criteria, and the
vessel is built in accordance with plans approved on the basis of the results of such analysis. The vessel
will be distinguished in the Record by the notation, SFA(years). The notation, SFA(years), denotes that
the designated fatigue life value is equal to 20 years or greater. The (years) refers to the designated fatigue
life equal to 20 years or more (in 5-year increments), as specified by the applicant.
Project management databases are provided by ABS to be used in the design phase, in the construction
phase, and during the operation phase of the ship. These databases are all web based, and they are used by
designers, shipyards, ABS engineers, and surveyors. The ABS Eagle Web Calc Structures and
Machinery software packages provide the engineers with instant calculations for hull and machinery
according to the ABS rule requirements.
The Rule Manager is a hypertext based rule search that allows to quickly identify the applicable ABS
Rules and IMO requirements for a project based on parameters including vessel type, service, scantling
length, and contract date. This software is used in all the design phases as well as in the monitoring of the
ship during its whole life.
491
The ABS Eagle Engineering Manager is a PLM tool for managing plan approval information as
drawing comments, outstanding comment status, certificates, missing information for the approval of the
projects, and drawings throughout the approval process. The software can also be used by owners and
vendors, and it has been designed for managing documents of all types of ships. During the operation
phase of the ship, the ABS Eagle survey manager helps the owners and shipyards, as well as the engineers
and the surveyors of the classification society, to manage all documents regarding the fleet information
and survey history, due, and status.
ABS also provides a tool for the Lifecycle Management of the ship NS5 Enterprise. The software
helps the user in the management of ship operations, from regulatory requirements and crewing to
payroll, purchasing, and planned maintenance. ABS does not provide any special class notation related to
the usage of these databases during the design and construction phases.
5.3.2
Bureau Veritas (BV)www.bureauveritas.com
VeriSTAR Hull 5 is the latest release of Bureau Veritas (BV) software for structural analysis by means
of a 3-D FE model calculation. It is based on a customization of the general FE tool FEMAP. Both NX
Nastran and MSC Nastran are supported as solvers. Using FEMAP as pre- and post-processors allows the
user to access all functionalities for geometry modeling, which are available through the standard FEMAP
interface. It can apply rules for not only CSR and CSR-H for bulk carriers and double hull oil tankers, but
also for BV steel structures and offshore units. Loads and boundary conditions are automatically applied
in accordance with the rules requirements. VeriSTAR can carry out the global strength assessment of
cargo hold structures according to yielding and buckling criteria, stress assessment of local detailed using
local refinement, and fatigue strength assessment.
For ship structural assessment based on complete ship model description, BV offers the tool
VeriSTAR CSM (Complete Ship Model), integrating the BV Rules standard calculation and FE analysis
under a common interface. It is mostly intended for container and RORO vessels and is valid for both
new building and existing ships, and it stores data in a ship-dedicated database. VeriSTAR CSM is a tool
thought to be used from an early stage of design and maintained during the ships life. MARS 2000,
freely distributed, offers functions to approve scantlings of plating and stiffeners of any cross-section or
transverse bulkhead according to BV Rules and IACS Common Structural (CS) Rules for Bulk Carriers
and Tankers.
Other stand-alone application for specific topics are provided below:
RUDDER: for scantling of rudders according to BV and IACS CS Rules for Bulk Carriers.
BULK: for transverse bulkheads and the double bottom of bulk carrier structural strength calculation
LIRA: a shaft alignment calculation tool based on the transfer matrix method, which includes reverse
calculations based on measurements.
STEEL: a 3-D beam analysis program based on the displacement method that is a general purpose
tool, not linked to specific rules or regulations, but suited to be used in various fields of the marine
industry.
Multiplicity of information related to approval and inspection in new construction activity is managed
by BV and BV costumers with the project management tool VeriSTAR Project Management (VPM), a
web-based collaborative platform in which all parties, depending on the assigned rights, can review and
exchange information during the ships design and construction. Design review, product certification, and
survey of construction are covered, and access to project files, status of design review, comments,
certification, etc. are easily available through a personalized interface.
VeriSTAR AIM is an open, web-based system designed to improve inspection management and
information control during the ships life after delivery. The system provides not only the storage and
management of asset documentation, anomalies and maintenance history, and statistics, but also access to
the available calculation models (e.g. structure, stability).
The service VeriSTAR-HLC (Hull Lifecycle) supports hull structure monitoring during the vessels
service life. Vessels operated under this program may be assigned the class notation VeriSTAR-HLC.
Based on a 3-D model representation of the ship, the core of the system is the management of the
inspection results such measurements, cracks data, and coating condition. The tools is based on the
OpenHCM format for the exchange of structural inspection results. The underlying 3-D models of the
ship structure may be created by the 3-D modeler out of ships main drawings manually or with the
possibility to import shipyard CAD files including IGES and DXF formats.
5.3.3
China Classification Society (CCS)www.ccs.org.cn
Aiming at safety management, energy saving, and environmental protection, the China Classification
Society (CCS) has been continuously developing a comprehensive set of software to facilitate product
492
lifecycle management for the marine industry. The software system, which is collectively named
COMPASS, can help designers, shipyards, and engineers of classification societies carry out rule-based
scantling calculations or corresponding design optimizations.
COMPASS-WALCS is 3-D wave loads calculation software with four modules: linear wave loads
analysis, nonlinear wave loads analysis, linear hydro-elastic analysis, and nonlinear hydro-elastic
analysis. The software can be applied to 3-D wave loads calculations for floating bodies with zero
forward speed and to ships with conventional forward speed, and it provides basic references for
determining wave loads design value for ocean engineering in all types of design conditions. It can also
be used for analysis of springing and whipping loads induced by nonlinear wave loads.
COMPASS-RULES is a computer assessing system for sea-going ships in accordance with CCS Rules
for Classification of Sea-Going Steel Ships, IACS URs; Regulations for Statutory Surveys of Ships and
Offshore Installations by MSA of PRC; relevant IMO conventions and codes; and other updated technical
standards. The software has been widely and successfully applied in plan approval, rules development
and research, auxiliary design, and shipping safety assessment. The software is comprised of more than
20 functional modules categorized into four sub-systems: performance calculation, structure calculation,
shafting calculation, and electrical calculation systems.
COMPASS-CSR-SDP is a structural design and assessment program for bulk carriers and oil tankers,
compliant with IACS CSR rules. The software effectively provides all CSR prescriptive rule-required
calculation capabilities for all main structures including the cargo hold region, bow, stern, machinery
space, and superstructures.
COMPASS-CSR-DSA is direct structural analysis software developed in accordance with CSR Rules
for Bulk Carriers and Oil Tankers based on the MSC.Patran platform. With the softwares powerful
capabilities of automatic loading, boundary constraining, corrosion deduction, mesh refinement, and
buckling panel definition and a rich structural database of marine structure profiles, it can greatly reduce
calculation time in FE modeling, increase analysis efficiency, and enable engineers to perform CSR rule
required FE analysis both quickly and accurately.
COMPASS-HCSR-SDP is prescriptive rule check software with full coverage of Harmonized Common
Structure Rules (HCSR) prescriptive requirements and of the whole ship. The software can be used to
perform rule required assessments of hull girder yielding, ultimate and residual strength, scantling
requirement, buckling strength, fatigue strength, bow impact, bottom slamming, tank sloshing, bulk
carrier grab, steel coil, and other special requirements. With an integrated platform of parametric
modeling, streamlined processing, and optimized numerical calculation methods, it effectively enables
engineers to perform HCSR rule required assessment both quickly and accurately.
COMPASS-HCSR-DSA is direct strength analysis software with full coverage of HCSR direct strength
analysis requirements including yielding, buckling, and fatigue assessment of the entire cargo area. The
software is based on the same platform and with same set of uniform capabilities and user interfaces as
CSR DSA software. It is designed in accordance with the industry tradition which achieves high levels of
practicability, usability, flexibility, and efficiency and can effectively enable engineers to perform HCSR
rule required FE analysis both quickly and accurately in all stages of ship design, ship development, plan
approval, and inspection. Additionally, two structural-based through-life support packages are offered:
COMPASS-ERS is a consulting service software to facilitate the societys emergency response service and
COMPASS-CAP is a condition assessment program. COMPASS-CAP is provided as a technical service
for owners with no relation to the class of ship.
5.3.4
Croatian Register of Shipping (CRS)www.crs.hr
The Croatian Register of Shipping (Hrvatski Registar Brodova, or CRS) has developed several tools for
the strength analysis of ship structures as well as product lifecycle management databases. CREST is a
software package that has been designed for carrying out 2-D analysis of ship structures. It allows the
user to verify the compliance of the ship structures with CRS rules in the preliminary design phases and
during the approval process. The tool supports the structure design by performing calculations for the
scantling of ship structures. The geometric model can either be defined manually by the user in the
software environment or imported from MAESTRO. Automatic reports can be created after the analyses.
CREST CSR BC and CREST CSR T are two releases of the CREST software that are designed for
verifying the compliance with the CSR of the cross sections of bulk carriers and double hull oil tankers,
respectively. The ARGOS tool has been developed for performing intact and damage stability calculations
and longitudinal strength calculations. Both analyses are carried out in accordance to the CRS rules. The
software is used by designers, shipyards, and class plan approval and can be used for any type of ship.
As for the PLM databases, ZEON is a web-based software for the management of plan approval
process. It has been designed to simplify and improve the documents exchanges among the designers and
shipyards, the CRS plan approval department, and the CRS surveyors. SPP is another tool provided by
493
the CRS. This tool is a web-based service, useful for sharing information and documents about the ship
status throughout its life. The software is used by the ship managers, owner, CRS surveyors, and flag
authority, and it allows users to share and review documents and certificates, such as the data about the
corrosion of plating, during the through-life inspections. Both ZEON and SPP software are designed to be
used for any type of ship, and all data gathered using the software are stored in web-based servers.
5.3.5
DNVGL
5.3.5.1 Introductory comment

The following information is based on a guided interview on the premises of DNVGL in July 2014 and
on an additional study of publications related to the software tools being documented. At that time, the
merger of the two former classification societies was approved but not completed. The future strategy
regarding the software tools offered to customers and being used in-house was not developed. Therefore,
it was decided to fully focus on the software that was available at that time and to discuss the software
provided by DNV and GL in separate sections.
5.3.5.2 Det Norske Veritas (DNV)www.dnvgl.com
Det Norske Veritas (DNV) has developed a series of software packages for supporting the assessment
process of ship structures.
Nauticus Hull consists of a series of modules that are combined in a workbench. In its basic
configuration, the Nauticus Hull system provides for the rule check software that allows the user to verify
the ship according to DNV rules for both the classification of ships and the IACS Common Structural
Rules for Tankers and Bulk Carriers. All modules of the Nauticus Hull workbench are designed to be
used in several design phases, from the pre-contractual phase to the detail design and approval process
phases. It is used by engineers of design offices and shipyards, engineers of the classification society, and
the owners. In the following paragraphs, the different Nauticus Hull modules are briefly discussed.
The Rule Check software is a generic design tool for the initial hull girder design and optimization. It
provides engineers with calculations for ship hull structures according to DNV rule requirements and to
the IACS Common Structural Rules for tankers and bulk carriers. The initial information about the ship
structures is entered into the concept model of Nauticus Hull and can then be used in all the other
modules of the system. The design of the ship cross sections and transverse bulkheads is performed using
the Section Scantlings Program. By means of this tool, the user can perform a 2-D analysis of the ship
structures. The hull girder longitudinal strength, the local strength, and the buckling of plates and
stiffeners can be verified according to the rule requirement. Moreover, shear flow calculations can be
carried out by means of that module. All model data defined using the Section Scantlings Program can be
imported in other modules for defining a 3-D model for FE analysis. Fatigue analyses of longitudinals are
carried out directly in the Section Scantlings Program environment.
The High Speed Light Craft Software package provides engineers with rule checking and structural
analysis carried out according to DNV rules for high speed and light crafts. The software supports monohull as well as catamaran designs, including non-symmetrical cross sections, and allows for calculating
the section properties and shear flow distribution for vertical force, horizontal force, and torsional
moment.
For modeling and analyzing the 3-D beam structures and for carrying out a first assessment of ship
structures, DNV has developed 3DBeam software. This tool can be used as a module of Nauticus Hull or
as a stand-alone program. Nonlinear responses from structures such as wires (tension only) or gaps
between members of the structure can also be considered. The geometry can be also defined in a
Microsoft Excel spread sheet and then imported into 3DBeam. Loads are generated during the preprocessing phase of the model definition, and loads due to hydrostatic pressure, inertia, wind, snow, and
temperature can be taken into account. All loads applied to the model, as well as boundary conditions, can
be graphically shown. In the post-processing phase, the results of the analyses (i.e. deflections, shear
force, and bending moments) can be displayed graphically in the geometric model or numerically in a
table. All results can be exported in a Microsoft Excel format. Moreover, reports can be created
automatically.
For analyzing and modeling methods and devices for securing containers on ships, and for assessing
forces on containers during transport voyages, DNV has developed the Container Securing Software.
This tool performs the analyses of the container blocks and securing methods, according to DNV Rules
for Ships, once the structure has been defined and the environmental load combinations (wind forces and
accelerations) have been applied to the structure. The geometrical model can also be defined using an MS
Excel spreadsheet and then imported into the Nauticus Hull environment. The analytical solution is
carried out by means of 3DBeam software, and the results are post-processed in Nauticus Hull. Different
solutions of securing methods and devices can be compared performing the analyses using this tool.
494
The Buckling AssessmentPULS is a software package that has been developed by DNV for carrying
out buckling strength evaluation of ships and offshore structures. The tool is able to perform a strength
assessment of stiffened thin plate panels and stiffened and unstiffened cylindrical shells according to
DNV rules. The analyses take into account the simultaneous action of in-plane loads, such as bi-axial
tension, compression, and shear stresses, which can be combined with lateral pressures. The results of the
analyses are visualized in the PULS Advanced Viewer, a module that providing 3-D graphics of buckling
modes and redistributed stress patterns.
Direct strength analysis of ship structures can be carried out using Nauticus Hull and Sesam GeniE.
The latter is a software package provided for 3-D modeling and direct FE analysis of floating structures
made of beams and/or plates. The geometry of the cargo holds to analyze is defined directly in the Sesam
GeniE environment, or it can be defined according to the section data imported from the Section
Scantlings Program. The loads are applied directly in the Sesam GeniE interface; otherwise, rule load
cases, boundary conditions, and corrosion additions are defined in the Nauticus Hull environment. Results
of the analyses are graphically shown in the 3-D models, and reports are automatically generated. The
integrated code checks automatically verify the compliance of the structures with DNV Rules for Ships.
The FE model can be imported by other FE software. The analyses are carried out in Sestra, an FE solver
for linear structural analyses such as static analysis, free and forced vibration analysis in frequency and
time domain, and superelement analysis. The FE analysis can also be exported in other formats for
solving the calculation using external solvers such as Nastran.
DNV assigns the Nauticus class notation to the vessels whose structures of the midship area have been
designed using a procedure based on FE calculations. The class notation also covers extended
requirements such as fatigue calculations for longitudinal-frame connections in several critical areas of
the vessel. The notation is mandatory for oil tankers, bulk carriers, and container carriers, with lengths
more than 190 m.
The DNV Exchange is a web-based PLM software provided by DNV for exchanging and sharing
information among the different stakeholders of a ships construction. The database can be also installed
directly on a PC using a stand-alone license. It is designed to be used in all design phases. The
information that is taken into account are drawings, remarks, certificates, missing information for the
approval of the project, and in the stand-alone version, sketches and photos. Web access to the 3-D model
of the vessel hull structure allows the user to carry out calculations of coating areas and steel weights. The
3-D model is also used for viewing the coating status, thickness measurements, and renewal estimations.
The model is also designed to be used by surveyors for identifying areas to be given special attention
during the maintenance. The software creates reports automatically, documenting the results of analyses
and inspections.
5.3.5.3 Germanischer Lloyd (GL)
Germanischer Lloyd (GL) offers four integrated tools that can be applied in the modeling and analysis
during the ship design and the operational phases of a vessel. The tools are not specific to any ship type.
The modeling of the ship hull structure is performed with POSEIDON, which then determines or
evaluates the scantlings of all longitudinal and transversal structural components, taking into account
local and global loads. Yield, buckling, and fatigue assessments are performed. The dimensioning is
checked against all relevant GL rules for hull structure approval, including CSR. The initial 2-D modeling
approach in the form of hull cross sections allows a fast build-up of a structure concept model. For special
ships like container vessels and bulkers, parameterized templates serve to generate a full topological
description in a very short time by specifying few parameters and describing specific the ship the
calculations that will be performed. At this stage, the designer does not have to specify any scantlings of
plates and stiffeners because POSEIDON offers the unique approach of calculating the according values
by automatically observing all loading conditions, which include global (e.g. hull bending) as well as
local (e.g. pressure heads) loads. The calculations are performed in an automated, iterative process until
convergence is reached. All calculated scantlings can be further analyzed by the designer by applying the
built-in explanation function, which links to the relevant rules that have been applied, including the
specific results. This allows the designer to get a sound insight of the necessary scantlings of the design.
Due to the fully topological modeling approach, compartments are generated automatically while the user
must supply a minimum amount of additional data such as contents, maximum pressure head, etc.
Additionally, changes to the ship hull form, the ship internal subdivision, and the structural design, which
occur frequently in the early design process, can be incorporated easily, while the required updated
scantlings are recalculated automatically. Furthermore, integrated model checks guarantee that unrealistic
or error prone ship structure models are not being applied for the calculation of the scantlings. Interfaces
exist to import the hull form (IGES) and compartmentation information from the NAPA Steel system.
The 2-D cross sections (which can be copied to other sections and adapt to changes, e.g. changes to the
495
hull form geometry, automatically) serve as the basis for built-up 3-D cargo hold models that perform FE
analysis by checking yield, buckling, and fatigue. The 3-D modeling functions are not restricted to the
midship region, but rather, they allow for a complete 3-D modeling of the ship. For the design of hatch
covers, an integrated beam and grillage analysis can be performed. The POSEIDON data model is stored
in a single file, and an API allows the user to get external access to the data. The POSEIDON data model
also serves to calculate the remaining hull structure capacities in case of damage, in which case, a
function is applied in the emergency response service.
The special program ShipLoad calculates the loads on the structure in a seaway for which numerous
loading conditions and wave data are to be observed. The output is either used as input to POSEIDON 3D FE analysis or can be applied to a global FE model to be analysed by any external FE analysis package.
The program Pegasus can be applied to manage hull structure thickness measurements in the ship
operation phase. The program is capable of developing a ship specific measurement strategy that results
in a detailed measurement plan, which guides a persons measurement taking. Measurement data are
stored in a database that automatically links the measurements to structural components. By this process,
taking the actual scantlings into account, it is possible to reassess the capability of the ship hull structure.
Furthermore, maximum diminution parameters serve to rationally evaluate the status of the structure as
well as perform detailed statistical analysis.
ShipManager Hull is a tool that can aid the ship operator in monitoring the structural condition of the
ships in a fleet. The program builds upon a hull lifecycle model (HCM), which is derived from the
prescriptive model generated by POSEIDON. All data relevant to be managed during the ship lifecycle
concerning the hull structure can be stored and visualized as well as reported upon request. This includes
scantling values ranging from as built to any number of actual values in a timeline. Additionally, the
status of the coating can be documented with respect to specific structural components or regions. The
system is capable of handling data, documents, and pictures, and it shows the conditions found during
inspections. Regions can be highlighted where special attention should be given in future services.
5.3.6
Korean Register of Shipping (KR)www.krs.co.kr
Sea Trust-RuleScant is the main Korean Register of Shipping (KR) software used for supporting ship
structural design and strength assessment according KR Rules, including also IACS CSR for Bulk
Carriers and Oil Tankers and enabling certification with CSR-H class notation. Software functionalities
cover hull cross section generation and 3-D modeling of parts of the ship or of the whole ship. The input
of model geometry and data is supported by the possibility of import geometric models from IGES,
STEP, and DXF, but it is also assisted by tools (e.g. dialogs, menus, and spreadsheets similar to Microsoft
Excel), providing recommended scantlings. FE models may be imported from MSC NASTRAN using
RuleScant for the automatic loading generation based on rule requirements. The analysis is carried out by
external software such as MSC-NASTRAN or IPSAP. Post-processing capabilities include buckling and
fatigue assessment.
New ship designs requiring a complete ship model may take advantages from the extended capabilities
offered by Sea Trust-ISTAS (Integrated Structural Analysis System), an integrated software system that
includes the components for the direct structural analysis. The system includes specialized modules for
ship motion and wave loads analysis; stochastic analysis and design loads generation; solver for FEA
analysis; simulation of ship motions, waves, stress, and displacement; and buckling and ultimate strength
assessment. Characteristics of the ship, such as hull form (limited to symmetric types) and weight
distribution (generated with whole ship structural model), are used in the seakeeping analysis (linear
frequency domain) to obtain dynamic loads based on a 3-D Panel theory. With a return period
corresponding to the ships life, stochastic analysis is carried out to obtain the most probable extreme
value. Design loads obtained from hydrodynamic analysis are transferred automatically to the global ship
FE model.
Using Sea Trust-FANSYS, spectral fatigue analysis is performed to evaluate the fatigue life directly
with stress results from whole ship analysis at all heading angles and wave frequencies. Fatigue loads, in
conjunction with a very fine mesh for evaluation of the hot spot stresses, allows the calculation of fatigue
life of the vessels specific structure. Also, simplified fatigue analysis method is supported. Fatigue loads
are estimated from KR Rules formulas, and then the fatigue levels for longitudinals, connections, and
hopper knuckle parts are calculated. Sea Trust(FSA 1), Sea Trust(FSA 2), or Sea Trust(FSA 3) class
notations may be assigned to vessels assessed with this software.
Management of thickness measurement data and relevant survey information are supported by Sea
Trust-TM.
Sea Trust-SLM, or KR Ship Lifecycle Management System, provides a 3-D ship model in which all
information during ship design, construction, operation, inspection, repair, and recycling can be stored for
easy access. It can be interfaced with the customers management system, and ship builder, owners,
496
operators, and surveyors can share and use information such geometric model, drawings, product
specification, maintenance data, etc.
Lloyds Register (LR)www.lr.org
RulesCalc has been developed for the assessment of ship structural designs according to Lloyds
Register (LR) rules and regulations for the classification of ships. It can be used as a stand-alone system
or in conjunction with other design software packages, including NAPA, Tribon, and LRs ShipRight
Structure Design Assessment (SDA). Project information is recorded in a hierarchically organized using
the analogy of a family tree in which there are parent branches or folders and sub-branches or sub-folders
that are children of the parent. This architecture allows the program to perform various assessments, such
as local scaling or 2-D ship section analysis, on the basis of consistent data inputted once at the beginning
of the design. This data is editable, if necessary, during its development. Any calculation performed is
tracked and visible in the graphical interface with the clear reference to rules applied.
RulesCalc, used in conjunction with Rulefinder, an interactive PC-based software system providing a
searchable access to a library of the latest requirements from both LR and IMO, gives the user the
immediate evidence not only to calculation results but also to rules requirements and formulas.
The complete end-to-end assessment of ship structure against LRs ShipRight SDA notation and
Fatigue Design Assessment procedures (FDA notation) is supported by the ShipRight 3-D FE analysis
tool. It includes features such as an interface with NASTRAN and PATRAN FE model tools, enhanced
calculation of ship motions and hydrodynamic loads, advanced spectral analysis and incorporation of
100A1, and IACS North Atlantic wave environments.
Transferability of electronic ship models from commonly used CAD packages (e.g. SmartMarine 3D
Intergraph, NAPA Steel, Tribon-Aveva, AutoCAD, and IGES format for structures and hull forms) to LR
applications is also enabled by the software Interface Toolkit.
Owners and managers of an LR-classed fleet may take advantage of Class Direct, a free service that
provides data for a single ship or an entire fleet that is relevant to survey status and historical records.
Class Direct may also contain thickness measurements reports created with Argonaut, a dedicated tool
developed by LR for this purpose.
5.3.7
Nippon Kaiji Kyokai (ClassNK)www.classnk.com
Nippon Kaiji Kyokai (ClassNK) has a comprehensive system that aims to support a ships life. PrimeShip
is the collective name for the software systems, guidance, and data services developed and offered by
ClassNK. The software groups are categorized as Hull, Machinery, Operation, and Maintenance, and
software related to hull and structural design are described below.
PrimeShip-HULL (HCSR) supports prescriptive rules and direct strength calculations, and it creates
models and performs strength analyses from stem to stern in accordance with the harmonized CSR
(HCSR). It allows the designer and yard to rapidly carry out case studies of potential designs and
optimization. It is a new software, compared to the previous PrimeShip-HULL version, by combining
the rule calculation and DSA analysis software with newly developed interfaces for data linkage with
commercial CAD software. The interfaces are XML-based and developed by PrimeShip CAD Interface
Project composed with ClassNK and members from 13 Japanese shipyards and 3 software venders. These
have been opened on the ClassNK web site (https://www.ps-cad.jp/outside/index.html). And these
provide linkages between CSR calculation software PrimeShip-HULL and commercial naval architecture
design packages, such as AVEVA, NUPAS, CADMATIC and NAPA STEEL, and the duplication of
modeling work on both a package possessed by a yard and class rule software can be reduced. Geometry
models and additional data can be input manually as well. The rule calculation assists taking input from
rule calculation software and generating recommended scantlings. For DSA analysis, both HyperWorksbased and Patran-based PrimeShip-HULL (HCSR) are developed and the original functions of the base FE
software are available. DSA analyses are carried out using Nastran, Raddioss, or Optistruct. Outputs are
geometry, scantlings, and other data, and reports are automatically generated.
PrimeShip-HULL (Rules) is a rule calculation software for hull structures. This free software allows
designers to calculate the requirements for structural members in accordance with IACS-CSR for bulk
carriers and oil tankers and Part C of ClassNK Rules. It is comprised of three software programs for
specific rule sets: PrimeShip-HULL(Rules)/NK Rules for ClassNK Rules, PrimeShipHULL(Rules)/CSR Bulk Carrier for IACS CSR for Bulk Carriers, and PrimeShip-HULL(Rules)/CSR
Tanker for IACS CSR for Double Hull Oil Tankers. It has functionalities of a manually entered geometry
models and data, or it can import geometry models from CAD software and a quick calculation for
strength members from cross sections to specific local members. It can then create a report in ClassNK
required formats.
PrimeShip-HULL (DSA) is a Patran-based FE direct strength assessment system in compliance with
the IACS-CSR for Bulk Carriers and Oil Tankers and Part C of ClassNK Rules. It includes three
497
dedicated software programs for specific rule sets: PrimeShip-HULL(DSA)/CSR for IACS CSR for
Bulk Carriers and Tankers, PrimeShip-HULL(DSA)/Guidelines for ClassNK Guidelines for Container
Ships, etc., and PrimeShip-HULL(DSA)/Ore Carriers for the ore carrier edition in accordance with
ClassNK Rules. It supports strength assessment for bulk carriers, oil/chemical tankers, and container
ships for designers, yards, and the class approval process. It has functionalities of an automatic
identification of structural members and compartments. Load cases in accordance with the specific rules
and guidelines can be taken into account, and recommended scantlings are available.
PrimeShip-HULLCare is a hull maintenance information service developed by ClassNK. It provides
ship owners with plate thickness measurement result records and hull maintenance information on
ClassNKs web site. It helps make ship maintenance planning an easy process and supports an enhanced
survey program (ESP) for ships. Users are able to view the information with PrimeShip-HullCare/2D
Internet and PrimeShip-HullCare/3D Model. The former of these contains significant condition
assessment and repair data including steel thickness at measured locations, photographs taken during
inspections, specifications and plans for repairs, paint maintenance, condition assessment scheme reports,
and hold frame replacement requirements for existing bulk carriers of IACS S31 requirements. The latter
of these, which is an optional service, provides an easy understanding of the ships condition with a
rotatable and zoomable 3-D model. It also allows the evaluation of a current section modulus by using upto-date thickness measurement records.
NK-PASS is a web-based system for plan approval status service, which supports designers, yards, and
class to submit e-drawings and manage approval process. It can manage submitted drawings, comments,
and their responses. Different menus for shipbuilders, machinery manufacturers, and design firms are
offered, and users can be registered in a department/section unit.
NK-SHIPS, a web-based service for owners and operators, is designed to retrieve vital information for
ship maintenance and management, such as survey status, periodical survey items and due dates, survey
history, survey records, and ISM/ISPS audit status, that assists in the effective maintenance and
management of ships.
ClassNK Archive Centre (As-Built Drawings Storage Service) is a database for the management of
the ships drawings. It provides an integrated management system of the ships drawings using cloud
computing, which can be accessed through the internet. The information, like confidential drawings, can
be securely and effectively transferred to the interested parties in case of emergencies.
5.3.8
Polish Register of Shipping (PRS)www.prs.pl
The Polish Register of Shipping (POLSKI REJESTR STATKW S.A., or PRS) has developed in-house
software for supporting surveyors in the approval process of the related technical documentation. These
tools are used by the PRS plan approval department and by the surveyors during the management of the
documents and information about new constructions or ships in services. As for the Product Lifecycle
Management databases, the key programs provided to the surveyors and PRS plan approval department
for supporting their work are SurveyPRS, NNB, and INAD. These are web-based systems that allow a
hypertext-based rule search to be performed.
The SurveyPRS is oriented to support the surveyors in the exchange of information, drawings,
certificates, remarks, and missing information. It was designed to be used during the construction phase
of new ships and during the operational phase until it is dismissed. The NNB and INAD are used during
the entire life of the ship, starting from the pre-contractual design phase until its dismissal. The NNB tool
is used for the exchange of the certificates issued by PRS between the surveyors and the plan approval
department and between the shipyard and the owner. The INAD tool allows the exchange of information
between the surveyors and the approval department of the classification society of the drawings and of the
remarks and missing information about a ship under construction or in exercise.
As for the tools for the scantling assessment of ship structures, PRS developed two tools used by the
surveyors and approval department of the classification society as well as by the ship owner. The
CSR_PRS is designed for the strength calculation of ship structures according to the common structural
rules. The Research and Development department of PRS is developing in this period a new HCSR_PRS
tool that is designed for the scantling assessment of ship structures according to the new Harmonized
Common Structural Rules. Both tools can verify the hull structures of the oil tankers and bulk carriers and
allow the user to carry out the scantling assessments of all parts of the ship structures. Calculations for the
minimum scantling of the structures, in compliance with the rules, as well as the 2-D analysis of the cross
section of the ship hull, can be carried out. Moreover some tools have been implemented in the CSR_PRS
and HCSR_PRS software to carry out direct calculation of strength of local structure, fatigue analysis,
buckling analyses, and ultimate strength analysis. These tools allow the user to import an FE model
created with other software and to define its element parameters. The load condition of the FE model can
be defined according to the HCSR rules. The FE analyses are carried out by means of NASTRAN, and
498
the results can be analysed by the LOADER and LOADERBRIDGE applications. The software includes
automated reporting functions.
WINSEA is a tool for supporting the surveyors of PRS in the approval process of ship technical
documentation. It allows the user to carry out scantling assessment of ship structures according to PRS
rules. By means of this software, a 2-D analysis of the hull cross section can be carried out. The model
geometry can be input directly in the software and a limit state analysis can be performed. The load
conditions for all the analyses are defined according to the classification society rules. Automated
reporting functionalities are provided.
STAB_PRS is a tool that allows the owners to carry out stability calculation for RORO carriers,
container ships, and bulk carriers. The software provides immediate calculations for verifying the stability
condition and the strength calculation in the loading and unloading process in day-to-day operations. The
assessment can be carried out in both intact and damage conditions.
5.3.9
Registro Italiano Navale (RINA)www.rina.org
Leonardo Hull 2-D (LH2D), a Registro Italiano Navale (RINA) tool for 2-D structural analysis, has been
continuously updated, following rules development, including specific calculation capabilities such as
structure verification in ultimate strength conditions and torsion checks of ships with large deck openings,
as in the case of container ships.
Data inputted in Leonardo Hull 2-D may be directly imported into Leonardo Hull 3-D (LH3D), the
tool dedicated to FE analysis. The 2-D section is automatically extruded, and the user is enabled to
complete the model by specialized features for modeling structures, such as bulkheads, stools, and trunks,
specifically tailored for ship design. Loads are automatically generated. According RINA and IACS
common structural rules, the Finite Element Model (FEM) analysis is performed via a two stage
approach: the analysis of a hull portion extended over three cargo hold lengths through a coarse mesh
model and a more detailed check of specified areas where finer mesh may be automatically generated.
Rules checks are automatically performed through the post-processing results of calculations based on the
MSC-NASTRAN solver.
LH2D and LH3D in particular can used both in the design stage and also during ships life-cycle. They
support a program of planned maintenance management based on the programmed updating of hull
scantlings in accordance with survey measurements, thereby facilitating the accurate prediction of
corrosion rates and steel renewal dates.
The use of FEM is certified by RINA with the assignment of the additional class notation STAR
HULL NB for new building and with the additional class notation STAR HULL in case of the creation
of the model for structural maintenance management. Existing ships may also take advantage of the
additional service Technical Advisor, offered by RINA, in which an expert team, thanks to the
availability of ship information, may give immediate support in emergency situations such as collisions,
groundings, fire, explosions, and oil spills.
RINA also offers as a stand-alone tool the software application Advance Buckling Check (ABC), which
performs direct calculations for thin walled plated construction. The non-linear, large deflection plate
theory of Marguerre and von Karman is applied, as prescribed by CSR for Oil Tanker.
The approval process is supported by the software application Leonardo DRAW, a website that gives
shipyards access to updated information relevant to their drawing approval status.
Information relevant to RINA class fleets is available throughout Leonardo INFO. It is organized into
two environments: Public, which is available to everyone and provides general data, and Private,
which is available only to RINA clients showing technical data, status of survey, certificates, historical
survey status, and survey reports.
5.4
Conclusions and future challenges
The software overview presented, compared to the one reported in ISSC 2000, shows a clear development
of class society tools. Some tools have been simply updated according to new rule requirement. In
particular, the creation of both CSR and CSR-H have been an important challenge as entirely new rule
sets have been created. However, new features have also been introduced in order to extend the support to
designers, surveyors, and ship owners.
CSR and CSR-H also defined a new scenario that creates common rules, among which a class societies
compete in terms of support offered to shipyards and ship owners. As the rules are common, designers
may choose the software to use independently from the classification society in charge of the approval.
Criteria for such software selection would include accurate analysis criteria, evaluating additional
software capabilities, licensing strategies, required training and technical support. A benchmark study
comparing the options that successfully implement the CSR-H rules against these criteria would be an
interesting future challenge.
499
As highlighted at the beginning of this chapter, no tests or benchmarks have been performed with the
analyzed tools; therefore, a comparison among software functionalities able to identify the real capacities
cannot be provided. As a conclusion to this section, considerations about general aspects and the overall
philosophy are presented, leaving detailed analysis to future investigations. It is clear, however, that the
trends highlighted in Section 5.1increased use of 3-D FEA, on-line collaboration and extension of
data/model use throughout the vessels lifeis common to almost all major classification societies in
terms of services offered.
The 2-D section analysis and scantling definition according to rules still maintain their importance in
the pre-contractual and first design stages. Tools-based databases are very useful in a design phase in
which main design variable definition is in progress. However, standalone applications dedicated to
particular purposes allow users to obtain results very quickly imputing only a smaller set of strictly
relevant data.
Regarding 3-D FEM, while the standard approach is for the design tools to use an external solver
for calculations, different approaches have been found in modeling and post-processing functions. For
example, in a scenario in which most of the tools are standalone applications, some consideration
about the choice to develop an application as customization of a general FEM tool may be a suitable
approach. It has to be considered that 3-D FEM models are often used by designers for internal check
and analysis (e.g. normal modes analysis, vibrations) with commercial FEM tools since they are not
adequately supported by classification societies software. Developing class societies tools as
customization of commercial ones, may appear to be a constraint to users in the choice of commercial
FEM tool, but it extends the possibility of integrating commercial software capabilities and reducing
the problem of data transfer. Importing and exporting data capabilities and implementation may not be
always optimal solutions, in particular if the design process requires interactive changes. However, for
less complex vessels, the availability of a toll with specialized features for modeling typical structures
without involving data importing from external software and consequence training may remain the
quickest solution.
Dynamic loading approaches and spectral fatigue analysis are probably the biggest innovations
introduced in tools since the last committee review. This is clearly an attempt to go beyond the rules
checking by providing instruments to fit the design goals, rather than simple approval. In this field,
interaction between structural and hydrodynamics software is probably the most critical aspect.
Automatization of loads transfer and implementation of statistical criteria simplify users work, but it
may reduce control of the calculation process and relevant results if a proper training for users is not
foreseen.
During the software review, some information about optimization capabilities has been gathered. The
thorough evaluation of this strategy would require a careful analysis and tests that are not subject of this
chapter due to the limited resources available. A large potential of growth in this area may be found,
considering that rule requirements are already implemented and so are available as constraints.
The increasing number of tools offering support to phases of a ships lifecycle, other than only design,
is interesting, in particular, hull monitoring and maintenance, which introduce the structure reassessment
based on measurements of actual plating thickness performed by surveyors.
Class societies PLM databases for new buildings and existing ships appear to have very similar
functionalities. Main differences may be found in the possibility of interaction between class societies
and other users (in some cases, external users may only access the information) and in the way the
information is stored. Today, storage capabilities or data accessibility can be simply ensured, but the
focus still remains on the quality of information and the related possibility of a quick use them. The
risk is to save a great amount of files containing drawings, reports, certificates, and measurements, so
that even if properly organized, it may become difficult to use, particularly for emergency support.
Solutions based on 3-D models as a platform for most important data storage may be a way to
supersede these mentioned problems, especially if interaction with stability and structural models is
foreseen. Obviously, in that way, class societies role of guarantor of design intellectual property
rights rises to a higher level.
In this chapter, ten classification societies have been considered, and capabilities of approximately
eighty types of software tools and applications have been briefly described. Despite all that has been
discussed, this chapter represents only a part of what the class societies software has to offer, and further
study of these tools is highly warranted.
6.
STRUCTURAL LIFECYCLE MANAGEMENT
6.1
Introduction
This chapter illustrates the importance of structural lifecycle management, including lifecycle data
management, Operation & Maintenance (O & M), and condition monitoring systems influence on the
500
design process and integrity of marine systems. Lifecycle data analysis can be used as a feedback loop for
the design team to test and check their designs. It can also assist with the identification of the most
common failures and their consequences for future designs. This chapter contains four major sections:
Tool Development, Data Interchange and Standards, Integration with Repair, and Integration with
Structural Health Monitoring Systems. All four sections contain widespread and relevant information on
each topic from the latest research papers. At the end of each section, a critical review is performed,
identifying potential gaps.
6.2
Tool development
Over the last few years, various maintenance methodologies and tools have been developed that assist the
overall lifecycle structural management. These methodologies and tools help minimize the risk and
maximize the availability of the systems. They also provide valuable feedback on system reliabilities for
the design team. One type of this advanced smart maintenance system was developed by Lee et al.
(2012b), which provides onsite engineering data to field engineers including 3-D CAD design
information during the work process. Radio-frequency identification (RFID) technology is applied to
derive exact information into the smart maintenance system where it has been incorporated into mobile
devices. Subsequently, a framework was presented by Temple and Collette (2013b) to schedule
maintenance cycles for naval vessels in order to minimize the lifetime costs of the structure.
Another maintenance strategy was introduced by Lazakis et al. (2012a), which is called Reliability and
Criticality Based Maintenance (RCBM). This methodology was applied for creating optimum
maintenance system on board a Diving Support Vessel (DSV). Reliability and critical analysis of the
main systems of the vessel were the starting points of this approach. Propulsion, lifting, anchoring &
hauling, and diving systems were the subsystems analyzed in this case study. Furthermore, by using the
Dynamic Fault Tree Analysis (DFTA) tool and the Birnbaum (Bir), Criticality (Cri), and Fussell-Vesely
(F-V) reliability importance measures, the results of the above analysis have been validated. This created
valuable information regarding component behaviors to be added into the lifecycle data management for
the design process.
A paper by Ruede et al. (2012) also practiced a semi-automatic Fault Tree (FT) synthesis method but in
conjunction with an innovative tool called Hierarchically Performed Hazard Origin & Propagation
Studies (HiP-HOPS), which expanded to use the simulation platform SimulationX. This approach has
the benefit of implementing a system model for the double purpose of behavioral simulation and
reliability analysis. Additionally, the method features automation and simplification of reliability and
availability prediction. This provides further advantages of enhancing speed and efficiency of the design
modifications without any requirement for updating simulation and reliability models separately. As a
result, it simplifies iteration of the design process in complex systems.
Schleder et al. (2012) presented an application of Bayesian Network (BN) to analyze different event
scenarios using the parent marginal probability distribution of each component. This required
computation of the posterior joint probability distribution of component subsets and function of the set of
all nodes. This model has been implemented on a Liquefied Natural Gas (LNG) Regasification System on
a Floating Storage and Regasification Unit (FSRU). Gazis (2012) also investigated another type of
probabilistic response analysis and reliability assessment on subsea free spanning pipeline systems. These
systems were exposed to random, wave-induced hydrodynamic forces. Monte Carlo simulation
methodology was the chosen analysis tool for demonstrating the advantages of integrating a reliabilitybased design.
Multi Attribute Decision Making (MADM) methodology was described by Lazakis et al. (2012b) for
selection and design of the optimum maintenance strategy for a given system. The TOPSIS ranking tool
was used in combination with the rating and aggregation for converting unclear verbal terms from experts
into more practical values. Markov Decision Process (MDP) is another Decision Support System (DSS)
tool that achieves optimum design and maintenance strategy for time-dependent environmental agreement
by using a consecutive decision-making frame work. A non-stationary MDP system, developed by Niese
and Singer (2013), considers ballast water exchange and treatment policy changes. Therefore, MDP to
model lifecycle decisions on this research determines the summary of the initial approach, outcomes, and
conclusions resulting from the implementation of the mentioned framework.
The implementation of lifecycle assessment (LCA) and lifecycle costing (LCC) in marine systems
design is proven. Aspen et al. (2012) have adopted this methodology in their work and discuss the future
opportunities overcoming the difficulties facing it. The environmental effect from shipping is becoming
an important factor for stakeholders in the maritime industry. Therefore, a holistic approach has been
developed by Fet et al. (2013) that compares existing environmental assessment tools and introduces
systems engineering as an approach to lifecycle designs. ABB has also prepared an insight into holistic
501
performance management and optimization of any vessel type by recognizing energy efficiency,
availability, and safety of the vessel (Ignatius et al., 2013).
Embrey (2012) examined the effect of human activities on errors in operations or maintenance on
major marine accidents. Consequently, a novel regime was executed in an organized manner with a
specific requirement regarding human caused risks. This resulted in the creation of a set of guidelines for
addressing regulatory requirements for safety critical maintenance activities in the marine sector. Another
safety related study was performed by New (2012) on the policy for Reliability Centred Maintenance
(RCM) in the Royal Navy fleet. This study illustrates the engagement technique of RCM into the Safety
Case regime of the First Sea Lords Safety Argument of the UK Royal Navy rules. It supplies an
integrated approach to safety and maintenance management. It also debates the potential paybacks of
RCM by addressing deficiencies.
Tanizawa et al. (2012) proposed a new experimental methodology for self-propulsion testing of a
marine diesel engine simulator in waves on a model ship. A marine diesel engine simulation program in
this study provided a real-time control system of propeller rotating speed, imitating the characteristics of a
marine diesel engine.
Reliability analysis in the marine industry can be also regulated and computerized. Hence, Linton
(2011) created a management system for the RCM program of the Coast Guard surface fleet. This
provides the standards and abilities desirable for the development of preventive maintenance systems that
satisfy operational RCM requirements and standards called SAE JA-1011. Another study by Tiusanen
et al. (2012) researches the application of management of Reliability, Availability, Maintainability, and
Safety issues (RAMS) on offshore wind farms in the initial stage of turbine design. This paper presented
guidelines for comparison of critical components on different offshore assets based on system availability
and safety. Classification and evaluation criteria for different RAMS+I (Reliability, Availability,
Maintainability, Safety, and Inspectability) factors are outlined and discussed as well. Additionally,
Brown (2012) introduced overall life Reliability and Maintainability (R & M) management and
assessment for defense programs. Capability and cost drivers are the major aspects of this study. It also
discusses the management of R & M in topics of derivation of requirements, use of modelling, conduct of
R&M audits, failure reporting, analysis and corrective action system (FRACAS), reliability
demonstration testing, R & M risk management, and the R & M and dependability case. The
dependability case itself identifies the sub-systems that have low reliability and are more prone to
failures.
The use of decision making tools for repair planning was considered by Niese and Singer (2013). In
this regard, the optimal maintenance strategy for time-dependent environmental agreement has been
considered with regards to the implementation of the chronological decision-making framework MDP for
ballast water exchange and treatment policy changes.
All of the above research papers present the development of tools by incorporating lifecycle analysis
and maintenance in order to maximize efficiency of the overall system and minimize risks, thus ensuring
safe system operation. These improvements can consequently be used as feedback for further enhancing
system performance from the initial design stage. This means that invaluable information learned from
failure modes, causes, and consequences can provide insight into the behavior of different systems and
their components. In this respect, various tools are presented, which are also related to maintenance
methodologies, such as RCBM, assisting in overall maintenance methodology development and lifecycle
analysis. A major part of the complementary tools presented above is related to reliability analysis tools
such as DFTA and BN. These reliability tools provide fundamental information and an analysis platform
for measuring system behaviors and availability. Another important aspect is related to methodologies on
the implementation of DSS. Two major types of DSS presented in this section are MADM and MDP.
Both reliability and DSS tools specify the overall system performance and highlight the challenging areas
related to the application in the lifecycle data management stage.
Moreover, this section also shows that there is continuous development in reliability-based approaches
for marine applications including tools such as RCM and RAMS methods. A number of case studies have
been identified, which are used for validating and testing the applicability of such approaches. However,
the review of the existing research and applications so far has revealed potential gaps that can be tackled
through future research. One of these gaps is related to the incorporation of condition monitoring systems
within reliability and criticality analysis tools to obtain up-to-date and real-time results for decision
making. Another gap identified through the examination of different methodologies is related to the
enhancement of the relationship between business aspects with risk and safety. In this respect, the
financial resources, the level of risk and safety should be considered in a combined way in order to
achieve a more accurate representation of the overall system lifecycle. Another noticeable issue is the
emerging combination of various maintenance tools into hybrid tools and methodologies that can be used
on lifecycle data analysis and management. A good example is presented through the potential
502
combination of DFTA and dynamic Bayesian networks (DBN). In this case, although DFTA is a widely
applied tool on systems with large number of components, DBN can demonstrate the inter-component
effects and inter-connectivity more efficiently while also considering the time variance effect among
different system components.
6.3
Data interchange and standards
As many players are involved in the lifecycle management of marine structures, efficient data exchange
between parties is a perquisite for successful lifecycle management. Foremost, it is beneficial to have a
data exchange system between the design office and the Operation & Maintenance (O & M) team. This
provides useful complementary information for future designs. In the recent decade, relays, push buttons,
and light-bulbs have been replaced by processors, graphical user interfaces, keyboards, and track balls.
Therefore, high-level computer languages like C++ and JAVA are becoming standard in the marine
industry. As a result, this section illustrates all the data interchange forms and standards, which make it
possible to maintain electronic copies of vessels history throughout its service life in a centralized
electronic location. Thus, Scherer and Cohen (2011) have discussed the Naval Surface Warfare Center
Ship Systems Engineering Station (NAVSSES) as a center for collecting and managing machinery system
data. Siemens Industry has also created software using their own light-weight 3-D neutral format JT for
shipbuilding (Malay, 2012). This uses Visualization & Digital Mock-Up (DMU), Documentation &
Archiving, and Data Exchange. JT format can facilitate data exchange among all stakeholders in a
shipbuilding project. This can be implemented for Product Lifecycle Management by providing a shared
platform for efficiently saving, representing, organizing, recovering, and recycling product-related
lifecycle information.
Bureau Veritas (BV) developed a Lifecycle Management (LCM) system that uses some of its own inhouse tools such as the VeriSTAR-HLC, VeriSTAR-AIMS, VeriSTAR-Hull, and Hydrostar modules
(Renard, 2012). Using this system, shipyards are able to recall the drawings, technical notes, and
computerized models of ships, as delivered, in an electronic Ship Construction File (SCF). SCF would
also contain a structural 3-D model to be used by ship owners for classification societies and thickness
measurement companies.
Tetrault (2012) presented the remote monitoring of engine rooms in marine industry. He highlighted
that remote monitoring is well developed in land based power production systems, but the approach for
the marine industry is not as simple. Thus, the offshore wind industry can be used as a starting point for
developing such a system for the marine industry. A paper by Goni and Jambrina (2013) discusses the
CAD design language implemented in ship design concepts by using three-level software architecture.
The development framework has been considered in Windows 8 by using web applications, such as
HTML5, CSS3, and JavaScript, and other native applications using C/C++. NET applications can be also
developed using C#, VB, and F# programming languages. The user interface and experience for non-web
application is described with XAML and the 3-D API for games and design applications is called
DirectX.
In this case, the project by INCASS (2014c) focused on using different database standards for marine
and shipping fields. One of the database standards defined in this project is called ISO 10303-11, which
uses EXPRESS data format on XML schema to define its data structure. Another standard explained in
this project is the Hull Condition Monitoring (HCM) 2.0 standard which focuses on data import strategy.
This includes what type of data from structural analysis is recorded in the database. Their overall database
model has two major sections, machinery and structural, with Java APIs to connect it to other analysis
methods such as a structural reliability analysis tool. Stone (2012) described how Maritime Training
Systems (MTS) achieved success on a Contracting for Availability (CFA) framework, which means it
keeps a system in the agreed condition. This can be achieved getting both contracting parties to agree to
the processes and methods to be implemented, which enable the Key Performance Indicators (KPI) to be
assessed. He also reflected his concerns for gaining decent quality data to support the CFA initiative. A
web-enabled tool called RAMtr@ck was adopted for application on MTS. This supplied user-friendly,
semi-automated data collection and analysis tools for instantaneous response of utilization and
availability of the system.
Standards are important in every stage of a vessels lifecycle. That is why Shin et al. (2012) developed
a prototype of a ship basic planning system based on the advanced IT systems for small and medium
sized shipyards. For this analysis, a standardized development environment and tools were selected.
These tools were used in system development for increasing competitiveness of small and medium sized
shipyards in the 21st century industrial environment. Thomson and Renard (2013) expressed the
importance of ship design standards in their research. Incorporating advanced information management
technologies, they studied the 3-D models of the as-built assets. They also delivered an inclusive
indication of challenges, solutions, and most suitable practice in the handover from shipbuilder to
503
operator of a complete digital information asset. Finally, a paper from Mebane et al. (2011) discussed the
variations of a government-led design on design approach, including schedule, organization structure, and
design methodology. A new technique, called Set Based Design (SBD), has been added for assessing
various systems and ship alternatives, although formal links to lifecycle management remain to be
explored in this framework.
Hifi and Barltrop (2013) used a central database platform for forecasting structural defects. This
platform assisted in the development of a reliability model that can consider individual components.
However, this provided an insufficient representation of the overall reliability of the ship. Therefore, a
method to adjust the reliability models using the data from experience-based methods was moreover
produced. Thus, the critical structural details have been used by the inspection companies, class
surveyors, ship managers, and ship designers for the calibration of the inspection planning for the
decision support tool.
For regulating structural elements of the vessels and marine structures, it is important to have both
safety policies and decision systems in place. Consequently, Forbes et al. (2012) have demonstrated an
end to end safety argument for the fleet of Royal Navy. This safety policy is a Joint Service Publication
430, which applies to all Ministry of Defense maritime activity. This paper also considers the knowledge
achieved from the loss of the Nimrod XV230 case study for endorsing the recent development and
implementation of the Navy Safety Improvement Plan. Another study by Ostuni et al. (2013) looked into
the application of DSS in structural analysis management and their application in shipboard security for
supporting crew members in the effective conduct in failure events. This is a knowledge-based DSS
integrated within a Damage Control System (DCS) for navies.
Summarizing the above, there are various formats that can be used for data interchange. One of the
major issues is related to coding and system compatibility because software used in one system may not
be applicable in another platform. This can be solved either by standardization or by creating plug-in
programs. However, creating plug-ins for each program could be both challenging as well as relatively
time consuming. This raises the importance of standardization of data formats. Standardized data formats
can be used to store useful system O & M information as well. As a result, the design team can use the
information created to identify the critical system areas that need further improvement. Another,
important issue with data exchange models is storage of confidential data. Most companies would not like
their data to be obtained by their competitors, and some of the data exchange models may not provide
substantial protection from information piracy. However, it is useful to enable the sharing of data sources
from various industrial organizations in one place in order to enhance the overall performance and safety
of the marine sector. Consequently, it is vital to find a system that can both save enough data to achieve
the previous goal without compromising confidential product information from different companies.
6.4
Integration with repair
Design processes may specify repair strategies and their integration with the overall lifecycle of the
system. One of the most important areas in repair integration with maintenance is the inventory planning
and organization. A paper by (Lutjen and Karimi, 2012) presented an optimized inventory system
simulation approach for a single-echelon used in offshore wind turbine installations. For this purpose,
they performed a heuristic reactive scheduling for synchronizing the installation vessels in different
weather conditions in supporting the planning of offshore logistics systems. In this case, for logistics in
marine environment, it is important to have a decent data gathering of the environmental conditions and
degradation processes.
In this case, in a study by Lee et al. (2012c), a Multiplicative Decomposition (MD) method was used to
achieve optimal results for a test miner tracked vehicle, considering noise variables in the Korea Deep
Ocean Study (KODOS) area of the north-eastern equatorial Pacific. MD precisely determined the
responses variance caused by deep-sea environmental variables. This information was then used in the
optimization of the test miner tracked vehicle for deep-sea manganese modules collection. A project by
INCASS (2014b) has also been working on creating a overall system for condition monitoring and a
historical data gathering system that connects repairs, maintenance, and operational results to the design
parameters of three different types of merchant vessels (container ships, oil tankers, bulk carriers). This
could identify problematic areas and consequences of these failures to the overall performance of the
vessels based on the particular design parameters.
Outfitting and replacement are other important factors in repair planning in maintenance. A study by
Ruy et al. (2012) confirms that. This study is based on a scenario related to an outfitting division while
not having any direct expertise to oversee the structural panel conditions on the installation of holes for
the outfitting equipment. A challenging part of the overall outfitting equipment installation is related to
the determination of the position and types of holes needed for a particular panel as well as the location of
the interconnected silhouette of panels. As a result, most of these processes require an effective, but time-
504
consuming, communication and discussion between the divisions, specifically hull experts and retrofitting
experts. Therefore, this paper presented an automated hole-plan handling system for decreasing time
wastage. Environmental conditions are also important in organizing a repair schedule, as harsh
environmental conditions can cause difficulties in personnel and inventory transport and outdoor repairs.
This harsh environment repair scheduling has been studied by Berg et al. (2012) where they looked into
subsea repairs in the eastern part of the Barents Sea with harsh environmental conditions and seasonal ice.
They suggested that during the harsh weather, an all-year Construction and Intervention Vessel (CIV)
must be used in conjunction with ice management vessels.
Reliability analysis of components and structures during the operation also provides useful information
regarding the lifecycle analysis of marine structures for the design team. For example, Wilkman et al.
(2012) introduced structural analysis measurements of Aker Arctic. Data was gathered from 95 different
vessels that were tested on 140 occasions. More specialized structural and data analysis tools were also
developed in industry such as a simplified fatigue assessment rooted in beam theory with a spectral-based
fatigue analysis procedure in MAESTRO introduced by Hunter et al. (2013). This method can generally
screen for fatigue damage on vessels hull. The MAESTRO software presented in this paper is open
framework, and it helps by implementing a fully functional lifecycle framework for maintenance,
monitoring, and reliability of ship structures.
A study by Dlugokecki and Hepinstall (2014) focuses on the Design for Production (DFP) concept and
its integration with other concepts such as Design for Maintenance. This research emphasizes the fact that
the design stage maintainibility of the vessel and ease of repair should be considered in order to decrease
the operational costs of the ships lifecycle. This should also include lowering the number of components
and minimizing akward repair areas.
Overall lifecycle structural management can be integrated with repair in order to provide advantageous
information on most common failure areas and their consequences to the design team. Additionally, it can
also represent further information on any difficulties faced by the ship on-board crew during repair jobs.
This information can be in the form of service cost, overhaul time, and unwanted failures caused by the
repair process. Thus, the design team can use the information for suggesting additional measures in terms
of maintainability aspects and timely decision making in order to overcome these issues. However, the
type of information and data to be stored regarding repairs need to be further scrutinized. This would also
require further research on tools and methodologies in order to connect the marine operational team with
the design team. In some cases, this connection is rather difficult as challenges may arise, such as in the
case of sharing information between the marine lifecycle management team and the design team.
6.5
Integration with structural health monitoring systems
Condition and health monitoring of the structural components is one the most important aspects of
evaluating the lifecycle of a vessel. An initial study presenting the integration of a structural health
monitoring with maintenance was performed by Ferrese et al. (2011). In this study, a particle swarm
optimization algorithm was applied to obtain an ideal control for a desired eigenstructure. Nonlinear
power system modeling has been used to establish an algorithm to be highly effective in the maintenance
of the system output related to the specified eigenstructure, which provides the importance of structural
integrity in planning. As a result, Caldwell (2012) illustrated the hull integrity management of Floating
Offshore Installation (FOI) as well as techniques used to monitor this structural aspect regarding class
requirements for the hull structure. One of the major methods of hull inspection methodology for vessels
is dry-docking, but FPSOs cannot be dry-docked. Therefore, structural integrity management has been
chosen to develop innovative techniques to support risk-based inspection of FPSOs without the need for
the time-consuming dry-docking. Complimentary to the above, a study by Kvarme et al. (2012)
demonstrated the structural integrity assessment being implemented for investigating the integrity of
pipelines based on information from external structural inspections.
McCarthy and Buttle (2012) discussed failures in the tensile armor wires of flexible risers. This was
detected through the reduction in applied stress in failed wires. A magnetic system was also introduced
for measuring the stress in all the tensile armor wires for determining failures due to cyclic fatigue. A ship
fatigue routing procedure was also investigated in a report by Mao et al. (2012) via a simple spectral
method in long-term fatigue assessment of container vessels. The advantage of applying this method has
been investigated on a container ship, which was proven that the fatigue life can be increased by at least
50% using an optimal ship routing system.
Stress/strain and ultimate strength are other factors that must be considered for structural analysis of
the maintenance planning and lifecycle management. Subsequently, a paper by DuToit et al. (2012)
introduces an advanced method of monitoring the stress/strain of power umbilicals. Fiber optic strain and
temperature sensors were installed for this study. Distributed temperature and strain measurements were
examined for manufacturing, verification, and continuous integrity monitoring. Ultimate longitudinal
505
strength of hull girder was also investigated in a study by Nam and Choung (2012), which delivered an
estimation of a Very Large Crude Carriers (VLCC) probabilistic damage, magnified due to collision and
grounding accidents based on Goal Based Standards (GBS) of International Maritime Organization
(2003). The extent of the damage, expressed as a function of the ultimate strength, has been calculated
using probability density functions with non-dimensional damage variables via the in-house software
Ultimate Moment Analysis of Damaged Ships (UMADS), which uses the progressive collapse method.
Longitudinal bending of the progressive collapse behavior of the ships hull girder was considered by
Alie and Fujikubo (2012) who used an incremental formula based on Smiths method. Then, a case study
was performed using this method on residual strength analysis of bulk carriers and a tanker with collision
damages. Focus area of analysis has been on the effect of the rotation of neutral axial on the residual hull
girder strength in the case of biaxial bending. INCASS (2014a) has created an innovative method of using
strain gauges to obtain periodic loading conditions on structural elements of a vessel. These data are fed
into an innovative inverse Finite Element Analysis (iFEM) where hydrodynamic and overall structural
reliability of the system are modeled. Subsequently, these results are used for validation of the
hydrodynamic calculations, reliability analysis, and FEM of the initial design predictions.
Another important area of stress/strain analysis of ship hull structure is the strength of welds. This has
been highlighted in a paper by Piovesan et al. (2012), which discusses, compares, and optimizes the
plastic constraint loss between Single Edge Notch Bending (SENB) and Single Edge Notch Tension
(SENT) specimens for base plate and weld joints. This required a correction factor generated using
corresponding Crack Tip Opening Displacement (CTOD) ratio by the critical Weibull stress criterion.
This was also defined by the internal stress distribution in plate thickness, considering critical fracture
volume along the crack tip.
Corrosion is one of the most vital phenomena that have been taken into account in structural analysis in
the marine environment. Thus, Htun et al. (2013) researched random field models for demonstration of
corroded surfaces. The surface geometry of corroded plates has been considered using an innovative
random field model called the Kerhunen-Loeve Expansion method. This is an alternative methodology to
more common uniform corrosion models. Another study on corrosion was based on the impact of
bacterial activity (Comanescu et al., 2012). This study was performed in the pipeline systems of the oil
and gas transport systems. The corrosion lifecycle has been modeled as a function of Microbiologically
Influenced Corrosion (MIC) risk with influence from water quality and bacterial population and types.
As mentioned previously, water quality is a critical factor in corrosion analysis. As a result, Gentile
et al. (2012) studied water quality and material property effects of the Baltic environment on the quality
of the steel material based on the ISO15156 Ed. Concentration of H2S and pH have been used for the
safety criteria of the Baltic environment. In general, both corrosion and fatigue can cause crack growth in
hull structure. Therefore, Volling et al. (2012) have debated about ductile fracture arrest in line pipes.
Subsequently, they have created an innovative approach on crack-arrest prediction using both analytical
and numerical approaches. For the analytical approach, energy based criteria were used to capture
pressure projections on material toughness values. In contrast, the numerical approach was used to
demonstrate material resistance by energy based cohesive zone model on material damage created via
Finite Element Analysis (FEM).
Lifecycle analysis is another criteria on ship structural monitoring and management which was
evaluated on a study by Ohba et al. (2013). This study considers sustainability assessment vessel structure
on risk of accidents based on environmental, economical, and societal factors. As a result, lifecycle
structural optimization of the midship section of a double hull tanker was carried out in this work. Five
optimization problems were evaluated: minimization of construction cost, optimization of Lifecycle
Benefit (LCB) based on the oil outflow hazard, maximization of LCB considering the CO2 emissions
hazard, expansion of LCB considering the risk of failure, and optimization of LCB considering all the risk
factors called the holistic risk.
The final important subjects to be considered for this study were the structural monitoring and testing
methodologies and tools. One example of these type of tools is illustrated by Grasso et al. (2011). The
paper presents the design of two hull monitoring system prototypes using temperature compensated laser
based optical sensors to be installed on a double sided bulk carrier and an ice class tug boat. This required
finite element analysis on defining the optimum sensors installation areas on vessels hull. Another
monitoring tool, developed by Todorov et al. (2012), was called the Automated Ultrasonic Testing (AUT)
data processing technique. This uses computer simulation for enhancing defect detection and mitigating
the inspection actions. Probability of Detection (POD) and accuracy of defect sizing was determined for
six illustrative AUT systems with demonstrative girth weld embedded defects. Quantification
methodology guidance was built, and measures to improve the performance AUT was then assessed by a
built-in quantification methodology.
506
In summary, this section provides extensive background research on different structural health
monitoring and analysis systems that can assist in optimizing structural integrity. The above review
includes various fatigue and stress analysis models developed in different sectors while also referring to
different structural monitoring and detection systems. Moreover it highlights the importance of
integrating FEM results with lifecycle analysis for the design team. However, a study on the usefulness of
different monitoring and detection equipment on FEM results of the marine structures would be beneficial
to include in future work. Another interesting area that has been highlighted in this section relates to the
analysis of collision damage in the structural integrity of vessels, although further work is needed in order
to suggest additional recommendations to the design team on minimizing structural damage and
optimizing structural integrity related to collision accidents. Finally, this section examines studies on the
effects of corrosion on structural integrity. Ship design needs to further consider mitigation measures
including coatings; therefore, a study on the behavior of coatings, paints, and anti-fouling systems could
prove crucial.
6.6
Summary of the lifecycle structural management systems
In summary, this chapter has looked into the relationship between LCM tools, data interchange and
standards, integration of structural health monitoring systems, structural repairs, and analysis with the
design process in marine structures. It has demonstrated that valuable information obtained from
maintenance, repairs, and structural analysis can be used to identify weak points in designs and
furthermore, suggest additional improvements. It is useful to consider that the design team needs to
further incorporate the ease of repairs and maintenance in ship design in order to increase maintainability
and address extensive operational expenses. In brief, it is vital to have an integrated lifecycle system that
connects the overall design process with other operational parameters such as repairs and maintenance.
However, there are challenges that need to be considered in the development of lifecycle data
management for ship design. One of the major tasks is the application of such methodology on a full
ship/marine environment. As no full-scale trial of such a system has been made yet, an accurate
assessment of the increased performance and benefits for the vessels structure of this methodology is not
currently possible. Additionally, the studies that consider areas of LCM do not incorporate all its phases
including the combination of novel research with industrial applications for design, survey, inspection
scheduling, maintenance, and repairs. Another issue is the defragmentation of different, existing, and
upcoming lifecycle management platforms, databases, and tools inside a common framework and
working platform. This could result in incompatibility between different platforms, which may cause
difficulties in sharing information.
Moreover, potential gaps have been identified in terms of interconnectivity of tools that may assist in
the prioritization of maintenance and repairs such as the combination of DFTA and Dynamic Bayesian
Network (DBN). The above may consider the time-dependent degradation of structural elements which
can be modeled accordingly. Additional consideration regarding the standardization of DSS would be
also beneficial. Moreover, most maintenance methodologies either consider operational or business
aspects with no integration between two major features of achieving effective maintenance regime.
No data storage security standard is fully addressed, especially when considering the confidential
manner and sensitivity of shipping data. Accordingly, a central database that can be used across the
marine industry by operators and other stakeholders would be essential in order to enhance performance,
safety, and environmental credentials of the sector. However, this cannot be achieved without a safe and
confidential information sharing protocol. Moreover, the type of the data to be saved is important such as
maintenance intervals, repair reports, and structural condition monitoring data. This would require more
comparative research on different structural monitoring and analysis tools. Finally, a comprehensive
study on using more automated systems could be beneficial in reducing the overall cost and accelerate the
overall analysis process in structural monitoring systems.
7.
OBSTACLES, CHALLENGES, AND FUTURE DEVELOPMENTS
Throughout this report, several recurring themes emerge that contain either foci of ongoing research or a
notable gap between the industrys desired capability and the current state of the art. Several of these
themes have elements present in both the research foci and the gap categories, indicating areas worthy of
further analysis and discussion to identify higher-level obstacles, challenges, or future development
needs. Four main themes are identified here to structure this discussion: increased lifecycle considerations
in structural design, the continued focus on risk- and goal-based design approaches, the rapid growth in
numerical models available to the designer, and the increasingly-important role played by classification
society software.
Perhaps one of the largest themes revealed by this committees work is that of increased consideration
of the structures entire lifecycle. At the design stage, interest in better understanding the overall lifecycle
507
has resulted in new approaches to directly simulate or optimize repair and maintenance strategies during
initial structural design. Such approaches range from formal optimization using detailed through-life
simulations to more higher-level commentary on the trade space between design-for-production and
design-for-maintenance under the aegis of design-for-X. Additionally, scenario-based design methods,
such as epoch-era analysis, are growing in popularity for the general marine platform design problem.
Such specific scenario-based analysis further opens the door to consider specific repair and maintenance
impacts on a vessels earning potential and safety through-life. At the current time, most of the
approaches reviewed were in the early stages of adoption and development, with the origin of the
developments resting primarily in academic-only or academic-industry joint projects. Also currently,
there appears to be a gap between the theory and the practical state of the art available for vessel
designers. For specific analysis types, such as LCA and LCC, industry-standard frameworks and common
databases required for practical analysis on individual projects do not yet exist. The transition of these
approaches to applied state of the art remains anticipated, not observed. However, given the success in
using direct lifecycle simulation of hydrodynamic responses for operability, fuel economy, and comfort in
the last decade, perhaps this transition is not far off for the structural domain.
The theme of increased consideration of the structures entire lifecycle is present far beyond design
stage activities. One manifestation is in the growing demand for through-life support of marine structural
systems. There is long-standing interest in getting design-stage data readily available to support on-board
inspection and updating, technologically, web-enabled storage, and portable viewing of this design data
appears feasible today based on the references presented in this report. The committees work revealed
that almost all major design software vendors now offer in their design suites some sort of PLM add-on
that are configured to solve this problem. Likewise, all major classification societies are touting their
through-life management software along with their initial design approval software. However, what is
largely missing is an agreed intellectual property protection framework for the data within these systems.
Shipyards are rightly concerned about exposing detailed production information to all downstream users.
However, without an accurate as-built description of the structure, many through-life monitoring and
updating techniques become impractical. Continued interest in the use this type of data for emergency
responses after a collision, fire, or grounding is further driving efforts to arrive at a workable solution to
this IP problem.
An area where technological challenges still remain is through-life structural health monitoring. Unlike
rotating machinery, where clear systems have been developed to interpret real-time monitoring data, the
structure community struggles with the data-to-decision problem. Current systems are able to record
accelerations, strains, and displacements throughout the structure. Translating this data into improved
operations and repair decisions remains difficult. Like many other fields, this represents a big data
challenge. This is a field where several new systems were proposed during the time period of this
committees mandate, and future development in this area is anticipated.
The development of both risk-based design and goal-based standards has been chronicled by the ISSC
over the past two decades, and it represents the second major theme for this chapter. These topics
continue to generate new publications and approaches during the current ISSC period. While early efforts
mainly focused on setting up the frameworks for these systems, the current publications are split between
larger philosophical issues and the details of attempting to implement practical risk-based or goal-based
design approaches. Overall, many rule sets are now moving toward a probabilistic foundation, and the
explicit inclusion of uncertainty in design approaches is now commonplace. For example, as discussed in
Chapter 4 of this report, risk and reliability measures are now often built into complex optimization and
decision support systems. Setting safety as a design objective has been widely discussed in the past
though only a few examples of multi-objective optimization frameworks for structural safety have been
presented in the past. During this committees mandate several papers have now shown Pareto fronts
between safety and cost or weight of the resulting system. This indicates that such trade-space data can be
developed in a practical manner today. However, the limits of these approaches are still not clear. While
they have their origin in safety-related concerns, is it appropriate to establish risk and goals for other
aspects of designs such as profitability, maintainability, or producibility? Example and complete systems
to implement these approaches in ship and platform design are currently sparse; much like LCC and LCA
models, the theory has been well established, but ready-made tools for practicing designers are still
lacking.
The continued rapid growth in the types of numerical modeling required in new design projects is the
third theme for this chapter. The types of models possible and the level of detail continue to grow, with
new modeling frameworks proposed almost yearly. Selecting the correct level of detail to model the
various aspects of each design is now an essential step in setting up a process to develop a new design.
Within this committees report, references note that the continued expansion of such modeling might be
reaching a point of diminishing or negative returns in terms of making better design decisions.
508
Additionally, new tools were reported whose primary purpose was to implement simpler versions of
complex calculations that were already available in their full form. The idea of these programs is that the
modeling would be faster and less error-prone if the user had a more limited set of modeling options. The
question of selecting the correct level of modeling detail for each design does not seem to have been
formally addressed to date in the literature, nor has a systematic approach to validating such models been
widely discussed. This is especially true if the number of structures designed to date by such novel
models is limited. This problem is one of many hierarchal levels, as once a modeling approach is selected
(e.g. FEA), the level of detail of that particular approach is normally adjustable by the designer.
A key component of managing the growth in numerical models is in efficient transfer of data among
different modeling programs, or among a central design repository and each modeling program. Such data
transfer has been a common theme of this committees reports for many years, and continued progress
can be seen in this area as well. While no major standard developments (e.g. STEP) were reported during
the current committees mandate, continued refinement of data exchange and numerous ad-hoc exchange
approaches relying on application API and common data definition languages (e.g. XML) were reported.
Given the rapid growth in the types of models available to the designer, the development of a universal
file format or format converter is likely entirely improbable at this point in time. Thus, efficient one-off
integration may become a topic of interest for integrating such toolsets.
The growth in modeling complexity and the sheer number of models have led to increased computational
demands to complete design simulations. This is especially true when such models are tied into optimization
frameworks. This committee performed a first review of parallel processing options for such frameworks
and found wide availability and adoption of cluster-style computing today. Coupled with the growth of
population-based optimization approaches, such computing resources now allow the designer to perform
wide-ranging trade studies in a time efficient manner. There is growing interest in the use of GPUs to further
accelerate such models; however, developments in this area are much more limited to date. Whether GPUs
will be able to overcome the economies of scale now present in cluster-style computing is an interesting
question for the next ISSC. The use of surrogate models to further reduce the computational burden of
detailed numerical models was first discussed by this committee in 2012. In this committees report, a more
detailed overview was presented, along with some initial recommendations for modeling type. However,
wide-scale application of surrogates in industrial optimization seems to be lacking at the moment, as the
cited papers were almost universally academic in origin. At the current time, there appears to be little
preventing the adoption of such models by industry, with extensive open-source toolkits available in popular
languages such as MATLAB. If wider adoption of these models is not seen in the future, it may indicate that
the time and uncertainty associated with building such a model does not make it cost competitive with the
falling cost of direct simulation via clustered computational resources.
Finally, the growth in complexity of classification society software is the last major theme for this
chapter. The updated review of the classification society software revealed in detail how complex these
applications have become. Almost all societies now include links to advanced simulation such as spectral
fatigue and through-life data access from their systems. Classification societies are well-positioned to
support through-life data access, and as a disinterested third party, these may have a role to play in
resolving the IP issues surrounding through-life data access. However, given the focused role of
classification in safety certification, other elements, such as optimization for production cost or
production detailing, are absent from such systems and unlikely to be added. This requires detailed design
information to exist in (at least) two systems simultaneously. Thus, one key area where smooth data
transfer is becoming increasingly important is between classification society software packages and
external design-focused CAD packages. While the review of classification society offerings revealed that
data exchange is important to all societies, this is an area where the industry would appreciate smoother
data exchange. As class society software continues to grow in both complexity and importance to the
structural design process, the role this software plays and the links between this software and the rest of
the design environment is expected to become increasingly important.
8.
CONCLUSION
Reviewing and analyzing the last three years of literature have revealed several constant themes in design
methods. As discussed in Chapter Two, there is continued interest in risk-based design and uncertainty
approaches in design. Additionally, the inclusion of various lifecycle events, from conventional lifecycle
costing to more advanced, scenario-based design methods, remain a topic of interest to the marine
structures community. However, the marine structures community is still not at the point where a single
methodology dominates any of these discussions. Most of these trends are equally reflected in the design
509
tools reviewed in Chapter Three. Again, a strong interest in higher fidelity and through-life models was
observed, alongside the normal progress made in numerical methods and design software. Additionally,
the problems of data exchange and lifecycle data management remain at the forefront in this community.
The issues to be addressed continue to extend beyond technical data interface challenges, with unresolved
IP issues and access controls high on the list of challenging problems that remain.
Optimization continues to attract significant interest. Indeed, the synthesis process required for a
complex marine structural design is complicated enough that assistance from a computer is widely
viewed as beneficial. Chapter Four reviewed numerical methods used for both optimization and
enhancing optimization via techniques like surrogate modeling and GPU programming. However, that is
potentially not the most exciting of the developments in this field. Instead, the ongoing work on
transforming optimization into a design support methodology instead of an answer-giving black box is
very notable. This marks the second Committee IV.2 report in a row in which extensive discussion of
design frameworks and procedures for optimization has played a significant part in the discussion of
optimization. Continued work on integrating optimization into the overall structural design methodology
is expected.
Classification societies have unique roles in influencing the initial design of ship structures and
managing them through-life. As tools influence the design process and methodology, this report reviewed
classification society tools for structural design, analysis, approval, and through-life support. Comparing
against an initial review from the year 2000, Chapter Five revealed not only the expected increase in
modeling fidelity and automation but also detailed extensions into through-life data management by
almost all classification societies. However, it is equally clear that much work remains to be done in
ensuring smooth data transfer between such proprietary tools, other design tools, and data management
systems.
Continuing on from the large LCM review in the preceding ISSC, this committee examined lifecycle
modeling and data management in Chapter Six. It is clear that rapid progress is occurring in this area,
with new tools appearing regularly and increased research intensity on the topic. This is especially true
for design-stage prediction tools as well as tools that can integrate data from structural health monitoring
systems. Equally exciting is the increased emphasis on integration of such approaches into decision
making and repair. LCM is expected to receive intense research attention in the future. Finally, the
Committee was able to extract some key themes and current gaps from the topics reviewed and present
them in Chapter Seven. This review was highly encouraging. While it is clear that new tools and
approaches are being developed constantly, it is equally clear that there is room for future generations of
researchers to improve structural design methods to better suit the developing needs of industry.
ACKNOWLEDGMENTS
The Committee would like to acknowledge the work of Mr. Kazim Tosayev and Ms. Teresa LaForest,
students at the University of Michigan, in coordinating references and providing grammatical editing of
the report.
REFERENCES
2012. Proceedings of the 18th International Ship and Offshore Structures Congress. In: Fricke, W. & Bronsart,
R., eds. 18th International Ship and Offshore Structures Congress, 0913 September 2012 2012 Rostock,
Germany. Hamburg: Schiffbautechnische Gesellschaft e.V., 559.
Alie, M. Z. M. & Fujikubo, M. Residual Longitudinal Strength of a VLCC Considering Probabilistic Damage
Extents. 22nd International Offshore and Polar Engineering Conference (ISOPE), 2012/06/17/22 2012.
Rhodes, Greece, 831838.
Alonso, V., Gonzalez, C. & Perez, R. Efficient Use of 3D Tools at Early Design Stages. Computer and IT
Applications in the Maritime Industries (COMPIT), 2013/04/15/17 2013. Cortona, Italy, 195207.
Anderson, K., Boulares, J. & Melsom, S. ABS Achieving Service Life Program and Navy Life Cycle
Maintenance. ASNE Day, 2013/02// 2013. Crystal City, VA, USA.
Andrews, D., Papanikolaou, A. & Singer, D. 2012a. Design for X. IMDC12The 11th International Marine
Design Conference. Glasgow, Scotland.
Andrews, D., Percival, V. & Pawling, R. 2012b. Just how valid is the ship design spiral given the existence of
cliffs and plateaux? IMDC12The 11th International Marine Design Conference. Glasgow, Scotland.
Andrews, D. J. 2003a. A Creative Approach to Ship Architecture. International Journal of Maritime
Engineering.
Andrews, D. J. Marine DesignRequirement Elucidation rather than Requirement Engineering. IMDC, May
2003 2003b Athens.
Apostolidis, A., Kokarakis, J. & Merikas, A. 2012. Modeling the Dry-Docking CostThe Case of Tankers.
Journal of Ship Production & Design, 28, 134143.
510
Asmara, A., Coenen, J., Van Loenen, M. & Van De Berge, E. Development, Implementation and Testing of a
3D Information Viewer in Shipbuilding. International Marine Design Conference, 2012/06/11/14 2012.
Glasgow, Scotland, 437448.
Aspen, D. M., Fet, A. M. & Johansen, B. A. Life Cycle System Analysis Methods in Ship DesignA Review of
Current Status, Opportunities and Challenges. International Marine Design Conference, 2012/06/11/14 2012.
Glasgow, Scotland, 3948.
Balland, O., Erikstad, S. O., Fagerholt, K. & Wallace, S. 2013. Planning vessels air emission regulations
compliance under uncertainty. Journal of Marine Science and Technology, DOI 10.1007/s007730130212
7.
Bayatfar, A., Amrane, A. & Rigo, P. 2013. Towards a Ship Structural Optimization Methodology at Early
Design Stage. International Journal of Engineering Research and Development, 9, 7690.
Beesemyer, J. C., Fulcoly, D. O., Ross, A. M. & Rhodes, D. H. 2011. Developing Methods to Design for
Evolvability: Research Approach abd Preliminary Design Principles. 9th Conference on Systems
Engineering Research. Los Angeles, CA.
Berg, T. E., Berg, B. O., Borgen, H. & Hannnen, S. Study of Operational Characteristics for an All Year
Intervention Vessel for the Barents Sea. 22nd International Offshore and Polar Engineering Conference
(ISOPE), 2012/06/17/22 2012. Rhodes, Greece, 13801385.
Bergstrm, M., Ehlers, S., Erikstad, S. O., Erceg, S. & Bambulyak, A. 2014. Development of an Approach
Towards Missio-Based Design of Arctic Marine Transport Systems. OMAE 2014. San Francisco.
Bertram, V. & Peric, M. Advanced Simulations for Offshore Industry Applications. Computer and IT
Applications in the Maritime Industries (COMPIT), 2013/04/15/17 2013. Cortona, Italy, 720.
Bharadwaj, U. & Wintle, J. 2011. Risk-Based Optimization of Inspection Planning in Ships. Journal of Ship
Production & Design, 27, 111117.
Bitner-Gregersen, E. M. & Toffoli, A. 2014. Occurrence of rogue sea states and consequences for marine
structures. Ocean Dynamics.
Black, F. & Scholes, M. 1973. The Pricing of Options and Corporate Liabilities. Journal of Political Economy,
81, 637654.
Boulougouris, E. & Papanikolau, A. 2013. Risk-Based Design of Naval Combatants. Ocean Engineering, 65,
4961.
Bresler, S., Strucko, A. & Sicking, P. Next-generation design paradigm for shipbuilders. 2013 2013. Royal
Institution of Naval Architects, 97108.
Brown, J. I. The Implementation Of The R&M/Dependability Case For Naval Platforms. Managing Reliability
and Maintainability in the Maritime Industry, 2012/01/25/26 2012. London, UK: Royal Institute of Naval
Architecture, 8894.
Cadmatic, N. 2012. NUPAS-Cadmatic News 2012. NUPAS-Cadmatic.
Cai, B., Liu, Y., Liu, Z., Tian, X., Ji, R. & Li, H. 2011. Reliability-Based Load and Resistance Factor Design of
Composite Pressure Vessel under External Hydrostatic Pressure. Composite Structure, 93, 28442852.
Caldwell, R. Hull Inspection Techniques & Strategy. Managing Reliability and Maintainability in the Maritime
Industry, 2012/01/25/26 2012. London, UK: Royal Institute of Naval Architecture, 3945.
Caprace, J., Da Silva, C., Rigo, P., & Pires. 2011. Discrete Event Production Simulation & Optimisation of Ship
Block Erection Process. COMPIT. Berlin, Germany.
Chaves, V. T., Tancredi, T. P. & De Andrade, B. L. R. 2014. Optimal structural design of small ships with
response surface. Developments in Maritime Transportation and Exploitation of Sea Resources. Guedes
Soares & Lpez Pea (eds) ed.: 2014 Taylor & Francis Group, 297307.
Chen, J., Tang, Y. & Huang, X. 2013a. Application of surrogate based particle swarm optimization to the
reliability-based robust design of composite pressure vessels. Acta Mechanica Solida Sinica, 26, 480490.
Chen, N.-Z., Wang, G. & Bond, J. 2013b. Hull girder reliability assessment of FPSO. In: Soares, G. &
Romanoff, J. (eds.) Analysis and Design of Marine Structures. London: Taylor & Francis Group, 535543.
Chen, T., Zheng, J. & Bae, D. M. 2012. A rapid estimation method of structural fatigue analysis for a 17k ton
DWT oil tanker. Journal of Marine Science and Application, 11 (2), 184190.
Collette, M. 2011. Hull Structures as a System: Supporting Lifecycle Analysis. Naval Engineers Journal, 123,
4555.
Collette, M. 2014. Effective optimization: How numerical optimization tools enable improved design
performance. Marine Technology, 2023.
Comanescu, I., Melchers, R. E. & Taxen, C. Life Time Assessment of Offshore Water Injection Pipelines as a
Function of Microbiologically Influenced Corrosion. 22nd International Offshore and Polar Engineering
Conference (ISOPE), 2012/06/17/22 2012. Rhodes, Greece, 252258.
Coraddu, A., Figari, M., Savio, S., Villa, D. & Orlandi, A. Integration of seakeeping and powering
computational techniques with meteo-marine forecasting data for in-service ship energy assessment.
511
Developments in Maritime Transportation and Exploitation of Sea ResourcesProceedings of IMAM 2013,

15th International Congress of the International Maritime Association of the Mediterranean, 2014. 93101.
Coyne, R. D., Rosenman, M. A., Radford, A. D., Balachandran, M. & Gero, J. S. 1990. Knowledge-Based
Design Systems, Reading, Massachusetts, Addison-Wesley.
Cui, H., Turan, O. & Sayer, P. 2012. Learning-based ship design optimization approach. Computer-Aided
Design, 44, 186195.
Cui, J.-J. & Wang, D.-Y. 2013. Application of knowledge-based engineering in ship structural design and
optimization. Ocean Engineering, 72, 124139.
Danese, N. & Aasen, R. Exploiting Weight Data to Support Engineering and Corporate Decision-Making
Processes. Computer and IT Applications in the Maritime Industries (COMPIT), 2013/04/15/17 2013.
Cortona, Italy, 334347.
Daniels, A. S., Parker, M. C. & Singer, D. J. 2011. Intelligent Ship Arrangements (ISA) Passage Variable
Lattice Network Templating Application. ICCAS10th International Conference on Computer and IT
Applications in the Maritime Industries. Berlin.
De Jonge, T. & Kramer, H. Integrating CAD/CAM in a PDM/ERP Environment. COMPIT 2014, 2014/05//
2014. Redwoth Hall, U.K., 158166.
De Neufville, R. & Scholtes, S. 2011. Flexibility in Engineering Design, MIT Press.
Devine, T. E. & Collette, M. Application of the Bayesian Optimization Algorithm to Naval Structural Design.
12th International Symposium Practical Design of Ships and Other Floating Structures, 2013/10/20/25 2013.
CECO, Changwon, Korea, pp. 694701.
Dlugokecki, V. & Hepinstall, L. 2014. Optimizing design processes to drive down total ownership cost.
Manitenance Technology, 3037.
Dong, F., Deglise-Hawkinson, J., Van Oyen, M. P. & Singer, D. J. 2013. Analytical Approach to a Two-stage
Queuing Network for the Planning of Outfitting Processes in Shipbuilding. Journal of Ship Production and
Design, 29, 136141.
Dong, P., Nie, C., Yang, X., Song, S. & Decan, L. 2012. A Math-Based Design-for-Producibility Evaluation of
Titanium Applications in Ship Hull Structures. SNAME Transactions, 120, 299305.
Dutoit, D., Nikles, M., Rochat, E., Willemoes, P. & Little, H. Distributed Fiber Optic Strain and Temperature
Sensor for Subsea Umbilical. 22nd International Offshore and Polar Engineering Conference (ISOPE),
2012/06/17/22 2012. Rhodes, Greece, 349355.
Ehlers, S. & Kujala, P. Cost optimization for ice-loaded structures, Analysis and Design of Marine Structures.
MARSTRUCT 2013, 2013/03/25/27 2013. Espoo, Finland, 111118.
Electric Boat, C. 2013. Prototype Implementation of a Visualization System for 3D Models of CPC Parts.
National Shipbuilding Research Program. Electric Boat.
Elsen, H. The Spectral Method Re-Engineered: High-Performance Finite-Element- Based Fatigue Assessment
Processes for Ship Structures. Computer and IT Applications in the Maritime Industries (COMPIT),
2013/04/15/17 2013. Cortona, Italy, 413424.
Embrey, D. How to Reduce Maintenance Error And Maximize Reliability In Maintenance Operations.
Managing Reliability and Maintainability in the Maritime Industry, 2012/01/25/26 2012. London, UK: Royal
Institute of Naval Architecture, 19.
Erikstad, S. O. & Ehlers, S. 2014. Simulation-Based Analysis of Arctic LNG Transport Capacity, Cost and
System Integrity. OMAE 2014. San Francisco.
Erikstad, S. O. & Levander, K. 2012. System Based Design of Offshore Support Vessels. IMDC12The 11th
International Marine Design Conference. Glasgow, Scotland.
Evans, J. H. 1959. Basic design concepts. ASNE Journal.
Fagerholt, K., Hvattum, L. M., Esbensen, E. F. & Nygreen, B. 2010. Using Decision Trees for a Stochastic
Maritime Routing Problem. EURO XXIV. Lisbon, Portugal.
Ferrese, F., Dong, Q., Nataraj, N. & Biswas, S. 2011. Optimal Feedback Control of Power Systems Using
Eigenstructure Assignment and Particle Swarm Optimization. Naval Engineers Journal, 123, 6776.
Fet, A. M., Aspen, D. M. & Ellingsen, H. 2013. Systems Engineering as a Holistic Approach to Life Cycle
Designs. Ocean Engineering, 62, 19.
Forbes, J. D. M., Dick, C. A. & Kelly, J. A. How Does The First Sea Lord Ensure The Royal Navy Is Safe?
Ensuring Safety In The Context Of Military Capability. Managing Reliability and Maintainability in the
Maritime Industry, 2012/01/25/26 2012. London, UK: Royal Institute of Naval Architecture, 2331.
Frangopol, D. M., Bocchini, P., Dec, A., Kim, S., Kwon, K., Okasha, N. M., Saydam, D. & Salvino, L. W.
Life-Cycle Ship Reliability Assessment, Damage Detection, and Optimization. 11th International
Conference on Fast Sea Transportation FAST 2011, September 2011 2011 Honolulu, Hawaii, USA. 555
560.
Freimuth, J. & Ma, M. Oil Tanker Cargo Hold Structural Optimization Using Ultimate Limit States. SNAME2014, 2014 Houston, USA.
512
Fu, S.-Y., Huang, H.-Y. & Lin, Z.-X. 2012. Collaborative Optimization of Container Ship on Static and
Dynamic Responses. Procedia Engineering, Volume 31, 613621.
Gaspar, H., Ross, A. M. & Erikstad, S. O. 2012. Handling temporal complexity in the design of non-transport
ships using epoch-era analysis. International Journal for Maritime Engineering (RINA Transactions Part A),
(AP).
Gazis, N. A Probabilistic Approach for Reliability Assessment and Fatigue Analysis of Subsea Free Spanning
Pipelines. 22nd International Offshore and Polar Engineering Conference (ISOPE), 2012/06/17/22 2012.
Gentile, M., Fehervari, M., Ragazzoni, S., Cattalini, M., Tomaselli, L. & Gjedrem, T. Nord Stream Project
Baltic Sea Environment: Hydrogen Damage Assessment for Line Pipe Steel and Anode Material Selection.
22nd International Offshore and Polar Engineering Conference (ISOPE), 2012/06/17/22 2012. Rhodes,
Greece, 4353.
Georgiev, P. 2011. Safety analyses for bulk carriers using metamodels of still water loads. In: Soares, G. &
Fricke (eds.) Advances in Marine Structures. London: Taylor & Francis Group.
Gillespie, J. W., Daniels, A. S. & Singer, D. J. Approaching Ship Arrangements from a Nonspatial Point of
View Using Network Theory. International Marine Design Conference, 2012/06/11/14 2012. Glasgow,
Scotland, 355364.
Goldghate, S. & Yamaguchi, S. A Study on IT Solution for Ship Design Based on Virtual Ship Concept. 22nd
International Offshore and Polar Engineering Conference (ISOPE), 2012/06/17/22 2012. Rhodes, Greece,
944951.
Goni, A. R. & Jambrina, L. F. A CAD Development Strategy for the Next Years. Computer and IT Applications
in the Maritime Industries (COMPIT), 2013/04/15/17 2013. Cortona, Italy, 143156.
Grasso, A., Vergine, A., Dimou, D., Samuelides, M., Tsouvalis, N. & Ferrari, M. 2011. Sensors location and
data processing algorithms of an optical fibers hull strength monitoring system. In: Soares, G. & Fricke
(eds.) Advances in Marine Structures. London: Taylor & Francis Group.
Gregor, J. A. 2003. Real options for naval ship design and acquisition : a method for valuing flexibility under
uncertainty. MSc, MIT.
Hagen, A. & Grimstad, A. 2010. The Extension of System Boundaries in Ship Design. International Journal of
Maritime Engineering, 152, 1728.
Halata, P. S., Friedewald, A. & Ldding, H. Augmented Reality Supported Information Gathering in One-of-akind Production. COMPIT 2014, 2014/05// 2014. Redwoth Hall, U.K., 489503.
Helle, S., Korhonen, S., Euranto, A., Kaustinen, M., Lahdenoja, O. & Lehtonen, T. Benefits Achieved by
Applying Augmented Reality Technology in Marine Industry. COMPIT 2014, 2014/05// 2014. Redwoth
Hall, U.K., 8697.
Hifi, N. & Barltrop, N. D. P. Improving Techniques For Ship Inspection And Maintenance Planning.
MARSTRUCT 2013, 2013/03/25/27 2013. Espoo, Finland, 4655.
Hifi, N., Xu, L. & Barltrop, N. Risk-Based Inspection Planning and Data Management Tool. International
Marine Design Conference, 2012/06/11/14 2012. Glasgow, Scotland, 501512.
Hirdaris, S. & Cheng, F. 2012. The role of technology in green ship design. IMDC2012. Glasgow, Scotland.
Hegh, M. W. 1998. Options in Shipbuilding Contracts. MSc, MIT.
Hoffmann, P. N., Eide, M. S. & Endresen, . 2012. Effect of proposed CO2 emission reduction scenarios on
capital expenditure. Maritime Policy & Management, 39, 443460.
Htun, M. M. & Kawamura, Y. A Study on the Method to Estimate the Stochastic Property of the Strength of
Corroded Plate by Response Surface Method with Polynomial Chaos Expansion. 12th International
Symposium Practical Design of Ships and Other Floating Structures, 2013/10/20/25 2013. CECO,
Changwon, Korea, pp. 731738.
Htun, M. M., Kawamura, Y. & Ajiki, M. 2013. A study on random field model for representation of corroded
surface. In: Soares, G. & Romanoff, J. (eds.) Analysis and Design of Marine Structures. London: Taylor &
Francis Group, 545553.
Hughes, O., Ma, M. & Paik, J. K. Applications of Vector Evaluated Genetic Algorithm (VEGA) in Ultimate
Limit State Based Ship Structural Design. ASME 2014 33th International Conference on Ocean, Offshore
and Arctic Engineering, 914 June 2014 2014 San Francisco, USA.
Hughes, O. F. 1988. Ship Structural Design: A Rationally-Based, Computer-Aided, Optimization Approach,
Jersey City, Society of Naval Arhitects Marine Engineers (SNAME).
Hunter, S., Freimuth, J. & Danese, N. Utilizing a Robust Fatigue Screening Process for Initial Design and
Throughout the Ship Life-Cycle. Computer and IT Applications in the Maritime Industries (COMPIT),
2013/04/15/17 2013. Cortona, Italy, 466479.
Ignatius, J., Rasanen, J. E., Tervo, K., Stoker, J. J. & Ellis, T. A Comprehensive Performance Management
Solution. Computer and IT Applications in the Maritime Industries (COMPIT), 2013/04/15/17 2013.
513
Incass 2014a. Inspection Capabilities for Enhanced Ship Safety, Deliverable 3.3 Hydrodynamic and structural
analysis. EC FP7 European Project.
Incass 2014b. Inspection Capabilities for Enhanced Ship Safety, Deliverable 5.1 Structural Database. EC FP7
European Project.
Incass 2014c. Inspection Capabilities for Enhancing Ship Safety, Deliverable 1.1 Annual technical report. EC
FP7 European Project.
International Maritime Organization 2003. Revised interim guidelines for the approval of alternative methods of
design and construction of oil tankers under regulation 13F(5) of annex I of MARPOL 73/78. Resolution
MEPC, 110.
Ji, Z.-H. & Wang, D.-Y. Multi-objective Optimization for Ultimate Strength and Stiffness of Ship Hull.
International Offshore and Polar Engineering Anchorage, 2013/07/05/ 2013. Anchorage, Alaska, USA.
Kajiwara, K. H. 2013. Application of Pull Concept-based Lean Production System in the Ship Building
Industry. Journal of Ship Production & Design, 29, 105116.
Kang, H. J., Choi, J., Lee, J.-K. & Yang, Y. A Conceptual Design of Ship Safety Assessment Model for Ship
Systems Safety Assessment. International Marine Design Conference, 2012/06/11/14 2012. Glasgow,
Scotland, 161168.
Kawamura, Y., Ohba, Y. & Kaede, Y. Lifecycle structural optimization of mid-ship of double hull tanker based
on holistic risk evaluation. MARSTRUCT 2013, 2013/03/25/27 2013. Espoo, Finland, 555563.
Keane, R. G. 2012. Reducing Total Ownership Cost: Designing Inside-Out of the Hull. Naval Engineers
Journal, 124, 6780.
Kim, T.-W., Storch, R. L., Hopman, H. & Erikstad, S. O. 2012. Applications in ship and floating structure
design and analysis. CAD Computer Aided Design, 44, 163165.
Kitamura, M., Kunihiro, H., Takezawa, A. & Uedera, T. 2011a. Shape Optimization System of Bottom Structure
of Ship Incorporating Individual Mesh Subdivision and Multi-Point Constraint. International Journal of
Offshore and Polar Engineering, 21, 209215.
Kitamura, M., Uedera, T., Hamada, K. & Takezawa, A. Shape and Size Optimization of the Double Bottom
Structure of Bulk Carrier at the Initial Design Stage with Finite Element Analysis. International Offshore
and Polar Engineering Conference, 2011/06/24/ 2011b. Maui, Hawaii, USA.
Koelman, H. J. A Mid-Term Outlook on Computer Aided Ship Design. Computer and IT Applications in the
Maritime Industries (COMPIT), 2013/04/15/17 2013. Cortona, Italy, 110119.
Kunakote, T. & Bureerat, S. 2013. Surrogate-assisted multi-objective evolutionary algorithms for structural
shape and sizing optimization. Mathematical Problems in Engineering, 2013.
Kvarme, S. O., Nordsve, A. C., Andersson, P. H. & Alsos, P. H. Advanced Structural Integrity Assessment of
Subsea Pipelines in Operation. 22nd International Offshore and Polar Engineering Conference (ISOPE),
2012/06/17/22 2012. Rhodes, Greece, 373380.
Larkins, D., Morais, D. & Waldie, M. Democratization of Virtual Reality in Shipbuilding. 12th International
Conference on Computer Applications and Information Technology in the Maritime Industries, 2013 2013a.
Larkins, D., Waldie, M. & Morais, D. A Scalable Approach to 3D Visualization and Review Throughout the
Shipbuilding Lifecycle. International Conference on Computer Applications in Shipbuilding, 2013 2013b.
Busan, Korea: Royal Institute of Naval Architects.
Lazakis, I., Turan, O. & Judah, S. Establishing The Optimum Vessel Maintenance Approach Based On System
Reliability And Criticality Analysis. Managing Reliability and Maintainability in the Maritime Industry,
2012/01/25/26 2012a. London, UK: Royal Institute of Naval Architecture, 5667.
Lazakis, I., Turan, O. & Olcer, A. Determination of the Optimum Ship Maintenance Strategy Through Multi
Attribute Decision Making. International Marine Design Conference, 2012/06/11/14 2012b. Glasgow,
Scotland, 473487.
Lee, D. K., Shin, J. G., Kim, Y. & Jeong, Y. K. 2014. Simulation-Based Work Plan Verification in Shipyards.
Journal of Ship Production & Design, 30, 4957.
Lee, J.-K., Lee, G. J., Park, B. J. & Kang, H. J. Risk-Based Approaches in Design for Ship Safety. International
Marine Design Conference, 2012/06/11/14 2012a. Glasgow, Scotland, 6372.
Lee, J., Seol, D., Rhee, S. H., Chi, H. R., Jun, D. S. & Ryu, M. C. 2010. Automatic Tool Development for Ship
Hull Form Parametric Study. PRADS2010 - 11th International Symposium on Practical Design of Ships and
Floating Structures. Rio de Janeiro.
Lee, J. M., Lee, K. H., Kim, D. S. & Lee, G. An Enhanced Smart Maintenance of Piping System for the
Offshore Plants. 22nd International Offshore and Polar Engineering Conference (ISOPE), 2012/06/17/22
2012b. Rhodes, Greece, 657660.
Lee, T. H., Cho, S., Lee, M., Lim, W., Park, S., Choi, J. S., Kim, H. W. & Hong, S. Robust Optimization of Test
Miner Considering Deep-sea Environmental Variables. 22nd International Offshore and Polar Engineering
Conference (ISOPE), 2012/06/17/22 2012c. Rhodes, Greece, 122129.
514
Levander, K. 2006. System Based Ship Design. TMR 4110 Marine Design and Engineering, Basic Course,
NTNU. Trondheim.
Linton, E. M. 2011. Coast Guard Reliability Centered Maintenance. Naval Engineers Journal, 123, 2530.
Liu, P.-L. & Jin, J.-H. Ship CAE modeling based on Tribon XML file. 2012 2012. IEEE Computer Society,
425428.
Lodding, H., Friedewald, A., Heining, M. & Schleusener, S. 2011. Virtual Reality Supported Assembly
Planning in the Shipbuilding Industry. Journal of Ship Production & Design, 27, 146152.
Lundquist, E. 2012. Virtual Shipyard: Learn the Yard Before You Get Your Badge. Naval Engineers Journal,
124, 3740.
Lutjen, M. & Karimi, H. R. Approach of a Port Inventory Control System for the Offshore Installation of Wind
Turbines. 22nd International Offshore and Polar Engineering Conference (ISOPE), 2012/06/17/22 2012.
Ma, M., Freimuth, J., Hays, B. & Danese, N. Hull girder cross section structural design using Ultimate Limit
States (ULS) based multi-objective optimization. 13th International Conference on Computer Applications
and Information Technology in the Maritime Industries, 2014a.
Ma, M., Hughes, O. & Mcnatt, T. Ultimate Limit State Based Ship Structural Design Using Multi-Objective
Discrete Particle Swarm Optimization. Proceedings of the ASME 34th International Conference on Ocean,
Offshore and Arctic Engineering, May 2015 2013a St. Johns, Newfoundland, Canada.
Ma, M., Hughes, O. & Paik, J. K. Ultimate strength based stiffened panel design using multi-objective
optimization methods and its applications to ship structures. PRADS 2013, 2013b.
Ma, M., Zhao, C. & Hughes, O. 2014b. A Practical Method to Apply Hull Girder Sectional Loads to Full-Ship
3D Finite Element Models Using Quadratic Programming. Ships and Offshore Structures, 9, 257265.
Malay, P. 2012. Easier access to shipbuilding product data. Naval Architect, 4244.
Mao, W., Ringsberg, J. W. & Rychlik, I. What is the Potential of Using Ship Fatigue Routing in Terms of
Fatigue Life Extension? 22nd International Offshore and Polar Engineering Conference (ISOPE),
2012/06/17/22 2012. Rhodes, Greece, 681687.
Matsuo, K. & Shiraishi, K. Introduction of AR Applications for Shop Floor in Shipbuilding. COMPIT 2014,
2014/05// 2014. Redwoth Hall, U.K., 318326.
Mccarthy, J. C. & Buttle, D. J. MAPS-FR Structural Integrity Monitoring for Flexible Risers. 22nd International
Offshore and Polar Engineering Conference (ISOPE), 2012/06/17/22 2012. Rhodes, Greece, 137144.
Mckenney, T. A., Buckley, M. E. & Singer, D. J. Differentiating Set-Based Design from Other Design Methods
and the Cultural Challenges of Implementation. International Marine Design Conference, 2012/06/11/14
2012. Glasgow, Scotland, 283296.
Mckenney, T. A., Kemink, L. F. & Singer, D. J. 2011. Adapting to Changes in Design Requirements Using SetBased Design. Naval Engineers Journal, 123, 6777.
Mcnatt, T., Ma, M. & Hunter, S. 2013. Historical perspective on the structural design of special ships and the
evolution of structural design methods. Ships and Offshore Structures, 8, 404414.
Mebane, W. L., Carlson, C. M., Dowd, C., Singer, D. J. & Buckley, M. E. 2011. Set-Based Design and the Ship
to Shore Connector. Naval Engineers Journal, 123, 7992.
Morais, D., Waldie, M. & Larkins, D. Leveraging Engineering Information. COMPIT 2014, 2014/05// 2014.
Redwoth Hall, U.K., 327337.
Motta, D., Caprace, J.-D., Rio, P. & Boote, D. Optimization of Hull Structures for a 60 meters MegaYacht. 11th
International Conference on Fast Sea Transportation, FAST 2011, 2011/09// 2011. Honolulu, Hawaii, USA.
Nam, J. M. & Choung, J. Residual Longitudinal Strength of a VLCC Considering Probabilistic Damage Extents.
22nd International Offshore and Polar Engineering Conference (ISOPE), 2012/06/17/22 2012. Rhodes,
Greece, 815822.
New, C. RCM In The Royal NavyDeveloping A Risk Based Policy For Integrating Safety And Maintenance
Management. Managing Reliability and Maintainability in the Maritime Industry, 2012/01/25/26 2012.
London, UK: Royal Institute of Naval Architecture, 3238.
Nienhuis, Y. W. 2012. Automatic Generation of Assembly Sequence for the Planning of Outfitting Processes in
Shipbuilding. Journal of Ship Production and Design, 28, 4959.
Niese, N. D. & Singer, D. J. 2013. Strategic life cycle decision-making for the management of complex Systems
subject to uncertain environmental policy. Ocean Engineering, 72, 365374.
Oers, B. V. 2011. A Packing Approach for the Early Stage Design of Service Vessels. PhD, TU Delft.
Oers, B. V., Stapersma, D. & Hopman, H. 2007. Development and Implementation of an Optimisation-Based
Space
Allocation Routine for the Generation of Feasible Concept Designs. COMPIT 2007. Italy.
Ohba, Y., Kawamura, Y. & Kaede, Y. 2013. Lifecycle structural optimization of mid-ship of double hull tanker
based on holistic risk evaluation. In: Soares, G. & Romanoff, J. (eds.) Analysis and Design of Marine
Structures. London: Taylor & Francis Group, 555563.
515
Ostuni, L., De Pascalis, A., Calabrese, F., Catado, M., Mancarella, L., Zizzari, A. A. & Corallo, A. An On-board
Expert System for Damage Control Decision Support. 2013/04/15/17 2013. Cortona, Italy, 238247.
Pahl, G., Beitz, W., Blessing, L., Feldhusen, J., Grote, K.-H. & Wallace, K. 2007. Engineering Design: A
Systematic Approach, London, Springer-Verlag London Limited.
Pajunen, S. & Heinonen, O. 2013. Automatic design of marine structures by using successive response surface
method. Structural and Multidisciplinary Optimization.
Papanikolaou, A., Harries, S., Wilken, M. & Zaraphonitis, G. Integrated Ship Design and Multiobjective
Optimization Approach to Ship Design. International Conference on Computer Applications in Shipbuilding,
2011.
Papanikolaou, A. & Soares, C. G. 2009. Risk-Based Ship Design: Methods, Tools and Applications, Springer.
Parker, M. C. & Singer, D. J. The Impact of Design Tools: Looking for Insights with a Network Theoretic
Approach. Computer and IT Applications in the Maritime Industries (COMPIT), 2013/04/15/17 2013.
Pawling, R. & Andrews, D. 2011. Design SketchingThe Next Advance in Computer Aided Preliminary Ship
Design? COMPIT 201112th International Symposium on Computer and IT Applications in Shipbuilding.
Berlin, Germany.
Payer, H. G. & Schellin, T. E. 2013. A class societys view on rationally based ship structural design. Ships and
Offshore Structures, 8, 319336.
Pedersen, R. A., Molnes, D. A. & Stokkeland, L. S. 2015. Structural optimization for ice-strengthened vessels.
In: Santos, I. C. G. S. T. A. (ed.) Maritime Technology and Engineering. Taylor & Francis Group.
Pegg, N., Roberts, P. & Macadam, T. Improving the Integration of Ship Model Data in Simulation Based Design
and Management. International Marine Design Conference, 2012/06/11/14 2012. Glasgow, Scotland, 409
425.
Penas, R. An advanced step in integrating database-controlled shipbuilding CAD with PLM systems. 2013 2013.
Royal Institution of Naval Architects, 171179.
Piovesan, C. M., Fogleman, K., Green, H., Meade, M. & Peterson, M. Deepwater High Pressure and
Temperature Pipeline Asset Integrity Inspection. 22nd International Offshore and Polar Engineering
Piric, K., Kitarovic, S. & Zanic, V. Validation and verification of modified DRM with respect to its application
in the reliability based design approach. In: Soares, G., ed. IMAM 2013 Conference: Developments in
Maritime Transportation and Exploration of Sea Resources, 2013 London. Taylor & Francis Group, 779
787.
Pradillon, J. Y., Chen, C. P., Collette, M., Czaban, Z., Erikstad, S., Giuglea, V. & Jiang, X. 2012. Committee
IV.2 Design Methods. Proceedings of the 18th International Ship and Offshore Structures Congress.
Hamburg, Germany: Schiffbautechnische Gesellschaft e.V.
Prebeg, P., Zanic, V. & Vazic, B. An Application of a Complex System Optimization Techniques to the Ship
Structural Design. International Marine Design Conference, 2012/06/11/14 2012. Glasgow, Scotland, 371
390.
Prebeg, P., Zanic, V. & Vazic, B. 2014. Application of a surrogate modeling to the ship structural design. Ocean
Renard, P. 2012. LCM drives safety and efficiency advances. Naval Architect.
Roberts, P., Macadam, T. & Pegg, N. Multiple Simulation Assessments from a Single Ship Product Model.
Computer and IT Applications in the Maritime Industries (COMPIT), 2013/04/15/17 2013. Cortona, Italy,
301315.
Roldan, D. W. & De Souza, N. Risk-Based Analysis of LNG Carriers Loading and Unloading Operations. 22nd
International Offshore and Polar Engineering Conference (ISOPE), 2012/06/17/22 2012. Rhodes, Greece,
716724.
Ross, A. M. & Rhodes, D. H. 2008. Architecting systems for value robustness: Research motivations and
progress. 2008 2nd Annual Ieee Systems Conference, 110117.
Ross, A. M., Rhodes, D. H. & Hastings, D. E. 2008. Defining changeability: Reconciling flexibility,
adaptability, scalability, modifiability, and robustness for maintaining system lifecycle value. Systems
Rox, N. & Astrup, O. C. Streamlining the Steel Design Process by Linking Design and Rule Scantling Tools.
267273.
Ruede, E., Walker, M., Papadopoulos, Y., Securius, P. & Parker, D. Experience from Application of SemiAutomatic Fault Tree Synthesis to Reliability and Availability Analysis of Ship Systems. 22nd International
Offshore and Polar Engineering Conference (ISOPE), 2012/06/17/22 2012. Rhodes, Greece, 619626.
Ruy, W.-S., Ko, D.-E. & Yang, Y.-S. 2012. The implementation of the integrated design process in the holeplan system. International Journal of Naval Architecture and Ocean Engineering, 4, 353361.
516
Sanchez, L., Ringsberg, J. W. & Johnson, E. 2011. Optimization of composite maritime structureseffects of
uncertainties on design criteria and limits. In: Soares, G. & Fricke (eds.) Advances in Marine Structures.
London: Taylor & Francis Group.
Scherer, T. & Cohen, J. 2011. Coast Guard Reliability Centered Maintenance. Naval Engineers Journal, 123,
85109.
Schleder, A. M., Martins, M. R. & Modarres, M. The Use of Bayesian Networks in Reliability Analysis of the
LNG Regasification System on a FSRU under Different Scenarios. 22nd International Offshore and Polar
Engineering Conference (ISOPE), 2012/06/17/22 2012. Rhodes, Greece, 881888.
Sekulski, Z. 2011a. Multi-objective optimization of high speed vehicle-passenger catamaran by genetic
algorithm. Part I. Theoretical background on evolutionary multi-objective optimization. Polish Maritime
Research, 18, 318.
Sekulski, Z. 2011b. Multi-objective optimization of high speed vehicle-passenger catamaran by genetic
algorithm. Part II. Computational simulations. Polish Maritime Research, 18, 330.
Sekulski, Z. 2011c. Multi-objective optimization of high speed vehicle-passenger catamaran by genetic
algorithm. Part III. Analysis of the results. Polish Maritime Research, 18, 330.
Sekulski, Z. 2012. Selected problems of multi-objective optimization in the seagoing ship hull structure
preliminary design, Szczecin, Poland, Wydawnictwo Uczelniane Zachodniopomorskiego Uniwersytetu
Technologicznego w Szczecinie, Szczecin.
Sekulski, Z. 2013. Multi-objective optimisation of ship hull structure by genetic algorithm with combined
fitness function. In: Soares, G. & Romanoff, J. (eds.) Analysis and Design of Marine Structures. London:
Taylor & Francis Group, 515524.
Sekulski, Z. 2014. Ship Hull Structural Multiobjective Optimization by Evolutionary Algorithm. Journal of Ship
Research, 58, 4569.
Sekulski, Z. 2015. Scalarising of optimization criteria proposal for multi-objective optimization of ship hull
structure by evolutionary algorithm. In: Santos, C. G. S. T. A. (ed.) Marine Technology and Engineering.
Taylor & Francis Group.
Shin, S.-C., Lee, S.-S., Kang, D.-H. & Lee, K.-H. 2012. The development of internet based ship design support
system for small and medium sized shipyards. International Journal of Naval Architecture and Ocean
Silva, J. E., Garbatov, Y. & Soares, C. G. 2012. Reliability assessment of a plate subjected to randomly nonuniform corrosion and compressive load. Maritime Engineering and TechnologyGuedes Soares et al.:
Taylor & Francis Group, London, 451458.
Sname 2011. Risk-Based Optimization of Inspection Planning in Ships. Naval Engineers Journal, 123, 3642.
Sobey, A. J., Blake, J. I. R. & Shenoi, R. A. 2013. Optimisation of composite boat hulls using first principles
and design rules. Ocean Engineering, 65, 6270.
Sdal, S., Koekebakker, S. & Aadland, R. 2008. Market switching in shippingA real option model applied to
the valuation of combination carriers. Review of Financial Economics, 17, 183203.
Stenius, I., Rosn, A. & Kuttenkeuler, J. 2011. On structural design of energy efficient small high-speed craft.
Marine Structures, 24 (1), 4359.
Stone, B. Performance-Based Contracting Data Collection & Analysis. Managing Reliability and
Maintainability in the Maritime Industry, 2012/01/25/26 2012. London, UK: Royal Institute of Naval
Architecture, 95102.
Suh, H.-W., Lee, D.-H., Jung, J.-Y., Kim, S.-M., Shin, S.-H., Lee, J.-J., Lee, H.-Y. & Lee, J.-H. Development of
light-weight model based erection clearance check system for shipbuilding and ocean engineering. 2013
2013. Royal Institution of Naval Architects, 8993.
Sun, L. & Wang, D. 2012. Optimal structural design of the midship of a VLCC based on the strategy integrating
SVM and GA. Journal of Marine Science and Application, 11 (1), 5967.
Taflanidis, A. A., Loukogeorgaki, E. & Angelides, D. C. 2013. Offshore wind turbine risk
quantification/evaluation under extreme environmental conditions. Reliability Engineering & System Safety,
115, 1932.
Tanizawa, K., Kitagawa, Y., Takimoto, T. & Tsukada, Y. Development of an Experimental Methodology for
Self-Propulsion Test with a Marine Diesel Engine Simulator. 22nd International Offshore and Polar
Engineering Conference (ISOPE), 2012/06/17/22 2012. Rhodes, Greece, 921928.
Temple, D. & Collette, M. 2013a. Optimization of structural design to minimize lifetime maintenance cost of a
naval vessel. In: Soares, G. & Romanoff, J. (eds.) Analysis and Design of Marine Structures. London: Taylor
& Francis Group, 525532.
Temple, D. W. & Collette, M. Optimum Lifetime Maintenance Schedule for Naval Vessels subjected to Fatigue
and Corrosion. 12th International Symposium Practical Design of Ships and Other Floating Structures,
2013/10/20/25 2013b. CECO, Changwon, Korea, pp. 395402.
517
Tetrault, P. J. The Future Of Marine Engine Remote Monitoring In Marine Applications. Managing Reliability
and Maintainability in the Maritime Industry, 2012/01/25/26 2012. London, UK: Royal Institute of Naval
Architecture, 111113.
Thomassen, P. E., Bruheim, P. I., Suja, L. & Froyd, L. A Novel Tool for FEM Analysis of Offshore Wind
Turbines With Innovative Visualization Techniques. 22nd International Offshore and Polar Engineering
Thomson, D. & Renard, P. The Digital Handover Shipyards as Producers of Life-Cycle Maintenance Models.
363379.
Tiusanen, R., Jannes, J. & Liyanage, J. P. Identification and Evaluation of RAMS+I Factors Affecting the
Value-Added by Different Offshore Wind Turbine Concepts in Nordic Context. 22nd International Offshore
and Polar Engineering Conference (ISOPE), 2012/06/17/22 2012. Rhodes, Greece, 451457.
Todorov, S. R., Porter, N. & Lozev, M. Technologies and Methodologies for Automated Ultrasonic Testing
Quantification. 22nd International Offshore and Polar Engineering Conference (ISOPE), 2012/06/17/22
2012. Rhodes, Greece, 364372.
Tokola, H., Niemi, E. & Remes, H. 2013. Block Erection Sequencing in Shipbuilding With General Lifting and
Joining Times. Journal of Ship Production and Design, 29, 4956.
Ulstein, T. & Brett, P. O. Critical Systems Thinking in Ship Design Approaches. International Marine Design
Conference, 2012/06/11/14 2012. Glasgow, Scotland, 365383.
Vasconcellos, J. M., Harduim, M. & Araujo, P. 2015. Multi-criteria optimization applied to tankers preliminary
design. In: Santos, C. G. S. T. A. (ed.) Marine Technology and Engineering. Taylor & Francis Group.
Vassalos, D. Design for Safety: Risk-Based Design Life-Cycle Risk Management. International Marine Design
Conference, 2012/06/11/14 2012. Glasgow, Scotland, 83106.
Viana, F. a. C., Simpson, T. W., Balabanov, V. I. & Toropov, V. 2014. Metamodeling in Multidisciplinary
Design Optimization: How Far Have We Reallly Come? AIAA Journal, 52, 670690.
Volling, A., Peppler, M. E., Kalwa, C., Brauer, H., Ouaissa, B. & Mauser, H. Predicting Crack Arrest in Line
Pipes. 22nd International Offshore and Polar Engineering Conference (ISOPE), 2012/06/17/22 2012.
Wagner, J. & Bronsart, R. Build-Up Scenarios for Ship Life-Cycle-Methods for Descriptor Analysis.
International Marine Design Conference, 2012/06/11/14 2012. Glasgow, Scotland, 251262.
Wang, C., Cao, Y., Lin, W., Wang, L., Tang, Y. & Zhao, C. 2014. An Automated Mesh Generation Algorithm
for Curved Surfaces of Ship Longitudinal Structures.
Wang, G. G. & Shan, S. 2006. Review of metamodeling techniques in support of engineering design
optimization. Journal of Mechanical Design, 129, 370380.
Wilkman, G., Ranki, E., Leiviska, T. & Heinonen, Y. On Full-scale Onboard Ship Measurements in Various Ice
Conditions, a Review of Existing Database at Aker Arctic Technology. 22nd International Offshore and
Polar Engineering Conference (ISOPE), 2012/06/17/22 2012. Rhodes, Greece, 13651372.
Won, S.-I., Choi, B.-K. & Kim, D.-S. Development of the Integrated System for DLA of the Hull Structure.
12th International Symposium Practical Design of Ships and Other Floating Structures, 2013/10/20/25 2013.
Yang, H. & Wang, A. 2012. Fatigue Reliability Based Design Optimization of Bending Stiffener. Journal of
Ship Research, 56, 120128.
Yu, L., Callet, M. L., Tong, J., Li, T. & Reynolds, E. 2012. A process-centric software solution for container
ship design to procedural requirements. IMDC12The 11th International Marine Design Conference.
Glasgow, Scotland.
Zagkas, V. & Bralic, S. Real-Time Collaborative Design of Ship and Offshore Structures using BVB CAFE.
COMPIT 2014, 2014/05// 2014. Redwoth Hall, U.K., 468479.
Zanic, V. 2013. Methods and concepts for the multi-criteria synthesis of ship structures. Ships and Offshore
Zanic, V., Andric, J. & Prebeg, P. 2013a. Design synthesis of complex ship structures. Ships and Offshore
Zanic, V., Das, P. K., Pu, Y. & Faulkner, D. 1993. Multiple criteria synthesis techniques applied to reliability
based design of SWATH ship structure. In: Faulkner, D., Das, P. K., Inceci & Cowling (eds.) Integrity of
Offshore Strucutres. Glasgow: EMAS Scientific Publications, 387415.
Zanic, V., Kitarovic, S. & Prebeg, P. Safety as Objective in Multicriteria Structural Optimization. ASME 2010
29th International Conference on Ocean, Offshore and Arctic Engineering, 2010 Shanghai, China.
Zanic, V., Piric, K. & Kitarovic, S. Reliability and Robustness Based Design Attributes for Multi-Criteria
Decision Making. ASME 2013 32nd International Conference on Ocean, Offshore and Arctic Engineering,
06/09/2013 2013b Nantes, France.
518
Zanic, V., Prebeg, P. & Andric, J. Method for Structural Concept Design of Open Ships. 12th International
Symposium Practical Design of Ships and Other Floating Structures, 2013/10/20/25 2013c. CECO,
Changwon, Korea, pp. 280289.
Zhao, C., Ma, M. & Hughes, O. F. Applying Strip Theory Based Linear Seakeeping Loads to 3D Full Ship
Finite Element Models. OMAE 2013, 2013 2013. Nantes, France: ASME, OMAE201310124.
Zheng, J., Gao, L., Qiu, H., Shao, X. & Jiang, P. 2013. Variable fidelity metamodel-based analytical target
cascading method for green design. International Journal of Advanced Manufacturing Technology, 114.
Zhu, B. & Frangopol, D. M. 2013. Reliability assessment of ship structures using Bayesian updating.
Engineering Structures, Volume 56, 18361847.
Zhu, J. & Collette, M. Reliability Based Structural Failure Modeling through Bayesian Networks. 12th
International Symposium Practical Design of Ships and Other Floating Structures, 2013/10/20/25 2013.

710 SEPTEMBER 2015
CASCAIS, PORTUGAL
VOLUME 2
COMMITTEE V.1
ACCIDENTAL LIMIT STATES

COMMITTEE MANDATE
Concern for Accidental Limit States (ALS) of ship and offshore structures and their structural components
during design. Types of accidents considered shall include fire, explosion, dropped object, collision and
grounding. Attention shall be given to hazard identification and related risks, assessment of accidental loads
and nonlinear structural consequences including residual strength. Uncertainties of ALS models for the use
in design shall be highlighted. Consideration shall be given to practical application of design methods and to
the development of ISSC guidance for implementation of ALS principles in engineering.
COMMITTEE MEMBERS
Chairman:
J. Czujko, Norway
L. Brubak, Norway
J. Czaban, Canada
M. Johnson, UK
G.S. Kim, Korea
S.J. Pahos, UK
K. Tabri, Estonia
W.Y. Tang, China
J. Wgter, Denmark
Y. Yamada, Japan
KEYWORDS
Accidental Limit States, hazard identification, safety levels, accidental loads, accidental load effects, fires,
explosions, dropped objects, ship collision, structural safety, material models for non-linear structural
analysis, fire design benchmark study.
520
ISSC committee V.1: ACCIDENTAL LIMIT STATES
CONTENTS
1.
INTRODUCTION ..................................................................................................................... 523
2.
FUNDAMENTALS OF ALS DESIGN ................................................................................... 524

2.1 Introduction ..................................................................................................................... 524
2.2 Codes and standards ....................................................................................................... 525
2.3 Updates of codes and standards...................................................................................... 527
2.4 Uncertainties in ALS in Design...................................................................................... 527
2.5 Practice for ships ............................................................................................................. 527
3.
HAZARD IDENTIFICATION ................................................................................................. 528

3.1 Introduction ..................................................................................................................... 528
3.2 Hazard identification....................................................................................................... 530
4.
SAFETY LEVELS IN ALS DESIGN ...................................................................................... 532

4.1 Introduction ..................................................................................................................... 532
4.2 Safety level of offshore structures in ALS ..................................................................... 532
4.2.1 General ............................................................................................................... 532
4.2.2 Discussion of new ISO standards for offshore structures ................................ 532
4.2.3 Characterization of hazards ............................................................................... 533
4.2.4 Accidental design situations .............................................................................. 533
4.2.5 ALS safety levels implied in structural codes .................................................. 533
4.3 Safety level of Ship Structures in ALS .......................................................................... 535
4.3.1 General ............................................................................................................... 535
4.3.2 GBS of Ship Structure Design .......................................................................... 535
4.3.3 Safety level in ULS in CSR............................................................................... 536
4.3.4 Safety level in ALS in CSR-H .......................................................................... 536
5.
ASSESSMENT OF ACCIDENTAL LOADS.......................................................................... 538

5.1 Introduction ..................................................................................................................... 538
5.2 Explosion Load Assessment ........................................................................................... 538
5.2.1 Deterministic approach...................................................................................... 539
5.2.2 Probabilistic approach ....................................................................................... 539
5.2.3 Definition of explosion loads for design ........................................................... 542
5.3 Fire Load Assessment ..................................................................................................... 542
5.3.2 Risk-based and probabilistic approach ............................................................. 543
5.4 Load Assessment for Collision accidents ...................................................................... 544
5.4.2 Risk-based and probabilistic approach ............................................................. 545
5.5 Load Assessment for Dropped Object accidents ........................................................... 546
5.5.2 Risk-based approach .......................................................................................... 547
6.
DETERMINATION OF ACTION EFFECTS ......................................................................... 547

6.1 Introduction ..................................................................................................................... 547
6.2 Review of Numerical Tools............................................................................................ 549
6.3 Modelling Geometries .................................................................................................... 550
6.4 Modelling Loads ............................................................................................................. 552
6.4.1 Ship Collision .................................................................................................... 552
6.4.2 Dropped Objects ................................................................................................ 553
6.4.3 Explosions .......................................................................................................... 553
6.5
6.6
6.7
6.8
521
6.4.4 Fire ..................................................................................................................... 554

Material Models .............................................................................................................. 554
6.5.1 Plasticity Model ................................................................................................. 557
6.5.2 Stress-Strain Curve ............................................................................................ 557
6.5.3 Failure Criteria ................................................................................................... 557
Uncertainties of ALS Models ......................................................................................... 560
Probabilistic methods...................................................................................................... 560
APPENDIX A ................................................................................................................. 560
6.8.1 True stress-strain curve for Ls-Dyna ................................................................ 560
7.
BENCHMARK STUDY. RESISTANCE OF TOPSIDE STRUCTURES

SUBJECTED TO FIRE ............................................................................................................. 561
7.1 Scope of work ................................................................................................................. 561
7.2 Strategy of benchmark study .......................................................................................... 562
7.3 Input ................................................................................................................................ 562
7.3.1 Geometry of target structure ............................................................................. 562
7.3.2 Material data ...................................................................................................... 563
7.3.3 Boundary conditions .......................................................................................... 564
7.3.4 Loads .................................................................................................................. 564
7.4 Results ............................................................................................................................. 566
7.4.1 Static analysis .................................................................................................... 566
7.4.2 Push-down analysis ........................................................................................... 567
7.4.3 Fire analysis ....................................................................................................... 568
7.4.4 Design of PFP .................................................................................................... 570
7.4.5 Effects of boundary conditions ......................................................................... 571
7.4.6 Methods of controlling numerical instability for beam element model ........... 571
7.4.7 Effects of local heat flux.................................................................................... 573
7.5 Conclusion from the benchmark study .......................................................................... 575
8.
REFERENCES .......................................................................................................................... 576
9.
ANNEX 1. MATERIAL MODELS FOR NON-LINEAR FINITE ELEMENT

ANALYSIS ................................................................................................................................ 579
9.1 Introduction ..................................................................................................................... 579
9.2 Guidelines and standards ................................................................................................ 580
9.3 Material model database ................................................................................................. 580
9.3.1 Steel .................................................................................................................... 580
9.3.2 Aluminium ......................................................................................................... 583
9.3.3 Foam, Isolator, Rubber ...................................................................................... 584
9.3.4 Ice ....................................................................................................................... 584
9.3.5 Air ...................................................................................................................... 585
9.3.6 Water .................................................................................................................. 586
9.3.7 Explosives .......................................................................................................... 586
9.3.8 Risers, Umbilical or Power Cable ..................................................................... 587
9.3.9 Composites ........................................................................................................ 587
9.3.10 Concrete ............................................................................................................. 588
9.3.11 Soil ..................................................................................................................... 588
9.4 References ....................................................................................................................... 589
522
Nomenclature:
ALARP
CFD
DAL
DOP
FEES
F&G
FW
FR
HAZID
HC
HSE
LEL
NFPA
PSA (PTIL)
PFP
QRA
TRA
TRR
TR
As Low As Reasonably Practical

Computational Fluid Dynamics
Dimensioning Accidental Loads
Dropped Object Protection
Fire, explosion and escape strategy according to ISO 13702
Fire and Gas
Fire Water
Functional Requirements (of the Contract)
Hazard Identification Study
Hydrocarbon
Health, Safety & Environment
Lower Explosion Limit
National Fire Protection Association
Petroleum Safety Authority Norway
Passive Fire Protection
Quantitative Risk Analysis
Total Risk Analysis
Tubing Replacement Rig
Temporary Refuge
Definitions:
Action: external load applied to the structure (direct action) or an imposed deformation or acceleration
(indirect action). A hazard creates actions that are used to determine action effects from a hazard.
Accidental Action: load originated from identified hazards during the design phase; it is the outcome of
QRA.
Action Effect: effect of action(s) on the structure or its components.
Design Accidental Load: chosen accidental load that is to be used as the basis for design.
NOTE 1 The applied/chosen design accidental load may sometimes be the same as the
dimensioning accidental load (DAL), but it may also be more conservative based on other input and
considerations such as ALARP. Hence, the design accidental load may be more severe than the DAL.
NOTE 2 The design accidental load should as a minimum be capable of resisting the
dimensioning accidental load (DAL).
Dimensioning Accidental Event (DAE): Accidental events that serve as the basis for layout, dimensioning
and use of installations and the activity at large.
Dimensioning Accidental Load (DAL): Most severe accidental load that the function or system shall be
able to withstand during a required period of time, in order to meet the defined risk acceptance criteria.
NOTE 1 DAL is normally defined based on DAE.
NOTE 2 The dimensioning accidental load (DAL) is typically generated as a part of a risk
assessment, while the design accidental load may be based on additional assessments and considerations.
NOTE 3 The dimensioning accidental load (DAL) is typically established as the load that
occurs with an annual probability of 10-4.
Explosion load: time dependent pressure or drag forces generated by violent combustion of a flammable
atmosphere.
523
Drag load: Drag force is caused by expanding hydrocarbon gas and air after explosion impinging upon an
object. The drag force is a function of the fluid velocity and density along with the object's reference area
and drag coefficient. The drag coefficient may further be a function of the Reynolds number. Reynolds
number depends on the fluid density, viscosity, and velocity as well as the object's characteristic length.
Smaller objects like piping which are inside an exploding gas cloud will be subjected to drag force.
Failure Strain: strain level at which the material is no longer providing any stiffness.
Fire load: Heat flux, normally defined in kW/m2 for a specified duration.
Frequency: The number of measurements or (expected) observations having a certain value, or
characteristic, during a certain observation period (e.g. expected annual frequency is a number of
expected observations during one year observation period).
Hazard: potential for human injury, damage to the environment, damage to property, or a combination of
these.
Hydrocarbon gas explosion: A process where combustion of a premixed gas cloud, i.e. fuel-air or fueloxidiser, is causing rapid increase of pressure. Gas explosions can occur inside process equipment or
pipes, in buildings or offshore modules, in open process areas or in unconfined areas.
Integrity: the ability of a structure to perform its required function effectively and efficiently over a
defined time period, while protecting health, safety and the environment.
Jet fire: Ignited release of pressurized, flammable gas and fluids.
Limit State: state beyond which the structure no longer fulfils the relevant assessment criteria.
Pool fire: Combustion of flammable or combustible fluids spilled and retained on a surface.
Probability: the relative frequency with which an event occurs, or is likely to occur
1.
INTRODUCTION
Accidents do happen. Structures like offshore facilities and ships are especially at risk as the presence of
high-pressure gas and hydrocarbons involves high potential risk of leaks, explosions and fires. Heavy ship
traffic, on the other hand, is associated with risk of ship impact or dropped objects on offshore structures,
as well as collisions between the ships. Keeping in mind the risks involved, offshore and shipbuilding
industries focus on diminishing accident hazards and protecting industrial facilities against them already
in the design phase.
Offshore facilities can only be put into operation based on national requirements. In order to obtain
such permissions, offshore facilities have to be designed for safe operation. Therefore, design standards
for offshore facilities include requirements and procedures for design against hazards. These standards
describe methods for the assessment of hazards and actions, as well as structural consequences emerging
from these hazards. In contrast to the requirements for offshore structures, ships have to obtain a class
from a classification society in order to operate on international waters. To get certifications from
classification societies, ships must be designed according to the classification rules and approved by the
societies. Contrary to offshore industry, current ship design rules do not include requirements for design
against hazards that may occur during ship operation.
Since offshore industrial systems are designed using a number of different standards and procedures,
special design tools have been established. Generally, two design philosophies are widely used in
structural engineering; the working stress method WSM (allowable stress method ASM) and the load
resistance design factors (LRDF). LRDF design method is based on the limit state concept, a concept that
involves the idea that a structure has a limit state beyond which it does not perform its functions anymore.
There are several limit states used in design that are meant to address hazards acting on industrial
facilities. An example are Accidental Limit States, which have been developed as a guidance in design
process. Accidental limit state for a structure defines a state beyond which the damage caused by accident
inevitably leads to collapse as a result of domino effect. As long as resistance is an inherent property of
the structure, loads rising from accidents are very uncertain and can be best presented as expected values
or in the form of probabilistic models.
This report is focused on the Accidental Limit States following Committees conviction that defining
ALS is a fundamental factor in the safe offshore and ship structure design. The scope of work includes
hazard identification, fundamentals of ALS design, safety levels in ALS design and the assessment of
accidental loads for hydrocarbon fires and explosions. A review of international and national standards
for offshore structures and ships has been presented in the section on Fundamentals for ALS design. The
524
section on Safety Levels in ALS Design gives basis and evaluation of these levels, e.g. reliability based
calibration of ALS codes. The Assessment of Accidental Loads section gives insight into methods of load
assessment to be used in ALS design.
The Committee has also carried out a benchmark study. Its objective was to assess the capability and
accuracy of available techniques for the prediction of structural response and strength of topside
structures without and with Passive Fire Protection (PFP) when subjected to define fire loads. In the study
the definition of the requirements for PFP using existing standards and numerical calculations are
compared, and the capabilities of modern software to simulate fire loads and responses are evaluated.
2.
2.1
FUNDAMENTALS OF ALS DESIGN

Introduction
The limit state design (LSD), also known as the load and resistance factor design (LRFD), refers to a
design method used in structural engineering. A limit state is a condition of a structure beyond which it
no longer fulfills the relevant design criteria. The condition may refer to a degree of loading or other
actions on the structure, while the criteria refer to structural integrity, fitness for use, durability or other
design requirements. A structure designed by LSD is proportioned to sustain all actions likely to occur
during its design life, and to remain fit for use, with an appropriate level of reliability for each limit
state. Building codes based on LSD implicitly define the appropriate levels of reliability by their
prescriptions.
From the viewpoint of a structural designer, four types of limit states are considered for steel structures,
namely:
serviceability limit state (SLS);

ultimate limit state (ULS);
fatigue limit state (FLS); and
accidental limit state (ALS).
The primary aim of the ALS design for steel structures may be characterized by the following three
broad objectives, namely:
to avoid loss of life in the structure or the surrounding area;
to avoid pollution of the environment; and
to minimize loss of property or financial exposure.
In ALS design, it is necessary to achieve a design such that the main safety functions of the structure
must not be impaired during any accidental event or within a certain time period after the accident. Since
the structural damage characteristics and the behavior of damaged structures depend on the types of
accidents, it is not straightforward to establish universally applicable structural design criteria for the
ALS. Typically, for a given type of structure, design accidental scenarios and associated performance
criteria must be decided upon the basis of risk assessment.
In many cases in design for accidental actions, prescriptive criteria derived from experience and
related studies are used. Prescriptive criteria may be assumed accidental effects, such as damage extent
due to collision, loss of a major bracing member, etc. Over the last two decades, the focus on ALS
design criteria in the design requirements in different design standards has increased, especially within
offshore structures. In order to replace or support the prescriptive criteria, there has been greater
recognition of the use of risk based procedures for offshore structures where the consequences of
structure failure are considered. Different approaches are used in different design standards and the
complexity of the approaches is varying. Due to the development of high performance computers, more
and more advanced methods such as nonlinear finite element analysis are used and consequently
increase the reliability of the results.
ALS in design consists of several phases and a short overview is presented in a flowchart in Figure .
The first step is the performance of a Quantitative Risk Assessment (QRA) which is a formalised
specialist method for calculating individual, environmental, employee and public risk levels for
comparison with regulatory risk criteria. In the design process action effects for every design hazard
action are calculated in order to evaluate risk and to determine structural safety.
525
Figure 1. Flowchart for application of the ALS in Design.
2.2
Codes and standards
In design standards, it is required that damage from accidental actions with reasonable likelihood of
occurrence shall not lead to complete loss of integrity of the structure, and the load-bearing function of
the structures must be maintained. The design must be dimensioned such that critical parts for the overall
strength are strong enough to withstand an accidental action, or alternatively, can be dimensioned in order
to minimize the consequences with a certain redundancy without causing failure. For ALS design criteria,
the most important standards for the offshore industry can be listed as:
ISO 19900 series

ISO 19900 General requirements for offshore structures
ISO 19902 Fixed steel offshore structures
ISO 19901-3 Topside structures
Norsok standards
N-001 Integrity of offshore structures
N-003 Actions and action effects
N-004 Design of steel structures
Z-013 Risk and emergency preparedness assessment
API Recommended Practice
14J Recommended Practice for Design and Hazards Analysis for Offshore Production Facilities,
2007
2FB Recommended practice for the design of offshore facilities against fire and blast loading, 2006
DNV-GL standards
DNV-RP-C204 Design against accidental load, 2010
DNV-RP-C208 Determination of Structural Capacity by Non-linear FE analysis Methods, 2013
ABS standard
Accidental load analysis and design for offshore structures, 2013
The design standards provide an overview of approaches that can be used to identify and assess the
effects of accidental structural loads arising from four accidental actions. These are dropped objects,
vessel collision, fire and blast/explosion. Typically, the approaches in the design standards for these
actions may either be determined by non-linear finite element analysis or energy considerations combined
with simple elastic-plastic methods. Since the computer resources are increasing continuously, it is more
and more common to use non-linear FE analysis.
A good introduction to the complete process may be obtained by studying the DNV Offshore standards
series and Norsok Standards.
DNV-OS-A101 (DNV-OS-A101, 2014) provides a standard for the safety and arrangements of
offshore installations by defining requirements for design loads, site arrangements, classification, shut
526
down logic, alarms, escape methods and communication. It serves as a contractual reference document
between suppliers and purchasers and provides guidelines for designers, suppliers, purchasers and
regulators as well as specifying procedures and requirements for units or installations subject to DNV
classification. More specifically, it outlines simplistic Design principles and defines accidental loads;
arrangements; hazardous area classification; emergency Shutdown (ESD) principles and requirements;
escape and communication requirements for the design and operation of offshore structures.
The DNV-OS-C101 (DNV-OS-C101, 2014) outlines the general design of offshore steel structures
using LRFD methods and considers hazards that create accidental loads to include impact from ship
collisions; impact from dropped objects; fires; explosions; abnormal environmental conditions; and
accidental flooding. Their design principle considers that within their prescribed accidental limit states
(ALS), structures with damage to their components arising from accidental events or operational failure
can exceed ductile limits but that the ultimate resistance of damaged structures will maintain structural
integrity after local damage or flooding with possible loss of station keeping (free drifting). However,
DNV policy on accidental limit states notes that since the onus for design remains subject to the inherent
uncertainty of the frequency and magnitude of the accidental loads, as well as the approximate nature of
the methods for determination of accidental load effects, that sound engineering judgement and pragmatic
evaluations in the design are essential. Consequently, many initiatives remain underway to better define
and reach agreement on reasonable design limit load definitions to establish rational standards.
DNV-RP-C204 (DNV-RP-C204, 2010) recommended practice for the analysis of loads against the
offshore installation design so as to maintain the load-bearing function of structures during accidental
events. The general philosophy is based on developing a design such that accidental loads will not impair
the main safety functions of the installation. The explicit design guidelines and analysis methods
presented by this standard enable very detailed and thorough design development. More specific
requirements for probabilistic based methods to demonstrate structural reliability compliance with
offshore standards are referred to DNV Classification Note 30.6 (DNV 1992). The standard also describes
two other methods for the design of offshore steel structure including the Load and Resistance factor
Design Method (LRFD) and a design method assisted by testing (quasi-deterministic). The design by load
resistance factor method (LRFD) defines the basic variables as being the loads acting on the structure
compared against the resistance of the structure. A target safety level is determined based on the
anticipated variation in the load and resistance and the reduced probabilities that the loads will act
simultaneously at their expected values. The level of safety of a structural element is considered to be
satisfactory if the design load effect does not exceed the design resistance.
Germanicher Lloyd (GL, 2013) provides guidelines for hazard identification and risk assessment of
offshore structures. They indicate hazard identification requires consideration of external hazards, such as
ship collisions, extreme environmental conditions and helicopter crashes. The risk assessment of the
hazards requires consideration of their tolerability to personnel, the facility and the environment. The
assessment involves identifying the initiating events, the possibility of escalation, estimating their
probability and their consequences.
Norsok standard N-001 (NORSOK-N-001, 2010) specifies general principles and guidelines for the
design and assessment of offshore facilities, and the verification of load bearing structures subjected to
foreseeable actions and related maritime systems. For accident limit state verification, the standard
requires design checks, per ISO 19900, to ensure that the accidental action does not lead to complete loss
of integrity or performance of the structure and related maritime systems. It requires that design actions
and resistances be calculated using deterministic computational models with uncertainties in the
computational models being covered by partial factors. However, the standard also allows design
verification to be based on a more complete reliability design method, provided it can be documented that
the method is suitable from a theoretical point of view, and that it provides adequate safety in typical
known cases. This opens for use of reliability methods which entail calibration of action and material
factors against a given failure probability level, or direct design by means of such methods. The safety
level can be calibrated directly against the safety of known structure types and be based on corresponding
assumptions. Direction on using reliability based design methods requires only that the results are on the
safe side. It does not define any specific hazard or limit state requirement.
Norsok standard N-003 (NORSOK-N-003, 2007) defines permanent and variable actions pertinent
to offshore structures. It provides a good overview of sea states, waves, wind, storms, ocean currents,
ice, snow, earthquakes, temperature and their effects on structures. It defines accidental actions as those
caused by abnormal operation or technical failure and includes, fires, explosions, ship collisions,
dropped objects, helicopter crashes and changes to intended pressure differentials. Accidental
assessments need to be supported with due account made for personnel qualifications, operational
procedures, system design features, safety systems and control procedures. The accidental action design
review applies to each accidental action that corresponds to an annual exceedance probability of 10-4.
527
Detailed descriptions and considerations for appropriate assessment of each hazard is presented as well
as methods to assign probabilistic values for combinations of events arising from the same phenomenon
like hydrocarbon gas fires and explosions. The ALS design check requires evaluation of the structural
damage caused by accidental actions; the ultimate capacity of structures with damage. The large
uncertainties associated with determining the accidental actions, normally justify the utilisation of
simplified nonlinear analyses methods both to calculate the damage and the global ultimate strength of
the damaged structure. Such methods may be based on plastic mechanisms (yield hinge or line
methods) with due recognition of possible premature rupture. Non-linear FE analyses are recognized to
determine the ultimate capacity of damaged structures.
Norsok standard N-006 (NORSOK-N-006, 2009) provides detailed guidance on the selection and
application of risk based and probabilistic methods for the design assessment of offshore structures, but
does not address methods specific to hazards nor accidents.
Norsok standard Z-013 (NORSOK-Z-013, 2010) provides a very comprehensive description of risk
and emergency preparedness analysis associated with exploration drilling, exploitation, production and
transport of petroleum resources as well as all installations and vessels that take part in the activity.
2.3
Updates of codes and standards
The design standards are continuously updated in order to be more accurate and suited for todays
accidental actions that may arise. As mentioned above, the trend is to use more advanced methods such as
nonlinear finite element method. Recently, a new guideline on how to establish the structural resistance
by the use of non-linear FE methods was developed (DNV-RP-C208, 2013). This guideline gives a
relatively detailed description on FE modelling (material, mesh, boundary conditions, etc.), analysis
procedure and post-processing of the results.
As a continuation of the development of DNV-RP-C208, a new Joint Industry Project was initiated in
2014 with the working title Nonlinear FE analysis of offshore structures Further development and
extension of design guidelines. This project will specifically address analyses related to accidental
actions, especially what is required when considering high energy ship impacts. This is necessary since
the acceptance criteria and analysis methodologies are specified in design standards written decades ago.
These are normally based on representative offshore supply vessel (OSV) design in the order of 5000 tons
displacement and characterized by a traditional raked bow. Modern OSVs are often designed with a
bulbous bow and are up to 8000-10000 tons displacement with a consequent increase of the impact
energy to be absorbed in the event of a collision. As a consequence, the Petroleum Safety Authority
Norway is considering to increase recommended ship impact design energies (Kvitrud, 2013).
2.4
Uncertainties in ALS in Design
There are several uncertainties in ALS design due to the approximate nature of the methods for
determination of action effects. It is therefore essential to apply sound engineering judgment in the
design.
For determination of actions effects, the usual approach is either to use simple elastic-plastic methods
or the non-linear FE analysis. Typically, the simple elastic-plastic methods are more conservative
compared to the more sophisticated non-linear FE analysis, and these simple methods are typically not
able to redistribute forces after failure of structural elements. On the other hand, the non-linear FE
analysis is more complex and is able to redistribute forces without complete loss of integrity of the
structure if a non-critical structural member fails. However, the non-linear FE analysis requires
experienced users and there are several factors that must be accounted for in order to achieve reliable
results. These factors can typically be material model, solution procedure, mesh refinement, etc. which
are discussed in more detail in the consecutive sections. For detailed review of action effect assessment in
ALS design see Sec. 6.
2.5
Practice for ships
Design incorporating ALS considerations has not been widely adopted by the shipping industry. This is
largely due to the evolutionary nature of the industry over hundreds of years and the development rather
of prescriptive rules for design; but also the challenges presented by being by definition structures which
must move in different ocean regions and be exposed to a range of environmental loads that are not easy
to quantify at the design stage. Nevertheless, there is growing interest in the use of alternative methods
and this is also being pushed by IMOs long term target for Goal Based Standards; though this may be
seen as reliability based safety limit approach rather than purely LSD.
It is therefore not surprising that there are few examples of whole ship design incorporating ALS in the
literature. However, there are several examples of ship design where specific aspects of ALS design are
considered.
528
Levander (2010) shows adoption of a range of safety features for the cruise liner Oasis of the Seas. A
major design feature is that the ship itself should be its own best lifeboat, leading to adoption of a great
deal of redundancy in the propulsion system, and location of prime movers in separate fire zones. These
are methods that have been historically used for naval vessels. Damage stability of the liner was proven
using probabilistic studies involving thousands of possible damage cases.
The specific design area of (damage through) collision and grounding is receiving considerable
attention at the moment. Pedersen (2010) presents a review of available collision and grounding analysis
procedures which supports the adoption of in risk based design methods. The mains steps are
Estimation of the grounding and collision probability

Models for calculation of the resulting grounding and collision damage
Analysis of the conditions of the damaged vessels
Estimation of costs associated with the accidents
Regarding the theoretical models, the classic approach is separate calculation of the main elements,
namely the ship motion during the incident, the external dynamics, the structural response and the internal
mechanics. Energy is transferred between these elements and damage estimated through a forcepenetration curve. This approach suffers the drawback of requiring a priori knowledge of the penetration
path which is only easily estimated for simple right angle collisions where the struck ship is assumed
motionless. The current research trend is for methods to couple two or more of these elements in the
analysis; FEM may be used for the structural response.
Ehlers and Tabri (2012) present such a combined model which avoids direct use of FEM, a numerical/
semi-analytical procedure for collision assessment including traffic density, striking angles and collision
velocities, using the example of a RoPax vessel; that work is extended in Montewka et al. (2014).
Contributing to establishing statistics for reliability approach, Cho et al. (2014) made a parametric
study of double hull tanker designs in collision resistance. Hussein and Soares (2009) also contribute in
this area with a study of three double hull tankers designed from the IACS Common Structural Rules;
they go on to argue the benefit of 20% increase in deck plate thickness in collision scenarios. Also
Saydam and Frangopol (2013) use reliability methods to study the performance of a single skin oil tanker
in sudden damage specifically grounding and collision scenarios. The effect of ship speed and heading
to the waves, and hydrodynamic analysis based on the sea states are included; the operational conditions
are shown to have significant effect.
Specifically regarding grounding damage, Kim et al. (2013) present a probabilistic design approach
grounding damage index R-D diagram, where R is residual strength and D is damage index. This is
based on the damage method proposed by Paik et al. (2012). In this study 50 grounding scenarios were
modelled for VLCC, Suezmax, Aframax and Panamax oil tankers. This method is seen as a temporary
first approach to the problem rather than a high quality model for design.
3.
3.1
HAZARD IDENTIFICATION
Introduction
Christou and Konstantinidou (2012) provide a good summary of existing databases from regulatory
authorities concerning analysis of past accidents in offshore oil and gas operations including Hydrocarbon
release, collisions and marine accidents including blowouts. Of particular interest, they report and
quantify the nature of structural defects responsible for seven major offshore accidents between 1977 and
2010 and list lessons learned. It is apparent from a review of their findings and stated lessons learned, that
the significant consequences, albeit possibly initiated by extreme loads (e.g. Alexander Kielland capsize
1980) were a direct result of the Ton Vrouwenvelder (2014) fifth column of human errors ranging from
failure to inspect welds for cracks, failure to monitor for fatigue, improper drilling procedures (Ixtoc
blowout 1979) (Adriatic Blowout 2004), miscommunication between shift workers (Piper Alpha
explosion 1988), incorrect safety valve (Ekofisk Blowout 1977), improper cementing following pipe
rupture (Montara Blowout 2009, Macondo blowout 2010). Their findings revealed a series of
organization and management failures rather than deficiencies in structural design. Their
recommendations stressed the need to address inadequate hazard identification procedures and separate
the operational from regulatory body responsibilities.
Nevertheless, it remains important to better quantify and codify minimum design requirements for marine
structures to withstand expected environmental and accidental loads. They include statistics from the DNV
world offshore accident dataset (WOAD) to European and American ocean areas and note a lack of data
from other areas. The highest numbers of incidents, in order, are related to FPSO/FSU Jackets, Jack-up and
helicopter operations, followed by semi-submersibles, pipelines and well support structures.
(a) Drilling operations
529
(b) Production operations
Figure 2. World accident statistics for drilling and production operations.
The WOAD statistics indicate that 86% (5323) accidents are non-human related but attributed to either
unsafe procedures or the absence of procedure, with about 1 percent due to acts of war or sabotage and
8% attributed to some form of design defect.
The total losses during various accidents in offshore operations between 1970 and 2007 are
given by OGP-434-17 (2010) and illustrated in Figure 2 Between 1970 and 2014 352 people were
killed and 173 injured in fire and explosion accident associated with offshore operations.
Figure 3. Percent distribution of total losses during offshore operation accidents between 1970 and
2007, OGP 2010.
Of additional interest, in cases of equipment malfunction, less than 0.18% was attributed to safety system
malfunction or cataclysmic earthquakes or volcanic activity, with weather and fires responsible for about
25% respectively, 8% for structural failures and 34% from natural causes (wear-out etc.). Of additional
530
interest, fixed units tended to have three to four times the number of fluid or gas spills and fires than
mobile units. In terms of total loss from a single cause, capsizing and loss of buoyancy (~80 respectively),
followed by fire (41) and collisions (33).
The U.S. Bureau of Safety and Environmental Enforcement http://www.bsee.gov maintains a very
useful database concerning details of offshore incidents and related statistics.
3.2
Hazards refer to situations or events with the potential to cause human injury, damage to the environment
or property. Operational hazards can take the form of monthly activities such as a shuttle tanker as it
approaches an offshore facility, or daily activities like a crane that lifts equipment off the laydown area.
What is distinct on the above hazards are the expected frequencies at which an offshore structure is
exposed to these hazards, and the consequences on structural integrity they might have. In practice, the
term hazard is often associated with probabilities of occurrence and expected consequences, the product
of the two is defined as the risk.
A comprehensive hazard identification (HAZID) is often part of the wider update of the quantified risk
assessment (QRA) and offshore safety case. It is defined by DNV (2002) as the process used to identify
hazards as an essential first step in a risk assessment. Since hazards are diverse, there are many methods
available for their determination. In general a structured approach is necessary to avoid overlooking
possibilities and to encourage a creative identification of previously neglected scenarios. The process
makes use of experience with accidents and draws from the expertise of a group of people with diverse
backgrounds to avoid biasedness and benefit from brainstorming. The various techniques include hazard
and operability studies (HAZOP) that apply a series of guidewords to systematically consider possible
operational deviations from design intent. Failure mode effects and criticality analysis (FMECA) is
another systematic method used to identify failure modes of electrical and mechanical systems. Failures
are often rated critical if they have a high frequency or severity rating. The structured what-if technique
(SWIFT) is similar to HAZOP but based on creative brainstorming rather than a formal list of
guidewords.
Ultimately, identified hazards become part of a hazard register to support initial, and through-life safety
management. In general, hazard assessments are best reported using a risk assessment matrix in line with
qualitative or quantitative risk assessment. A typical matrix includes a severity scale, ranging from
negligible, through critical to catastrophic, a frequency range from frequent, through probable, remote or
incredible and a risk class that includes, intolerable, undesirable, and tolerable with or without mitigation
measures. Variations of risk matrices are presented by ISO-17776 (2000).
Generally, Classification Rules require that structures comply with extreme permanent loads
superimposed on regular environmental conditions, as well as for extreme environmental conditions
superimposed on large permanent loads. The aim of Accidental Limit State (ALS) design is to provide
adequate structural reliability to prevent catastrophic failure in the event of extreme and unusual
environmental load conditions that arise from collisions, groundings, fires, explosions, system failures,
dropped objects, terrorism and other cataclysmic events including earthquakes, hurricanes and tsunamis.
Extreme and unusual environmental loads arise from events that are collectively considered to be hazards.
In contrast to SLS, where design loads require elastic behaviour, materials subjected to action effects can
be expected to deform beyond their elastic limits to the point of failure. The definition of actions and
action effects in the context of ALS is given in Section 6. HAZID studies often comply with policies of
various regulatory authorities and/or with the requirements set out in the operators guideline on HAZID
studies whereas available. It is the fundamental requirement for operators to perform an appropriate
assessment to clearly demonstrate that envisaged risks have been identified and prevention measures put
in place.
More specifically , in his overview of reliability based structural design, Ton Vrouwenvelder (2014)
expands the list of hazards to broach five categories that include normal loads, natural accidents, manmade accidents, human influences and human errors. While it is possible to develop rational design for
normal and accidental loads, and while human reliability analysis methods continue to develop, it is
difficult to design against terrorism, vandalism, misuse, maintenance and operational mistakes.
Particularly the hazard of terrorism was proved to be beyond operational hazards of marine structures,
where even the recommended target reliability indexes by Pahos (2009) are dependent on the location of
the panel and should be applied after a HAZID study. When applied to offshore structures, the
requirement for the ALS is based on the philosophy that small damages, which inevitably occur, should
not cause disproportionate consequences (Moan et al., 2002).
531
Given their diverse range of characteristics, absolute magnitudes of loading functions arising from
accidents are based on the use of load exceedance curves. Such curves plot the magnitude of the load
against the probability of exceeding it as a function of time. NORSOK-Z-013 (2010) provides a
procedure to generate such load exceedance curves for offshore structures. Exceedance curves are
intended to rationalize load limits. Figure 3 suggests a typical 10-fold increase in magnitude between the
1/100 year statistics compared to the 1/1000 year likely event. In the limit, such arguments are used to
support establishing rationalized design requirements. The resultant accidental load that a facility can
withstand during a required period of time is then called the Dimensioning Accidental Load or DAL.
The ABS (2013) guidance notes on accidental load analysis and design for offshore structures provide
a design philosophy and hazard evaluation overview that considers four specific, non-environmental,
hazards that need consideration in the design and operation of offshore structures and facilities. These
include: i) ship collision hazards, ii) dropped object hazards, iii) fire hazards, and iv) blast hazards. In this
context, ABS defines accidents to be: unintended events that arise during the course of installing,
operating, or decommissioning an offshore oil and gas facility. ABS describes the process for assessing
accidental loads as one that follows traditional risk and structural assessment methods. Low risk and
likelihood events, based on the owners risk tolerance are removed and a refined assessment of the
remaining accidents is undertaken to determine appropriate mitigation. They indicate that the structural
performance criteria for each accident type need to be defined by the owners in keeping with
requirements specified by regulatory, classification and corporate entities. These need to include the basis
for localized structural member failure, such as yielding, buckling, plastic hinges, excessive deformation
and connection failures; a global structural failure including hull girder collapse for SSFF, damaged
stability and topside structural collapse; and consideration of safety critical elements including fire/blast
containment walls, escape routes and muster areas and containment equipment failures.
Standard methods for identification of accident hazards are presented by API (2007), IMO (2002)
Appendix 3, and ABS (2013). The guidance notes describe: what-if methods that use brainstorming
techniques to postulate potential mishaps and ensure that appropriate safeguards are in place; Hazard and
operability analysis (HAZOP) methods that use a systematic process to ensure that system design intent
has appropriate performance safeguards; failure modes and effects analyses (FMEA) methods that use
inductive reasoning to assess each system component failure mode and safeguard its effect on system
performance; fault tree analysis (FTA) methods that use deductive analysis techniques to establish the
relationship and consequences between human error, equipment failure and external events; and event
tree analysis (ETA) methods using inductive analysis techniques to model possible outcomes of and event
using decision trees to define multiple safeguards for events such as explosions and toxic releases.
Hazard curves like that shown in Figure 4 are used to describe the variation of hazard (loading)
magnitude with the return period. The inverse of the return period (years) is referred to as the annual
probability of exceedance. The 100 and 10,000 year returns correspond to the ISO-19900 (2013) ULS and
abnormal ALS design events respectively. The slope of the curve presents the ratio between the 10,000
and 100 year return period. Accordingly, it provides a relative measure of the difficulty in reducing a
given risk, in that the greater the slope the more challenging the hazard.
Figure 4. Hazard Curve (OGP-486, 2014).
532
In Figure 5, the slope for sea ice ranges between 1.4 and 1.5, wave actions between 1.5 and 1.9 and
earthquakes ranging between 2.5 and 5. Iceberg phenomenon is grouped with earthquakes. The variation
in the curves arises from regional differences, with northern North Sea curves typically forming the upper
bounds compared to southern North Sea locations. Correspondingly, the hazards from earthquakes and
icebergs are more challenging than wave loads, which in turn are more challenging than sea ice activity.
4.
4.1
SAFETY LEVELS IN ALS DESIGN

Introduction
Safety levels in ALS design are normally not given specifically. It is rather a risk related design approach
based on rare events that is applied. Such risk based procedures are structure specific since ships and
other floating installations, in addition to structural strength requirements, must fulfill requirements with
respect to stability and capsizing. The structure must be designed so that it is robust in relation to the
considered accidental actions by allowing local damage but preventing the collapse of the whole
structure.
Additionally, ALS design must ensure that the installation is able to maintain structural integrity for a
period of time under specified (usually reduced) environmental conditions sufficient to enable evacuation
of personnel and avoid pollution.
4.2
Safety level of offshore structures in ALS
4.2.1 General
The load and resistance factor design codes, at least for offshore structures, specify limit states such as
ULS, FLS, SLS and ALS to be considered in the design process. The safety level in such codes is
normally calibrated by structural reliability analysis (SRA). Such calibrations are associated with a
notional probability of failure. However, it should be noted that these formal probabilities of failure
neither consider the consequences of accidents, nor provide a true measure of the actual probability of
failure.
In more general terms accidental actions are associated with hazards due to errors in operation or
system failure. They are typically influenced by human involvement in their cause and dominate the
recorded risk picture of correctly designed offshore structures. A rational design against accidental
hazards must involve and be based on quantitative risk analysis performed for the installation.
4.2.2
Discussion of new ISO standards for offshore structures
The new ISO standards for offshore structures, see e.g. ISO-19902 (2007) and ISO-19906 (2010), offer a
practical implementation of the design approach against accidental and abnormal actions through the
identification of relevant hazards and subsequent design using ALS criteria.
When using either the partial factor design format or a nonlinear pushover analysis, the designer
quantifies the structures ability to resist overload relatively to a chosen reference action, which
depends on the limit state considered. In ISO-19902 (2007) the reference action is associated with
either the 100-year environmental situation, or with accidental design situations with a
significantly lower probability of occurrence. The first case is the regular design situation, and the
second case represents accidental design situations.
All structures and their structural components must be designed to satisfy particular limit states. Each
limit state is verified by defining a number of design situations, and requiring that the associated action
effects shall meet given design criteria. It is noted that the acceptance criteria are met by demonstrating
compliance with specified design procedures, not via a direct comparison with a target probability of
failure.
Design situations are classified into three categories:
1. persistent situations, with a duration similar to the design life of the structure;
2. transient situations, with a much shorter duration and varying levels of intensity;
3. accidental situations, which are of short duration and low probability of occurrence.
Persistent and transient situations are typically considered in the regular ULS design situation using an
environmental action with a return period of 100 years. Accidental situations consider hazards with a
return period of 10 000 years in the ALS design situation.
533
4.2.3 Characterization of hazards

Accidental situations are treated by considering hazards. A hazard is any potential for human injury,
damage to the environment, damage to property, or a combination of these. Using such a broad definition
of a hazard, several of the usual hazards to a platform (e.g. extreme storms) are treated in the regular
design situation, while hazards considered in accidental design situations are abnormal and accidental
situations with a low, but not neglectable probability of occurrence.
In ISO-19902 (2007) hazards are categorized based on probability of occurrence and are divided
into three main groups.
Group 1: hazards with a probability of occurring or being exceeded of the order of 10-2 per annum (return
period of the order of 100 years);
Group 2: hazards with a 10 to 100 times lower probability of occurring or being exceeded, i.e.
probabilities of the order of 10-3 to 10-4 per annum (return periods of the order of 1000 to 10 000
years);
Group 3: hazards with a probability of occurring or being exceeded markedly lower than 10-4 per annum
(return periods well in excess of 10 000 years).
Designing for Group 1 hazards is normally treated by the regular design process and is incorporated in
the verification of ULS limit states for persistent and transient design situations.
Other hazards belonging to Group 1 and not treated by the regular design process, as well as hazards
belonging to Group 2, are addressed by a requirement that the structure shall satisfy particular accidental
limit state (ALS) requirements.
Hazards falling into Group 3 are sometimes referred to as residual accidents and may normally be
neglected in design.
4.2.4
Accidental design situations
ISO-19902 (2007) requires that a structure shall incorporate robustness through consideration of the
effects of all hazards and their probabilities of occurrence to ensure that consequent damage is not
disproportional to the cause. The intention of the associated ALS limit states is to ensure that the structure
can tolerate specified accidental situations and, if damage occurs, that it subsequently maintains structural
integrity for a sufficient period under specified environmental conditions to enable evacuation to take
place. Accidental situations relate to two types of hazards:
Hazards associated with identified accidental events. These hazards belong to group 1 or group 2 and are
not included in the regular design process.
Hazards associated with abnormal environmental actions. These hazards can occur due to very rare and
abnormally severe environmental conditions. They correspond to a significantly longer return period
than the ordinary design actions.
The two types of hazards are different by nature. In principle accidental events can in some cases be
avoided by taking appropriate measures to eliminate the source of the event or by bypassing and
overcoming its structural effects. In contrast to this, the possible occurrence of abnormal actions cannot
be influenced by taking such measures.
An accidental design situation is considered in an accidental limit state (ALS) and normally comprises
the occurrence of an identified accidental event or an abnormal environmental action, in combination with
expected concurrent operating conditions and associated persistent and transient actions.
In ISO-19902 (2007) the regular design situation (ULS) for an L1 platform corresponds to an annual
probability of 3*10-5 based on the reliability model given in Efthymiou et al. (1997), but ISO 19902 gives
no direction for the determination of probabilities of failure associated with accidental situations.
After the issue of ISO-19902 (2007) the standard for arctic structures ISO-19906 (2010) was issued.
This standard deals with abnormal ice action and calibrated the regular ULS design situation and the
abnormal ALS ice action to the same safety level (10-5/year) based on SRA approaches (OGP-434-16,
2010). They concluded that the safety level thus determined is of the same order of magnitude as the one
used in ISO-19902 (2007).
4.2.5 ALS safety levels implied in structural codes

4.2.5.1 General
In a rational ALS criterion the accidental action should be defined as a characteristic value preferably
defined in probabilistic terms. This has been done both in ISO-19902 (2007), ISO-19906 (2010) and in
534
NORSOK-N-003 (2007), where the characteristic accidental action for offshore structures is specified by
an annual exceedance probability of 10-4.
The ALS criterion also applies to abnormal environmental conditions such as hazards associated with
abnormal environmental actions, e.g. wave actions. In this connection focus should also be given to
abnormal waves with high crest or unusual shape especially in such cases where the 10-2 wave might
not reach the platform deck, but the 10-4 wave crest hits the deck and causes a significant increase in the
wave loading.
4.2.5.2
Introduction of risk based acceptance criteria
The probability of failure associated with the accidental action A can be estimated as:
PF = p( F | A) p( A)
(1)
where p(F | A) is the conditional probability of failure given A, p( A) is the probability of the accidental
action, p(F | A) is normally determined using either structural reliability analysis (SRA) or by Monte
Carlo simulation, while p(A) is determined using quantitative risk analysis (QRA).
In order to obtain a true reliability value by means of (1), both terms must be accurate reflections of
reality, but generally any deficiencies in the determination of the two terms p(A) and p(F | A) makes it
challenging to realistically estimate probabilities of failure for accidental actions.
Structural reliability analysis determines notional failure probabilities which can be quite different from
real failure probabilities, and which should only be accepted as real probabilities of failure if they are
calibrated to known conditions, and that is rarely the case. An alternative approach which has attracted
increasing interest in recent years is to determine p(F | A) by Monte Carlo simulation, as it may hold
potential to provide a more accurate estimate of p(F | A) , provided a huge number of simulations are
carried out. In addition, the determination of p(A) is no trivial task as it ideally must consider all relevant
failure causes and modes to closely reflect reality.
Therefore, (1) is at best believed to estimate an order of magnitude rather than a precise figure. Under
ideal conditions the probabilities of failure determined by (1) should match actuarial values of failure
collected over sufficiently long time of service for a given type of structure.
Ultimate failure consequences include fatalities, environmental damage and loss of assets, but most
regulatory regimes have their main focus on limiting fatalities, and annual failure probabilities are favoured
to ensure the same fatality risk of individuals at any time. Generally, it is accepted to treat different hazards
and failure modes separately and assume the total failure probability equal to the sum of the individual
probabilities.
Current offshore standards and regulatory regimes do not specify ALS acceptance criteria directly in
probabilistic terms, and only ISO-19906 (2010) specifies acceptance criteria for ALS abnormal
environmental action.
However, Moan (2009) provides some insight in the failure probability implied by the NORSOK-N001 (2010) ALS criteria. The main focus is on fatalities, and acceptance criteria are derived based on
acceptance of risk levels historically accepted by society. The risk model used, the FN diagram, was
originally developed by Whitman (1981) based on early accident data for offshore structures and ships.
The FN curve depicted below is based on more recent data. It compares experienced overall accident rates
with respect to fatalities in the offshore and shipping industries. Floating platforms are not included
because of limited experience with such platforms.
Figure 5. Frequency-fatality diagram (Whitman 2009).
535
In this diagram the horizontal axis represents the consequences in terms of fatalities while the vertical
axis represents the annual occurrence rate of N or more fatalities. This diagram shows that the annual
frequency of losses with 50-100 fatalities which could be considered as total losses is of the order of
6 10-5 for fixed (production) platforms and 10-3 for mobile units. Based on this data, the annual target
failure probability of structural collapse due to each accidental action was chosen to be 10-5 for a fixed
production platform.
Moan (2007) found that for structures designed to NORSOK requirements p(F | A) is of the order 0.1.
He thus concluded that the probability of failure due to an accidental event with an annual probability of
10-4 would be of the order 0.1 10-4 = 10-5 per year. The steps in this ALS design approach are illustrated
in Figure 6 below.
Figure 6. Norsok Accidental Limit State procedure (Moan, 2007).
DNV and ISO specify accidental and abnormal loads at an annual probability of 10-4, but are not giving
specific values for the annual probability of total loss.
4.3
Safety level of Ship Structures in ALS
4.3.1 General
Most of the past codes of ship structures have not been based entirely on the reliability analysis, neither
on accident limit state requirements. But there is still a safety level implied in these rules. Teixeira and
Soares (2010) discussed the reliability of intact and damaged ships and presented the relevant
considerations that should be addressed when formulating the reliability problem of damaged ships.
Particularly, the effect of the damage on the ship ultimate strength and the relevant changes on the still
water and wave induced loads following damage, including the environmental conditions and duration of
exposure that are applicable for the damaged ship were considered (Teixeira and Soares, 2010).
IMO proposed Goal Based Standard (GBS) requirements, meaning safety level requirements for
ship structure design, which should reach the target safety level.
4.3.2
GBS of Ship Structure Design
After the 89th council of IMO in 2002, the GBS was implemented in the marine safety field. Later, the
GBS framework with 5-tiers structures was presented by IMO and regarded as standard of codes.
Relating contents were included in previous ISSC reports.
The GBS has been developed in two parallel directions. One is safety level approach (SLA), which
performs risk evaluation on current ship safety standards using overall analysis and formulates the risk
acceptance criteria through formal safety assessment (FSA). The other is called deterministic approach,
and formulates the quantitative functional requirements based on abundant practice experience on bulk
536
carriers and crude oil tankers. The 87th MSC adopted two documents that are important to the
development of GBS: International goal-based new ship construction standards for bulk carriers and oil
tankers and Goal-based standards compliance validation guidelines. The former includes the general
goals and 15 functional requirements of ship construction, corresponding to the first and second tier of the
5-tier structured GBS; while the latter explains validation standards and program of the 15 requirements
in detail, corresponding to the third tier. The adoption of the two documents symbolized the breakthrough
of the deterministic approach, indicating that GBS is operable in major ship domain. However, the risk
level of requirements in GBS hasnt been evaluated and quantified, which means that the risk level of
ship satisfying GBS is still unknown. Thus the current GBS is only based on historical experience instead
of risk level.
In order to find sensible risk acceptance criteria and implement them into the GBS system,
quantification of the risk level of current specification system needs to be accomplished via FSA. While
this approach goes hand in hand with the development of FSA and historical statistics, due to the
limitation of these two factors, the development of SLA will take a long time.
Following the implementation of GBS, IACS has identified gaps in the Common Structural Rules
(CSR) for bulk carriers and tankers analysis that need to be mitigated. Recently, Harmonized Common
Structural Rules (CSR-H) deal with the residual strength in this context, as no requirement to residual
strength is given in the CSR. Some societies have their own criteria, e.g. (DNV-OS-A101, 2014) and
(ABS, 1995b, ABS, 1995a), and these may be considered as potential requirements to be implemented in
the CSR.
The residual strength after collision or grounding, including flooding, should be adequately covered by
rule criteria, and in compliance with GBS.
4.3.3
Safety level in ULS in CSR
The hull girder ultimate capacity check is categorized as an ultimate limit state. The ultimate strength
criterion is given in a partial safety factor format, and has been calibrated using structural reliability
analysis techniques.
In the CSR, partial safety factors are calibrated for different target probability levels. A target level
based on existing structures should be somewhere between 10-3 and 10-4.A target level based on tabulated
values, such as those used in DNV Classification Note 30.6 (DNV, 1992) as included in Table 1, indicates
a stricter target level. In the CSR, the real failure probability of hull girder ultimate capacity for oil
tankers is 10-3 (IACS, 2006).
Table 1. Acceptable Annual Probability of Failure (DNV, 1992).
Consequence of failure
Less serious
Serious
I-Redundant structure
10-3
10-4
II-Significant warning before the occurrence of failure in
10-5
10-4
a non-redundant structure
III-No warning before the occurrence of failure in a non10-6
10-5
redundant structure
Class of failure
4.3.4
Safety level in ALS in CSR-H
In CSR-H, the residual hull girder ultimate criterion is given in a partial safety factor format as well. It
has also been calibrated using structural reliability analysis techniques. A comparative approach
illustrated in Figure 7 was used. The ship following a collision or grounding damage should have the
same safety level as intact ship for hull girder strength (Figure 7).
Two steps should be carried out to compare the safety level between intact ship and damaged ship
(IACS, 2014):
(a) If the failure probability of the damaged case turns out to be lower than for the intact case, this
would be an indication that a residual strength criterion is unnecessary. If a higher failure
probability is obtained for the damaged case, this would indicate the need for a residual strength
criterion.
(b) Irrespective of the results from (a), it is decided to have a residual strength criterion in the rules. The
evaluations as illustrated and described in Figure 7 will be made to support the proposal of such a
criterion. The intention is to make the residual strength criterion so that it corresponds to the same
target reliability level as the intact criterion.
537
Figure 7. Comparative approach, structural reliability, intact and damaged condition (IACS, 2014).
Figure 8. Principle for the development of a residual strength criterion (IACS, 2014).
IACS (2014) had made the comparison of reliabilities between the intact hull structure and the damaged
hull structure. The hull girder strength of the bulk carrier was first adjusted to accurately meet the
ultimate strength criterion in the rules. Then a structural reliability analysis, considering failure of the hull
girder of an intact ship in North Atlantic environmental conditions, was performed. The failure
probability could be calculated, following the same methodology as used for the development of the hull
girder ultimate strength criterion for tankers (IACS, 2006). Thereafter, the probability of failure following
collision was calculated for the same ship. Compared to the intact case, the following key items were
accounted for:
Annual probability of damage (collision)

Probability of damage size, in the case of collision
Reduction in capacity as a function of damage size
Increase in still water bending moment due to damage
Environmental conditions when the damage occurs
Exposure time to environment, before rescue to shore
The results in (IACS, 2014) show that the annual probability of failure is 810-4 for the intact case.
This compares well with the results obtained for tankers (IACS, 2006). It is seen that the failure
probability during three months in the so-called collision environment is higher (510-3) when calculated
conditional on the event of collision. Assuming that the annual probability of a collision event is 0.01, the
538
annual probability of hull girder failure following collision damage is then 510-5. This is lower than the
failure probability for the intact case. Now the residual strength criterion will be considered in isolation,
and adjusted to meet the same target reliability level of 810-4 from the intact case. This is done by
redesigning the hull girder as illustrated in Figure 9, and is implemented by reducing the capacity until the
target failure probability is obtained.
5.
5.1
ASSESSMENT OF ACCIDENTAL LOADS

Introduction
The subject of this chapter is the assessment of accidental loads which are implemented in the limit state
based structural design. It is not only the calculation procedures of accidental loads but also determination
procedures of the design accidental loads that are described in this chapter.
The characteristics of accidental loads and various methods to calculate them, that is, theoretical or
empirical formulae, FEA (Finite Element Analysis), CFD (Computational Fluid Mechanics), and so on,
were well described in previous ISSC reports. Collision and grounding accidents for ship structures are
found in ISSC (2006), and explosion, fire, dropped object and collision accidents for offshore structures
in ISSC (2012). Therefore this report focuses on the research and new findings in the field of accidental
loads presented in recent literature.
In practical Accidental Limit State (ALS) design as well as other limit state designs, the design loads
need to be determined methodically and effectively, taking into account the degree of accuracy and
economic feasibility, as well as safety inherent in the each design phase. It is impossible to guarantee
absolutely safe structures at the design stage because of numerous uncertainties, such as loadings,
materials, structural behavior or human and environmental factors. An accident is usually a chain of
unpredictable events and often a multi-physics phenomenon. The challenges, however, do not free
designers from the responsibility of minimizing potential risks. Therefore, ALS design requires the design
concepts based on risk assessment and probabilistic approach.
Recently, the risk based probabilistic assessment for accidental loads is increasingly being used and
continuously developed, especially in the design of offshore facilities. That is, the state-of-the-art concept
in determining design accidental loads is the probabilistic approach rather than the deterministic one
which is usually based on the worst-case scenario. Therefore, the design standards and rules which are
practically used in the ALS design of offshore are summarized in this chapter. Additionally, the generally
accepted practices of industries for risk based or probabilistic determination of design loads are
introduced. Identification of accidental situations or actions to be considered in the ALS design can be
found in the relevant design standards and rules. Collision, dropped objects, explosion and fire accidents
for offshore structures are clearly mentioned as accidental design cases in ISO, API, NORSOK standard
or offshore codes from class societies (DNV, ABS, LR, BV, etc.). The specific description of accidental
actions for ship structures is hardly found in the ship rules of class societies. Therefore, only the above
accidental actions are considered in this chapter.
5.2
Explosion Load Assessment
The safety assessment of offshore facilities against explosion accidents should be performed in the design
phase to prevent loss of lives or catastrophic failure of structures, and be preceded by the process of
determining the design accidental load for explosion.
The design explosion load can be derived by the worst-case scenario, and the results are often too large
to be accommodated by the facilities. Therefore, some reasonable and practicable level of loads to assist
in determining design load is used. The Dimensional Accidental Load (DAL) which is usually established
as the load that occurs with an annual exceedance frequency of 10-4 , as well as two levels of explosion
loading, that is Ductility Level Blast and Strength Level Blast (DLB and SLB) loads were suggested by
some regulation authorities (API, 2006, Oil&GasUK, 2007, NORSOK-Z-013, 2010). These are now
widely used in offshore projects and well explained in ISSC (2012).
There are some references for the derivation of accidental loads. For instance, DNV-OS-A101 (2014)
defines generic design loads for some accidental cases, and API (2006) refers to nominal loads and
empirical formulas for fire and explosion. NORSOK-Z-013 (2010) recommends the probabilistic
approach with CFD simulation for explosion, and similarly FABIG (2014) recommends the probabilistic
CFD simulation for fire. Oil&GasUK (2007) and FABIG (1992) including its Technical Notes are an
overall guidance for explosion and fire engineering. However, in offshore projects client requirements or
specification of the facilities and best practices of industry developed from engineering works are
commonly used.
Explosion loading scenarios are numerous and specific to the facility or calculation model, and usually
developed on the base of risk analysis which is already described in section 2, 3, and 4 of this report. The
539
explosion load can be calculated by the nominal load method, simplified method based on theoretical
formulas, and numerical simulation.
5.2.1
Deterministic approach
The easiest way to get the design accidental loads is to bring the prescriptive loads by referring to the
relevant rules, standards or industrial guidance. The nominal load is a space averaged, peak explosion
overpressure for typical concept types of installations, and based on limited design data set and operating
experiences (API, 2006, DNV-OS-A101, 2014). Therefore, it is often a conservative value and used in the
early project phase.
In the case of simplified calculation models for explosion loads, there are some empirical models such
as TNT method for the high explosive and Multi-energy methods (Lees, 1996) or B-S-T model (Tang and
Baker, 1999) for vapor cloud explosion. These are based on correlations with experimental data, and
usually used to predict far field blast effects. Another simplified calculation model is the
phenomenological method, which is based on a physical process more than the empirical model and has
better accuracy with consideration of actual geometry using a simplified system. However, these
simplified models are gradually substituted by the numerical simulation mentioned below and
occasionally used as means of assistance.
The numerical simulation model for explosion load assessment requires consideration of likely sources
and magnitudes of leaks, ignition and consequent explosion development. These are presently well
addressed by CFD which is the most fundamentally based method and has the best potential for accurate
prediction of gas explosion phenomenon. These tools solve the conservation equations of mass,
momentum and energy including turbulence and combustion. A comparison of computational methods is
presented by (Ledin, 2002) and includes extensive literature references and theoretical algorithms to
describe gas explosions. The ventilation and dispersion analysis is needed to calculate the location and
size of the gas cloud. The overpressure depends on the type of gas, the volume and concentration of the
gas cloud, ignition source type and location, etc. The geometric details including process equipment,
piping or vessels are modelled exactly as far as possible, because the explosion load is also affected by
the degree of congestion and confinement of layout and geometrical shapes of the impacted structures.
Therefore, the processing and building methods of 3D CAD data or CFD mesh for various objects and
situations are also needed.
5.2.2
Probabilistic approach
The explosion accident is a complex phenomenon derived by a number of random variables, which have
many uncertainties. Figure 9 shows steps in explosion modelling and intermediate results. The design
random variables considered in calculating explosion loads are gas leakage rate and direction; wind speed
and direction; locations of ignition; gas clouds size, location and concentration.
Establish
leakagescenarios
Freq.
Kg/s
Dispersion,Early ignition
Establishcloud
sizedistribution
Freq.
Kg/s
Varycloudlocation,Ignition,
PerformCFDexplosionsimulations
Establish
explosionloads
Freq.
P max
Accidentallimitstatecriteria,
Personnelacceptancecriteria,etc.
Establish
consequences
Freq.
Cons.
Exceedance freq.,Farvalue,
Impariment freq.,ect.
Presentriskpicture
Figure 9. Schematics of procedure for calculation of explosion risk (NORSOK-Z-013, 2010).
540
The distribution of explosion loads on the target facilities is defined based on numerous explosion
scenarios using sophisticated analysis methods, like CFD. However, the cost of simulating all of possible
scenarios is very high. The probabilistic approaches in the present industrial practices were presented by
Czujko (2001):
Simulating a very high number of scenarios using simple models (e.g. PHAST/DNV),
Performing a low number of CFD simulations and extrapolating the results based on previous project
data
Semi-probabilistic and risk-based approach implemented into an industry standard (NORSOK-Z-013,
2010) where accidental scenarios for gas explosion analysis are selected based on frequencies of
parameters affecting scenarios. Resulting probabilistic models of explosion loads, i.e. exceedance
curves, are derived from frequencies of scenarios implemented. This approach brings more accurate
CFD calculations into risk analysis in a consistent and transparent way. The order of 300 ~ 400 CFD
simulations is typically performed in this approach.
Generic probabilistic approach where a selection of scenarios for explosion analysis is based on a
probabilistic procedure, and a probabilistic model of explosion loads, i.e. exceedance surface, is
developed based on numerical and probabilistic simulations.
While faster results are acquired in the first two approaches above, very conservative estimates for the
risk as well as wrong trends with regard to sensitivity studies can be expected. The third approach, risk
based probabilistic explosion loads assessment, is still under development with better understanding and
modelling of the physics.
For the last approach, the overall concept (Czujko, 2001) and a few remarkable studies have recently
been presented (Czujko and Paik, 2010, Paik and Czujko, 2012) However, this method has not yet been
introduced in the industry and is under development, especially as far as the exceedance surface is
concerned. The explosion loads are expressed not by a singular value but two components: explosion
peak pressure and its duration (or impulse). In the case with two load-parameters, design values may be
obtained from the joint probability distribution by contour curves. That is, the design load shall be
established based on the probabilistic relation between the predicted explosion pressure and its duration
from numerous scenarios (NORSOK-N-003, 2007, NORSOK-Z-013, 2010, LR, 2014c). Moreover, each
load component has a great effect on structural safety. When one component of an explosion load is
changed, the dynamic response of a structure under explosion loads shows inconsistent tendency (Biggs,
1964). Therefore, selection of the adequate duration corresponding to the defined pressure is an important
issue in determination of design explosion loads.
In principle, the probabilistic explosion load distribution should be presented as a frequency
distribution of overpressure and impulse, i.e., a P-I surface in a frequency space. Because of the two
parameters (P and I) it is not possible to cut the P-I surface in a unique way such that a unique loadexceedance curve can be established as long as the design has not been finalised and the response
characteristics of the structure are unknown (not decided upon). However, for a quasi-static or impulsive
response case the P-I surface can be simplified to an overpressure-frequency or an impulse-frequency
relation respectively. For the quasi-static case a simple overpressure exceedance curve can thus be
established where the DAL pressure is determined by the risk acceptance (i.e., frequency cut-off)
criterion. The associated range of impulse (or duration of overpressure pulse) for the DAL pressure can
then be found from the scenarios representing the DAL pressure. For use in the impulsive or dynamic
domain of the structural response the DAL can be presented as a triangular pressure pulse with a duration
representative for the cut-off frequency. The duration can typically be defined by the 10 per cent
overpressure points before and after the pressure maximum.
The explosion loads of all scenarios can be listed in descending order and the value whose cumulative
exceedance frequency corresponds to risk acceptance criteria, i.e. 1.0 10-4/year is selected as the
dimensioning accidental load. However, in the present industrial practices for offshore explosion
analyses, pressure and duration (or impulse) are considered separately and not treated as a combined term
when design explosion loads are determined with given acceptance criteria (Figure 10).
Kim and Kim (2014) proposed a method to better integrate overpressure and impulse using an
exceedance surface approach. The expected explosion response from a series of 8 random variables
including wind and leakage data that determine the gas cloud, its concentration, position and size were
used to combine overpressure and impulse (P-I) parameters into a joint probability density function and
thereby provide a more realistic definition of the explosion load criteria (Figure 9). Applying the general
risk acceptance criteria for the exceedance frequency of 1 10-4/year allows developing sets of pressure
and impulse values as a curve shown in Figure 10. Within the load limits by asymptotic points, the
response frequency of the structure, T can be then used to determine one of design points rather than the
greater loads suggested by the P-t and dual curve methods.
(a) Dual curve method
(b) P-t relationship curve method

Figure 10. Determination of dimensioning explosion load with exceedance curves (Kim and Kim, 2014).
Figure 11. Exceedance frequency surface for design explosion load (Kim and Kim, 2014).
Figure 12. P-I design load curve from exceedance surface (Kim and Kim, 2014).
541
542
5.2.3

Definition of explosion loads for design
The design explosion pressures for a topside structure calculated using CFD analyses are expressed as
representative values for each area of the topside. However, depending on the size and shape of structures
and equipment in the area, these explosion pressures generate different types of loads, that is,
overpressure, drag pressure and differential pressure. Their detailed definition or implicated structural
objects are well described in ISSC (2012). An example application of these loads is shown below, and the
coefficients can be different in real projects.
Blast load for Structural members (wall, deck, beam, column, panel)
Overpressure and drag load
Blast load for Piping
Drag load, FD = Dynamic pressure Drag coefficient (Cd = 1.2) = A v2 Cd
where, is the fluid density, A is the maximum cross sectional area of the object in a plane normal to
v, and v is the large scale fluid velocity ignoring spatial fluctuations in the vicinity of the object.
- or Drag load, FD = Local overpressure 1/3
Blast load for Vessels
- Pressure differential load, FP = P section area PDF
where, PDF is pressure distribution factor(for cylinders, 2/) and P is the differential pressure across the
obstacle.
Blast load for Grating floor

Drag load, FG = Dynamic pressure Drag coefficient (Cd =2.0) Permeability factor
Blast load for large items
Reflection load, FR = Overpressure Reflection coefficients (2.0 for front wall)
5.3
Fire Load Assessment
The determination of the rational design loads for fire has the highest priority for coherent and reasonable
evaluation of structural redundancy and life safety under serious thermal loads. A fire accident is also
caused by combined effects of many random variables, which have high uncertainties and affect the
incidence and progress of a fire accident. Therefore, the probabilistic approach is needed instead of the
deterministic approach. The design fire load satisfying given acceptance risk criteria shall be determined
in terms of heat flux and its corresponding duration because the structural response is highly influenced
by both of the two load components. However, at present, there is no practical way to deal with two load
components at once in the offshore fire design. Treating heat flux and duration independently could result
in a mistaken prediction due to the omission of correlation existing between them.
5.3.1
Fires present a major hazard to offshore platforms. Typical heat load limits are reported by (Moan et al.,
2002), NOSOK S-001 and FABIG to generate in the order of 200 to 300 kW/m2 and last between 15
minutes and two hours. DNV-OS-A101 rules also define requirements for critical items in liquid and gas
collection areas and specify that fire load requirements for structures which have critical items shall be to
withstand a jet fire (250 kW/m2) for 30 minutes and a pool fire (150 kW/m2) for the following 30
minutes. In areas with only oil or condensate containing equipment, critical items shall be designed to
withstand a pool fire (150 kW/m2) for 60 minutes. Areas with only gas containing equipment shall have
critical items designed to withstand a jet fire (250 kW/m2) for 30 minutes. Additional classical algorithm
based fire spread modelling methods that determine probabilities, fire growth, temperature time histories
and correlations for full room involvement in marine structures are reported by Sprague and Dolph
(1996).
The nominal fire loads similar to explosion case were suggested in various documents (API, 2006,
NORSOK-S-001, 2008, DNV-OS-A101, 2014, FABIG, 2014). In these documents, tabulated guidance
for each fire type (pool fire & jet fire) are provided including their typical properties like flame size, heat
flux, temperature, etc. In the case of numerical models for consequence of a fire load, PHAST software is
widely used with 2D simplified model and KFX or FLACS software can be used with 3D CFD model.
5.3.2
543
Risk-based and probabilistic approach
The probabilistic fire load for offshore facilities is usually determined in the process of fire risk
assessment. The scope of the fire risk analysis is to identify potential major fire hazards on the
installation. The specific objectives are to:
Identify the fire scenarios resulting from process or non-process failure cases
Evaluate the frequency of fire scenarios
Quantify the consequence of the identified scenarios (flame length, pool diameter, heat flux contours)
Identify the targets vulnerable to the fire scenarios and evaluate their consequences
Identify the escalation potential of each accidental event and the possible damage to the asset.
Identify the possible requirement for additional risk reduction measures to help prevent and/or
mitigate the effects of the identified fire scenarios (safety distance of the restricted Area, Passive Fire
Protection, fire wall requirement, etc.)
The topside process segments are divided into isolatable segments bounded by Emergency Shutdown
Valves (ESDV), and these are further broken down based on the operating conditions, compositions,
phases and location etc. In the context of the fire risk analysis, the ESDVs and SDVs are crucial to ensure
a safe isolation and shutdown of the process system in the event of an emergency (i.e. detection of
hydrocarbon release or fire).
The leak failure frequency of each isolatable segment is based on historical data and derived by
counting the associated equipment and fittings e.g. vessels, heat exchangers, valves, flanges, instruments
and piping, etc. The fire frequency from a leak event is determined considering the probabilities of
ignition, early detection, isolation and blowdown.
The consequence of the fire event can be modelled by 2D simplified method (e.g. PHAST) or 3D CFD
method (e.g. KFX) and expressed with flame length, flame diameter, heat flux contour or impairment
frequency, etc. These loads are assessed against design criteria. The design fire load for the target can be
determined by comparing the fire loads with the likely fire resistance of the potential targets in order to
assess the likely frequency of credible damage scenarios.
For probabilistic fire modelling, it is needed to establish dimensioning fire loads. The fire load have
two parameters, that is, incident heat flux and duration, and have strong spacial and time dependency.
One problem is that there is no unique and unambiguous way to derive dimensioning fire load
considering these two load parameters ((NORSOK-N-003, 2007, LR, 2014b).
Recently, semi-probabilistic and risk based approach for fire design was suggested (FABIG, 2014) as
the explosion case. In this procedure, 110-4/year DAL dose and dimensing scenario are introduced from
exceedance curve of heat dose. however, this heat dose must be converted back to a flux and duration,
and therefore, there is still some weakness that the time history profile of the fire load or correlation
between heat flux and duration is not exactly expressed. Huser and Vianna (2010) and Huser (2013)
outlined a probabilistic procedure for fire loads based on the same principles used by NORSOK-Z-013
(2010) for explosion analysis, and also explaned the dimensioning heat dose application of FABIG (2014)
as shown in Figure 12. Paik and Czujko (2012) showed the probabilistic procedure for fire loads with
exceedance curves of both temperature and heat dose (Figure 13). Kim and Kim (2015) propose a method
to better integrate heat flux and duration with maintaing the time history shape of fire loads by using joint
probability distributions and exceedance surfaces among heat flux, heat dose and leak duration similar to
their previous work(Kim and Kim, 2014) for explosion loads (Figure 13).
Figure 13. The exceedance curve of heat dose and DAL fire scenario Huser (2013).
544
Figure 14. The exceedance curve of temperature and heat dose (Paik and Czujko, 2012).
Figure 15. The joint probability distribution for heat flux, heat dose and leak duration (Kim and Kim, 2015).
5.4
Load Assessment for Collision accidents
A ship collision with an offshore structure takes place when a support, off-load or trading vessel lose their
station keeping ability for various reasons and collide with the offshore facility. Ensuing damage to both
structure and ship include large deformation of primary structures as well as cracked joints, tears in
plating and ruptures of various tanks or other stability control systems. The preliminary assessment
defines the acceptance criteria, the collision events, and collision risk. The detailed assessment selects
more specific pertinent preliminary scenarios and identifies mitigation measures to meet the criteria and
re-assess the event using less conservative analysis techniques to evaluate the local and global expected
performance and re-evaluate the risk to determine whether additional mitigation is required.
Departing from the conventional structural design concept where functional loads are mainly
concerned, considering the accidental loads in design phases is getting increased to reflect its serious
impact to determine the required structural integrity. Collision accidents are one of the important potential
hazards, to which ships and offshore structures are easily exposed. Therefore, the assessment of their
consequences to ultimately minimize or prevent potential damage should be performed and it is preceded
by defining design collision load. Generally, the collision design load is determined by deterministic or
probabilistic approach.
One of the practical ways to determine design collision loads is using design codes produced by the
various regulatory authorities or classification societies. However, no standards are universally accepted
and requirements of specification are different for each offshore project. Therefore, procedures to
determine design loads should be properly selected project by project. Another deterministic approach is
based on a few scenarios, which are the worst-case type or the cases of some typical collision accidents.
In this way, analysis method of external mechanics is employed to estimate the amount of kinetic energy
available to be dissipated to deformation energy during collision.
Over the past decades, risk assessment methodologies for ship and offshore collision problems have
increasingly been applied and more realistic design collision loads have been determined reducing the
conservatism of the deterministic approach. In this approach, the frequency and consequence of each
collision scenarios are determined. The frequency of the collision is related to the visiting or passing
frequency of vessels which sail or operate near the field where the installation is located and the
545
probabilities of colliding conditions (i.e. approaching angle and speed, wave and wind conditions, etc.).
The consequence of each accident is determined in terms of impact energy by the external mechanics.
The results of the risk based approach can be explained by employing the risk matrix whose row and
column represent the collision frequency and consequence respectively, or the exceedance frequency
curve of collision impact energy.
5.4.1
The process involved in collision is very complicated and many predictive calculation procedures have
been developed for predicting collision impact energy and its response. Collision mechanics can be
classified into two parts, namely external mechanics and internal mechanics (ISSC, 2006, ISSC, 2012).
Those two parts are often treated separately, but in some cases, they are solved together.
The first point of reference for the designer is the codes of practice produced by the various Regulatory
Authorities and/or Classification Societies. The codes for collision accidents follow approximately the
same approach that is the assessment of impact energy to be absorbed by the struck objects. Collision
with 5,0007,000 ton supply boats with speed of 0.52 m/s has been investigated as the most probabilistic
event. Thus, the codes are explicit with this type of collision. A tabulated summary of the codes was
given in the HSE report by MSL-Engineering (2000), and the overview of existing guidance on collision
assessment is introduced in LR (2014a). In practical offshore projects, projects specification states the
design value and acceptance criteria in detail and these are based on the experience of previously-built
installations, relevant rules, regulations and the results of engineering research.
5.4.2
Risk-based and probabilistic approach
The objectives of collision risk analysis are to identify potential ship collision scenarios, evaluate the
likelihood of their occurrence, determine their consequences and impact energy, and to assess the risk of
ship collision. Therefore, design accidental loads for collision can be selected by assessing the impact
energy against the design acceptance criteria for risk or performance. The LR guidance notes for collision
analysis (LR, 2014a) provide an overview of the methods and standards to perform risk based
assessments of collision events.
Collision scenarios define the typical collision events and their probabilities to be considered in design.
Parameters needed for defining collision scenarios include the type and geometry of a colliding vessel,
colliding speeds, colliding angles, collision locations, cargo loading conditions, drafts and so on.
Colliding ships may be categorized into passing vessels (not related to the installation like shipping
traffic including merchant vessels, fishing vessels, passenger vessels, naval vessels, etc.) and visiting
vessels (to serve the installation like supply vessels, export tankers, and product tankers, etc.). Installation
dimensions and orientation, shipping data (traffic density, type and size), visibility, wave height, vessel
speed distribution, collision risk management system, i.e., standby vessel, etc. are factors influencing
collision frequency of passing vessels. The collision risks associated with visiting vessels are different
from passing vessel collisions, as they occur more frequently, and usually have lower energy impacts.
However, the relatively higher frequency of these low energy impacts can increase the cost of repairs over
the lifetime of the FPSO to a substantial level. It is also possible to differentiate ship collisions according
to whether or not the colliding vessel is under power at the time of the collision, i.e., Powered or Drifting.
Impact energy calculations for passing and visiting vessels are used to assess vulnerable areas of the
installation. The total impact energy of a colliding vessel with a static installation is equal to the kinetic
energy of the vessel immediately prior to impact, i.e. E = mv2, where E = colliding kinetic energy; m =
the effective mass of the colliding vessel, m = ms + dms; v = the velocity of the colliding vessel
immediately prior to impact; ms = displacement of the colliding vessel; dms = added mass of the colliding
vessel. For the rigorous expression refer to Pedersen and Zhang (1998). The added mass term is a factor
that takes account of the hydrodynamic forces that act on a vessel during a collision. This factor depends
upon a considerable number of parameters but the simple approximations are commonly adopted: dms =
0.1ms for a bow/stern on collision, dms = 0.4ms for a side-on collision. Some of this kinetic energy will be
absorbed in elastic/plastic deformation of the colliding ships structure and some will be absorbed in
elastic/plastic deformation of the structure. For glancing blows, the colliding ship may also have a
residual kinetic energy after the collision.
The impact energies of all scenarios can be listed in descending order and the value of which
cumulative exceedance frequency corresponds to risk acceptance criteria, i.e. 1.0 10-4/year is selected as
the dimensioning accidental load. This dimensioning load is assessed against design criteria. The design
impact load for the target can be determined by comparing the impact energies with the likely impact
resistance of the potential targets in order to assess the likely frequency of credible damage scenarios.
Youssef et al. (2014a) outlined methods to address hazard identification and probabilistic scenario
selection for ship collision accidents. Collision scenarios are then described using a set of parameters that
546
are treated individually as random variables and analyzed by statistical methods to define the ranges and
variability to formulate the probability density distribution for each scenario. As the consideration of all
scenarios would not be practical, a sampling technique is applied to select a certain number of prospective
collision scenarios. Applied examples for different types of vessels are presented to demonstrate the
applicability of the method. Youssef et al. (2014b) also reported a probabilistic risk assessement for shipship collisions, in which not only the total collsion risk but also the exceedance curves are established that
can be used to define the collision design loads in association with various design criteria (Figure 16).
Figure 16. Collision exceedance curve in terms of exceedance of collision frequency versus absorbed energy
(Youssef et al., 2014b).
5.5
Load Assessment for Dropped Object accidents
The dropped events generally arise during the transfer of cargoes by cranes, and these are rarely critical to
the global integrity of the installation and will typically cause local damage. However, these events may
bring about serious consequences such as the personnel fatalities or the system shutdowns, so it is
important to evaluate the structural strength to prevent these losses.
The process for assessing accidental loadings from dropped objects follows in principle the risk
assessment traditionally used by the offshore industry, in which expected dropped events are developed
with their likelihood and then critical scenarios and design impact energy is decided considering the
consequences of those events. The impact energy is characterized by its kinetic energy determined by its
mass including hydrodynamic added mass if it falls through water and its velocity. The ALS design is
performed for this impact energy and has to satisfy the design requirements. The kinetic energy has to be
dissipated as strain energy in the impacted component and possibly also in the dropped object itself.
Generally, this involves large plastic strains and significant damage to the impacted component, which
shall comply with ductility and stability requirements.
5.5.1
Its hard to find the specific references for the impact energy of dropped objects. Rules or standards show
only general requirements and can be summarized with two criteria: no penetration of impacted deck
structure and no safety critical systems shutdown due to large deformation. For example, the class rule
(BV, 2014) requires that the energy of the dropping object at the moment of the contact with the deck is
to be lower than the absorbed energy by the deck deformation at the limit state. The limit state
corresponds to the first rupture of a plate in the deformed deck area. NORSOK-N-004 (2004) states that
the masses of the dropped objects are normally based on operational hook actions of the platform crane
and critical areas of dropped objects are to be determined based on crane operation sectors, crane reach
and actual movement of actions. DNV-OS-A101 (2014) advises that the weights of the dropped objects to
be considered for design of the structure are normally taken as the operational hook loads in cranes. The
impact energy at sea level is normally not to be taken less than 5 MJ for cranes with maximum capacity
more than 30 tonnes. The impact energy below sea level is assumed to be equal to the energy at sea level.
In practical offshore projects, their specification sometimes states the minimum impact energy (usually
547
hundreds kJ) and acceptance criteria and these are based on the experience of previously-built
installations, relevant rules, regulations and the results of engineering research.
5.5.2
Risk-based approach
The purpose of dropped object risk analysis is to assess potential dropped object scenarios with frequency
and their consequences. The dropped object risk analysis includes:
A comprehensive register of all lifts including the type of object, size, weight, frequency, etc.
Identification of the hazardous areas and safety critical elements susceptible to be damaged due to
dropped object from mechanical handling activities
To assess the likely frequency of dropped objects onto potentially vulnerable targets, both on the
installation and overboard/subsea
Estimation of the frequency of dropped objects onto potentially vulnerable targets on the installation
or overboard
Assessment of the impact energies and of the effects of such impacts
Evaluation of the results against the risk acceptance criteria; when necessary proposition of mitigating
actions, including restricted areas for normal crane operations, to reduce the risks to as low as
reasonably practicable (ALARP).
Frequencies of dropped objects can be taken from the failure databases (OREDA, 2009, OGP-434-8,
2010). The subsea analysis is performed usually in accordance to the methodology provided in the
DNV-RP-F107 (2010). The impact energy is calculated according to the kinetic energy of the dropped
=
, where Ek = Kinetic energy; g = Gravitation acceleration; h = Height of the
object, i.e.
dropped object. The impact energies of all scenarios can be listed in descending order and the value of
which cumulative exceedance frequency corresponds to risk acceptance criteria, i.e. 1.0 10-4/year is
selected as the dimensioning accidental load. This dimensioning load is assessed against design criteria.
Figure 17 shows the typical exceedance probability distribution of impact energy for the dropped
objects.
Figure 17. Impact energy distribution of dropped objects on production installations, 2002~2001, NCS (Vinnem,
2014).
6.
6.1
DETERMINATION OF ACTION EFFECTS

Introduction
The determination of action effects is the next step in ALS process as a continuation after identifying
accidental actions. Actions are necessary to improve the design and further develop design standards and
safety limits. A successful determination of action effects is therefore necessary for the evaluation of
structural integrity either as a direct consequence of accidental actions, or as an escalation of the initial
event; the decision if the structure is compliant can then be made. A proper evaluation should include the
548
appropriate analytical model or software capable of capturing the action effects, and their non-linear
attributes. The determination of action effects may lead to the definition of appropriate metrics, or the
adoption of existing limiting criteria.
For the evaluation of action effects the analysed structure has to be represented with a mathematical
model including the description of geometry, material and loads. Depending on the problem this
mathematical model can be a closed form analytical equation, simplified beam model based on rigidplastic analysis, or a large numerical model consisting of shell, beam and solid elements suitable for nonlinear analysis. Simplified models often have limited usage and accuracy making them suitable for studies
where a large number of analyses are to be conducted. Typical examples, where such simplified models
are used are risk analysis studies for ship collision in a certain sea region, see for example Goerlandt et al.
(2012), Montewka et al. (2014) or Qu et al. (2011). Goerlandt et al. (2012) were applying simplified
consequence models to study the probability of oil outflow in tanker collisions in the Gulf of Finland.
They concluded that using more accurate collision energy and/or structural capacity models for purposes
of maritime traffic risk analysis is currently not worthwhile, as the uncertainty regarding impact scenario
is overwhelming- thus indicating the sufficiency of simplified tools. Another model focusing on the
simplified methods and collisions in the Gulf of Finland by Montewka et al. (2014) presents a
probabilistic model for collision accidents and successfully validates it with historical data. Qu et al.
(2011) studied the collision probabilities in the Singapore Strait and proposed a series of measures, such
as speed limits and passage guidelines, to reduce collision risks. Simplified consequence models for
collision can be based on the deformation mechanisms or on numerical simulations. Haris and Amdahl
(2013) and Gao et al. (2014) are examples of simplified models that sum the crushing resistance of
different structural components to form the collision response of the whole side structure. By combining
numerical and analytical approaches, Ehlers and Tabri (2012) presented a method that calibrates a semianalytical collision model based on single numerical simulation allowing for a rapid assessment of
different collision scenarios for a given ship.
Recent work on the subject of nonlinear structural response analysis of offshore installations under
explosive loads, with coupled load mapping between CFD solvers and nonlinear structural analysis is
presented in Paik et al. (2014). This work shows that rule-based uniform load application can result in an
overestimation of structural damage, and in other cases it may lead to an underestimation. The traditional
SDOF approach is also known for returning different nonlinear structural responses depending on the
blast load profiles (usually triangular load for dynamic structural analysis when considering hydrocarbon
explosion accidents). Simplification of the topside structure when assessing blast effects often overlooks
the uncertainties associated with load profile idealisations (triangular, rectangular, symmetric triangular,
or linearly decaying).
Dropped objects analysis is closely related to collision analysis following general guidance notes for
impact actions. Impact scenarios can be identified from risk assessment considering crane lifting capacity,
crane reach, and laydown area capacities, where lift frequencies and mass of lifted object are considered.
Acceptance criteria take several forms not only based on Class Rules but also on the operators
operational requirements. BV (2014) recommend safety principles dictating that structural elements may
suffer permanent deformations without any rupture. NORSOK N-004 Annexes A.4, K4.4.4, L.8.1 and
M.6.2 provide guidance with simplified methods for resistance/energy dissipation expressions and
selection of critical areas of dropped object analysis.
Requirements from the operators point of view may be specific in terms of impact energy on specific
areas on deck or equipment, particular laydown capacity on deck plating, or even deck plating designed to
be capable of withstanding explicit impact energy based on a standard ISO container from a specific
height with the assumption that a percentage of the overall impact energy is to be transferred to the
structure through elastoplastic deformation.
If no suitable simplified model is available or when a specific structure has to be studied at higher
detail, numerical models can be used. Numerical models are more suitable for treating material and
geometrical non-linearity, highly dynamic loads and the coupling between loads and structural response.
A numerical model aimed at establishing safety levels should be in appropriate balance between precision
and time efficiency, yet results should be benchmarked against established, well-documented and
accepted modelling and industry criteria. Academic studies usually provide more precise models that can
be used to prepare benchmarks for the validation of software and modelling approaches. The challenge is
in finding the right balance between sensible analysis time and results that couple continuous design
updates of hazard analysis with design iterations.
Several challenges in the determination, analysis and evaluation of action effects using
numerical models are discussed in this section with emphasis on uncertainties associated with ALS
conditions.
6.2
549
Review of Numerical Tools
The structural consequences of accidental loads, as identified in Section 2, may be assessed in various
ways; by energy considerations combined with simple elastic-plastic methods (DNV-RP-C204, 2010), or
by coupled advanced numerical methods. Accidental loads, and ultimately their possible escalation
into an accident disproportional to the initiating event can be addressed in the design stage as shown in
Annex 3. This section is focused on the most popular numerical tools used nowadays.
Software often used is based on linear elastic analysis, and it extends in the non-linear domain with
advanced strength and failure models. The finite element method is the most popular among structural
analyses when accidental loads are investigated, while more advanced analyses may include fluid-structureinteraction solutions, gas dispersion models and thermal effects. Despite advanced models, closed-form
solutions are preferred where applicable, due to their simplified use, and the avoidance of convergence
setbacks.
From modelling point of view the challenge is not in using numerical methods but rather using the
available tools appropriately. A balance between structural detail and confidence in results should
characterise a model.
In ALS design where non-linear action effects are dominant, FE analysis is practically axiomatic. Implicit
and explicit algorithms can be used with the two terms referring to the time integration schemes. The former
is solving for displacements at the nodes through the inversion of the stiffness matrix, and applied to the
nodal out-of-balance forces; while the latter solves for nodal accelerations by dividing summation of forces
by nodal mass. The specifics of time integration schemes can be found in the documentation of the FE code.
Several limitations apply to both schemes but explicit solvers are relevant to short duration, high pressure
gradient and large strain applications; often many times above the materials ultimate capacity. The
experienced strain rate witnessed by the material determines the appropriateness of the solver. Table 2 is an
unofficial guide showing the transition from implicit to explicit as strain rate levels increase.
Table 2. Solver Applicability per Hazard.
Solution
Impact Vel
(m/s)
Implicit
<10
<50
Explicit
Strain Rate (1/s)
>50
-5
Effect
Hazard/Analysis
Static/Creep
Fire/ Push Over
-5
-1
Elastic
Ship collision
-1
Elastic-Plastic
Collision, dropped object
10 - 10
10 - 10
Due to the large degree of non-linearity involved in ship collision analyses, implicit solvers are to be
avoided as significant time is needed to achieve convergence. Explicit solvers may be used but the
analysis time can be demanding with extensive modelling effort needed in the underlying assumptions
and preparation (stabilisation energy and non-linear parameters often needed). Simplified expressions and
industry-specific software that predict the collapse load of a member after a collision and the joint
capacity are often used avoiding the intricacies of FE codes that can be demanding to master. Non-linear
solvers like USFOS, Ls-Dyna and Abaqus are preferred for these problems where the loss of stiffness at
large displacements and member collapse is considered. The output of concern is usually a forcedeflection, or energy dissipation-deflection curve. Most effort should be spent on establishing credible
failure criteria and creating a mesh size based on recommended practices where applicable. Most
commercial codes nowadays support a series of failure criteria that remove elements as soon as certain
limits are exceeded. Typical failure criteria are based on geometric strain, plastic strain, element
distortion, or time step.
Gas dispersion and fire simulation are often analyses that operators undertake in the context of fire
safety design to judge the integrity of the structure. Additional need for fire response analysis other
than design purposes can be a result for accident investigation, part of risk analysis, or the need to
study escalating effects of fires; often explosions. Heat loads from fires and structural fire response
analysis in commercial codes use the k-epsilon model for turbulence modelling, the eddy dissipation
concept for combustion modelling and a radiation model based on the discrete transfer method.
Typically, the loads are presented as time-histories of ambient temperature around the structure or the
heat flux acting at the structure. These thermal loads are applied to the structure either via using
directly the temperature as a boundary condition, via heat flux boundary condition or using radiation
and convection with prescribed ambient temperature history. Thermal problems can be solved with
implicit solvers in the time domain. For steel structures under thermal loads it is the deformation of the
heated members of primary concern, and not so much the ultimate strength as long as the maximum
temperature remains below 400C. If the temperatures increase further, the ultimate strength and the
550
load carrying capability of the structure becomes of interest. Explicit solvers can be used for thermal
analyses as well, but this approach can be computationally expensive. Implicit solvers are other
preferable for collapse analysis under the thermal loads as the simulations tend to have long durations
and the dynamic effects are neglectful. Thermal solvers implemented in Abaqus and USFOS are among
the most popular tools used in the industry to assess thermal effects. A relatively new thermal solver in
the market is FDS with the advantage of being free of charge as of 2014. The main drawback of FDS is
that it is developed for low momentum applications of domestic fires and there is need for user
interaction to use it for oil and gas applications. Finally, FLACS and FLACS Fire are well established
tools for fire propagation and fire load assessment that have been used extensively over the years.
DNV-GL Phast is another tool for hazard analysis for several stages until the action is fully established.
Table 3. summarises the software options per accidental load.
Table 3. Software applicability per accidental load.
Load Type
Analysis Type
Solver
Software
Ship Collision
Non-Linear
USFOS, Ls-Dyna, Abaqus
Dropped Object
Non-Linear
Fire (thermal)
Linear/Non-Linear
Explosion & Dispersion
Non-Linear
Implicit/Explicit
Explicit,
Hand Calcs
Implicit/Explicit
Explicit,
Hand Calcs
Ls-Dyna, Autodyn, Abaqus

FDS, FLACS, CFX, USFOS, FAHTS
Ls-Dyna, Autodyn, FLACS, AutoReaGas
Anyone involved in engineering analysis is aware that numerical tools have a theoretical basis and are
formulated to conform to experimental data. The failure mode in a ship collision or dropped object
analysis should be of the expected type based on experiment observation, experience, and engineering
judgement. The challenge in using advanced numerical methods is not so much in getting results, but
rather to present an estimate based on valid engineering decisions made upon modelling. The challenge
lies with the analyst in deciding if the simulated solution is representing a realistic response. Built-in
features of non-linear solvers are constantly developing, and they are by no means free of defects. One
should be cognisant of the underlying principles and assumptions made in the software. Todays
software will implement by default settings requiring minimal necessity for user-input values; a
simple model should be subject to a well-defined load/deflection curve in order to calibrate the results
against any underlying models. Choosing the appropriate material model based on large deflections
with a given rupture strain, is only an indication as guidelines and recommendations are not
considering the mesh size of the model. A sensitivity analysis should follow the first solution to make
sure that the output value is converging.
6.3
Modelling Geometries
Any structure needs to be modelled with sufficient detail so that a satisfactory response is captured when
loads are applied, while the degree of detail depends on the type of analysis, as shown in Table 4. Thus,
the detail level of the model depends on the scope of the analysis, the load character and the anticipated
response. While the analysis under operational loads has to capture linear stress distribution, accidental
and ultimate limit state analysis has to capture geometrical and material non-linearity with a finer mesh
resolution at stress concentration areas. Some general guidelines for modelling can be found in DNV-RPC208 (2013) and DNV (1999).
Furthermore, the action orientation and application method can sometimes determine the way a
model needs to be meshed. Explicit analyses for example are less tolerant and highly sensitive to
distorted elements (CFL limit), while implicit analyses can proceed with an inferior mesh quality.
Advanced mesh algorithms are nowadays available, but certain fundamental criteria need to be met
when meshing. It cannot be stressed enough that when trying to capture non-linear effects in implicit
analyses, or in cases where action effects require the implementation of explicit solvers, reusing the
implicit mesh is grossly erroneous. Engineers should also be aware that the implicit mesh is not
suitable to capture the dynamic phenomena expected during high strain rate phenomena such as
thinning, hardening, or thermal softening; the model should be re-meshed when working with
inherited or legacy models. Internal checks are carried out where comments and warnings regarding
the mesh quality are registered in the .out file of most codes. Messages contained in the .out file
should be checked even at the end of a successful analysis. The higher the degree of detail in a
model, the higher the likelihood to violate mesh criteria as a higher number of elements and nodes is
551
created. Typical mesh quality criteria are presented in Table 4 although a deviation from the values
shown may be possible across FE codes.
Table 4. Element quality criteria.
Quality
Criterion
Description
Warpage
Less than 15deg
Deviation from planar plane
Aspect Ratio
Less than 5:1
Ratio of the longest edge to shortest one
Skew
Less than 60deg
Angle between two lines joining opposite mid-sides of the element
Jacobian
Greater than 0.6
Measure of the deviation of an element from an ideally shaped element
The type of accidental effect being studied greatly determines the degree of detail needed and the time
spent to ensure that the mesh is satisfactory for the intended analysis, this is given epigrammatically in
Table 4.
Table 5. Geometry Discretisation.
Item/Load Type
Operational Loads
Fatigue Loads
Accidental Loads
Ult Limit State
Analysis Type
Linear
Linear/Hot Spot
Very fine in SC
areas
Non-Linear
Fine to allow for
buckling, tripping
4-6 elements
between stiffeners
Non-Linear
Fine to capture
local failure
4-6 elements
between stiffeners
Mesh Size
Coarse
Characteristic
Mesh Size
Single element
between stiffeners
t t in SC areas
Most commercial FE codes support a direct import from CAD packages but industry-specific software is
often limited to few geometry formats that limit the options when facing problems upon import.
Removing any unnecessary details, unlikely to contribute to the stiffness of the model, is recommended
not only for successful imports but also for avoiding singularities in the contour plot. In cases where CAD
models consist of solids, the FE analysis may only require a surface model. Thus, a middle surface
application is required. The benefit from working with shell instead of solid elements is the convenient
way in which thickness is defined. Any thickness changes can be easily accommodated without the need
to remodel. Often the non-linear model is developed from a model used for linear analysis. Linear model
should be complemented with non-linear material properties and often the mesh resolution should be
increased in the areas where non-linear deformations and effects are assumed to occur i.e. the collision
zone or the drop zone for dropped objects analysis. Furthermore, in linear models the stiffeners are often
replaced with beam models although this approach cannot consider buckling or tripping of the stiffeners.
A shell-beam combination to model a stiffeners, or girders, is also possible where girder webs are
modelled by means of shell elements and flanges may be modelled using beam elements.
A discretised model should describe the three dimensional geometries of the structure so that the nonlinear solver can yield reliable results based on a geometry that can respond to the expected behaviour and
allow for the anticipated failure modes to develop. Common failure types of structures include tension
yielding, bending/compression failure, stability failure (the main failure mode in jacket structures), postbuckling load shedding in neighbour members, and interaction with local buckling. The implemented
elements should capture these failure modes at all times. Yield and fracture modelling requires careful
consideration on element size in order to capture high stress concentrations in the facture zone. Typically,
several elements should be present in the yield zone to achieve good strain estimates (DNV-RP-C208,
2013). However, this discretisation level is hard to achieve with large structures and material failure
criteria should be calibrated for larger element size as discussed in Section 6.5.3. Detailed FE models used
for collision analysis can be seen for example in Storheim and Amdahl (2014) and Gao et al. (2014).
In the context of modelling for accidental effects the degree of detail varies. Beam models are by no
means as detailed as those required in fatigue or stress studies where inclusion of joints and stiffeners is
necessary. Beam members are often modelled with centre-to-centre length properties for conservative and
simplicity reasons. Modelling with face-to-face properties reduces the buckling length, thus increasing the
members capacity. Hellan (1995) reported that the effect of face-to-face modelling can increase the
capacity by 5%. The number of beam elements per span should be carefully selected as excessive number
of elements can return spurious plastic hinges particularly in non-linear analyses. An exception to the
above is modelling for buckling response where multiple beam elements, usually 4 to 6, per span are
used. In short, the structural members needed are the ones that provide stiffness and strength.
552
Stiffened panels can be fully modelled or simplified by using orthotropic modelling techniques, or the
combined stiffener method. Asymmetric stiffeners, like HP profiles, are often modelled as flat bars with
equivalent properties. Any offsets are to be addressed by flushed stiffeners or beams and this is treated as
shown in Table 5 based on the studied hazard.
Secondary members consist of members that do not contribute to the global stiffness or strength of the
structure. Should secondary members attract any loads, these should be modelled with sufficient detail so
that the loads are transferred onto the primary structure realistically. A typical example of this is the
conductor framing and the boat landing fenders. Any inclusion of secondary members in the global model
should be justified as they can affect the results with excessive deformations (in the load-deflection curve)
when heavily loaded. Members that contribute to the overall capacity of the structure should be modelled.
Members not included in the model can attract hydrodynamic forces and additional loading due to their
own weight. In practice, omission of secondary members is usually accounted for by factoring up the
material density but this technique may not be always applicable.
Table 6 gives the modelling requirements per hazard considered in this section as the degree of detail
and structural arrangement varies. Particularly in advanced analyses where CFD analysis precedes the
structural assessment, the layout of the deck area needs to be modelled accurately to allow the formation
of poor fire if that is possible. In most cases a beam/stiffened panel assembly is simplified to a beam-only
model. In the full structural configuration the stiffened panel may, or may not, have its stiffeners or beams
flushed to plate level.
Another detail pertinent to beam-only models, often used in thermal analyses, is the applied boundary
conditions. In a fully fixed condition the beam node is fixed in all 6 degrees of freedom, while in a shell
model the upper, lower flange and web nodes can be fixed if necessary. Shell elements are not rare as all
surfaces (flanges and web) in the model can be considered to be engulfed (worst situation) at the same
degree of heat flux. Another benefit is that radiation among neighbour surfaces can also be considered in
a shell model. The benchmark study of this Committee showed that in heat flux application USFOS and
LS-Dyna agree very well.
The application of passive fire protection (PFP) is not explicitly modelled but rather considered
through the effective heat transfer coefficient of U-value in advanced fire analysis.
Table 6. Modelling Requirements.
Action
Collision
Plated structure with flushed members
Dropped Object
Beam/plated structure with flushed members
Blast
Plated structure with flushed members, stools, structural foundations

Plated structure or beam-only structure
with, or without, PFP
No flushed members
included.
Fire/Thermal
6.4
Structural Detail
Modelling Loads
The way loads are applied can greatly alter the structural response of a structure. ALS analysis implies
that the actions are of dynamic and non-linear nature, which could significantly affect the initial structural
configuration and topology.
Considering the nature of accidental loads such loads are characterised by the following aspects:
a. Kinetic energy governed by the mass of the striking body (ship including added mass, dropped object,
projectile mass)
b. Impact energy of the striking body, and the struck body if necessary
c. Impulse of overpressure
d. Temperature rise and duration of thermal effects
The following sections discuss modelling loads per hazard in detail.
6.4.1
Ship Collision
Ship collision analysis is often conducted in displacement controlled manner i.e. the analysis of external
motion dynamics is decoupled from the evaluation of the structural consequences. Such analysis assumes
a certain prescribed penetration path. It has been shown that such assumption is valid for symmetric right
angle collisions while in non-symmetric collisions the penetration depth depends on the ship structural
configuration, masses, collision location etc. Differences in damage description and penetration depth can
553
be significant. Brown (2002) compared the coupled model SIMCOL to the decoupled model of Pedersen
and Zhang (1998) and concluded that, while the total energy is similar in both approaches, the
decomposition of the total energy into transverse and longitudinal energy differs significantly. The same
conclusion is drawn by Tabri and Broekhuijsen (2011) using coupled and decoupled finite element
simulation to conclude that the decoupled simulation can result in erroneous description of penetration
depth in oblique angle collisions.
Regarding ship collisions, the prescriptive scenarios exist for ship-platform or ship-FPSO collisions,
see for example recent guidelines by LR (2014a). In offshore ship collision studies in design phase the
speed of the striking vessel is not to be less than 2.0m/s while the most probable impact location should
be determined by risk analysis with due account of the factors that affect the exact location like tidal
changes, and vessel motions sea states (NORSOK-N-003, 2007). In the absence of specific information
about the impact zone, values between 10m below LAT and 13m above HAT ((NORSOK-N-003, 2007)
are to be considered. Force-indentation, energy-deformation and force-deformation curves are available in
DNV-RP-C204 (2010) for different ship sizes and impact location (bow, stern, broad side). Should a
more detailed investigation be required than the guidelines provided in classification rules, then an local
explicit FE analysis should be considered with emphasis on material properties and failure criteria.
Such prescriptions typically do not exist for ship-ship collision. The accidental scenarios are defined
from a risk analysis considering a limited sea region (Qu et al., 2011, Goerlandt et al., 2012, Montewka
et al., 2014) or considering neighbouring ships sailing in the region, (UN, 2013).
6.4.2
Dropped Objects
Due to the short duration event, dropped object analyses are usually solved with explicit solvers where the
kinetic energy of the dropped object converted to strain energy in the struck and the striking body is
captured. Material failure is also considered in these studies, where implicit element formulations would
be inadequate to accommodate. These models are often local with sufficient detail as shown in structural
drawings. Special attention should be paid to the enforced boundary conditions; fixed conditions to the
struck body often applied at a distance away from the striking point are typical. The striking body can be
modelled wholly or partially. In the latter case the initial velocity corresponding to the whole body should
be applied. Certain codes like AUTODYN can only return pressure results when using just solid
elements, while Ls-Dyna can output pressure contours with shell elements as well. This subtle difference
can have consequences for both modelling and computational effort.
Convergence in these analyses is not a primary concern but the results can be highly sensitive to
plasticity models and mesh size. The kinetic energy of the dropped object is converted to strain energy in
the struck and the striking bodies. These analyses should be accurate in the defined impact velocity and
the dissipation of strain energy through the developed plasticity. Material failure of the struck body in the
form of rupture or crack is the primary concern as the impact can escalate to secondary hazards (gas
dispersion, fire, or explosion). Different impact angles should be considered, often in a parametric style of
definition, as this can influence plate rupture. In a numerical analysis when using solid elements the
assumption of a head-on impact (theta = 90deg) and impact at an angle (theta 90deg), can be the
difference of a surface-to-surface type of contact in the former case, while in the latter case can be an
edge-to-surface type with highly localised strains that drive failure. The above contact definitions are not
a concern when shell elements are used to model the stiffened plate and the dropped object.
The above paragraph concerns striking and struck bodies in air. The striking velocity of bodies that
experience free fall in water is different than in air ( = 2 ) and needs to consider their mass and area,
the height h in air prior to contact with water, the hydrodynamic added mass, water density and drag
coefficient. Guidance is given in DNV-RP-C204 (2010) but the subject of sinking dynamics deserves a
thorough treatment.
6.4.3
Explosions
Explosive loads on fire walls, louvered panels and bulkheads are usually considered uniformly distributed
over the exposed structure on a component level analysis. For global response analysis the overpressure
distribution is highly non-uniform with local maxima and minima throughout the structure. A global
analysis can reveal multiple locales where overpressure pockets are developed and numerous cases could
be studied as a result. In design level recommended practices can approach this type of loads
deterministically, by means of a suitable shape function for the displacements in the elastic and elasticplastic range. Several assumptions are made in the boundary conditions of explosively-loaded panels and
the systems fundamental periods. Recommended practices are based on Biggs theory with its suitability
discussed by Heywood and Martland (1999). Symmetric triangular pressure-time profiles, characterised
by rise time, maximum pressure and pulse duration, are often preferred for dynamic structural analyses.
Several other idealisations can be applied (rectangular, linearly decaying, gradually increasing, but
554
practitioners should be aware that nonlinear structural response can be different depending on the blast
load profiles. Applying uniformly distributed loads on design may be inadequate in terms of capturing a
representative response as deformations tend to be underestimated. A more accurate, non-uniform,
pressure application based on coupled CFD-structural models is presented in Paik et al. (2014) where the
torsional behaviour at the topside connection is considered and compared to results without any
connections. The main benefit from integrated analyses is the enhanced precision with which the
nonlinear structural response, with time and location-dependent explosive loads, is performed.
Models of blast resistant members like blast walls, bulkheads, decks and partition walls need to include
their supporting members (stools, shedder plates and brackets) onto the deck plating where they are
supported. These structures are often modelled with shell elements while beam elements may be present
as well. Due to the high strain rates experienced by the material, strain rate effects need to be considered
as well as temperature dependency. NORSOK N-003 and DNV-RP-C208 suggest the use of temperature
dependent stress-strain relationships as given in Eurocode 3, Part 1.2, Section 3.2 (EUROCODE, 2005).
6.4.4
Fire
As it is not feasible to design for a worst case scenario accounting for a prohibitively high number of
cases, probabilistic approaches based on quantitative risk analysis (QRA) are often implemented. A
formal QRA often provides the necessary input in the form of leak location, wind direction, leak rate
(kg/sec) and ventilation that help in determining the number of cases for analysis. A specific leak rate can
yield numerous scenarios for analysis as it returns different consequences and different leak rates form
explosive atmospheres with different frequencies of ignition. A successful QRA will also consider
escalation of events that might lead to explosion through a combination of probabilistic fire analysis, CFD
simulations, and thermal coupling to the structural model.
The use of FE analysis is at most times the preferred option due to the coupled effects that thermal
loads induce to the structure particularly when originated from accidental scenarios. Software packages
often analyse the accidental scenarios where a combustible plume is formed based on a CFD code and
ignited based on probabilities identified in QRA. The structural assessment is done based on the coupled
solution between a hydrodynamic code and the structural model. The output of these analyses takes the
form of predicting the collapse of the structure when subjected to fire. A new procedure that introduces
and benchmarks a coupling tool between KFX and Ls-Dyna is presented by Paik et al. (2013). The main
benefit from this integrated analysis is the enhanced precision with which the nonlinear structural
response, with time and location-dependent heat loads, is performed. This advantage is in line with a later
study from Paik et al. (2014) where the benefits of integrated analysis are highlighted.
Different methods to apply heat loads are available. Thermal loads defined in the form of heat flux
(radiative and convective) from hydrocarbon fires such as gas jet fires and pool fires, based on specific
mass release rates (kg/sec), can be found in NORSOK-S-001 (2008) and FABIG TN11(FABIG, 2009).
Standard fire/time temperature curves ISO-834-1 (1999) and EN 1363-1(EUROCODE, 1999a)) are
typical curves for cellulosic combustibles (wood, textile). A hydrocarbon curve ISO-834-2 (2009) and
EN 1363-2(EUROCODE, 1999b)) can also be used to simulate hydrocarbon fires if required.
Additional input for hydrocarbon fires can be found in UL1709(UL, 1994) and ASTM E1529(ASTM,
2014) where the temperature rises rapidly and levels out at 1093C for the remaining of the simulation.
These thermal loads are to be applied either via using directly the temperature as a boundary condition,
via heat flux boundary conditions, or using radiation and convection with prescribed ambient
temperature history.
Thermal analyses are solved with an implicit solver in time domain. Explicit solvers should be avoided
as unwanted response will be registered and analysis time can be prohibitive for long duration events in
the time frame needed.
The appropriate temperature-dependent properties of the materials involved in thermal analyses need to
be considered. Steel and aluminium have different properties particularly after the heat load application,
while most aluminium alloys are more vulnerable than steel when exposed to elevated temperatures (see
Paragraph 6.5 for further details).
6.5
Material Models
ISO-19902 (2007) dictates that the expected non-linear effects, including material yielding, buckling of
structural components and pile failures, should be adequately modelled and captured. Carbon steel is the
most popular of the materials used in ship and offshore structures. Stainless steel is also found particularly
in the topside structure together with aluminium alloys.
Material properties vary with the carbon content that affects ultimate strength, yield strength and
tensile failure levels. Carbon steel exhibits a linear stress-strain relationship until the materials yield
strength, unlike stainless steel that exhibits a rounded response in its stress-strain curve adding uncertainty
555
to the extraction of a yield strength value. Generally, when using stainless steel more than two points
should be used in the stress-strain curve around yield strength. This difference makes the use of design
rules necessary so that uncertainty is reduced and industry accepted values are referenced in the analysis.
Material properties will also vary in thermal analyses where thermal properties need to be considered
as a function of temperature. Significant differences in the thermo-physical properties (conductivity,
specific heat capacity and thermal strain) of carbon steel and aluminium are noted when exposed to high
temperatures; changes in the mechanical properties (Youngs modulus and yield stress) are also taking
place and should be addressed. More specific properties for different steel qualities are found on SCI
(2006), FABIG TN11(FABIG, 2009), FABIG TN6(FABIG, 2001), and Eurocode 3 (EN 19931:2(EUROCODE, 2005)). Properties for aluminium alloys can be found in Eurocode 9 (EN 19991:2(EUROCODE, 2007)). A succinct pool of information for materials used in the context of ALS design
can be found in the Material Database presented in this work in Annex 1.
For conducting a thermal analysis, the thermal properties of the materials have to be included in the
analysis. These include the thermal expansion parameters and the reduction of mechanical properties as a
function of temperature. For common metallic materials, the typical values for these properties can be
found from many engineering handbooks, material databases and design rules. For steel and other
metallic materials the typical properties to be included as a function of the temperature are the following:
Reduction of Youngs modulus
Figure 18. Elastic modulus of steel as predicted by different models and as measured in different tests
(Kodur et al., 2010).
Reduction of Yield Strength
Figure 19. Yield strength of steel as predicted by different models and as measured in different (Kodur et al., 2010).
556
Specific Heat Capacity
Figure 20. Specific heat of steel as predicted by different models and as measured in different tests (Kodur et al.,
2010).
Thermal Conductivity
Figure 21. Thermal conductivity of steel as predicted by different models and as measured in different test programs
(Kodur et al., 2010).
Thermal Elongation
Figure 22. The coefficient of thermal elongation of steel.
557
Surface Emissivity
Temperature dependent, typically 0.250.8.

From the modelling point of view, three material characteristics are of primary concern, namely the
plasticity model, the stress-strain curve, and the defined failure criteria; these are discussed in the
following paragraphs.
6.5.1 Plasticity Model

Plasticity is a property that has to be considered with the corresponding hardening model (kinematic,
isotropic, or a combination of both). Steel is generally known to exhibit kinematic hardening. In analyses
where the ultimate strength is sought, a piecewise linear isotropic plasticity model can be used while
kinematic hardening is more appropriate to cyclic loading. Permanent displacements from impulsive
loads are known to be unaffected by the hardening model; typically a combination of kinematic-isotropic
is being used (=0.5). A wide-ranging research on the effects of high strain rates on material properties is
given in OTI 92 602.
Plasticity with hardening exhibited beyond the yield strength can be described with a number of
expressions. Five common models in FE analysis are:
a.
b.
c.
d.
e.
Elastic-perfectly plastic
Elastic-linear hardening
Elastic-piecewise hardening
Elastic with exponential hardening
Ramburg-Osgood equation
Most FE codes will assume that true values (not engineering) are being used in the material input data.
It is the responsibility of the engineer to convert engineering to true values; guidance is provided in DNVRP-C208 and in Appendix A. As a rule of thumb, true stresses will be higher than engineering (nominal)
when plotted against strain.
Typical yield strength values for carbon steel are in the range of 420MPa with ultimate strength in the
range of 500-600MPa and failure strain of 19%. For stainless steel the corresponding values are 460MPa,
640MPa and 25%. Modulus of elasticity can be considered in the range of 200-210GPa with Poissons
ratio of 0.3. The variation on material strength as a result of thickness change should also be considered in
the analysis; values can be found in EN 10025:3(EUROCODE, 2004). Additional information can be
found in the Material Database presented in this work in Annex 1.
6.5.2 Stress-Strain Curve

Acceptance criteria often dictated by classification societies determine the material properties or provide
some guidance as to which material models should be used in non-linear analyses. In ALS design it is
highly recommended that the material properties are obtained from appropriate material certificates, or
laboratory tests, which present the whole stress-strain curve that can be converted to true strains and
stresses and used as a basis to define a failure criterion. In lack of specific figures the ultimate strain can
be taken as 15%, or 100 strain at yield point y (Eurocode 3). Failure based on plastic strain should
always consider mesh sensitivity and strain rate effects.
In impulsive loads where the material undergoes large degrees of membrane stretching, shell thinning
needs to be considered as the plasticity level increases. This is usually combined when true stress-strain
values are used; in the opposite situation where engineering values are used, it is more appropriate to
neglect membrane stretching.
6.5.3 Failure Criteria

ALS design often involves simulating non-linear deformations and ductile fracture in large complex
structures. In these analyses, shell elements are preferred over solid elements for computational reasons.
A key question in such analyses is to establish a proper material curve for plastic region and a reliable
fracture criterion that could cover the length scales from large structural shell elements to the material
behaviour in the crack tip including localization in the form of necking. There are two main reasons,
which complicate establishing a reliable fracture criterion (Krgesaar et al., 2014):
Shell elements with plane stress formulation are incapable of representing the 3D stress state during
necking and localisation of deformation preceding fracture initiation.
558
Large shell elements with an element length Le ( Le = A , where A is the element area) of several
times the sheet thickness cannot capture the high strain gradients in the localization band, which
reduces the accuracy even more.
Localisation in the form of necking causes mesh size effects, whereby fracture strain increases, as the
mesh is refined.
For a practical approach to material modelling several topics presented above cannot be considered
precisely. DNV-GL has recently published and is still developing further the recommended practices for
the determination of structural capacity of non-linear FE analysis methods (DNV-RP-C208). Among
other issues the DNV-RP-C208 suggests material curves both for engineering stress-strain and true stressstrain space. The material curves are defined until the necking that is assumed to initiate once the ultimate
strain value is reached, see Figure 23. Several studies have shown that after the material possesses
significant resistance after the necking and the material curves and the failure strain should be scaled
according to the element size, see for example Ehlers and Varsta (2009) in Figure 23 and Hogstrm et al.
(2009) in Figure 24. Typical approach is to calibrate a constant failure strain (for example using an
empirical expression by Scharrer et al. (2002) after which the fracture is assumed to occur and the
element is removed from the simulation.
Figure 23. DNV-GL true stress-strain curve for S235 and S355 (DNV-RP-C208, 2013).
Figure 24. True stress-strain curve obtained by optical measurement (Ehlers and Varsta, 2009).
559
Figure 25. True stress-strain curve obtained by optical measurements for NVA and Domex 355 grade steel
(Hogstrm et al., 2009). VE stands for Virtual Extensimeter and describes the length over which the
stress-strain relation is evaluated. Smaller VE value corresponds to smaller elements.
While the constant ultimate plastic strain is a popular criterion, more advanced failure criteria and scaling
models are being developed. Employing equivalent plastic strain measured with tensile tests as a fracture
criterion and scaling it based on some law (See Barbas Law or some empirical relation) to account for
different mesh sizes, is sufficient in cases where the stress state corresponds to uniaxial tension and does
not change significantly during the analysis. However, in real structures such idealistic conditions rarely
occur. Under multi-axial stressing, fracture depends on the pressure or the stress triaxiality as noticed by
several authors and shown in many experimental studies, see Bao and Wierzbicki (2004), Barsoum and
Faleskog (2007), Wierzbicki et al. (2005), Haltom et al. (2013), Hopperstad et al. (2003). Bai and
Wierzbicki (2010) showed that depending on the stress state, the fracture strain could vary several times
for the same element length. Therefore, a reliable fracture criterion used for large structures must
incorporate the effect of the stress state in addition to element size. One of the criteria accounting for that,
available in LS-Dyna is the so-called RiceTraceyCockcroftLatham criterion (RTCL) presented by
Trnqvist (2003). This criterion includes the effect of the stress state, but the calibration of the criterion to
account for the mesh size effects is solely based on the equivalent plastic strain determined with the
uniaxial tension test. In the same criterion the fracture strain can be adjusted for different element lengths
according to Barbas Law without the influence of the stress state.
A novel theoretical framework for adjusting failure strain based on the stress triaxiality and element
size has been given by Walters (2014). Krgesaar et al. (2014) studied the validity of Walters adjustment
of the fracture strain by exploring its ability to model fracture in large-scale shell structures by using
Walters framework to calibrate a stress triaxiality based fracture criterion formulated by Lou et al.
(2012). Calibrated criterion was used in simulating large-scale panel indentation experiments carried out
by Alsos and Amdahl (2009). The comparison of the measured force-displacement curves with simulated
ones indicates that the approach of scaling fracture strain based on both, stress state and mesh size,
accurately accounts for the reduction in ductility with increasing element size. Furthermore, comparison
with the fracture criterion adjusted only based on the uniaxial tension test demonstrated that the new
approach reduces the element size dependency. Krgesaar and Romanoff (2013) studied the effect of
element softening while simulating the structural deformations using relatively large elements. However,
these more advanced approaches by Walters (2014) and Krgesaar et al. (2014) are not yet implemented
into commercial software packages.
560
6.6

Uncertainties of ALS Models
Addressing uncertainties associated with action effects should be given due consideration particularly for
actions that require dynamic analyses where the structural response is nonlinear. Dynamic actions are
highly sensitive to inertial and damping characteristics, particularly when the excitation source coincides
with the natural frequencies of the structure. Specific actions like earthquake and ship collision relate to
such sources where uncertainties are partly taken care of by safety factors. Numerous definitions on what
uncertainty in modelling is exist in the literature, see Ayyub and McCuen (2011) and DNV (1992). In
engineering analysis uncertainty in design can be addressed with probabilistic design analysis (PDA)
tools; essentially a combination of probability theory and FE methods that can help estimating reliability
indices. In very broad terms PDA provides an insight on the effect of loads and strength parameters, to
the output values, e.g. plastic strain. By the time engineers set up a numerical model. Most of the random
variables have been estimated, either from Class Rules, previous work, operator feedback, ongoing
research, or open literature. The remaining uncertainties consist of those involved in model input
parameters and those involved in modelling techniques. Modelling uncertainties can take the following
form:
Structural discretization.
Material properties and models.
Description and application of loads.
Coupling between load and structural response (or coupling between action and action
effects).
Efforts to reduce uncertainty in modelling and improve structural response are focused on using
coupled models with integrated solutions like the ones presented by Paik et al. (2013), Paik et al. (2014)
in fire and explosion analyses. Decoupled solutions can underestimate deformations, spatial distribution
of heat and over-pressures or overestimate deformations occasionally. Decoupled solutions can also be
inadequate in capturing the torsional behaviour at the topside connections and returning unreliable
thermal loads.
Probabilistic models can address uncertainties based on reliability theory and introduce metrics like
that of Robustness Index based on probability of failure under uncertain loads as presented in Czujko and
Paik (2014). The advantage of probabilistic modelling of loads is that a frequency of load exceedance or,
more precisely, frequency of exceedance of load parameters can be obtained. Uncertainties can take the
form of exceedance curves represented by probability density function derived from curve fitting.
Lingering uncertainties remain in QRA undertaken prior to any analysis. These uncertainties will
always remain in place as long as they are deemed to remain within allowable risk levels.
6.7
Probabilistic structural analysis was defined by Ditlevsen and Madsen (1996) as being the art of
formulating a mathematical model within which one can ask and get an answer to the question What is
the probability that a structure behaves in a specified way, given that one or more of its material
properties are of a random or incompletely known nature, and/or that the actions on the structure in
some respects have random or incompletely known properties?
6.8
APPENDIX A
6.8.1 True stress-strain curve for Ls-Dyna
Typically, for non-linear finite element simulations true stress-strain curve is needed. For that engineering
stress-strain curve obtained from the tensile test is transformed into true stress-strain curve. Here, a
procedure is presented for a conventional tensile test. More advanced techniques based on optical
measurements can be seen for example in Ehlers and Varsta (2009).
From the tensile test the engineering stress is obtained as a function of engineering strain. Engineering
strain is transformed into true strain by
true =
dL
L0
L +
= ln(L) ln(L0 ) = ln(L0 + ) ln(L0 ) = ln( 0
) = ln(1 + engin)
L0
where the following relations are used:
(A1)
engin =
561
L L0
=
L0
L0
Equation (A1) is valid through the whole test process. Transforming engineering stress into true stress
is more complicated than it was in the case of strains. As the cross-sectional area before the test, A0, and
after the test, AT, are given, the true stress can be evaluated at the end of the test using relation:
T = Y
A0
AT
(A2)
where y is yield stress (0,2 for example). Equation (A2) can be used only at the last point of the test
when broken specimen can be measured, as AT values are not given for other test points. For the rest of
the test region another approximation should be used. It is assumed that tensile pressure causes even
decrease of cross-sectional area through the whole gauge length:
AT ( L + dL) = A0 L
A0 L + dL
dL
=
=1+
= 1 + engin
AT
L
L
Now the true stress can be approximated as follows:
engin
(1 + engin )
(A3)
Equation (A3) is valid only until necking occurs and true stress values between the necking point and
the last point are still undetermined. Undetermined values should be evaluated by using given true stress
values. This is explained in Figure 26. Engineering stress-strain is obtained directly from the tensile test.
As a next step, true stress value at the end of the test is calculated by using Equation (A2). In the graph it
is marked with the circle at the far right of Figure 26. Approximation using eq. (A3) connects this
extreme stress-strain value with the true stress-strain values before the necking, when the true stress-strain
curve calculated from the engineering values starts to fall.
Figure 26. Stress-strain curves.
7.
7.1
BENCHMARK STUDY. RESISTANCE OF TOPSIDE STRUCTURES

SUBJECTED TO FIRE
Scope of work
The objective of the benchmark study is to predict the strength of topside structures subjected to fires and
compare different techniques assessing the strength of these structures. The capabilities of modern
software to simulate such complex loads are evaluated and Passive Fire Protection (PFP) design using
numerical predictions are assessed.
The following committee members have contributed to the benchmark:
562
Table 7. Committee members contributed to the benchmark.

Participation
J. Czujko
L. Brubak
G.S. Kim
K. Tabri
Y. Yamada
W.Y. Tang
J. Czaban
7.2
Affiliation
Nowatec AS, Norway
DNV GL, Norway
HHI, Korea
MEC, Estonia
NMRI, Japan
Shanghai Jiao Tong University, China
Department of National Defence, Canada
Analysis software
USFOS, LS-DYNA
ABAQUS
USFOS, LS-DYNA
LS-DYNA
LS-DYNA, Marc
ABAQUS
LS-DYNA
Reference on Figures
Czujko
Brubak
Kim
Tabri
Yamada
Tang
Czaban
Strategy of benchmark study
The capability and accuracy of available techniques for the prediction of structural response of topside
structures under fire loads are evaluated in terms of:
1. Temperature development on topside structures.
2. Deflections at 16 locations of the topside structure surface.
The strategy of benchmark study is presented in Table 8.
Table 8. Strategy of the benchmark study.
Title
Primary study
Parametric study
Descriptions
1. Static analysis for all in-place loads
3. Push-down analysis, uniform pressure until collapse
4. Fire analysis
4.1 Application of fire load defined by standard hydrocarbon fire curve
4.2 Application of fire load by predefined constant global heat flux
5. Design of PFP
1. Effects of boundary conditions
2. Methods of controlling numerical instability for beam element model
2.1 Element remove method
2.2 Temperature remove method
3. Effects of modelling assumption, shell vs beam element models
4. Effects of local heat flux
7.3
Input
7.3.1 Geometry of target structure
Nowatec AS prepared a 3D CAD geometry model of topside deck structures to facilitate this benchmark
study. The target structure was a part of an offshore topside deck structure, Figure 27. The thickness of
deck plate was 8mm. In the first step the deck plate was not modeled explicitly but the weight of plate was
included in the total weight of the deck. Geometrical details of the structure are given in Annex 2. All inplace loads acting on the deck structure were given to all participants in form of Excel and STAAD Pro
input files.
Figure 27. The deck structure. Geometry used in the benchmark study.
563
7.3.2 Material data

7.3.2.1
Strength properties of steel material
Material model for strength assessments was assumed to be bilinear elastic plastic, described by the
material constants.
European grade structural steel S355 was considered for this benchmark study. The yield stress and
elastic modulus of steel at room temperature were 355 MPa and 210 GPa, respectively, and Poisson ratio
was 0.3.
To perform the thermal and structural response analysis of structures under fire loads, material
properties should be defined as temperature dependent. Figure 28 presents the reduction factors for yield
stress and elastic modulus at elevated temperature defined according to Eurocode (EUROCODE, 2005).
Figure 28. Reduction factors at elevated temperatures.
7.3.2.2
Thermal properties of steel material
The carbon steel material curves for elevated temperatures specified in Eurocode were used in the thermal
and structural analysis. Figure 29 shows the curves for specific heat capacity and thermal conductivity,
respectively (EUROCODE, 2005). The surface emissivity of steel was set to 0.8 for all structural
members exposed to fire.
Figure 29. Thermal properties of steel: specific heat capacity (left) and conductivity (right).
7.3.2.3
PFP thermal characteristics
The fire protection material applied to a structural member was to remain unaffected and retain its fire
performance if subjected to fire loads. The chosen spray-on fire protection material was Chartek VII. The
thermal characteristics of Chartek VII shown in Figure 30 was used in this study (IPC, 2000).
564
Figure 30. Thermal characteristics of Chartek VII: specific heat (left) and conductivity (right).
7.3.3
Boundary conditions
Figure 30 shows the structure was fixed at the supporting points (A, B, C, D, E, F, G, H) in the case of
beam element model. Figure 31 and Table 9 illustrate the boundary conditions of finite element model.
For finite element model, supporting points along the cross-section of the beam were all fixed. In this
study simulations considering two types of boundary conditions, such as fixed in-plane (B.C_1) or free
(B.C_2), were performed.
A
A
C
D
C
D
E
F
E
F
G
H
Figure 31. Boundary condition of beam

element model.
G
H
Figure 32 Boundary condition of shell element model.
Table
9. Summary of boundary conditions.
__________________________________________________________________________________________________________
Boundary condition Descriptions (fixed =1)
__________________________________________________________________________________________________________
B.C_1 The structure is fixed supported at the supporting points
(A,B,C,D,E,F,G,H)
__________________________________________________________________________________________________________
B.C_2
A (UX, UY, UZ, RX, RY, RZ =1)

B, C (UZ, RY, RZ =1)
D (UZ, RX, RY, RZ =1)
E (UX, UY, UZ, RX, RY, RZ =1)
F, G, H (UY, UZ, RY, RZ =1)
__________________________________________________________________________________________________________
7.3.4
7.3.4.1
Loads
Mechanical action
Three basic types of mechanical actions were applied in the model. However, in order to calculate the
collapse strength of the deck model, push-down analysis was performed by using a push-down load case
as shown in Table 10.
Table
10. Mechanical load case summary.
___________________________________________________________________________________________________________________
Load case Loading condition Based on Remarks
___________________________________________________________________________________________________________________
LC1 Gravity =1.1*9.81 (m/s2) Gravity loads should be applied individually. Mass of deck plate (27472kg) is
considered in the gravity load
LC2 Live load =5.63 (kN/m2) live loads should be converted to shell and beam loads respectively
LC3 Piping and equipment Given in STAAD/USFOS piping and equipment loads should be applied as nodal loads
(for application of LC3 see Annex 2)
LC4* Uniform pressure = Live load*scale factor Live load (LC2) is scaled up
____________________________________________________________________________________________________________________
*LC4 is used for push-down analysis
7.3.4.2
565
Fire action
The target deck structure was subjected to the following fire actions;
1. Standard hydrocarbon fire curve
2. Predefined constant global heat flux
In case of hydrocarbon fire, Eurocode provides the standard hydrocarbon fire equation as follows
(EUROCODE, 2005);
g = 20 + 1080(1 0.325e0.167t 0.675e2.5t )
(1)
where g = gas temperature near the steel member in C; and t = time in minutes.
Figure 33 shows the standard hydrocarbon fire curve.
1200
Temperature (C)
1000
800
Standard Hydrocarbon Fire Curve
600
400
200
0
0
10
20
30
40
Time (min)
50
60
Figure 33. Standard hydrocarbon fire curve.
In the case of constant global heat flux loading, the global heat flux on the target deck structure was
QG=100 KW/m2 and the local heat flux was QL=350 KW/m2. Figure 35 shows the application area of
global heat flux and Figure 34 presents the four locations of local heat flux on the target deck structure.
Location
1
Local heat flux: 350kW/m2
Global heat flux:

100kW/m2
Location
2
Location
4
Location
3
Figure 34. Global heat flux.
7.3.4.3
Figure 35. Local heat flux.
Monitoring of results
Figure 36 presents the location of monitoring points for reporting of deflections/damage during fire and
residual deflections after fire over the target structure surface. The monitoring point considered was at the
centre of the web of I-beam. From practical point of view, flange is more important, but it was decided to
566
report the results on the centre of the web for simplicity reasons and to compare with the beam element
model.
A1
X1
Y1
B1
A2
X2
Y2
B2
A3
X3
Y3
B3
A4
X4
Y4
B4
Constrained nodes
Monitoring points
Figure 36. Location of monitoring points on the target structure, deformation and temperature.
7.4
Results
7.4.1 Static analysis
Static analysis was performed by considering all in-place loads (self-weight, live load, piping &
equipment load). Table 11 summarizes the modelling approach and reaction forces for in-place loads
calculated by the participants of the benchmark study. Figure 37 shows the deflection at monitoring
points due to in-place loads reported by all participants. Table 11 summarizes the mean and standard
deviation at monitoring points.
Table 11. Modeling approach for benchmark study.
Participant
Software
Brubak
ABAQUS
USFOS
LS-DYNA
USFOS
LS-DYNA
LS-DYNA
ABAQUS
LS-DYNA
Marc
Czujko
Kim
Tabri
Tang
Yamada
Element
type
Shell
Beam
Shell
Beam
Shell
Shell
Shell
Shell
Shell
Load
LC1
7.31E+05
1.02E+06
1.03E+06
1.04E+06
1.00E+06
7.34E+05
7.27E+05
7.32E+05
7.32E+05
Boundary
condition
Fixed
Fixed
Fixed
Fixed
Fixed
Fixed
Fixed
Fixed
Fixed
Static analysis (A2)
Mean
Brubak Czujko Czujko

Kim
Kim
Tabri
Tang
Yamada
(ABAQUS) (DYNA) (USFOS) (DYNA) (USFOS) (DYNA) (ABAQUS) (DYNA)
LC2
2.37E+06
2.46E+06
2.46E+06
2.39E+06
2.37E+06
2.39E+06
2.42E+06
2.48E+06
2.49E+06
LC3
3.38E+05
3.38E+05
3.38E+05
3.38E+05
3.38E+05
3.38E+05
3.38E+05
3.38E+05
3.38E+05
Static analysis (X3)
Mean

Kim
Kim
Tabri
Tang
Yamada
567
Static analysis (Y3)
Mean

Kim
Kim
Tabri
Tang
Yamada
Figure 37. Deflection at monitoring points subjected to in-place loads.

Table 12. Modeling approach for benchmark study.
Mean
Standard deviation
Monitoring point
A2
31.14
2.36
X3
50.87
4.18
Y3
27.84
2.40
7.4.2
7.4.2.1
Push-down analysis
Shell element models
In order to calculate the collapse strength of deck model, push-down analysis was performed. All in-place
loads (LC1~LC3) acting on the model did not induce the global collapse. Therefore, addition load
(uniform pressure) was applied in the model and the collapse strength was reported by each participant.
Figure 38 shows the comparison of results for push-down analysis among all participants using shell
element model. The results indicate that the collapse strength of the deck model documented by each
participant had good agreement with each other in the linear state, but had big differences in the plastic
state. For any specific vertical force (e.g. 11600kN) the displacement reported by each participant was
compared and the results differed about 20% with the mean value, which is significant.
Figure 38. Vertical force versus deflection at monitoring point A2 and X3, shell element model.
7.4.2.2
Shell vs. beam element models
Figure 39 shows the comparison of results for push-down analysis between beam element model and shell
element model. The results shown in Figure 39 indicate that the beam and shell element models have
good agreement in elastic state but beam element model collapsed early.
12000
12000
10000
10000
8000
8000
Force (kN)
Force (kN)
568
6000
6000
4000
4000
Monitoring Point (A2)

Beam element model
Shell element model
2000
Monitoring Point (X3)

Beam element model
Shell element model
2000
100
200
300
Z-displacement (mm)
400
500
100
200
Z-displacement (mm)
300
400
Figure 39. Comparison of results for push-down analysis between shell and beam element models at monitoring
point A2 and X3.
7.4.3
Fire analysis
The objective of the benchmark study was to assess the capability and accuracy of available techniques
for the prediction of structural response of topside structures subjected to standard hydrocarbon fire curve
and constant heat flux loads.
7.4.3.1
Heat transfer analysis
Before to perform the structural response analysis subjected to fire, the results of heat transfer analysis
should be compared. Figure 40 shows the results of heat transfer analysis for 30 minutes fires.
Standard hydrocarbon fire
Figure 40. Comparison of results of heat transfer analysis for 30 minutes fires at monitoring point A2.
7.4.3.2
Standard hydrocarbon fire curve
The loading condition was the application of standard hydrocarbon fire curve on the whole structure.
Figure 41 shows the comparison of temperature and Figure 42 shows the comparison of deflection at
monitoring point A2 and X3.
Temperature (oC)
Temperature (oC)
569
Figure 41. Comparison of temperature for standard hydrocarbon fire at monitoring point A2 and X3.
Figure 42. Comparison of deflection for standard hydrocarbon fire at monitoring point A2 and X3.
Shell vs beam element models
Temperature (oC)
Figure 43 and Figure 44 show the comparison of results for standard hydrocarbon fire between beam
element model and shell element model. The results of temperature and deflection have good agreement.
Figure 43. Comparison of temperature for

standard hydrocarbon fire at monitoring point A2.
7.4.3.3
Figure 44. Comparison of deflection for standard

hydrocarbon fire at monitoring point A2.
Constant global heat flux
The loading condition was the application of constant global heat flux (100kW/m2) on the whole
structure. Figure 45 shows the comparison of temperature and Figure 46 shows the comparison of
deflection at monitoring point A2 and X3.
570
Figure 45. Comparison of temperature for constant global heat flux at monitoring point A2 and X3.
Figure 46. Comparison of deflection for constant global heat flux at monitoring point A2 and X3.
Shell vs beam element model
Figure 47 and Figure 48 show the comparison of results for constant global heat flux between beam
element model and shell element model. The results of temperature and deflection have good agreement.

constant global heat flux at monitoring point A2.
7.4.4
Figure 48. Comparison of deflection for constant

global heat flux at monitoring point A2.
Design of PFP
In order to consider the effect of PFP, only the thermal properties were considered in the heat transfer
analysis. In the case of USFOS program (USFOS, 2013), the input of PFP material is defined in terms of
resultant heat conduction. Therefore, USFOS considers only the thickness and conductivity of PFP in
571
order to calculate the steel temperature of the PFP insulated structure. As specific heat capacity is not
defined, it is assumed that there is no heat absorbed by PFP material. On the other hand, there is no
possibility to calculate directly the effect of PFP in LS-DYNA program (LS-DYNA, 2014). One of the
possible solutions is to make a model of PFP by solid elements and assign PFP properties on those solid
elements. However, this procedure is very time consuming and cumbersome in the case of big and
complicated FE models. Another possible way is to calculate the steel temperature through simple
equation developed by Eurocode (EUROCODE, 2005) and apply that steel temperature in the FE model
manually. The second procedure was adopted in the current benchmark study.
Figure 48 shows the comparison of temperatures for standard hydrocarbon fire with PFP at monitoring
point A2 and X3. The deflection of standard hydrocarbon fire with PFP is presented in Figure 49.

standard hydrocarbon fire with PFP at monitoring
point A2.
7.4.5
Figure 50. Comparison of deflection for standard

hydrocarbon fire with PFP at monitoring point
A2.
Effects of boundary conditions
The boundary conditions are very important for the proper calculation of the structural responses. In the
case of beam element model, the center of cross section (node located at the center of the web) was used
as the boundary node, and for shell element model the nodes located along the cross section of the beam
were considered as boundary nodes (see Figure 51). Figure 51 presents the difference in deflection due to
the boundary conditions. Standard hydrocarbon fire curve has been applied for this comparison study.
The deflection is reported for monitoring point A2 and X3.
0
-100
Z-Displacement (mm)
-100
-200
-200
-300
-300
-400
Monitoring Point (X3)

B.C_1
B.C_2

B.C_1
B.C_2
-400
-500
0
100
200
Time (s)
300
400
100
200
300
400
Time (s)
Figure 51. Comparison of deflection for standard hydrocarbon fire between B.C_1 and B.C_2.
7.4.6
Methods of controlling numerical instability for beam element model
In the case of beam element model, the numerical simulation sometimes stopped due to secondary
member collapse. Loads at a secondary member resulted in large displacements and distorted elements,
which caused the numerical problems. To solve this problem, two possible methods were adopted in the
572
benchmark study and compared with each other in terms of effectiveness and accuracy. The first one was
named the element remove method and the second one the temperature remove method.
7.4.6.1
Element remove method
In this method the secondary members were removed when fire response analysis was performed. In the
case of element remove method, the X and Y directional displacement was different than the original
value at primary members as shown in Figure 52. Also, the global reaction force under in-place loads was
different due to removed elements.
USFOS (Original)
USFOS (Elements remove)
Figure 52. Comparison of deflection between original and elements removed model.
7.4.6.2
Temperature remove method
In the case of temperature remove method, the secondary members were not exposed to the fire loads.
The heat transfer analysis was performed considering only the primary members exposed to fire and then
structural response analysis was conducted using the results of heat transfer analysis. Figure 51 shows the
comparison of deflection between element remove method and temperature remove method. In this
benchmark study the temperature remove method was used in all fire analyses (both beam and shell
element model) in order to keep consistency between beam and shell element model.
0
Z-Displacement (mm)
-200
-400
-600

USFOS_Orignial
LS-DYNA
USFOS_Temperature remove
USFOS_Element remove
-800
-1000
0
100
200
300
400
Time (s)
Figure 53. Comparison of deflection between element and temperature remove method.
7.4.7
7.4.7.1
573
Effects of local heat flux

Local heat flux (location 1)
This load case considers both global and local heat flux on the target deck structure. The local heat flux
(350kW/m2) was applied only in location 1 and global heat flux (100kW/m2) was applied in the rest of deck
structure. This case was analyzed to check the effect of local heat flux on the target deck structure. Figure 52
and Figure 54 show the results of local heat flux (location 1) analysis at monitoring point A2 and X1.
Figure 54. Comparison of temperature for local heat flux in location 1 at monitoring point A2 and X1.
Figure 55. Comparison of deflection for local heat flux location 1 at monitoring point A2 and X1.
7.4.7.2
(350kW/m2) was applied only in location 2 and global heat flux (100kW/m2) was applied in the rest of
deck structure. Figure 56 and Figure 57 show the results of local heat flux (location 2) at monitoring point
A2 and X2.
574
Figure 57. Comparison of deflection for local heat flux in location 2 at monitoring point A2 and X2.
7.4.7.3
(350kW/m2) was applied only in location 3 and global heat flux (100kW/m2) was applied on the rest of
A2 and X4.
Figure 59. Comparison of deflection for local heat flux in location 3 at monitoring point A2 and X4.
7.4.7.4
575
(350kW/m2) was applied only in location 4 and global heat flux (100kW/m2) was applied in the rest of
A2 and Y3.
Figure 60. Comparison of temperature for local heat flux in location 4 at monitoring point A2 and Y3.
Figure 61. Comparison of deflection for local heat flux in location 4 at monitoring point A2 and Y3.
7.5
Conclusion from the benchmark study
The objective of the benchmark study is to predict the strength of topside structures subjected to fires and
compare different techniques assessing the strength of these structures. The capabilities of modern
software to simulate such complex loads are evaluated and the needs for Passive Fire Protection (PFP)
design using numerical predictions are assessed.
The presented benchmark study consists of a relatively simple structural arrangement, i.e. a part of an
offshore topside deck structure, subjected to fire loads. However, the study proved to be sufficiently
complex enough to cause significant scatter in results when analyzed by a group of experts. This scatter is
attributed to the underlying simulation assumptions made by the analysts. These results provide
invaluable insight into the variability in predictions when different values are used for influential
parameters, one of which is the analysts themselves.
The conclusions of primary study and parametric study are as follows:
Static analysis
This phase unveiled the influence of the individual approximations including the assumed in-place
loading, geometric discretization, and boundary conditions. It was found that USFOS, ABAQUS and LSDYNA were able to predict the static deflection with good accuracy.
576
Push-down analysis
To check the collapse strength of deck, push-down analysis was performed. The results indicate that the
collapse strength of the deck model documented by each participant had good agreement with each other
in the linear state, but had big differences (20%) in the plastic state.
Fire analysis
In the case of fire analysis, there were some differences. The temperature was different and as a result the
deflection was also different. Therefore, in the case of fire analysis, thermal loading should be applied
correctly in terms of the number of exposed sides, thermal properties and heat transfer. Design of PFP
through numerical simulation was evaluated for beam and shell element model.
Effects of boundary conditions
The boundary conditions are an important factor for the proper calculation of the structural responses.
The results indicate that very exact identification and modelling of boundary condition is necessary to
correctly predict structural behavior in fire conditions.
Effects of local heat flux
To document the behavior of structure subjected to both global and local heat flux, present study was
performed for 4 cases of local fires. The results indicated that local heat flux gives a larger damage than
the constant/global heat flux.
8.
REFERENCES
ABS 1995a. Guide for assessing hull girder residual strength for bulk carriers. American Bureau of Shipping.
ABS 1995b. Guide for assessing hull girder residual strength for tankers. American Bureau of Shipping.
ABS 2013. Accidental load analysis and design for offshore structures. American Bureau of Shipping.
Alsos, H. S. & Amdahl, J. 2009. On the resistance to penetration of stiffened plates, Part I Experiments.
International Journal of Impact Engineering, 36,799807.
API 2006. Recommended Practive for the Design of Offshore Facilities against Fire and Blast Loading. American
Petroleum Institute.
API 2007. API 14J: Design and Hazards Analysis for Offshore Production Facilities. American Petroleum Institute.
ASTM 2014. Standard Test Methods for Determining Effects of Large Hydrocarbon Pool Fires on Structural
Members and Assemblies. American Society for Testing and Materials.
Ayyub, B. M. & Mccuen, R. H. 2011. Probability, statistics, and reliability for engineers and scientists, CRC press.
Bai, Y. & Wierzbicki, T. 2010. Application of extended MohrCoulomb criterion to ductile fracture. International
Journal of Fracture, 161, 120.
Barsoum, I. & Faleskog, J. 2007. Rupture mechanisms in combined tension and shearexperiments. International
Journal of Solids and Structures, 44, 17681786.
Biggs, J. M. 1964. Introduction to structural dynamics, McGraw-Hill College.
Brown, A. J. 2002. Collision scenarios and probabilistic collision damage. Marine Structures, 15, 335364.
BV 2014. Rules for the Classification of Offshore Units. Bureau Veritas.
Cho, S. R., Kim, J. M. & Roh, M. I. 2014. Towards more rational design of ship structures against collisions.
Quantitative Methods and Socio-Economic Applications in GIS, 187.
Christou, M. & Konstantinidou, M. 2012. Safety of offshore oil and gas operations: Lessons from past accident
analysis. Institute for Energy and Transport.
Czujko, J. 2001. Design of offshore facilities to resist gas explosion hazard, engineering handbook, Trondheim,
Norway, CorrOcean ASA.
Czujko, J. & Paik, J. K. 2010. Explosion and fire engineering of FPSOs (Phase II): Definition of gas explosion
design loads. Final Report No. EFEF JIP-04-R1, Research Institute of Ship and Offshore Structural Design
Innovation, Pusan National University, Busan, Korea.
Czujko, J. & Paik, J. K. 2014. A new method for ALS design of thin walled structures subjected to hydrocarbon
explosion loads. the ICTWS 2014 7th International Conference on Thin-Walled Structures. Busan, Korea.
Ditlevsen, O. & Madsen, H. 1996. Structural reliability methods, Chichester, John Wiley & Sons.
DNV-OS-A101 2014. Safety Principles and Arrangements. Det Norske Veritas.
DNV-OS-C101 2014. General design of offshore steel structures (LFRD Method). Det Norske Veritas.
DNV-RP-C204 2010. Desgn against accidental loads. Det Norske Veritas.
DNV-RP-C208 2013. Determination of structural capacity by non-Linear FE analysis methods. Det Norske Veritas.
DNV-RP-F107 2010. Risk Assessment of Pipeline Protection. Det Norske Veritas.
DNV 1992. Structural reliability analysis of marine structures. Classification Note 30.6. Det Norske Veritas.
DNV 1999. Ultiguide: best practice guidelines for use of non-linear analysis methods in documentation of ultimate
limit states for jacket type offshore structures. Det Norske Veritas.
Efthymiou, M., Van De Graaf, J. W., Tromans, P. S. & Hines, I. M. 1997. Reliability-based criteria for fixed steel
offshore platforms. Journal of Offshore Mechanics and Arctic Engineering, 119, 120124.
Ehlers, S. 2010b. Strain and stress relation until fracture for finite element simulations of a thin circular plate. ThinWalled Structures, 48, 18.
577
Ehlers, S. & Tabri, K. 2012. A combined numerical and semi-analytical collision damage assessment procedure.
Eurocode 1999a. Fire-resistance Tests General Requirements. European Committee for Standardization CEN.
Eurocode 1999b. Fire-resistance Tests Alternative and Additional Procedures. European Committee for
Standardization CEN.
Eurocode 2004. Hot Rolled Products of Structural Steels- Part 3. Technical Delivery Conditions for
Normalized/nomalized Rolled Weldable Fine Grain Structural Steels. European Committee for Standardization
CEN.
Eurocode 2005. Design of steel Structures- Part 1-2. General Rules- Structural Fire Design. European Committee
for Standardization CEN.
Eurocode 2007. Design of aluminum structures- Part 1-2. Structural Fire Design. European Committee for
Standardization CEN.
FABIG 1992. Interim Guidance Notes for the Design and Protection of Topside Structures against Explosion and
Fire. Steel Construction Institute.
FABIG 2001. TN-6. Design Guide for Steel at Elevated Temperatures and High Strain rates. Steel Construction
Institute.
FABIG 2009. TN-11. Fire loading and Structural Response. Steel Construction Institute.
FABIG 2014. TN-13. Design Guidance for Hydrocarbon Fires. Fire And Blast Information Group.
Gao, Z., Hu, Z., Wang, G. & Jiang, Z. 2014. An analytical method of predicting the response of FPSO side
structures to head-on collision. Ocean Engineering, 87, 121135.
GL 2013. IV Rules for the Certification and Constructio of Industrial services. Part 6 Offshore Technology, 3 Fixed
Offshore Structures. Part 7 Offshore Substations, 3 General Safety. Germanicher Lloyd
Goerlandt, F., Sthlberg, K. & Kujala, P. 2012. Influence of impact scenario models on collision risk analysis.
Ocean engineering, 47, 7487.
Haltom, S. S., Kyriakides, S. & Ravi-Chandar, K. 2013. Ductile failure under combined shear and tension.
International Journal of Solids and Structures, 50, 15071522.
Haris, S. & Amdahl, J. 2013. Analysis of shipship collision damage accounting for bow and side deformation
interaction. Marine Structures, 32, 1848.
Hellan, . 1995. Non-linear pushover and cyclic analyses in ultimate limit state design and reassessment of tubular
steel offshore structures. PhD Thesis, Norwegian Institute of Technology.
Heywood, M. & Martland, R. The suitability of the Biggs method for the dynamic analysis of axially restrained blast
loaded beams and panels. Conference of Safety on offshore installations, 1999.
Hogstrm, P., Ringsberg, J. W. & Johnson, E. 2009. An experimental and numerical study of the effects of length
scale and strain state on the necking and fracture behaviours in sheet metals. International Journal of Impact
Hopperstad, O. S., Brvik, T., Langseth, M., Labibes, K. & Albertini, C. 2003. On the influence of stress triaxiality
and strain rate on the behaviour of a structural steel. Part I. Experiments. European Journal of MechanicsA/Solids, 22, 113.
Huser, A. 2013. Probabilistiske brannrisikoanalyser til bruk i design. Det Norske Veritas.
Huser, A. & Vianna, S. 2010. Fire CFD modelling applied to offshore design. Det Norske Veritas.
Hussein, A. & Soares, C. G. 2009. Reliability and residual strength of double hull tankers designed according to the
new IACS common structural rules. Ocean Engineering, 36, 14461459.
IACS 2006. Background document of IACS common structural rules for double hull oil tankers. Section 9/1
DESIGN VERIFICATION HULL GIRDER ULTIMATE STRENGTH. International Association of Classification
Societies.
IACS 2014. Harmonized CSR TB report hull girder residual strength. International Association of Classification
Societies.
IPC 2000. Chartek 7 fireproofing fire exposure case studies. International Protective Coatings.
ISO-834-1 1999. Fire-resistance Tests Elements of Building Construction Part 1: General Requirements.
International Organization for Standardization.
ISO-834-2 2009. Fire-resistance Tests Elements of Building Construction Part 2: Guidance on Measuring
Uniformity of Furnace Exposure on Test Samples. International Organization for Standardization.
ISO-17776 2000. Petroleum and natural gas industries Offshore production installations Guidelines on tools and
techniques for hazard identification and risk assessment. International Organization for Standardization.
ISO-19900 2013. Petroleum and natural gas industries General requirements for offshore structures. International
Organization for Standardization.
ISO-19902 2007. Fixed steel offshore structures. International Organization for Standardization.
ISO-19906 2010. Arctic offshore structures. International Organization for Standardization.
ISSC 2006. V.1. Collision and Grounding. The 16th Int. Ship and Offshore Structures Congress 2.
ISSC 2012. Damage Assessment following Accidents. The 16th Int. Ship and Offshore Structures Congress 2.
Kim, D. K., Kim, H. B., Mohd, M. H. & Paik, J. K. 2013. Comparison of residual strength-grounding damage index
diagrams for tankers produced by the ALPS/HULL ISFEM and design formula method. International Journal of
Naval Architecture and Ocean Engineering, 5, 4761.
Kim, M. G. & Kim, G. S. 2015. Probabilistic Design Accidental Load for Fire Accidents in Offshore Topside
Structures. ESREL2015: European Safety and Reliability Conference. Switzerland.
578
Kim, M. K. & Kim, G. S. Determination of the Design Load for Structural Safety Assessment against Gas Explosion
in Offshore Topside. PSAM12-Probabilistic Safety Assessment and Management, 2014 Hawaii.
Kodur, V., Dwaikat, M. & Fike, R. 2010. High-Temperature Properties of Steel for Fire Resistance Modeling of
Structures. Journal of Materials in Civil Engineering, 22, 423434.
Krgesaar, M. & Romanoff, J. 2013. Influence of softening on fracture propagation in large-scale shell structures.
International Journal of Solids and Structures, 50, 39113921.
Krgesaar, M., Romanoff, J. & Tabri, K. 2014. Simulating ductile fracture in large scale shell structures influence
of mesh size and damage induced softening. In revision.
Kvitrud, A. Modifications of the PSA regulations based on case studies of stability accidents. ASME 2013 32nd
International Conference on Ocean, Offshore and Arctic Engineering, 2013. American Society of Mechanical
Engineers, V001T01A005-V001T01A005.
Lademo, O. G., Hopperstad, O. S., Berstad, T. & Langseth, M. 2005. Prediction of plastic instability in extruded
aluminium alloys using shell analysis and a coupled model of elasto-plasticity and damage. Journal of materials
processing technology, 166, 247255.
Langseth, M., Hopperstad, O. S. & Hanssen, A. G. 1998. Crash behaviour of thin-walled aluminium members. Thinwalled structures, 32, 127150.
Ledin, H. S. 2002. A review of the state of the art in gas explosion modeling. Buxton, U.K: Health and Safety
Laboratory.
Lees, F. P. 1996. Loss prevention in the process industries, Butterworth Heinemann.
Levander, K. Goal Based Ship Safety Application in large cruise ship design. 5th International Conference on
Collision and Grounding of Ships (ICCGS), 2010 Espoo, Finland.
Lou, Y., Huh, H., Lim, S. & Pack, K. 2012. New ductile fracture criterion for prediction of fracture forming limit
diagrams of sheet metals. International Journal of Solids and Structures, 49, 36053615.
LR 2014a. Guidance notes for collision analysis. Lloyds Register.
LR 2014b. Guidance notes for fire loadings and protection. Lloyds Register.
LR 2014c. Guidelines for the calculation of probabilistic explosion loads. Rules and Regulations for the
classification of offshore units. Lloyds Register.
Ls-Dyna 2014. Users manual for LS-DYNA, Version 971, Livermore Software Technology Corporation.
Moan, T. 2007. Development of accidental collapse limit state criteria for offshore structures. Development of
Accidental Collapse Limit State Criteria for Offshore Structures. Stanfrod University.
Moan, T. 2009. Development of accidental collapse limit state criteria for offshore structures. Structural Safety, 31,
124135.
Moan, T., Amdahl, J., Wang, X. & Spencer, J. Risk assessment of FPSOs with emphasis on collision. Annual
Meeting, the Society of Naval Architects and Marine Engineers (SNAME), 2002.
Montewka, J., Ehlers, S., Goerlandt, F., Hinz, T., Tabri, K. & Kujala, P. 2014. A framework for risk assessment for
maritime transportation systemsA case study for open sea collisions involving RoPax vessels. Reliability
Engineering & System Safety, 124, 142157.
MSL-Engineering 2000. Collision Resistance of Ship-Shaped Structures to Side Impact, HSE Books.
NORSOK-N-001 2010. Structural design. Standard Norway.
NORSOK-N-003 2007. Action and action effects. Standard Norway.
NORSOK-N-004 2004. Design of Steel Structure. Standard Norway.
NORSOK-N-006 2009. Assessment of structural integrity for existing offshore load-bearing structures. Standard
Norway.
NORSOK-S-001 2008. Technical Safety. Standard Norway.
NORSOK-Z-013 2010. Risk and Emergency Preparedness Assessment. Standard Norway.
OGP 2010a. Major accidents. International Association of Oil & Gas Producers.
OGP 2010b. Mechanical lifting failures. International Association of Oil & Gas Producers.
OGP 2010c. Ship / Installation collisions. International Association of Oil & Gas Producers.
OGP 2014. Reliability of offshore structures. Current design and potential incosistiencies. International Association
of Oil and gas Producers.
Oil&GasUK 2007. Fire and Explosion Guidelines. Oil & Gas UK.
OREDA 2009. Offshore Reliability Data Handbook, SINTEF.
Pahos, S. J. 2009. Reliability-based vulnerability requirements for asymmetric threats in the design of ship
structures. University of Strathclyde.
Paik, J. K. & Czujko, J. 2012. Engineering and Design Disciplines Associated with Management of Hydrocarbon
Explosion and Fire Risks in Offshore Oil and Gas Facilities. Transactions of Society of Naval Architects and
Marine Engineers, 120.
Paik, J. K., Czujko, J., Kim, J. H., Park, S. I., Islam, M. S. & Lee, D. H. 2013. A new procedure for the nonlinear
structural response analysis of offshore installations in fires. Transactions of Society of Naval Architects and
Marine Engineers, 121.
Paik, J. K., Czujko, J., Kim, S. J., Lee, J. C., Kim, B. J., Seo, J. K. & Ha, Y. C. A new procedure for the nonlinear
structural response analysis of offshore installations in explosions. The Society of Naval Architects and Marine
Engineers (SNAME 2014), 2014 Houston, USA.
Paik, J. K., Kim, D. K., Park, D. H., Kim, H. B. & Kim, M. S. 2012. A new method for assessing the safety of ships
damaged by grounding. International Journal of Maritime Engineering, 154, A1A20.
579
Pedersen, P. T. 2010. Review and application of ship collision and grounding analysis procedures. Marine
Pedersen, P. T. & Zhang, S. 1998. On Impact Mechanisms in Ship Collisions. Journal of Marine Structures 11,
403451.
Qu, X., Meng, Q. & Suyi, L. 2011. Ship collision risk assessment for the Singapore Strait. Accident Analysis &
Prevention, 43, 20302036.
Saydam, D. & Frangopol, D. M. 2013. Performance assessment of damaged ship hulls. Ocean Engineering, 68, 6576.
Scharrer, M., Zhang, L. & Egge, E. D. 2002. Collision calculations in naval design systems. Report Nr. ESS.
Sprague, C. M. & Dolph, B. L. 1996. Performance-Based Firesafety Analysis of the Uscgc Vindicator (WMEC3).
Transactions of Society of Naval Architects and Marine Engineers, 104.
Storheim, M. & Amdahl, J. 2014. Design of offshore structures against accidental ship collisions. Marine Structures,
37, 135172.
Tabri, K. & Broekhuijsen, J. Influence of ship motions in the numerical prediction of ship collision damage. 3rd Int.
Conf. on Mar. Struct.MARSTRUCT, 2011. 3917.
Tang, M. J. & Baker, Q. A. 1999. A new set of blast curves from vapor cloud explosion. Process Safety Progress,
18, 235240.
Teixeira, A. P. & Soares, C. G. 2010. Reliability assessment of intact and damaged ship structures. Advanced Ship
Design for Pollution Prevention 9780.
Ton Vrouwenvelder, A. C. W. M. 2014. Reliability based structural design, Safety, Reliability and Risk
Management, London, Taylor & Francis Group.
Trnqvist, R. 2003. Design of crashworthy ship structures. Technical University of Denmark Kgns Lyngby,
Denmark.
UL 1994. Standard for rapid rise fire tests of protection materials for structural steel. Underwriters Laboratories.
UN 2013. The european agreement concerning the international carriage of dangerous goods by inland water-ways.
Part 9. Rules for construction.
Usfos 2013. Users manual for USFOS, Version 8-7, USFOS AS.
Vinnem, J.-E. 2014. Offshore risk assessment: principles, modelling and applications of QRA studies, Springer
Science & Business Media.
Walters, C. 2014. Framework for adjusting for both stress triaxiality and mesh size effect for failure of metals in
shell structures. International Journal of Crashworthiness, 19, 112.
Wierzbicki, T., Bao, Y., Lee, Y. W. & Bai, Y. 2005. Calibration and evaluation of seven fracture models.
International Journal of Mechanical Sciences, 47, 719743.
Youssef, S., Ince, S. T., Kim, Y. S., Paik, J. K., Cheng, F. & Kim, M. S. 2014a. Quantitative risk assessment for
collisions involving double hull oil tankers. Int J Maritime Eng, 156, 157174.
Youssef, S., Kim, Y. S., Paik, J. K., Cheng, F. & Kim, M. S. 2014b. Hazard identification and probabilistic scenario
selection for shipship collision accidents. Int J Maritime Eng, 156, 6180.
9.
9.1
ANNEX 1. MATERIAL MODELS FOR NON-LINEAR FINITE ELEMENT

ANALYSIS
Introduction
Offshore structures exposed to hazards as defined above may undergo highly non-linear structural
deformations, including rupture. Therefore, finite element analyses of these events require the input of
appropriate material relations including failure representing the local material behaviour. Depending on the
hazard to be analysed and the materials found on the offshore structures a selection of recommended
material models can be made, see Table 13. The physical origin of these material models will be briefly
presented, followed by numerical implementation possibilities as well as comments, hints and shortcomings
arising from the use of those models as well as concerns of guidelines and standards. However,
Table 13. Recommended material models and associated hazards.
Material
Foam, Isolator,
Risers, umbilical
Steel Aluminium
Ice Air Water Explosives
Composite Concrete Seabed
Hazard
Rubber
or power cables
Hydrocarbon
explosions
Hydrocarbon fires
Underwater
explosions
Wave Impact
Water-In-Deck
Dropped Objects
Ship Impact
Earthquakes
Ice, Iceberg
Flooding
- recommended, - recommended where applicable
580
hazard simulations utilizing the recommended material models and input parameters can be used for basic
physical checks, but they may not be applicable in general.
The material modelling represents a crucial part of all numerical simulations, because it predefines how
the material is assumed to behave during the simulations. Hence, the ability of the material model to
represent the physical behaviour accurately directly influences the accuracy of the simulation results and
their reliability. Furthermore, the correct physical behaviour may be represented well by the underlying
assumptions of the material model, because it can correspond well to the physical experiment done to obtain
the properties of the material in question. However, whether or not this experiment or the correspondence
represents the true material behaviour remains often a question, e.g. a classical tensile experiment is a
material test by agreement even though a structural test is carried out. Hence, the utilization of such
experimentally based material models using small structural tests can lead to inconsistent results when
applied to general structures. Furthermore, it remains often questionable whether the obtained material
model corresponds to the discrete mathematical model, i.e. the finite element mesh, of the structure to be
analysed. Hence, a material model should be unique and usable for any mesh size or conditions and should
therefore not affect the results with a change in discretization of the simulation domain. In the past, often the
term true material model was utilized, which is however misleading as it implies that it is true by all
means and could be universally applied. In fact, all material measures are true with respect to their
determination scale, i.e. the engineering measure obtained by a tensile experiment is true with respect to the
specimens gauge length.
Hence, this chapter seeks to provide appropriate guidance to identify the material model to be used
with the associated hazard according to Offshore structures exposed to hazards as defined above may
undergo highly non-linear structural deformations, including rupture. Therefore, finite element analyses
of these events require the input of appropriate material relations including failure representing the local
material behaviour. Depending on the hazard to be analysed and the materials found on the offshore
structures a selection of recommended material models can be made, see Table 13. The physical origin of
these material models will be briefly presented, followed by numerical implementation possibilities as
well as comments, hints and shortcomings arising from the use of those models as well as concerns of
guidelines and standards. However, hazard simulations utilizing the recommended material models and
input pa-rameters can be used for basic physical checks, but they may not be applicable in general.
Table 13 in such a way that it is consistent with the discretized, respectively meshed, simulation
domain. Furthermore, engineering based best practices are provided as well as the associated
shortcomings. The nomenclature of the numerical implementation used in the material input cards can be
found in Hallquist (2007). The effects the material models account for, e.g. strain rate, temperature or
damage criteria, will be provided alongside a selection of references relevant to the given material.
Thereby, this database of material models will clarify common questions and uncertainties associated
with the use of material models.
9.2
Guidelines and standards
ISO 19902 Ed 1 requires that the expected non-linear effects, including material yielding, buckling of
structural components and pile failures, should be adequately modelled and captured. Strain rate effects
should be considered as well as temperature dependency. NORSOK standard N-003 and DNV
Recommended Practices DNV-RP-C204 suggest the use of the temperature dependent stress-strain
relationships given in NS-ENV 1993 1-1, Part 1.2, Section 3.2. To account for the effect of residual
stresses and lateral distortions compressive members should be modelled with an initial, sinusoidal
imperfection with given amplitudes for elastic-perfectly plastic material and elastic-plastic material
models. General class rules or CSR commonly state that an appropriate material model should be used;
possibly in the form of a standard power law based material relation for large deformation analysis of
steel structures. Additionally, some specify critical strain values to be used independent of the mesh size,
which should, however, be sufficient, may be specified.
Hence, these guidelines and standards fail to provide a clear guidance for the analyst and may easily
lead to diverse results simply by choosing different, yet not necessarily physically correct, material
parameters.
9.3
Material model database
9.3.1
Steel
Commonly, the nonlinear material behaviour is selected in the form of a power law; see, for example,
Alsos et al. (2009) and Ehlers et al. (2008). The power law parameters can be obtained from standard
tensile experiments; see Paik (2007). However, with this approach agreement between the numerical
581
simulation and the tensile experiment can only be achieved by an iterative procedure for a selected
element size chosen a priori. Hence, the procedure needs to be repeated if the element size is changed.
Furthermore, the determination of the material relation alone does not necessarily suffice, as the failure
strain, i.e. the end point of the stress versus strain curve, depends in turn on the material relation.
However, a significant amount of research has been conducted to describe criteria to determine the failure
strain, for example by Trnqvist (2003), Scharrer et al. (2002), Alsos et al. (2008), and to present their
applicability (e.g. Tabri et al. (2007) or Alsos et al. (2009). However, all of these papers use a standard or
modified power law to describe the material behaviour, and none of these papers identifies a clear relation
between the local strain and stress relation and the element length.
Relations to obtain an element length-dependent failure strain value are given by Peschmann (2001),
Scharrer et al. (2002), Trnqvist (2003), Alsos et al. (2008) and Hogstrm et al. (2009). These curvefitting relations, known as Barbas relations, are obtained on the basis of experimental measurements.
However, they define only the end point of the standard or modified power law. Hence, Ehlers et al.
(2008) conclude that the choice of an element length-dependent failure strain does not suffice in its
present form.
Therefore, Ehlers and Varsta (2009) and Ehlers (2010b) presented a procedure to obtain the strain and
stress relation of the materials, including failure with respect to the choice of element size using optical
measurements. They introduced the strain reference length, which is a function of the discrete pixel
recordings from the optical measurements and corresponds to the finite element length. As a result, they
present an element length dependent material relation for NVA grade steel including failure, see Figure 9.1.
Moreover, Ehlers et al. (2012) identified that a constant strain failure criterion suffices for
crashworthiness simulations of ship structures and that the strain rate sensitivity of the failure strain and
ultimate tensile force is less than three per cent, see Figure 9.2. Hence, for moderate displacement speeds
the strain rate influence is negligible.
An example input card following the LS-DYNA nomenclature for a piece wise linear material
(mat_24) is given in Table 14.
Table 14. Piece wise linear steel material model.
However, the material behaviour, that is the change in the yield stress, at higher strain rates, , can be
calculated according to the Cowper-Symonds relation
&
where C, p are the strain rate parameters and may be chosen as 40.4/sec and 5 for mild steel, respectively.
Additionally, effects on elevated temperatures may be accounted for by scaling the global yield stress as a
function of the temperature, see Figure 61. The increase in yield- and ultimate strength at cryogenic
temperatures, i.e. -100 and -163 C, is presented by Yoo et al. (2011) for mild stainless steel.
a)
b)
Figure 62. NVA grade steel: measured local strain and stress relation (a) and failure strain (b) (Ehlers, 2010a).
582
1
1
0.8
Yield stress scale factor
True failure strain
0.975
0.95
0.925
0.6
0.4
0.2
0.9
16
8
4
Displacement speed [mm/min]
32
Figure 63. Influence of the displacement speed on

the failure strain (Ehlers et al., 2012).
100
200
300
400
500
600
700
800
900
Temperature [C]
Figure 64. Global yield stress scale factor versus

temperature for mild steel.
Definitial of thermal properties of materials requires additional keycards compared to the definition of
basic material properties. A working example is presented in Table 15. *PART keyword should include
the definitions for the basic material properties (marked with 355 in Table 9.3) and additional thermal
material definition (marked with 2 in Table 15). Basic material definition is via
*MAT_ELASTIC_PLASTIC_THERMAL, that defines gives the mechanical properties such as steel
density and temperature dependent Youngs moduli, Poissons ratio, coefficients of thermal expansion,
yield stresses and plastic hardening moduli. A maximum of eight temperatures with the corresponding
data can be defined with a minimum of two points is needed. Keyword
*MAT_THERMAL_ISOTROPIC_TD_LC allows additional isotropic thermal properties such as steel
conductivity (tclc) and steel specific heat (hclc) to be specified by load curves. Finally, keyword
*MAT_ADD_THERMAL_EXPANSION is used to apply the thermal expansion to a certain part
according to a specified curve (curve no. 100 applied to the part no. 1 in Table 15). It should be noted that
the
latter
overwrites
the
thermal
expansion
coefficients
defined
in
*MAT_ELASTIC_PLASTIC_THERMAL keyword.
Table 15. Definition of thermal properties for steel (temperature in K).
*PART
Part 1 (Section 1, MAT 355, thermal material 2)
$# pid secid
mid eosid hgid grav adpopt tmid
1
1
355
0
0
0
0
2
*MAT_ELASTIC_PLASTIC_THERMAL
$# mid
ro
355 7.8500e-9
$# t1
t2
t3
t4
t5
t6
t7
t8
273
293
373
673
773
973 1073 1473
$# e1
e2
e3
e4
e5
e6
e7
e8
2.1e5 2.1e5 2.1e5 1.47e5 1.26e5 2.73e5 1.89e5
201
$# pr1
pr2
pr3
pr4
pr5
pr6
pr7
pr8
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
$# alpha1 alpha2 alpha3 alpha4 alpha5 alpha6 alpha7 alpha8
0
0
0
0
0
0
0
0
$# sigy1 sigy2 sigy3 sigy4 sigy5 sigy6 sigy7 sigy8
355
355
355
355
277
82
39
1
$# etan1 etan2 etan3 etan4 etan5 etan6 etan7 etan8
200
200
200
200
200
200
200
200
*MAT_THERMAL_ISOTROPIC_TD_LC
$# mid
tro tgrlc tgmult
2 7.8500E-9
hclc
tcl
3
4
*DEFINE_CURVE_TITLE
Steel CONDUCTIVITY (TCLC), temperature in K
$# lcid sidr
sfa
sfo offa offo dattyp
4
0 1.0000 1.000000
273 0.000
0
0,54
583
....
1400,27.3
*DEFINE_CURVE_TITLE
Steel SPECIFIC HEAT (HCLC), temperature in K
$# lcid sidr
sfa
3
0 1.0000 1.000000
273 0.000
0
0,4.25E+08
....
1400,6.50E+08
*MAT_ADD_THERMAL_EXPANSION
1,100,1
*DEFINE_CURVE_TITLE
Steel THERMAL EXPANSION, temperature in K
$# lcid sidr
sfa
100
0 1.0000 1.000000
273 0.000
0
0,1.20E-05
....
1400,2.00E-05
9.3.2
Aluminium
Various thin-walled aluminium structures under crash behaviour, i.e. large deformations including
rupture, have been analysed experimentally and numerically in the past.
Langseth et al. (1998) uses an elasto-plastic material model with isotropic plasticity following the von
Mises yield criterion and associated flow rule, see Berstad et al. (1994). Strain rate effects are often
neglected for aluminium alloys, such as AA6060, in the strain rate range of 104 to 103 s-1, see for
example Lindholm et al. (1971). As a result, Langseth et al. (1998) are able to obtain good
correspondence in terms of deformed shape, and shape of the force-displacement curve.
However, if high strain rates are to be expected, then the yield stress scaling according to CowperSymonds may be used. Ngre et al. (2004) study the crack extension in aluminium welds using the
GursonTvergaardNeedleman (GTN) model and obtain reasonable correspondence in terms of force
versus crack mouth opening displacement (CMOD). However, the GTN model requires a vast amount of
input parameters whose physical origin cannot be directly provided. Furthermore, Ngre et al. (2004) use
8-node brick elements, which are not suitable for large complex structures at present. Hence, from an
engineering viewpoint this model does not suffice.
Lademo et al. (2005) utilize a coupled model of elasto-plasticity and ductile damage based on Lemaitre
and Lippmann (1996) using the critical damage as an erosion criterion. They are able to simulate
aluminium tensile experiments numerically with very good agreement using co-rotational shell elements
and an anisotropic yield criterion Yld96 proposed by Barlat et al. (1997).
Such advanced material models can be easily implemented into numerical codes, and further increase
in yield and ultimate strength at cryogenic temperatures, i.e. -100 and -163 C, can be considered
following the results by Yoo et al. (2011) for mild aluminium. Furthermore, a strain reference lengthbased approach using optical measurements as proposed by Ehlers (2010b) for steel may be used to
obtain a consistent material relationship. However, for most analyses a consistent determination of the
global material behaviour, see Figure, together with a Von Mises yield criterion will suffice.
An example input card following the LS-DYNA nomenclature for a piece wise linear material
(mat_24) is given in Table 16.
Table 16. Piece wise linear aluminium material model
584

300
250
Stress [MPa]
200
150
100
50
0
0
0.02
0.04
0.06
Strain []
0.08
0.1
0.12
Figure 65. Example of a global strain versus stress curve from experiments.
9.3.3
Foam, Isolator, Rubber
Gielen (2008) presents an isotropic polyvinyl chloride (PVC) foam model, which exhibits elasto-damage
behaviour under tension and elasto-plastic behaviour under compression. His damage model is consistent
with the physical behaviour of the foam, a full-scale application and verification is however missing.
Cui et al. (2009) present a model for uniform foam based on Schraad and Harlow (2006) for disordered
cellular materials under uni-axial compression. As a result, they obtain various influencing parameters
affecting the energy absorption capacity under impact. Hence, functionally graded foams may be used to
increase impact resistance.
In the case of rubber, a simplified rubber/foam material model (mat_181) may be used, which is
defined by a single uni-axial load curve or by a family of uni-axial curves at discrete strain rates, see
Figure 64. An example input card following the LS-DYNA nomenclature for such rubber material is
given in Table 17.
Table 17. Simplified rubber/foam material model.
2500
Force [N]
2000
1500
1000
500
0
0
10
20
30
Displacement [mm]
40
50
Figure 66. Exemplary force-displacement curve for rubber referenced as LC/TBID in mat_181.
9.3.4
Ice
One of the main difficulties when modelling ice is the prediction of ice failure, i.e. fracture, under loading
at temperatures around the melting point of the ice. Thus the local ice-structure interaction includes
transitions between the different phases. The failure process of ice begins when the edge of the moving
ice hits the structure. This contact induces loads to the edge of the ice causing a stress state in the ice.
When the stresses exceed the strength of ice, it fails. Ice becomes ductile with visco-elastic deformations
during low loading rates and brittle during high loading rates.
Polojrvi and Tuhkuri (2009) developed specialized simulations tools utilizing the boundary element
method, whereas Forsberg et al. (2010) utilizes the cohesive element method (CEM) to model ice failure.
The latter is however of highly stochastic, or even random, nature and eventually results in reasonable
agreement if experimental validation data becomes available.
However, Liu et al. (2011) treat the ice in a coupled dynamic ship ice berg collision as an isotropic
material, see Riska (1987), using the well-known Tsai-Wu strength criterion, see Tsai and Wu (1971). As
585
a result, the obtained numerical results give an indication of the structural damage of the ship structure.
However, their model erodes the ice at failure in an unphysical fashion resulting in purely numerical
pressure fluctuation in the contact surface.
Therefore, the underlying material models and ice properties are in need to be defined consistently to
account for the possible scatter and thereby to result in reliable design methods for ships and offshore
structures. Hence, unless material model data is not available explicitly for tension and compression
including an appropriate failure criterion for brittle ice failure based on micro-crack growth, a simple
elastic model may be employed. The latter is however only valid to some extent, if, e.g. the flexural
strength of an ice sheet is of interest.
Therefore, as a first attempt, ice may be modelled as a volumetric body following non-iterative
plasticity with a simple plastic strain failure model (mat_13). However, therein the yield- and failure
stress is note rate or pressure dependent and the temperature is assumed constant. An example input card
following the LS-DYNA nomenclature for Baltic Sea ice is given in Table 18.
Table 18. Simplified ice material model.
9.3.5
Air
For numerical simulations of structures subjected to underwater explosions, where the target is airbacked, the air needs to be modelled. The main material parameters are the mass density and the equation
of state (EOS). The latter can be expressed as a linear polynomial defining the pressure in the gas as a
linear relationship with the internal energy per initial volume. The ideal gas EOS is an alternative
approach to the linear polynomial EOS with a slightly improved energy accounting algorithm. In most
cases, the mass density is the only parameter defined for the air. The same material properties were used
in Trevino (2000) and Webster (2007).
An example input card for air following the LS-DYNA nomenclature is given in Table 19 according to
Webster (2007).
Table 19. Air material model.
The EOS example input following the LS_DYNA nomenclature is given in Table 20 according to
Webster (2007) in the most common form, which defines the parameters such that it is an ideal gas
behaviour.
Table 20. Linear polynomial equation of state for air.
Do (2009) describes the calculation process of e0, which can be used to define an initial pressure within
the air. Additionally, an example input card for the ideal gas EOS following the LS-DYNA nomenclature
is given in Table 21 according to Martec-Limited (2007).
Table 21. Ideal gas equation of state for air.
586
The ideal gas EOS is the equivalent of the linear polynomial with the C4 and C5 constants set to a value
of ( 1).
9.3.6
Water
When conducting simulations of structures subjected to underwater explosions, water models are
required.
The primary mechanical property to be defined is the mass density and in some cases the pressure cutoff and dynamic viscosity coefficient is needed. The cut-off pressure is defined to allow the material to
numerically cavitate when under tensile loading. This is usually defined as a very small negative number,
which allows the material to cavitate once the pressure goes below this value.
Additionally, the equation of state (EOS) needs to be defined, most commonly as a Gruneisen EOS
with cubic shock-velocity-particle velocity defining the pressure for compressed materials. The constants
in the Gruneisen EOS are found from the shock wave velocity versus particle velocity curve. Two
example input cards following the LS-DYNA nomenclature for water (mat_009) are given according to
Trevino (2000) and Webster (2007) in table 22 and 23 respectively.
Table 22. Material model for water (Trevino, 2000).
Table 23. Material model for water (Webster, 2007).
Additionally, Gruneisen EOS is the most commonly used EOS for defining the water behaviour with
underwater explosion events. An example input card following the LS-DYNA nomenclature is given in
Table 24 according to Webster (2007).
Table 24. Equation of state for water.
9.3.7
Explosives
An explosive material requires two keywords to define the behaviour of the material. These include the
material keyword and the equation of state (EOS). The mechanical properties to be considered are the
mass density, the detonation velocity in the explosive and the Chapman-Jouguet pressure. Furthermore,
the bulk modulus, shear modulus and yield stress may be required depending on the model.
For the EOS, there are three possibilities to define the pressure for the detonation products. All of these
EOS define the pressure as a function of the relative volume and the internal energy per initial volume.
The most commonly used EOS for explosive behaviour is the standard Jones-Wilkins-Lee (JWL). This
EOS was modified by Baker and Stiel (1997) and has the added feature of better describing the highpressure region above the Chapman-Jouguet state.
In addition to the material and EOS definitions in LS-DYNA, the INITIAL_DETONATION keyword
is required to define the position and time of the initiation of the detonation process. This is the point at
which the detonation initiates and the time for the remaining explosive to detonate is determined by the
distance to the centre of the element divided by the detonation velocity. In the material definition for
MAT_HIGH_EXPLOSIVE_BURN (mat_008) the value of BETA determines the type of detonation. If
beta burn is used, any compression of the explosive material will cause detonation. For programmed burn,
the explosive material can act as an elastic perfectly plastic material through the definition of the bulk
modulus; shear modulus, and the yield stress. In this case, the explosive must be detonated with the
INITIAL_DETONATION keyword.
587
An example input card following the LS-DYNA nomenclature for TNT (mat_008) is given in Table 25
according to Webster (2007).
Table 25. Explosive material model
Furthermore, the most commonly used Jones-Wilkens-Lee EOS is given in Table 26 according to the LSDYNA nomenclature (Webster, 2007).
Table 26. Equation of state for the explosive material model
Keywords *LOAD_BLAST and *LOAD_BLAST_ENHANCED allow indirect modelling of the

explosive and the propagation of blast wave without the need of actual discretization of the explosive or
the air mesh around it. These keywords allow to define an airblast function for the application of pressure
loads due to explosives described via equivalent mass of TNT. While *LOAD_BLAST only models the
incident wave, the *LOAD_BLAST_ENHANCED includes enhancements for treating reflected waves,
moving warheads and multiple blast sources. The loads are applied to facets defined with the keyword
*LOAD_BLAST_SEGMENT. Example of indirect modelling is given in Table 27.
Table 27. Indirect modelling of explosive loading.
*LOAD_BLAST_ENHANCED
$# bid
m
xbo
ybo
zbo
tbo unit blast
1
30 -250000
0 6850 -0.53
5
2
$# cfm
cfl
cft
cfp nidbo death negphs
2.205e3 3.28E-3 1e+3
145
0
9.3.8
Risers, Umbilical or Power Cable
What all these structures have in common is the fact that they are typically very long, therefore slender.
Their global mechanical properties to be defined are the bending-, torsional- and axial stiffness.
Furthermore, the main aspect to be covered when modelling such structures is their stiffness dependency
with respect to tension, torsion and curvature, i.e. stick-slip effects.
Therefore, experimental measurements of the global and local behaviour as well as a local analysis of
the cross-section are needed. Typical numerical implementations would utilize elasto-plastic and viscoelastic material models considering friction, contact formulation (lift-off) as well as torsion/rolling effects
on pipes.
Svik (2011) studied the local behaviour of stresses in flexible pipes with a detailed model considering
the cross-section build-up. However, for global analysis of an offshore structure, where the support effect
of the slender structure is of interest, a simpler discretisation using beam elements with local stiffness
properties can be used, see Rustad et al. (2008).
For a typical 8 flexible riser the following global parameters can be found: EI=200 kNm2, EA=7.7 108
N, GIt=5.9 106 Nm2.
An example input card following the LS-DYNA nomenclature for a visco-elastic material (mat_117) is
given in Table 28.
Table 28. Visco-elastic riser material model.
9.3.9
Composites
Composite materials can be of various types, such as classical fibre-reinforced plastics or various stacks
of materials, i.e. sandwich like structures. Therefore, their material parameters are very specific to the
exact type of composite found in the offshore structure.
Menna et al. (2011) simulate impact tests of GFRP composite laminates using shells and provide the
material parameters for a Mat Composite Failure Option Model (mat_059) of LS-DYNA. Feraboli et al.
588
(2011) present an enhanced composite material with damage (mat_054) for orthotropic composite tape
laminates together with a series of material parameters.
Most orthotropic elastic materials can be described until failure according to:
=
where is the compliance matrix besides the six stress and strain components. Hence, the compliance
matric can be composed of the extensional stiffness coefficients, the extensional-bending stiffness
coefficients and the bending stiffness coefficients.
An example input card following the LS-DYNA nomenclature for a composite matrix material
(mat_117) using such compliance matrix formulation is given in Table 29 for an equivalent stiffened
plate.
Table 29. Composite material model.
9.3.10 Concrete
Concrete material requires two keywords to define the behaviour of the material. These include the
material keyword and the equation of state (EOS). The mechanical properties to be considered are the
mass density, the shear modulus and an appropriate measure of the damage, respectively softening. The
EOS describes the relation between the hydrostatic pressure and volume in the loading and unloading
process of the concrete uncoupled from the deviatoric response. These parameters are typically obtained
by experimental testing of the concrete under different loading directions and rates. Thus, the damage
includes strain-rate effects.
Markovich et al. (2011) present a calibration model for a concrete damage model using EOS for
tabulated compaction and a concrete damage, release 3, model (mat_72r3) and provide the required input
parameters. Tai and Tang (2006) studied the dynamic behaviour of reinforced plates under normal impact
using the JohnsonHolmquist Concrete equivalent strength model with damage and an EOS, which
requires less input parameters and allows for easier implementation with good accuracy.
An example input card following the LS-DYNA nomenclature for concrete material (mat_111) is given
in Table 30 according to Tai and Tang (2006).
Table 30. Concrete material model.
9.3.11 Soil
For some simulations of hazard the seabed has to be included. However, the material parameters for
seabed, respectively soil, are fairly location dependent and may vary significantly within close
proximities. Therefore, it is of utmost importance to obtain experimental data for the site in question.
Typically those experiments should identify the soil stiffness in different directions, the friction, the
break out resistance and a cycling behaviour (trenching). Henke (2011) presents numerical and
experimental results for Niederfelder sand and uses a hypoplastic constitutive model, assuming cohesion
589
less linear elastic behaviour, to achieve good correspondence. Vermeer and Jassim (2011) use a SPH
approach with an elastic-plastic Mohr-Coulomb model to simulate drop anchors and present the utilized
material parameters. Furthermore, solid elements can be used to represent sandy soils or granular
materials following the Mohr-Coulomb behaviour.
An example input card following the LS-DYNA nomenclature for a Mohr-Coulomb material
(mat_173) is given in Table 31 according to the material parameters from Vermeer and Jassim
(2011)Vermeer and Jassmin (2011).
Table 31. Soil material model.
Another alternative for soil modelling is an isotropic material with damage that is available for solid
elements. The model has a modified Mohr-Coulomb surface to determine the pressure dependent peak
shear strength. It was developed for applications involving roadbase soils by Lewis (1999) for the Federal
Highway Administration (FHWA), who extended the work of Abbo and Sloan (1995) to include excess
pore water effects. Table 32 presents an example of FHWA soil model for compressed sand with the
material properties obtained from Wang (2001) and FHWA (2004).
Table 32. Isotropic soil material model with damage.
*MAT_FHWA_SOIL
$# mid
ro nplot spgrav rhowat
vn gammar intrmx
2 2.35e-9
1 2.65 1e-9
1.1
0
4
$#
k
g phimax ahyp
coh eccen
an
et
19
11 0.524 5.37e-4 6.2e-3
0.7
0
0
$# mcont pwd1 pwksk pwd2 phires dint vdfm damlev
0.034
0
0
0 1e-3 0.00001 6e-5 0.99
$# epsmax
2
9.4
References
Abbo, A. & Sloan, S. 1995. A smooth hyperbolic approximation to the Mohr-Coulomb yield criterion. Computers &
structures, 54, 427441.
Alsos, H. S., Amdahl, J. & Hopperstad, O. S. 2009. On the resistance to penetration of stiffened plates, Part II:
Numerical analysis. International Journal of Impact Engineering, 36, 875887.
Alsos, H. S., Hopperstad, O. S., Trnqvist, R. & Amdahl, J. 2008. Analytical and numerical analysis of sheet metal
instability using a stress based criterion. International Journal of Solids and Structures, 45, 20422055.
Baker, E. L. & Stiel, L. I. Improved Quantitative Explosive Performance Prediction Using Jaguar. 1997 Insensitive
Munitions and Energetic Materials Technology Symposium, Tampa, FL, 1997.
Barlat, F., Maeda, Y., Chung, K., Yanagawa, M., Brem, J. C., Hayashida, Y., Lege, D. J., Matsui, K., Murtha, S. J.
& Hattori, S. 1997. Yield function development for aluminum alloy sheets. Journal of the Mechanics and Physics
of Solids, 45, 17271763.
Berstad, T., Langseth, M. & Hopperstad, O. S. Elasto-viscoplastic constitutive models in the explicit finite element
code LS-DYNA3D. Second International LS-DYNA3D conference, San Francisco, 1994.
Cui, L., Kiernan, S. & Gilchrist, M. D. 2009. Designing the energy absorption capacity of functionally graded foam
materials. Materials Science and Engineering: A, 507, 215225.
Do, I. H. P. 2009. LS-DYNA ALE Addvanced Application Course Notes, Livermore, CA.
Ehlers, S. 2010a. A procedure to optimize ship side structures for crashworthiness. Proceedings of the Institution of
Mechanical Engineers, Part M: Journal of Engineering for the Maritime Environment, 224, 111.
Ehlers, S. 2010b. Strain and stress relation until fracture for finite element simulations of a thin circular plate. ThinWalled Structures, 48, 18.
Ehlers, S., Broekhuijsen, J., Alsos, H. S., Biehl, F. & Tabri, K. 2008. Simulating the collision response of ship side
structures: a failure criteria benchmark study. International Shipbuilding Progress, 55, 127144.
Ehlers, S., Tabri, K., Romanoff, J. & Varsta, P. 2012. Numerical and experimental investigation on the collision
resistance of the X-core structure. Ships and offshore structures, 7, 2129.
Ehlers, S. & Varsta, P. 2009. Strain and stress relation for non-linear finite element simulations. Thin-Walled
590
Feraboli, P., Wade, B., Deleo, F., Rassaian, M., Higgins, M. & Byar, A. 2011. LS-DYNA MAT54 modeling of the
axial crushing of a composite tape sinusoidal specimen. Composites Part A: Applied Science and Manufacturing,
42, 18091825.
Forsberg, J., Hilding, D. & Grtner, A. 2010. A homogenized cohesive element ice model for simulation of ice
action a first approach. In: Ehlers, S. & Romanoff, J. (eds.) 5th International Conference on Collision and
Grounding of Ships. Espoo, Finland.
Gielen, A. W. J. 2008. A PVC-foam material model based on a thermodynamically elasto-plastic-damage
framework exhibiting failure and crushing. International Journal of Solids and Structures, 45, 18961917.
Hallquist, J. O. 2007. LS-DYNA keyword user's manual, Version 971, California, Livermore software technology
cooperation.
Henke, S. Numerical and experimental investigations of soil plugging in open-ended piles. Proceedings of the
Workshop Ports for Container Ships of Future Generations, J. Grabe, ed, 2011. 97122.
Hogstrm, P., Ringsberg, J. W. & Johnson, E. 2009. An experimental and numerical study of the effects of length
scale and strain state on the necking and fracture behaviours in sheet metals. International Journal of Impact
Lademo, O. G., Hopperstad, O. S., Berstad, T. & Langseth, M. 2005. Prediction of plastic instability in extruded
aluminium alloys using shell analysis and a coupled model of elasto-plasticity and damage. Journal of materials
processing technology, 166, 247255.
Langseth, M., Hopperstad, O. S. & Hanssen, A. G. 1998. Crash behaviour of thin-walled aluminium members. Thinwalled structures, 32, 127150.
Lemaitre, J. & Lippmann, H. 1996. A course on damage mechanics, Springer Berlin.
Lindholm, U. S., Bessey, R. L. & Smith, G. V. 1971. Effect of strain rate on yield strength, tensile strength and
elongation of three aluminum alloys. J MATER, 6, 119133.
Liu, Z., Amdahl, J. & Lset, S. 2011. Plasticity based material modelling of ice and its application to shipiceberg
impacts. Cold regions science and technology, 65, 326334.
Markovich, N., Kochavi, E. & Ben-Dor, G. 2011. An improved calibration of the concrete damage model. Finite
Elements in Analysis and Design, 47, 12801290.
Martec-Limited 2007. Numerical Study of Soil Modelling Approaches using LS-DYNA. Unclassifed: Defence
R&D Canada Valcartier.
Menna, C., Asprone, D., Caprino, G., Lopresto, V. & Prota, A. 2011. Numerical simulation of impact tests on GFRP
composite laminates. International Journal of Impact Engineering, 38, 677685.
Ngre, P., Steglich, D. & Brocks, W. 2004. Crack extension in aluminium welds: a numerical approach using the
GursonTvergaardNeedleman model. Engineering Fracture Mechanics, 71, 23652383.
Paik, J. K. 2007. Practical techniques for finite element modeling to simulate structural crashworthiness in ship
collisions and grounding (Part I: Theory). Ships and Offshore Structures, 2, 6980.
Peschmann, J. 2001. Energy absorption computations of ship steel structures under collision and grounding
(translated from German). Doctoral Dissertation. Technical University of Hamburg.
Polojrvi, A. & Tuhkuri, J. 2009. 3D discrete numerical modelling of ridge keel punch through tests. Cold Regions
Riska, K. 1987. Modelling ice load during penetration into ice. Joint report of the technical research center of
Finland and the National Research Council of Canada.
Rustad, A. M., Larsen, C. M. & Srensen, A. J. 2008. FEM modelling and automatic control for collision prevention
of top tensioned risers. Marine Structures, 21, 80112.
Svik, S. 2011. Theoretical and experimental studies of stresses in flexible pipes. Computers & Structures, 89,
22732291.
Scharrer, M., Zhang, L. & Egge, E. D. 2002. Collision calculations in naval design systems. Report Nr. ESS.
Schraad, M. W. & Harlow, F. H. 2006. A stochastic constitutive model for disordered cellular materials: Finitestrain uni-axial compression. International Journal of Solids and Structures, 43, 35423568.
Tabri, K., Maattanen, J. & Ranta, J. 2007. Model-scale experiments of symmetric ship collisions. Journal Marine
Tai, Y. S. & Tang, C. C. 2006. Numerical simulation: The dynamic behavior of reinforced concrete plates under
normal impact. Theoretical and Applied Fracture Mechanics, 45, 117127.
Trnqvist, R. 2003. Design of crashworthy ship structures. Technical University of Denmark Kgns Lyngby,
Denmark.
Trevino, T. 2000. Applications of arbitrary lagrangian eulerian (ALE) analysis approach to underwater and air
explosion problems. DTIC Document.
Tsai, S. W. & Wu, E. M. 1971. A general theory of strength for anisotropic materials. Journal of composite
materials, 5, 5880.
Vermeer, P. & Jassim, I. 2011. On the dynamic elastoplastic material point method. Ports for container ships of
future generations workshop. Hamburg.
Webster, K. G. 2007. Investigation of close proximity underwater explosion effects on a ship-like structure using the
multi-material arbitrary Lagrangian Eulerian finite element method.
Yoo, C. H., Kim, K. S., Choung, J., Shim, C. S., Kang, J. K., Kim, D. H., Suh, Y. S., Shim, Y. L., Urm, H. S., Kim,
M. S. & An, G. B. 2011. An experimental study on mechanical, fatigue and crack propagation properties of IMO
type B CCS materials at room and cryogenic temperatures. In: Soares, C. G. & Fricke, W. (eds.) Advances in
Marine Structures. London, UK: Taylor & Francis Group.

710 SEPTEMBER 2015
CASCAIS, PORTUGAL
VOLUME 2
COMMITTEE V.2
NATURAL GAS STORAGE AND TRANSPORTATION

COMMITTEE MANDATE
Concern for the safety and design of containment systems for the storage and transportation of natural gas in
connection with floating platforms and terminals, and onboard ships. This is to include assessing the
performance of various containment systems for gas under compression (CNG), liquefaction under cooling
(LNG), and combinations of the two methods. Particular attention shall be given to the integrity and safety
aspects of containment systems under pressure and thermal loads, and the interaction between fluid and
structure under static and dynamic conditions. Consideration shall be given to the installation and safety
matters of gas and dual fuel propulsion systems, fitted onboard conventional vessels. Needs for revision of
current codes and regulations shall be addressed.
COMMITTEE MEMBERS
Chairman:
R. H. M. Huijsmans, Netherlands
M. Arai, Japan
M. D. Feirreira, Brazil
M. K. Ha, Korea
M. Lindgren, Norway
T. Rahman, Australia
S. Schreier, Germany
P. Sharma, USA
O.Valle, Mexico
Longbin Tao, UK
A. Zakky, Indonesia
Z. Zhan. China
KEYWORDS
Cargo containment systems, Liquefied Natural Gas Carrier, Floating Liquid Natural Gas, Floating Storage
and Regasification Unit, Membrane Tank, Spherical Tank, Prismatic Tank, Compressed Natural Gas,
Sloshing, Offshore Terminal, Arctic, Structural Integrity, Collision, Flooding, Fatigue, Vibration, Fire
Safety, Corrosion, Boil Off Gas, Cryogenic Spillage, Fuel Liquefied Natural Gas.
592
ISSC committee IV.2: NATURAL GAS STORAGE AND TRANSPORTATION
CONTENTS
1.
INTRODUCTION
593
2.
BACKGROUND
593
3.
SAFETY AND DESIGN

595
3.1 Cargo containment
595
3.1.1 Non-self Supporting TanksMembrane Tanks
595
3.1.2 Independent tanks
595
3.1.3 New Development of CCS
596
3.2 Structural Integrity and rules
596
3.3 Sloshing
598
3.3.1 Global Flow and Sloshing-Ship Motion Coupling, Online Sloshing Prediction 598
3.3.2 Long-Term Assessment
599
3.3.3 Experimental Methods, Benchmark
600
3.3.4 Sloshing Model Test Benchmark
600
3.3.5
Sloshing Physics, Scaling ELPs, dominating physics and relevant
scaling laws
600
3.3.6
Numerical Methods
601
3.4
Leakage
602
3.5
Fatigue
602
3.6
603
3.7
Sloshing Control
605
3.8
605
4.
LNG AS FUEL
4.1
Why LNG as Fuel
4.2
LNG Supply Chain
608
608
608
5.
SAFETY AND DESIGN SPECIAL APPLICATIONS

5.1
5.2
Side by side or Tandem mooring?
5.3
Arctic
610
610
611
612
6.
CONCLUSIONS
612
REFERENCES
612
ISSC committee V.2: NATURAL GAS STORAGE AND TRANSPORTATION
1.
593
INTRODUCTION
The Committee V.2 is a follow up of a new specialist committee for natural gas storage and transport
whose mission was to outline the safety and design aspects of containment systems used for natural
gas storage and transportation on the ocean. With the increase in the worldwide demand for natural
gas as a relatively clean energy source compared to other fossil fuels, new concepts and technologies
related to the storage and transportation of natural gas have emerged. Based on the committees
mandate and the specialties of its members, the Committee has reviewed the performance of existing
and new containment systems and has discussed their safety.
The initial section of the report describes the safety records, transportation and market trends as the
background of the Committees work. Next, the safety aspects of LNG are discussed and an overview
of Cargo Containment Systems (CCSs) and operational features related to the safety and design of
natural gas storage and transportation systems are described.
The following chapters deal with the measures that must be taken to assure the safety of the Cargo
Containment System per mode of failure. Also structural integrity management issues are outlined.
Possible failure modes caused by several incidents such as sloshing, collision, fatigue, and the like,
together with the measures to mitigate them, are discussed. The necessity of establishing new rules
and regulations is emphasized with regard to the new concepts of natural gas storage and transportation, for example in applications such as Floating Liquefied Natural Gas (FLNG), Arctic and for applications of LNGas fuel.
2.
BACKGROUND
According to The IEA WEO2014 data, world primary energy demand rises about one-third between 2012
and 2035, or 1.2% per year on average. This compares with 2% per year over the previous 27-year period.
Oil and coal consumption grow more slowly than the overall rise in energy demand (12% and 16%), while
natural gas, nuclear and modern renewables rise much more quickly (44%, 74% and 134%) (Fig. 1 &
Table 1).
Table 1. World energy consumption by primary energy sources, IEA WEO2014 data (Source: IEA WEO2014
(New Policies).
1990
2000
2012*
Oil
3231
3663
4158
Gas
1668
2072
2869
Coal
2230
2357
3796
526
676
642
Nuclear
Hydro
184
225
313
Bioenergy***
893
1016
1318
Other renewables 36
60
142
Total (Mtoe)
8768
10069
13238
*2012 data are preliminary estimates
** Compound average annual growth rate
2020
2025
2030
2035
2012 - 2035**
4469
4545
4600
4666
0,50%
3234
3537
3824
4127
1,60%
4137
4238
4309
4398
0,60%
869
969
1051
1118
2,40%
391
430
466
501
2,10%
1488
1598
1718
1848
1,50%
311
432
566
717
7,30%
14899 15749 16534 17375 1,20%
*** Includes traditional and modern biomass uses
Mtoe = million tonnes of oil equivalent
Figure 1. World primary energy demand in Mtoe (million tonnes of oil equivalent)
(EA WEO2014 (New Policies)).
594
The Fukushima effect is an opportunity for the development of natural gas energy. After the catastrophe, Japan and many other countries decided to limit the use of nuclear energy and to turn back
to LNG. Because of low CO2 and NOx emissions and the reduced impact of natural gas on environment compared to oil or nuclear energy, development of offshore gas fields becomes one of the most
suitable solutions to sustain our needs in energy and to reduce the use of dangerous and polluting materials. Finally, the evolution of the technologies with the implementation of new installations such as
LNG Regasification Vessel (LNG RV), LNG Floating Production Storage and Offloading (LNG
FPSO) or Floating Storage Regasification Unit (FSRU) allows gas companies to explore and develop
not only deeper fields but also remaining fields, exploited during the last decades for oil but abandoned because of the lack of technology to exploit gas, or marginal fields which were considered not
profitable a few years ago. Nowadays, it exists a wide range of offshore solutions that allow gas companies to fit as close as possible to the world needs (Benyessaad et al, 2014).
On the other hand there are two issues regarding fuel of ships high fuel oil price and carbon emission control. Particularly, The IMO (International Maritime Organization) requires 90% of SOx and
80% of NOx reduction in exhaust gas by 2015 and 2016, respectively. Essentially, both of these issues
have equal economic significance since expensive low sulfur fuel oil or additional exhaust gas treatment is required to meet the emission regulation. Therefore, it is natural that many ship owners take a
profound interest in LNG fueled large commercial ship these days (Kim et al, 2013).
Figure 2 shows the trend of long term energy price. Compared to rapid increase of crude oil, gas
price is almost stable with the discovery of shale gas.
Figure 2. Long term energy price trend (Kim et al, 2013).
Figure 3. Comparison of emissions for ME diesel engine (left) and gas fueled engines (right).
Figure 3 shows the characteristics of LNG fuel in the view point of gas emission. In this figure, 6570
ME-C and 6570ME-GI represent the use of oil and LNG as fuel, respectively. It shows that LNG fueled ship can reduce nearly all of SOx emission and some amount of NOx emission, and besides it can
also reduce CO2 by 23%. ME-GI with EGR (Exhaust-Gas Recirculation) system reduces NOx up to
80% (Kim et al, 2013).
3.
SAFETY AND DESIGN
3.1
Cargo containment
3.1.1
Non-self-Supporting TanksMembrane Tanks
595
Membrane tanks are non-self-supporting structures with design vapor pressure of 0.25~0.7 bar. These
tanks consist of a thin membrane layer which is supported by insulation such as plywood, triplex, reinforced polyurethane foam and mastic to keep the temperature. This type has benefits in efficiency of
storage capacity (no void space) and containment weight reduced with strong and simple structure of
cargo tanks. However LNG membrane tanks may have structural problems on partial filling caused by
large impact loads due to sloshing, and first of all, safety for collision, being stranded and sloshing
should be approved for arctic LNGC, LNG Floater and LNG fuelled vessels (Bang et al, 2012).
The sloshing problem on membrane type tanks has become a big issue in LNG tanks industries
(LNGC, FSRU and FLN). The massive researches have been done to solve the problem. The challenge
in sloshing assessment is how to accurately predict the sloshing loads due to the motion of structure,
calculate the impact on structure (structural strength) and validate with full scale measurements. Considerable research efforts have been made and more details will be presented in the following subsection.
3.1.2
Independent tanks
According to IGC code, the independent tanks are classified as Type A, Type B and Type C. The advantage of the independent tank of LNG especially for IMO type is structural safety to withstand the
sloshing, internal flow and external impact. Bang et al. (2012) have evaluated structural stability of the
independent tank focusing on the cargo tank support structure between cargo tank and hull structure. It
is noted that there is no significant research on LNG cargo independent tank only limited studies have
been reported on IMO type B.
3.1.2.1 Compress Natural Gas Tanks

CNG is one of the methods for storage of natural gas in high-pressure tanks at pressure levels of 2025
MPa (200 to 250 bars, or 2900 to 3600 psi). Migration from fuel oil to fuel gas recently has encouraged
the invention of new concept for storage and transport of natural gas including CNG concept. In beneficial, CNG is easily available from existing infrastructure without liquefaction process and easy to store in
room temperature (no cryogenic hazard) and short bunkering time. However the safety aspect regarding
pressured storage is the main drive to use CNG method for storage of natural gas. Nevertheless CNG is
more feasible for users with short routes. The table below shows a comparison of CNG vs. LNG (Wang
et al, 1992).
While LNG dominates the market for sea transport of natural gas, a number of recent studies have
shown that compressed natural gas (CNG) is economically more attractive than LNG for sea transport
of relatively smaller volumes of gas over shorter distances (Wang et al. 2009).
Vernengo et al. (2013) established an automatic and integrated preliminary design procedure to generate
and evaluate feasible technique for a trade of CNG and a first trial application has been carried out for a
specific case. The procedure covers the typical aspects of ship design: selection of main hull dimensions,
identification of weight (steel, outfitting, pressure vessel and cargo), propulsion and fuel consumption at
nominal speed, trim and stability checks. Related to the gas containment systems, an issue was identified in
combination of the properties of the compressed gas with very small specific gravity with characteristics of
the gas containment system which are large pressure vessel in steel, with significant weight and dimensions. Further developed of the procedure is necessary to consider different solutions for the gas containment systems including the use of innovative materials. This will allow an evaluation of such solutions in a
cost/benefit perspective.
Table 2. Process & cargo differences between CNG and LNG (Wang et al., 2008).
Aspect
Physical state
Pressure
Temperature in tank
Loading
Terminals
Ships
Receiving
Loading/offloading
Compression ratio
Containment D/t (diameter/wall thickness?)
Material
CNG
Gas
100 50 bar
30C to 40C
Dehydrate, compress
Jetty or buoy
Simple, like bulk-carrier
Heat & decompress utilize
energy released
Gas under pressure
~200 250:1
~25 60
Fine grain normalized C-Mn
steel, FRP
LNG
Liquid
1 bar
163C
Treat, liquefy, store
Jetty, or regas offshore
Sophisticated, efficient
Store, regasify
As liquid
~600:1
~1000
Aluminum, Stainless,
Ni steel
596
3.1.3

New Development of CCS
The new IGC Code has been totally revised by IMO during MSC 97 with effective date on 1 July
2016. There are many changes including the requirements related to the cargo containment system
(CCS). It has specific requirements based on the types of the tanks related to the design, structural
analysis, ultimate design condition, fatigue design condition, accidental condition and testing. For
membrane tanks there are new requirements related to the potential incidents that could lead to loss of
fluid tightness over the life of the membranes to be evaluated, these include:
Ultimate design events:

1.
2.
3.
4.
5.
6.
7.
tensile failure of membranes;

compressive collapse of thermal insulation;
thermal ageing;
loss of attachment between thermal insulation and hull structure;
loss of attachment of membranes to thermal insulation system;
structural integrity of internal structures and their supporting structures; and
failure of the supporting hull structure.
Fatigue design events:

1.
2.
3.
4.
fatigue of membranes including joints and attachments to hull structure;

fatigue cracking of thermal insulation;
fatigue of internal structures and their supporting structures; and
fatigue cracking of inner hull leading to ballast water ingress.
Accident design events:

1.
2.
3.
4.
accidental mechanical damage (such as dropped objects inside the tank while in service);
accidental over pressurization of thermal insulation spaces;
accidental vacuum in the tank; and
water ingress through the inner hull structure.
Table 3 shows the new requirements for independent tank types of cargo containment system.
3.2
Structural Integrity and rules
Rules and Standards, Partial Fillings, Structural Assessment Procedures

All major classification societies have reacted to the need for guidance on the treatment of partial filling levels of LNG Carriers and FLNGs. Bureau Veritas places a strong emphasis on numerical sloshing analyses whereas several other classes recommend experimental investigation. Regardless of the
hydrodynamic part of the sloshing investigations, the main emphasis of the class recommendations are
focused on the structural assessment with the aim of equal or even higher safety levels against sloshing
damages as for the proven concepts of the existing fleet of LNG Carriers. The individual approaches
differ slightly among the major classification societies, but in general a comparative approach as well
as a direct assessment of sloshing loads and responses is covered by the updated Rules.
Structural integrity Management is a means of ensuring the ability of an asset or structure and its
component to perform its planned function effectively and efficiently for its intended application
throughout its service life. Survey and inspection are important inputs in the structural integrity management of the structure or system. In case of floating structure unit (FPSO, FLNG, FSRU, etc) maintenance methods, the prescriptive rules with frequent intervals (5 years) in dry docks are to be used if
the structure is based on ship shaped type. However, floating offshore structure is subject to operate
without dry dock for inspection, maintenance and repair, so that integrity management system with
Risk Based Inspection (RBI) method may be considered to establish rational maintenance program
(Survey and Inspection) for floating hull.
In case of FPSO structure, many methodologies and concepts of structural integrity management
based on Risk Based Inspection have been introduced. Bisotto et al. (2004) have presented the methodology to establish inspection and survey plans combining RBI analysis and the industry expertise.
The general simple work-flow of the inspection plan is illustrated in Fig.5. The effort to set inspection & survey plan is highly linked to the choice of the risk assessment method: qualitative and/or
quantitative. The research in the area of integrity structural management or sometime called asset integrity management is ongoing especially in offshore industry to maintain safe production and to ensure
the lifting of oil & gas production. Wisch et al. (2009) have derived various methods regarding to the
structural integrity management of FPSO structure which can be used as references for FLNG and
FSRU. Furthermore the recent study on structural integrity management of FPSO by Ku et al. (2012)
proposed the structural reliability methodology and demonstrated its application to the RBI planning
597
of an FPSO. Structural reliability based methods can assist in providing a framework for assessing
site-specific loading and degradation mechanisms (such as fatigue and corrosion) through a systematic
consideration of the uncertainty in each degradation mechanism.
Table 3. Benchmark of The new IGC Code requirements for Independent tanks.
Item
Design
Basis
Independent tanks
Type A
Based on classical
structural analysis
procedure using recognized standard
Design vapour pressure
(P0) < 0.07 Mpa
Full Secondary barrier
for 55C T< 10C
Structural
Analysis
Internal pressure and

the interaction loads
shall be applied.
Direct strength calculation shall be conducted
for parts which are not
cover by the codes.
The tanks with supports shall be designed
for the accidental loads
Ultimate
design
condition
Nominal membrane
stress (primary and
secondary members)
Buckling check
N/A
Fatigue
design
condition
Accidental
design
condition
The tanks and tanks

suports shall be
designed for accidental
loads and design
condition.
Type B
The tanks designed using model tests,
refined analytical tools and analysis
methods to determine stress level, fatigue life and crack propagation characteristics.
Design vapour pressure P0 < 0.07 MPa
Partial Secondary barrier with small
leak protection system shall be provided
for T< 10C
The structural analysis (Plastic deformation, buckling, fatigue failure, crack
propagation) shall be determined using
all of dynamic and static loads.
FEA or similar method and fracture
mechanics analysis, three dimensional
analysis shall be carried out.
A complete analysis of the particular
ship accelerations and motions in irregular waves, and of the response of the
ship and its cargo tanks to these forces
and motions shall be performed, unless
the data is available from similar ships.
Plastic deformation check
Buckling check
Type C
The tank designed
based on pressure vessel criteria modified to
include fracture mechanics and crack propagation criteria.
Fatigue and Crack propagation analysis
For Tanks with temperature below 55C it is

to be ensured an initial
surface flaw will not
propagate more than
half the thickness of the
shell during the lifetime
of the tanks
The tanks and tanks
suports shall be
designed for accidental
loads and design
condition.
The tanks and tanks suports shall be

designed for accidental loads and design
condition.
N/A
Plastic deformation
check
Buckling check
(New methodology FLNG)

FLNG and FSRU are the typical types of floating offshore structures with increased order in recent
years. With same basis as FPSO, FLNG/FSRU has more complicated topside process for handling the
gas processing on the topside (Production, liquefaction and regasification). The challenge is how to
build gas plant on limited space and ensuring efficient and safe operations and that is the main reason
for the choice of such system. FLNG as LNG production platforms and FSRU as LNG regasification
platforms, imply a blend of technology from land-based LNG industry, offshore oil and gas industry
and marine transportation technology (Arronson 2012).
In contrast to trading ships, FLNG & FSRU is not easy to drydock, and they have additional specific issues regarding to handling the gas (storage and gas process on the topside). These reasons will be
a challenge in Structural Integrity Management in FLNG and FSRU structure in way to conduct Risk
assessment with limited rules and codes related. Arronson (2012) has identified the risks relevant to
FLNG. The risks are compared to conventional LNG carriers and whether or not regulatory alignment
possibilities exist. To identify the risks, a risk analysis was performed based on the principles of formal safety assessment methodology. In table 4, the list of hazards of FLNG with the risk level have
been presented by Arronson (2012).
The list described in Table 4 was the outcome of a brainstorming event and followed the IMO FSA
guidelines. However the study was limited to the first step of the FSA, risk identification and the
598
results have been used to estimate a rough value. Nevertheless the study provides a preliminary analysis for risk assessment of FLNG & FSRU with less experience in design and operation in the world.
The further study is needed.
Asset Subdivision
TAG System
Review of
Documentation
Components
Selected
Qualitative Assessment
(HAZID)
Quantitative
Assessment (FDF)
Detailed RBI
(FDF <10)
Probability of
Failure
Consequences of
Failure
Risk
Ranking
Inspection Program
Figure 4. Work-flow of simple maintenance program based on Risk Assessment, Bisotto et al. (2004).
Table 4. List of hazard to FLNG & FSRU by Arronson (2012).
Area
Feedstock*
Gas processing and

Liquefaction
/Regasification
Cargo handling
Offloading And
vessel overall
Hazard
Blow out
Hydrocarbon release from
the turret
Hydrocarbon release in
process area
Cryogenic spill in liquefaction area (regasification plant
in case of FSRU)
Spill of hazardous substance
Fire/explosion in process
area
Fire/explosion in containment area
Inert gas release in containment area
Sloshing in cargo tanks
Ship collision
Cryogenic leak during offloading (loading in case
of FSRU)
Frequency
3
2
Consequence
1
3
Risk level
4
5
2
2
1
3
3
5
4
1
1
1
2
3
5
3
4
*) The area and list of hazards are possible only for FLNG
3.3
Sloshing
3.3.1
Global Flow and Sloshing-Ship Motion Coupling, Online Sloshing Prediction
One of the critical areas in the design of FLNG/FSRU systems as well as floating LNG terminals is the
transfer of LNG between the floating structures. During this transfer, intermediate filling levels at least
in one tank at a time are inevitable. Thus sloshing can occur in the partially filled tanks and affect the
599
motions of the vessels in the seaway. Furthermore the close proximity of the LNG Carrier and the
FLNG or floating terminal has a significant influence on the motion behaviour as compared to the
open water situation. Therefore dedicated motion analyses have been carried out including the coupling effect of the ship motion and the motion of the liquid inside the tanks as well as accounting for the
presence of two floating bodies in close proximity. Park et al. (2013), for example, investigated the
motion behaviour of an FLNG with tank capacity of 220,000 m and an LNG Carrier with a capacity
of 145,000 m in side-by-side configuration with a separation distance of 5 m. The RAOs of the vessel
motions were calculated by three-dimensional potential theory. The liquid motion in the tanks was
also determined by linear potential theory, reasoning that the global fluid motion and the resulting
forces relevant for the vessel motions were covered by this approach. In this study one-row and tworow tank arrangements for the FLNG were investigated.
The coupled motions of an LNG Carrier and the sloshing liquid are also investigated by Zhao et al.
(2014). The authors use linear potential theory for the ship motions and a non-linear potential theory
approach for the fluid motion inside the tank. Furthermore, a number of references to other studies of
coupled motions using different approaches are given there.
Huang et al. (2013) transform the hydrodynamic coefficients and wave loads of the outside flow
from frequency domain to the time domain and calculate the liquid sloshing motion inside the tank
based on incompressible Reynolds-averaged Navier-Stokes equations and the Volume of Fluid method
for the interface capturing using the OpenFOAM framework.
There are several approaches using computationally highly efficient methods to predict the occurrence of severe sloshing conditions. Godderidge et al. (2012) present a Rapid Sloshing Model based
on a pendulum equation. Cao et al. (2013) used in-house software LTS (Linear Time-domain Sloshing) based on Desingularized Rankine Singularities. Kayal and Berthon (2013) employed an analytical
approach augmented by a set of designated model tests for tuning the analytical model. These methods
cannot simulate the highly complex violent sloshing flows in the tanks, but they effectively determine
the onset of severe sloshing conditions in the tanks based on the current ship motions in faster than
real time. Thus these methods are likely to form the basis of sloshing prediction systems and decision
support systems on board of LNG Carriers and other vessels operating with large partially filled tanks.
Zheng et al. (2013) follow a different approach and determine the so-called Sloshing Severity Index
SSI as a pre-screening tool to limit the number of sloshing model tests to those sea conditions likely to
result in the most severe sloshing loads. To this end, a coupled linear potential theory method for the
ship motions and sloshing is employed. The SSI is calculated from a global parameter, the total energy
of the sloshing liquid in a tank, and a local parameter, the maximum vertical velocity of the sloshing
liquid at the free surface.
Park et al. (2012) studied the coupled sloshing and motion analysis of FLNG in Side-by-Side Arrangement. In this study, both coupled effects between two-body motion and sloshing in the LNG
tanks are considered. The methodology for coupled ship motion analysis is based on 3-dimensional
potential theory in the frequency domain. For the sloshing analysis, the violent liquid motion inside
tank is treated with SHI-SLOSH CFD code to calculate sloshing impact pressure. The time series of
ship motion is generated with the ship motion results considering both two-body and sloshing coupled
effect.
In a study on coupling effect between seakeeping and sloshing for membrane-type LNG carrier,
Wang et al. (2012) evaluate the performance of membrane-type LNG carriers under half-loading conditions. The simulations for the coupled ship motion and sloshing have been carried out by using a
newly developed time domain scheme. The external fluid field was solved by potential theory, while
the internal fluid field was solved by a 3D finite difference method. The ship motions in different
wave direction and frequencies were calculated. The coupling effect was found to have significant
influence on transverse motions, while the longitudinal motions were not influenced. The slosh flow
inside tanks was also investigated.
3.3.2
Long-Term Assessment
The long-term assessment of sloshing loads is usually carried out based on a limited number of sloshing model tests, where a three-dimensional model of an LNG tank under consideration is subjected to
time series of irregular ship motions in all six degrees of freedom. The difficulties of this approach
comprise the proper selection of the test conditions by pre-screening methods, as described above, and
the meaningful stochastic interpretation of the test results. The aim of this stochastic interpretation is
the reliable prediction of the maximum sloshing loads in an LNG tank and their probability of occurrence. Therefore the probability of exceedance is calculated for the peak sloshing pressures from the
model tests. The estimation of peak pressures at a certain probability level is then achieved by fitting
stochastic distribution functions such as the Weibull distribution or a Generalized Pareto distribution
to the test results. The influences of the data preprocessing and the selection of the fitting distribution
for sloshing pressures have recently been discussed by Fillon et al. (2013) as well as Dematteo and
600
Gervaise (2014). Diebold et al. (2013) investigated the statistical behavior of global sloshing parameters such as the tank accelerations due to the ship motions and the sloshing global forces as well as
additional local parameters such as the fluid velocity normal to the tank boundaries at a certain distance from these boundaries. In this context Dematteo and Ratouis (2013) present a method to stochastically simulate the spatial pressure distribution over the tank boundaries in the vicinity of the limited
number of physical pressure sensors. This approach may help to identify high pressure sloshing impacts that have been missed by the physical sensors.
Numerical simulations in combination with model tests have been carried out by Park et al. (2014)
to optimize the cross-sectional shape of an LNG tank for unrestricted filling levels. Here the Weibull
distribution is employed for the stochastic interpretation of sloshing pressures. Oh et al. (2014) used
model tests and the Generalized Pareto distribution to identify the relevant sloshing pressures for an
FLNG design and subsequently carried out the structural assessment of the FLNG cargo containment
system based on these data.
The sloshing model tests for the long-term assessment are usually carried out using water and air at
ambient conditions and Froude-scaling is employed for the motions and the resulting pressures. This
approach might result in overly conservative pressure estimates, see Scaling section.
3.3.3
Experimental Methods, Benchmark
As mentioned in the previous section, the long-term assessment of sloshing pressures mostly relies on
experimental model testing, since this is the only commonly accepted source of pressure data, as
pointed out by Pistani and Thiagarajan (2012). Given the importance of the sloshing model test data
for the design of LNG tanks and their cargo containment systems great efforts are undertaken to obtain
reliable data. Pistani and Thiagarajan (2012) assess the complete measurement chain from the test rig
to the data acquisition system including first steps of data processing. The authors address the influence of each component of the chain on the final data and identify the susceptibility of the employed
pressure sensors as one important challenge to overcome. Ahn et al. (2013) investigated the behaviour
of four different models of pressure sensors commonly in use for sloshing model tests in great detail.
Baudin et al. (2012) present the peak memory cards, originally dating back to the 1980s, now revisited
for sloshing model tests as an alternative means to capture the maximum peak pressures of sloshing
impacts at reduced sampling rates.
3.3.4
Sloshing Model Test Benchmark
The Sloshing Model Test Benchmark (SMTB) was initiated in 2011 and is now organized by the
ISOPE Technical Committee on Sloshing Dynamics and Design. The long term aim of the benchmark
is to advance the quality and comparability of sloshing model tests and thus increase the fidelity of the
results in the sloshing assessment of new LNG tank designs or applications. This SMTB recognizes
that model tests are still the state of the art in the sloshing assessment of LNG tanks on floating vessels. The tank for the benchmark tests is a simple rectangular tank subject to exciting motions with
reduced number of degrees of freedom. In the period of 2011 to 2012 nine institutions participated in
the benchmark. The test conditions and results are summarized by Loysel et al. (2012). Based on the
spreading in the results of 2012 the focus for the period of 2012 to 2013 was laid on the repeatability
of single-impact waves in the same rectangular tank. The results are summarized again by Loysel et al.
(2013) and were discussed during the ISOPE 2013 conference. The Sloshing Model Test Benchmark
will now be continued on a biennial schedule with the next round to be concluded in 2015.
3.3.5
Sloshing Physics, Scaling ELPs, dominating physics and relevant scaling laws
Important results for the phenomenological and physical interpretation of breaking wave impacts such
as sloshing impacts on a vertical tank wall have been reported by Lafeber et al. (2012). The authors
identified three elementary loading processes (ELP) that can be related to different physical processes
in sloshing impacts. The first ELP is associated with the direct impact of the liquid on the tank wall.
This highly localized phenomenon is named the direct impact ELP (ELP1). The next pressure pattern
derives from the formation of a jet of liquid travelling along the tank wall. This ELP2 is called the
building jet ELP. The third discrete process identified from the test results is the oscillating pressures
of a gas pocket entrapped by the overturning wave crest and the tank wall. This loading process is
titled the pulsating gas pocket ELP (ELP3).
Starting from the ELPs analytical and numerical investigations have been carried out in order to
find scaling laws for the different loading processes. The results for pulsating gas pockets based on an
extension of Bagnolds piston model are reported by Ancellin et al. (2012) and Brosset et al. (2013).
Several other authors have also been working on scaling effects.
601
Furthermore, the first results of full-scale onboard sloshing measurements of an LNG Carrier were
presented by Pasquier and Berthon (2012). Several papers have been published on the comparison of
results obtained at different scales.
Karimi et al. (2013) studied the global and local effects of gas-liquid density ratio on shape and kinematics of sloshing waves and their scaling considerations. The authors concluded that the effects of
gas-liquid density ratio (DR) on sloshing wave shapes when investigated globally (far from impact
zones) by 2D sloshing model tests with irregular excitations at two different scales and for low-fill
levels appear to be small for the tested range of density ratios which implies similar wave shapes far
from impact zones. When repeating the same irregular motions, global flow keeps the same phase
regardless of tested DRs which enables to recognize an accurate impact-by-impact relation between
model tests at similar and different scales and adds a deterministic side to post-processing model test
results, The local effects of DR (right before impact) on breaking wave shapes were also investigated
by 2D sloshing model tests with single breaking waves at two different scales and for low-fill levels.
The local effects of DR clearly modify the impact geometry before gas compressibility interference
with significant consequences on induced pressures.
Brosset et al. (2013) presented a Generalized Bagnold Model in which experimental and numerical
studies have shown that the behavior of gas pockets entrapped by a breaking wave when impacting a
wall is well described by the piston model first modelized with a single Ordinary Differential Equation
(ODE) by Bagnold under the assumption of a perfect gas and isentropic conditions. As for a sloshing
impact inside the tank of LNG vessels, an inertial acceleration is always involved during the impacts
and as several authors observed some evidence of the influence of liquid compressibility during wave
impact tests. A 1D model of the liquid piston problem including a constant inertial acceleration is proposed based on isentropic compressible Euler equations, as an extension of the previous 0D model. A
parametric study is performed with the first one looking at the influence of each dimensionless number
on the maximum pressure at wall. The scaling of the 1D liquid piston model is studied giving insight
on the scaling process when several similarity laws are at work.
In Rafiee et al. (2012) numerical simulations of 2D liquid impact benchmark problem using two
phase compressible and incompressible methods were compared. The authors simulated the benchmark impact problem (as proposed in ISOPE 2009) using a twophase compressible Smoothed
Particle Hydrodynamics (SPH) method and an incompressible LevelSet method. A 2D impact
benchmark problem is simulated for two different liquidgas combinations, namely airwater and
NGLNG. The results highlight the effect of aforementioned parameters on the impact pressure and
the capability of the proposed SPH and LevelSet schemes in accurately predicting complex impact
pressure.
3.3.6
Numerical Methods
A whole myriad of papers have emerged that treat sloshing using SPM methods or related methods
such as e.g. MPS, moving particle semi-implicit methods. In Zhang et al. (2013), liquid sloshing is
simulated based on Moving Particle Semi-Implicit (MPS) method, which is a meshless method. However, the traditional MPS method suffers from strong unphysical pressure oscillation. To overcome
this, some improvements had been made, such as: nonsingular kernel function, mixed source term for
pressure Poisson equation (PPE) and an accurate surface detection method. Smooth pressure field is
obtained based on the present MPS method. The predicted pressure at resonance on the wall of the
LNG tank by MPS method shows a good agreement with experimental data and other numerical results. The impact behavior induced by liquid sloshing is accurately predicted. In addition, violent free
surfaces are observed.
To include hydro-elastic effects, methods based on finite particle methods have shown progress as
reported by Baeten et al (2013). Here one observes a LNG tank shape in terms of sloshing impact
pressure and focuses on an innovative modelling approach of elastic tank walls. This approach is
based on finite particles, which provide similar properties as finite elements in classical structure mechanics. It features a particle-based liquid model and provides time accurate hydrodynamic pressure
results in three dimensions and 6 Degrees-of-Freedom. Based on the results of specific tank geometries using a rigid wall model, unsteady sloshing loads are computed for elastic tank walls. The
stresses and strains are discussed to optimize the structure in terms of impact pressure loading. Thereby, the LNG tank geometry is flexible in terms of wall friction, elasticity laws and damping
coefficients.
Hwang et al. (2014) reports on a simplified impinging jet model for practical sloshing assessment of
LNG cargo containment. A simplified numerical modelling is conducted to create an assessment
procedure for sloshing impact in partially-filled LNG cargo tanks. A computational procedure based on
an impinging jet of LNG acting on an insulation panel is employed in order to evaluate the sloshing
impact on a real cargo tank. The velocity of the impinging jet is determined by a similarity rule that
602
scales up the impact velocity from a small scale to that of actual scale. Several procedural components
are introduced for a structural response calculation based on transient pressure.
A variety of numerical methods are compared by Fossa et al. (2012), using FEM-CFD and FEMFSI approaches. Simulations were done in two-dimensions, i.e. an infinitely long tank. Investigation
was carried out first by Finite Element Method (FEM) Computational Fluid Dynamics (CFD) approach, considering the structure as rigid wall. Thereafter, fluid structure interaction (FSI) model using
finite element method for both fluid and structural degrees of freedom was implemented. Typical test
cases were considered, for which experimental results were available in open literature and already
compared with other CFD results in previous works.
3.3.6.1 Smooth Particle Hydrodynamics

Cao et al. (2014) study sloshing in a rectangular tank. Their study focuses on the SPH core issues,
such as the accuracy and the stability of the kernel function and boundary treatments. Firstly, the accuracy and computational stability of four common SPH kernel functions are simply investigated by two
simple cases, and a more appropriate kernel function is selected. Secondly, the dummy particles and a
novel boundary treatment considering the boundary motion are applied. Furthermore, the laws of impact pressure of the two-dimensional tank under forced rolling with different excitation frequencies
and excitation angles are studied. Then, the influences of a baffle for the liquid sloshing in a twodimensional tank under forced surging are analyzed, and the action mechanisms of the baffle are
summarized. Finally, the coupled motion of swaying and surging for a three-dimensional tank is studied, which aims to lay a foundation for further study on the influence of sloshing loads on real ship
motions.
Chowdhury et al. (2014) studied a -SPH model that has been proven to be superior to standard
SPH for solving fluid flow problems covering a broad range of Reynolds (Re) numbers including violent fluid motion involving free surface fragmentation. Analysis has been carried out to identify major
factors affecting the present -SPH model from predicting higher order harmonics. The role of XSPH
factor as used in SPH in smoothening of velocity has further been revisited. The model prediction for
different physical quantities of sloshing has been found to be in good agreement with results from
other experimentally, numerically and analytically based studies. Suitable scales for the tuning factors
are proposed for the development of robust SPH models which are expected to simulate sloshing
waves defined by relatively broader d/L (ratio of initial water depth to tank length) ranges.
Gotoh et al. (2014) present two schemes for enhancement of Incompressible SPH (Smoothed
Particle Hydrodynamics)-based methods in simulation of violent sloshing flows, and in particular,
sloshing induced impact pressures. The enhanced schemes include a Higher order Laplacian and an
Error-Compensating Source of Poisson pressure equation, abbreviated as HL and ECS, respectively.
These two schemes correspond to those derived within the framework of MPS (Moving Particle Semiimplicit) method and are proposed for the first time within the framework of SPH. The enhancing
effects of HL and ECS schemes are shown by simulating violent sloshing flows induced by sway excitations and rotation alone. The significance of dynamically adjusted coefficients in the ECS-related
schemes is highlighted by considering a previously applied scheme comprising of a constant coefficient and a corresponding newly proposed one which incorporates a dynamic coefficient. Concise
insights are presented on appropriate choice of kernel function.
3.4
Leakage
In the analysis of two main LNG CCS (cargo containment system) insulation boxes for leakage safety,
Woong et al. (2012) conducted an experiment to measure the thermal conductivity of insulation materials used in a LNG CCS under various temperature conditions ranging from room temperature to
cryogenic temperature. Using the experimentally determined thermal properties a steady-state thermal
analysis of a CCS insulation box was investigated. In addition, a thermal analysis was conducted for
the hull part of a carrier ship in a scenario in which LNG leaked through the secondary barrier of the
insulation due to an accident. The safety of the hull was considered based on the ductile to brittle transition temperature of the hull.
Choi et al. (2013) analyzed the flow and thermal characteristics of leaked LNG in glass wool for the
Mark III CCS. They evaluated pressure distribution of the glass wool with the condition that cryogenic
liquefied material leaked into the insulation box with constant pressure of leaked liquid. Furthermore,
flow analysis was carried out with different glass wool to investigate how the leaked cryogenic liquid
behaves in the glass wool. Simulation result was compared favorably to experimental results to realize
cryogenic liquid in the glass wool.
3.5
Fatigue
Song et al. (2013) present a study to investigate the load effects due to sloshing impact on the
fatigue assessment of an independent type B LNG tank. The selection of the cargo containment
603
system type for an FLNG (Floating LNG) is mainly related to the sloshing problem. FLNG are
considered to be operating at an intermediate loading condition prone to sloshing inside the containment system. One of the most popular LNG containment systems for the FLNG is an independent prismatic tank categorized as IMO (International Maritime Organization) type B tank, in
which the ship hull supports the LNG tanks. It is recognized that the independent type B tank is
free from possible liquid sloshing inside the tank from the standpoint of its strength due to its
stiffened panel structure. In the present study, the sloshing impact pressures on the internal structures of an independent type B tank are predicted based on scaled sloshing model tests using a 2D
rectangular tank with internal structures. Finally, a fatigue assessment based on the sloshing model results is carried out to quantify the effects of sloshing loads on the fatigue strength of an independent type B tank.
Kim et al. (2011) study cryogenic fatigue assessment for Mark III insulation system. The authors
investigated the typical failure mode and obtained the stress range versus number of cycles to failure
(S-N) data of MARK III type liquefied natural gas (LNG) insulation system under the fatigue loading
at actual cryogenic environment. A systematic experimental research is carried out for the assessment
of the fatigue strength of MARK III insulation system at cryogenic temperature. Three different types
of test specimens are tested for the evaluation of fatigue performance of MARK III insulation system.
Test specimens are determined considering the fatigue vulnerable locations such as mastic area, slit
area, and top bridge pad area inside the actual LNG cargo tanks. All test specimens are fabricated as
close as possible to the actual yard practice. A series of fatigue test results is represented as S-N
curves. Cyclic fatigue loadings were carefully considered similar to the actual sloshing loads. The
effect of sloshing impacts is considered by selecting the stress ratio (R=10). Different cryogenic
temperatures are employed according to the test locations in consideration of temperature gradient
within the insulation system. Consistent S-N curves of MARK III insulation system at both room and
cryogenic temperatures are obtained and compared. It is observed that the slopes of S-N curves from
both fatigue test results are almost identical, and the fatigue strengths are found to exhibit similar
trend.
3.6
Based on the previous ISSC(2012) report, collision and flooding are one of the dominant issues related
to hull-ice interaction and iceberg collision. To continue the previous research, there are several studies addressing collisions and flooding which are presented in this report. Lee et al. (2013) carried out
full scale collision simulation to estimate a more reliable and realistic collision impact response, using
Fluid-Structure Interaction (FSI) analysis technique of LS-DYNA code, and current load to iceberg
instead of current velocity.
In the new IGC Code 2014 which is effective by 1 July 2016, there are some fundamental
amendments related to the ships intended to operate for periods at a fixed location in re-gasification
and gas discharge mode or gas receiving, processing, liquefaction and storage mode (it may mean
temporary liquefied gas terminals). The risk of collision during berthing maneuvers shall be ensured
with appropriate steps by the administrator and port administrations. Additional requirements shall
be established based on the principle of the IGC Code. Also there is some modification of requirement for tank location based on tank types which shall meet the collision criteria (shall be protectively located and withstand the collision loads without deformation of the supports, or the tank
structure in way of the supports, likely to endanger the tank structure) and also flooded compartment
criteria causing buoyancy on tank (the anti-flotation arrangements shall sustain the upward force and
there shall be no endangering plastic deformation to the hull). Additional accidental loads included
collision loads and loads due to flooding on ship shall be determined as consideration in the design
as well as some accident design that are potential accidents which could lead to loss of fluid tightness over the life of the membrane tanks included accidental mechanical damage (such as dropped
objects inside the tank while in service). The comprehensive requirements have been provided in the
new IGC Code.
Hu et al. (2011) presented a verification of a simplified analytical method for the prediction of
structural performance during ship groundings over seabed obstacles from numerical simulations. This
simplified analytical method was developed by Hong and Amdahl and calculates grounding characteristics, such as resistance and distortion energy, for double-bottomed ships in shoal grounding accidents. Two finite-element models are presented. One was built for a hold, and the other was built for a
hold and a ship hull girder also considering sectional properties, ship mass, added mass and the hydrodynamic restoring force. The verification was completed by comparing horizontal and vertical resistances and the distortion energy between seven numerical simulation cases and a set of corresponding
cases computed by a simplified analytical method. The results showed that the resistances obtained by
the simplified analytical method are close to the mean values of the resistance curves obtained by numerical simulations. The comparisons demonstrated that the energy dissipation prediction capability of
604
the simplified analytical method is valuable. Thus, the simplified analytical method is feasible for
assessing ship groundings over seabed obstacles with large contact surfaces and trapezoidal crosssection. Finally, a new method for predicting the structural performance of the time-consuming
complete-ship model by applying a combination of normal numerical simulations and ship-motion
calculations is proposed and proven.
In Yu et al. (2013) an investigation is reported on structural performance predictions of doublebottom tankers during shoal grounding accidents. This investigation is carried out for the predictions
of structural performance of double-bottom tankers during ship grounding over the shoal type
seabed obstacles. Hong and Amdahl developed a simplified analytical model for the unstiffened
double bottom. This method is carefully studied, verified and then used as the first stage of the prediction. The second stage is concerned with stiffeners since stiffeners are indispensable components
for double-bottom tankers. A prevailing way to handle is to smear stiffeners onto their attached plating known as the smeared thickness method. However, the effective ratio in this method is dubious in
such shoal grounding accidents. Proper values of this parameter are determined in stage two, and then
together with the method in stage one, constitute a reliable and efficient tool for structural performance predictions of double-bottom structures in shoal grounding accidents. A double-bottom tanker
is chosen as object for the case study. Finite element models of the hold both stiffened and unstiffened are created for numerical simulations using the LS_DYNA software. Simulation cases cover a
wide range of slope angles of the indenter and indentations. Numerical results show that Hong and
Amdahls model in stage one is capable of predicting energy dissipation with high precision but poor
accuracy for grounding resistances, and a possible reason may be the neglect of vertical resistance.
The updated smeared method proposed in stage two is also proved to be capable of grasping major
characteristics of stiffeners. Results and conclusions drawn from this paper can be conveniently applied for assessments of the performance of ship double-bottom structures during shoal sliding
grounding scenarios, and will benefit the application of accidental limit state design concept in the
ship design stage.
Figure 5. Overview of steps in a comprehensive collision risk analysis, from Pedersen et al. (2010).
Hogstrm et al. (2012) reported an extensive study of a ships survivability after collision. The International Maritime Organization strives towards a more risk-based view on addressing the damage
stability of ships. The study addresses the survivability following a ship collision by the use of a sequential (de-coupled) computational methodology. The methodology is comprised of structural
605
analysis of a collision scenario followed by dynamic damage stability simulations of the struck ship in
order to establish the time to capsize of the struck ship.
According to the ISSC committee V.ICollision and grounding (2006), the following elements outline the principles of collision and grounding design standards.
The emphasis of the investigation is on the structural computations of the collision event; explicit
finite element analyses are presented for a case study of a collision scenario. In particular, uncertainties of input parameters in the finite element simulations and their impact on the shape and size of the
damage opening area, and time to capsize of the struck ship, are addressed. Material modelling aspects are studied including material properties within a material class as well as damage modelling.
In addition, the effects of using a deformable or rigid striking bow section, the friction coefficient, the
collision angle and the speed of the striking ship are studied. Recommendations for a sufficient level
of simplifications for arriving at reliable results in numerical simulation of ship collisions are made.
Based on these findings, recommendations on how ship collision analyses should be set up have been
made: 1. The dispersion of the material: The analyst should be aware that the properties of a material
may have a dispersion that gives large consequences on the outcome of the analysis and account for
this. 2. Failure criteria: The study has shown that the choice of failure criterion has a significant influence on the outcome of the analysis. 3. Modelling of striking bow section: In the variance analyses
this factor is shown to have the highest influence on the results of the surveyed factors. Using a rigid
bow section is a very crude assumption that should only be used in comparative studies. When an
analysis of a real collision scenario is to be carried out, a deformable bow section should be used if
the results are to have any significance. 4. Friction coefficient in contact conditions: This factor has
minor influence on the outcome of the analyses. However, a conservative analysis should use a low
friction coefficient; a realistic value for wet plates is 0.1. 5. Collision angle and striking ship speed:
The speed of the striking ship has major influence on the results since it determines the amount of
kinetic energy that needs to be absorbed by the structures. The effect of the collision angle has no
statistical significance for the geometries used in the current study. A conclusion is that the solution
is highly dependent on geometrical effects that are hard to predict. Thus, different collision angles
should be analysed in order to find which one gives the most severe results for the geometry being
studied.
In Lee et al. (2012a) a Safety Assessment of Ship Collision using FSI Analysis Technique is presented. Structural safety assessment was performed for the collision accident of specialized ship structure and its cargo. To ensure reasonable and reliable safety assessment, accurate and correct full-scale
ship collision simulation was carried out using FSI analysis technique of LS-DYNA code and propulsion force instead of velocity in collision simulation.
Lee et al. (2012b) used this approach for iceberg collisions. In this study, full-scale collision simulation was carried out to estimate a more reliable and realistic collision impact response. For the investigation of the effects of the type, mesh size, material model and property of iceberg on the collision
responses, full-scale collision simulations were also performed in the void (air) condition.
3.7
Sloshing Control
Kim et al. (2013) studded the use of an anti-sloshing blanket system to prevent sloshing. A comparative study is presented for the development of a new device for the reduction of the sloshing
impact load on an LNG cargo containment system. A floater-type blanket is proposed as a new device. The floater floats and deforms on the free surface inside an LNG tank, and the floaters material can work properly at extremely low temperature. To observe the efficiency of this device, a series
of 3D model tests is performed at 1/50 model scale. Slosh-induced impact pressures are measured
on the inner side of the tank wall in irregular motion excitations, with varying filling depths and sea
states. It is observed that this floater-type device can reduce the slosh-induced impact pressure in
most conditions.
3.8
Li et al. (2012) reported the application of Safety Science and Technology to LNG ships, in which
analyses were carried out for the fire fatalness of LNG ships. The risk evaluations of fire and explosion in the LNG ship were made at a different angle by use of the Dow method, BLEVE (boiling
liquid expanding vapour explosion) model and VCE (vapour cloud explosion) model. The clear, accurate, comprehensive evaluation findings were obtained, in which the fire and explosion hazard index
F&EI is 168 and the degree of risk is very large before the security measure compensations. However, after security measure compensations, fire and explosion risk index F&EI falls to 67 and the degree
of risk is reduced. The results could be applied as references to the scientific management and the
adoption of safety measures for the shipping enterprises that plan and participate actively to import
LNG project.
606
In addition to their 2004 (Sandia Laboratories) report on spillage of LNG, Blanchet et al. (2013)
presented a study on LNG ship insulation experiments using large LNG pool fire boundary conditions.
The tests were conducted on a large scale in an LNG pool, using liquid methane.
Figure 6. Surface Emissive Power vs. Pool Diameter for various hydrocarbon fuels Reference should be
given for this and the next figure.
Figure 7. LNG pool fire test siteSandia Lab.
From the Fire Test for Tank Insulations, the following observations were taken: The steel plate approached 1100C and the insulation layers closest to the steel plate reached 800C; and all insulation
systems suffer degradation and reduction in mechanical strength.
From the effect of Large Spill as described in table 5, the following conclusions were taken:
Significant reduction in strength
Simultaneous multi-cargo tank cascading damage spill are unlikely
Risk mitigation categories identified:
o Options to help reduce the possibility of spills, especially large spills,
o Options to reduce ship damage if a large spill did occur, and
o Ship design modifications that could help protect the ship from damage and the crew from
danger from a large spill and fire
607
Table 5. spill.
Figure 8. Progression of fire.
Petti et al. (2013) analyzed the structural and thermal consequences of LNG vessel cascading damage.
The authors examined susceptibility of an LNG ship and cargo tanks to cryogenic and fire damage
from a large LNG spill. Moss and Membrane type LNG ship design were investigated.
As a conclusion the structural integrity can be severely compromised for large spills with significant
sections of the hull cracked and no longer capable of effectively resisting any loads. Both ships are
judged to be disabled and severely damaged due to a large cargo tank breach and spill. Neither ship
would be capable of movement and would need to have any remaining LNG cargo transferred. Simultaneous multi-cargo tank cascading damage spill scenarios are judged to be unlikely, though sequential
tank spills are possible. These sequential spills are not expected to increase the hazard distances, but
could increase the duration of the fire.
Moss LNG Ship cross-section
Membrane LNG Ship cross-section

Figure 9. Moss and membrane tanker.
608
4.
LNG AS FUEL
A note from the GL and LR brochure.
4.1
Why LNG as Fuel
Using liquefied natural gas (LNG) as ship fuel has recently gained more attention not only in Europe,
but also in Asia and the USA. There are three drivers visible which make LNG as ship fuel one of the
most promising new technologies for shipping:
First, using LNG as ship fuel will reduce sulphur oxide (SOx) emissions, which are created when
using fuel with high sulphur content, by between 90% and 95%. This reduction level will become
mandatory within the so-called Emission Control Areas (ECAs) from 2015. A similar reduction will
be enforced for worldwide shipping from 2020 on, pending a review at the IMO which may shift
the introduction to 2025.
In addition, the reduction of nitrogen oxide (NOx) emissions to comply with IMO Tier III limits,
applicable in ECAs from 2016, can be achieved for both pure gas engines and four-stroke dual-fuel
engines, which are typically used on board ships engaged in short sea and coastal shipping.
Second, due to the lower carbon content of LNG compared to traditional ship fuels, a 20%25%
reduction in carbon dioxide (CO2) emissions is possible. The actual reduction depends on engine
type and the range of possible measures for reducing the unwanted release of unused methane.
Third, current LNG prices in Europe and the USA suggest that LNG could be offered at a price
comparable to heavy fuel oil (HFO), taking into account its energy content and the costs of smallscale LNG distribution (not yet available). This means that LNG will certainly look commercially
attractive compared with the low-sulphur marine gas oil (MGO) which will be required to be used
within the ECAs if no other technical measures are implemented to reduce SOx emissions.
A more systematic analysis of costs and benefits for LNG-fuelled container vessels was recently
completed by GL and leading engine manufacturer MAN, who is currently preparing to launch a range
of LNG-fuelled two-stroke engines in 2013. Five container vessel sizes and three different operational
profiles were investigated by systematically changing the price of LNG and the proportion of time
spent operating within ECAs.
Results for using LNG as fuel were compared with the principal other technical alternatives, the
implementation of exhaust gas cleaning systems (scrubbers), to reduce emissions to required levels.
The decision to use LNG as ship fuel as opposed to using a scrubber system would depend on the
price differential between LNG and HFO, the proportion of time spent operating within ECAs and the
starting year for the vessel. Most LNG systems are expected to become profitable in 2020, when a
global low-sulphur fuel standard is expected to be implemented.
4.2
LNG Supply Chain
A number of initiatives are taken regarding LNG-bunkering stations in order to make shipping more
environmentally friendly and competitive. One possible supply chain for LNG as fuel as an example is
described by Wang et al. (2014) as follows:
A small LNG tanker could load, say, 10,000 m of LNG at an LNG terminal, e.g. Zeebrugge, and
transport her cargo to an intermediate LNG terminal. Several small LNG tankers already exist and are
thought to be available to fulfil this task. Intermediate terminals have been built, such as the one near
Stockholm, and more are planned for Gothenburg and several other locations in the Baltic region. The
port of Hamburg also considers this to be an option for the future supply of LNG to ships.
The last step of supplying LNG to the end-user is performed by LNG bunker vessels, none of which
have yet been built. These would be small LNG tankers, too, and would deliver the LNG to the ship in
a fashion similar to the present bunker delivery of HFO. It is noted, however, that technical and regulatory challenges need to be overcome before LNG bunkering in ports becomes standard.
Regulatory Development
Class society GL has been working on a number of projects involving gas as a ship fuel and is participating in the development of the IMO rules for gas as a ship fuel on behalf of the German government.
Ship Design
The development of rules and design concepts of LNG-fuelled vessels
The Bit Viking is the worlds first vessel in service whose main machinery has been converted to
burn LNG as fuel. She is also the largest commercial vessel which is not an LNG tanker, to use LNG
609
as fuel. The conversion saw the main engines replaced and two large LNG tanks installed on deck,
each with a capacity of 500 m, giving the vessel a range under LNG of 12 days. Conversions of existing vessels to use LNG as ship fuel, however, are assumed to be limited due to a lack of engine type
retrofit options and available space for LNG tanks. The first vessels to use LNG as ship fuel were
large LNG tankers using boil-off gas as fuel. Later generations of LNG tankers also used traditional
ship fuels such as HFO. The first vessels built to use LNG as fuel, and to thus benefit from lower
emissions levels, were a ferry (in 2000) and two offshore service vessels (in 2003) operating in Norwegian waters. To date, another 13 such vessels have been commissioned, making Norway the country with the largest fleet of LNG-fuelled vessels. At the same time, an LNG supply infrastructure has
been established along the Norwegian coast.
At present, there are a number of vessels on order which are designed to use LNG as ship fuel. A
ferry has been ordered by Viking Line for operation between Finland and Sweden, and a general cargo
ship has recently been ordered by Norline for operation in the North Sea. In addition, a number of
offshore service vessels are on order, two of which are intended for an operation in US waters off the
Gulf of Mexico.
The first study on an LNG-fuelled container feeder vessel was published by GL in early 2009 and
demonstrated the concepts technical feasibility and commercial attractiveness for operations within
an emission control area (ECA), as compared with a standard vessel of today. In addition, GL has
issued approval in principle (AIP) certificates, on the basis of the GL guidelines, to two container vessels designed to use LNG as ship fuel and with capacities of 4,100 TEU and 14,000 TEU, respectively.
Noteworthy article in Der Spiegel English edition: http://www.spiegel.de /international/business/newimo-regulations-push-shipping-industry-toward-cleaner-fuel-and-lng-a-916811.html.
In the Netherlands a strong initiative is taken by the inland ship owners to look for LNG as fuel alternative. This is heavily supported by the Harbour of Rotterdam in order to meet its exhaust goals for
2020. (in dutch) http://www.deenshipping.com/nl_NL/nieuws.html/nieuwsbericht/29
In a recent study Wang et al. (2014) showed the pitfalls in regulatory issues to use LNG as fuel for
both sea going as well as inland vessel. There is no international rule recognising that LNG can be
used as a marine fuel, apart from the IGC code which allows LNG carriers to use boil-off gas as a part
of the ships propulsion. In order to fill this gap, the IMO has started to draft the international code of
safety for ships using gases or other low flashpoint fuels (IGF code) which will cover safety and operational issues for gas-fuelled seagoing vessels. The code is expected to be finalised by 2014. In addition, the lack of a set of comprehensive LNG bunkering regulations is one of the key barriers to the
new application. So far, no international standards have been established which incorporate minimum
requirements for the bunkering procedures, training and equipment necessary to ensure safe LNG handling for gas-fuelled ships via both shore-based and ship-to-ship bunkering operations.
The ISO is working on/issued a first draft of such international guidelines for harmonising LNG
bunkering standards.
Other hindrances are related to the use of LNG on inland vessels in Europe. In line with the relevant
European agreement concerning inland shipping e.g. the international carriage of dangerous goods by
inland waterways (ADN) and the Rhine vessel inspection regulations (RVIR); the regimes prohibit the
installation on inland ships of combustion engines that use a fuel with a flashpoint below 55degC. This
means LNG is restricted to be used as a fuel since its flashpoint is -180degC. To close this regulatory
gap, the EU authority has started to establish a specific permit process for LNG-powered inland vessels and later may develop appendices under the existing regulatory framework.
(http://www.porttechnology.org/images/uploads/technical_papers/LNG_LR.pdf). A conclusion from
their study is that one can see that LNG as a clear clean fuel can reach all environmental targets without any abatement technology. However, the current lack of bunkering infrastructure and operation
standards imply that the use of LNG as a ship fuel is expected to first gain momentum in niche markets, like small ferry routes and regional liner traffic. In the longer run (perhaps from 2020) the adoption of LNG as a ship fuel on a global scale rests on three main factors: the price difference between
LNG and low sulphur fuel oil; the global emission regulations e.g. the global SOx limits enforced in
2020 or 2025; the availability of LNG bunkering facilities in a global context.
For sea-going vessel, Kim et al. (2013) have conducted a study for structural development of LNG
as fueled in a large container ship. There is a growing interest in the appearance of LNG fueled large
commercial ship with rising oil price and stronger emission regulation of these days. Though small or
middle size LNG fueled vessels have been already operated, the application of LNG fuel to large
commercial vessels is now under development. The report describes engineering works such as structural analysis, crack propagation & leakage analysis and temperature distribution analysis for the
application of ACT-iB (Aluminum Cargo Tank Independent type B ) to LNG fueled vessel. From this
development, it was concluded that LNG fueled technologies are ready to large container ship application in structural point of view. However, the engineering works are limited to the design safety construction of LNG tank without risk of collision. The specific research in collision issue for LNG tanks
is needed in the future.
610
Figure 10. Positive and negative factors in the adoption of LNG as a ship fuel. The percentages refer to the share
on a total of 33 studies.
5.
SAFETY AND DESIGN SPECIAL APPLICATIONS
5.1
Chen et al. (2013) reported the analysis of advanced hydrodynamic of FLNG terminals. They consist
of the complex interaction of multiple bodies and the coupling effect of seakeeping (wave diffraction
and radiation around bodies) and sloshing (liquid motions in tanks). Based on the recent development
to introduce the dissipation in potential flows and new formulations of boundary element method, the
seakeeping analysis is enhanced to be able to make accurate predictions of gap resonances and major
dynamic effect of liquid motion in tanks.
The report has presented the numerical models that BV has developed in order to properly predict
the relative motions between the FLNG and LNGC during offloading by taking into account all the
complex hydrodynamic phenomena. According to BV there are three focus operations, which must be
examined and result in a series of advance studies for FLNG Terminals:
Liquefaction process with regards to limited deck space and floating unit motions and acceleration (FLNG & FSRU)
Containment system with regards to offshore conditions for continuous operation.
LNG offloading from the FLNG to LNGC or from the FSRU to LNGC.
Regarding the containment system, one of major issues to be addressed is the operation
at all filling levels on the contrary to standard LNGC which can navigate only at low (below 10% of the height) and high filling levels (above 70% of the height) and not at partial
fillings (between 10% and 70% of the height). In LNG Terminal both of FSRU or FLNG as
BV experience with LNGRV that can operate at all filling levels for some given locations.
Furthermore BV has issued new guidelines ([NI554], [NI564]) for sloshing assessment for
LNG terminals.
[NI554] seakeeping analysis, CFD calculations and sloshing model test which allow to
determine the design sloshing loads on the cargo containtment system, the inner hull
structure and the pump mast.
[NI564] two levels of strength assessment (a rule based approach and more refined approach using non-linear and dynamic FEA) on the cargo containtment system, the inner
hull structure and the pump mast.
611
The LNG offloading operability and reliability (FLNG to LNGC) is one of the critical point
for the FLNG development.
If FLNG is turret moored, this FLNG should be weathervaned. A heading analysis is then
mandatory to determine the dominant wave heading.
Multibody interaction with a small gap between both units may induce some unexpected
motions.
Coupling between seakeeping and liquid motions inside the tanks is to be taken into account
for each unit.
In advance, Clauss et al. (2013) have presented the study about side by side LNG transfer at rough
sea since the side by side LNG transfer method is limited regarding flexible pipe design. It is because
of hydrodynamic characteristics of this multibody system, one key aspect is the analysis of the exciting forces and motions due to wave amplification between the ships. In the gap between the hulls, the
incoming wave field is amplified and changes dramatically. Clauss et al. (2013) conducted the numerical approaches in frequency domain of investigating the gap effects. Finally suitable transfer configurations that ensure safe LNG offshore transfer up to Hs = 3 m have resulted with typically offloading
scenario.
Also, Wilde et al. (2013) have addressed the major findings in terms of weather limitations, tugboat
requirements and other critical aspects for the berthing and offloading operation with hydrodynamic
and nautical studies for offshore LNG operations. These were based on full mission bridge simulations, model tests campaigns, time domain simulations, fast time maneuvering simulations and downtime assessments.
5.2
Side by side or Tandem mooring?
The choice of mooring the LNG carrier close to the FLNG is quite different from traditional mooring
of FPSO and shuttle tankers. The latter can operate in tandem which gives quite some operational benefits. There is a tendency to develop a tandem cryogenic solution for offloading LNG.
Mauries et al. (2014) describe the development of an LNG Tandem Offloading System using floating cryogenic hoses. Their system is composed of a compact hose storage system on the LNG Terminal allowing to store the hoses between two offloading operations, of a connection head (hose end
termination piece) to ease the deployment/retrieval of the hoses and of a storage and maintenance platform for the connection head at the aft of the LNG Terminal, also allowing to replace a hose section in
offshore conditions. On the LNG Carrier side, a bow loading platform is installed to ensure hoses connection even in exposed environmental conditions.
Figure 11. Possible LNG-FLNG tandem configuration using Dynamic Positioning next to traditional side-byside solution (Prelude) (see www.eshiptrading.com)
Zhao et al. (2013) describe the hydrodynamics of an FLNG system in tandem offloading operation.
In their study, the hydrodynamic characteristics of a single point turret-moored FLNG system in tandem offloading operation are investigated by using a time-domain coupled dynamic analysis program
and a set of comparative model tests. The numerical simulation model features well the hydrodynamic
performance of the coupled system obtained from the experiments. The influence of distance between
the two vessels on their hydrodynamic performance has been investigated. Furthermore, two different
ways of connecting FLNG vessel and LNG carrier have also been comparatively studied. The numerical results show that there is significant difference at the hydrodynamic performance in the two ways
of tandem offloading operations, which means that the connection between FLNG vessel and LNG
carrier plays an important role. The outcome of this study would offer better understanding on the
hydrodynamics of multi-bodies, which can further lead to more practical applications for the design
and operation of FLNG.
612
5.3

Arctic
Developments of arctic LNG projects

At present there are 11 arctic LNG projects at various stages of development. Some have been put on
hold, e.g. extension of Snohvit (NOR), Shtokman (RUS), others are on track or even accelerating, e.g.
Pechora (RUS), Yamal (RUS), or Kitimat (CAN). Western Russian projects intend LNG export
through arctic waters to European markets. Therefore there is demand for specialized arctic LNG carriers. Designs and actual vessels are already under development. With the start of shale gas production
in North America, project interests have been shifted but still demand for arctic LNG shipping
remains.
Oh et al. (2013) studied the safety assessment of membrane LNG carrier under ice impact. This
study described the new technique of numerical analysis and safety assessment procedure for repeated
ice impact. Tryaskin et al. (2012) presented the mathematical modeling of the liquid impact on the
tank walls of the membrane-type LNG carrier after impact with an ice barrier. Apart from the semiempirical methods described in the Rules of classification and construction an alternate method based
on OpenFoam CFD modeling is presented. In this paper the problem of jamming of the ship hull into
the heavy ice conditions is solved, that allowed to estimate the parameters of the ship stopping. The
simulation of the liquid gas flow in the prismatic membrane-type LNG tank after impact with ice barrier is presented. Unfortunately no validation experiments were shown. The results of the pressure
loads on the wall of the tank and flow patterns are presented.
6.
CONCLUSIONS
For the safety of the transport of LNG there is a need for development of consistent regulations and
LNG spillage and safety handling and protection are areas where further investigation work may also
be beneficial.
Further study is required on LNG Offshore Process and systems operability of LNG. Accurate prediction and evaluations of sloshing loads at any filling levels is necessary for the safety assessment of
the containments. Semi-empirical approach for LNGC operation cannot be generalized and the safety
aspects should be further investigated. LNG spillage and safety handling and protection are areas
where further investigation work may also be beneficial. For the FLNG development, the LNG offloading operability and reliability (FLNG to LNGC) is one of the critical issues. The nautical aspects
of side by side mooring need further investigation. Further study on the viability of tandem offloading
solutions for LNG transfer should be pursued.
Implication of the new IGC code
There are many new LNG containment systems under development that do not fit into the established
IGC code definitions of tanks. New generic regulations need to be established to handle new innovative containment system designs.
LNG as fuel both for inland as well as seagoing vessels
LNG as fuel is a new area coming quickly as an attractive fuel alternative in shipping. LNG will be
applied for any type of ships and LNG fuel tanks may be located in other areas than in cargo areas.
A new IGF code is currently under development and its implications should be studied. Also
regulatory gaps e.g. in LNG bunkering should be revisited. One can see that LNG as a clear fuel can
reach all environmental targets without any abatement technology. However, the current lack of
bunkering infrastructure and operation standards imply that the use of LNG as a ship fuel is expected
to first gain momentum in niche markets, like small ferry routes and regional liner traffic. In the longer
run (perhaps from 2020) the adoption of LNG as a ship fuel on a global scale rests on three main
factors: the price difference between LNG and low sulphur fuel oil; the global emission regulations
e.g. the global SOx limits enforced in 2020 or 2025; the availability of LNG bunkering facilities in a
global context.
In-situ measurements and inspections
Hull structures and containment inspection systems still need a connection with the operation of the
vessel e.g. those used on FPSOs.
REFERENCES
Ancellin, M., Brosset, L., Ghidaglia, J.-M. (2012). Influence of Phase Transition on Sloshing Impact Pressures
Described by a Generalized Bagnold's Model. Proceedings of the Twenty-second (2012) International Offshore and Polar Engineering Conference, Rhodes, Greece, June 1722, 2012.
Andr Baeten: Finite Particle based Elastic Structure Dynamics Modeling for LNG Sloshing Analysis, Proceedings of the 2013 ISOPE conference.
613
Ansari, M.R., R. D. Firouz-Abadi, M. Ghasemi: Two phase modal analysis of nonlinear sloshing in a rectangular
container Ocean Engineering 38 (2011) 12771282
Antuono, M., B. Bouscasse, A. Colagrossi and C. Lugni: Two-dimensional modal method for shallow-water
sloshing in rectangular basins Two-dimensional modal method for shallow-water sloshing in rectangular basins Journal of Fluid Mechanics/Volume 700/June 2012, pp 419440
Ashkan Rafiee1, Nitin Repalle2 and Frederic Dias: Numerical Simulations of 2D Liquid Impact Benchmark
Problem Using Twophase Compressible and Incompressible Methods, Proceedings of the 2013 ISOPE conference.
Bang, Jun-Kee. Heo, Joo-Ho. Jeong, Tae-Soek Lee, Cheol-Ho. Pyun, Jang-Hoon. Park, Jung-Kyu (2012). Structure Safety Assessment for Independent LNG Cargo Tank Considering the Support Structure, Proceedings of
Twenty-second (2012) International Offshore and Polar Engineering Conference (ISOPE), Rhodes, Greece.
Baudin, E, B. Fillon and L. Diebold (2012). An Alternate Impact Measurement for the Sloshing Impact Pressures. Proceedings of the Twenty-second (2012) International Offshore and Polar Engineering Conference,
Rhodes, Greece, June 1722, 2012.
Bechara Kayal and Christian-Frdric Berthon (2013). Analytical Approach to Predict Sloshing Severity in LNG
Membrane Tanks Based on Optimized Series of Model Tests. Proceedings of the Twenty-third (2013) International Offshore and Polar Engineering Conference, Anchorage, Alaska, USA, June 30July 5, 2013.
Benyessaad, Olivier. and Ruf, Diane. (2013). Risk and Reliability Analyses for Innovative LNG Offshore Floating Units, OMAE 2013, Paper number 10349.
Bernhard Godderidge, Stephen R. Turnock, Mingyi Tan (2012). A rapid method for the simulation of sloshing
using a mathematical model based on the pendulum equation. Computers & Fluids 57 (2012) 163171.
Bisotto, Paulo. and Rouhan, Antoine. (2004). Survey and Inspection Management for FPSOS. International
Conference on Ocean, Offshore and Artic Engineering (OMAE 2004), Paper number 51433.
Blanchet, C. Morrow M. Hightower LNG ship insulation experiments using large LNG pool fire boundary
conditions, LNG 17 Conference, Texas 2013
Brosset, L., Ghidaglia, J.-M., Guilcher, P.-M., Le Tarnec, L. (2013). Generalized Bagnold Model. Proceedings of
the Twenty-third (2013) International Offshore and Polar Engineering Conference, Anchorage, Alaska, USA,
June 30July 5, 2013.
Brosset, L., W. Lafeber, H. Bogaert, M. Marhem, P. Carden, J. Maguire: A Mark III Panel Subjected to a Flipthrough Wave Impact: Results from the Sloshel Project, Proceedings of the 2013 ISOPE conference.
Cao, X. Y. F. R. Ming, A. M. Zhang: Sloshing in a rectangular tank based on SPH simulation,, Applied Ocean
Research, Volume 47, August 2014, Pages 241254
Carden, E. P. C., Zegos, N. J. Durley-Boot, J. R. Maguire, D. Radosavljevic, S. Whitworth: Modelling of Full
Scale Wave Impacts on the Sloshel Mk III Test Panel, Proceedings of the 2013 ISOPE conference.
Chen Z., Z. Zong, H. T. Li, J. Li: An investigation in to the pressure on solid walls in 2D sloshing using SPHmethod Ocean Engineering 59 (2013) 129141
Chen, Xiao-Bo. Diebold, Louis. and Guillaume de-Hautecloque. (2013). Hydrodynamic Issues of FLNG Systems. OMAE 2013, Paper Number 11450.
Clauss, Guenther F. Dudek, Matthias and Testa, Daniel. (2013). Gap Effects at Side-by-Side LNG-Transfer
Operations. OMAE 2013, Paper Number 10749.
Daniel T. Valentine: Internal-Wave Sloshing In Triangular-Shaped Containers: A Numerical Study, Proceedings
of the ASME 2013 International Conference on Ocean, Offshore and Arctic Engineering
Daniel Villa Mike Hightower: LNG Vessel Cascading Damage structural and Thermal Analyses. Sandia lab rep.
2013, LNG 17
De Chowdhury, S., S.A. Sannasiraj: Numerical simulation of 2D sloshing waves using SPH with diffusive terms,
Applied Ocean Research, Volume 47, August 2014, Pages 219240
Dematteo, A and A. Ratouis (2013). Stochastic Simulation of the Pressure Field Induced by Sloshing Impacts.
Proceedings of the Twenty-third (2013) International Offshore and Polar Engineering Conference, Anchorage, Alaska, USA, June 30July 5, 2013.
Dematteo, A. and E. Gervaise (2014). Assessing the Risk Induced by Sloshing Using Generalized Pareto Distributions. Proceedings of the Twenty-fourth (2014) International Ocean and Polar Engineering Conference, Busan, Korea, June 1520, 2014.
Dong Young Lee, Yunho Kim and Hang S. Choi: A Study of the Sloshing Mode on the Motion of 2D Rectangular Tanks in Regular Waves, Proceedings of the 2013 ISOPE conference.
Fabrizio Pistani and KrishThiagarajan (2012). Experimental measurements and data analysis of the impact pressures in a sloshing experiment. Ocean Engineering 52 (2012) 6074
Fabrizio Pistani, KrishThiagarajan: Experimental measurements and data analysis of the impact pressures in a
sloshing experiment Ocean Engineering 52 (2012) 6074
Fabrizio Pistani_,Krish Thiagarajan Dominique Roddier Tim Finnigan: Comparison Of Sloshing Impacts For
Rectangular And Chamfered Lng Tanks, Proceedings of the ASME 2012 International Conference on Ocean,
Offshore and Arctic Engineering
Fillon, B., J. Henry, L. Diebold and Q. Derbanne (2013). Extreme Values Theory Applied to Sloshing Pressure
Peaks. Proceedings of the Twenty-third (2013) International Offshore and Polar Engineering Conference, Anchorage, Alaska, USA, June 30July 5, 2013.
Fu-kun Gui and Sheng-chao Jiang: Numerical Simulation of Liquid Sloshing Problem under Resonant Excitation, Advances in Mechanical Engineering, Volume 2014,
614
Fuxin Huang, Renchuan Zhu, Chi Yang and Francis Noblesse (2013). Numerical Studies of Coupling Effects
between Liquid Sloshing and Ship Motions. Proceedings of the Twenty-third (2013) International Offshore
and Polar Engineering Conference, Anchorage, Alaska, USA, June 30July 5, 2013.
Guilcher P.M., L. Brosset, N. Couty and D. Le Touz: Simulations of Breaking Wave Impacts on a Rigid Wall at
Two Different Scales with a Two Phase Fluid Compressible SPH Model, Proceedings of the 2011 ISOPE
conference.
Hamid Alemi Ardakani And Thomas J. Bridges: Shallow-water sloshing in vessels undergoing prescribed rigidbody motion in three dimensions Journal of Fluid Mechanics, Volume 667, January 2011, pp. 474519
Hannes Bogaert, Miroslaw Lech (Mirek) Kaminski Laurent Brosset: Full And Large Scale Wave Impact Tests
For A Better Understanding Of Sloshing Results Of The Sloshel Project, Proceedings of the ASME 2011
International Conference on Ocean, Offshore and Arctic Engineering
Hirotsugu Dobashi and Akio Usami: Dynamic Amplification Factor of NO96 Insulation Structures of Membrane
System, Proceedings of the 2013 ISOPE conference.
Hitoshi Gotoh, Abbas Khayyer, Hiroyuki Ikari, Taro Arikawa, Kenichiro Shimosako: On enhancement of Incompressible SPH method for simulation of violent sloshing flows, Applied Ocean Research, Volume 46,
June 2014, pp. 104115
Hoonkyu Oh, Whasoo Kim, Byeongjae Noh, Claude Daley and Jaemyung Lee: Safety Assessment of Membrane
Type LNG Carrier under Repeated Ice Impact 2013, Proceedings of the 2013 ISOPE conference.
Hu In Choi, Sun Hong Kwon Ki Hyun Lee Sang Beom Lee Young Jun Yang: An Experimental Study On Hydro-Elasticity In Sloshing, Proceedings of the ASME 2013 International Conference on Ocean, Offshore and
Arctic Engineering
Hu Zhiqiang, Amdahl Jrgen, Hong Lin: Verification of a simplified analytical method for predictions of ship
groundings over large contact surfaces by numerical simulations, Marine Structures 24 (2011) 436458
Huirong Jia Torgeir Moan: The Effect Of Sloshing In Tanks On Motions And Hull Girder Responses Of Damaged Vessels, Proceedings of the ASME 2013 International Conference on Ocean, Offshore and Arctic Engineering
Hwang, Se Yun. Lee, Jang Hyun and Kim, Sung Chan. (2013). Structural Response Analysis of LNG CCS Experiencing The Sloshing Impact Determined by Both Convolution and Fluid Structure Interaction Method.
OMAE 2013, Paper Number 11336.
IMO, International Code for the Construction and Equipment of Ships Carrying Liquefied Gases in Bulk (IGC
Code) 2014 by the International Maritime Organization
Jason P. Petti Carlos Lopez Victor Figueroa Robert J. Kalan Gerald Wellman James Dempsey, ISSC Report
2006 Collision and grounding, Southampton.
Jeong Hwan Kim, Joong Hyo Choi, Sun Gun Park, Dong Kwon Lee, Yeong Tae Oh: Structural Development of
LNG Fueled Large Container Ship, Proceedings of the ASME 2013 32nd OMAE International conference. I
Jerome de Lauzon, Louis Diebold, ime Malenica: Sloshel Project Strength Assessment of LNG Cargo Containment Systems under Sloshing Loads, Proceedings of the 2013 ISOPE conference.
Jess Gmez-Goi, Carlos A. Garrido-Mendoza, Jos Luis Cercs, Leo Gonzlez: Two phase analysis of sloshing in a rectangular container with Volume of Fluid (VOF) methods Ocean Engineering 73(2013)
208212
Jong Jin Park Hiroshi Kawabe Mun Sung Kim Byung Woo Kim Jae Kwang Eom: Numerical Sloshing Analysis
of FlNG and LNGC Considering Two-Body Motion Effect, Proceedings of the ASME 2012 International
Conference on Ocean, Offshore and Arctic Engineering
Jong Jin Park, Jang Hoon Seo, Chang Hun Jin, Ki Hun Joh, Byung Woo Kim, Yong Suk Suh and Yonghwan
Kim (2014). Sloshing Assessment of LNG Vessels for Unrestricted Tank Filling Operation. Proceedings of
the Twenty-fourth (2014) International Ocean and Polar Engineering Conference, Busan, Korea, June 1520,
2014.
Jong Jin Park, Jang Hoon Seo, Munsung Kim, Byung Woo Kim and Yong Suk Suh (2013). Coupled Sloshing
and Motion Analysis of FLNG in Side-by-Side Arrangement. Proceedings of the Twenty-third (2013) International Offshore and Polar Engineering Conference, Anchorage, Alaska, USA, June 30July 5, 2013.
Jong Jin Park, Jang Hoon Seo, Munsung Kim, Byung Woo Kim and Yong Suk Suh:
Jun-Kee Bang, Joo-Ho Heo, Tae-Seok Jeong, Cheol-Ho Lee, Jang-Hoon Pyun and Jung-Kyu Park: Structure
Safety Assessment for Independent LNG Cargo Tank Considering the Support Structure, Proceedings of the
2012 ISOPE conference.
Jun-Kee Bang, Joo-Ho Heo, Tae-Seok Jeong, Cheol-Ho Lee, Jang-Hoon Pyun and Jung-Kyu Park: Structural
Safety Assessment for Support Structure of Independent LNG Cargo Tank (IMO Type B) Considering
High Level Reaction Force, Proceedings of the 2013 ISOPE conference.
Kimmoun, O., A. Ratouis and L. Brosset: Influence of a Bubble Curtain on the Impact of Waves on a Vertical
Wall, Proceedings of the 2013 ISOPE conference.
Ku, A. Krzonkalla, V. Wang, X. Chen, J and Howser, Rachel. (2012). Structural Reliability Application in Risk
Based Inspection Plan for Semi-Submersible Floating Structure. Deep Offshore Technology (D.O.T) Conference.
Lafeber, W, L. Brosset and H. Bogaert: Comparison of Wave Impact Tests at Large and Full Scale: Results from
the Sloshel Project, Proceedings of the 2011 ISOPE conference.
Lafeber, W., Brosset, L., Bogaert, H. (2012). Elementary Loading Processes (ELP) involved in breaking wave
impacts: findings from the Sloshel project. Proceedings of the Twenty-second (2012) International Offshore
and Polar Engineering Conference, Rhodes, Greece, June 1722, 2012.
615
Lanquetin, B. Rouhan, A. and Gourdet, G. (2007). Risk Based Inspection on F(P)SOs Hulls: Case Studies on
New Built Units. OMC 2007.
Laurent Brosset, Jean-Michel Ghidaglia, Pierre-Michel Guilcher and Louis Le Tarnec: Generalized Bagnold
Model, Proceedings of the 2013 ISOPE conference.
Lee, Sang-Gab. Zhao, Tuo. (2013). Structural Safety Assessment of LNGC CCS under Iceberg Collision using
FSI Analysis Technique, ISOPE 2013.
Li, H. Zhenghu: Fire and explosion risk analysis and evaluation for LNG ships, Procedia Engineering 45 (2012)
7076 2012
Louis Diebold, Quentin Derbanne and Thomas Gazzola (2013). Statistical Behavior of Global and Local Sloshing Key Parameters. Proceedings of the Twenty-third (2013) International Offshore and Polar Engineering
Conference, Anchorage, Alaska, USA, June 30July 5, 2013.
Louis Diebold, Quentin Derbanne and Thomas Gazzola: Statistical Behavior of Global and Local Sloshing Key
Parameters, Proceedings of the 2013 ISOPE conference.
Loysel, T., Chollet, S., Gervaise, E., Brosset, L., De Seze, P.-E. (2012). Results of the First Sloshing Model Test
Benchmark. Proceedings of the Twenty-second (2012) International Offshore and Polar Engineering Conference, Rhodes, Greece, June 1722, 2012.
Loysel, T., Gervaise, E., Moreau, S., Brosset, L. (2013). Results of the 2012-2013 Sloshing Model Test Benchmark. Proceedings of the Twenty-third (2013) International Offshore and Polar Engineering Conference,
Anchorage, Alaska, USA, June 30July 5, 2013.
Mateusz Graczyk, Kjetil Berget and Joachim Allers: Experimental Investigation of Invar Edge Effect in Membrane LNG Tanks, J. Offshore Mech. Arct. Eng. 134(3), 031801 (Feb 01, 2012) (7 pages)
doi:10.1115/1.4005183
Matthieu Ancellin, Laurent Brosset, Jean-Michel Ghidaglia: Influence of Phase Transition on Sloshing Impact
Pressures Described by a Generalized Bagnolds Model, Proceedings of the 2013 ISOPE conference.
Mattia Fossa, Cesare M. Rizzo, Giorgio Tani and Michele Viviani: Simulations of a Sloshing Experiment by
FEM CFD and FEM FSI Approaches, Proceedings of the 2013 ISOPE conference.
Mauries, B, F. Benoit F. Lirola V. Lagarrigue: Development of an LNG Tandem Offloading System Using
Floating Cryogenic Hoses Breaking the Boundaries of LNG Transfer in Open Seas, Proceeding of the Offshore Technology Conference 2014 OTC-25342-MS
Miaozi Zheng, Yonghwan Kim, Sang-Yeob Kim and Yangjun Ahn (2013). Study on the Prediction of Sloshing
Severity. Proceedings of the Twenty-third (2013) International Offshore and Polar Engineering Conference,
Miaozi Zheng, Yonghwan Kim, Sang-Yeob Kim and Yangjun Ahn: Study on the Prediction of Sloshing Severity, Proceedings of the 2013 ISOPE conference.
Miguel A. Celis, Juan B. V. Wanderley: A Numerical Investigation On 2d And 3d Sloshing Effects, Proceedings
of the ASME 2011 International Conference on Ocean, Offshore and Arctic Engineering
Min Han Oh, Jong Min Kim, Joong Soo Moon, Woo Seung Sim and Hyun Soo Shin (2014). Sloshing Assessment of FLNG Cargo Containment System due to Sloshing Loads in Bi-modal Seas Effect of Wave
Intensity. Proceedings of the Twenty-fourth (2014) International Ocean and Polar Engineering Conference,
Busan, Korea, June 15-20, 2014.
Mitra S., C. Z. Wang, J. N. Reddy, B.C. Khoo: A 3D fully coupled analysis of nonlinear sloshing and ship motion Ocean Engineering 39 (2012) 113
Mun Keun Ha, Deog Jin Ha and Dong Hyun: Challenges and New Technologies for Worlds Largest Floating
LNG, Proceedings of the 2013 ISOPE conference.
Myung Hyun Kim, Min Sung Chun, Wha Soo Kim, Jang Ho Yoon, Min Soo Kim, Hang Sub Urm, Byung Jae
Noh, Yong Suk Suh, Yoon Pyo Kil and Jae Myung Lee: Cryogenic Fatigue Strength Assessment for MARKIII Insulation System of LNG Carriers, J. Offshore Mech. Arct. Eng. 133(4), 041401 (Apr 12, 2011)
Myung-Jae Song Yang-Jun Ahn, Sang-Yeob Kim and Yonghwan Kim Jun-Hyung Jung and Jeong-Hwan Kim:
Effects of Sloshing Loads on Fatigue Strength of Independent Type B LNG Tank International Journal of
Offshore and Polar Engineering (ISSN 1053-5381) Vol. 23, No. 2, June 2013, pp. 112119
Myung-Jae Song, Yang-Jun Ahn, Sang-Yeob Kim, Yonghwan Kim, Jun-Hyung Jung, Jeong-Hwan Kim:
Effects of Sloshing Loads on Fatigue Strength of Independent Type B LNG Tank, International Journal of
Offshore and Polar Engineering, Volume 23, Number 2, 2013
NI 554 (2011). Design Sloshing Loads for LNG Membrane Tanks. Guidance note NI 554 DT R00 E, Bureau
Veritas.
NI 564 (2011). Strength Assessment of LNG Membrane Tanks under Sloshing Loads. Guidance note NI 564
DT R00 E, Bureau Veritas.
Nikita Tryaskin, Igor Tkachenko, Alexander Dukarskiy, Vladimir Yakimov, Vladimir Tryaskin and Dimitriy
Kiselev: Simulation of the Sloshing in the Prismatic Gas Tank after Impact Interaction of the Vessel with Ice
Barrier, Proceedings of the 2012 ISOPE conference.
Odd M. Faltinsen and Alexander N. Timokha: Multimodal analysis of weakly nonlinear sloshing in a spherical
tank Multimodal analysis of weakly nonlinear sloshing in a spherical tank, Journal of Fluid Mechanics/
Volume 719/ March 2013, pp. 129164
Odd M. Faltinsen and Alexander N. Timokha: Analytically approximate natural sloshing modes for a spherical
tank shape, Journal of Fluid Mechanics Volume 703 July 2012, pp. 391401
Odd M. Faltinsen and Alexander N. Timokha: On sloshing modes in a circular tank On sloshing modes in a
circular tank, Journal of Fluid Mechanics/Volume 695/March 2012, pp. 467477
616
Olivier Benyessaad, Diane Ruf, Karina Forte De Souza: Risk and Safety Studies for New LNG Floating Units
Projects, Proceedings of the 2013 ISOPE conference.
Pasquier, R., Berthon, C.-F. (2012). Model Scale Test vs. Full Scale Measurement: Findings from the Full Scale
Measurement of Sloshing Project. Proceedings of the Twenty-second (2012) International Offshore and Polar
Engineering Conference, Rhodes, Greece, June 1722, 2012.
Peder A. Tyvand and Touvia Miloh: Incompressible impulsive sloshing, Journal of Fluid Mechanics/Volume
708/ October 2012, pp. 279302
Pedersen, T. P.: Review and application of ship collision and grounding analysis procedures, Marine Structures
23 (2010) 241262
Per Hogstrm, Jonas W. Ringsberg: An extensive study of a ships survivability after collision A parameter
study of material characteristics, non-linear FEA and damage stability analyses, Marine Structures 27 (2012)
128
Ping-jian Ming, Wen-yang Duan: Numerical simulation of sloshing in rectangular tank with VOF based on unstructured grids Journal of Hydrodynamics, Ser. B, Volume 22, Issue 6, December 2010, Pages 856864
Pirker, S., A. Aigner, G. Wimmer: Experimental and numerical investigation of sloshing resonance phenomena
in a spring-mounted rectangular tank Chemical Engineering Science, Volume 68, Issue 1, 22 January 2012,
Pages 143150
Reza, M, Karimi, C. Kosinski, L. Brosset: Comparison of Sloshing Model Test Results at Scales 1:10 and 1:40,
Proceedings of the 2013 ISOPE conference.
Romain Pasquier and Christian-Frdric Berthon: Findings from the Full Scale Measurement of Sloshing
Project, Proceedings of the 2011 ISOPE conference.
Sang-Gab Lee and Hong-Anh Nguyen: LNGC Collision Response Analysis with Iceberg Considering Surrounding Seawater, Proceedings of the 2013 ISOPE conference.
Sang-Gab Lee and Tuo Zhao: Structural Safety Assessment of LNGC CCS under Iceberg Collision using FSI
Analysis Technique, Proceedings of the 2013 ISOPE conference.
Sang-Gab Lee, Jae-Hyung Nam, Jin-Kyung Kim, Tuo Zhao and Hong-Anh Nguyen: Structural Safety Assessment of Ship Collision Using FSI Analysis Technique, Proceedings of the 2013 ISOPE conference.
Sang-Gab Lee, Jin-Kyung Kim, Hong-Anh Nguyen and Jae-Hyung Nam: Structural Safety Assessment of
LNGC MARK III Membrane type CCS under Sloshing Impact Loading, Proceedings of the 2013 ISOPE conference.
Sang-Woo Lee, Young-Woo Kwen, Young-Min Kim and Man-Soo Kim: Introduction of Structural Design and
Construction of FLNG, Proceedings of the 2013 ISOPE conference.
Sang-Yeob Kim, Yangjun Ahn, Kyong-Hwan Kim and Yonghwan Kim: Experimental Studies on Sloshing in a
STX Independence Type-B Tank, Proceedings of the 2013 ISOPE conference.
Se Yun Hwang(1), Jang Hyun Lee(2) and Sung Chan Kim(3): Simplified Impinging Jet Model for Practical
Sloshing Assessment of LNG Cargo Containment, Proceedings of the 2012 ISOPE conference.
Sebastian Schreier and Bernhard Mehl: Experimental Investigation of 3D Sloshing Effects in Thin Rectangular
Tanks, Proceedings of the 2012 ISOPE conference.
Se-min Jeong, Sung-Chul Hwang, Jung-Woo Nam Jong-Chun Park: Numerical Simulation of Impact Loads by
Sloshing in a 3-D Rectangular Tank Using Eulerian and Lagrangian Approaches, Proceedings of the 2013
ISOPE conference.
Se-min Jeong1, Sung-Chul Hwang1, Jung-Woo Nam1 and Jong-Chun Park: Numerical Simulation of Impact
Loads by Sloshing in a 3-D Rectangular Tank Using Eulerian and Lagrangian Approaches, Proceedings of the
2013 ISOPE conference.
Seok Kyu Cho, Hong Gun Sung, Bo Woo Nam and Hang Shoon Choi: Numerical Study on the Motion and
Drift Forces of Side-by-Side Moored FSRU and LNGC Containing Sloshing, Proceedings of the 2013 ISOPE
conference.
Seok Kyu Cho, Hong Gun Sung, Sa Young Hong and Seok Won Hong, Yong Su Kim, Moon Keun Ha, Young
Dal Choi, Byoung Suk Yu, Rae Dae Jang: Experimental Study on the Side-by-Side Offloading Operation of
FSRU and LNGC, Proceedings of the 2013 ISOPE conference.
Seung-Hee Lee, Young-GillLee, Kwang-LeolJeong: Numerical simulation of three-dimensional sloshing
phenomena using a finite difference method with marker-density scheme Ocean Engineering 38 (2011) 206
225
Seyyed M. Hasheminejad, M. M. Mohammadi: Effect of anti-slosh baffles on free liquid oscillations in partiall
filled horizontal circular tanks, Ocean Engineering 38 (2011) 4962
Sharen Cummins Murray Rudman Krish Thiagarajan: The Effect Of Pressure Solution In Sph Simulations Of
Sloshing Flow, Proceedings of the ASME 2011 International Conference on Ocean, Offshore and Arctic Engineering,
Souto-Iglesias, A, E. Botia-Vera, G. Bulian: Repeatability and Two-Dimensionality of Model Scale Sloshing
Impacts, Proceedings of the 2013 ISOPE conference.
Sung Woong Choi. Sung Ha Kim, Jeung U Rho and Woo Il Lee: Analysis of Flow and Thermal Characteristics
of Leaked LNG in Glass Wool for the Mark 3 CCS, Proceedings of the 2013 ISOPE conference.
Sung Woong Choia, Jeong U. Roha, Moo Sun Kimc, Woo Il Leea: Analysis of two main LNG CCS (cargo containment system) insulation boxes for leakage safety using experimentally defined thermal properties, Applied
Ocean Research Volume 37, August 2012, Pages 7289
Sven Hoog, Joachim Berger, Johannes Myland: The Mooring Bay Concept For Long Loading In Harsh And Ice
Conditions, Proceedings of the ASME 2031 International Conference on Ocean, Offshore and Arctic
Engineering
617
Takashi Ikeda, Raouf A. Ibrahim, Yuji Harata and Tasuku Kuriyama: Nonlinear liquid sloshing in a square tank
subjected to obliquely horizontal excitation Nonlinear liquid sloshing in a square tank subjected to obliquely
horizontal excitation Journal of Fluid Mechanics/Volume 700/June 2012, pp. 304328
Thiagarajan, K.P., D. Rakshit, N. Repalle: The airwater sloshing problem: Fundamental analysis and parametric
studies on excitation and fill levels Ocean Engineering 38 (2011) 498508
Thibaut Loysel, Eric Gervaise, Sylvestre Moreau and Laurent Brosset: Results of the 20122013 Sloshing Model
Test Benchmark, Proceedings of the 2013 ISOPE conference.
Thibaut Loysel, Sabrina Chollet, Eric Gervaise, Laurent Brosset and Pierre-Emmanuel De Seze: Results of the
First Sloshing Model Test Benchmark, Proceedings of the 2013 ISOPE conference.
Turner, M. R. and T. J. Bridges: Nonlinear energy transfer between fluid sloshing and vessel motion Nonlinear
energy transfer between fluid sloshing and vessel motion, Journal of Fluid Mechanics/Volume 719/March
2013, pp. 606636
Tzabiras, G., G.M. Katsaounis, V. Papakonstantinou and S.A. Mavrakos: Viscous Flow Calculations around
Transverse Sections of a Floating LNG Storage Terminal in Heave and Roll Motions, Proceedings of the 2013
ISOPE conference.
Vernengo, Giuliano. (2014). Ship synthesis model for Preliminary design of fleet of compressed natural gas
carriers, Elsevier.
Wang X, Arai M: A numerical study on coupled sloshing and ship motions of a liquefied natural gas carrier in
regular and irregular waves, Proceedings of the Institution of Mechanical Engineers, Part M: Journal of Engineering for the Maritime Environment (2014)
Wang, B., Y. Shin and X. Wang: Reliability-Based Sloshing Assessment of Containment Systems in LNGCs and
FLNGs, Proceedings of the 2013 ISOPE conference.
Wang, S., T. Notteboom (2014), http://www.porttechnology.org/ images/uploads/technical _papers/LNG_LR.pdf.
Wang, Xiuli. (2009). Advanced Natural Gas Engineering, University of Houston, Gulf Publishing Company,
Houston, Texas.
Wenhua Zhao, Jianmin Yang, Zhiqiang Hu, Bin Xie: Hydrodynamics of an FLNG system in tandem offloading
operation, Ocean Engineering, Volume 57, 1 January 2013, Pages 150162
Wenhua Zhao Jianmin Yang Zhiqiang Hu and Tao Peng: Numerical And Experimental Investigation On Hydrodynamic Characteristics Of Flng With Account Of The Inner-Tank Sloshing, Proceedings of the ASME 2012
International Conference on Ocean, Offshore and Arctic Engineering
Wen-hua ZHAO, Jian-min YAN, Zhi-qiang HU, Long-fei XIAO. Experimental investigation of effects of innertank sloshing on hydrodynamics of an flng system, Journal of Hydrodynamics, Ser. B Volume 24, Issue 1,
February 2012, Pages 107115
Wenhua Zhao, Jianmin Yang, Zhiqiang Hu, Longfei Xiaoa, Longbin Tao: Hydrodynamics of a 2D vessel including internal sloshing flows, Ocean Engineering Volume 84, 1 July 2014, Pages 4553
Wenhua Zhao, JianminYang, ZhiqiangHu, LongbinTao (2014). Coupled analysis of nonlinear sloshing and ship
motions. Applied Ocean Research 47 (2014) 8597.
Wilde, J de and Dekker, J. (2013). Review of Hydrodynamic and Nautical Studies for Offshore LNG Operation.
OMAE 2013, Paper Number 11366.
Wisch, D. J. and Mc Master, F. J. (2009). Floating System Integrity Management: Developing a Process. OTC,
Paper number 20184.
Wisch, D. J. and Mc Master, F. J. (2009). Floating System Integrity Management: Developing a Process. OTC,
Paper number 20184.
Xin Wang and Makoto Arai: A Study on Coupling Effect Between Seakeeping and Sloshing for MembraneType LNG Carrier International Journal of Offshore and Polar Engineering (ISSN 1053-5381) Vol. 21, No. 4,
December 2011, pp. 256263
Yangjun Ahn, Kyong-Hwan Kim, Sang-Yeob Kim, Sang-Woo Lee, Yonghwan Kim and Jae-Hoon Lee (2013)
Experimental Study on the Effects of Pressure Sensors and Time Window in Violent Sloshing Pressure Measurement. Proceedings of the Twenty-third (2013) International Offshore and Polar Engineering Conference,
Yangjun Ahn, Sang-Yeob Kim, Kyong-Hwan Kim, Sang-Woo Lee, Yonghwan Kim and Jong-Jin Park: Study
on the Effect of Density Ratio of Liquid and Gas in Sloshing Experiment, Proceedings of the 2013 ISOPE
conference.
Yonghwan Kim and Sang-Yeob Kim Yangjun Ahn Kyong-Hwan Kim Sang-Eon Jeon, Yong-Suk Suh, Jong-Jin
Park and Seung-Myun Hwangbo: Model-Scale Sloshing Tests for an Anti-Sloshing Blanket System International Journal of Offshore and Polar Engineering (ISSN 1053-5381) Vol. 23, No. 4, December 2013, pp. 254
262
Young Dal Choi. Byoung Suk Yu, Rae Dae Jang: Experimental Study on the Side-by-Side Offloading Operation
of FSRU and LNGC, Proceedings of the 2013 ISOPE conferenc
Young MooJi a, YoungSupShin b, JunSangPark b,n, JaeMinHyun: Experiments on non-resonant sloshing in a
rectangular tank with large amplitude lateral oscillation Ocean Engineering 50 (2012) 1022
Yunho Kim, DongYoung Lee, Hong G. Sung and Hang S. Choi: A Study on the Effect of Sloshing on the Roll
Motion of a 2D Rectangular Box in Regular Waves, Proceedings of the 2012 ISOPE conference.
Yusong Cao, Fuwei Zhang and Stergios Liapis (2013). A Computer Code for Fast Simulations of Liquid Tank
Sloshing. Proceedings of the Twenty-third (2013) International Offshore and Polar Engineering Conference,
Yuxin Zhang and Decheng Wan: Apply MPS Method to Simulate Liquid Sloshing in LNG Tank, Proceedings of
the 2013 ISOPE conference.
618
Zhao, W., J. Yang, Z. Hu, B. Xie: Hydrodynamics of an FLNG system in tandem offloading operation, Ocean
Engineering, Volume 57, 1 January 2013, Pages 150162
Zhao, W. H., J.M. Yang, Z. Q.Hu, Y. F. Wei: Recent developments on the hydrodynamics of floating liquid
natural gas(FLNG) Ocean Engineering 38 (2011) 15551567
Zhaolong Yu a, Zhiqiang Hua, Jrgen Amdahl b, Yi Liu: Investigation on structural performance predictions of
double-bottom tankers during shoal grounding accidents, Marine Structures 33 (2013) 188213
Zhijun Wei, Qianjin Yue and Shilun Ruan: An Experimental Investigation of Liquid Sloshing Impact Load in a
Rectangular Tank, Proceedings of the 2013 ISOPE conference.

710 SEPTEMBER 2015
CASCAIS, PORTUGAL
VOLUME 2
COMMITTEE V.3
MATERIALS AND FABRICATION TECHNOLOGY

COMMITTEE MANDATE
The committee shall give an overview about new developments in the field of ship and offshore materials and
fabrication techniques with focus on trends which are highly relevant for practical applications in the industry
in the recent and coming years. Particular emphasis shall be given to the impact of welding and corrosion
protection techniques on structural performance, on the development of lighter structures and on computer
and IT technologies and tools, which are meant to link design and production tools and to support efficient
production.
COMMITTEE MEMBERS
Chairman:
J.-D. Caprace, Belgium

B. Mishra, USA
F. Pires, Brazil
F. Roland, Germany
I. Schipperen, Netherlands
J. Andric, Croatia
L. Li, China
N. Osawa, Japan
P. Lindstrom, Norway
R. Doig, Peru
H. Remes, Finland
M. H. Kim, South Korea
KEYWORDS
Lightweight, composites, bonding, welding, residual stress, distortions, line heating, fatigue, productivity,
corrosion, discrete event simulation, virtual and augmented reality.
620
ISSC committee V.3: MATERIALS AND FABRICATION TECHNOLOGY
CONTENTS
1.
INTRODUCTION
622
2.
GENERAL TRENDS
2.1
Developments in the Maritime Markets and their impact on the Trends in
Fabrication and Materials Technologies
2.1.1
Korea
2.1.2
Japan
2.1.3
China
2.1.4
Europe
2.1.5
Brazil
2.2
2.2.1
Korea
2.2.2
Japan
2.2.3
China
2.2.4
Europe
2.2.5
Brazil
2.2.6
USA
622
622
624
624
624
624
625
625
625
626
626
626
627
628
3.
STRUCTURAL MATERIALS
3.1
Metallic materials
3.1.1
Aluminium alloys
3.1.2
Titanium
3.1.3
Metal foam
3.1.4
Application of metals in low temperatures
3.2
3.2.1
Fire resistant materials
3.2.2
Bio-composites
3.2.3
Influence of sea water on non-metallic materials
3.2.4
Recycling and disposal
3.2.5
Application of non metallic materials at low temperatures
3.3
Hybrid materials
629
629
629
630
630
631
633
634
634
636
636
637
637
4.
JOINING AND FABRICATION TECHNOLOGY

4.1
4.1.1
Welding automation and recent developments in joining technologies
4.1.2
Underwater welding
4.1.3
Frictions stir welding of steel
4.2
4.2.1
Plate bending with line heating
4.2.2
Post-treatment of welded joints and plate edges
4.2.3
Hybrid structures and joints
4.3
4.3.1
Weld geometry and misalignments
4.3.2
Effect residual stress and distortions
4.3.3
Utilisation of high strength steel and thin plates
4.4
637
637
637
638
638
640
640
640
641
642
642
643
643
644
5.
CORROSION PROTECTION
5.1
Protection rules
5.2
Coating and paints
5.2.1
Epoxy-based coating systems
5.2.2
Zinc-rich paints
5.2.3
Thermal spraying and deposition
5.2.4
Antifouling (AF) coatings
5.2.5
Self Healing Coatings
5.2.6
Intelligent Coatings
5.2.7
Ice-breaker Coatings
5.3
Cathodic protection
5.4
5.5
Corrosion Monitoring
644
644
645
645
645
645
646
646
646
646
647
647
648

5.6

5.6.1
Visual Inspection of Welds:
5.6.2
Inspection for Delayed (Hydrogen Induced) Cracking
5.6.3
Methods of Inspection
5.6.4
Under film corrosion detection
621
648
648
648
649
649
6.
MANUFACTURING SIMULATION
6.1
Discrete Event Simulation and production optimization
6.1.1
Layout planning
6.1.2
Production planning
6.1.3
Outfitting and customization
6.1.4
Logistic simulations
6.2
Virtual and Augmented Reality
649
650
650
651
651
652
652
7.
WELDING SIMULATION
7.1
Computation Welding Mechanics
7.2
Arc Welding Simulation Methodologies
7.2.1
Sequentially coupled thermos-mechanical models
7.2.2
Thermo-mechanical staggered coupled
7.3
Heat source models
7.4
Material Models
7.5
Thermal- and Mechanical Boundary Conditions
7.6
Mesh size
7.7
7.8
7.9
Benchmark case
653
653
653
653
653
654
655
656
657
657
657
658
8.
659
REFERENCES
660
622
1.
INTRODUCTION
Due to the past crises, the shipbuilding and offshore industry has realised that new innovative designs and
design and production methods are necessary to decrease operational costs, production costs and emissions, while meeting the changing rules and regulations. This ISSC-V.3 report is discussing recent development in materials and fabrication technology applied to ship and offshore structures.
Chapter 2 focuses on worldwide trends in materials and fabrication methods. Developments in metallic
and non-metallic structural materials are dealt in Chapter 3. Advances in fabrication and joining technologies such as welding are increasing. Some main areas of applications and research in those areas are
described in Chapter 4. Innovative development about corrosion protection systems are presented in
Chapter 5 while Chapter 6 give an overview about the application of production simulation and virtual
reality to improve the production management of ship and offshore structures.
The ISSC-V.3 technical committee has performed a benchmark to define a Best Practice Guideline to
use Computational Welding Mechanics tools (CWM) in shipbuilding and offshore industry. To achieve
this objective various experimental welding tests have been performed in order to give a reference point.
Both the residual welding distortions and residual stresses have been compared between numerical simulations and welding experiments for a common T welded assembly used in the shipbuilding industry.
However, it has been decided to publish the results of this study in a separate document. Nevertheless,
Chapter 7 of this report presents the state of the art as well as the experimental test case that has been
analysed.
2.
2.1
GENERAL TRENDS
Developments in the Maritime Markets and their impact on the Trends in Fabrication
and Materials Technologies
With the exception of a decline in 2009 seaborne trade has seen a steady growth at around 4-6% per annum which is expected to last until 2016, (Hamers, 2013). In addition, the demand for passenger ships
and ferries remains stable at a lower level than before the crisis. New markets related to the exploitations
of the oceans (offshore oil and gas, offshore renewable energies and marine resources) develop continuously and call for new ships and marine structures.
Despite of slightly positive developments in market volumes, remaining transport over capacities have
led to historically low order intake for global shipbuilding at around 16 million CGT annually, which is
about 35% below the global capacity Figure 1.
Figure 1. World Commercial Shipbuilding Activity in Compensated Gross Tonnage (CGT). Source: SEA Europe,
based on IHS Fairplay database
623
However, this situation does not equally affect all market segments, (Hamers, 2013). In
particular, the following trends provide new market chances for the future:
The sharp increase of bunker costs (about 3 times increase since 2008) as well as upcoming environmental legislation forces ship owners to increase energy efficiency and reduce the environmental foot
print of their ships this leads to an increased demand for new green ships as well as to an increased market volume for retrofitting;
Fierce competition on the transport market requires the reduction of operational costs of ships taking
the entire life cycle into consideration. This as well as the development of new market segments has
increased the demand for specialized vessels with a comparatively high degree of outfitting, which
accounted for 75% of the total investments made in 2012 Figure 2;
The two previous trends also lead to increased demolition rates, with a market volume increasing by
3.5 times since 2008. This again results in decreased average age of the tonnage scrapped.
Figure 2. Global Investment in New buildings Market by Vessel Types. Source: SEA Europe, based on Clarkson
database
In the frame of those global trends, the shift of shipbuilding capacities away from Europe and the US
towards Asia and South America has continued during the last years. However, this does not equally apply to all market segments and a clear specialization can be observed in particular in high value added
niche markets, see Figure 3.
Typeofships
Crudeoiltanker
Bulkcarriers
Product&chemicaltankers
LNGandLPGcarriers
Containervessels
RoRoandothercargo
Cruise,Paxandferries
MegaYachts
Offshorevessels
Navalvessels
China
Korea
Japan
Europe
Brasil
USA
Figure 3. Simplified visualization of market foci in key regions of the world maritime industry, where means moderate and means important. Source: Center of Maritime Technology based on world order book end
of 2013 from Clarkson
624
The following paragraphs discuss some specific market strategies and related research and development trends in the key maritime regions of the world. The discussion focuses on selected research areas
with particular relevance to materials and fabrication and related structural reliability. Next sections intend to explain why different areas are putting the focus of their work on different subjects.
2.1.1
Korea
Korea has traditionally produced various types of ships and offshore structures, such as oil tankers, bulk
carriers, cargo carriers, container ships and LNG carriers. With the increasing demand for liquefied natural gas (LNG), the construction of advanced LNG carriers and floating LNG production vessels has become an important market segment. Recently, the shipbuilding industry in Korea is accounting for some
80% of the international order backlog for LNG carriers, and is expecting a further increase of this market.
This market focus has triggered a number of research initiatives in the field of innovative steel materials and insulation materials for extremely low temperatures. Major shipyards in Korea have carried out
research related to low nickel and high manganese steels to apply for the new LNG tanks, in a joint venture with POSCO and Nippon Steel & Sumitomo Metal Corporation (NSSMC).
2.1.2
Japan
The total order book (order intake) of Japanese shipyards (reported by Japan Ship Export Association)
in 2013 comprised around 26.4 million GT, including primarily bulk carriers (56.9%), LNG tankers
(8.1%) car carriers (4.6%) and container ship (3.4%). While this is a downward trend, as compared to
the previous years, the weakening of the Japanese Yen and the anticipation of an upturn in the market
resulted in a significant increase in orders during the second half of 2013, bringing the total order book
(order intake) close to 13 million GT, from 9-10 million GT during the last couple of years. However,
as the total annual new build capacity in Japan is approximately 16 million GT, there is still excess
capacity.
In these days, over 80% of all new build contracts are for bulk carriers. However, it is considered that
the future of Japanese shipbuilding does not lie in the mass production of bulkers, but in high-value tonnage and new advanced-design vessels capable of meeting environmental and fuel efficiency requirements. Many Japanese yards have recently revealed new fuel efficient and environmentally friendly designs, claiming a reduction in fuel consumption of more than 30% compared to traditional designs delivered only a few years ago.
Japanese yards are being encouraged to move into building more technically advanced and specialized
offshore units. Kawasaki Heavy Industries (KHI) and Japan Marine United (JMU) have established themselves in Brazil by acquiring significant shares in yards focusing on the local Brazilian offshore market,
and orders of specialized tonnage from foreign owners are increasing in Japanese yards.
2.1.3
China
According to China Association of the National Shipbuilding Industry (Cansi, 2014), China maintained
the leading position in global shipbuilding in 2013, in terms of deliveries (41.4% of total GT), order intake (47.9%) and total order book (45%). After a decline in 2012, new orders have been increased in
2013, but the economic benefit of shipbuilding in China still continues to decrease. Shipbuilding enterprises face difficulties in delivery of vessel and financing. According to statistics, the whole shipbuilding
industry realized profit 25.2 billion Yuan in 2013, a drop by 13.1 percent.
From January to August 2014 the hand-held orders in China increased by 34.1%. Incoming new orders
continued to grow as well as the export of ships and main business income of shipbuilding. Nonetheless,
the price of new ships fell under the influence of slow-down of global economy and low order shipping
market. Moreover, the economic efficiency of key shipbuilding yards was on decline, and the market
prospects in China remain stringent.
In response to that, measures to deal with the current shipbuilding situation were introduced. New ships
will be developed along energy saving and environmental protection, and the government will give support for research and development of new ships and new ocean engineering equipment.
2.1.4
Europe
European shipbuilders have concentrated their RDI efforts to maintain and strengthen their position in
high value added niche markets, such as cruise and passenger vessels, mega yachts and naval ships. Europe has a strong network of ship equipment suppliers, which are serving global markets. The close interaction between ship owners, shipyards and equipment suppliers provides the basis for the development
of new customer specific ship generations and a holistic assessment of life cycle cost efficiency and
625
environmental impact, which has become an integral part of most RDI activities. The need to increase
payload to weight ratio leads to the development on lighter structures, made of traditional and innovative
materials. Using the developments in material sciences, the use of smart and adaptive materials and structures to improve the ability of ships to adapt to a wide range of operational scenarios is a new trend observed during the last years. To improve cost efficiency in ship production, multi-material joining techniques as well as improved planning, resource management and logistics in outfitting along the supply
chain have remained another focus of research.
Many small and medium sized European shipyards have successfully engaged in the growing market
for repair, retrofit and conversion of ships, e.g. with energy saving devices. This has triggered research in
reverse engineering, design for recycling and retrofit and the organization of retrofit processes.
The political decision to move towards renewable energies such as offshore wind calls for new materials and efficient processes for the pre-fabrication, erection, maintenance, repair and retrofit and for dismantling of structures and ships for offshore wind parks. First research projects related to that have been
started.
Increased transport demands within European and with neighbouring regions in line with the political
pressure towards environmentally friendly means of transport has started to initiate some research on
innovative inland ship concepts and retrofit solutions for existing ships, although the focus of ongoing
research in this sector is currently more on traffic management, energy efficient and environmentally
friendly fuels and propulsion.
2.1.5
Brazil
The Brazilian shipbuilding industry is formed by three groups of shipyards, complemented by a huge
number of small and sometimes improvised facilities engaged in construction of small crafts.
The first group includes a few shipyards, constructed in the 1970s, capable to build ships up to Panamax size. These companies are mainly build oceangoing ships to coastal operation (most of them ordered
by the Brazilian state owned oil company Petrobras); offshore supply vessels and some minor offshore
works;
The second group corresponds to a number of yards dedicated to the offshore supply (OSV) market
green field plants or existing old fashioned facilities. Most of the OSV new buildings are ordered by private owners that have long term time charter with Petrobras;
The shipyards in the two first groups use conventional processes and show low productivity levels. The
engineering content in their products and processes is relatively low. The shipyards usually subcontract
both ship design and detailing drawings. There are a very small number of engineers in those companies
and the engagement in research and development to introduce innovative processes and solutions is very
limited.
The third group is composed by the yards engaged in offshore construction. This sector stands for a
very large share of the ship and offshore construction market in Brazil in terms of value.
The five principal shipyards belong to this group are Keppel Fels Brazil, Estaleiro Atlantico Sul, Jurong, Enseada and Ecovix. All of them are owned by foreign companies or have long term cooperation
with Japanese shipbuilders which are supposed to support the Brazilian shipyards through consultancy
services, providing software, procedures and standards, training people and undertaking technical and
managerial duties.
The order books of the main five shipyards consist of ships (tankers, FPSO, drill ships) and rigs owned
by Petrobras or its subsidiaries.
The main shipyards have large production capacity and are planned to operate at a higher technological
level than the usual in Brazilian industry, employing e.g. pre-erection of mega-blocks and intensive preoutfitting. However, the four new facilities are facing some difficulties due to the lack of technological
and managerial capability and scarcity of qualified workforce.
2.2
2.2.1
Korea
In Korea, R&D funds related to naval architecture and ocean engineering are mostly handled by the Ministry of Trade, Industry and Energy (Motie, 2014), (Keit, 2014) and the National Research Foundation
(NRF), (Nrf, 2014). These institutes are responsible for planning long term strategy and to develop
projects that best reflect the strategy and enable R&D to be conducted efficiently.
MOTIE has supported the following research programs associated with maritime transportation system:
Development of materials and drilling steel pipe for shale gas production (2014 ongoing; approximately 2.81 million US$);
626
Development of core technology for a LNG ship-to-ship bunkering shuttle (2013 ongoing; 0.7 million US$ per year);
Development of manufacturing technique for high manganese hot-rolled steel and pipe parts for LNG
storage and flow line (2013 ongoing; 1.4 million US$ per year);
Development of technologies for platform supply vessels under polar environments (2012 ongoing;
0.8 million US$ annually);
Since 2013, the development and assessment of fatigue and fracture performance of new high manganese steel is under investigation by HHI, SHI, DSME and POSCO (funded by MOTIE with 1.4 million
US$).
The Global Core Research Centre for Ships and Offshore Plants (GCRC-SOP) at Pusan National University, funded by NRF, carries out research in the field of ships and offshore plants in extreme environment such as in deep sea, Polar Regions, or extremely low temperature. R&D funding of material related
research by NRF has increased from 24.6 million US$ to 58.8 million US$ between 2007 and 2011.
2.2.2
Japan
The Japan Ship Technology Research Association (JSTRA) carries out two large projects subsidized by
Nippon Foundation related to manufacturing technologies:
Formation of strategic base for laser-arc hybrid welding technology in marine industry and development of innovative welding technique. In this project, a Center of Excellence (COE) of laser-arc hybrid
welding was established in Kyusyu University. New welding techniques such as one-side full penetration welding are studied.
Development of monitoring system for advanced construction management. Integration methods of
monitoring data to generate virtual job shop are being studied.
JSTRA also conducts Application of powered exoskeleton for shipbuilding work and Development
of alternative anticorrosion techniques for PSPC/WBT coating system. In the later project, inorganic
zinc-rich coating system and corrosion resistant steel for WBT are studied by ClassNK.
2.2.3
China
The National Natural Science Foundation (Nsfc, 2014) will support some research programmes for ship
and ocean engineering in China. Some recent projects focus on the following topics:
Prediction and control of wrinkling of rolling sectional material of ship under forming
Approach for multi-axial fatigue life assessment of welded structures based on critical plane theory
Design research of pre-deformation of composite propeller
Method and mechanism to welding deformation and crack resistance of marine thick plate with adaptive control of laser shock under welding process
Research of accumulative damage and total life coupling fatigue and corrosion
Interaction between fatigue crack growth and redistribution of welding residual stress of FPSO
structure.
2.2.4
Europe
The main source of R&D funding, at European level, are the Research and Development Framework Programmes (FP) of the European Commission. Research projects related to ship production and materials
have been primarily funded under the Priority Sustainable Surface Transport since 1998 with more widespread options to cover maritime aspects jointly with other sectors in the areas of key enabling technologies, such as ICT or materials.
The new European research and innovation funding scheme HORIZON 2020 has started in 2013 and
will last for seven years. In addition to research projects, the framework includes instruments which foster
innovation at higher Technology Readiness Levels (TRL) and specific funding instruments for small and
medium enterprises, aiming to promote the market uptake of research result. While global challenges, like
climate change and emission reduction have been dominating strategic goals in the previous programme,
HORIZON 2020 clearly adds industrial competitiveness as a strategic goal of European research policy.
Since FP7 (2010) research topics related to offshore operations have been partly funded under a joint
initiative The Oceans of Tomorrow. The funding available is increasing in this initiative and the
follow-up priority Blue Growth, but most of the budget is used for marine sciences rather than maritime
technologies.
627
Studies financed through the EU framework programmes Marpos (2011) and Mesa (2013) have analysed the topics of research co-financed by the FPs as well as the amount of funding for maritime research, see Figure 4.
Estimates have shown that this corresponds to about 900 researchers financed annually. This comes in
addition to national R&D programmes, which can be either generic or specifically dedicated to maritime
technologies and reach the same order of magnitude as European funding in some member states,
(Martec, 2014).
(a) Number of projects and funding from FP5 to H2020 regarding EU research transport topic
(b) Number of topics and funding for Oceans of Tomorrow (FP7) and Blue Growth (H2020)
Figure 4. Research Funding under European Framework Programmes for Maritime Research
Parts of the national research funds are used for trans-national projects among a limited number of participants. Considering additional regional development funds for R&D infrastructure and networks across
industry sectors and private investments, the strong involvement of the European maritime industry in
research and development becomes evident, although this engagement is not equally spread across companies, sectors or regions.
All in all, long term strategic research and development without any doubt has produced significant results in the field of production technologies and materials in the maritime sector, the introduction of low
distortion laser welding, sophisticated planning and logistics for highly complex outfitting intensive ships
or the application of advanced design methods being only some example for that.
More importantly perhaps is the intense R&D network between European companies, research centres
and academia which has developed during the projects but stretches far beyond research into rule making
and commercial activities. This network increasingly also involves smaller and medium sized companies
and less developed regions of the European Union and is probably unique in the world at least in the maritime industry.
2.2.5
Brazil
The Brazilian shipbuilding industry has never had much R&D involvement and at present seems to rely
on technology transfer from abroad. However, some research institutions in Brazil have recognized expertise in the field of naval architecture and ocean engineering:
The Graduate School of Engineering of Rio de Janeiro Federal University COPPE/UFRJ,
Polytechnic School of So Paulo University, Institute for Technological Research IPT,
Petrobras Research Centre CENPES.
628
There are also some other universities developing research activities in related fields. On the other
hand, a recent project, initiated by a group of scientific, professional and entrepreneurial associations,
aims at creating a research institution dedicated to shipbuilding technologies. The creation of this institute
was not accomplished until now, but has sound prospects of being created in the short term.
The main research programmes on Materials and Fabrication of Ships and Offshore Structures are
developed at COPPE/UFRJ, and have been funded by Petrobras and Ministry of Science and Technology
(Shipbuilding and Water Transportation Research Fund):
Shipbuilding Processes Simulation and Optimization and

Distortion and Residual Stress Evaluation in Ship Construction
The main source of R&D funds related to shipbuilding in Brazil is the Shipbuilding and Water Transportation Research Fund, managed by the Ministry of Science and Technology. This Fund has invested a
significant amount of money in R&D projects proposed by research institutions, in the last decade. Nevertheless, due to the lack of engagement of the shipbuilding or engineering companies, there is no relevant
log of projects on the field of materials and fabrication of ships and offshore structures.
2.2.6
USA
Research & Development expenditures in US are the highest per capita in the world, but represents only
2.8% of the GDP. Majority of the support comes from the industry at 308 billion US$ while the federal
government provides approx. $ 36 billion. The industrial support translates to an average of 3.5% of revenues, (Battelle, 2014).
Sectors to receive particular attention are technology platforms like robotics, high-performance computing, social media, software, cost-effective energy sources and nano-biotechnology.
US expects to spend approx. 12.5 billion US$ on aerospace, defence and security sectors, while the
energy sector will receive 7.3 billion US$ a significant part of this funding is related to materials engineering. R&D on chemistry and advanced materials will be supported at 12.2 billion US$.
In relation to the maritime sector, the American Bureau of Shipping (ABS) and the National Science
Foundation (NSF) fund fundamental research in several areas that involve materials and fabrication technologies pertaining to ship structures. The prominent research programs are being conducted through two
main programs at NSF:
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
Materials Processing & Manufacturing

Deformation Mechanics and Microstructure Evolution during Micro Forming of Metals;
Towards Room Temperature Formability in Magnesium Alloys;
Damage Tolerant 3D Periodic Interpenetrating Phase Composites with Enhanced Mechanical Performance - Design, Fabrication, Analysis and Testing;
Structure-Property-Processing Correlations in Ice template Materials;
Fundamentals of Bonding in Kinetic Consolidation Processes;
Fundamental Studies on Ultrasonic Vibration Assisted Laser Surface Modification;
Friction Stir Process for Joining Dissimilar Metals;
Enhance the Formability of Advanced Steels and Aluminium Alloys;
Characterization, Design and Modelling of Shape Memory Composites;
Mechanical Behaviour, Quasi-Static and Dynamic Crushing of Cellular Materials;
High Fidelity Numerical Investigations of Tailored Magnetic Fields for Defect Reduction in Continuous Casting of Steel;
Novel Manufacturing of Bio-inspired Metal Matrix Composites by Semisolid Forming-Joining.
Structural Materials & Mechanics
Design and Development of Fire-Resistant Ferritic Steels for Structural Applications;
Enhancing the Life Cycle of Plastic Pipes Through Nano-reinforcement;
Ultra-High Performance Fibre Reinforced Concrete Structures;
Advanced Interpenetrating Networks for Structural Applications.
In addition, ABS has established many research programs for marine and offshore structures in collaboration with several international groups.
3.
3.1
629
STRUCTURAL MATERIALS
Metallic materials
The lightweight metals in shipbuilding industry, such as aluminium allows and titanium, have been traditional used for hull constructions of high speed vessel, superstructures and outfitting. Due to increase
demand for evaluation of total life cycle costs of the ship, compare to just acquisition costs, the new opportunity for increased usage of lightweight metals have been investigated. Low temperature applications,
such as LNG storage, were already studied in the 70s, however recently attention focused into fatigue
behaviour and fracture toughness at low temperatures as well as the use of new materials.
3.1.1
Aluminium alloys
The review of the benefits and cost impact of aluminium alloys for naval ship structure is given in Lamb
et al. (2011). The traditionally excepted operational and performance advantages/disadvantages of aluminium are given in Table 1:
Table 1. Operational and performance advantages/disadvantage of aluminium (Lamb et al., 2011)

Advantages
High strength- to-weight ratio
Density one-third that of steel
Excellent corrosion resistance in marine environments
Weldability
Ease of forming, bending, and machining
Availability and diversity of functional semifinished products
Similarity of structural details and building
approaches
High recycling value
Environmental compatibility
Nonmagnetic
Disadvantages
Higher raw material cost versus steel
Lower stiffness (one-third that of steel)
Lower melting temperature than steel requiring additional insulation for fire protection
Less industry experience with aluminium
Shortage of trained aluminium welders
Smaller number of shipyards that have facilities for production with aluminium
Limited number of shipyards with facilities
and experience to repair aluminium ships.
The paper by Lamb et al. (2011) presents a comparative analysis of the acquisition cost of aluminium
and steel ship equivalent designs. The life-cycle cost advantage of aluminium ships are presented, as
well as advances in aluminium materials, design approaches, and manufacturing methods that are improving the cost and reliability of these ships. This paper illustrates that even though the cost of the
equivalent aluminium ship structure is 40% more than the steel structure, the equivalent aluminium
naval ship can be built within just 7.5% of the acquisition price of the steel ship. This is possible because of the cascading benefits of the aluminium ships significantly lighter weight. Advances in aluminium technology (material and fabrication) and new facilities in the shipyards for aluminium production are further improving the acquisition cost of aluminium ship. From a total life cycle cost perspective, aluminium ships enjoy a clear advantage over steel ships. Based on the findings presented in the
paper, it is suggested that the US Navy should consider broadening its use of aluminium ships. As also
stated in the paper, the materials section, close and constant collaboration between material supplier
and customer is essential to materials and systems breakthroughs. By understanding the Navys and
industries design and system performance requirements, the next generation of marine alloys and advanced aluminium designs can be developed.
As maritime industry increased their use and constructions of high speed vessel made of aluminium
a practical reference of key design consideration is needed for rational design of those vessels. The
(SSC-464, 2012) project Design and Detailing for High Speed Aluminium Vessels Design Guide and
Training provides a good overview of aluminium alloys and manufacturing and construction techniques that should be considered during the vessel design and construction processes. Specifically the
guide describes the wrought and cast alloy and temper designation systems; the rolling, casting, extruding, and machining processes, riveting, bolting, welding and adhesives as joining techniques; general
and high speed loading; the ultimate strength; and fatigue, fracture, fire and corrosion as specific considerations for aluminium vessels. Guidelines as these enable the extended use of aluminium in the
maritime sector.
630
3.1.2
Titanium
As stated in the past ISSC-V.3 reports (2003, 2006, 2009) titanium is the best engineering material for
marine environment, but its application is still limited mainly to pumps, piping and sea-water-cooled
heat exchangers and other specific parts mainly in Navy vessels. Mountford Jr. and Scaturro (2010)
estimated that 3000 to 4000 tons of titanium tubing and pipe are being used on offshore oil platforms in
fire-main, service water, and other systems solving seawater corrosion and erosion problems. Structural
application of titanium is still limited to specific structural parts in navy vessel (masts, radar structure,
cargo, compartment and emergency doors, torpedo bay doors, vertical launch systems, etc.). They presented main attributes, benefits, use and application of titanium in the marine market. All of the explained attributes and benefits suggest strong reasons for evaluating the design and greater use of titanium in systems for eliminating problems associated with seawater and marine environments, to improve efficiencies, to increase payloads and survivability, to reduce weight, maintenance, labour, and
fuel consumption, and to improve survivability and finally to decrease life-cycle costs. Dong et al.
(2013b) investigated titanium and its alloys as ship hull materials for reducing life-cycle cost in highspeed sea transportation. Besides the increased interest for its application significant challenges remain
in structural design and production. This is in large part due to the fact that titanium and its alloys
represent a drastic change in material behaviour from traditional ship hull materials such as steel or
even aluminium alloys. On one hand, their high strength to weight ratio can offer a significant weight
reduction in hull design. On the other, some of the today's ship design rules, such as proportionality
requirements for structural forms, prevent structural designers from taking a full advantage of this class
of new ship hull materials. Additionally, a better understanding of some of the material characteristics
and their impact on manufacturability in ship construction becomes important in order to devise effective build strategies from piece part fabrication to structural assembly. In this paper, some of the design
and production challenges are highlighted based on a comprehensive investigation into the manufacturability and structural performance of a titanium mid-ship section. A math-based approach using advanced computational structural modelling and process simulation is described for addressing some of
these challenges. Solutions for mitigating some of the design and construction concerns uniquely associated with titanium ship hull have been presented along with experimental validation results ranging
from process development to fabrication trials on a series of selected component examples. These examples include a quantitative evaluation of weld sizing requirements for achieving static and fatigue
strength requirements and panel welding distortion control. Finally, implications of authors findings
on building titanium ship hulls in today's shipyard environment and future areas of research and development have been discussed. It is advocated that welding-fabricated structural forms can now be
viewed as a viable alternative to expensive extrusions which have been shown to be cost-prohibitive for
ship hull applications.
3.1.3
Metal foam
Metal foam cores in metal sandwiches provide a light alternative to solid steel. It minimizes weight,
while maximizing stiffness. Next to that also the damping and energy dissipation increases. Smith et al
(Smith 2012) provide a good overview of the state of the art of steel foam materials. Although the paper focuses on applications in civil engineering, several applications are named that can also be beneficial to the maritime industry, e.g. the use of metal foams for crash applications or in floor and wall
systems. The paper provides an overview of manufacturing techniques including pros and cons. The
mechanical properties of the material are lower than of normal steel of course, with e.g. young modulus, depending on the manufacturing technique about 1/1000th to 1/20th of standard steel. Some of the
materials do have a compressive strength that is in the same order of magnitude of solid steel. Standards are being developed, e.g. Japanese and ISO for the testing of these materials. The article concludes with an overview of analyses models. Jasion et al. (2012) and Szyniszewski et al. (2012) both
provide analytical formulation for the determination of the buckling strength of these types of materials. Jasion focuses on the critical loads in beams and circular plates with aluminium facings and aluminium foam core. Due to the use of metal foam, shear effects in the core cannot be disregarded. When
including these shear loads in the analytical model, Jasion finds analytical results that are within 6% of
the finite element analyses and experiments. Szyniszewski provides a design method for the in plane
compressive strength of steel sandwich panels with steel facings and a steel foam core. The formulation
show that foaming the middle 30% of a solid steel plate leads to optimal strength gains for slender
plates, which can be in excess of 200% of the strength of a solid steel sheet with the same mass
(Figure 5).
Different aluminium sandwich structures have been further developed and evaluated in Crupi et al.,
(2013). The aim of this paper was the comparison of static and low-velocity impact response of two
aluminium sandwich typologies: foam and honeycomb sandwiches. Quasi-static indentation tests showed
631
Figure 5. Strength normalised to solid plate of the same weight as function of the initial width to thickness ratio
that the indentation resistance depends on the nose geometry and is strongly influenced by the cell diameter and by the skin core adhesion for the honeycomb and aluminium foam sandwich panels, respectively. The static bending tests, performed at different support span distances on sandwich panels with the
same nominal size, produced two different collapse modes (mode I more bending like and mode II a more
shear like failure mode) and simplified theoretical models, using plastic hinge representations, were applied to explain the observed collapse modes (Figure 6). The results with these theoretical models were in
good agreement with the experiments. The capacity of energy dissipation under bending loading is affected by the collapse mechanism and also by the face-core bonding and the cell size for aluminium foam
sandwiches (AFS) and honeycomb panels, respectively. A series of low-velocity impact tests showed that
the aluminium honeycomb sandwiches fail due to the buckling of the cells and is strongly influenced by
the cell size, whereas the aluminium foam sandwiches collapsed for the foam crushing and their energy
absorbing capacity depends by the foam quality. The experiments clearly show that aluminium sandwiches are lightweight and can provide good energy dissipation properties. From the literature studied it can
be concluded that metal foamed materials could be considered as a lightweight alternative for solid metal
parts.
Figure 6. Collapse modes of investigated honeycomb and aluminium foam sandwiches obtained by static
bending tests (Crupi et al., 2013)
3.1.4
Application of metals in low temperatures
Recently, the demand for liquefied natural gas (LNG) continues to increase due to the rise of environmental issues and the significant increase in crude oil price. In addition, the strict air emissions legislation by
International Maritime Organization (IMO) and other local air quality controls led to the attention of vessel owners toward the use of LNG as a fuel. These events have resulted in increasing construction of LNG
storage tanks such as LNGC (Liquefied Natural Gas Carrier), LNG FPSO (Liquefied Natural Gas Floating Production Storage and Offloading), FLNG (Floating Liquefied Natural Gas) and LNG bunkering
632
system. Then, the construction of vessels and storage tanks leads to an increased demand for materials
with specified low temperature properties.
A number of materials are available that are engineered specifically for service at a temperature of 163C and lower. Initial studies and measurements of low temperature material properties already date
from the 70s. However, past research did not include test results of fatigue and brittle fracture resistance
at cryogenic environment. The weldability and the ease of fabrication are also important factors for selecting the optimum material. In Korea, extensive studies have been performed to evaluate the fatigue
performance for various welded joints and the fracture toughness in weld metal and Heat Affect Zone
(HAZ) at cryogenic temperature. In particular, there have been few JIP (Joint Industry Project) to evaluate
cryogenic material properties. This paragraph provides a brief overview of JIP for fatigue and fracture
performance assessments of low temperature materials.
The LNG storage tanks are exposed to cryogenic temperatures as low as -163C, and low temperature
alloys are required because their cryogenic mechanical properties exhibit superior performance such as,
high strength, ductility and toughness at low temperatures. In this regard, 9% nickel alloy steel, stainless
steel, Invar alloy and aluminium are commonly employed for low temperature applications as they possess good weldability and toughness in addition to the benefit of relatively low cost of construction. Recently, low nickel alloy steels and high manganese steel have been developed in order to replace existing
commonly used low temperature materials due to its cost effectiveness. The most common low temperature materials and the newly developed materials are as followings:
9% nickel alloy steel

Stainless steel (SUS 304L, SUS 316L)
Invar alloy
Aluminium alloy (Al 5083)
Low nickel alloy steels (7%, 5%, 3.5% nickel steels)
High manganese steel
3.1.4.1 Low nickel alloy steel

One of the most common materials for LNG storage systems has been 9% nickel alloy steel over the last
50 years because it provides excellent mechanical properties at cryogenic temperatures. It has been used
mainly for the inner shell of LNG storage tanks as a ferritic cryogenic material. Brittle fracture properties,
which are closely related to the safety of structures, of 9% nickel alloy steel have been extensively investigated by many researches. In recent years, the increase of nickel price led to the development of low
nickel alloy steels to cope with the risk of sudden increase of the 9% nickel steel price. In this respect,
low nickel alloy steel has been developed for replacing 9% nickel alloy steel with the motivation of cost
effectiveness. Hyundai Heavy Industry (HHI), Pusan National University (PNU) and Nippon Steel &
Sumitomo Metal Corporation (NSSMC) (2009) studied material properties of 9% nickel steel for the application of IMO Type-B LNG Cargo Containment System (CCS). They performed FCGR (Fatigue
Crack Growth Rate), fatigue and fracture toughness tests for 9% nickel steel. In 2006, Korea Gas Corporation (KOGAS) built the largest above-ground LNG storage tank with a gross capacity of 200,000m3 at
Tongyoung and Pyeongtaek LNG receiving terminals in South Korea. The LNG storage tank was made
of 9% nickel alloy steel inner tank, pre-stressed concrete outer tank, suspended ceiling deck, secondary
bottom and corner protection system, and bottom heating system. Furuya et al. (2011) investigated material properties and fracture toughness of 6% nickel alloy steel. As the fracture performance evaluation
including of ESSO test and the wide plate tensile test, it was demonstrated that 6% Ni steel has good characteristics regarding brittle fracture initiation and propagation in base metal plates and welded joints. In
addition, ethane has the boiling point at temperature of -87C, shipyards have a considerable interest for
employing 3.5% nickel alloy steel in replace of 9% nickel steel or stainless steel in order to reduce the
cost of weld materials. In this respect, HHI and PNU are planning to examine the tensile strength, fatigue
and fracture performance of 3.5% nickel alloy steel.
3.1.4.2 Aluminium and stainless steel

Kim et al. (2014) performed an experimental study on the fatigue performance of cryogenic metallic
materials for IMO Type-B tank. This study was supported by DSME, HHI, SHI, ABS, DNV, Lloyd and
POSCO group. In particular, the project investigated tensile property, fatigue performance and fracture
toughness assessment of three materials SUS 304L, 9% nickel steel and Al 5083. These materials are
being considered as possible material candidates for IMO Type-B LNG carriers. In order to ensure the
structural reliability of IMO Type-B tank, it is very important to evaluate the material properties at
cryogenic environment. Al 5083 is considered to be the most suitable material for IMO type B LNG
633
tank compared to SUS304 and 9% Ni steel. This is because the material characteristics of the Al 5083
at cryogenic temperatures turned out to be comparable to other materials. SUS304 is also considered to
be a good choice once the potential problem of relatively lower CTOD value is overcome by suitable
design.
3.1.4.3 High manganese alloy steel

Many industries are considering the use of high manganese alloy steel that can replace the commonly
employed low temperature materials because of the high cost of nickel alloy. In this regard, the material
properties and fatigue strength assessments of high manganese alloy steel are carried out by Lee et al.
(2014b). The fatigue strength of the high manganese alloy steel at room temperature is examined and
appears to be as good as other low temperature materials. However, the fatigue strength at cryogenic temperature is slightly less than that of other materials. POSCO and KAIST (2014) have developed a high
manganese steel grid-pattern LNG storage tank that can dramatically increase the storage capacity of
LNG tanks. This new technology employing high manganese steel can enhance the storage capacity up to
20,000 cubic meters from the existing 1,000 cubic meters. High manganese steel can be welded better
than stainless steel, making it easier to manufacture storage tanks.
3.1.4.4 Invar alloy

Invar alloy (Fe-Ni 36%) possesses low thermal expansion, moderately high strength and good toughness
at low temperature. These properties and good weldability are usefully employed in the many cryogenic
applications. In particular, Invar alloy is widely used in the primary and secondary barriers of membrane
type (No. 96 type) of LNGC. No. 96 type LNG tank is designed with two identical metal membranes
forming two independent thin plates of Invar alloy in order to minimize the possibility of LNG leakage in
the second barrier. Han et al. (1994) investigated fatigue strength of overlap welded joint for Invar alloy.
This study performed fatigue test for Invar/Invar overlap welded joint and Invar/stainless steel overlap
welded joint. Oh et al. (2014) studied fatigue performance of various Invar weld joint such as raised edge
specimen and overlap welded joint specimen. In addition, this research proposed the fatigue design curve
of Invar alloy based on the effective notch stress approach.
3.2
After a long use of composites in the sporting and leisure ships and an increased application in navy vessels according to Hellbratt (2008) the time is now ready for light weight composite materials to enter the
merchant shipbuilding. A study was done in which the life cycle costs of a 127 m long ferry were assessed. The ferry was designed in steel, aluminium and composite. A full life cycle cost assessment was
performed to show the assets of composites. The life cycle cost analysis included the effects of weight
savings of aluminium and composites, manufacturing costs, maintenance and operational costs including
disposal. The analysis substantiates that although composites are more expensive in production, the life
cycle costs are considerably lower than the metal alternatives. In this particular case the composite ferry
became less expensive than the steel version after only 2 years of operation. A breakeven point with aluminium was reached after 10 years of operations. The authors emphasize that although most advantages
will be seen in high speed vessels, also slower vessels can benefit from the use of composite materials for
example via lower weight superstructures, masts etc. Selvaraju and Ilaiyavel (2011) comment on the benefits of composites due to the ability of combination of functional requirements with these types of materials. They see an increased application of composites in the navy due to among others the low weight,
corrosion resistance and low signature value and noise dampening. Also in the commercial and leisure
market the effect of low weight and noise and vibration damping is an important asset in the increased
application.
Since composites tend to be more expensive in production it is necessary to look at life cycle costs. In
the Through life project (Throughlife, 2014) several applications were considered. The results from the
life cycle cost calculations show that after around 4.5 years the best performing sun deck design alternative starts to have less life cycle costs than a reference steel design. A car deck design in composite was
more profitable from the beginning.
In offshore the corrosion resistance and low weight are the main reason for increased use in low pressure pipes, tanks, cable ladders, gratings and trays. Edvardsson (2013) showed in a presentation within Elass EU project (Elass, 2014) the already existing application of FRP gratings, piping, pressure vessels
etc. Some investigations are going on in the use of composites for propellers. The better fatigue characteristics and theoretically better cavitation performance combined with the possibility of directional property optimisation make composites suitable for propeller application.
Due to the increased application in the maritime sector the effect of sea water exposure on composites
is an area of increased interest. Recycling remains a topic of interest as well as the assessment of
634
composite structures and repairs using composites. The latter is studied for example in the Co-Patch EU
project (Co-patch, 2010).
Next to the more conventional fibre materials such as glass and carbon an interest is seen in the use of
natural fibres such as hemp, flax and bamboo. These fibres are abundantly available, from a renewable
resource and bio-degradable. However, the hydrophilic nature of these fibres and the larger spreading in
mechanical properties are areas to be investigated. In general fire remains an issue in non-metallic materials. Therefore fire retardant composites are of interest. The increased attention for LNG has led to an
interest in the use of composites in cryogenic application.
3.2.1
Fire resistant materials
Fire issues remain an important research area for composite materials. To comply with the Solas regulations research into flame retardant materials is ongoing. Next to Solas the high speed code is the only
other code that allows for the use of composite (combustible) materials via prescriptive rules. The inland
waterway codes for now, as far as the authors know, do not permit the use of combustible materials.
However, some work is going on in that area to adapt the HSC for this purpose. Next to that different
national regulations exist. It would be beneficial to have regulations for inland and short sea shipping in
Europe, to open up this market for lightweight materials.
In September 2012 the EU FP7 Nanocore project was started (Nanocore, 2011). The aim of this
project is the development of a new flame retardant system for use in PVC-based polymeric foams for
sandwich core materials with low fire, smoke and toxicity (FST) values and enhanced mechanical
properties. This goal will be reached by insertion of nanoparticles modified with a phosphorous fire
retardant. Kandola and Krishnan (2014) studied the reduction of flammability of unsaturated polyester
(UP) resin by blending it with less combustible in char-forming resins such as phenol-formaldehyde,
melamine-formaldehyde and furans. Fire retardant fibre reinforced composites can be made using 3
different methods. Adding an outer protective material or the application of fire retardant paints of
coatings are passive protections. They have no effect on the combustibility of the fibre reinforced material. The latter can be done by chemical of physical modification of the basic resin material. Most of
the systems used in the latter approach either lead to the release of toxic gases or could decrease the
mechanical properties of the resin. The authors of the paper are studying several alternatives to identify
the best candidates for blending with UP. Several tests have been performed on co-cures unsaturated
polyester with a char forming resin, such as phenolic, melamine-formaldehyde or furan. The thermal
stability of all three options was higher than that of the UP. Depending on the test chosen to quantify
the different aspects of a materials fire performance, the alternatives rank in a different order. However,
overall it was concluded that the phenolic resin and its blend with PU shows the highest fire safety
ranking. In FP7 the project FIRE-RESIST was started. Information on this project can be found in a
presentation given by Paajanen and Hakkarainen at the September 2013 kick-off meeting of e-lass
(Paajanen and Hakkarainen, 2013). In the project 5 different types of fire retardant measures are assessed, e.g. multi-micro-layered structural composites, hybrid thermoset composites and particle-doped
polymer fibres for fire-retarded commingled composites. In the BESST, Breakthrough in European
Ship and Shipbuilding Technology) project (BESST, 2013) among others full scale flash-over fire tests
were performed on the demonstrator developed in the EU DE-Light project.
3.2.2
Bio-composites
Due to the increased environmental awareness of costumers and governments and rising prices of petroleum based products, the demand for natural bio-composites increases. In 2010 the global natural fibre
market share was 2.1 Billion US Dollar, growing to 3.8 Billion Dollar in 2016 according to Brief (2011).
Most of its growth is seen in the automotive and construction industry. According to Lucintel, the maritime industry, specifically the leisure boat industry, is only a small part of the total composites market,
although much of this market (68%) is already composite materials based. Also Cicala et al. (2010) state
that most application of bio-composites was in automotive applications. According to them this was because the bio-composites were mostly in the shape of short random fibre reinforced parts that were used
for non-structural applications, such as car roofs, covers and door panels. The studies that were done on
long fibre structural reinforced members were mainly based on housing applications. Fernberg (2014)
gave a presentation at the final Throughlife workshop in Papenburg, providing an overview on bio-based
composites in which fibres, resins, mechanical properties and costs are discussed. He also concludes that
application is mainly in small products although he does mention some 6 and 7 m sailing boats made
from bio composites. According to Jose et al. (2012) bio-composites can be divided into 2 groups, the
bio-fibres embedded in petroleum based plastics such as PE or PP and bio fibres embedded in bio-plastic
(PLA). The full green composites, so natural fibres embedded in a bio-plastic, are fully degradable and
sustainable. They may be used effectively in short life cycle mass production consumer goods. Lucintel
635
further divides the natural fibres into two main groups, wood and non-wood fibres. The Figure 7 shows
the subdivision made by Lucintel.
Figure 7. Subdivision of natural fibres according to (Brief, 2011)
Jose et al. (2012) discusses the main properties of several common synthetic and natural fibres. Although
the density of both groups is normally in the same order of magnitude, the tensile strength and Youngs
modulus are normally one order or more smaller for natural fibres. This is confirmed by data given by
Westman et al. (2010) and Cicala et al. (2010). Typical strength values for natural fibres such as hemp,
sisal, banana, flax, etc. are in the range of 300-700 MPa, whereas glass and carbon fibres normally show
values between 2000-4500 MPa. Also for the Youngs modulus the same trend is seen. Typical values for
glass are in the order of 80 GPa and for carbon even in the 230 GPa, while the natural fibres, according to
polymer composites volume 1, shows values of 10-30 GPa. From a strength point of view, banana and
sisal fibres are the best options. According to Jose et al. (2012) and Cicala et al. (2010), fibres will show
an increasing strength with increasing cellulose content and decreasing spiral angle with respect to the
fibre axis.
One of the main issues with respect to natural fibres is the hydrophilic nature of most of these materials. This means that the fibres will absorb water from the air leading to degradation in the fibre properties, but also to difficult and weak bonding between fibres and matrix. To limit these effects, treatment of
the fibres is needed. Several processes can be used for this, such as cold plasma chemistry, oxidative and
non-oxidative chemical treatment, acytilation and alkali treatment, (Jose et al., 2012), (Westman et al.,
2010), (Cicala et al., 2010). The effect of chemical treatments on the natural fibres is twofold. While most
chemical treatments lead to a decrease in the fibre strength due to the delignification and degradation of
cellulosic chains during the treatment (Cicala et al., 2010), polymer composites shows an increase of tensile strength of a sisal/PP composites after immersion in water, see Figure 8. This is probably due to the
better bonding between fibre and matrix after chemical treatment.
Figure 8. Effect of chemical treatment on the tensile strength of sisal/PP composites after immersion in water, Fibre
loading 2s0%, temperature 20 degrees C (Jose et al., 2012)
A study on chopped Kenaf fibre mats impregnated by a vinyl ester resin was carried out by Westman
et al. (2010). There was no specific treatment of the fibre mats. Specimens were tested in tensile and
636
flexural tests and immersed in water for 24 hours to assess the water uptake. The Kenaf fibres absorbed
approximately 10-12% of water, while a standard glass fibre specimen only absorbed around 0.2%. Without significant water absorption the specific strength of the Kenaf specimens was close to half of that for
the glass fibre specimens both in tensile and in flexural tests. The glass specimens showed a decrease of
roughly 10-15% in the tensile and flexural strength after 24 hours soaking in water, while the Kenaf specimens only had 50% (tensile) or 33% (flexural) of their specific strength left. The initial difference in
specific modulus was less, but the drop in specific modulus after immersion in water of the Kenaf specimen was again considerable (50-67% for tensile and flexural tested specimens respectively). The authors
expect that fibre treatment can reduce the water absorption and as such decrease the material degradation
after immersion. It can be concluded that although natural fibres provide an environmentally friendly and
cheap alternative to synthetic fibres. The lower mechanical properties, the larger spreading in material
properties and the susceptibility of the materials to sea water will probably be restricted to small, non
structural parts unless these issues are successfully countered.
3.2.3
Influence of sea water on non-metallic materials
The studies on the influence of sea water on composites focus on the effect of the all or single sided exposure to sea water of both static, fatigue and dynamic properties. Siriruk and Penumadu (2014) studied a
carbon vinyl either composite in air, fully immersed and exposed to sea water on one side. They noticed a
severe degradation of the fatigue life in exposed conditions. According to the authors the effect is due to
the growth of voids between fibres and matrix due to the contact with sea water, which leads to an early
fatigue failure. When fully immersed a reduction of fatigue life of 85% was seen, for one sided exposure
50% decrease was found. Poodts et al. (2013) tested glass fibre reinforced plastic also for the effect of sea
water on fatigue. They compared fully immersed specimens with specimens in air. A saturation of the
specimen in water was seen after 22 weeks. No clear difference was found for the submerged and dry
samples in fatigue testing. Comparing the two studies it seems that the choice of materials has a large
effect on the sea water degradation influence on fatigue. Also vibrational tests done by Poodts et al.
(2013) showed no differences. For a static 3 point bending test a small decrease was seen in the quasi
static strength in the first month of exposure, after this time the results remained constant. Korach et al.
(2014) not only studied the effect of sea water exposure but also of humidity and UV radiation. For their
research they exposed carbon fibre reinforced vinyl ester coupons to different conditions. The first group
was tested in a test chamber subjected to certain UV, temperature and humidity conditions. In these test
chambers an elevated temperature was maintained. The second group of coupons was tested in dark containers, fully submerged in fresh, sea or salt water. A final group was placed in a tidal pool, either just
above the water surface, in the tidal zone or fully submerged. A three point bending test has been performed on the specimen to see the influence of the different exposure conditions over time (6 months and
1 year of exposure). The various types of exposure do not have a significant effect on the flexural modulus. However, the flexural strength does show a decrease up to 18% for both indoor immersion and outdoor exposure. It should be noted that there is little difference between immersions in fresh or salt water.
The decrease in flexural strength is attributed to the degradation of the matrix fibre interface and the formation of micro-cracks on the coupon surface, which was also stated by Siriruk as the reason for the fatigue characteristics degradation on this type of material. A decrease in the energy release rate is seen in the
tidal pool and indoor immersed specimens. Most degradation in this respect was seen for the coupon exposed to 25% of sea air and 75% immersed in the tidal pool. Xu et al. (2012) emphasise the need to do
exposure studies only for one sided exposed specimens. Although they cannot make a clear comparison
with other authors due to a difference in test conditions, they find an impact strength reduction of only
10% after 29 months of exposure of an E-glass/vinyl ester composite, whereas other authors show values
of 30 up to even 70% for all-sided exposed panels. After 10 months the authors see a saturation setting in.
The reducing effect of one sided exposure found by Xu et al. (2012) is in line with the results seen by
Siriruk and Penumadu (2014).
3.2.4
Recycling and disposal
In the previous ISSC report attention was paid to the different recycling techniques and residual properties of recycled materials. Petrovic et al. (2012) looked at the tensile strength and tensile modulus of 20
mm short fibre reinforced glass epoxy material. The recycling was done using a chemical solvent to remove the epoxy. Nearly 6% of the glass fibres were also lost in this process. From the remaining recycled
fibres and from new fibres similar tensile test specimens were made. The recycled samples showed a
mean Youngs modulus and strength that was approximately 14% lower. This is due to the decreased
bonding of fibres and matrix due to the solvents, which damage the fibre surface layer.
Before recycling is opportune, the composite vessel or structure has to be scrapped first. Hellbratt
(2008) describes in his article the scrapping of a 54 m long Danish standard flex 300 patrol ships.
637
The scrapping took place in a floating dock in 7 days. Total costs of the operation were less than
0.1 million euro. In this example, the complete composite scrap was burned for the energetic value.
3.2.5
Application of non metallic materials at low temperatures
Due to the broad range of materials and material combinations composites are a good material for combining different functionality. As stated in section 3.1.4, industry is looking into alternative materials for
low temperature applications. Fibre reinforced composites could be one of these alternatives. Yu et al.
(2012) studied the effect of fibre addition to the insulation material and adhesive layers in a LNG containment system to delay or avoid crack forming. Normal insulation materials for LNG containment systems will crack when the primary containment barrier fails. This will lead to exposure of the second barrier to LNG and failure of that barrier. Within the paper an experimental and numerical study is described
into the effect of adding short glass fibres in both the insulation material and the adhesive layer between
the secondary containment system and the insulation material. From tests it was seen that if the adhesive
layer was reinforced with a volume fraction of more than 15.4% of glass fabric, the failure strain of the
second barrier, made of aluminium, was a factor 4 higher than for a non reinforced situation. This was
due to the spreading of plastic deformation.
3.3
Hybrid materials
Combining metals and composites in ship design leads to an optimised solution with the potential to reduce manufacturing costs associated with complex bows and sterns, while leading to a lower weight and
better non-magnetic properties. In his paper, Barsoum (2003) discusses different types of hybrid combinations of metal and composite (GFRP) and advantages both in normal operations and in naval applications
are discussed. The paper gives references to some research done in the area of hybrid joints, slamming
and deck buckling.
The ADAM4EVE EU project (Adam4eve, 2014) focuses on the development and assessment of applications of adaptive and smart materials (or combinations) and structures to enable ships to react more
flexible to the changing operational and environmental conditions. The use of smart and intelligent materials will also offer new functionality.
In 2011 the EU FP7 project MOSAIC was started (Mosaic, 2011). The main aim of this project is to
investigate if high strength low allowed steels could be used in specific structural details and to see if
composites could be used as a replacement for steel in parts of the ship, such as the superstructure, transverse bulkheads etc. The main reasons for the project partners to look into these replacements are improved structural response, reduced corrosion, reduced maintenance and operational costs and weight
reduction.
The US navy is building two ships with a composite superstructures and hangers, the DDG1000 and
DDG1001 for weight saving reasons. However, according to the US navy possible weight savings elsewhere in the ship were possible, so the third sister ship DDG1002 will be build, for a lower price, completely in steel again (Cavas, 2013).
The Oshima shipbuilding Co. together with DNV has developed an open hatch bulk carrier for the future. The concept is equipped with large composite hatch covers consisting of single skin glass-reinforced
plastic and a sandwich construction (Noury, 2014). All these applications show that it is a good development to combine the best of different types of materials to reach the most optimum design.
4.
4.1
4.1.1
JOINING AND FABRICATION TECHNOLOGY

Welding automation and recent developments in joining technologies
With the development of microelectronics and power electronics technologies, various advanced control
methods have been successfully implemented to the arc welding systems with superior dynamic performance. This enables to enhance quality, productivity, and usability in the automatic welding system for
the heavy industry field. The sampling time for the control of the automation welding system has been
significantly reduced from 10 ms to 1 ms with the application of high performance microprocessor. In
addition, the external interfaces of the conventional system including CC-Link, Profibus and DeviceNet
have been replaced by the high speed interface method such as Ethercat. It leads the development of the
automatic welding system requiring high accuracy position control of welding pool such as TIG welding
process. Recently, various studies have also been performed to develop the lightweight automatic welding
system with a universal robot, which can be applied to welding in narrow spaces during shipbuilding; see
e.g. (Ku et al., 2010), (Sanders et al., 2012) and (BESST, 2013). Some examples are shown in Figure 9.
The technologies including an intuitive user interface and high performance teaching pendent are replacing off-line teaching system, based on PC, and have accelerated the usability of the automatic welding
638
system as well. The recent application topics in the arc welding machines have been mainly focused on
the reduction of the energy consumption and the improvement of the welding quality by using high-speed
inverter control technology. Unlike the conventional welding machines (SCR type and low-speed inverter
type), high-speed advanced inverter technology also facilitates the light weight and compact design of the
welding machine. It provides also a speed regulation up to ten times higher compared to the conventional
inverter technology. A shipbuilding company has launched Hybrid welding power source, which
changes the polarity DCEN (Direct Current Electrode Negative) to DCEP (Direct Current Electrode Positive) and vice versa by electric circuit. It means that GTAW and FCAW process together are applicable in
a single power source. Furthermore, prototype Ethernet port integrated power source is on testing to monitor the welding parameters and machine error messages. In the future, the supervisor will be able to
check the welding conditions in real time regardless of working sites.
Self-driving mobile welding robot
for double-hull structure
Flexible and modular laser

equipment
Figure 9. Examples of recently developed automated welding robots, (Ku et al., 2010) and (BESST, 2013)
4.1.2
Underwater welding
The offshore platforms and pipelines require continuous operation and, thus, reliable underwater welding
on-situ. The underwater welding is applied also for ships, when ship is repaid afloat. During the last three
decades, underwater welding has been developed as a solid fabrication and repair technology; see e.g.
(Reynolds, 2010). At the same time, inspection has developed to ensure the quality of underwater welds.
A comprehensive review of the field was provided by the international workshop on the underwater welding and inspection technology (Liu et al., 2010). Traditional arc and hyperbaric welding have been improved and new techniques such as friction processes have been further developed. Richardson et al.
(2010) summarised the development in dry hyperbaric welding. In past years, the focus has been primarily on the development of remotely controlled welding for the deep water technologies. The gas metal arc
and flux cored arc processes have received considerable attention over the past years, since they can produce stable welds at depths to 2.5 km without any pressure dependent process limitation. In addition,
welding experiment with solid and metal cored wires has proven their robustness on a range of materials
including low and high alloy steels. During last decades, the friction welding was also successfully applied for underwater environment; see (Couch, 2010), (Gibson et al., 2010), (Cui et al., 2014b) and (Cui
et al., 2014a). Benefit of friction stud welding is that it is a solid phase process and, thus, the absence of a
liquid weld pool avoids potential problems with hydrogen absorption. In addition, there is no change in
welding parameters and weld chemistry with increased water depth. However, the process is presently
limited to the welding of studs and small hot tapping fittings. Friction stir welding avoid this limitation,
(Liu et al., 2011) and (Zhao et al., 2015). Fitness for service assessment and inspection of underwater
welds were discussed in Dong (2010), Tern et al. (2014), Dong et al. (2012), Jia et al. (2013), Cridland
(2010), Perez (2010), Goldberg and Marshall (2010), Al-Abbas (2010) and Olson et al. (2010). In fitness
for service assessment, the main challenges are weld quality definition considering various defects, fracture mechanics based stress intensity solutions for complex joint configurations, the long-term load modelling and residual stress distributions that can be significantly different from air welds. Holdsworth and
Reynolds (2010) provided a comprehensive review about the standards for underwater welding. During
the recent years, the standards have been further developed. However, further development is required to
make them better to support the variety of industries that apply underwater welding and inspection.
4.1.3
Frictions stir welding of steel
Friction stir welding (FSW) process was invented in 1991 (Thomas et al., 1991), and it is nowadays widely used for lightweight metals such as aluminium in deck structures for ships (Shah &Tosunoglu, 2012).
During the last decade, a significant effort has been made on the development of cost effective and
639
durable tools for harder alloys such as steels; see e.g. (McPherson et al., 2013) and (Cam, 2011). Until
recently, the transfer of FSW capability into the steel application has been very limited due to the relatively poor performance and high cost of the tools required (Azevedo et al., 2014). This situation is changing
with FSW tools that are capable of producing industrially useful lengths of welds in steel, (Miyazawa
et al., 2012), (Cater, 2013) and (Toumpis et al., 2014). The lower heat input associated with FSW should
produce less metallurgical change in the HAZ, smaller distortion and residual stresses in comparison to
fusion welding processes, see e.g. (McPherson et al., 2013) and (Mahoney et al., 2010). Furthermore, the
problems with hydrogen cracking in steels as well as welding fumes would be eliminated (Gomes et al.,
2012). Most of the studies on FSW of steels reported that FSW achieves grain refinement in the stir zone
of the carbon steel similar to aluminium alloys, (Cam, 2011), (Lienert et al., 2003) and (Reynolds, 2010).
Additionally, complex phase transformations also occurred in the FSW process. The microstructural evolution of steels during FSW is more complicated than that of aluminium alloys due to the occurrence of
transformation, recrystallization, as well as grain growth at high temperature. The effect of the carbon
content and the transformation on the mechanical properties and microstructures of the FSW carbon steel
joints was investigated by Mahoney et al. (2010) and Fujii et al. (2006). They concluded that the microstructures and mechanical properties of the carbon steel joints are significantly affected by the welding
conditions; the strength of the steel joints increases with a decreasing heat input. Unlike the FSW of aluminium alloys, the FSW of steels may require the use of a shielding gas mainly due to protect exposed
portion of the tool from oxidation; see e.g. (Shah and Tosunoglu, 2012). New FSW tools based on iridium
are almost unaffected by oxidation at elevated temperatures (Miyazawa et al., 2012). This may discard the
need for shielding gas in future applications.
Tool design and life are major considerations in FSW, (Thompson et al., 2013), (Meilinger & Trk,
2013), (Meilinger & Trk, 2013) and (Rai et al., 2011). Rai et al. (2011) presented an insight on the several important aspects of FSW tools such as tool material selection, geometry and load bearing ability,
mechanisms of tool degradation and process economics. The hybrid polycrystalline cubic boron nitride
(pcBN)/W-Re tool families are currently amongst the best performing systems for FSW of steels, (Cater,
2013) and (Mahoney et al., 2010). Tool life may become a critical factor, although the process cost can be
influenced by other factors. The typical failure mode for the current tools is for the shoulder to wear back
at the junction of the shoulder and the probe, and for the probe then to break off. By using iridium (Ir) and
rhenium (Re) the recrystallization temperature and high- temperature strength can be increased. Application of an Ir-10%Re welding tool to AISI 304 stainless steel was studied in Miyazawa et al. (2012).
Mechanical properties of FWS steel joint were studied e.g. by McPherson et al. (2013) and Azevedo
et al. (2014). McPherson et al. (2013) investigated a series of 4, 6 and 8 mm DH36 steel welds, which
were produced using optimum welding conditions. The mechanical properties of FSW joint were compared to SAW joint typical shipbuilding production process route. Overall, the performance of the FSW
material was superior to the SAW joints. Distortion and fatigue were particularly positive in the FSW
joints. An 8 mm thick plate was also produced using two FSW passes, one from either side, and it was
found to have superior toughness and fatigue performance compared to the single sided 8 mm FSW material. Good fatigue properties were reported also by Azevedo et al. (2014). They studied fatigue behaviour of FSW welded steel shipbuilding joints in the GL-36. The 4-mm-thick plates were welded with
several different parameters, tools and pre-welding conditions. The results showed the feasibility of welding shipbuilding steel by FSW producing good quality welds. Micro-hardness results showed an increased
tendency from the base material to the stirred zone and, thus, all tensile specimens broke outside the
welded region, see
Figure 10. The fatigue test results showed significantly higher fatigue strength in comparison to FAT80
design curve, which is typically used for arc welded butt joints.
PM
Figure 10. Examples of a FSW tool, welded steel joint and hardness distribution of weld, (Toumpis et al., 2014) and
(Meilinger and Trk, 2013)
640
4.2
4.2.1

Plate bending with line heating
Line heating has been used as a method for forming curved shell plate for more than half a century. Traditional flame heating is strongly related to the experience and know-how of a worker. During last two decades, significant effort has been made in order to automate the line heating method. Recently, the fullyautomated plate bending system was implemented in practise; see
Figure 11. This automated bending system (IHIMU- ) consists of a heating plan creation program,
robots to handle the heating, and plate turnover system (Tango et al., 2011). The system calculates heating plans by using the Finite Element Method and inherent strain theory. Saenz et al. (2011) studied the
factors affecting inherent deformation during plate forming by line heating. They focused on the plate
edge effect caused by fact that inherent deformation at plate edge does not behave similarly in the central
region of the plate. Saenz et al. proposed two new methods for line heating to minimize the plate edge
effect. Osawa et al. (2011) developed method for numerical estimation of inherent deformation induced
by induction line heating. An electromagnetic-transient analysis of induction heating test was carried out
and a transient conduction simulation process was proposed. The developed method was successfully
validated with experiments. Since induction line heating can provide more controllable heat input in comparison to traditional flame heating, increasing interest has been shown to it during the last years; see e.g.
(Lee, 2012). In addition, high-frequency induction heating can heat thick steel plates to a target temperature through the desired depth in short time. Lee (2012) studied the plate bending by multidisciplinary
analyses. The new simulation program for induction heating was proposed to overcome the restriction of
shape of coil, clearance between coil and plate, and consideration of the variance of parameters dependent
on temperature. Lee and Hwang (2014) and Lee et al. (2013) utilized laboratory-scale high-frequency
induction-based line heating to investigate the influence of heating patterns on the permanent deformation
behaviour of an SS400 thick plate. They concluded that proper design of triangular heating pattern appeared to be the most important factor in determining the final shape of the thick plate. A numerical model to predict a 2-dimensional circular heat input for triangular heating was proposed in Bae et al. (2012).
The parameter influencing on induction heating has been studied and discussed by Zhang et al. (2011)
and Zhang et al. (2012). The prediction and evaluation of curved shapes after heating is one of the fundamental issues for automated plate bending system, (Seong et al., 2013) and (Nguyen et al., 2014). An
approach to hull plate forming was developed by Seong et al. (2013). An inverse solution to threedimensional plate forming was proposed. Bending angles and shrinkages were firstly achieved and they
were converted to forming parameters through the database map, which represented the relationship between geometrical parameters and forming parameters. Experiments were performed for validation. The
alternative approach for heat-path prediction was developed by Nguyen et al. (2014). They developed an
artificial neural network system, where the vertical displacements and transverse shrinkage of the plate
are used as inputs. The induction heating positions and parameters are the output of the models. The system utilised laminated plate theory and with Finite Element Method.
Figure 11. IHIMU- automated steel plate bending system (Tango et al., 2011): deformation caused by local
heating, automated process flow and robotic heating system.
4.2.2
Post-treatment of welded joints and plate edges
The utilisation of the high-strength steel materials in marine applications requires use of advanced production technology. If the traditional production technology is applied, it is assumed that the crack-like
defect occurs. Then, the increased fatigue strength for high strength steel is lost due to increased notch
sensitivity. Because of the higher notch sensitivity, the stress concentration factor is higher for high
strength steel than that of normal strength steel (de Jesus et al., 2012). However, post-treatment of critical
641
structural details such as welded joint or plate edge can remove surface defects and, thus, allow better
utilisation of high strength steel. Remes et al. (2013) and Korhonen et al. (2013) investigated the influence of surface integrity on the fatigue strength of high-strength steel used in large marine structures e.g.
the balcony openings of a cruise ship; see Figure 12. The investigation utilised large-scale specimens,
which have yield strength of 355 MPa, 460 MPa, or 690 MPa. After plasma cutting, the specimens were
treated by grinding or by grinding followed by sandblasting, i.e. using post-cutting treatments that are
suitable for shipyard conditions. The fatigue strength was found to increase with increasing material
strength for the grinded specimens. The slope of the S-N curve was influenced by both the material yield
strength and the surface treatment condition. Based on the results it was concluded that a significant increase in fatigue strength can be achieved using manufacturing methods suitable for shipyard conditions.
The increase in fatigue strength was determined to be due to the reduced surface roughness and compressive residual stresses. Sandblasting after grinding increases the surface roughness, but reduces the fatigue
strength only slightly since sandblasting produces a favourable defect shape that has a larger notch radius
when compared to grinding.
Fatigue critical
opening corner
Nominal stress range [MPa]
1000
800
700
600
R=0.1
y=690 MPa
Grinded surface
500
400
y=460 MPa
300
y=355 MPa
Plasma-cut edge
200
FAT160
(m=4)
100
4
1010
000
5
100
10000
1 000
106000
10 000
107 000
Number of cycles to final failure
Figure 12. Influence of post-cut treatment and material yield strength on the fatigue strength. The reference design
curve FAT 160 is given for comparison, (Remes et al., 2013).
Post treatment has also been known to have positive influence on fatigue strength of welded joints. However, the main challenge in their utilisation in practise has been the lack of the design recommendation.
Recently, the utilization of post-weld treatment has been supported by several investigations. In particular, the high-frequency mechanical impact (HFMI) treatment is observed to be a reliable and effective
method for post-weld treatment (Yildirim & Marquis, 2013). The benefits of the HFMI technique are low
induced noise level and small tools, which produces a smooth rounding to weld fusion zone, compressive
residual stresses and hardened surface layer at weld; see e.g. (Leitner et al., 2014), (Le Quilliec et al.,
2013) and (Tai & Miki, 2014). Due to this, the fatigue crack initiation life is increased and the fatigue
strength of welded joint has been significantly improved, (Leitner et al., 2014), (Tai & Miki, 2014),
(Yildirim & Marquis, 2012), (Yildirim et al., 2013), (Yildirim & Marquis, 2013) and (Yildirim &
Marquis, 2014). Based on the experimental and numerical investigations, design guideline is proposed for
fatigue strength improvement depending on material yield strength; see Figure 13. In comparison to aswelded joint, the improvement achieved with HFMI varies between 60% and 150% depending on material yield strength (Marquis et al., 2013). The quality control issues have been investigated by Yekta et al.
(2013) and the quality assurance guidelines is proposed by Marquis and Barsoum (2014). This proposal
considers steel structures of plate thicknesses of 5 to 50 mm and for yield strengths ranging from 235 to
960MPa. Since the post-weld treatment is possible only for weld toe side, the advantages of the treatment
can be fully realized if other potential failure modes such as weld root failure are avoided, (Marquis,
2010). The fatigue strength improvement is influenced also by preload and mean stress since the compressive stress is considered as one of the main reasons for fatigue improvement, (Marquis, 2010) and
(Okawa et al., 2013). Consequently, Deguchi et al. (2012) proposed that the utilisation of HFMI treatment
should be focused on the structural members for which extensive compressive load rarely occurs. To further develop guidelines for marine structures, the future research should be directed towards large-scale
structures and variable amplitude loading with varying mean stress.
4.2.3
Hybrid structures and joints
Use of composite is alternative approach to improve the strength to weight ratio. Typically in marine
structures, composite has been utilised for selected weight or strength critical areas or joints. Thus, these
hybrid structures or joints consist of both traditional steel and composites; see e.g. (Kabchea et al., 2007).
Composite has been successfully applied as repair path for corroded and damaged onshore pipelines and
642
ship deck structures; see e.g. (Grabovac & Whittaker, 2009). To extend these repairs to offshore pipes
such as risers, (Alexander & Ochoa, 2010) investigated the composite repair under tension, bending, internal and external pressure. Based on limit analysis and strain-based design, a new carbonfibre composite repair system was introduced. The effectiveness of composites for underwater steel pipeline repairs is
discussed also in Shamsuddoha et al. (2013). Composite-to-metal adhesively bonded lap joint joints were
studied by Anyfantis and Tsouvalisn (2013). They concluded that the effect of the adhesive thickness and
stiffness ratio is negligible compared to the effect of the overlap length to the stiffness and strength of the
joints. Errouane et al. (2014) studied cracked aluminium plate and developed a numerical model for path
design considering the patch height, width and thickness, as well as adhesive thickness. The results show
that the stress concentration at the crack tip can be strongly reduced by using patch repairs. Fatigue behaviour of carbon fibre reinforced polymer plate was studied by Yu et al. (2013). Based on the experimental
and theoretical investigations for different degrees of fatigue damage, significant fatigue life extension
was observed for the strengthened cracked steel plates. Thus, the over-lamination of the welded joint provides an alternative solution to improve fatigue strength of the welded joints. Although the composite is
most widely applied for repair path in marine application, the larger composite structures are coming to
obtain more weight saving. For instance, Kabchea et al. (2007) studied bolted composite metal hybrid
panel assembly, where deck plates were replaced by light weight solution. Furthermore, composite was
recently used in superstructures of two navy ships, (Cavas, 2013). However, due to higher cost US Navy
switched from composite back to steel in the next new buildings.
HFMI Equipment
Before the treatment
Fatigue resistance
After the treatment
Figure 13. An example of high frequency mechanical impact (HFMI) equipment, treated weld and the recommended
fatigue resistance curves for HFMI treated welds in comparison to the curve for weld without treatment,
(Yildirim and Marquis, 2014) and (Marquis et al., 2013)
4.3
The good production quality and workmanship are the key issues for structural safety and also for successful utilisation of new materials. The latest developments on quality standards were discussed extensively in the previous ISSC Committee V3 report (Chapter 5). Thus, this report is aimed to summarise
recent findings in the field of production quality and strength.
4.3.1
Weld geometry and misalignments
The joining and fabrication processes induce unfavourable defects, which can have significant effect on
the strength of the structures. Thus, the quality of production would directly influence the strength. The
most common joining technology in shipbuilding industry is welding. The variations in geometry, generated at the weld toe, create local stress concentrations. The mechanical behaviour of welded structures can
be significantly affected by the effects of the employed joining process. Performing technological examinations for different materials and welding tasks is directly connected with the weld qualities and
strengths produced depending on the type of joint and application, (Neubert & Kranz, 2013), (Fricke
et al., 2012a) and (Fricke et al., 2012b). Different weld type and process could influence geometrical parametric of weld joint, and the stress concentration factors and fatigue lifetime could be different,
(Ahmadi & Lotfollahi-Yaghin, 2012), (Pasqualini et al., 2013), (Ahmadi et al., 2013), (Fricke &
Tchuindjang, 2013) and (Remes, 2013). Design of block joints in ship structures of permissible tolerances
for axial and angular misalignment is particularly important for fatigue assessments. The partly large imperfections such as axial and angular misalignments were extensively recorded in order to obtain the characteristics of typical pre-deformations. For plate thickness below 5 mm, the production induced distortion increases and thus, it is recommended to extend the tolerance values in the production standards for
643
angular distortion, (Eggert et al., 2012) and (Lillemae et al., 2012). The optimised weld processes are
required to achieve low distortions and good weld profile, and thus, improvement in the fatigue strength,
(Remes & Fricke, 2014). Stoschka et al. (2013) studied a relation between experimental fatigue life and
the weld parameters. The experimental results showed that in case of high quality welds with negligible
geometric notch factor, a small, but distinct influence of the filler metal on fatigue is observable.
4.3.2
Effect residual stress and distortions
Heating and the subsequent cooling also modify the material and mechanical properties in this specific
area and introduce a residual stress field. Consequently, welded joints become critical areas regarding
fatigue damage, and most fatigue cracks initiate at the weld toe. For fatigue strength of welded structure,
the influence of welding residual stresses has to be considered. Pan et al. (2013) emphasizes the important
role played by residual stresses in determining the failure location. For an accurate estimation of the fatigue performance of welded joints, not only the initial residual stress field but also its variation under load
is decisive. The relaxation of residual stresses during fatigue loading, however, reduces this hazard to the
structural health, (Baumgartner & Bruder, 2013). That is, before considering the influence of residual
stresses in fatigue, the effect of fatigue on residual stresses should be understood, (Farajian, 2013). Heating and the subsequent cooling has also significant influence on material properties particularly in aluminium structures, where the material properties in the (HAZ) can be reduced even 50% in comparison to in
base material strength, (Sensharma et al., 2010).
Most fabrication processes introduce distortions and affect the material properties. Shapes and magnitudes of the initial geometrical imperfections were observed to affect the ultimate load (Xu & Soares,
2012). The existence of a heat affected material showed no influence on the ultimate strength of the tested
panels, (Paulo et al., 2013). Concerning the effect of the residual stresses on the collapse load of the
plates, it can be concluded that this stress field leads to a decrease in the plate collapse load, when comparing with a stress free model, (Paulo et al., 2014). Gannon et al. (2012) researched the threedimensional distribution of welding-induced residual stress and distortion, and the results showed the
residual stresses reduced the ultimate strength of the stiffened plate by 11% with a consequent reduction
in hull girder ultimate moment of 3.3%. Cerik and Cho (2013) investigated the effects of residual stress
on the ultimate strength behaviour of stiffened cylinders, and it is shown that rather than cold bending
stresses, especially yield level tension zone induced by welding has a significant effect on both strength
and stiffness. The presence of residual stresses in between stringers causes considerable loss in axial stiffness of stiffened cylinders. It can be concluded that this fact should be taken into account.
4.3.3
Utilisation of high strength steel and thin plates
High strength steels has showed increasing interest in shipbuilding industry in order to reduce steel
weight and increase energy efficiency. Nevertheless, the sensitivity to mechanical property degradation
by hydrogen increases dramatically with strength. This phenomenon leads to hydrogen-assisted cold
cracking. It was proved that this can be avoided by the established sufficient post-weld heat treatment,
(Steppan et al., 2013) and (Stein & Kallage, 2008). In order to utilise high strength steel in fatigue loaded
marine structures, a better surface integrity than that obtained with traditional production process is
needed. The investigation shows that post-cutting treatments suitable for shipyard conditions can considerably increase the fatigue strength of high-strength steel used in opening corners of a large-scale structure. Sandblasting after grinding increases the surface roughness, but reduces the fatigue strength only
slightly, (Remes et al., 2013) and (Korhonen et al., 2013). The fatigue strength improvement as a function
of yield strength was observed also for post-weld treated joints in Yildirim and Marquis (2012). Furthermore, recent investigations showed that increased fatigue strength can be obtained for cut plate edges
without treatment, when cutting quality is good, (Sperle, 2008) and (Laitinen et al., 2013).
New light-weight structures in shipbuilding aim at thinner plates in combination with welding
processes characterized by low heat-input, i.e. laser and laser-hybrid welding. The distortions are an key
issue particularly with thin plates with thickness of less than 5 mm, (Eggert et al., 2012), (Lillemae et al.,
2012), (Lillemae et al., 2013) and (Fricke et al., 2013). For thin plates and slender ship structures, the
secondary bending stress due to angular misalignment plays an important part and it can changes as the
function of the applied tension load. Therefore, it is important to consider the plate straightening effect in
structural stress analysis and fatigue strength assessment, (Lillemae et al., 2012). The lower heat-input
of laser-based welding methods is able to produce joints with lower distortion, which increases
fatigue strength noticeably in terms of nominal stress range, (Lillemae et al., 2012), (Fricke et al., 2013)
and (Selle et al., 2013). The quality of the welds is also importance, when the fatigue strength of
thin plate is considered. Thin plate structures are sensitive to welding-induced flaws such as undercut,
(Remes & Fricke, 2014). However, with optimised welding parameters, these challenges can be avoided
644
and similar fatigue strength properties can be obtained as those of thick plates, (Remes & Fricke, 2014),
(Fricke et al., 2013) and (Selle et al., 2013).
4.4
Dimension control is an important part of efficient and reliable production. Especially for thin and
slender structures, the control of the production-induced deformation is one of the main challenges.
Chen et al. (2011) introduced an approach for measuring weld-induced initial distortions. The approach
utilises photogrammetry, which generates the displacement field from sets of digital images. The procedure was validated against direct measurements of the deflections of box girders and statistical analyses of the error between the measurements and photogrammetric evaluations were presented. Hiekata
et al. (2011) developed a laser-based measuring system for the accuracy evaluation of a curved shell
plate. In this system, the point cloud data of a curved shell plate was measured by a laser scanner, and
the displacement errors between the measured data and design data expressed with NURBS are compared to provide feedback for quality control. The production process has significant influence on the
amount of deformation. Consequently, reliable methods for prediction of deformation and shrinkage are
essential for dimensional control. Effect of welding sequence on residual stress and distortion in flatbar stiffened plates were investigated by Gannon et al. (2010). The simulation considered thermal and
structural analyses using an element birth and death technique to model the addition of weld metal to
the work piece. A smaller distortions and residual stresses were observed to welding sequence, where
welding was started from the middle. In addition, it has been shown that distortions predicted by the
finite element model agree well with experiments. Camilleri et al. (2010) studied the influence of different tack welding fabrication procedures on the deformations. The length of the tacks, number of
tacks, and position of tacks with respect to the seam weld were altered, and their sensitivity was investigated. It was concluded that tacking the plate at fewer, but longer tack welds led to less deformation.
Initial clamping during tack welding had also significant effect on the final distortion. Reduction of
welding distortion for a hatch forming production was studied in Wang et al. (2013b). It was shown
that optimised assembly process and welding sequences can reduce distortion significantly. Weld
shrinkage predictions can be predicted using semi-analytical regressions models or finite element (FE)
analysis. The regression model is based on experimental and simulated data (Wang et al., 2013a), while
finite element simulation is done using linear-elastic or plastic analysis, (Wang et al., 2013b), (Kim
et al., 2012) and (Yang et al., 2011). (Yang et al., 2013) and (Yang et al., 2014) developed a regressionbased model for overall distortion prediction. The input data of model is panel design parameters including parameters such as material type, plate thickness, stiffener shape, spacing, and length, and
overall panel dimensions. In addition, fabrication details such as the welding process, weld sizes, welding parameters, and the use of fixtures can be considered. Accurate prediction of distortions, however,
requires the utilisation of finite element method. Wang et al. (2013a) applied elastic FE analysis based
on the inherent deformation theory. The inherent deformations of different types of welded joints included were evaluated beforehand using thermal elastic plastic FE analysis. Kim et al. (2012) developed equivalent load method based on inherent strain for welding distortion analysis of hull blocks.
The equivalent load was determined by integrating inherent strain components, which were calculated
in HAZ using the highest temperature and the degree of restraint. Then, welding distortion was calculated by elastic analysis. Yang et al. (2011) also introduced two-step distortion modelling methods for
large welded structures to reduce the computation time. To improve overall quality management in
shipyard, Buksa et al. (2013) introduced quality improvement system considering all stages of the production process in a shipyard. They utilized a differentiation-based approach to quality management,
and by taking targeted corrective and proactive actions into account they identified points of the manufacturing process, in which errors are likely to occur. Improvement for planning and scheduling of retrofitting for old ship were discussed by Krause et al. (2014).
5.
5.1
CORROSION PROTECTION
Protection rules
Germanischer Lloyd AG, Hamburg, has established some guidelines and rules for corrosion protection in
2010, (GL, 2010). These Guidelines contain technical fundamentals on corrosion and the rules applying
to corrosion protection on ships, structural parts, components and structures under maritime environmental and application conditions.
Under the condition that the corresponding boundary conditions are observed, they can also be applied
to other systems, structural parts and components. These Guidelines are intended to supplement the
GL Rules for Hull Structures (I-1-1), Section 35 and the GL Rules for Coating of Ballast Water Tanks
(VI-10-1) and for Corrosion Protection of Crude Oil Cargo Tanks (VI-10-3) which are limited to only
those aspects which are imperative from the classificatory point of view and which shall always be
645
complied with for the construction of ships with GL Class. National or international provisions and rules
are to be observed in addition.
Similarly, NORSOK has developed standards and guidelines for ship structure corrosion issues, such
as Petroleum and natural gas industries - Floating offshore structures - Part 1: Monohulls, semisubmersibles and spars (ISO 19904-1:2006), Petroleum and natural gas industries -- Arctic offshore structures (ISO 19906: 2010), Common Structural Rules for Double Hull Oil Tankers with Length 150 metres
and above - DNV Rules for Classification of Ships (DNV Ship Rules Part 8 Chapter 1:201207), Assessment of structural integrity for existing offshore load-bearing structures. Edition 1 (NORSOK N-006) ,
Design of Offshore Steel Structures, General (LRFD Method) - DNV Offshore Standards (DNV-OSC101:201104), Concrete LNG Terminal Structures and Containment Systems - DNV Offshore Standards
(DNV-OS-C503:201010), (Norsok, 2014).
5.2
5.2.1
Coating and paints

Epoxy-based coating systems
Ingle et al. (2011) examined the databases that track the corrosion/coating failure conditions in specific
tanks of US Navy vessels since the late 1990's. They reported that the legacy solvent-based coating failures includes higher levels of coating loss as a function of time as compared that the ultra-high-solids
(UHS) coatings. It was also shown that the 11 year UHS data did not contraindicate the US Navys goals
for a 20 year life span from seawater ballast tank coatings. Baere et al. (2013) compared the conventional
construction method of ships ballast tanks, grade A steel and protected with a standard epoxy coating
backed up with sacrificial zinc anodes with some feasible alternatives. The considered alternatives are: (i)
an increase of the scantlings, (ii) application of the novel and more durable TSCF25 coating (iii), the use
of corrosion resistant steel or (iv) a standard PSPC15 coating combined with lifetime lasting aluminium
sacrificial anodes. They concluded that the durable coating and the use of lifetime lasting aluminium
anodes are bound to improve the actual basic tank concept. Corrosion resistant steel becomes attractive
depending upon the evolution of the international steel market.
5.2.2
Zinc-rich paints
Shi et al. (2009) studied the corrosion behaviour of an epoxy zinc-rich paint on interface-contaminated
carbon manganese-silicon steel. They found that the corrosion resistance of the contaminated paint was
significantly influenced by diffusion of zinc corrosion products during the initial stage of immersion, and
diffusion of iron corrosion products at the end of immersion. Three transmission line models were applied
to account for the corrosion process of the uncontaminated and contaminated zinc-rich paints.
Gergely et al. (2014) developed zinc-rich hybrid paint coatings using nano-size particles composed
of alumina hydrate modified with polystyrene-sulfonate (PSS) doped polypyrrole (PPy) and either purifiedor functionalised multi-walled carbon nanotubes (MWCNTs). They demonstrated the improved
barrier and galvanic function of the hybrids over traditional zinc-rich paints (ZRPs) by examining performance metrics.
Hao et al. (2013) studied the inhibitive properties of epoxy coatings containing different volume fractions of zinc phosphate. They showed that zinc phosphate could improve the protection ability of epoxy
coatings and its best volume fraction was 30%. They also showed that the presence of zinc phosphate can
form an inhibiting film which was composed of the phosphating film of FePO4, Fe2O3, and FeO, as well
as the shielding film of zinc phosphate on the steel surface.
Kowalczyk and Spychaj (2014) investigated the corrosion protection of a steel substrate using zinc-rich
epoxyalkyd paints (ZRPs) modified with ionic liquids (HMIMPF6 or BMPyrrTFSI). They found that the
pyrrolidinium IL significantly reduces the corrosion rates of steel and zinc but the ZRPs with HMIMPF6
offered better cathodic protection of steel than unmodified ZRP and ZRPs with BMPyrrTFSI.
Schaefer and Miszczyk (2013) investigated the influence of nanosized particles on electrochemical
action of standard zinc-rich paints by means of SEM as well as potential and impedance measurements. Samples with different concentration of nanoparticulate zinc (0%, 5%, 10% and 40%) were
tested, and they found that addition of 510 % nanoparticulates extended the galvanic action of the
coating.
5.2.3
Thermal spraying and deposition
Hausbrand et al. (2012) developed the thin silica coatings (1560 nm) by applying a flame assisted chemical vapor deposition (CVD) process. These coatings showed good corrosion properties in cyclic humidity
testing (bare) and salt spray testing (painted). It was found that the corrosion started at pin-holes and proceeded then further under galvanic coupling with the surrounding regions, and the corrosion rate was
reduced by a factor 10 or more with respect to uncoated steel.
646
5.2.4
Antifouling (AF) coatings
Acevedo et al. (2013) investigated the antifouling activity of organic extracts from some epibiont-free
Colombian Caribbean Sea sponges and a sea-cucumber. Extracts were incorporated into hard stable gels
and into soluble matrix antifouling paints. They demonstrated that the greatest antifouling activity was
contained in extracts of A. Tabulata and Holoturia Glaberrima.
Bellotti et al. (2012) explored the possible application of a natural and abundant product, such as tara
tannin (TT), in the preparation of antifouling (AF) coatings. They found that the AF performance of
tara tannin derivative was better than that of quebracho tannin, and the use of the fruit pods of the TT
avoided the cutting down of the trees as it occurs in the case of quebracho trees.
Chambers et al. (2014) investigated a combination of the antifouling compound (a natural product extract) and the delivery system (control depletion polymer) by using surface engineering techniques and
analysis. They examined the effect of incorporated natural product on the matrix, and potential issues in a
working environment, in some form of binder on a substrate in the ocean, were discussed. Flemming et al.
(1998) have evaluated a system for fouling and biofouling in marine systems.
5.2.5
Self Healing Coatings
Within the European project THROUGHLIFE, (Throughlife, 2014) self healing model coatings have
been developed, (Scharf et al., 2013) and (Molter et al., 2014) and tested at lab scale as well as in a ballast
water tank of a ferry with good success. The tests showed good effects of the embedded micro capsules
on corrosion life time at about 10% higher cost compared to standard coatings, see Figure 14. Discussions
to turn the research results into commercial products are ongoing as well as long term tests.
Figure 14. Damages in the model coatings caused by the Erichsen cupping after 1000 hours of salt spray test
(source: Fraunhofer IFAM), (Throughlife, 2011)
5.2.6
Intelligent Coatings
The European project ADAM4EVE (Adam4eve, 2014) is developing and testing coatings which change
their colour and reflection properties in dependence of the temperature. Those coatings can be applied to
steel surfaces or windows to reduce the energy demand for air conditioning of reefer or cruise ships,
(Lackieren, 2014). The project is ongoing and practical tests on board of ships are foreseen at a later
stage.
5.2.7
Ice-breaker Coatings
Ice strengthening on the other hand is found much more commonly in ships designed for Arctic or Antarctic work. There is no actual universal definition of what needs to be done to a ship to be "officially
ice strengthened" and it can be applied to all manner of ships, whether supply ships, tankers, container
ships, warships etc. Commonly ice-strengthened ships can cope with continuous one year old ice about
50cm - 100cm thick. Breaking ice by any ship is not a case of forcing the ice aside, but by the ship riding up and over the ice in front of it, with the weight of the ship then breaking the ice, this may be a
continuous process or can result in a lot of back-and-forth in particular thick places, (Coolantarctica,
2014).
These ships are heavy for their size, to make them more effective at breaking through ice when they are
pushed up above it by their engines. Hull made from special steels designed for optimum strength at low
temperatures Air bubbling systems to assist ice-breaking. Air is forced under pressure from 2m or so below the water line where ice is met, helping to break it and move it out of the way. Extra thick steel is
used at the bow, the stern and at the waterline. An "ice horn" is used to protect the rudder and propeller
when in reverse, and an "ice knife" in front to protect it when in forwards motion.
Rounded keels and a lack of stabilizing fins means that progress is quicker and smoother through ice
and that there aren't any parts to be ripped off. A further discomfort comes from breaking through
continuous thick ice with constant vibration, noise and jarring against the ice. The hard coatings used to
protect the knife is described, (International, 2014).
647
These coatings have been used on ice breakers for many years. Such coatings require specific film
properties and application conditions to achieve their unique anti ice abrasion characteristics. Class
societies recognise these unique properties and give allowances to reduce steel scantling thickness if
these coatings are used. The reduction of steel weight is a key driver in the design of offshore structures
by engineers.
These low friction coatings are able to retain their smooth surfaces even when moving through heavy
ice up to 2m. This is demonstrated by in field measurements. The very best of coatings have a track
record going back 35 years.
Ice abrasion resistant coatings are now a key requirement for the hulls of polar class drill ships.
The Stena DrillMAX ICE currently under construction in SHI Korea makes use of such a coating.
With an upgrade to Polar Class 4 for its classification it will be able to operate in 1st year ice up to
2m thick.
Some yards are not keen to apply special ice abrasion resistant coatings as specialist equipment is
needed. However, for an industry that prides itself on using the most appropriate and state of the art technology this is a poor excuse to allow yards to select inappropriate coatings. Achieving the maximum performance from ice abrasion coatings has already been reviewed, (Park et al., 2006). The review reinforces
that true ice abrasion resistant coatings have very quick hard dry times and require hot twin feed airless
spray application.
5.3
Cathodic protection
Surkein et al. (2009) presented a review of offshore vessel and fixed structure CP (cathodic protection)
design, contrasting the effectiveness of impressed current and sacrificial anode designs for the hull. They
concluded that a galvanic anode is superior to ICCP (impressed current cathodic protection) for stationary
FPSs except in cases where an FPS is relatively small or has a fairly short design life.
Sun et al. (2013) studied the corrosion behaviour and working performance of AlZnInMgTi sacrificial anode in simulated deep water environment. They found that the corrosion of AlZnInMgTi
was accelerated by the cathodic process, and the efficiency dropped dramatically in deep water, and at
least 22% more AlZnInMgTi anode should be required in deep water environment.
5.4
This specification covers structural steel plates, shapes, bars, and rivets for use in ship construction. Materials under this specification can be categorized as ordinary strength and higher strength. Plates in all
thicknesses shall be normalized or thermo-mechanical control processed while shapes and bars in all
thicknesses shall be heat treated and rolled. Heat analysis of ordinary strength structural steel shall be
used to determine the required chemical composition for carbon, manganese, phosphorus, sulphur, nickel,
chromium, molybdenum, silicon, and copper. Same analysis shall be used to higher strength structural
steel to determine the required chemical composition for carbon, manganese, phosphorus, sulphur, silicon, vanadium, aluminium, titanium, nickel, chromium, molybdenum, silicon, copper, nickel, and niobium. Materials shall conform to the required metallurgical structure which shall be evaluated by determining the average grain size. Mechanical properties such as elongation and toughness shall be evaluated
using tension test and Charpy V-notch impact test.
The ASTM Spec. covers the requirements of ship steel forms as below, (ASTM-A131/A131M-14,
2014):
1.1 This specification covers structural steel plates, shapes, bars, and rivets intended primarily for use
in ship construction.
1.2 Material under this specification is available in the following categories:
1.2.1 Ordinary StrengthGrades A, B, D, and E with a specified minimum yield point of 34 ksi [235
MPa], and
1.2.2 Higher StrengthGrades AH, DH, EH, and FH with a specified minimum yield point of 46 ksi
[315 MPa], 51 ksi [350 MPa], or 57 ksi [390 MPa].
1.3 Shapes and bars are normally available as Grades A, AH32, and AH36. Other grades may be furnished by agreement between the purchaser and the manufacturer.
1.4 The maximum thickness of products furnished under this specification is 4 in. [100 mm] for plates
and 2 in. [50 mm] for shapes and bars.
1.5 When the steel is to be welded, it is presupposed that a welding procedure suitable for the grade
of steel and intended use or service will be utilized. See Appendix X3 of Specification A6/A6M for
information on weldability.
1.6 The values stated in either inch-pound units or SI units are to be regarded separately as the standard.
Shiotani et al. (2012) have presented a new development for superior corrosion resistant steel for
ballast tank application. First, the actual environment at the upper deck in a ballast tank of a bulk
648
carrier was measured. The upper deck environment was characterized by cyclic repetition of water
evaporation and condensation accompanied with daily temperature increases and decreases. The corrosion resistance of steels was evaluated by the cyclic corrosion test with salt spray using conditions
found from the environmental measurement results. The degradation area of coating films at a
scratch in the developed steel was about 70% that in the conventional steel. The fine rust particles
which formed under the coating film on the developed steel were considered to contribute to its excellent corrosion resistance. The mechanical properties of the developed steel, such as tensile properties and Charpy impact properties satisfied the KE36 specification. The effect of corrosion on the
structural reliability of steel offshore structures has also been recently evaluated by NKK as a guideline, (Melchers, 2005).
5.5
Corrosion Monitoring
To reduce labour cost and increase the quality of edge preparation prior to coating, German research
projects developed technologies for thermal edge rounding by means of laser of plasma sources. This
method was pre-approved by DNV of structural shipbuilding steels, (Jasnau et al., 2011). Within this
project, various edge geometries and preparation technologies have been systematically investigated.
Results showed that there is no statistical evidence that a preparation of edges with a 2mm radius will
improve corrosion protection of the edges compared to alternative edge geometries, (Heinemann &
Roland, 2011).
5.6
Non-destructive testing (NDT), also referred to as NDE (non-destructive evaluation/examination) and

NDI (non-destructive inspection), is a family of specialized technical inspection methods that provide
information about the condition of materials and components without destroying them, (DNV, 2014).
Weld Surface Appearance Welding in hull construction is to comply with the requirements of Section 24-1 Hull Construction of the ABS Rules for Materials and Welding (Part 2) and IACS Recommendation
No. 47 Shipbuilding and Repair Quality Standard. Methods used for preparing and cleaning welds and
non-destructive test procedures are to be to the satisfaction of the Surveyor.
5.6.1
Visual Inspection of Welds:
Welds are to be visually inspected to the satisfaction of the Surveyor. Visual inspection acceptance criteria are contained in Section 8 of the Guide, (ABS, 2011).
Visual inspections of welds may begin immediately after the completed welds have cooled to ambient
temperature. However, delayed cracking is a concern for extra high strength steels, 415 N/mm2 (42
kgf/mm2, 60,000 psi) yield strength or greater. When welding these high-strength steels, the final visual
inspection shall be performed not less than 48 hours after completion of the weld and removal of preheat.
Refer to 1/1.5 below for requirements for delayed cracking inspection.
5.6.2
Inspection for Delayed (Hydrogen Induced) Cracking
Non-destructive testing of weldments in steels of 415 N/mm2 (42kgf/mm2, 60,000 psi) yield strength or
greater is to be conducted at a suitable interval after welds have been completed and cooled to ambient
temperature.
Minimum 48 hours of interval time for steels of 415 MPa (42 kgf/mm2, 60,000 psi) yield strength or
greater but less than 620 MPa (63 kgf/mm2, 90,000 psi) yield strength.
Minimum 72 hours of interval time for steel greater than or equal to 620 MPa (63 kgf/mm2, 90,000 psi)
yield strength.
At the discretion of the Surveyor, a longer interval and/or additional random inspection at a later period
may be required. The 72 hour interval may be reduced to 48 hours for radiography testing (RT) or ultrasonic testing (UT) inspection, provided a complete visual and random MT or PT inspection to the satisfaction of the Surveyor is conducted 72 hours after welds have been completed and cooled to ambient
temperature.
When delayed cracking is encountered in production, previously completed welds are to be reinspected for delayed cracking to the satisfaction of the Surveyor. At the discretion of the Surveyor,
re-qualification of procedures or additional production control procedures may be required for being free
of delayed cracking in that production welds.
5.6.3
649
Methods of Inspection
Inspection of welded joints is to be carried out by approved non-destructive test methods, such as visual
inspection (VT), radiography (RT), ultrasonic (UT), magnetic particle (MT), liquid penetrant (PT), Thermal Imaging, etc.
A plan for non-destructive testing is to be submitted. Radiographic or ultrasonic inspection, or both, is
to be used when the overall soundness of the weld cross section is to be evaluated Magnetic particle or
liquid penetrant inspection or other approved method is to be used when investigating the outer surface of
welds or may be used as a check of intermediate weld passes such as root passes and also to check back
gouged joints prior to depositing subsequent passes.
Surface inspection of important tee or corner joints in critical locations, using an approved magnetic
particle or liquid penetrant method, is to be conducted to the satisfaction of the Surveyor. Where a method
(such as radiographic or ultrasonic) is selected as the primary non-destructive method of inspection, the
acceptance standards of that method govern. However, if additional inspection by any method should
indicate the presence of defects that could jeopardize the integrity of structure, removal and repair of such
defects are to be to the satisfaction of the Surveyor. Welds that are inaccessible or difficult to inspect in
service may be subjected to increase the levels of non-destructive inspection.
The extent and locations of inspection and selection of inspection method(s) are to be in
accordance with (ABS, 2011):
The applicable ABS Rules;

The material and welding procedures used;
The quality control procedures involved;
The results of the visual inspection, and
The discretion of the Surveyor;
5.6.4
Under film corrosion detection
Osawa et al. (2014) have studied the effectiveness of fluorescent ferric ion indicator for under film corrosion detection of coated steel panels in water ballast tank. (1S)-3'-(diethylamino)-N,N-diethyl-3-oxo3',9a'-dihydro-3H-spiro[isobenzofuran-1,9'-xanthen]-6'-amine oxide RhoNox-1, has been chosen as
Fe2+ indicator.
Under film corrosion sensing semi-transparent modified epoxy films, which contains RhoNox-1, are
prepared in the process. SS400 steel test panels are blasted and the prepared semi-transparent Fe2+ sensing paints are applied. Some panels are scribed after coating, and NaCl solution is dropped onto the steel
surface before coating in other panels. These panels are subject to immersion tests in NaCl solution and
cyclic corrosion tests by using artificial seawater. Changes in fluorescence are observed during these
tests. The following behaviour is noted:
1) RhoNox-1 is not responsive to amine-based hardener, and "prematurely fluorescent can be prevented; 2) RhoNox-1is functioning as an early detection indicator for steel corrosion; 3) Steel's anode
reactions in underfilm corrosion that progressed during the cyclic corrosion tests had been detected clearly by using the developed Fe2+ sensing paint. 4) It is enough to apply 1wt% RhoNox-1 containing paint
as a mist coat for non-destructive monitoring of anode reaction when semi-transparent paint used in this
study is applied as a top coat.
6.
MANUFACTURING SIMULATION
Nowadays the shipbuilding industry is facing a high competitive environment due to production overcapacity and competitors that are applying hostile low labour costs. Under these circumstances, the optimization of design, production and operation processes are vital to keep and increase their competitive power in global extent.
In recent years, engineering departments of shipyards and universities have been attempting to
tackle this problem through the:
improvement of the quotation and design steps in order to quickly provide an optimum solution in
term of structure, production and cost,
minimization of costs and production time through automation and improved schedules,
Maximization of product quality.
To overcome this gap, strong emphasis has recently been placed on the development of Computer Integrated Manufacturing (CIM) systems to support the design and production of ships by linking the
650
design system with the production support system. Indeed, progress in these fields has led to the improvement of quality and accuracy, which are two essential preliminaries for effective production.
Amongst CIM methodologies used in shipbuilding industry, recent developments about Discrete Event
Simulation (DES), production optimization, Virtual Reality (VR) and Augmented Reality (AR) are presented in the next sections.
The main advantage of these CIM tools is that they allow anticipating possible problems in the production and operation phases. This can thus provide a virtual prototype without the costs of a physical model
and available from an early design stage.
One of the major problems in the shipbuilding processes simulation and scheduling today is the lack of
interoperability between data systems and software applications, (Lee et al., 2014a). Ship data is complex
and stored throughout multiple applications that do not automatically interface with each other. The data
must then be manually integrated by gathering information from multiple sources and verifying individual
results. Lee et al. (2014a) states in his paper that it is tedious and the most time consuming part of that
process.
However, Enterprise Resource Planning (ERP) systems have been developed for shipbuilding and are
currently playing a major role in optimizing resources in a shipyard's supply chain, value chain and information chain. These systems also link the necessary ship data to help decision makers retrieve information and make educated management decisions.
6.1
Discrete Event Simulation and production optimization
Discrete Event Simulation (DES) is a simulation which describes the target system with state variables,
transition functions and time advance function. Its operation is carried out by changing its states with the
response to event occurrence. Discrete Event Simulation (DES) only takes points in time (events) into
consideration. Such events may, for example, be an element entering a station or leaving it, or moving on
to another machine. Any movements in between have little interest for the simulation itself. What is important is that the entrance and the exit events are displayed correctly. When the element enters a material
flow object, the software calculates the time until it exits that object. Finally, the simulation software
makes a list of all the important events where each event is programmed and executed step by step. DES
programs allow the mobilization of virtual plants or supply chain such as shipyards where product data
contains all geometrical and methodical information about the ships and offshores structures while the
simulation model includes all parameters describing the production facilities, resources (machines, humans, etc.) and processes. One of the major advantages of the DES is that it is possible to integrate the
operating rules of each workshop or transportation activity and simulate the complex interactions between
the different actors (human and material resources, transportation, machinery and tools, etc.). The DES is
particularly effective to tackle phenomena such as the surface management, transport management, flow
management (identification of bottlenecks), management of failures and hazards that a simple analytic
workload simulation cannot integrate.
Creating a DES model, based on generic process descriptions and properties, takes a considerable effort and includes a wide range of potential configurations. This is acceptable for factory design and layout
planning simulations for example, where the cost of creating the model accounts only for a small part of
the total investment and where the involvement of trained experts can be easily afforded. However, when
it comes to reflecting actual shipyard configurations and processes on a day to day basis including capturing changes over time (production planning) this effort has been found to be quite high for production
planners and engineers. Major shipyards in Europe and Korea (Lee et al., 2014a) are able to afford
specialists focusing on these tasks. Both, layout planning and production planning are presenting recent
developments. Recent research activities about DES have been observed and detailed in the following
sections.
6.1.1
Layout planning
A layout planning DES has been recently developed by Kim et al. (2013). A stochastic shipyard simulation has been introduced to evaluate the capacity of executing master production schedules in a shipyard
as it simulates Integrated Hull and Outfitting Plan. The results highlighted the production problems such
as capacity shortages and throughput of each resource. Then, the production capacity of each facility is
evaluated. Similarly, Lee et al. (2014a) presented recently a simulation based manufacturing in order to
check the feasibility of master production schedule of big shipyards. He advocates that this type of simulation avoid strategic errors at the upper stage of the planning and strongly reduce the risks of having bad
surprises during the manufacturing of complex products.
Song et al. (2010) developed an interesting simulation approach to select the best shipyard layout alternative amongst 70th possible generated arrangements. Each arrangement represents a scenario which will
651
be assessed in term of efficiency by simulation. Then the best alternative is selected based on key performance indicators such as working volume per day, work in progress, required space, etc.
6.1.2
Production planning
Goo et al. (2013) presented a layered ship production scheduling system based on DES. He introduced a
Layered DES framework for simulation with mixed level of information for shipbuilding in order to better take into account the dissimilarities between the intermediate products in shipbuilding industry. The
model is divided by level of details of usable information (layer) and it is designed to facilitate crosslinked information between different layers.
Ozkok and Helvacioglu (2013) presented a case study of a DES applied to the production of double
bottom blocks. He compared various improvement strategies by production simulation and has demonstrated that simulation were able to largely improve the throughput of the shipyard workshops (about
100% for the best strategy). Thereafter a list of production recommendations is proposed to improve the
productivity of the shipyard. Similarly, Chen et al. (2013) demonstrated the feasibility to use DES on
block production line of a small shipyard size in order to promote the lean manufacturing. He demonstrated that simulation is able to provide a powerful decision tool in order to improve the planning by identifying bottlenecks, poor space utilization and resource utilization.
However the most promising developments regarding DES concern the coupling of simulation with
optimisation algorithm. Tokola et al. (2013) recently introduced a DES model coupled to a local search
heuristic optimization algorithm. The objective function is focus on minimizing the lead-time of the erection process of a suezmax tanker using various erection strategies such as aft-to-fore, bottom-up and pyramid. Then, simulation is used to study how much benefit can be obtained by the optimized sequencing
in different capacity conditions.
Another issue that has shown a regain of interest during the past years is the scheduling and optimization of shipyard space allocation. Today, shipyards change their design method in order to increase the
number of simultaneous tasks with the use of more structural blocks (modular construction strategy). The
assembly of big elements requires a necessary available area within the fabrication workshop to perform
the production. As the blocks become larger and heavier, production space in the shipyard becomes a
constraint. Therefore it is important to properly use and optimise this space avoiding unnecessary moves
that results in non-value-added costs. Cipriano and Barreto (2014) recently proposed a resource constrain
project scheduler in order to solve that problem. However he deals with a limited number of production
constraints, which hardly reflect the realistic production situation. Conversely, Caprace et al. (2013) developed a heuristic algorithm solver efficiently generate and compare multiple space allocation alternatives in a reduced time with the ultimate goal of maintaining the critical ship erection schedule. He shows
on a real scheduling case (about 150 blocks during 1.5 year) that an improvement of 12% can be reached
between a schedule obtained by the solver and a manual schedule made by an experienced planner.
6.1.3
Outfitting and customization
Outfitting, and more specifically, piping seem to have a renewed interest among researchers these past
years probably because the assembly work of pipe unit is currently carried out by skilled workers, guided
by complicated two dimensional drawings, and with their long experiences. This is also a high interest
topic for offshore industry (offshore platforms, FPSOs, etc.) and naval industry (submarines) where outfitting cost is higher than in other segments.
Complexity of pre-outfitting and outfitting structures, delays in final design delivery and unpredictable
events on shop floor trigger scheduling conflicts that arise because of compartment space shortages, resource constraints and unfeasible assembly sequences. The concept of 4D planning have been recently
introduced by Bouvet (2013). 4D planning stands for 4-Dimension Planning and enables to display the
progress of an assembly sequence step-by-step. It has been developed primarily to reduce the cost of rework during on-board outfitting. A large part of rework is due to unexpected lack of space in crowded
compartment where outfitting work has already started. When a piece of equipment cannot be fitted because too many items are already installed, disassembly work is required and that will cause unexpected
delays. This kind of rework can be avoided by getting equipment fitted right the first time, following a
feasible sequence validated beforehand through 3D simulation.
In addition, Wei et al. (2010) and Wei and Nienhuis (2012) proposed two methodologies to generate
automatic schedules for outfitting processes within shipbuilding and offshore industry. She concludes that
further research and developments should be performed to obtain more practical and realistic outfitting
schedules algorithms. Noda et al. (2013) recently proposed an improvement of the automatic scheduling
system of outfitting process planning provided Wei et al. They introduce a new system which divides the
pipes into several groups considering assembling order that skilled workers adopt. Then concluded that
652
the work efficiency is improved comparing the existing solution without degrading the on-site assembling
flexibility.
To improve the shipbuilding system efficiency in the presence of variation (unexpected delays), a twostage strategic-level outfitting planning model has been developed by Dong et al. (2013a). The results of
the model provide the optimal percentage of outfitting work that should be completed at each stage of the
assembly process (on-board outfitting, pre-outfitting, etc.). A closed formed equation for system cycle
time is provided by applying Kingmans equation from queuing theory. Using the closed form equation,
the optimal percentage of total outfitting work processed at each stage can be calculated given any
scenario.
A long-term scheduler module applied to the outfitting cabin installation for cruise ships have been recently developed by Sasaki et al. (2013). This simulation module focuses only on the bottleneck processes
and uses a constrain-based scheduling system. Sasaki states that the developed approach was useful to
find work improvement plan to shorten the lead time of the whole cruise ship project.
6.1.4
Logistic simulations
Discrete-event simulation has been used as an analysis tool to evaluate the new production system concepts and has also been used in the operation and planning of manufacturing facilities. However, an important aspect that could also be analysed with this method is the supply chain and intern logistic of shipyards.
Woo et al. (2010) developed a logistic based simulation to improve the decision-making system for
ship blocks production of a big size shipyard. This system verifies if the resource allocation and planning
for the block stock area, the transporter, transportation jigs, and roads that come with logistics flow of the
yard are suitable for the established production planning. Both, logistic transport and storage areas are
modelled using DES. The authors made a comparison between two logistics alternatives and concluded
that production simulation as largely improving the productivity of the shipyard. They state that the simulation can be used for both long-term and short-term planning decisions. Similarly, Solano-Charris and
Paternina-Arboleda (2013) made a sensitivity analysis on a DES model of the supply chain of a naval
shipyard. This model supports the operations planning and the decisions on capacity in naval shipyards.
6.2
Virtual and Augmented Reality
Further step to production simulation through DES would be realization of the virtual shipyard by using
of both virtual reality and augmented reality (von Lukas et al., 2014).
Virtual Reality (VR) is defined as an immersive and interactive computer simulation of real or imagined world that are 3 dimensional and calculated in real-time. Most current VR environments are primarily empirical experiences, displayed either on a computer screen or with special stereoscopic displays,
and some applications include additional sensory information such as sound, smell, touch (haptic), etc.
The VR application applied to shipbuilding industry is not new, but it is more extended now thanks to
the important improvements in software and hardware. From a user point of view, it is possible now to
find a wide range of solutions to meet the most demanding requirements, to achieve measurable results, in
terms of efficiency and costs, (Fernandez & Alonso, 2015) and (Alonso et al., 2014).
Maybe the engineering department of shipyards is where it is more current to use this kind of solutions.
And production department is maybe where it should be extended because is where there are the most
costly errors. They are not usually working with a VR solution but now it is clear that it is very useful to
check the model in order to avoid interferences and collisions between parts, schedule and sequence errors and any inconsistencies. VR can also be use to improve the ergonomic aspect during construction and
operation of the ships. At the end, the great impact in cost reduction is possible thanks to the early error
detection, being much more expensive the modifications in manufacturing and production stages.
VR can also improve the assembly sequence planning by providing a tool to check the robustness of
the schedule in terms of ergonomy and interferences. This is especially interesting for outfitting of complex marine structures where delays and re-planning are frequent. A prototype coupling VR to assembly
sequence planning have been recently developed by Lodding et al. (2011). He shows that to simplify the
use of VR in shipbuilding industry the data preparation effort should be greatly reduced in order to be
profitable. These prototypes further supports the finding and verification of assembly sequences by providing necessary information, e.g. about the next part to assemble. Also an automatic model preparation
for collision control is procured.
Friedewald and Schleusener (2014) studied the coupling of Physics Engines (PE) to VR simulations in
the research project Power-VR. Results show that there is an interest to use PE in VR to improve the
quality of simulations in the case of suspended loads at sea. For instance the movement of a safety boat
suspended to the davit before launching can be simulated considering the swell. Similarly, Ham et al.
(2014) and Ha et al. (2014) developed a simulation of the motion of the erected block and floating bodies
653
on the multibody system dynamics. Then, the calculated motion and wire rope tension are used for dynamic effect estimation. Various applications are presented such as, a block turnover simulation by gantry
crane, a block turnover simulation by floating crane, a flare tower turnover simulation, a crane transportation simulation by two synchronised floating cranes, connexion between two floating structures at sea and
a life raft launching simulation. In order to reduce the tedious time data preparation of this kind of simulations, Ku et al. (2014) developed a scenario management of multibody dynamic simulator. He presented a
prototype with a test case on lifting blocks using cranes. Due to his modular behaviour this system drastically reduced the modelling time.
Augmented reality (AR) system combines the real world with virtual information and provides expanded and information-rich view of the environment to the user. Recent advances in technology, lower
cost of otherwise expensive equipment and especially development of the smart handheld-device market
have propelled the rapid development of the AR. Head Mounted Displays (HMDs) are probably the most
promising because they provide the best immersive experience. HMDs are worn on the users head and
have a small display device in front of each eye presenting virtual objects or information over the view of
the real world. As stated by Vasilijevic et al. (2011), many fields already benefit from AR technology and
there is actually some developments in maritime industry such as in ship construction, electronic navigational aids, maintenance assistance, diving assistance, etc.
7.
7.1
WELDING SIMULATION
Computation Welding Mechanics
The exponential growth in computer performance combined with equally rapid developments in numerical methods and geometric modelling have enabled Computational Weld Mechanics (CWM) to reach the
stage where it can solve an increasing number of problems that interest the shipbuilding and offshore
industries, (Goldak and Akhlaghi, 2006), (Lindgren, 2007) and (Babu, 2010).
In a scientific context Finite Element (FE) based on thermo-mechanical simulations to determine the
temperature and stress fields present during welding have been performed for about forty years. The FE
technology and the computer capacity (computations and storage) have developed significantly in the
recent years, and now allow including detailed weld process parameters in the simulations.
Recently published CWM standards (AWS, 2013) indicates that CWM is in the progress to become an
established and mature engineering process in the field of Residual Stresses and Distortion Predictions
(RSDP) (Radaj, 2003). It also indicates that CWM analyses can be used as a fairly reliable tool at the
assessment of Welding Procedure Specifications (WPS) and proposed weld process parameters at the
manufacturing engineering stage. It has also been reported that CWM analyses has been used to optimise
the effects of welding induced residual stresses (WRS) at the structural integrity design phase (Cox,
2011), (Goldberg and Marshall, 2010) and (Pahkamaa et al., 2012).
Finite Element Analyses (FEA) of arc weld processes is quasi-dynamic multi physical problems that
generally require High Performance Computer (HPC) with Massively Parallel Processors (MPP).
To be accurate, welding simulation should consider carefully the use of realistic material properties and
welding heat source, as described in the next sections.
7.2
Arc Welding Simulation Methodologies
Arc weld processes are characterised by transient thermal and dynamic material phenomena of the interacting weld and base material. During the weld metals solidification and contraction process the weld
melt pools surrounding base material is first expanding and thereafter contracting.
7.2.1
Sequentially coupled thermos-mechanical models
Finite Element weld simulations are commonly executed by the use of sequentially coupled thermomechanical models, (Brickstad & Josefson, 1998), (Murnsky et al., 2012) and (Fu et al., 2014). It implies
that first is the entire transient thermal simulation solved and secondly is the transient mechanical
simulation solved by the use of the temperature field solution obtained from the thermal simulation, see
Figure 15.
7.2.2
Thermo-mechanical staggered coupled
Thermo-mechanical staggered coupled may also be used. There the Finite Element weld simulations are
executed as Non-linear transient thermo-mechanical staggered coupled implicit FEA (Lindgren, 2007).
It implies that the transient thermal and mechanical simulations are solved parallel to each other. There
the two FEA-processes provide essential input data to each other, see Figure 16. At the end of the process,
transient temperatures, stresses, thermal strain, elastic strain and plastic strain are provided. It should be
noted that a computational time reduction of about 85 % compared to ABAQUS for a fully 3D transient
thermal TIG-bead on plate benchmark simulation has been reported (Lindstrm, 2013).
654
Figure 15. Schematic description of a sequentially coupled thermo-mechanical Finite Element weld simulation (Fu
et al., 2014)
Figure 16. Schematic description of a staggered coupled thermo-mechanical Finite Element weld simulation
(Lindstrm, 2015)
7.3
Heat source models
Because the welding process consists of a highly coupled thermal-mechanical processes, accurate simulation of the thermal transfer analysis in the welding process is an important pre-requisite to ensure reliability in numerical simulation. Over recent decades, several analytical models have been developed to
determine an appropriate heat source model for estimating the temperature fields.
Rosenthal (1941) firstly developed the travelling instant point heat source model to research the temperature fields in welding process. Eagar and Tsai (1983) modified the Rosenthal's model including 2D
Gaussian distributed heat source model.
More recently Josefson (1983), a square shaped 2D transient equivalent heat flux density function was
used that 15 years later was (Brickstad & Josefson, 1998) modified to the shape of an triangle. This approach was adopted by TWI, after a modification by (Wei & He, 2010) and by DNV GL after additional
modifications (Lindstrm, 2013).
In another approach formulated by Dong and Hong (2003) adopted by the Swedish Radiation Safety
Authority (2009) (Zang et al., 2009a), (Zang et al., 2009b), (Mullins & Gunnars, 2009) and in continuous
use by Dong et al. (2014) the 2D weld heat flux is applied in the form of a constant liquid weld metal
temperature during the time it takes for the weld head to pass by the cross-section plane. However, this
approach has been reported (Lindstrm, 2015) to result in a weld heat flux distribution with a fairly unrealistic shape. Figure 17 compare five different method to distribute the same amount of weld heat energy (Q = 6.86 MJ) as a 2D heat flux over time relative to the weld cross-section.
In order to obtain the temperature field as realistic as possible, some volumetric heat source model was
developed. One of them has been considered as 3D continuous double ellipsoidal moving heat source
model (Goldak et al., 1984). This model commonly denoted Goldak weld heat source is still a reference
for numerous welding simulation code (Shapiro, 2003) and (Oddy et al., 1990).
655
Later, a number of numerical and analytical models were developed to predict temperature fields in
welding process based on these heat source models. Gery et al. (2005) and Shanmugam et al. (2010) investigated the effects of welding parameters on the temperature distribution in a single butt welding and
T-joints welds, respectively. Attarha and Sattari-Far (2011) investigated the temperature fields in thin
welded plate through the experimental measurements and numerical simulations based on the double
ellipsoidal heat source model. Winczek (2010) and Winczek (2011) proposed analytical solutions to model the temperature fields in a half-infinite body during welding process.
More recently Fu et al. (2015) presented a neural network model developed to predict the four double
ellipsoidal weld heat source starting from the Voltage, Amperage and Speed of the welding machine. The
main advantage of the process is a reduction of the computation time required to setup the parameters of
equivalent heat flux of big simulation models.
Figure 17. Diagram illustrating how five different methods are distributing the same amount of weld heat energy (Q
= 6,86 MJ) as a 2D heat flux over time relative to the weld cross-section (Lindstrm, 2015)
Figure 18. Illustration of the double ellipsoidal weld heat source model
7.4
Material Models
Arc welding technology has been continuously improved during the last decades becoming an established
process for the manufacturing, maintenance, and repair of metal structures. Today, it is understood that
the ultimate mechanical properties of a welded joint depends upon an intricate relationship between several contributing factors, see Figure 19.
Both the thermal and mechanical material model should be designed to be used for modelling of the
base material, solidified weld material and weld pass material to be activated in a later sequence of the
multi-pass weld simulation process. There the mechanical material model is a non-linear thermo-elasticplastic model that can employ a mixed hardening with a range from 100% isotropic hardening to 100%
linear kinematic hardening that allows residual stress release, triggered by temperature.
The residual stress release function (also known as annealing function) eliminates the materials prior accumulated hardening history. It implies that the base and weld material only can retain and/or build up residual stresses at temperatures below the residual stress release temperature.
656
Weld material should be present from the very first simulation sequence in the form of quiet material
and are gradually activated during the time it takes for the thermal energy of the heat source to heat up the
quiet weld filler material between its solidus (TS) and liquidus temperatures (TL).
Figure 19. Flowchart of the intricate relation between the variables that affect the mechanical properties and distortion of a welded joint (Lindstrm, 2015)
The quiet materials represent volumes empty from metal i.e. vacuum or fluids (gas and liquid). A quiet
material stress should never reach the yield point; therefore shall the Youngs modulus (E) value be small
enough to not influence the surroundings but still be large enough to avoid numerical problems.
The quiet thermal materials initiation temperature intervals are the solidus (TS) and liquidus temperature (TL). The quiet mechanical materials initiation temperature is intended to be set a couple of hundred
Celsius degrees above the TS and TL. This to reflect that the liquid weld filler material is entering the
weld melt pool in an overheated state (Lindstrm & Ulfvarson, 2003) and thereby facilitate the capture of
thermo-mechanical contributions from the liquid, solidification and solid contraction (Campell, 2004), see
Figure 20. Above TL is the thermal conductivity value () 300 W/(mC) and between TS and TL is =
150 W/(mC) to mimic the heat and mass transfer of the weld melt pool (Belytschko et al., 2014).
Figure 20. Illustration of the liquid, solidification and solid contraction in relation to TL and TS for an austenitic
stainless steel of the type 316L, (Campell, 2004)
7.5
Thermal- and Mechanical Boundary Conditions
Simulation of weld processes there the object (weld model) is exposed to still standing air are the thermal
boundary conditions of the surfaces described by an apparent thermal convection. Describing the total
657
amount of heat transfer from the surfaces due to convection and radiation Equations 1 and 2, (Argyris
et al., 1983). This approach is recommended by the Swedish Radiation Safety Authority (Zang et al.,
2009a).
h = 0.0668 T
h = 0.231 T 82.1
20C T 500C
T > 500C
(1)
(2)
In the case a surface is exposed to a forced flow of fluid will the thermal boundary conditions be calculated by the use of the methodology presented in Lindstrm (2009).
Contacts between material surfaces are modelled as thermo-mechanical conditions by the use of a penalty
based coupling approach, with a heat transfer conductance of 75 kW/(m2C) and a static friction coefficient
of 0.15.
7.6
Mesh size
At thermo-mechanical FEA is elements of the 1st order type used in one and the same mesh for both the
thermal- and mechanical simulations. There the polynomial describing the total strain field is one order
lower than the polynomial of the thermal strain field. Anyhow, by the use of a very fine mesh (small mesh
size) in areas with steep temperature gradients i.e. the Fusion Zone (FZ) and the Heat Affected Zone
(HAZ) inconsistent stress and strain results are avoided (Oddy et al., 1990).
Element size in the way of the weld joint i.e. the weld metal and about 30 mm base material on each
side of the weld metal shall have a very fine and preferably square shaped mesh. The element size should
for the 1st order element type be about 0.3 0.5 mm. And the preferred element side length ratio is 1:1:1
and in worst case the element side length ratio of 1:2:3 may be used. At distance of about 50 mm from the
fusion line or the regions of interest the element size can be rapidly increased. A typical mesh used at
parametric 2D CWM analyses of a critical ship hull plate weld joint is presented in Figure 21.
(a) Complete view of the FEM model where the

dimensions of the left plate is 300x200x30 mm
and the dimension of the right plate is
300x200x20 mm
(b) Local view of the FEM model around the butt

welded joint
Figure 21. 3D transient thermal butt weld model made up of 131716 solid elements and 140793 nodes (Lindstrm,
2013)
7.7
The computational cost of a CWM analysis depends on the total number of degrees of freedom (DOF),
the transient time and the amount of physical processes that are simulated.
Lindstrm (2015) mentioned that the computational time for fully 3D CWM analysis constituting of
2142782 DOF (thermal and mechanical) costs about 8 000 CPUh whilst the 2D CWM analysis (5227 DOF)
of the same problem constitutes of costs about 2 CPUh. On the other side can the cost of the fully 3D CWM
analyses be reduced to 4 000 CPUh if the use of symmetry boundary conditions is feasible.
7.8
Parallel to the development of the CWM technology there has been research activities related to the measurement of residual stresses in metallic materials. Residual stress measurement methods fall into one of
the following two categories destructive or non-destructive. Some destructive methods are argued to be
semi-destructive examples on that are the hole drilling and deep hole drilling method. It can be discussed
if the stakeholder is prepared to take the risk to operate a structure with an 10 mm hole drilled through
a critical component. Overviews of the most commonly used residual stress measurement methods are
presented in Schajer and Ruud (2013), Cheng and Finnie (2007) and Fitzpatrick and Lodini (2003).
658
Recently industrial standards demonstrates that the X-ray diffraction method is the most matured method
followed by the hole drill and strain gauge method, and the slitting method (ASTM-E1426-98(2009)e1,
2009), (ASTM-E915-10, 2010), (ASTM-E2860-12, 2012), (ASTM-E1928-13, 2013), (ASTM-E837-13a,
2013).
7.9
Benchmark case
The ISSC-V.3 technical committee has performed a benchmark to define a Best Practice Guideline to use
Computational Welding Mechanics tools (CWM) in shipbuilding and offshore industry. To achieve this
objective various experimental welding tests have been performed in order to give a reference point. Both
the residual welding distortions and residual stresses have been compared between numerical simulations
and welding experiments for a common T welded assembly used in the shipbuilding industry. However, it has been decided to publish the results of this study in a separate document.
The experiments were performed using the low carbon grade DH36 ferric steel plate typically employed in shipbuilding industry. The geometric model of the T-joint weld is shown in Figure 22. The
length of the T-joint is 500 mm, the width of the flange is 500 mm with a thickness of 9 mm, and the
height of the web is 300 mm with a thickness of 9 mm.
Figure 22. Welding experiment of a T joint assembly and position of thermocouples (I-M), residual stress measurements (A-F) and distortion measurements points (sections A-A, CL-CL and B-B)
Two mechanical boundary conditions have been analysed. The first one corresponds to free supported
side edges while the second one correspond to clamped side edges.
The filler material corresponds to an AWS A5.20 E71T-1C/9C/-9C-J where the electrode diameter was
1.2 mm. The Flux Core Arc Welding (FCAW) single-side one pass welding procedure is employed with
two different weld legs of respectively 5.5 mm and 7 mm. The second fillet weld is started when the temperature of specimen cools down to the ambient temperature. The welding conditions have been set-up
respectively to 184 A, 28 V and 290 mm/m for 5.5 mm weld leg and 215 A, 28 V and 230 mm/min for
the 7 mm weld leg.
Five thermocouples were mounted on the plate (at the half thickness) prior to the welding process and
connects directly to the multi-channels data logging system. In this way, the temperature during the welding process at any specified time can be conveniently stored for later data processing and comparison
with numerical results. The thermocouples (I to M) are at 20 mm, 20 mm, 30 mm, 15 mm and 15 mm
from the welding seam and 50 mm, 25 mm, 0 mm, 25 mm and 50 mm from the centreline of the plates
(CL), respectively. Thermocouples are located to the side of the first weld.
A portable Coordinate Measuring Machine (CMM) was employed to measure the deflection of the upper surface of the flange after welding. A list of target points have been defined along 3 planes (A-A, CL
and B-B) 125 mm, 0 mm and 125 mm from the centreline of the plates, respectively. The coordinates
of the target points are measured before and after weld. The deflection can be obtained by using the
659
differences of the coordinates before and after the welding in vertical direction. The angle can also be
calculated by using the coordinates of the target points on the flange and web. The angular distortion is
known as the difference of the angle before and after weld.
Residual stresses are measured by X-ray diffraction 0.2 mm below the surface of the base plate at the
center line on the opposite side of the welds. Five target points (A to F) have been defined at 10 mm,
0 mm, 10 mm, 20 mm, 50 mm and 100 mm from the weld seam. Point F corresponds to the side of the
second weld.
The preliminary results of the benchmark show some discrepancies between the various simulation
models (various simulation softwares and methodologies) and the physical experiment. Biggest difficulty is probably the calibration of the equivalent heat flux source model that required high skilled experts and standards procedures elaboration. Moreover the CWM simulation is time consuming for both
modelling and computing. Obtaining reliable material database for this kind of simulation is another
big challenge. We consider that practical use of that methodology is still reserved for big companies
and big research centres. Further research is still required to establish fully reliable procedures. However, we noted that some small software companies are really active in this domain and the gap will be
soon filled.
8.
A state of the art review of the recent research and development on materials and fabrication technology
has been carried out. Findings and recommendations for further research are summarized as follows.
There exists an increasing demand for lighter ships. This can be reached either by use of lighter materials (composites, light metals or hybrids) or lightweight design. It has been demonstrated that even
though the cost of the equivalent aluminium ship structure is 40% more than the steel structure, the
equivalent aluminium naval ship can be built within just 7.5% of the acquisition price of the steel ship due
to the cascading benefits of the lightweight material. Some recent publications showed a regain of interest
to use titanium and its alloys for ship hull materials to reduce life-cycle cost in high-speed sea transportation. However some design and production challenges are still remaining to be viable. Metal foam cores
in metal sandwiches provide a light alternative of solid steel for decks and bulkheads. It minimizes
weight, while maximising stiffness but it is still immature for practical use.
The demand for liquefied natural gas (LNG and LNG FPSO) continues to increase due to the rise of
environmental issues. A number of materials are available that are engineered specifically for service at a
temperature of -163C and lower. New researches present results of fatigue and brittle fracture resistance
of welded assemblies at cryogenic environment. High nickel (9%) alloy steel is common for the inner
shell of LNG storage tanks. However the increase of nickel price lead actually to the development of low
nickel alloy steels for these structures.
After a long use of composites in the sporting and leisure ships and an increased application in navy
vessels the time is now ready for light weight composite materials to enter the merchant shipbuilding.
Since composites tend to be more expensive in production it is necessary to look at life cycle costs. Due
to the increased application in the maritime sector the effect of sea water exposure on composites is an
area of increased interest. Recycling remains a topic of interest as well as the assessment of composite
structures and repairs using composites. Fire issues remain an important research area for composite materials. To comply with the Solas regulations research into flame retardant materials is ongoing. Due to
the increased environmental awareness research on bio-composites has been started. Bonding of
lightweight materials is still an issue. Also the connection of lightweight structural parts to conventional
ship parts remains an area of further study.
Welding has been and will continue to be the most important joining technique for shipbuilding. Residual stress and distortion associated with welding are the key issues to be overcome. Recent publications
highlight the development of automated guided robots able to weld in confined space such as double bottom of ships. The offshore platforms and pipelines require continuous operation and, thus, reliable underwater welding on-situ. Traditional arc and hyperbaric welding have been improved and new techniques such as friction processes have been further developed for underwater applications. Laser welding
is still facing important developments, and at present actual application is extending based on the rapid
development of the fibre laser. A prototype of mobile laser equipment has recently been developed. Huge
progress has been performed in the development of Friction Stir Welding for industrial application of
steel while it is traditionally used for Aluminium. New tools are now capable of producing industrially
useful lengths of welds in steel.
Line heating has been used as a method for forming curved shell plate for more than half a century.
However is strongly related to the experience and know-how of the workers. Recently, a fully-automated
line heating plate bending system was implemented in practice. Post-treatment of critical structural details
such as welded joint or plate edge can remove surface defects and, thus, allow better utilisation of high
660
strength steel reducing appearance of fatigue cracks. New studies show the influence of surface integrity
on the fatigue strength of high-strength steel used in large marine structures.
The good production quality and workmanship are the key issues for structural safety and also for
successful utilisation of new materials. The joining and fabrication processes induce unfavourable defects, which can have significant effect on the strength of the structures. Thus, the quality of production
would directly influence the strength. Different weld type and process could influence geometrical parametric of weld joint, and the stress concentration factors and fatigue lifetime could be different. Recent
studies shows that optimised weld processes are able to achieve low distortions and good weld profile,
and thus, improvement in the fatigue strength. Welding distortion hinders automation and mechanization
of shipbuilding, and also requires additional corrective work, causing worse quality and less productivity.
Vast literature exists in terms of residual stress measurement, numerical simulation of welding and its
effect on distortion, and effects of distortion on the structural strength. Concerning the effect of the residual stresses on the collapse load of the plates, it can be concluded that this stress field leads to a decrease
in the plate collapse load, when comparing with a stress free model. From practical point of view, accuracy management is important, and it should be dealt with in terms of not only the reduction and control of
welding distortion, but also the accuracy of cutting, 3D measurement and design modifications to reduce
distortions.
Corrosion remains one of the main challenges of the shipbuilding and offshore industry. The goal of
corrosion protection is to achieve safety by controlling thickness diminution. It is considered that there
are many ways to achieve this goal, and many alternatives to the IMO PSPC and novel ideas are being
researched. Recent studies have been done for the development of zinc-rich paints sometime including
nanoparticles. Extended galvanic action of the coating is generally observed. Self healing model coatings
have been developed and tested at lab scale as well as in a ballast water tank of a ferry with good success.
New developments for superior corrosion resistant steel for ballast tank application have been presented.
Simulation based production is already widely applied in many shipyards. Further step would be rationalization of the virtual shipyard including outfitting as well as overall steel construction stages, coupling
optimization with simulation, and wider application of visible tools and on-site computing such as wearable computers, RFID and augmented reality systems.
The literature review on Computation Welding Mechanics Simulation shows that there are no existing
standard procedures to construct this kind of complex models. Biggest difficulty is probably the calibration of the equivalent heat flux source model that required high skilled experts and standards procedures
elaboration. Moreover the CWM simulations are time consuming for both modelling and computing.
A benchmark to define a Best Practice Guideline to use Computational Welding Mechanics tools
(CWM) in shipbuilding and offshore industry have been realised. Results of the benchmark will be published in a separate document but presented during the congress.
REFERENCES
Abs 2011. Guide for nondestructive inspection of hull welds. American Bureau of Shipping.
Acevedo, M. S., Puentes, C., CarreO, K., Len, J. G. M., Stupak, M., Garc-A, M. N., PRez, M. &
Blustein, G. 2013. Antifouling paints based on marine natural products from Colombian Caribbean. International
Biodeterioration & Biodegradation 83, 97104.
Adam4eve 2014. Adam4eve European Project.
Ahmadi, H. & Lotfollahi-Yaghin, M. A. 2012. Geometrically parametric study of central brace SCFs in offshore
three-planar tubular KT-joints. Journal of Constructional Steel Research 71, 149161.
Ahmadi, H., Lotfollahi-Yaghin, M. A. & Yong-Bo, S. 2013. Chord-side SCF distribution of central brace in internally ring-stiffened tubular KT-joints: A geometrically parametric study. Thin-Walled Structures, 70, 93105.
Al-Abbas, F. 2010. Inspection Challenges for Offshore Facilities. International Workshop on the State of the Art
Science and Reliability of Underwater Welding and Inspection Technology.
Alexander, C. & Ochoa, O. O. 2010. Extending onshore pipeline repair to offshore steel risers with carbonfiber
reinforced composites. Composite Structures 92, 499507.
Alonso, V., Perez, R., Sanchez, L. & Transtad, R. 2014. Advantages of Using a Virtual Reality Tool in Shipbuilding. Proceeding of the 13th International Conference on Computer Applications and Information Technology in
the Maritime Industries.
Anyfantis, K. N. & Tsouvalis, N. G. 2013. Loading and fracture response of CFRP-to-steel adhesively bonded joints
with thick adherents Part I: Experiments. Composite Structures 96, 850857.
Argyris, J. H., Szimmat, J. & William, K. 1983. Finite element analysis of the arc welding process, Numerical methods in thermal problems. Proceedings of the third international conference, Seattle WA, USA. Pineridge press,
Swansea UK.
Astm-A131/A131m-14 2014. Standard Specification for Structural Steel for Ships.
Astm-E83713a 2013. Standard Test Method for Determining Residual Stresses by the Hole-Drilling Strain-Gage
Method. West Conshohocken, PA.
661
Astm-E91510 2010. Standard Test Method for Verifying the Alignment of X-Ray Diffraction Instrumentation for
Residual Stress Measurement. West Conshohocken, PA.
Astm-E142698(2009)E1 2009. Standard Test Method for Determining the Effective Elastic Parameter for X-Ray
Diffraction Measurements of Residual Stress. West Conshohocken, PA.
Astm-E192813 2013. Standard Practice for Estimating the Approximate Residual Circumferential Stress in
Straight Thin-walled Tubing. West Conshohocken, PA.
Astm-E286012 2012. Standard Test Method for Residual Stress Measurement by X-Ray Diffraction for Bearing
Steels. West Conshohocken, PA.
Attarha, M. J. & Sattari-Far, I. 2011. Study on welding temperature distribution in thin welded plates through experimental measurements and finite element simulation. Journal of Materials Processing Technology, 211, 688
694.
Aws 2013. AWS A9.5:2013Guide for Verification and Validation in Computation Weld Mechanics. AWS.
Azevedo, J., Infante, V., Quintino, L. & Dos Santos, J. 2014. Fatigue Behaviour of Friction Stir welded steel joints.
Advanced Materials Research, 891892, 14881493.
Babu, S. S. 2010. Introduction to Integrated Weld ModeModel, MMetal Process Simulation. ASM Handbook. ASM
international, 411431.
Bae, K.-Y., Yang, Y.-S. & Hyun, C.-M. 2012. Analysis of triangle heating technique using high frequency induction
heating in forming process of steel plate. International Journal of Precision Engineering and Manufacturing, 13,
539545.
Baere, K. D., Verstraelen, H., Rigo, P., Passel, S. V., Lenaerts, S. & Potters, G. 2013. Reducing the cost of ballast
tank corrosion: an economic modeling approach Marine Structures 32, 136152.
Barsoum, R. G. S. 2003. The best of both worlds: hybrid ship hhull using composites and steel. AMPTIAC, 7, 55
61.
Battelle 2014. Battelle R&D Magazine Annual Global Funding Forecast Predicts R&D Spending Growth will Continue While Globalization Accelerates.
Baumgartner, J. & Bruder, T. 2013. Influence of weld geometry and residual stresses on the fatigue strength of longitudinal stiffeners. Welding in the World, 57, 841855.
Bellotti, N., Del Amo, B. & Romagnoli, R. 2012. Tara tannin a natural product with antifouling coating application
Progress in Organic Coatings 74, 411417.
Belytschko, T., Liu, W. K., Moran, B. & Elkhodary, K. 2014. Nonlinear Finite Elements for Continua and Structures.
Besst 2013. Breakthrough in european ship and shipbuilding technologies (BESST) European Project.
Bouvet, C. 2013. 4D planning : production-scheduling optimization for hull outfitting. Proceeding of the 12th International Conference on Computer Applications in Shipbuilding (ICCAS).
Brickstad, B. & Josefson, B. L. 1998. A parametric study of residual stresses in multi-pass butt-welded stainless
steel pipes International Journal of Pressure Vessels and Piping 75, 1125.
Brief, L. 2011. Opportunities in natural fiber composites.
Buksa, T., Pavletic, D., Sokovic, M. & Buksa, J. 2013. A Differentiation-Based Approach to Quality Management
in Shipbuilding Taking into Consideration Errors in Manufacturing Processes. BRODOGRADNJASHIPBUILDING, 4, 488503.
Cam, G. 2011. Friction stir welded structural materials: beyond Al-alloys. International Materials Reviews, 56, 1
48.
Camilleri, D., Gray, T. G. F. & Mcpherson, N. 2010. Optimizing Tack Welding Fabrication Procedures Using Numerical Finite Element Models. Journal of Ship Production and DesignJSPD, 26, 117134.
Campell, J. 2004. Castings.
Cansi 2014. CANSI.
Caprace, J.-D., Petcu, C., Velarde, M. G. & Rigo, P. 2013. Optimization of shipyard space allocation and scheduling
using a heuristic algorithm. Journal of Marine Science and Technology, 18, 404417.
Cater, S. 2013. Forge welding turns full circle: friction stir welding of steel. Ironmaking & Steelmaking, 40, 490
495.
Cavas, C. P. 2013. Navy Switches from Composite to Steel.
Cerik, B. & Cho, S.-R. 2013. Numerical investigation on the ultimate strength of stiffened cylindrical shells considering residual stresses and shakedown. Journal of Marine Science and Technology, 18, 524534.
Chambers, L. D., Wharton, J. A., Wood, R. J. K., Walsh, F. C. & Stokes, K. R. 2014. Techniques for the measurement of natural product incorporation into an antifouling coating Progress in Organic Coatings 77, 473484.
Chen, B. Q., Garbatov, Y. & Soares, C. G. 2011. Measurement of Weld-Induced Deformations in ThreeDimensional Structures Based on Photogrammetry Technique. Journal of Ship Production and Design, 27, 51
62.
Chen, N., Zhenjiang, Wang, Z. & Wu, J. 2013. Simulation-Based Research on Adjustment Technology of Ship
Block Production Plan. International Conference on Information Technology and Applications (ITA).
Cheng, W. & Finnie, I. 2007. Residual stress measurement and the slitting method, Springer.
Cicala, G., Cristaldi, G., Recca, G. & Latteri, A. 2010. Composites based on natural fibre fabrics, woven fabric
engineering.
Cipriano, W. & Barreto, D. 2014. Project Scheduling Problems with Resource Constraints for Assembly in Shipbuilding. Proceeding of the 13th International Conference on Computer Applications and Information Technology in the Maritime Industries.
Co-Patch 2010. Co-patch EU project.
662
Coolantarctica 2014. Icebreakers and Ice Strengthened Ships.

Couch, W. J. C. 2010. Friction Stud Welding Testing for Submarine Rescue. International Workshop on the State of
the Art Science and Reliability of Underwater Welding and Inspection Technology.
Cox, D. 2011. Repair of the NRU Reactor Vessel: Technical Challenges and Lesson Learned. Proceeding of the
International Conference on Research Reactors: Safe Management and Effective Utilization.
Cridland, M. 2010. Impact of Material Quality Control on Underwater Welding, Inspection and Reliability. International
Workshop on the State of the Art Science and Reliability of Underwater Welding and Inspection Technology.
Crupi, V., Epasto, G. & Guglielmino, E. 2013. Comparison of aluminium sandwiches for lightweight ship structures: Honeycomb vs. foam. Marine Structures 30, 7496.
Cui, L., Yang, X., Wang, D., Cao, J. & Xu, W. 2014a. Experimental study of friction taper plug welding for low
alloy structure steel: Welding process, defects, microstructures and mechanical properties. Materials & Design,
62, 271281.
Cui, L., Yang, X., Wang, D., Hou, X., Cao, J. & Xu, W. 2014b. Friction taper plug welding for steel in underwater
wet conditions: Welding performance, microstructures and mechanical properties. Materials Science and
De Jesus, A. M. P., Matos, R., Fontoura, B. F. C., Rebelo, C., Da Silva, L. S. & Veljkovic, M. 2012. A comparison
of the fatigue behavior between and steel grades. Journal of Constructional Steel Research 79, 140150.
Deguchi, T., Mouri, M., Hara, J., Kano, D., Shimoda, T., Inamura, F., Fukuoka, T. & Koshio, K. 2012. Fatigue
strength improvement for ship structures by Ultrasonic Peening. Journal of Marine Science and Technology, 17,
360369.
Dnv 2014. Non-Destructive Testing (NDT) and Evaluation (NDE).
Dong, F., Deglise-Hawkinson, J., Oyen, M. P. V. & Singer, D. J. 2013a. Analytical Approach to a Two-stage
Queuing Network for the Planning of Outfitting Processes in Shipbuilding. Journal of Ship Production and Design, 29, 136141.
Dong, P. 2010. Fitness-for-Service Assessment of Underwater Welds in Offshore Structures. International Workshop on the State of the Art Science and Reliability of Underwater Welding and Inspection Technology.
Dong, P. & Hong, J. 2003. Recommendations on residual stresses estimate for fitness for-service assessment. Welding Research Council Bulletin.
Dong, P., Song, S., Zhang, J. & Kim, M. H. 2014. On residual stress prescriptions for fitness for service assessment
of pipe girth welds. International Journal of Pressure Vessels and Piping, 123124, 1929.
Dong, P. A., Nie, C. B., Yang, X. B., Song, S. A. & Decan, L. A. 2013b. A math-based design-for-produceability
evaluation of titanium applications in ship hull structures. TransactionsSociety of Naval Architects and Marine
Engineers, 120, 299305.
Dong, W., Moan, T. & Gao, Z. 2012. Fatigue reliability analysis of the jacket support structure for offshore wind
turbine considering the effect of corrosion and inspection. Reliability Engineering & System Safety 106, 1127.
Eagar, T. & Tsai, N. 1983. Temperature fields produced by traveling distributed heat sources. Welding Journal, 62,
346355.
Edvardsson, J. 2013. Offshore and lightweight. E-lass.
Eggert, L., Fricke, W. & Paetzold, H. 2012. Fatigue strength of thin-plated block joints with typical shipbuilding
imperfections. Welding in the World, 56, 119128.
Elass 2014. Elass EU project.
Errouane, H., Sereir, Z. & Chateauneuf, A. 2014. Numerical model for optimal design of composite patch repair of
cracked aluminum plates under tension International Journal of Adhesion and Adhesives 49, 6472.
Farajian, M. 2013. Welding residual stress behavior under mechanical loading. Welding in the World, 57, 157169.
Fernandez, R. P. & Alonso, V. 2015. Virtual Reality in a shipbuilding environment. Advances in Engineering Software 81, 3040.
Fernberg, P. 2014. Bio-based composite, an overview. throughlife final workshop 26th March.
Fitzpatrick, M. E. & Lodini, A. 2003. Analysis of residual stress by diffraction using neutron and synchrotron radiation, CRC Press.
Flemming, H.-C., Tamachkiarowa, A., Klahre, J. & Schmitt, J. R. 1998. Monitoring of fouling and biofouling in
technical systems Water Science and Technology 38, 291298.
Fricke, W., Remes, H., Feltz, O., Lillemae, I., Tchuindjang, D., Reinert, T., Nevierov, A., Sichermann, W., Brinkmann, M., Kontkanen, T., Bohlmann, B. & Molter, L. 2013. Fatigue strength of laser-welded thin-plate ship
structures based on nominal and structural hot-spot stress approach. Ships and Offshore Structures, 0, 16.
Fricke, W. & Tchuindjang, D. 2013. Fatigue strength behaviour of stud-arc welded joints in load-carrying ship
structures. Welding in the World, 57, 495506.
Fricke, W., Von Lilienfeld-Toal, A. & Paetzold, H. 2012a. Fatigue strength investigations of welded details of stiffened plate structures in steel ships International Journal of Fatigue 34, 1726.
Fricke, W., Zacke, S., KoAk, M. & Eren, S. 2012b. Fatigue and Fracture Strength of Ship Block Joints Welded
with Large Gaps. Welding in the World, 56, 3039.
Friedewald, A. & Schleusener, S. 2014. Coupling Virtual Reality and Physics Engines for Ships. Proceeding of the
13th International Conference on Computer Applications and Information Technology in the Maritime
Industries.
Fu, G., Gu, J., Lourenco, M. I., Duan, M. & Estefen, S. F. 2015. Parameter determination of double-ellipsoidal heat
source model and its application in the multi-pass welding process. Ships and Offshore Structures, 0, 114.
663
Fu, G., Lourenco, M. I., Duan, M. & Estefen, S. F. 2014. Effect of boundary conditions on residual stress and distortion in T-joint welds Journal of Constructional Steel Research 102, 121135.
Fujii, H., Cui, L., Tsuji, N., Maeda, M., Nakata, K. & Nogi, K. 2006. Friction stir welding of carbon steels. Materials Science and Engineering: A 429, 5057.
Furuya, H., Saitoh, N., Takahashi, Y., Kurebayashi, K., Kayamori, Y., Inoue, T., Uemori, R. & Okushima, M. 2011.
Development of 6% Nickel Steel for LNG Storage Tanks. 30th International Conference on Ocean, Offshore and
Arctic Engineering.
Gannon, L., Liu, Y., Pegg, N. & Smith, M. 2010. Effect of welding sequence on residual stress and distortion in flatbar stiffened plates. Marine Structures, 23, 385404.
Gannon, L., Liu, Y., Pegg, N. & Smith, M. J. 2012. Effect of welding-induced residual stress and distortion on ship
hull girder ultimate strength Marine Structures 28, 2549.
Gergely, A., Paszti, Z., Mihaly, J., Drotar, E. & Tamachkiarowa, A. S. T. 2014. Galvanic function of zinc-rich coatings facilitated by percolating structure of the carbon nanotubes. Part II: Protection properties and mechanism of
the hybrid coatings. Progress in Organic Coatings 77, 412424.
Gery, D., Long, H. & Maropoulos, P. 2005. Effects of welding speed, energy input and heat source distribution on
temperature variations in butt joint welding Journal of Materials Processing Technology 167, 393401.
Gibson, D., Paculba, N. & Grey, I. 2010. Friction Stud Welding Underwater in the Offshore Oil and Gas Industry.
International Workshop on the State of the Art Science and Reliability of Underwater Welding and Inspection
Technology.
Gl 2010. Rule for Classification and ConstructionCorrosion ProtectionGuideline for corrosion protection and
coating systems. Germanischer Lloyd.
Goldak, J., Chakravarti, A. & Bibby, M. 1984. A new finite element model for welding heat sources. Metallurgical
Transactions B, 15, 299305.
Goldak, J. A. & Akhlaghi, M. 2006. Computational Welding Mechanics.
Goldberg, L. & Marshall, P. W. 2010. The Role of Performance Demonstration Initiatives to Qualify NDT Methodologies and Inspectors for In-Service Inspections Using Risk Based Inspections Structural Integrity Management. International Workshop on the State of the Art Science and Reliability of Underwater Welding and Inspection Technology.
Gomes, J. F. P., Paula Cristina Silva Albuquerque, Rosa Maria Mendes Miranda & Vieira, M. T. F. 2012. Determination of airborne nanoparticles from welding operations. Journal of Toxicology and Environmental Health, Part
A: Current Issues, 75, 747755.
Goo, B., Choi, M. & Chung, H. 2013. A Study on Modeling and Simulation of Layered Ship Production Scheduling
System. Proceedings of the Twenty-third (2013) International Offshore and Polar Engineering (ISOPE).
Grabovac, I. & Whittaker, D. 2009. Application of bonded composites in the repair of ships structuresA 15year
service experience. Composites Part A: Applied Science and Manufacturing, 40, 13811398.
Ha, S., Ku, N., Roh, M.-I., Kim, K. S., Ham, A. S.-H., Li, X., Hong, J.-W. & Lee, H. 2014. DEVS-based scenario
manager of multibody dynamics simulator for shipbuilding production process (WIP). Proceedings of the Symposium on Theory of Modeling & SimulationDEVS Integrative (DEVS14)Spring Simulation Multiconference.
Ham, S.-H., Park, K.-P. & Lee, C.-Y. 2014. Crane lifting simulation for production planning in shipbuilding. Proceeding of the 12th International Conference on Computer Applications in Shipbuilding (ICCAS).
Hamers, G. 2013. SEASupplying, Building and Maintaining the Future.
Han, M. S., Han, J. M. & Han, Y. S. 1994. A study on the fatigue strength and Allowable Stress of Invar(Fe-36%Ni)
steel Overlap Joint Applied to Cargo containment of LNG Carrier. Journal of Korea Welding Society, 12, 102
115.
Hao, Y., Liu, F., Han, E.-H., Anjum, S. & Xu, G. 2013. The mechanism of inhibition by zinc phosphate in an epoxy
coating Corrosion Science 69, 7786.
Hausbrand, R., Bolado-Escudero, B., Dhont, A. & Wielant, J. 2012. Corrosion of flame-assisted \CVD\ silica-coated
steel sheet. Corrosion Science 61, 2834.
Heinemann, M. & Roland, F. 2011. Corrosion protection as per IMO resolution MSC.215(82) Practical experience with the per-formance of different edge preparations. Proceeding of the IMO Sub-Committee on Ship Design and Equipment (DE) 55th session. IMO.
Hellbratt, S. E. 2008. Time for light weight composite materials to enter the merchant shipbuilding. European conference on composite materials ECCM13.
Hiekata, K., Yamato, H., Enomoto, M., Oida, Y., Furukawa, Y., Makino, Y. & Sugihiro, T. 2011. Development and
Case Studies of Accuracy Evaluation System for Curved Shell Plates by Laser Scanner. Journal of Ship Production and Design, 27, 8490.
Holdsworth, R. D. & Reynolds, T. J. 2010. Review of Standards and Certification for Underwater Welding and
Inspection. International Workshop on the State of the Art Science and Reliability of Underwater Welding and
Inspection Technology.
Ingle, M., Slebodnick, P., Martin, J., Ellor, J. & Cassidy, P. 2011. High Solids Coatings Performance and Service
History. Proceeding of the NACE Internacional Corrosion Conference.
International 2014. The arctic when coatings really matter.
Jasion, P., Magnucka-Blandzi, E., Szyc, W. & Magnucki, K. 2012. Global and local buckling of sandwich circular
and beam-rectangular plates with metal foam core. Thin-Walled Structures 61, 154161.
Jasnau, U., Krause, M., Roland, F. & Buchbach, S. 2011. Technologie fur die Herstellung beschichtungsgerechter
Kanten mit einem Festkorperlaser. Schweissen im Schiffbau und Ingenieurbau, 2631.
664
Jia, C., Zhang, T., Maksimov, S. Y. & Yuan, X. 2013. Spectroscopic analysis of the arc plasma of underwater wet
flux-cored arc welding. Journal of Materials Processing Technology 213, 13701377.
Jose, J. P., Malhotra, S. K., Thomas, S., Joseph, K., Goda, K. & Sreekala, M. S. 2012. Polymer composites. 1.
Josefson, B. L. 1983. Stress Redistribution During Annealing of a Multi-Pass Butt-Welded Pipe. Journal of Pressure Vessel Technology, 105, 165170.
Kabchea, J. P., Caccesea, V., Berubea, K. A. & Thompsonb, L. 2007. Analysis of a hybrid composite/metal ship
hull structural system with removable panels. Ships and Offshore Structures, 2, 227240.
Kandola, B. & Krishnan, L. 2014. Fire performance evaluation of different resins for potential application in fire
resistant structural marine composites. Proceedings of the eleventh international symposium fire safety science.
Keit 2014. KEITKorea Evaluation INstitute of Industrial Technology.
Kim, K., Lee, J. & Hwang, H. 2013. Initial Plan of Shipbuilding Capacity Based on the Stochastic Discrete Event
Simulation. Proceedings of the Twenty-third (2013) International Offshore and Polar Engineering (ISOPE).
Kim, Y. T., Kim, T. J., Park, T. Y. & Jang, C. D. 2012. Welding Distortion Analysis of Hull Blocks using Equivalent Load Method Based on Inherent Strain. Journal of Ship Research, 56, 6370.
Kim, Y. W., Lee, J. M., Kim, M. H., Noh, B. J., Sung, H. J., Ando, R. & Matsumoto, T. 2014. An Experimental
study for fatigue performance of 7% nickel steels for type B LNG carriers. Proceeding of ASME 2014 33rd International Conference on Ocean, Offshore and Arctic Engineering.
Korach, C. S., Afshar, A., Liao, H. T. & Chiang, F. P. 2014. Challenges In Mechanics of Time-Dependent Materials
and Processes in Conventional and Multifunctional Materials. challenges in mechanics of time-dependent materials and processes in conventional and multifunctional materials.
Korhonen, E., Remes, H., Romanoff, J., Niemel, A., Hiltunen, P. & Kontkanen 2013. Influence of surface integrity on the fatigue strength of high strength steel in balcony openings of cruise ship structures. Proceedings of the
4th International Conference on Marine Structures.
Kowalczyk, K. & Spychaj, T. 2014. Zinc-free varnishes and zinc-rich paints modified with ionic liquids Corrosion
Science 78, 111120.
Krause, M., Hbler, M., Roland, D. F. & Mller, M. 2014. Efficient RetrofittingHow planning tools and reverse
engineering methodologies can improve repair shipyards performance. Transport Research Arena.
Ku, N., Cha, J.-H., Lee, K.-Y., Kim, J., Kim, T.-W., Ha, S. & Lee, D. 2010. Development of a mobile welding robot
for double-hull structures in shipbuilding. Journal of Marine Science and Technology, 15, 374385.
Ku, N., Ha, S., Hwang, H.-J., Kim, K.-S., Lee, K.-Y. & Roh, M.-I. 2014. Program of Scenario Management for
Simulation of Lifting Block Using Crane. Proceedings of the Twenty-fourth (2014) International Ocean and Polar Engineering Conference.
Lackieren, B. 2014. Aktuelle Forschungsprojekte im Fokus: ADAM4EVE bringt die Entwicklung intelligenter
Lacke voran.
Laitinen, R., Valkonen, I. & Komi, J. 2013. Influence of the base Material Strength and Edge Preparation on the
Fatigue Strength of the Structures Made by High and Ultra-high Strength Steels Procedia Engineering 66, 282
291.
Lamb, T., Beavers, N., Ingram, T. & Schmieman, A. 2011. The Benefits and Cost Impact of Aluminum Naval Ship
Structure. Journal of Ship Production and Design, 27, 3549.
Le Quilliec, G., Lieurade, H.-P., Bousseau, M., Drissi-Habti, M., Inglebert, G., Macquet, P. & Jubin, L. 2013. Mechanics and modelling of high-frequency mechanical impact and its effect on fatigue. Welding in the World, 57,
97111.
Lee, D. K., Shin, J. G., Kim, Y. & Jeong, Y. K. 2014a. Simulation-Based Work Plan Verification in Shipyards.
Journal of Ship Production and Design, 30, 4957.
Lee, J. S., Kim, K. S., Kim, Y., Yu, C. H., Park, J. & Ho Kang, B. 2014b. Fatigue Strength Assessment of High
Manganese Steel for LNG CCS. Journal of the Society of Naval Architects of Korea, 51, 246253.
Lee, K. S., Eom, D. H. & Lee, J.-H. 2013. Deformation Behavior of SS400 Thick Plate by High Frequency Induction Heating Based Line Heating. Metals and Materials International, 21419, 315328.
Lee, K. S. & Hwang, B. 2014. An approach to triangular induction heating in final precision forming of thick steel
plates. Journal of Materials Processing Technology 214, 10081017.
Lee, Y. H. 2012. Prediction of Plate Bending by Multi Divisional Analysis in Induction Heating. Journal of Ship
Research, 56, 146153.
Leitner, M., Stoschka, M. & Eichlseder, W. 2014. Fatigue enhancement of thin-walled, high-strength steel joints by
high-frequency mechanical impact treatment. Welding in the World, 58, 2939.
Lienert, T. J., W. L. Stellwag, J., B. B. Grimmett & Warke, R. W. 2003. Friction Stir Welding Studies on Mild
Steel. SUPPLEMENT TO THE WELDING JOURNAL,, 82, 19.
Lillemae, I., Lammi, H., Molter, L. & Remes, H. 2012. Fatigue strength of welded butt joints in thin and slender
specimens International Journal of Fatigue 44, 98106.
Lillemae, I., Remes, H. & Romanoff, J. 2013. Influence of initial distortion on the structural stress in 3 mm thick
stiffened panels. Thin-Walled Structures 72, 121127.
Lindgren, L. E. 2007. Computational Welding Mechanics.
Lindstrm, P. R. M. 2009. Heat Transfer Prediction of In-Service Welding in a Forced Flow of Fluid. ASME Journal
of Offshore Mechanics and Arctic Engineering, 131, 03130410313046.
Lindstrm, P. R. M. 2013. DNV Platform of Computational Welding Mechanics. IIW Annual Assembly.
Lindstrm, P. R. M. 2015. Improved CWM platform for modelling welding procedures and its effects on structural
behaviour. PhD thesis, University West, Trollhttan, Sweden.
665
Lindstrm, R. M. & Ulfvarson, A. 2003. An Experimental Rig for Verification of the Mechanical Properties of
Weld Produced at In-Service Welding. ASME 2003 22nd International Conference on Offshore Mechanics and
Arctic Engineering, Volume 3: Materials Technology; Ocean Engineering; Polar and Arctic Sciences and Technology; Workshops, 5765.
Liu, H. J., Zhang, H. J. & Yu, L. 2011. Effect of welding speed on microstructures and mechanical properties of
underwater friction stir welded 2219 aluminum alloy. Materials & Design, 32, 15481553.
Liu, S., Olson, D. L., Else, M., Merritt, J. & Cridland, M. 2010. Proceeding of the International Workshop on the
State of the Art Science and Reliability of Underwater Welding and Inspection Technology. International Workshop on the State of the Art Science and Reliability of Underwater Welding and Inspection Technology.
Lodding, H., Friedewald, A., Heinig, M. & Schleusener, S. 2011. Virtual Reality Supported Assembly Planning in
the Shipbuilding Industry. Journal of Ship Production and Design, 27, 146152.
Mahoney, M., Nelson, T., Sorenson, C. & Packer, S. 2010. Friction stir welding of ferrous allows: Current status.
Materials Science Forum, 638642, 4146.
Marpos 2011. MARPOSMaritime Policy Support.
Marquis, G. 2010. Failure modes and fatigue strength of improved HSS welds. Engineering Fracture Mechanics 77,
20512062.
Marquis, G. & Barsoum, Z. 2014. Fatigue strength improvement of steel structures by high-frequency mechanical
impact: proposed procedures and quality assurance guidelines. Welding in the World, 58, 1928.
Marquis, G., Mikkola, E., Yildirim, H. & Barsoum, Z. 2013. Fatigue strength improvement of steel structures by
high-frequency mechanical impact: proposed fatigue assessment guidelines. Welding in the World, 57, 803822.
Martec 2014. MARTECMaritime Technologies.
Mcpherson, N. A., Galloway, A. M., Cater, S. R. & Hambling, S. J. 2013. Friction stir welding of thin DH36 steel
plate. Science and Technology of Welding and Joining, 18, 441450.
Meilinger, A. & Trk, I. 2013. The importance of friction stir welding tool. Production Processes and Systems, 6,
2534.
Melchers, R. E. 2005. The effect of corrosion on the structural reliability of steel offshore structures Corrosion
Science 47, 23912410.
Mesa 2013. MESAMaritime Europe Strategy Action.
Miyazawa, T., Iwamoto, Y., Maruko, T. & Fujii, H. 2012. Development of high strength Ir based alloy tool for friction stir welding. Science and Technology of Welding and Joining, 17, 213218.
Molter, L., Bergstrm, M., Norden, C., Roland, F. & Lehne, M. 2014. Application of unconventional technologies,
materials and business models for an improved life cycle performance of ships. Proceeding of the Transport Research Arena.
Mosaic 2011. Mosaic EU project.
Motie 2014. MOTIEMinistry of Trade, Industry and EnergySouth Korea.
Mountford Jr., J. A. & Scaturro, M. R. 2010. TitaniumAttributes, Benefits, Use, and Applications in the Marine
Market. Journal of Ship Production and Design, 26, 1319.
Mullins, J. & Gunnars, J. Influence of Hardening Model on Weld Residual Stress Distribution. JUNE 2009. Swedish
Radiation Safety Authority, Stockholm, Sweden.
Murnsky, O., Smith, M. C., Bendeich, P. J., Holden, T. M., Luzin, V., Martins, R. V. & Edwards, L. 2012. Comprehensive numerical analysis of a three-pass bead-in-slot weld and its critical validation using neutron and synchrotron diffraction residual stress measurements. International Journal of Solids and Structures 49, 10451062.
Nanocore 2011. Nanocore EU project.
Neubert, J. & Kranz, B. 2013. Characteristics and strength behaviour of laser hybrid welds on T- and butt joints
considering European and international standards. Welding in the World, 57, 373382.
Nguyen, T.-T., Yang, Y.-S. & Kim, J.-W. 2014. An artificial neural network system for heating-path prediction in
induction heating process for concave curved surface forming. International Journal of Precision Engineering
and Manufacturing, 15, 259265.
Noda, K., Shindo, S. & Kimura, H. 2013. A material distribution scheduling for rigging ship-hull blocks with pipes.
Proceeding of the 12th International Conference on Computer Applications in Shipbuilding (ICCAS).
Norsok 2014. Norsok Standards. NORSOK.
Noury, P. 2014. Fire risk assessment of FRP composite hatch cover for panamax bulk carrier.
Nrf 2014. NRF.
Nsfc 2014. NSFCNational Natural Science Foundation of China.
Oddy, A. S., Mcdill, J. M. J. & Goldak, J. A. 1990. Consistent Strain Fields in 3D Finite Element Analysis of
Welds. Transactions of the ASME Journal of Pressure Vessel Technology, 112, 309311.
Oh, D. J., Lee, J. M. & Kim, M. H. 2014. Fatigue strength assessment of Invar alloy weld joints using the notch
stress approach Engineering Failure Analysis, 42, 8799.
Okawa, T., Shimanuki, H., Funatsu, Y., Nose, T. & Sumi, Y. 2013. Effect of preload and stress ratio on fatigue
strength of welded joints improved by ultrasonic impact treatment. Welding in the World, 57, 235241.
Olson, D. L., Jones, Z. S., Abdullah, A. A., Madeni, J. C. & Liu, S. 2010. Friction Stud Welding Testing for Submarine Rescue. International Workshop on the State of the Art Science and Reliability of Underwater Welding and
Inspection Technology.
Osawa, N., Kojima, R., Takada, A. & Takano, S. 2014. Fundamental Study on the Non-Destructive Under-Film
Corrosion of Coated Steel Panels in Water Ballast Tanks. Proceeding of the 28th Asian-Pacific Technical
Exchange and Advisory Meeting on Marine Structures (TEAM).
666
Osawa, N., Saenz, A. V., Murakawa, H., Ishiyama, M. & Tango, Y. 2011. Numerical Estimation of Inherent Deformation Induced by Induction Line heating. International Workshop on Thermal Forming and Welding Distortion.
Ozkok, M. & Helvacioglu, I. H. 2013. A Continuous Process Improvement Application in Shipbuilding. Brodogradnja, 64, 3139.
Paajanen, A. & Hakkarainen, T. 2013. FIRE-RESISTDeveloping Novel Fire-Resistant High Performance Composites. E-lass kick-off meeting september 2013.
Pahkamaa, A., Wrmefjord, K., Karlsson, L., Sderberg, R. & Goldak, J. 2012. Combining Variation Simulation
With Welding Simulation for Prediction of Deformation and Variation of a Final Assembly. Journal of Computing and Information Science in Engineering, 12, 021002021007.
Pan, L., Athreya, B., Forck, J., Huang, W., Zhang, L., Hong, T., Li, W., Ulrich, W. & Mach, J. 2013. Welding residual stress impact on fatigue life of a welded structure. Welding in the World, 57, 685691.
Park, C. K., Maho, K., Chi, S. K., Lee, J. H., Won, J. H., Kim, S. H. & Y, J. W. 2006. High Performance AntiAbrasing Coating. Proceeding of the NACE conference.
Pasqualini, O., Schoefs, F., Chevreuil, M. & Cazuguel, M. 2013. Measurements and statistical analysis of fillet weld
geometrical parameters for probabilistic modelling of the fatigue capacity Marine Structures 34, 226248.
Paulo, R. M. F., Carlone, P., Valente, R. a. F., Teixeira-Dias, F. & Palazzo, G. S. 2014. Influence of friction stir
welding residual stresses on the compressive strength of aluminium alloy plates Thin-Walled Structures 74, 184
190.
Paulo, R. M. F., Teixeira-Dias, F. & Valente, R. a. F. 2013. Numerical simulation of aluminium stiffened panels
subjected to axial compression: Sensitivity analyses to initial geometrical imperfections and material properties
Thin-Walled Structures 62, 6574.
Perez, R. 2010. Reliability of Underwater Wet Welding for Offshore Structures. International Workshop on the
State of the Art Science and Reliability of Underwater Welding and Inspection Technology.
Petrovic, J. M., Ljubic, M. R., Stamenovic, M. R., Dimic, I. D. & Putic, S. S. 2012. Tension mechanical properties
of recycled glass-epoxy composite material. APTEFF, 189198.
Poodts, E., Minak, G. & Zucchelli, A. 2013. Impact of sea-water on the quasi static and fatigue flexural properties of
\GFRP\. Composite Structures 97, 222230.
Radaj, D. 2003. Welding Residual Stresses and Distortion: Calculation and Measurement.
Rai, R., De, A., Bhadeshia, H. K. D. H. & Debroy, T. 2011. Review: friction stir welding tools. Science and Technology of Welding and Joining, 16, 325342.
Remes, H. 2013. Strain-based approach to fatigue crack initiation and propagation in welded steel joints with arbitrary notch shape International Journal of Fatigue 52, 114123.
Remes, H. & Fricke, W. 2014. Influencing factors on fatigue strength of welded thin plates based on structural stress
assessment. Welding in the World, 58, 915923.
Remes, H., Korhonen, E., Lehto, P., Romanoff, J., Niemel, A., Hiltunen, P. & Kontkanen, T. 2013. Influence of
surface integrity on the fatigue strength of high-strength steels. Journal of Constructional Steel Research 89, 21
29.
Reynolds, T. J. 2010. Service History of Wet Welded Repairs and Modifications. Proceeding of International Workshop on the State of the Art, Science, and Reliability of Under-water Welding and Inspection Technology.
Richardson, I. M., N.J., W., Armstrong, M. A. & Berge, J. O. 2010. Developments in Dry Hyperbaric Arc Welding
A Review of Progress Over the Past Ten Years. International Workshop on the State of the Art Science and Reliability of Underwater Welding and Inspection Technology.
Rosenthal, D. 1941. Mathematical theory of heat distribution during welding and cutting. Welding Journal, 20, 220
234.
Saenz, A. V., Camano, A. L., Osawa, N., Ma, N., Rashed, S. & Murakawa, H. 2011. Influential Factors Affecting
Inherent Deformation during Plate Forming by Line heating: Edge ef-fect. International Workshop on Thermal
Forming and Welding Distortion.
Sanders, D., Tewkesbury, G., Ndzi, D., Gegov, A., Gremont, B. & Little, A. 2012. Improving automatic robotic
welding in shipbuilding through the introduction of a corner-finding algorithm to help recognise shipbuilding
parts. Journal of Marine Science and Technology, 17, 231238.
Sasaki, Y., Okada, K. & Inoue, S. 2013. Application of a factory simulation tool to the outfitting process. Proceeding of the 12th International Conference on Computer Applications in Shipbuilding (ICCAS).
Schaefer, K. & Miszczyk, A. 2013. Improvement of electrochemical action of zinc-rich paints by addition of nanoparticulate zinc Corrosion Science 66, 380391.
Schajer, G. S. & Ruud, C. O. 2013. Practical Residual Stress Measurement Methods, John Wiley & Sons, Ltd.
Scharf, S., Stenzel, V. & Nieradzik, J. 2013. Damit Korrosion nicht zum Ballast wird. Selbstheilende Beschichtungen sollen die Korrosion in Ballastwassertanks reduzieren. Farbe und Lack, 119, 2832.
Selle, H. V., Peschmann, J. & Eylmann, S. 2013. Implementation of fatigue properties of laser welds into classification rules. Analysis and Design of Marine Structures-Proceedings of the 4th International Conference on Ma-rine
Structures.
Selvaraju, S. & Ilaiyavel, S. 2011. Applications of composites in marine industry. Journal of engineering research
and studies, 2, 8991.
Sensharma, P., Collette, M. & Harrington, J. 2010. Effect of Welded Properties on Aluminium Structures. Ship
Structure Committee.
Seong, W.-J., Jeon, Y.-C. & Na, S.-J. 2013. Ship-hull plate forming of saddle shape by geometrical approach Journal of Materials Processing Technology 213, 18851893.
667
Shah, S. & Tosunoglu, S. 2012. Friction Stir Welding: Current State of the Art and Future Prospects. The 16th
World Multi-Conference on Systemics WMSCI 2012.
Shamsuddoha, M., Islam, M. M., Aravinthan, T., Manalo, A. & Lau, K.-T. 2013. Effectiveness of using fibrereinforced polymer composites for underwater steel pipeline repairs Composite Structures 100, 4054.
Shanmugam, N. S., Buvanashekaran, G., Sankaranarayanasamy, K. & Kumar, S. R. 2010. A transient finite element
simulation of the temperature and bead profiles of T-joint laser welds Materials & Design, 31, 45284542.
Shapiro, A. B. 2003. Heat Transfer in LS-Dyna. Proceedings of the 4th European LS-DYNA Conference. 22nd
23rd.
Shi, X., Nguyen, T. A., Suo, Z., Liu, Y. & Avci, R. 2009. Effect of nanoparticles on the anticorrosion and mechanical properties of epoxy coating Surface and Coatings Technology 204, 237245.
Shiotani, K., Samusawa, I., Tachibana, S. & Komori, T. 2012. Development of Corrosion Resistant Steel for Ballast
Tank. Proceeding of the international Conference on Ship and Offshore Technology (ICSOT)Developments in
Ship Design and Construction.
Siriruk, A. & Penumadu, D. 2014. Degradation in fatigue behavior of carbon fiber-vinyl ester based composites due
to sea environment. Composites Part B: Engineering 61, 9498.
Solano-Charris, E. L. & Paternina-Arboleda, C. D. 2013. Simulation model of the supply chain on a naval shipyard.
Int. J. of Industrial and Systems Engineering, 13, 280297.
Song, Y. J., Lee, D. K., Woo, J. H. & Shin, J. G. 2010. System Development and Applications of a Shipyard Layout
Design Framework. Journal of Ship Production and Design, 26, 144154.
Sperle, J.-O. 2008. Influence of Parent Metal Strength on the Fatigue Strength of Parent Material with Machined and
Thermally Cut Edges. Welding in the World, 52, 7992.
Ssc-464. High speed aluminium vessel design guide. 2012. Ship Structure Committee.
Stein, J. & Kallage, P. 2008. Induction assisted welding technologies in steel utilisationINDUCWELD.
Steppan, E., Mente, T. & Bllinghaus, T. 2013. Numerical investigations on cold cracking avoidance in fillet welds
of high-strength steels. Welding in the World, 57, 359371.
Stoschka, M., Leitner, M., Posch, G. & Eichlseder, W. 2013. Effect of high-strength filler metals on the fatigue
behaviour of butt joints. Welding in the World, 57, 8596.
Sun, H., Liu, L., Li, Y., Ma, L. & Yan, Y. 2013. The performance of Al-Zn-In-Mg-Ti sacrificial anode in simulated
deep water environment. Corrosion Science 77, 7787.
Surkein, M. B., Lafontaine, J. P. & Tanner, R. E. 2009. Corrosion Protection of Deep Water Permanently Moored
Floating Production Systems using Cathodic Protection. Proceedings of the Nineteenth International Offshore
and Polar Engineering Conference (ISOPE).
Szyniszewski, S., Smith, B. H., Hajjar, J. F., Arwade, S. R. & Schafer, B. W. 2012. Local buckling strength of steel
foam sandwich panels. Thin-Walled Structures 59, 1119.
Tai, M. & Miki, C. 2014. Fatigue strength improvement by hammer peening treatmentverification from plastic
deformation, residual stress, and fatigue crack propagation rate. Welding in the World, 58, 307318.
Tango, Y., Ishiyama, M. & Suzuki, H. 2011. IHIMU-alphaA Fully Automated Steel Plate Bending System for
Shipbuilding. IHI Engineering Review, 44, 611.
Tern, G., Cuamatzi-Melndez, R., Albiter, A., Maldonado, C. & Bracarense, A. Q. 2014. Characterization of the
mechanical properties and structural integrity of T-welded connections repaired by grinding and wet welding.
Materials Science and Engineering: A, 599, 105115.
Thomas, W. M., Nicholas, E. D., Needham, J. C., Murch, M. G., Temple-Smith, P. & Dawes, C. J. 1991. Improvements relating to friction stir welding.
Thompson, B. T., Seaman, J. M. & Eff, M. N. 2013. Friction Stir Welding Tool Life Development for Thick Section
Steel. Proceedings of the Twenty-third (2013) International Offshore and Polar Engineering.
Throughlife 2011. Throughlife project newsletter: Self Healing Coatings.
Throughlife 2014. Throughlife European Project.
Tokola, H. A., Assis, L. F., Freire, R. M. & Niemi, E. 2013. Optimization of the welding in the erection scheduling
of a suezmax tanker ship. Proceeding of the 12th International Conference on Computer Applications in Shipbuilding (ICCAS).
Toumpis, A., Galloway, A., Cater, S. & Mcpherson, N. 2014. Development of a process envelope for friction stir
welding of DH36 steelA step change. Materials & Design, 62, 6475.
Vasilijevic, A., Borovic, B. & Vukic, Z. 2011. Augmented Reality in Marine Applications. Brodogradnja, 62, 136.
Von Lukas, U., Vahl, M. & Mesing, B. 2014. Maritime Applications of Augmented Reality Experiences
and Challenges. Virtual, Augmented and Mixed Reality. Applications of Virtual and Augmented Reality. Springer
International Publishing, 465475.
Wang, J., Rashed, S., Murakawa, H. & Luo, Y. 2013a. Numerical prediction and mitigation of out-of-plane welding
distortion in ship panel structure by elastic \FE\ analysis Marine Structures 34, 135155.
Wang, J., Sano, M., Rashed, S. & Murakawa, H. 2013b. Reduction of Welding Distortion for an Improved Assembly Process for Hatch Coaming Production. Journal of Ship Production and Design, 29, 153161.
Wei, L. & He, W. Comparison of Measured and Calculated Residual Stresses in Steel Girth and Butt Welds. 2010.
TWI Ltd, Cambridge, UK.
Wei, Y. & Nienhuis, U. 2012. Automatic Generation of Assembly Sequence for the Planning of Outfitting Processes
in Shipbuilding. Journal of Ship Production and Design, 28, 4959.
Wei, Y., Nienhuis, U. & Moredo, E. 2010. Two Approaches to Scheduling Outfitting Processes in Shipbuilding.
Journal of Ship Production and Design, 26, 2028.
668
Westman, M. P., Fiefield, L. S., Simmons, K. L., Laddha, S. G. & Kafentzis, T. A. Natural fiber composites: a review. 2010. U.S. department of energy.
Winczek, J. 2010. Analytical solution to transient temperature field in a half-infinite body caused by moving volumetric heat source International Journal of Heat and Mass Transfer 53, 57745781.
Winczek, J. 2011. New approach to modeling of temperature field in surfaced steel elements. International Journal
of Heat and Mass Transfer, 54, 47024709.
Woo, J. H., Song, Y. J., Kang, Y. W. & Shin, J. G. 2010. Development of the Decision-Making System for the Ship
Block Logistics Based on the Simulation. Journal of Ship Production and Design, 26, 290300.
Xu, L. R., Krishnan, A., Ning, H. & Vaidya, U. 2012. A seawater tank approach to evaluate the dynamic failure and
durability of E-glass/vinyl ester marine composites Composites Part B: Engineering 43, 24802486.
Xu, M. & Soares, C. G. 2012. Assessment of the ultimate strength of narrow stiffened panel test specimens ThinWalled Structures 55, 1121.
Yang, Y.-P., Castner, H., Dull, R., Dydo, J., Huang, T. D., Fanguy, D., Dlugokecki, V. & Hepinstall, L. 2014. Complex-panel Weld Shrinkage Data Model for Neat Construction Ship Design Engineering. Journal of Ship Production and Design, 30, 1538.
Yang, Y.-P., Castner, H., Dull, R., Dydo, J. R. & Fanguy, D. 2013. Uniform-Panel Weld Shrinkage Data Model for
Neat Construction Ship Design Engineering. Journal of Ship Production and Design, 29, 116.
Yang, Y. P., Castner, H. & Kapustka 2011. Development of Distortion Modeling Methods for Large Welded Structures. Journal of Ship Research, 27, 2634.
Yekta, R. T., Ghahremani, K. & Walbridge, S. 2013. Effect of quality control parameter variations on the fatigue
performance of ultrasonic impact treated welds International Journal of Fatigue 55, 245256.
Yildirim, H. & Marquis, G. 2013. A round robin study of high-frequency mechanical impact (HFMI)-treated welded
joints subjected to variable amplitude loading. Welding in the World, 57, 437447.
Yildirim, H. C. & Marquis, G. B. 2012. Fatigue strength improvement factors for high strength steel welded joints
treated by high frequency mechanical impact International Journal of Fatigue 44, 168176.
Yildirim, H. C. & Marquis, G. B. 2014. Fatigue design of axially-loaded high frequency mechanical impact treated
welds by the effective notch stress method Materials & Design, 58, 543550.
Yildirim, H. C., Marquis, G. B. & Barsoum, Z. 2013. Fatigue assessment of high frequency mechanical impact
(HFMI)-improved fillet welds by local approaches International Journal of Fatigue 52, 5767.
Yu, Q. Q., Chen, T., Gu, X. L., Zhao, X. L. & Xiao, Z. G. 2013. Fatigue behaviour of CFRP strengthened steel
plates with different degrees of damage. Thin-Walled Structures 69, 1017.
Yu, Y. H., Kim, B. G. & Lee, D. G. 2012. Cryogenic reliability of composite insulation panels for liquefied natural
gas (LNG) ships Composite Structures 94, 462468.
Zang, W., Gunnars, J., Dong, P. & Hong, J. K. Improvement and Validation of Weld Residual Stress Modelling
Procedure. JUNE 2009a. Swedish Radiation Safety Authority, Stockholm, Sweden.
Zang, W., Gunnars, J., Mullins, J., Dong, P. & Hong, J. K. Effect of Welding Residual Stresses on Crack Opening
Displacement and Crack-Tip Parameters. JUNE 2009b. Swedish Radiation Safety Authority, Stockholm,
Sweden.
Zhang, X.-B., Liu, Y.-J., Yang, Y.-L., Ji, Z.-S. & Deng, Y.-P. 2011. Technical Parameter Analysis of HighFrequency Induction Heating Applied to Steel Plate Bending. Journal of Ship Production and Design, 27, 99
110.
Zhang, X.-B., Yang, Y.-L. & Liu, Y.-J. 2012. The Numerical Analysis of Temperature Field During Moveable Induction Heating of Steel Plate. Journal of Ship Production and Design, 28, 7381.
Zhao, Y., Lu, Z., Yan, K. & Huang, L. 2015. Microstructural characterizations and mechanical properties in underwater friction stir welding of aluminum and magnesium dissimilar alloys Materials & Design, 65, 675681.

710 SEPTEMBER 2015
CASCAIS, PORTUGAL
VOLUME 2
COMMITTEE V.4
OFFSHORE RENEWABLE ENERGY
COMMITTEE MANDATE
Concern for load analysis and structural design of offshore renewable energy devices. Attention shall be
given to the interaction between the load and structural response of fixed and floating installations taking due
consideration of the stochastic nature of the ocean environment. Aspects related to prototype testing and
certification shall be considered.
COMMITTEE MEMBERS
Chairman:
Zhen Gao, Norway

Harry B. Bingham, Denmark
Rachel Nicholls-Lee, UK
Frank Adam, Germany
Debabrata Karmakar, Portugal
Dale G. Karr, USA
Ivan Catipovic, Croatia
Giuseppina Colicchio, Italy
Wanan Sheng, Ireland
Pengfei Liu, Canada
Yukichi Takaoka, Japan
Johan Sltte, Norway
Hyun-Kyoung Shin, Korea
Spyros A. Mavrakos, Greece
Yu-Ti Jhan, China (Taiwan)
Huilong Ren, China
KEYWORDS
Offshore wind turbine, wave energy converter, marine current turbine, multiuse plat-form, design, integrated
dynamic analysis, model test, field test.
670
ISSC committee V.4: OFFSHORE RENEWABLE ENERGY
CONTENTS
1.
INTRODUCTION
671
2.
OFFSHORE RENEWABLE ENERGY RESOURCES

2.1 Offshore wind energy resources
2.1.1 Resource assessment
2.2 Wave energy resources
2.3 Tidal and ocean current energy resources
2.3.1 Physical resource assessment
2.3.2 Numerical resource modelling
671
671
672
673
674
674
674
3.
OFFSHORE WIND TURBINES

3.1 Recent industry and research development
3.2 Numerical modelling and analysis
3.2.1 Numerical tools state-of-the-art
3.2.2 Load and response analysis of bottom-fixed wind turbines
3.2.3 Load and response analysis of floating wind turbines
3.3 Physical testing
3.3.1 Laboratory testing
3.3.2 Field testing
3.4 Transportation, installation, operation and maintenance
3.4.1 Current industry and research development
3.4.2 Numerical simulations of marine operations
3.4.3 Guidelines on marine operations for offshore wind turbine transportation,
installation, operation and maintenance
3.5 Rules and standards
675
675
678
678
679
681
687
687
689
689
690
691
692
692
4.
WAVE ENERGY CONVERTERS

4.1 Numerical modelling and analysis
4.1.1 Load and motion response analysis
4.1.2 Mooring analysis
4.1.3 Power take-off analysis
4.2 Physical testing
4.2.1 Laboratory testing and validation of numerical tools
4.2.2 Field testing
4.3 Rules and standards
693
695
695
698
699
700
701
701
702
5.
TIDAL AND OCEAN CURRENT TURBINES

5.1 Development, modelling and testing of tidal current energy converters
5.1.1 Device development
5.1.2 Numerical modelling and experimental testing
5.2 Environmental impact
5.2.1 Marine planning
5.3 Economic feasibility
703
703
703
703
704
704
704
6.
COMBINED USE OF OCEAN SPACE
705
7.
CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK
707
REFERENCES
709
1.
671
INTRODUCTION
This Specialist Committee V.4 Offshore Renewable Energy was first introduced to the ISSC in 2006.
Since then, offshore wind technology has experienced a significant development, leading to commercial
deployment of large-scale offshore wind farms. Extensive national and international research work has
also been conducted regarding the development of Wave Energy Converters (WECs) and tidal and ocean
current turbines. The ultimate goal, from both an academic and an industrial standpoint, is to develop
cost-effective solutions for utilisation of offshore renewable energy. This report describes recent
developments in this area with emphasis on research activities.
The subject of offshore renewable energy is highly multidisciplinary. In view of the relevance to the
ISSC as well as the background and competence of the committee members, the focus of this report was
naturally given to the marine aspects. However, when discussing the offshore wind technology,
aerodynamic aspects are inevitably addressed because of the importance for design analysis. Similarly,
Power Take-Off (PTO) systems and control are included in the discussion of wave energy converters,
while issues of environmental impact and economic feasibility are discussed in the chapter of tidal and
ocean current turbines. On the other hand, electrical aspects and grid development are omitted from the
discussion in this report.
Chapter 2 presents a resource assessment for wind, wave and marine current energy; which is
complementary to the information presented in the previous ISSC reports. The methodologies used for
resource assessment, in particular the numerical models developed in recent years, are discussed in detail.
Offshore wind technology is the most developed of the three areas considered and it is also the focus of
this report. In particular, topics on numerical analysis and physical testing of Floating Wind Turbines
(FWTs) are emphasised. Design rules and marine operations relating to transportation and installation of
Offshore Wind Turbines (OWTs) are also discussed. The developments in wave energy converters and
tidal and ocean current turbines are summarised in Chapters 4 and 5, respectively. A separate chapter is
included to introduce some of the recent research projects on the integration of offshore renewable energy
with other potential use of ocean space.
In Chapter 7, the main findings and conclusions from this committee are stated with respect to the different offshore renewable energy technologies. The challenges for developing cost-effective solutions are
identified, and the recommendations for future work in this area are provided.
2.
OFFSHORE RENEWABLE ENERGY RESOURCES
In the previous ISSC reports (ISSC (2006), ISSC (2009), ISSC (2012)), resources of offshore renewable
energy (including offshore wind, wave, tidal and ocean current energy) were briefly introduced, with the
key focus being European waters. This current report discusses the resources in other geographical areas
around the globe. Moreover, detailed and accurate assessment of the resource for a given local area is
very important for selection of the most suitable and cost-effective technology in the development of
commercial offshore energy farms. It is crucial to reduce the uncertainty in the resource assessment
normally conducted in the planning phase in order to improve the overall economics of an offshore
energy farm. The numerical models and methods for resource assessment developed in recent years are
discussed in detail.
2.1
Offshore wind energy resources
The North Sea is the centre for offshore wind development in Europe. The water depths in this region are
not large, varying between 20m and 70m. The US maps illustrated in Figures 1 and 2 indicate the mean
annual wind speeds at 90m above the surface, and the corresponding water depths at these locations. It
can be seen that the water depths in most of the areas along the east and south coasts are moderate, while
deeper waters in excess of 100m dominate the west coast and the Great Lakes areas. An average wind
speed of at least 7m/s at this reference height is considered the minimum speed required for wind energy
resource suitability. The estimated potentially extractable energy is around 4,150GW considering the
areas within 90km from shore. Further details and information on the characteristics and validation
methods were detailed by Schwartz et al. (2010).
The potential of wind power in Japan is estimated as 144GW for onshore generation and 608GW for
offshore generation under the assumption that the offshore wind turbines are installed within 30km of the
shore, in 200m water depths and under 7.5m/s of average wind speed at an 80m height. According to the
Japan Wind Power Association (JWPA, 2012), this total offshore output is equal to around three times of
the total output by all power generation facilities in the country. In Japan, there are aspects of the
672
challenges regarding cost, technology, and social acceptance that differ widely from the natural and social
environments and other conditions surrounding offshore wind power development in Europe.
Consequently, it will be essential to utilize the proving research currently being advanced to establish
low-cost offshore wind power generation technologies compatible with the conditions in Japan, according
to the New Energy and Industrial Technology Development Organization (NEDO, 2013).
China has a long coastline which is a relatively rich resource for offshore wind energy. According to
Qin et al. (2010), the technical potential is 1,322GW for the offshore area within 100km of the coast; with
two-thirds of these areas located in shallow waters with depths from 10m to 50m. This is a similar energy
density to that found in the North Sea.
Figure 1. United States offshore wind resources (Schwartz et al., 2010).
Figure 2. United States bathymetry distribution (Schwartz et al., 2010).
2.1.1
Resource assessment
The most commonly used tool for wind resource predictions, on land as well as offshore, is the Wind
Atlas Analysis and Application Program (WAsP). WAsP is a computer program developed for predicting
wind climates and power productions from wind turbines and wind farms. Another method to predict the
wind resource is the mesoscale meteorological model (MM5). A comparison of these two methods was
carried out in 2007 for the German Bight by Jimenez et al. (2007).
A method of estimating the annual wind energy potential of a selected site using short-term
measurements related to one years recorded wind data at another reference site was published by
Bechrakis et al. (2004). They summarized a method which offers a reliable prediction of the wind
potential of an area using only short time period measurements in combination with longer term wind
measurements of a nearby station. The errors in the mean value estimation are tolerable and fall within the
expected percentages of uncertainty with respect to the annual variation of the wind conditions in an area.
673
As per Bechrakis et al. (2004) and Jimenez et al. (2007), it is important for an acceptable prediction of
the wind resource to know the climate conditions at the points of support for the prediction. For example,
an operational wind forecast system for the Portuguese pilot area of Aguadoura was presented by
Salvacao and Guedes Soares (2014).
Furthermore, some results regarding high-resolution reanalysis data and floating met-mast
measurements at deep-water locations influenced by coastal topography were summarized by Del Jesus
et al. (2014). Motion effects on Lidar (Light detection and ranging) wind measurement data of the
EOLOS buoy were highlighted by Bischof et al. (2014).
2.2
Wave energy resources
Figure 3 illustrates a map of global annual mean wave power (Mrk et al., 2008). It was based on the data
from the ECMWF WAM model archive and calibrated and corrected by Fugro OCEANOR against a
global buoy and Topex satellite altimeter database. This map indicates that the highest average levels of
wave power are found on the leeward side of temperate zone oceans due to the origins of oceanic wind
(and inherently waves), in particular, between 40 and 60 in both hemispheres.
Figure 3. Global annual mean wave power estimates in kW/m (Mrk et al., 2008).
High resolution numerical models are useful for resource assessment of a specific site. The FP7 EquiMar
project (http://www.equimar.org/) has delivered a suite of high level protocolsgeneral principles to
enable fair comparison of marine energy converter testing and evaluation procedures. In particular,
Venugopal et al. (2010) provided a full description of the numerical methods used for the wave resource
assessment. These methods can be divided into two categories: 1) the deterministic approach, which
describes the sea surface evolution accurately in both time and space; and 2) the spectral approach, which
provides a statistical description of the wave conditions in space and time. The best deterministic models
are able to capture all the relevant physics of wave/bottom/shoreline and non-linear wave-wave
interaction, but they are relatively computationally expensive. In practice, spectral methods are more
commonly used to characterize the sites of interest. Available spectral methods have reached their third
generation and provide an approximate description of the physical effects associated with nonlinear wavewave interaction, wind input, and wave-breaking/bottom dissipation modelled as source terms.
Even though most of these numerical models are open-source and user-friendly their accuracy and
reliability depends on the skill of the user, the computational power available and the accuracy of the
input data. For example, Strauss et al. (2007) compared the wave prediction along the Gold Cost of
Australia with two of these solvers MIKE 21 (MIKE-21 (2008), Johnson (1998)) and SWAN (SWAN,
2009). Both the models were shown to over-estimate the significant wave-height for high wind speed,
highlighting the importance of the correct input data of the wind field.
Venugopal et al. (2011) also produced a report for EquiMar which shows a comparison between the
models MIKE21, SWAN and TOMAWAC (2014), for the wave prediction along the Figueira da Foz
coastline in Portugal. The correlation coefficients obtained between the measured buoy data and the wave
model output for various wave parameters are above 0.8 for each of the models, indicating a very good
correlation. However, the same authors pointed out that Figueira da Foz is a rather simple location (the
bathymetry is smooth and the coastal geometry is nearly linear); which explains the good agreement
between the model results and measured data, despite the use of coarse bathymetric data. Nonetheless,
674
there are a few differences between three models outputs which could be attributed to slightly different
formulations and input model parameters (coefficients) selected for the simulations.
For verification of numerical models against measurement buoy data for local wave resource assessment, it is essential to compare statistical distributions (rather than only basic statistics) of relevant parameters (Edwards et al., 2014a) and to compare spectral values (Edwards et al., 2014b). Such level of
comparison is important to determine the reliability of and to reveal some potentially vital issues in the
numerical models for prediction of wave energy extraction.
In Liberti et al. (2013), the WAM solver (Gnther and Behrens, 2011) is used for the prediction of
wave energy, heights, periods, directions, seasonal and inter-annual variability in the Mediterranean Sea.
The fine spatial resolution of the solver and the good accuracy in the modelling of the bathymetry and of
the wind-field allowed a remarkable comparison with wave buoy data over a ten year period. On the
website http://utmea.enea.it/data/waves, the same authors displayed the results of the Operative Model
WAVES for the next five days of hourly forecasts for the Mediterranean Sea. The model is based on a
nesting between WAM used for the whole basin at 1/32 resolution, approximately 3 km, and SWAN that
zooms in to ten interesting areas at 1/200 resolution, approximately 800m.
Even when accurately and carefully handled, these models cannot completely describe the evolution of
the free surface and the kinematics of the wave field. It is now common practice to equip wind turbines
with Lidars to measure the speed and direction of the incoming wind, and it would be similarly beneficial
for WECs to have a means of now-casting the incoming waves to better position themselves (Cagninei
et al., 2013) or control the PTO settings (Sjolte et al., 2013). Although the description of the wave energy
resource is most relevant at the design stage when the features of the WEC are tuned to the exploitation
site, now-casting could become a means to improve productivity once the technology has advanced to the
fully commercial stage. A tool that may make this possible is SNOW, which is derived from ship science
where the now-casting represents a critical feature for the landing of aeroplanes or helicopters on ships
(Xiao et al. (2009), Yue and Solodonz (2008)).
2.3
Tidal and ocean current energy resources
Resource assessment is important and critical for marine current turbines to estimate energy yield and
evaluate extreme design conditions. The tidal and ocean current energy resources are strongly sitedependent. The understanding of in-situ flow conditions can be derived from a combination of numerical
modelling and survey data. Goundar and Ahmed (2014) performed a resource assessment and turbine
blade design for Fiji according to measurement of flow velocities, flow direction, wave height, wave
frequencies, and velocity profiles. Bahaj (2013) showed a site measurement of flow conditions close to
Alderney, Channel Islands, UK. Kim et al. (2012a) have assessed the ocean renewable energy resources
in Korea, concluding that tidal energy is likely to play an important role in their future energy mix.
Several, initially less energetic, sites around the globe are now attracting interest for development: these
include the southern coast of Iran (Rashid, 2012); the Aleutian Islands, Alaska (Wright, 2014); the Puget
Sound, USA (Polagye and Thomson, 2013); the Shannon Estuary, Ireland (ORourke et al., 2014);
Roosevelt Island, New York, USA (Gunawan et al., 2014).
2.3.1
Physical resource assessment
The flow conditions in an area with a large or strong tide are complex and physical assessment is often
imperative to assess the resource and determine the appropriate locations for various devices. Acoustic
Doppler Current Profilers are frequently used (Thomson et al. (2012), Polagye and Thomson (2013),
Stevens et al. (2012), Palodichuk et al. (2013), Bell et al. (2012), ORourke et al. (2014), Gunawan et al.
(2014)) both bottom mounted and on floating vessels or platforms. A common observation in various
locations around the globe is the presence of large scale, anisotropic eddies dominating the energy
spectra. Marine radar has also recently been utilised to determine surface currents at a few locations (Bell
et al., 2012) with a good level of success. Ultimately, the temporal variation of the current speeds, current
directions, turbulence intensities, and power densities attained through physical measurement provides
essential data for device, structure and array development with respect to uncertainties in energy yield and
structural fatigue.
2.3.2
Numerical resource modelling
Physical modelling of resources is both time consuming and costly, therefore numerical assessment of
resources is desirable to bring down both timescales and overall project costs. With the complexities in
flow conditions evident at tidal sites this is not a trivial matter. Large scale oceanographic models are well
675
developed (Yang et al. (2013), Kelly et al. (2012), Taguchi et al. (2014)) but tend not to capture the local
convolutions apparent in tidal races.
Two dimensional modelling is common, and well-practised in oceanography, and several have been
further developed to incorporate various aspects specific to tidal sites. Easton et al. (2012) showed that
60% of incoming tidal energy is dissipated via friction with the local environment in an energetic tidal
channel.
Three dimensional modelling is generally required in order to capture the erratic nature of tidal energy
sites for proper assessment and analysis for project development. Venugopal and Nemalidinne (2014)
indicated that a coupled 3D wave and tidal model performed well when compared to ADCP data for the
waters in the Pentland Firth. Pacheco et al. (2014) also used a combination of numerical modelling and
physical measurements for the Faro channel in the Ria Formosa, Portugal. The results indicated that the
channel would be a good test site for HATTs with power densities in the region of 5.7kW/m2 in currents
of 2.2m/s. It is stated that the methodology can be adapted for application to any other intel-tidal channel
or estuarine mouth. Ultimately, it is agreed that 3D models should be validated with data attained through
physical test methods (Work et al. (2013), Tang et al. (2014)).
Of particular interest is the Semi-implicit Eulerian-Lagrangian Finite-Element (SELFE) model used by
Behera and Tkalich (2014). The model solves the 3D shallow water equations with Boussinesq approximations. The model was used to assess the potential for tidal energy extraction in the Singapore Strait,
and concluded that a theoretical array could produce in excess of the 4% of the energy demand for
Singapore.
3.
OFFSHORE WIND TURBINES
3.1
Recent industry and research development
Development of demonstration and commercial offshore wind farms

Since 2008, offshore wind industry has experienced a significant growth in installed capacity, especially
in Europe. According to International Energy Agency (IEA, 2014a), the total installed offshore wind
capacity by end of 2013 reached 6,590MW from the 11 IEA Wind member countries and 2,011MW of
the total new capacity was added during 2013 (Table 1). A high average annual increase of about 35% is
also observed during recent years (20112013).
Table 1. Cumulative Installed Capacity (MW) of Offshore Wind Power 20112013

(IEA, 2014a).
Country
2011 Capacity
2012 Capacity
2013 Capacity
United Kingdom
1,838
2,679
3,653
Denmark
871
920
1,271
Germany
200
280
903
China
108
390
428
Netherlands
228
228
228
Japan
25
25
50
Finland
26
26
26
Ireland
25
25
25
Korea
0
2
2
Norway
2
2
2
Portugal
2
2
2
Total
3,325
4,579
6,590
Among these developments, Europe is the major player, accounting for 96.9% of the newly installed
capacity in 2013 and 92.7% of the total installed capacity by end of 2013. In terms of the cumulative
installed capacity in different European countries, the UK has a share of 56% by end of 2013, followed
by Denmark 19%, Belgium 9%, Germany 8%, Netherlands 4% and Sweden 3%.
According to European Wind Energy Association (EWEA, 2014), a monopile is still the most popular
support structure for bottom-fixed offshore wind turbines, with 75% of the substructures in European
waters being monopiles, 12% gravity-based foundations and 5% jackets.
676
However, as the International Energy Agency (IEA, 2013) indicated, the investment costs for offshore
wind can be two to three times higher than onshore wind developments due to the additional expense
associated with substructures, electric infrastructure and on-site installation. Moreover, the recent
offshore wind development also resulted in an increase in the capital cost per unit installed capacity,
Figure 4. This is because more and more wind turbines are being installed in deeper waters, further away
from shore. The main challenge for offshore wind is cost reduction and this is also true for development
of alternative offshore renewable energies.
Therefore, the tendency in offshore wind industry is to increase the size of individual turbines, as well
as the size of wind farms, in order to reduce the installation and maintenance costs. The average rated
power of offshore wind turbines increased from 3MW in 2008 to 4MW in 2013 and the average wind
farm capacity was around 485MW per wind farm in 2013.
Figure 4. Capital costs (EUR/W) of European offshore wind farms by year (IEA, 2013) (The figure was reproduced by
IEA, with the source from GL Garrad Hassen. The bubble diameter indicates wind farm capacity.)
According to the European Commission (EC, 2008), the EU has set a target for offshore wind
development, to achieve 40GW of installed capacity by 2020 and 150GW by 2030, which is equivalent to
4% and 14% of EU electricity demand, respectively. This implies a significant amount of marine
operations related to transportation and installation of wind turbine components in the European waters in
the next five years. Considering 4MW wind turbines, more than 1000 units will need to be installed every
year. However, the trend to develop and deploy large-scale wind turbines (6-8MW) would help to reduce
the total number of operations at sea.
While there are no commercial offshore wind farms operating in the United States as of this writing,
there are several projects at advanced stages of planning. Cape Wind in Nantucket Sound near the coast
of Massachusetts has plans for the construction of 130, 3.6MW, offshore wind turbines scheduled to
begin in 2015. Initial stages of construction have also begun on Deepwater's Block Island off the Rhode
Island coast. The Block Island Wind Farm has plans for a 30MW development consisting of 5 turbines. In
May 2014, the US Department of Energy (DOE, 2014) selected three demonstration projects for
additional funding to achieve commercial operation by 2017. Among them, Dominion Virginia Power
will test two 6MW direct-drive wind turbines on twisted jacket foundations, while Fishermens Energy of
New Jersey will install five 5MW direct-drive wind turbines also on twisted jacket foundation. Principle
Power are to install five 6MW direct-drive wind turbines on their semi-submersible floating foundation
WindFloat.
The offshore wind development in China has been slow (Korsnes, 2014) since the first major offshore
wind farm, Donghai Bridge 102MW installation, was completed off Shanghai in 2010. With the second
major wind farm of 150MW (Phase 1 and 2 of the Rudong Intertidal wind farm) completed, the total
installed offshore wind capacity reached 428.6MW by the end of 2013. On the other hand, according to
the 12th 5-year Plan for Renewable Energy, the Chinese Governments goal is to achieve 5GW of
installed offshore wind power by 2015, and 30GW by 2020. However, there is some uncertainty as to
whether the 2015 goal can be achieved (Korsnes, 2014).
In Japan, only 50MW of the total offshore wind capacity was installed by end of 2013, including two
2MW floating wind turbines. In 2015, another two 7MW floating wind turbines (one on an advanced spar
677
with heave plates and the other on a V-shape semi-submersible) will be deployed in the Fukushima area
as part of the Fukushima Floating Offshore Wind Farm Demonstration Project run by Fukushima
Offshore Wind Consortium (FOWC, 2011). In the long term, the Japan Wind Power Association (JWPA,
2014) has set up a roadmap to install 700MW of wind power offshore (100MW on floating foundations
and 600MW on bottom-fixed foundations) by 2020 and 10GW (4GW floating and 6GW bottom-fixed) by
2030.
Development of floating concepts
Floating wind turbines have been proposed and developed in recent years, but there are no commercial
farms yet. The Cost of Energy (COE) is much higher than that of offshore bottom-fixed wind turbines,
due to costly support structures. Experiences regarding design, installation and operation of floating wind
turbines are now being gathered through prototype tests. Three different methods for floating platform
foundation are defined by Butterfield et al. (2005)ballast, buoyancy (weighted waterplane area) and
mooring line stabilisation. A comprehensive overview of various global floating foundation development
projects is provided by both Bossler (2013) and Arapogianni and Genachte (2013).
Statoil installed the worlds first floating offshore foundation, the Hywind spar buoy, in 2009 which
supports a Siemens 2.3MW wind turbine (Statoil, 2014), Figure 5(a). A semi-submersible floating
platform, WindFloat, was launched with a 2MW Vestas turbine on board in 2011 (PrinciplePower, 2014),
Figure 5(b). Two Japanese floating foundations were commissioned, and subsequently installed, in the
second half of 2013. The semi-submersible foundation by Mitsui with a Hitachi 2MW downwind turbine
(FOWC, 2013), Figure 5(c), and a hybrid spar floating offshore wind turbine developed by a Kyoto
University and Toda Corporation with a Hitachi/JSW 2MW turbine (GOTO-FOWT-Website, 2014) have
been in operation since late 2013, Figure 5(d).
Figure 5. Prototypes of floating wind turbines (from left: a) Hywind (Statoil, 2014); b) WindFloat (PrinciplePower,
2014); c) Fukushima semi-submersible (FOWC, 2013); d) GOTO spar (GOTO-FOWT-Website, 2014)).
Other buoyancy stabilised ideas are in the phase of conceptual design, such as the IDEOL platform
(http://www.ideol-offshore.com/en/floater_overview), which is a ring-shape surface concrete floater with
a shallow draught and compact dimensions. Some have reached a more detailed level of development
such as the three column Mitsubishi structure intended for deployment with a 7MW turbine on board.
This platform has been fabricated and is planned for installation in 2015 (FOWC, 2011).
The last floating foundation concept, with several installed at sea and in operation as prototypes at
small scale, is the Tension Leg Platform (TLP). Such concepts for floating offshore wind turbine support
structures are currently at various stages of development. Iberdrolas TLP solution comprises a single
central body as per Rodriguez et al. (2014), Figure 6(a). Glosten Associates have developed a similar
concept, the PelaStar structure (Scott, 2013), Figure 6(b). Following the same principals as a TLP the
Tension Leg Buoy (TLB) has also been explored. A TLB (Myhr and Nygaard, 2012) has significantly
lower steel mass in comparison to other platform types that can be deployed in water depths of 50m and
more, Figure 6(c). However, the mooring line configuration and associated buoyancy distribution results
in higher horizontal loads when compared to a conventional TLP mooring spread. The angled
TLB mooring spread (Lee (2005), Myhr and Nygaard (2012)), is also utilised by the GICON-TLP (Adam
et al., 2014).
678
Figure 6. Concepts of TLP or TLB floating wind turbines (from left: a) Iberdrola TLP (Rodriguez et al., 2014); b)
Glosten PelaStar TLP (Scott, 2013); c) TLB concept (Myhr and Nygaard, 2012); d) GICON-TLP (Adam
et al., 2014)).
The concepts discussed above are floating platforms supporting one Horizontal Axis Wind Turbine
(HAWT). Vertical Axis Wind Turbines (VAWTs) on floating foundations were also studied in recent years.
In addition, floating concepts with multiple turbines have also been proposed. A small-scale multiple
turbine, hexagonal floating platform for an offshore wind power plant was installed in Hakata bay by
Kyushu University on December, 2011 (Bossler, 2013). The platform is 18m in diameter and has a
displacement of 140ton supports two 3kW wind turbines equipped with Flanged-Diffuser and solar panels
rated to 1.5kW. The platform is supported by six vertical cylinders connected by a truss. The hydrodynamic
forces and moments in waves were determined using interaction theory for multiple floating bodies. Extreme
environmental loading, wind and wave, needs to be considered when designing platforms for safety and
through-life performance. CFD (Computational Fluid Dynamics) has been undertaken and is being
developed for the multibody platform, ready for comparison with results attained from model tests under
both wind and waves. In this manner the principal dimensions were confirmed. The project subsequently
intends to undertake a verification test with two 200kW wind turbines equipped with flanged-diffusers and
solar panels rating up to 100kW on a 100m diameter, hexagonal platform floating in the ocean.
3.2
3.2.1
Numerical toolsstate-of-the-art
In order to develop cost-effective solutions of offshore wind turbines, uncertainties associated with the
estimation of loads/load effects in the design phase should be reduced. This requires accurate numerical
methods for dynamic response analysis. Offshore wind turbines are designed and analysed using
simulation tools which allow an account of the coupling effects between aerodynamics, hydrodynamics
and elastic responses as well as control of the turbine. Therefore, incident waves, sea current,
hydrodynamics and dynamics of the support structures should be considered in line with wind turbine
loads as per Jonkman and Musial (2010). Several integrated tools exist to simulate dynamic behaviour of
offshore wind turbines, both bottom-fixed and floating.
Integration of onshore wind technology and the technology developed in the offshore oil and gas
industry has resulted in a significant development of numerical tools for analysing offshore wind turbines.
These integrated tools (Jonkman and Musial, 2010) are developed either from the numerical codes for
onshore wind turbines by adding hydrodynamic modules, such as FAST, HAWC2, Bladed, 3DFloat, or
from a hydrodynamic and response analysis code coupled to a wind turbine aerodynamic module, such as
Simo-Riflex-AeroDyn.
As reported in ISSC (2012), the IEA initiated two benchmark studies on most of these toolsthe
Offshore Code Comparison Collaboration (OC3) and Offshore Code Comparison Collaboration,
Continuation (OC4). The purpose of the OC3 and OC4 projects was to verify the accuracy of offshore
wind turbine simulation tools via a code-to-code comparison of simulated responses of various offshore
structures (Vorpahl et al. (2014), Robertson et al. (2014)). The OC3 project considered monopile, tripod
and spar wind turbines; while a jacket wind turbine and a semi-submersible wind turbine were studied in
the OC4 project.
As a further continuation, the ongoing OC5 project (Offshore Code Comparison Collaboration,
Continuation, with Correlation) focuses on the validation of numerical tools against model tests or field
measurements over three phases. In the first two phases, the model test results of hydrodynamic loads on
679
a fixed cylinder and responses of a semi-submersible floating wind turbine are to be considered. The field
measurements of a jacket wind turbine will be used in the last phase.
3.2.2
Load and response analysis of bottom-fixed wind turbines
3.2.2.1 Integrated dynamic analysis and fatigue assessment

Reliable power production from a wind turbine is one of the key factors contributing reduction in the cost
of energy. Providing adequate reliability can help reduce the need for costly repairs and downtime. Most
of the shallow water offshore wind turbines are installed on bottom-fixed substructures such as
monopiles, concrete gravity bases, tripods or jackets. Water depth plays an important role to find a costeffective solution for foundations. Although the current industrial practice for design of bottom-fixed
wind turbines still relies on a decoupled analysis approach with wind turbines and foundations designed
by different companies, integrated analysis methods and tools have been developed (see Section 3.3.1)
and should be soon applied in the industry for structural design and power output estimation.
Bagbanci et al. (2012a) studied the influence of the environmental conditions on wind turbine design
loads for a monopile foundation using aero-servo-hydro-elastic code. Detailed analysis regarding the
bending moment at the tower base and blade root for various values of water depth, tower height, pile
diameter and differing turbulence model was presented. The maximum blade bending moment was
observed to increase with wind speed up to the rated wind speed of 11.5 m/s, and then decrease.
However, for each case the bending moment at the tower base is at a maximum for 30m and minimum for
10m of water depth.
Shirzadeh et al. (2013) performed an aeroelastic simulation of a Vestas V90 3MW wind turbine using
the HAWC2 aeroelastic code. The Operational Modal Analysis was used to identify the damping value of
the fundamental fore-aft mode of an offshore wind turbine using both real life measurements and
simulations. It was observed that the estimations of the total damping of an offshore wind turbine gave a
quantitative view of the stability characteristics of the wind turbine.
In the offshore oil and gas industry, fatigue criteria have been of key concern for fixed offshore
platforms since the early 1970s. It is an important consideration for the design of structures in areas with
repetitive hydrodynamic loading, storm conditions and especially for dynamically sensitive structures.
For offshore wind turbines, dynamic response under resonance is more significant than that of traditional
platforms used in the offshore petroleum industry due to the wind loading effect. The fatigue load and the
number of load cycles to be considered are much higher than those associated with a more traditional
platform. Thus, the fatigue performance of welded connections is a design-driving criterion for offshore
wind turbine support structures. On the other hand, long-term fatigue analysis requires a large number of
simulations considering different combinations of wind and wave conditions. This creates a requirement
for efficient yet accurate analysis methods.
Long and Moe (2012) performed a frequency domain analysis for a lattice type bottom fixed offshore
wind turbine. The 5MW NREL (National Renewable Energy Laboratory) wind turbine installed in water
of depth 35m was redesigned with respect to the Fatigue Limit State (FLS) for three-legged and fourlegged lattice towers. It was observed that the lattice towers designed in accordance with the ultimate
limit state could not meet the fatigue requirements, notably at the joints. In other words, the fatigue limit
state is the governing design criterion. In comparison with designs based upon ultimate strength, the mass
of both types of towers increased by maximum 30%.
Dong et al. (2012) presented the dynamic response of a jacket support structure due to wind and wave
loads, and predicted the fatigue reliability of welded multi-planar tubular joints of the same structure. A
site in the northern North Sea with a water depth of 70m was considered and the hot-spot stress approach
was used for fatigue analysis of failure-critical location of the related reference braces of the selected
tubular joints.
Yeter et al. (2014a) and Yeter et al. (2014b) developed a methodology by employing the finite element
method for coupled dynamic analysis in frequency domain, which indeed allowed performing more
sophisticated response analysis by the means of modelling of a tubular structural system by shell
elements. Yeter et al. (2015a) investigated the influence of the overload on the retardation under various
load cases that are specified by factors of retardation.
Detection of damage in the members of support structures prior to a catastrophic failure is also an
important area for investigation. Liu et al. (2014a) proposed an improved modal strain energy method for
damage localisation in jacket-type offshore wind turbines. The method defined a series of stiffnesscorrection factors that could be employed to calculate the modal strain energy of the measured model
680
without utilising the stiffness matrix of the finite element model as an approximation. The numerical
studies indicated that the proposed method significantly outperformed the traditional modal strain energy
and modal strain energy decomposition methods, especially for damage location.
3.2.2.1.1
Ice-induced load and response analysis of offshore wind turbines
Support structures for offshore wind turbines in cold climates are subject to ice actions that add risks and
increase costs of construction and maintenance (Salo and Syri, 2014). Facilities in the Baltic and the Gulf
of Bothnia, for example, must be designed to withstand static and dynamic ice loads (Hirdaris et al.,
2014). Popko et al. (2012) made an in-depth comparison for the available design standards for offshore
wind turbine support structures subjected to sea ice loads. The IEC standard 61400-3 provides several
Design Load Cases (DLCs) for consideration, as do standards such as DNV-OS-J101 (Sirnivas et al.,
2014). Both parked and power production modes are considered. Ice load scenarios must be addressed for
horizontal and vertical ice forcing from thermal expansion as well as ice floe impact, cyclic loading from
ice crushing, and pressure ridge encounters with support structures. Design scenarios and basic design
approaches for ice resistance are influenced by knowledge obtained from arctic offshore structures (ISO
(2010), Thomas and Graham (2014)).
As is the case for offshore oil production structures located in cold climates, the magnitude of ice loads
may be of the same order as wind and wave loading; in fact ice action may dominate in some situations
(McGovern and Bai (2014), Hou and Shao (2014)). Serious ice induced vibrations of offshore structures
have been seen to occur; for example in the Bohai Sea and Bothnia Bay (Xu et al., 2014). Severe
vibration conditions can arise at certain impact velocities in which the breaking patterns of the ice occur
at a frequency which matches the structural vibrations modes (Bjerks et al. (2013), Bjerks et al. (2014),
Xu et al. (2014)). This condition is referred to as frequency lock-in and is specifically addressed in the
IEC and ISO standards and class society guidelines. Frequency lock-in is important for determination of
force and displacement magnitudes and also the number of vibration cycles for fatigue assessments
(Hendrikse et al., 2014).
Ice loading is stochastic in nature (Zvyagin and Sazonov, 2014) and average ice pressures depend upon
contact area as well as the failure mode of the ice, which is in turn influenced by the ice and structural
geometry (Bekker et al., 2013). Local pressures are generally lower than global forces (Spencer et al.,
2014) and probabilistic descriptions of these effects have received considerable attention (Taylor and
Richard, 2014). These effects also play important roles in analysis of impact forces from level ice features
and pressure ridges (Molyneux et al. (2013), Norouzi et al. (2013)).
Conical structures located on the turbine support structure in the ice belt region are used for reducing
ice force magnitudes and ice-induced structural vibrations (Barker et al., 2014). Both upward and
downward sloped cones have been considered, each having a tendency to cause the ice to fail in bending
rather than crushing and thereby reduce the forces acting on the structure (Xu et al., 2014).
New advanced structural dynamics modelling capabilities have been developed to incorporate
combined loading from wind, wave and ice in a finite element model (Hilding et al., 2012) and timedomain simulation of wind turbine response. The simulation software FAST, developed by the National
Renewable Energy Laboratory, has a new modular framework which includes capability for modelling
multimember substructures (Song et al., 2013). New modules covering ice loading capability are also
being integrated (Yu et al. (2013), Yu et al. (2014), Barahona et al. (2015)).
3.2.2.2 Challenges for design and analysis of bottom-fixed wind turbines

Grouted connections
A grouted connection forms the connection between the transition piece and the monopile in offshore
wind turbine structures.
Such connection has been subjected to intensive research during the last couple of years based on
experiences in the offshore wind industry. It was found that an acceptable level of safety was not reached
with existing design methods, developed for traditional jacket structures that mainly are subjected to axial
loading. This was based on the fact that stresses in a monopile caused by bending moments from wind
loading can be more than one order of magnitude larger than those due to the axial load alone (Lotsberg
et al. (2012a), Lotsberg et al. (2013)).
DNV initiated two joint industry projects (JIPs) on grouted connections. The first was carried out from
20092011 to investigate the structural capacity of grouted connections made without using shear keys,
which have been the most common solution for monopiles. The second project was initiated in 2011,
studying the capacity of cylindrical shaped grouted connections with shear keys (Lotsberg et al. (2012b)).
681
The JIPs have resulted in the development of a new design methodology to account for large dynamic
bending moments on monopiles, and updates of the existing design standards.
Soil-pile interaction
The method commonly adopted in design codes for analysis of laterally loaded piles is denoted the p- y
approach and is based on the Winkler model. It has been a successfully proven method in the offshore oil
and gas industry. However, it has not been shown to be suitable for predicting responses for monopiles
with large diameters and loads that need to be accounted for in the offshore wind industry. Further
research in this direction is required.
The PISA project (Pile Soil Analysis project) (http://www.eng.ox.ac.uk/geotech/research/PISA ) is a
research project that is focusing on developing new and improved design methods for laterally loaded
piles, specifically targeting monopile foundations. The project is led by Dong Energy and run through the
Carbon Trusts Offshore Wind Accelerator programme. Onshore field tests were performed in 2014 to
assess the lateral response of monopiles in soils considered as representative for the North Sea, in both
sands and clays, involving piles with a diameter of approximately 2.1m.
Negro et al. (2014) identified the uncertainties compromising offshore wind farm structural design and
which were the design of the transition piece and the difficulties regarding characterisation of soil
properties. Alternative design criteria were proposed using heterogeneous soil characteristics, considering
the risks related to the requirement of considerable investment and time for a full soil campaign.
Damgaard et al. (2014) performed a numerical investigation based on nonlinear Winkler foundation
models. Experimental tests were carried out in order to evaluate the eigen-frequencies related to the
lowest eigen-mode of offshore wind turbines. The study showed that the permeability of the subsoil has a
strong influence on the stiffness of the wind turbine; and that this may, to some extent, explain the
deviations observed between experimental and computational eigen-frequencies.
Similarly, eigen-frequencies and therefore dynamic characteristics of monopile wind turbines will alter
due to the presence of scour. Myrhaug and Ong (2013) studied the scour around vertical pile foundations
of offshore wind turbines resulting from long crested (2D) and short crested (3D), nonlinear, random
waves. A stochastic method was used to derive the maximum equilibrium scour depth around the piles
exposed to the wave climate. The approach was based on the assumption that the waves were a stationary
narrow-band random process and adopted the Forristall wave crest height distribution.
Structural reliability analysis
In recent years, fatigue reliability assessments have been performed for different types of wind turbine
support structures; monopiles, jackets and tripods (Srensen (2012), Dong et al. (2012), Lee et al. (2014),
Yeter et al. (2015b)). As highlighted by Dong et al. (2012), future research needs to focus on the
identification and quantification of model uncertainties associated with external load analysis, dynamic
response assessment and fatigue damage calculation.
Mardfekri and Gardoni (2013) developed probabilistic models to predict the deformation, shear and
moment demands on offshore wind turbine support structures subject to operational and environmental
loadings. The proposed models were then used to assess the fragility of an offshore wind turbine
subject to day-to-day wind, wave and current loading. The conditional probabilities of exceeding the
specified performance levels were found to increase with the average wind speed, up to the rated wind
speed.
Based on a sensitivity analysis performed by Yeter et al. (2015b) for a tripod offshore wind turbine, it
was found that the uncertainty associated with wind loading dominates over that of wave loading in a
fatigue reliability assessment. For bottom-fixed offshore wind turbines, wind-induced dynamic responses
are much larger than wave-induced responses.
Thns et al. (2012) has established the model basis regarding the ultimate limit state consisting of
structural, loading, and probabilistic models of the support structure of offshore wind energy converters
together with a sensitivity study. The model basis is part of a risk based assessment and monitoring
framework, and is applied to determine the reliability as prior information for the assessment and as a
basis for designing a monitoring system.
3.2.3
Load and response analysis of floating wind turbines
3.2.3.1 Design aspects and analysis methods for floating wind turbines
As mentioned in Section 3.1 three primary foundations were defined for floating offshore wind turbine
support structures: spar, semi-submersible and TLP. Thiagarajan and Dagher (2014) presented a brief
682
review on the concepts of offshore floating platforms. The key studies considering floating foundation
structures for offshore wind were described. The simulation tools and experimental studies were
discussed to assist development of sustainable solutions. A detailed review of offshore floating wind
turbine concepts is also presented by Bagbanci et al. (2012b) and Guedes Soares et al. (2014a).
Bachynski and Moan (2012) discussed the design requirements for TLP structures with respect to five
key aspects:
-
the natural periods for surge and sway motions longer than 25s;
the heave, roll and pitch natural periods less than 3.5s;
the largest mean offset not exceeding 5% of the water depth;
the minimum tendon size according to the ULS (Ultimate Limit State) design;
and the minimum displaced volume of 2000m3.
Five baseline TLP designs supporting the NREL 5MW wind turbine were derived in their study, with
the displacement ranging from 2330m3 to 11866m3. Dynamic response analyses for assessment of the
designs were carried out using the coupled code Simo-Riflex-AeroDyn. A practical, single column, TLP
design with displacement ranging from 3500-6500m3 was developed, incorporating three pontoons, with
a pontoon radius of 28-35m. Such a platform may be capable of supporting the NREL 5MW wind turbine
safely in harsh environmental conditions.
A semi-submersible concept is likely to be more competitive in shallower waters when compared to the
spar and TLP concepts. A version of the WindFloat semi-submersible, proposed by Roddier et al. (2009),
has been under test in open water since 2011. In a comparative study between a generic WindFloat
(4500ton) and a generic spar (7500ton) supporting a 5MW wind turbine (Roddier et al., 2011), the
hydrodynamic performance of these two concepts was found to be similar; however WindFloat, being
equipped with heave plates, exhibits improved overall response. WindFloat also facilitates installation
and commissioning techniques, with the whole system being assembled at the quay side and then towed
to the final location.
Kvittem et al. (2012) used a semi-submersible wind turbine similar to the WindFloat concept in the
comparison of various hydrodynamic models. The Morison model is often used in aero-hydro-servoelastic codes developed from the wind industry such as HAWC2, while the potential flow model with
quadratic drag forces is commonly applied in time-domain global response analyses using the codes
developed in the offshore oil and gas industry, such as Simo-Riflex-AeroDyn. The Morison model can
provide a good representation of the motion responses, as compared to the potential flow model, if proper
added mass coefficients are selected and stretching of wave kinematics and dynamic pressure under the
columns are considered.
Wind turbine design typically involves a significant number of coupled dynamic analyses in the time
domain, and one important aspect of these is to directly obtain the responses in the floater members. Luan
et al. (2013) developed a modelling method that can be used in the time domain to obtain the loads in the
braces of a semi-submersible wind turbine structure. The semi-submersible floater model consists of
columns as rigid bodies, with hydrodynamic loads calculated based on the potential theory, and braces
connecting columns as beam elements with the Morison-type loads. A case study of the ULS design
check for the brace system of the OC4 semi-submersible wind turbine was performed to illustrate the
developed method.
3.2.3.2 Global response analysis

Floating wind turbines differ from bottom-fixed turbines by their larger rigid-body motions, the
importance of the coupling effects between wind and wave loads and induced responses, and increased
challenges regarding the automatic control system. These complexities require coupled tools for dynamic
analysis, which also take into account the control actions. In the last decade, various floating wind turbine
concepts have been proposed and developed. Different types of floaters represent different dynamic
behaviours in wind and waves, and therefore lead to different responses in the rotor, tower and wind
turbine drivetrain. Comparative studies on the dynamic behaviour of spar, semi-submersible and TLP
floating wind turbines have been conducted for both operational and survival conditions.
Jonkman and Matha (2011) compared three floating concepts (spar, TLP and barge), supporting the
NREL 5MW wind turbine. Dynamic responses under combined wind and wave loads were predicted by
coupled analysis using FAST. Wind turbine aerodynamics were modelled by blade element momentum
theory, and linear hydrodynamics and mooring spring models were also considered in the analysis. The
three concepts were compared based on the calculated ultimate loads, fatigue loads and instabilities. The
683
barge concept led to the highest load in the turbine, while the differences in the loads between the spar
and TLP concepts were not significant; except for the loads in the tower, which were greater in the spar
system.
Bagbanci et al. (2011a) studied the dynamic responses of a 5MW, spar-type, floating wind turbine. A
hydrodynamic study of the floater was performed and the added mass, damping coefficients and
excitation forces from linear potential theory and the hydrostatic data were obtained. Using the
hydrodynamic data, the aero-servo-hydro-elastic simulation for the spar-type floating wind turbine was
carried out. Further, a brief comparison of spar-type and barge-type offshore floating wind turbines was
performed by Bagbanci et al. (2011b) with simulation results for tower base motions and platform
motions for various wind conditions being obtained. It was observed that with the increase in wind speed
the pitch motion of the spar type floating wind turbine increased, whereas the pitch motion decreased for
the barge type floating wind turbine.
Roald et al. (2013) studied the effect of second-order hydrodynamic loads on a spar and a TLP based
wind turbine. Using the proposed method, the importance of the second-order effects was assessed for the
OC3-Hywind spar and the UMaine TLP. The comparison showed that the second-order forces were very
small for the OC3-Hywind spar.
Sweetman and Wang (2012) presented the theoretical developments underlying an efficient
methodology to compute the large-angle rigid body rotations of a floating wind turbine in the time
domain. The tower and Rotor-Nacelle Assembly (RNA) were considered as two rotational bodies in
space, for which two sets of Euler angles were defined and used to develop two systems of Euler dynamic
equations of motion.
Jeon et al. (2013) determined the dynamic response of a rigid cylindrical floating substructure moored
by three catenary lines under long-crested irregular waves using a fluid-rigid body interaction simulation.
A numerical investigation of the dynamic response of spar-type floating substructure due to wave
excitation was performed. The upper part, composed of wind turbine blades, hub and nacelle, was
simplified as a lumped mass, and a dynamic model of three catenary mooring lines was considered.
Bae and Kim (2014) assessed the global performance of a spar-type floating offshore wind turbine with
two different extreme environmental conditions. The coupled analysis included time-varying
aerodynamic loading and damping, tower-blade elastic deformation, blade-control-induced loading, and
the gyroscopic effect of the rotating blades. A numerical prediction tool was developed for the fully
coupled dynamic analysis of the system in the time domain including aero-loading, blade-rotor dynamics
and control, mooring dynamics, and platform motions so that the influence of rotor-control dynamics on
the hull-mooring performance could be evaluated.
Karimirad and Moan (2013) performed a stochastic analysis of a tension-leg spar wind turbine
subjected to wind and wave actions and the dynamic motions, structural responses, power production and
tension leg responses were obtained. The negative damping, rotor and tower shadow effects were
discussed when studying the power performance and structural integrity of the system.
Apart from the code-to-code comparison of different numerical tools (as discussed in Section 3.2.1),
validation against model test results or field measurements is an important step for developing such
numerical tools. Publically available field measurements are limited and the code-to-experiment
validation in the current stage focuses on the use of results from various laboratory tests, see Section
3.3.1.
Utsunomiya et al. (2013) developed an analysis tool to assess the dynamic response of floating
offshore wind turbines under extreme environmental conditions. The dynamic analysis tool consists of a
multi-body dynamics solver, considering aerodynamic loads, hydrodynamic loads and mooring loads.
Detailed comparison with the experimental results of a small size (100 kW) spar wind turbine at sea were
performed in order to validate the model; where the wind, current and wave actions were applied
simultaneously.
Coulling et al. (2013) compared FAST predictions with MARIN test data for a semi-submersible
floating wind turbine at 1:50 scale. The focus of the comparison were system global and structural
responses resulting from aerodynamic and hydrodynamic loads. The comparisons indicated that FAST
captures many of the pertinent physics in the coupled floating wind turbine dynamics problem. The
potential areas of improvement for the code FAST and for the experimental procedures to ensure accurate
numerical predictions were also highlighted.
Gueydon and Weller (2013) studied floating foundations for wind turbines and subsequently developed
a numerical model of a semisubmersible floater supporting a wind turbine. The model was calibrated
against model test results, including static loading and decay tests, and the simulation results obtained
684
were compared with the model test results for two sea states without wind: a white noise sea state and an
operational sea state.
Philippe et al. (2013a) and Philippe et al. (2013b) performed a coupled dynamic analysis of a semisubmersible floating wind turbine and compared it with their model test results. Hydrodynamic loads
were calculated using a linear frequency domain approach. In addition to the hydrodynamic damping and
hydrostatic restoring effects, the aerodynamic damping and gyroscopic stiffness were also considered.
The numerical analysis results agree well with the model test results when comparing the motion
Response Amplitude Operators (RAOs) in waves.
Suzuki et al. (2013) developed an analysis code, UTWind, for the assessment of rotor-floater-mooring
coupled response. In UTWind, blades and floater are modelled as frame structures using beam elements,
while mooring lines are modelled as lumped masses. The numerical results by UTWind were compared
with experimental results in wave and wind-wave coexisting conditions with/without blade pitch control.
A good agreement was obtained for these cases. The code also reproduced the response due to negative
damping discussed at design stage.
3.2.3.3 Design and analysis of wind turbine components

The drivetrain in a wind turbine converts mechanical energy from the rotor shaft to electricity via a
generator, and it is an integral part of a wind turbine system. Traditionally, high-speed gearboxes were
dominant drivetrain systems for wind turbines, with a market share above 85% (Kaldellis and Zafirakis,
2012). However, along with the development of offshore wind technology, a wide range of options of the
drivetrain technologies are now available, including medium-speed (hybrid) gearboxes, direct drives,
hydraulic drives, differential, multiple shaft and variable speed gearboxes.
Performance of a mechanical/hydraulic drivetrain must be tested at full scale (not at model scale) using
a test bench in a laboratory. In this manner the global loads on drivetrain shaft can be imposed more
accurately based on the knowledge regarding global behaviour of the whole wind turbine, or alternatively
from field testing of wind turbines (Oyague et al. (2009), Link et al. (2011), Helsen et al. (2011a)). The
high-speed gearbox developed by NRELknown as GRC (Gearbox Reliability Collaborative) 750kWhas
been used in many studies (Xing et al. (2013), Guo et al. (2014)). This drivetrain is of a high-speed
generator, one-stage planetary, two-stage parallel and three-point support type.
With the wind industry moving offshore into deeper waters, the drivetrains must be properly designed
with due consideration of various load effects in order to increase reliability and avoid costly offshore
repairs. Similar to that of a wind turbine system, a consistent approach should be applied, considering all
the phases of the drivetrain life cycle, i.e. design, manufacturing, operation and monitoring, fault
detections, life extension and decommissioning.
The current design approach for wind turbine gearboxes is based on the IEC 61400-4 guideline
(IEC, 2012) which refers to ISO 6336 series of standards (ISO, 2006) for gear components used in
alternative industries. In these design codes, gears are designed using a semi-probabilistic approach
incorporating safety factors or application factors. Wind turbines experience significant dynamic
loads due to wind turbulence, rational rotor excitations and turbine control. These also lead to dynamic
responses in the drivetrain system. A design approach based on direct calculation of load effects is
preferable. The eigen-frequencies of a gearbox (except that of the first torsional mode) are much higher
than the wind turbine excitation frequencies, therefore a decoupled analysis method can be
successfully applied (Xing and Moan, 2013). In such method, the main shaft loads are obtained from a
global analysis model with a simplified 1-DOF (torsional) drivetrain model and then applied as input
to a detailed MBS (multibody simulation) or FE (finite element) drivetrain model (Peeters et al.
(2006), Helsen et al. (2011b), Xing and Moan (2013)), together with the drivetrain accelerations to
represent the inertial loads.
Nejad et al. (2013) performed a long-term extreme value analysis of gear-transmitted loads due to the
main shaft torque with multibody simulations, and a simplified method was demonstrated for the gear
transmitted load calculation. The long-term extreme value of the gear transmitted loads for wind speeds
from the cut-in to the cut-out values was calculated and three statistical methods for long-term extreme
value analysis of the main shaft torque in the offshore wind turbines were presented.
Floating wind turbines may increase dynamic responses in the drivetrain due to motions excited not
only by wind loads but also by wave loads. Xing et al. (2013) studied the dynamics of the 750kW GRC
drivetrain on a spar floating wind turbine. Global aero-hydro-elastic-servo time-domain analyses were
performed using HAWC2, while the commercial code SIMPACK was used to carry out multi-body
drivetrain time-domain analyses. The comparison between a spar-type floating wind turbine and an
685
equivalent land-based turbine indicated that in general there are increases in the standard deviations of the
internal drivetrain responses (such as the tooth contact forces, bearing loads and gear deflections) in the
floating device. These changes are a result of increased main shaft loads in the floating wind turbine,
especially the non-torque loads.
3.2.3.4 Mooring analysis and design

Offshore wind turbines in deeper waters are exposed to stronger and steadier winds which can be used to
improve overall efficiency of the wind farm. However, in water depths greater than 50m fixed support
structures are not a cost-effective solution. During development of such wind farms, existing mooring
technology and numerical models from the offshore oil and gas industry may be applied.
Comparison of an uncoupled and a coupled approach for estimation of semi-submersible FWT
response was achieved by Masciola et al. (2013). In the study, the uncoupled model was formed using
FAST while the coupled model was enabled by linking FAST to OrcaFlex. A key finding of the study
was that the mooring line dynamics had a limited role on the influence of surge and heave motions, but
are significant when observing the tensions in the mooring lines in extreme sea states. The authors also
noted that in the case of a snatch load, there are large differences in the estimation of platform responses
between the two models. Masciola and Jonkman (2014) implemented a dynamic mooring line model into
the open-source Mooring Analysis Program (MAP) for use in the FAST modularisation framework. The
lumped mass model was selected with consideration of external forces due to hydrodynamic loading,
gravity and seabed contact.
Hall et al. (2014a) evaluated mooring line model fidelity for the OC3-Hywind, the ITI Energy Barge
and the NREL TLP type offshore floating wind turbines. The three standard floating wind turbine designs
were implemented and tested using a set of steady and stochastic wind and wave conditions.
An increased probability of slack line events is a particular problem within the design of FWTs (Hsu
et al., 2014). These events usually occur due to a combination of low pre-tensions, lightweight platforms,
shallow water depths and large platform motions in response to a survival storm condition. However,
such events have the potential to occur in various sea states, and are not necessarily limited to extreme
conditions. Hsu et al. (2014) used available model test data for a semi-submersible wind turbine to
investigate how snatch force of the mooring lines may influence the motion and vice versa. There was a
reasonably strong correlation between the tension spikes and wave-induced motion, as well as the vertical
motion of the fairlead. Robertson et al. (2013) noticed that several slack-line events occurred for waveonly and combined wind/wave cases with large wave heights, during the tests of a TLP FWT. Each of the
events was caused by the passing of a large wave that caused the tension in one of tendons to go
negativei.e. the tendon was in compression. The presence of these events, however, merely indicates that
this TLP design was not sufficiently robust.
A parametric study of the dynamic response of spar-type wind turbine with three catenary mooring
lines in irregular waves was undertaken by Jeon et al. (2013). Through numerical experiments, the
time and frequency domain responses were investigated with respect to the total length and
connection position of mooring lines. It was found that the surge and pitch motions were minimised
when the three catenary lines were connected at the centre of buoyancy or slightly above the centre of
buoyancy.
Fylling and Berthelsen (2011) presented a program for conceptual optimisation of FWTs support
structure of the spar buoy type, including the mooring system and power cable. In their approach, the
optimisation variables were spar buoy shape, height and diameter of column sections, mooring line and
power cable data and vertical position of mooring line fairleads. A nonlinear optimisation approach with
arbitrary constraints was used where the objective function was the spar buoy, mooring line and power
cable costs. A mooring system optimisation procedure for FWTs using frequency domain analysis was
proposed by Brommundt et al. (2012). The work emphasised the cost reduction of catenary mooring
systems for the semi-submersible wind turbine type. Benassai et al. (2014) conducted a similar study
focused on minimising the catenary mooring system weight of a FWT with a tri-floater, semi-submersible
support structure, with reference to ultimate and accidental loads.
Two main lines of research are apparent in recent studies with respect to mooring system for FWTs.
The first focused on development and validation of mooring system numerical models within the model
of FWT response. In general, development started from implementation of quasi-static models taking into
account only non-linear restoring forces of the mooring system. Subsequently, more sophisticated models
capable of describing dynamics of mooring lines have been developed based on experience from the oil
and gas industry. Within this line of research special attention has been given to slack line events of
686
mooring systems. The second line of research considers optimisation of the design process of the FWT in
order to reduce the cost of mooring systems.
3.2.3.5 Challenges for design and analysis of floating wind turbines

Refined versus simplified analysis methods
The majority of the coupled tools for analysis of floating wind turbines, discussed in the previous
sections, are applied in time domain in order to capture nonlinear effects from aerodynamic or
hydrodynamic loads, nonlinear geometrical effects due to large motions, and automatic control. Further
development and validation of these numerical tools should be considered in the future.
However, it is also interesting to develop simplified methods which are more useful at a preliminary
design stage, in view of a large number of simulations needed for long-term response analysis for ultimate
and fatigue design check, or for design optimisation.
Karmakar and Guedes Soares (2013) performed dynamic analysis using the environmental contour
method to estimate the long-term extreme loads for semi-submersible offshore floating wind turbine.
For fatigue limit state design, efficient and accurate analysis methods are required. Kvittem and Moan
(2015) compared the frequency domain models for the tower base bending moment with fully coupled,
nonlinear, time domain simulations. The predictions from this model agreed reasonably well with the
results from a fully coupled, nonlinear analysis in Simo-Riflex-AeroDyn, and it was assumed that
bending moments in the tower base were primarily caused by platform motions, accelerations and rotor
forces, and that fluctuations in the axial force were negligible.
Using the basis function approach Hall et al. (2014b) studied the optimisation of a FWT using a
hydrodynamics-based method. A frequency-domain model was used to evaluate the design space,
applying linear hydrodynamics for the floating platform and linearized representations of the wind turbine
and mooring system. Optimisations were run to find the platform design that minimised nacelle
accelerations for two different mooring systems, and reasonable results were obtained.
Behaviour of floating wind turbines under fault conditions

Modern large-scale wind turbines typically have variable-speed, pitch-regulated, upwind rotors. The
controller is designed and implemented to maximise the power extraction efficiency for below-rated wind
speeds via generator torque control, and to mitigate loads and achieve constant power output for aboverated wind speeds via blade pitch control (normally collective control of the three blades). Faults, or
failures, in wind turbine components (both hardware and software) may lead to less power production and
increased dynamic loads in the system. Although IEC 61400-1 (IEC, 2005) and 61400-3 (IEC, 2009)
define design load cases involving wind turbine faults, no detailed procedures for modelling and analysis
of faults, and the subsequent system responses, are specified.
Faults do, however, occur in wind turbine systems. The EU ReliaWind project analysed the long-term
operational data and fault records of 350 onshore wind turbines (Wilkinson and Hendriks, 2011) and
presented a reliability profile for wind turbine components. Among all of the components, the blade
pitching system has the highest failure rate. Pitch controller faults may reduce power production and
increase dynamic loads. Pitch actuator faults have the potential to cause a blade to be stuck at a certain
pitch angle and may increase the aerodynamic loads on that blade. For floating wind turbines, excessive
motions can be excited by the unbalanced aerodynamic loads on the rotor plane due to pitch actuator
faults (Johnson and Fleming (2011), Jiang et al. (2013)). This also increases the loads on the drivetrain of
a floating system, as the bending moments on the main shaft are strongly affected (Xing et al., 2013). For
emergency shutdown, it is common practice to pitch all of the blades to feather simultaneously at a
maximum pitch rate. The presence of pitch system faults will influence the effectiveness of emergency
shutdown.
So far, there has been limited research carried out in this direction and future work on wind turbine
faults is imperative. This may include physical understanding and numerical modelling of faults,
assessment of the effects of faults on system responses, effective alleviation methods or procedures,
condition monitoring for fault diagnosis and detection and fault-tolerant control.
Floating vertical axis wind turbines
Recently, there has been a resurgence in interest in vertical axis wind turbines (VAWTs) for offshore
applications with floating foundations. The EU DeepWind project (http://www.deepwind.eu/TheDeepWind-Project) considers a novel spar-type vertical axis wind turbine, with focus on the development
of tools for optimisation and design and the verification against model test results.
687
Wang et al. (2014) discussed the potential application of a VAWT on floating foundations. There are
several advantages to the use of a VAWT: power generation is independent of wind direction, they have a
low CoG for improved hydrodyanmic performance, and access to the drivetrain and generator is
facilitated for improved maintenance and repair strategies. However, compared to an equivalent 5MW
HAWT, the large yaw motions induced by the rotor torque impose challenges for design of the mooring
system.
Borg et al. (2014a) developed a numerical model for VAWT analysis. Aerodynamic theories suitable
for VAWTs were discussed. Borg et al. (2014b) focused on modelling of mooring systems and structural
behaviour of floating VAWTs, discussing various mathematical models and their suitability within the
context of developing a model of coupled dynamics. The advantages and downfalls of each model were
noted, from the simple force-displacement relation to the nonlinear finite element model.
VAWT development thus far is limited when compared to that regarding HAWTs. Several directions
might be considered in the future work: further development and validation of numerical tools,
development of an advanced control strategy to reduce power variation and dynamic loads, development
of safe procedures for start-up and shutdown and novel floater and mooring design to reduce system
motions, in particular yaw motions.
3.3
Physical testing
Laboratory testing at small scale is an important part of conceptual study for novel concepts of offshore
wind turbines. It enables detailed investigation of underlying aerodynamic and hydrodynamic
phenomena. Field testing of large scale or prototype of offshore wind turbines is also a crucial step to
demonstrate both the technological and economic feasibility for further development towards a full
commercialisation.
Both laboratory and field tests are very valuable for validation of numerical tools if the measurements
of both environmental conditions and system responses are properly performed. Laboratory testing has
the advantage of being well-controlled and having repeatable environmental conditions. For example,
extreme wind and wave design conditions can be re-produced in the lab, but they might not occur at sea
within a limited amount of testing time. However, the generation of wind fields with good quality in open
space (for example in a wave tank) is difficult. The scaling effect introduces another uncertainty to
interpret the laboratory test results. For an offshore wind turbine, it is important to scale both the
aerodynamic loads according to Reynolds law and the hydrodynamic loads using Froudes law.
However, it is not possible to simultaneously satisfy both scaling laws (Mller et al., 2014). For prototype
tests at full scale, there is no scaling problem, but the measurements of both wind and wave fields in
particular might not be accurate and sufficient for validation of numerical tools. Low long-term reliability
of the instrumentations might be another problem for testing at sea. For drivetrains and generators, fullscale tests are preferred.
3.3.1
Laboratory testing
Recent laboratory tests of offshore wind turbines are primarily concerned with floating concepts: spar,
semi-submersible and TLP. While model tests of bottom-fixed offshore wind turbines focused on
nonlinear wave loads on support structures in extreme wave conditions.
In order to study the dynamic responses of a monopile wind turbine under breaking wave loads, a
model tests using a representative 1:30 wind turbine tower and monopile with realistic flexibility was
performed in MARINs Shallow Water Basin (De Ridder et al., 2011). The water depth was 30m and the
diameters of the tower and the monopile were 5.4 m and 6m at full scale, respectively. The 1st and 2nd
bending modes were correctly modelled by adjusting the distribution of bending stiffness and weight
along the tower height. The test involving a focused breaking wave revealed that the horizontal
acceleration at nacelle height can reach up to 0.9g (g = 9.81 m/s2). This implies significant loads on the
wind turbine components, such as the drivetrain.
As part of the Danish Wave Loads Project, model tests of a flexible pipe in steep and breaking waves
were performed. The 1:36.6 test pipe represented the NREL 5MW monopile with a diameter of 6m at full
scale in a water depth of 20m. The test results were used to validate three numerical methods (Paulsen
et al., 2013), a linear potential flow solver, a nonlinear solver and a CFD method (the Navier-Stokes/VOF
solver). In general, the linear solver deviated significantly from the experiments for nonlinear steep
waves. The nonlinear solver performed equally well as the CFD method for most of the cases, but for the
near-breaking and breaking waves, the wave force is more accurately captured by the CFD method due
the ability to handle wave breaking.
688
A summary of model tests on floating wind turbines was presented by Mller et al. (2014). In most of
the model tests the floater was geometrically scaled, yet still modelled as one rigid body, while the
flexibility of the tower was realised in some tests in order to capture the elastic response. Mooring lines
were represented simply as springs, or as physical lines including the dynamic effect as well.
While model tests can be used to examine the uncertainties related to hydrodynamic and aerodynamic
load prediction for isolated effect, the lack of consistent scaling laws make it difficult to investigate the
effect of simultaneously acting wave and wind loads. Often very simple models of the aerodynamics have
to be applied in the model tests (Coulling et al., 2013). There exist a variety of methods for modelling the
rotor. A geometrically-scale rotor at model scale would produce less thrust force due to lower lift and
higher drag coefficients as compared to the full-scale rotor (Fowler et al. (2013), Philippe et al. (2013a)).
Therefore a re-design of the rotor might be necessary in order to achieve the corresponding thrust curve
(Martin et al., 2012). Alternatively, the rotor may be modelled simply as a disk to reproduce the thrust
force (Roddier et al., 2010), or a controlled fan providing active force in order to mimic the thrust force
(Azcona et al., 2014).
Roddier et al. (2010) discussed the development of the WindFloat concept and highlighted the
challenges for any offshore floating wind turbine foundations, related to turbine design considering
floater induced motions, fabrication and installation procedures and costs. Model tests were performed in
a wave tank facility, using fans and a drag disk placed on the model to generate wind loads and a rotating
rotor behind the disk to model gyroscopic effect. The numerical hydrodynamic results were compared to
the experimental measurements.
A series of model tests of three floating wind turbine concepts (spar, semi-submersible and TLP) at
1:50 scale were performed at MARIN through the DeepCwind project, see Figure 7. The main objective
of this research project is to develop a proper scaling methodology, to realise a model wind
turbine for laboratory testing and to validate numerical codes (Fowler et al. (2013), Martin et al. (2012),
Robertson et al. (2013)). Robertson et al. (2013) summarized the results and the experiences obtained
from the first phase of the tests as well as the comparison against the numerical code FAST. It was found
that using geometrically scaled rotor blades produced too low thrust force which is of significant
importance for pitch motion of the floating wind turbines, and therefore increased wind speeds need to be
considered in order to achieve the target thrust force. The comparison between the FAST simulations and
the test results showed a good agreement in general, but also revealed the importance to include a secondorder hydrodynamic model as well as a dynamic mooring line model in FAST. The same approach was
used in the test of a semi-submersible with brace systems in the Hydrodynamics and Ocean Engineering
Tank at ECN (Philippe et al., 2013a).
Figure 7. The three floating wind turbine concepts tested at MARIN (left: spar; right-top: TLP; right-bottom: semisubmersible) (Robertson et al., 2013).
In order to overcome the problems related to a geometrically scaled rotor, a performance-matched wind
turbine model with re-designed blade geometry was developed at University of Maine (Martin et al.,
2012) and built in 2013 at MARIN (De Ridder et al., 2014). The new turbine model can produce the
689
target thrust coefficients of the NREL 5MW wind turbine at model scale for Froude-scaled wind speed.
Together with an active pitch control mechanism and a new high-quality wind generator setup, this
turbine was then used in the additional tests of the DeepCwind semi-submersible wind turbine (Goupee
et al., 2014) and in the tests of the GustoMSC Tri-Floater at MARIN (Huijs et al., 2014).
Model tests of floating wind turbines have also been performed at other facilities with focus on
hydrodynamic behaviour. Several tests have been conducted considering different designs of spar-type
hull (Shin et al. (2013), Sethuraman and Venugopal (2013)). A 1:100 semi-submersible with a singlepoint mooring system was tested at the Ocean Basin of Institute of Industry Science, at the University
of Tokyo. Model tests of a TLB for offshore wind application were conducted by Myhr et al. (2011).
Wehmeyer et al. (2013) performed a model test at 1:80 scale for a tension leg moored floater as
support of a floating wind turbine in the 3D deep water basin of the Hydraulics and Coastal
Engineering Laboratory at the University of Aalborg. The test focused on the ULS behaviour and it
was shown that the inclusion of high-order hydrodynamics was essential, especially regarding TLPspecific responses such as ringing and slack line events. Calibration of a scaled finite element model
by comparing the scaled natural periods of a measured TLP system show that the hydrodynamic added
mass coefficients for complex structures are in contrast to the literature as per Adam et al. (2013) and
accepted design codes (e.g. DNV-GL). This was traced back to a semi-transparent characteristic of the
structure.
In addition, hybrid model testing has been undertaken in order to try and avoid the scaling problem, in
which either aerodynamic loads (Hall et al., 2014c) or hydrodynamic loads (Bottasso et al. (2014), Bayati
et al. (2014)) are calculated by numerical codes and reproduced by real-time controlled actuators.
However, such testing technique relies strongly on the accuracy of the applied numerical codes and the
response characteristic of the actuators. Further validation of such methods needs to be conducted.
3.3.2
Field testing
There exist several intermediate-scale floating wind turbines which are currently under test at sea. Such
tests are usually developed with the aim of gathering information for the development of large full-scale
floating turbines. The 100kW downwind spar floating wind turbine was launched in June 2012 near
Kabashima Island, Japan. Since then, the test turbine has been attacked by several severe typhoons and it
survived with no damage. The study by Utsunomiya et al. (2013) and Ishida et al. (2013) also showed a
good agreement between the numerical simulations and the field measurements for motion and structural
responses of the spar wind turbine in extreme wind and wave conditions.
A 1:8 scale prototype of a 6MW, semi-submersible, floating wind turbine (VolturnUS) was deployed
off the shoreline of Castine, Maine, USA in June 2013. The turbine has a rated power of 12kW.
Comparison of the measurement data with numerical results using FAST shows that the numerical code is
suitable for modelling a nearly full-scale floating wind turbine in both operational and extreme design
environments (Viselli et al., 2014).
There are two full scale, floating wind turbine prototypes that are currently being tested at sea; the
2.3MW Hywind deployed in 2009 and the 2MW WindFloat deployed in 2011. These two prototypes
have gone through several years of testing, and have demonstrated the feasibility of floating wind turbine
technology. In particular, the power performance of these two floating turbines are in excess of
expectations (Nielsen (2013), Maciel (2012)).
The capacity factor of Hywind was in the order of 4454% for the northern North Sea conditions
(Nielsen, 2013), much higher than 30-35% of typical offshore bottom-fixed wind turbines in the North
Sea. In addition, comparison with the measured data has shown the capability of numerical tools to
capture structural responses of the Hywind with a reasonable level of accuracy. The next step for Hywind
developer, Statoil, is to optimise the floater to support larger turbines in order to achieve a cost-effective
solution for a small farm of floating wind turbines.
However, there is no detailed technological information publically available regarding these projects as
they have been developed by industrial companies as opposed to research establishments. There is
therefore a need to develop research infrastructures for testing of full-scale wind turbines through which
the community at large can have full access to the measurement data in order for development of the
industry to progress.
3.4
Transportation, installation, operation and maintenance
Transportation, installation, operation and maintenance of offshore wind turbines has become an
important part of the economics associated with offshore wind farms. The key for cost reduction is to
690
increase the efficiency and the workability of relevant vessels that are involved in these phases;
however, in general, there exists limited research work published in this area. The focus in this section
of the report is relating to transportation and installation of offshore wind turbine components
(foundation, tower, nacelle and blades). Currently, installation of these components (individually or in
a pre-assembled manner) is carried out primarily using large jack-up vessels, or in some cases floating,
heavy lift, DP (Dynamic Positioning) installation vessels. New transportation and installation methods
and vessels have been proposed recently. It should be mentioned that wind turbine operation and
maintenance are not covered by this report. These aspects are discussed in detail in Rademakers et al.
(2011).
3.4.1
Current industry and research development
New transportation and installation vessels, methods and concepts

Kim et al. (2012b) described a jack-up wind turbine installation vessel as the worlds first vessel which
was a self-propelled and self-elevating platform. The vessel was designed to carry out lifting operations
using cranes onboard in jacked-up mode for water depths of up to 45 m, and also serve as transportation
and storage unit carrying wind farm structures and consumables.
Norwind Installer and Ulstein (stvik, 2012) have designed a new installation vessel incorporating a
DP system. The DP vessel is designed for world-wide operations with a focus on pre-piling and
jacket/tripod/transition piece installation for the offshore wind industry in North Europe. The vessel was
designed for maximum efficiency and cost effectiveness, and features an 800ton heave compensated
offshore crane on the starboard side, while a pre-piling template can be located on a support structure at
the stern. The vessel can carry 4 jackets each trip, or alternatively at least 24 piles or up to 12 transition
pieces on the large open aft deck.
Saipem SA proposed a new offshore wind turbine installation device Castoro Vento vessel (Ruer
et al., 2009). The vessel is based on the float-over technology which has been used in the offshore oil and
gas industry. The vessel is a U-shaped barge equipped with several ballast water compartments to allow
ballasting/de-ballasting operations. It is designed to transport wind turbines up to 6MW, for installation in
water depths between 2535 m. The vessel can carry a fully assembled turbine including the foundation
(gravity-based, steel tripods and jacket structures) to the offshore site and then install it on a pre-prepared
seabed utilising ballasting operations. Therefore, no offshore lifting operations are needed.
The Merlin concept (Bland, 2004) was proposed to install fully-assembled wind turbines offshore. The
complete turbine is intended to be lifted out onto a bespoke installation barge for transportation to the
offshore site where it is rotated to the vertical position and connected to a preinstalled foundation in a
single operation. The analysis results showed that the Merlin system could operate in higher sea states
than most conventional turbine installation vessels.
The Dutch company Ulstein Sea of Solutions and SPARCS have developed a new wind turbine
concept, called F2F (floating to fixed), aiming to reduce the high offshore installation costs associated
with the offshore wind industry (Ulstein, 2012). The concept is constructed, pre-assembled and
commissioned onshore, then floated to a stable working draft and towed to the offshore location. When
arriving at the location, it is lowered to the sea bed through a ballasting system and fixed in place by
suction anchors. When maintenance and repair is required, it can be re-floated again and subsequently
towed to shore.
Reinertsen AS have proposed a novel concept of a self-installing offshore wind turbine with gravitybased foundation (Wsj et al., 2013). The concept combines installation of the substructure and turbine
in one single operation by using two structurally connected standard barges. A cost saving potential in the
early phase of 17% was identified compared to a steel jacket solution.
From the concepts described above, it can be seen that there is a tendency to develop new vessels/methods which facilitate the installation of fully-assembled wind turbines thereby avoiding risky and
weather-sensitive offshore lifting operations. However, the studies on these new concepts are at a very
early stage of development, and more effort into detailed design and analysis should be performed in order to prove the feasibilities of the different concepts.
Transportation and installation of floating wind turbines
Up to now, only a few wind turbines in the world exist that are supported on a floating support structure.
The installation methods for floating wind turbines are quite different from those for bottom-fixed wind
turbines due to increased water depths. For turbine installation in shallow water, jack-up vessels are
normally used to provide stable platform for lifting and offshore bolting, which could hardly be carried
691
out by a floating vessel with motions in six degrees of freedom. However, for deep waters, if onshore
assembly is chosen, the installation can only be performed by floating vessels, which is in itself
challenging and costly. In order to reduce installation costs and to make floating wind turbines more costeffective, innovative installation strategies are required.
The Hywind demo (Statoil, 2014) has draft of 100 meters and installation of the turbines on top of the
spar in shallow waters is not possible. Therefore, the spar hull was wet-towed horizontally to a near shore
fjord, where it was temporarily moored and then upended through ballasting. An offshore heavy lift
vessel was subsequently used to transport and assemble the turbine components. The fully-assembled
wind turbine was then towed by three vessels to the offshore site, where it was connected to the preinstalled mooring lines.
The WindFloat prototype (PrinciplePower, 2014) was the first offshore floating wind turbine that was
deployed without the use of any offshore heavy lift vessel. Due to the low draft of the semi-submersible, the
complete structure was fully assembled onshore prior to being towed offshore. Once at the offshore site, a
conventional catenary mooring system was connected to the floating wind turbine.
WindFlip is a specialized barge for transporting and installing spar type floating wind turbines
(WindFlip, 2013). One fully-assembled turbine, lying in a nearly horizontal position on top of the deck of
the barge is transported at a time. After the barge has been towed to the offshore site, it is slowly ballasted
down, and the barge with the attached turbine flips 90o until both are in the vertical position. The design
of the barge is optimised to maintain stability throughout all phases of the flipping process. The turbine is
then connected to a pre-installed mooring system.
For TLP wind turbines, the key challenges are related to the installation of anchors and tendons, as
well as towing of the TLP floater itself.
3.4.2
Numerical simulations of marine operations
For validation and improvement of any installation method, careful numerical and experimental studies
are important. Numerical simulation may help to identify critical phases of different marine operations,
prior to the execution of these operations at sea. It will also provide a better basis to establish rational
limiting criteria for safe operation, and to evaluate operational weather windows and associated
probability of success. In a broader sense, numerical simulations in combination with weather forecasting
could also be useful for decision-making in actual operations at sea. A variety of marine operations might
be studied by numerical models. The focuses here are sea transportation analysis, lifting operation and
docking operation of service vessels.
Donaire (2009) presented a program to perform the sea transportation analyses for a given barge and
wind turbine configuration, which includes two modules: the sea keeping model of the barge and the sea
transportation model considering wind turbine aero-elastic responses. The criterion to check the
performance of the transportation was chosen as the acceleration limitation at the turbine nacelle, from
which the limiting weather condition can be determined.
Graczyk and Sandvik (2012) performed a numerical study of lift-off and landing operations for wind
turbine components on a ship deck using a frequency-domain approach for ship motion analysis. Two
different vessels (one large supply monohull vessel and one small catamaran vessel) were considered in the
comparative study. A limiting acceleration of the lift object was defined and used in combination with the
probability distribution of sea states to determine the weather window for such operations.
Wu (2014) performed numerical analyses of docking operations between service vessel and offshore
wind turbine for personnel and equipment transfer also using a frequency-domain method. In one case, a
medium-size service vessel with motion compensation devices was considered. The limiting criteria for
operability analysis of docking operation were defined as the compensation limits of the 6 degrees-offreedom relative motions, velocities and accelerations between the docking device and the offshore wind
turbine. In the second case, a small service vessel with a simple fender was considered. The limiting
criteria were defined as no slip between the service vessel and the wind turbine tower at the touch point,
which implies that the vertical force should be smaller than the friction force. The proposed frequencydomain approach was shown to be much more efficient than a time-domain method for the analysis of
long-term operability.
Li et al. (2014) performed a detailed numerical study on the crane operation for lowering the monopile
and jacket substructures through the wave splash zone to the sea bed using a floating installation vessel.
Transient time-domain analyses of the coupled system of the installation vessel and the lift object were
carried out considering the submergence-dependent hydrodynamic loads on the substructure. In
particular, the vessel shielding effect (that is the reduction of the wave kinematics on the leeward side of
692
the vessel) was studied and found to have a significant effect, depending on the vessel heading angle, on
the dynamic responses of the substructure motions and lift wire tension.
Sarkar and Gudmestad (2011) and Sarkar and Gudmestad (2012) proposed a novel method to install
monopiles by isolating the installation operation from the motion of the floating vessel using a preinstalled submerged support structure. A new substructure was designed to support the monopile against
waves and currents during the initial phase of driving into the seabed. A simplified numerical model in
Riflex was established to study the responses of the structures during the driving operation. The analysis
results indicated the submerged support structure probably can increase installation weather window as
compared to the existing installation technology.
3.4.3
Guidelines on marine operations for offshore wind turbine transportation, installation,

operation and maintenance
There are no specific design rules or guidelines regarding marine operations for offshore wind turbine
transportation, installation, operation and maintenance. The rules and guidelines established in the offshore
oil and gas industry are often utilised, since they are also applicable to similar operations in the offshore
wind industry. The objective of these guidelines is to provide methods and simplified formulations for
establishing design loads to be applied for planning and execution of marine operations.
DNV Recommended Practice H301 (DNV, 2010) gives guidance on modelling and analysis of marine
operations, in particular for lifting and towing operations. The GL Noble Denton Technical Guidelines
and Rules also cover many marine operations, including load-outs, marine transportations (GLNobleDenton, 2013a), marine lifting operations (GL-NobleDenton, 2013b), installation of concrete
offshore gravity structures (GL-NobleDenton, 2013c) and jacket structures (GL-NobleDenton, 2013d).
The NORSOK Standard J-003 (NORSOK, 1997) defines the basic requirements for vessels performing
marine operations and for the planning, execution and work associated with such operations on the
Norwegian Continental Shelf.
3.5
Rules and standards
Current standards or technical specifications for design of offshore wind turbines and their substructures
essentially consist of the following documents for bottom-fixed wind turbines:
IEC 61400-3 (IEC, 2009);

DNV-OS-J101 (DNV, 2013a);
GL (IV Part 2) (GL, 2005);
ABS #176 (ABS, 2013a);
and for floating wind turbines:
DNV-OS-J103 (DNV, 2013b);

ABS #195 (ABS, 2013b);
BV NI 572 (BV, 2010);
ClassNK Guideline (ClassNK, 2012);
Although there exist design standards for floating wind turbines, the industry is still in an early stage of
development, therefore further development and revision of these standards is necessary when the
relevant industry experience is available.
A recent assessment of standards has been provided by the NREL (Sirnivas et al., 2014) which,
although more focussed on US applications, includes review of institute standards API, IEC and ISO and
class society guidelines published by ABS, BV, DNV and GL. These standards and guidelines are
continually being assessed and updated (Karimirad (2014), Negro et al. (2014), Perveen et al. (2014),
Woebbeking and Argyriadis (2013)).
The most common rules to certify offshore wind turbines are the IEC 61400-3 rule and the DNV-GL
guidelines. The current edition of the IEC 61400-3 rule considers bottom-fixed offshore wind turbines.
IEC has established a Technical Committee (TC) 88, developing the design rule for floating wind
turbines. The guideline for floating sub-structures, however, is not yet published; a technical specification
of IEC 61400-3-2 will be published in the near future.
693
The DNV-OS-J101 rule deals with offshore wind turbines with fixed substructures (monopiles and
jackets). The DNV-OS-J103 rule regards to wind turbines with floating substructures. Similar to the
DNV-OS-J101 standard, the GL (IV Part 2) handles with offshore wind turbines for fixed sub-structures.
After the merge of DNV and GL in December 2013, all of the rules will be compiled and subsequently rewritten.
Most of the other standards, both national and international, have been developed in line with IEC
61400-1 (IEC, 2005) and 61400-3. For example, ABS #176, also an international standard, issued first in
2010 and revised in 2013. It is mostly aligned with IEC 61400-3.
ClassNK (2012) published Guideline for Floating Offshore Wind Turbine Structures in July 2012. It
applies to unmanned floating structures with a 20-year design life or over and to the tower and RNA as
options. The guideline was developed referring to IEC 61400-1 and IEC 61400-3 and the related ClassNK
Rules for the Survey and Construction of Steel Ships. The design principle implied in the guideline is to
maintain the structural integrity of floating structures throughout their design lifetime without the need for
docking and repair, and to consider the absolute minimum of damage stability and one line breaking in
order to prevent drifting.
4.
WAVE ENERGY CONVERTERS
Wave energy conversion technology has a large potential to contribute to the worlds renewable,
emission-free energy supply. A recent comprehensive study by the Electric Power Research Institute
(Jacobsen, 2011) concluded that the recoverable wave energy resource along the coastlines of the United
States is in the order of 1,000TWh/yr, which represents nearly 1/3 of the total electricity currently used in
that country. Similar potentials exist for many countries around the world. WEC technology is still
relatively immature when compared to wind, water, solar and even tidal power extraction, and it thus
faces significant hurdles to achieving economically competitive designs. This chapter will attempt to
summarise the main commercial and research developments that have taken place since the last ISSC
report in 2012 (ISSC, 2012).
In the past three years, the number of test centres, prototype and pre-prototype wave power
installations has exploded world-wide so that they are now too numerous to list in this condensed report.
However, the International Energy Agency has run a Technology Initiative called Ocean Energy Systems
(IEA, 2014b) since 2001, and they maintain a website listing world-wide Ocean Energy installations
including Wave, Tidal, Thermal Gradient, Salinity Gradient and combined installations. As of the end of
2013, there were 22 open-sea testing facilities around the world, 15 in Europe, 5 in North America, and 1
each in Asia and Oceania. The total installed and approved-for-installation wave energy power is listed at
1.5MW in North America, 76MW in Europe, 4MW in Asia and 43MW in Oceania.
The European Commission has established an industry trade association for ocean renewable energy with
the name Ocean Energy Europe (OEE, 2014). They have published an Industry Vision Paper which calls
for a dramatic expansion of ocean renewable energy over the next forty years culminating in up to 100GW
of online wave energy by the year 2050. The group outlines a number of barriers to achieving this goal, but
the EC has also established an Ocean Energy Forum run by European Energy Ministers and senior industry
figures which is actively working to advance the industry on three critical fronts: technology; finance; and
environment/consenting. These developments suggest the possibility of a dramatic expansion in the WEC
industry over the next few decades.
Concepts and development status
In contrast to wind and tidal energy, wave energy extraction devices span a large range of different
concepts with over a hundred different designs being proposed over the years, many of which are
currently under active development. Different types of WEC concepts have been studied, including:
Attenuator;
Point absorber;
Wave surge converter;
Oscillating Water Column (OWC);
Overtopping or Terminator;
Cycloidal lifting surface or cycloidal turbine wave absorber;
Submerged pressure differential;
Other.
694
According to Falco (2010), these devices can be classified into three main categories based on the
working principle, namely oscillating water column, oscillating bodies and overtopping. Falco (2010)
also listed the projects that have reached the prototype stage, or at least have received extensive
development efforts, Figure 8.
Figure 8. Wave energy technology classification (Falco, 2010).
Despite this variety of concepts, the differences in scaled metrics, such as absorbed power per device
volume or absorbed power per device surface area, are surprisingly small for a large number of designs
(Babarit et al., 2012). One device, however, stands out in terms of the theoretically predicted absorbed power
per device volume; the cycloidal lifting surface device originally invented by researchers at Delft University
of Technology in the Netherlands (Hermans et al., 1990) and is currently being commercialized by Atargis
Energy Corporation (Seigel, 2012). While predicting the ultimate COE for any WEC device is still rather
uncertain, it seems clear that no radical breakthroughs have emerged over the past three years.
There has however, been much activity in this time. The FP7 project MariNet has allowed for a large
number of ocean energy devices to be tested at the ever increasing network of European facilities. Several
of these projects deal with wave energy, and they are listed on the webpage http://www.fp7marinet.eu/access_user_projects_wave_energy.html. There are currently twenty-nine projects recorded
here. The development stage of each project is also stated, using the convention:
1)
2)
3)
4)
5)
Concept validation;
Design validation;
Sub-system validation;
Solo device validation;
Multi-device demonstration.
Only one of the concepts is at stage 5, two at stage 4 and the rest between stages 1 and 3.
The most critical topics of validation for these tests are:
-
Power production in real sea conditions (scaled duration at 20-30 minutes) (11 tests);
Mooring arrangement and its influence on the performance (9 tests);
Initial estimates of the full system load regimes and accurate simulation of PTO characteristics (8
tests);
Investigation of the physical processes governing the device response (some of which may not be well
defined theoretically nor numerically solvable);
Survival loading and extreme motion behaviour (7 tests).
695
Even though this collection is not exhaustive of the whole panorama worldwide, it gives a picture of the
main topics of research and an idea of the current stage of development for WEC technology.
Although the technology is still rather young, a number of developers claim that their ultimate COE
will be competitive with, or superior to, offshore wind energy once the mass-production stage has been
reached. Verification of these claims will require continued, and ideally increased, research and government support for this emerging technology.
4.1
The general approach to the analysis of a WEC concept is to apply a sequence of increasingly more
sophisticated models to capture more and more of the relevant physics, depending on the stage of
development of the concept. Potential flow models are used in the initial stages, often starting with a
purely theoretical model, followed by numerical calculations using a linear frequency-domain analysis
and finally moving to a weakly-nonlinear time-domain formulation which can include the most important
nonlinear effects from in particular the PTO and/or mooring system. These calculations are usually based
on solutions using the Boundary Integral Equation Method (BIEM), as exemplified by the panel code
WAMIT (Lee, 1995). In the case of simple geometries (e.g. vertical axisymmetric bodies), analytical
solutions can be used as well (Mavrakos and Konispoliatis, 2012). This analysis is relatively fast and is
suitable for refinement and optimisation of all aspects of the concept from the geometric shape to the
mooring system and the PTO control strategy. Suitable models should also be included here for important
viscous and possibly compressibility effects via empirical viscous damping coefficients and for OWCs air
flow models for the turbine (see for example Conde and Gato (2008)). Recent examples of this analysis
are reported by Rhinefrank et al. (2013) and Sheng et al. (2014).
In the later stages of design, fully nonlinear potential flow models and/or Euler or Navier-Stokes
solvers (CFD) can be used to capture wave breaking and other highly nonlinear phenomena so as to
perform ULS and survivability calculations. The computational effort for these more refined models is,
however, very high and relatively few examples can be found in literature (Luo et al. (2014), Hu et al.
(2011), Ashkan et al. (2013)). A recently completed Danish research project described at
http://www.sdwed.civil.aau.dk/ collects a number of tools available for this purpose, with Li and Yu
(2012) providing a description and an example of the basic analysis and design procedure.
Overtopping and bottom hinged devices are especially difficult to analyse, as they are generally
strongly dependent on wave breaking and highly nonlinear motions which can undermine their
survivability. For these devices, CFD solutions are normally required, but due to their prohibitive cost, it
is important to first identify the critical conditions using potential flow to provide initial conditions for
more comprehensive but more computationally expensive solvers.
4.1.1
Load and motion response analysis
Similar to the dynamic analysis of offshore wind turbines, the purpose of numerical modelling and
analysis of a WEC, or an array of WECs, is to address on one hand the serviceability (power production)
and on the other hand the safety (structural design with respect to ULS and FLS). Accurate numerical
models are of great usefulness and importance to predict and maximize power production, to make
acceptable and optimal component and structural design, and therefore to achieve reasonable economics
of WECs. In this section, we reference and briefly discuss a number of recent publications collected into
several broad focus areas.
Single device analysis
Li and Yu (2012) applies a full analysis procedure to a point absorber concept, running through all the
stages of analysis.
Babarit et al. (2012) present a numerical benchmark study of eight generic WEC concepts, including
seven oscillating bodies and one oscillating water column. In order to make a fair comparison of the devices,
three performance measures of the annual power production per unit cost were adopted based on total mass,
surface area and maximum PTO force. Time-domain numerical models were used for the analysis of global
motions and power output, based on the linear frequency-domain hydrodynamic coefficients, but including
the nonlinear effects from empirical viscous forces, PTO forces and end stops. Among the eight selected
concepts, the bottom-fixed oscillating flap was found to be the most promising, though the differences in the
total COE for the various devices was small.
Jeans et al. (2013) studied the performance of a 2D lift based cycloidal WEC (CycWEC) in unidirectional
irregular, deep water waves, as a follow-up to their previous numerical and experimental work on the same
696
concept in regular wave conditions. This concept consists of two hydrofoils attached by a cylindrical
framework parallel to a horizontal main shaft. The structure is submerged close to the free surface and
aligned parallel to the incoming wave crests. Potential flow theory is used, and each hydrofoil is modelled as
a point vortex moving under a free surface. The hydrofoil position and bound circulation are controlled
based on a feedback flow control using a sensor located up-wave of the device and wave state estimator, to
achieve high efficiency for irregular waves. Their results show that the capture width ratios for incident
wave fields consisting of 7 and 10 main regular wave components were 0.85 and 0.77, respectively. These
results were validated by 1:300 scale experiments, which also measured a capture width ratio of 0.77 for
similar irregular wave conditions.
Sakai et al. (2014) considered a Salter Duck-type device which uses a rotating internal pendulum as the
PTO. Linear frequency-domain calculations predict that the device can achieve a capture width ratio
between 0.75 and 1.
Lovas et al. (2010) used linear potential flow theory to evaluate the performance of a shore
mounted, circular OWC at the tip of a coastal corner of any angle. A semi-analytic solution was
obtained using eigen-function expansions, to give relatively simple solutions for the hydrodynamic and
power performance. The energy extraction efficiency for a convex and a concave corner were
compared to existing solutions for a thin breakwater and a straight coastline. The inherent advantage of
the pneumatic system was explored to show that the column radius should not be too small to yield a
broad bandwidth of high efficiency. In addition to the effects of system sizes, they have introduced a
simpler strategy of optimisation by assuming that the parameter representing the PTO device can take
two different values, one for the high-frequency range and one for the low-frequency range. This
strategy was found to be comparable to the theoretically ideal strategy where the parameter can be
controlled for any frequency.
Fonseca and Pessoa (2013) considered a floating OWC concept with a U-shaped chamber using a
coupled model for the floater and the interior chamber motions. By tuning the natural periods of the roll
motions and the interior chamber, they were able to obtain an average efficiency in irregular waves of
0.47. The average efficiency was defined as the ratio between the average air turbine power output and
the average wave energy flux associated to a wave front with the width of the device (15m). The
simplicity of this concept suggests that it could have a good COE.
Luo et al. (2014) presented an experimentally validated two-dimensional, fully nonlinear CFD model
to analyse the efficiency of a fixed OWC device with a linear power take off system. In contrast to the
linear wave case, for fully nonlinear waves, a substantial reduction of efficiency was observed with
increasing wave height. Moreover, the optimal pneumatic damping coefficient was found to be dependent
on the wave height. These results may lead to significant implications in the design and operation of
practical OWC systems.
Gkikas and Athanassoulis (2014) developed a nonlinear system for identification of the pressure
fluctuation inside the chamber of a 2D OWC device under regular wave excitation. A Wiener
Hammerstein cascade systemic scheme was assumed, which led to a truncated Volterra series
representation of the functional analogue. The nonlinear thermodynamic equations were investigated for
various combinations of excitation amplitudes and frequencies and it was found that the first three
harmonics were the most significant when explaining the response of the OWC.
Single device optimisation
Kurniawan and Moan (2013) performed a geometric optimisation of submerged wave absorbers oscillating
about a fixed horizontal axis, with the objective of minimizing two cost functions (i.e. the ratio of the
submerged surface area to the maximum absorbed power, and the ratio of the maximum reaction force to the
maximum absorbed power). Geometric configurations with uniform simple cross-sectional shapes (line,
circle, and elliptical sections), were considered. For each configuration, the body dimensions and
submergence, as well as the submergence of the rotation axis, were the optimised variables. They found that
most of the optimal geometries have their rotation axes close to the sea bottom and their bodies close to the
free surface. The optimal size of the geometries varies depending on the selected wave frequency range, but
the optimal cross-sectional dimensions are generally less than one third of the water depth when optimised
over a uniform distribution of wave frequencies from 0.4 to 1.3rad/s. Among the cross sections considered,
the elliptical one performed best.
Gomes et al. (2010) presented a geometry optimization of the acceleration tube of an IPS (Inter-Project
Service) buoy WEC, with the objective of obtaining of a good relation between the energy extracted from
a specific wave site and the submerged volume of the device, rather than the objective of power
697
maximisation. The annual average power was found to be insensitive to the PTO damping coefficient for
the tube with an optimal geometry. Taking the optimal PTO damping coefficient for each sea state gave
an increase of ~4% in the annual produced energy.
Falco et al. (2012) conducted a hydrodynamic analysis and optimisation of an OWC spar buoy; an
axisymmetric device consisting of a submerged vertical tail-tube fixed to an axi-symmetric floater that
oscillates mainly in heave. Linear potential flow theory in frequency domain was used. The tube geometry and
the air turbine characteristics were optimised. The adoption of a tail tube with a lower segment of larger inner
diameter will result in a significant reduction in the optimal tube length and in the turbine coefficient. The
volume of the air chamber also significantly affects the optimal tube length, the optimal tube length decreasing
with increasing air chamber volume, but not the optimal turbine coefficient.
Control
Control is very important for WECs to achieve good power absorption and acceptable loading under a
large range of environmental conditions. Recent developments in the control of WECs and arrays focus
on the application of various methods for improved performance in irregular wave conditions.
Latching control has been around since the 1970s and is well-known for improving wave energy
conversion efficiency. Current developments in control technologies have focused on how to practically
implement the technology. Falco (2008) proposed a new method to implement latching control and
determine the instant when the device is unlatched. Unlike conventional methods, where detailed future
information is always preferred, this new technique releases the device once the PTO force is larger than a
pre-set value. Similarly, Sheng et al. (2015a) and Sheng et al. (2015b) proposed a new method for
implementing latching control in which the latching duration is calculated purely based on sea state
parameters, such as the energy period, in such a way that detailed future information is not needed. These
new methods are currently sub-optimal and require development; however, they do indicate that the
resulting increase in wave energy conversion is significant.
Scruggs et al. (2013) applied Linear-Quadratic-Gaussian (LQG) optimal control theory to WEC control
in stochastic seas. Optimal causal controllers were designed for a point absorber and compared to the noncausal and static cases, in terms of the overall power generation performance and spectral characteristics.
It was found that for optimal causal controllers, it was necessary that they are gain-scheduled in
accordance with changes in the spectral content and direction of the sea state, in order to maintain
acceptable power performance.
Sichani et al. (2014) studied the dynamic behaviour of a point absorber considering non-linearities
arising from the buoyancy force, geometrical constraints on the motion of the WEC and the saturation
loads of the controller. They found that as the maximum available control force decreases the power that
could be extracted by the device reduces. For a controller with a high saturation load, there exists an
optimal value of PTO damping to obtain maximum power absorption. The inherent addition of stiffness
associated with the PTO system decreases the power absorption, but was necessary to constrain the
motion of the device.
Bacelli and Ringwood (2013) studied and compared two model-based control strategies for an array of
wave energy converters (point absorbers); Global Control (GC) and Independent Control (IC). GC is
based on a centralised control algorithm, which uses a complete hydrodynamic model of the array;
whereas with IC, each device is controlled independently without considering the hydrodynamic
interaction between WECs. Both control strategies apply constraints on the maximum allowable
oscillation amplitude and the PTO force. The results indicated that the use of GC in arrays gave an
improvement of up to 10% in power performance when compared to independent control. The benefits of
GC were greatest for arrays experiencing strong hydrodynamic interaction between individual devices,
and for individual devices with strong radiation properties.
Array of devices and park effect
Babarit (2013) reviewed literature related to the park effect for arrays of WECs. The theories and methods
that are applicable to wave interaction in arrays of WECs include:
-
Analytical methods, only applicable to heaving axisymmetric devices;

Asymptotic approximations of far field diffracted/radiated waves;
Numerical models based on boundary element methods;
Boussinesq and spectral wave models, originally developed for wave propagation over large domains
(a few kilometres to hundreds of kilometres).
698
For small devices (1020 m maximum dimension), in small arrays (<10 devices), with standard layouts
(regular or shifted grids with a separation distance of 1020 times the characteristic dimension), the park
effect was found to be negligible. For larger arrays, however, the number of rows should be minimised to
reduce the park effect.
Beels et al. (2010) applied the time-dependent mild-slope equation model MILDwave to study wave
propagation and transformation over a farm of five Wave Dragon devices. Based on an assessment of the
total power absorption and wave height reduction behind the farm, the optimal layout was found to be a
staggered grid (with three WECs in the first row and two WECs in the second row) with a separation
distance of 2D (with D the characteristic width of the device). They conclude that this layout does not
introduce any losses associated with park effects.
Renzi and Dias (2012a) developed a mathematical model to study the hydrodynamic behaviour and
wave power absorption of a bottom-hinged, flap-type, oscillating WEC in a channel. Within the
framework of linear inviscid potential-flow theory, analytical solutions were found and validated
against available numerical and experimental data. The transverse sloshing modes of the channel were
found to enhance the performance of the device near resonance. This theory has been further
developed and applied to study the wave interaction of a periodic array of flap-type WECs (consisting
of an infinite number of in-line WECs), based on an asymptotic analysis of the far-field waves
generated by the array (Renzi and Dias, 2012b). The analytical solutions agree well with the numerical
results based on a finite element formulation. Furthermore, a semi-analytical approach was developed
for a finite array of flap-type WECs and compared well with a finite-element numerical model based
on the software COMSOL Multiphysics for the 2-flap and 3-flap systems with an in-line configuration
(Renzi et al., 2014).
Analytical solutions for an analysis of arrays of OWC devices have been considered for the solution of
diffraction and pressure dependent radiation problems. Such methods have been used for stationary arrays
of interacting devices having cylindrical chambers (Konispoliatis and Mavrakos, 2013a) or annular air
chambers (Konispoliatis and Mavrakos, 2013b), as well as for free floating devices (Konispoliatis and
Mavrakos, 2014a). Numerical and experimental results concerning exciting wave forces, mean drift loads,
air volume flow rate, inner pressure and absorbed wave power were presented for parametric variation of
the distance between devices.
Nader et al. (2012) studied the scattered wave field around single and multiple OWCs devices using a
finite element formulation and linear wave theory. Power capture efficiency of the single device and of
the array was evaluated by parametric variation of the spacing between devices, the pneumatic damping
and direction of the incident waves. It was found that the presence of neighbouring OWCs has a
significant influence on the power capture efficiency of an individual device, even at large separations.
The optimal pneumatic damping for OWCs in an array was potentially found to differ from that of an
isolated OWC of the same dimensions.
ULS and FLS analysis
WEC devices are frequently designed to operate in a resonant condition in waves in order to improve
power capture which often leads to large motion responses and high risk of wave slamming on the WEC
structure near the free surface. De Backer et al. (2010) studied slamming effects on point absorbers in
operational conditions for three buoy shapes: two cones with dead rise angles of 45o and 30o, and a
hemisphere with a waterline diameter of 5m. The difference in peak slamming loads between the 45o cone
and the hemisphere was a factor of 2, while the difference in power absorption was only 48%. It is
therefore important to consider slamming during the shape design phase, alongside power absorption.
Rogne (2014) studied a hinged 5-body WEC with a shallow draft, cylindrical centre buoy attached to 4
semi-submerged spherical buoys, both numerically and experimentally. Both linear and weakly non-linear
hydrodynamic models were applied in the optimisation for power absorption. The weakly non-linear
model resulted in a better comparison with the experimental results.
4.1.2
Mooring analysis
There are two distinct types of mooring system for a floating WEC depending on whether it is an integrated part of the PTO system or not. The majority of floating WECs only use the mooring system for
station-keeping and therefore the focus in this section is on recent developments in this area.
Recent studies have focused on the influence of the moorings on the performance of the WEC. This is
to be expected, since the priority for every WEC is absorption of wave energy in an efficient and
profitable manner. It is difficult to draw general conclusions as the findings tend to be specific to a
699
particular type of WEC. The numerical tools used for the analyses are well-established in the offshore
industry, as discussed in previous ISSC reports.
A point absorber moored with three chain lines was investigated by Cerveira et al. (2013) to determine
the influence of the mooring system on power production. Measured climate data was used to define
the sea states. The device extracts energy during the surge and heave modes using two idealised PTO
systems represented by linear damping and stiffness coefficients. The dynamics are solved in the
frequency domain. Two mooring configurations were examined, slack and moderately slack, along with
the un-moored case. It was found that the mooring system reduced the annual energy capture by only
around 1%.
Bachynski et al. (2012) used a linearised frequency domain approach to analyse and optimise a
tethered WEC in irregular waves. Inertial and damping effects of the mooring system were not included.
The results showed that a relatively light mooring system had little effect on the power take-off, but did
introduce a low-frequency, coupled, pitch-surge resonance that can cause system failure if subjected to
long-period swells.
A two-body point absorber was studied by Muliawan et al. (2013) with the aim of estimating the
one-year power production both with and without moorings. A mooring configuration with four, semitaut, steel wire lines was considered. Two-body hydrodynamic analyses were carried out in the time
domain. The effect of the mooring system was implemented using a nonlinear spring. They observed
that, as long as the length of the mooring lines could accommodate the motions of the device due to
first order wave forces, the effect of the mooring was insignificant. Vicente et al. (2011) also
investigated a two-body point absorber. In this case, a slack bottom mooring spread of chains was
considered. They also concluded that the mooring system did not significantly affect the absorbed
power.
Vicente et al. (2009) assessed an array of point absorbers through a series of numerical simulations, in
regular and irregular waves. Three identical hemispherical buoys were considered in an equilateral
triangle configuration. The mooring system consists of three spread lines, with a further three lines
connecting the buoys to a centrally placed weight. Damping and inertia effects of all of the mooring lines
were neglected. The PTO of each floating converter was assumed to be activated by the buoy heaving
motion. The dynamics were solved in the frequency domain, without hydrodynamic interaction between
the three bodies. It was concluded that the performance was significantly affected by the presence of this
tight mooring system.
A positive effect on power conversion relating to the mooring system was discovered by Konispoliatis
and Mavrakos (2014b). The objective of the study was an array of closely spaced OWC devices. Three
configurations were examined; the devices were assumed to float independently, the devices were
connected together forming a freely floating (rigid) multi-device system, and the connected devices were
moored as a TLP. The calculations were done in the frequency domain with linearised mooring terms.
The authors found that the third configuration absorbs the most wave energy.
In general, energy absorption by oscillating bodies is achieved through various forms of damping. The
mooring system however, typically adds damping to the system and the associated energy loss reduces
the efficiency of the WEC. Most of the studies described have neglected the damping in the mooring
system when evaluating the efficiency of device. Moreover, those that did include it did not consider it to
be nonlinear. This is in conflict with the fact that most mooring lines produce non-linear damping
(Johanning et al., 2007). Therefore, to gain a more accurate representation of this problem, time-domain
methods including non-linear mooring forces are recommended for future analysis.
4.1.3
Power take-off analysis
The power take-off system is a sub-system for converting extracted hydrodynamic energy into useful
energy, typically in the form of electricity. Conventional PTO systems can be categorised into four types:
Hydraulic systems (oscillating bodies);

Air turbines (oscillating water columns);
Direct drive generators (oscillating bodies);
Water turbines (overtopping devices).
Salter et al. (2002) published a summary of PTO systems for wave energy conversion. The first
three types of PTOs have been described with respect to applications in the field of practical WECs
700
(Curran et al., 2008). PTO technologies for wave energy conversion are generally in a developmental
stage, with no dedicated mature technologies yet in place. Currently, conversion efficiencies and
reliability are both limited when compared to conventional power conversion technologies. The main
challenges for most WEC devices include the inherent low speed (roughly in the order of 1m/s), and
the oscillatory motion which leads to a large force variation over the cycle. Some devices do not
surfer from these problems, however, for example; overtopping devices, cycloidal turbines, and the
OWC, which all avoid one or both of these problems. Recent developments in PTO technology have
focused on improving the energy conversion efficiency by means of control technologies and
strategies.
Kamizuru (2012) compared a number of different applications and PTO developments. In a follow-up
paper (Kamizuru et al., 2013), they studied the Hydrostatic Drive Train (HDT) for improved wave energy
conversion.
PTO systems for OWCs are more mature than for other types of WECs. These systems are based on
self-rectifying (Wells-type/linear) turbines, impulse turbines (nonlinear), or the radial or bi-radial turbine
(Falco et al. (2011), Falco and Gato (2012)). The first two types have been tested in practical wave
energy plants (Takao and Setoguchi, 2012). A large amount of work has been carried out regarding the
design of guide vanes for such turbines, which can reach average energy conversion efficiencies of 60
70% for Wells and impulse turbines, and as much as 80% for radial turbines.
Direct drive generators may be applicable to some WEC devices, with similar technology developed
for the offshore wind energy industry. In principle, the main difficulties are similar for both applications;
a huge power conversion system is required to replace the mechanical gearbox. It should be noted that,
for wave energy conversion, direct drive means a linear, reciprocating motion (Mueller et al., 2007). As
such, direct drive technology for WECs is more difficult than for wind energy, but some solutions are
being developed (Crozier et al. (2011), Serena et al. (2012)).
4.2
Physical testing
Physical testing of WECs includes both laboratory tests at relatively small scales (typically from 1:100 to
1:20) and offshore field tests at larger scales (from 1:5 to 1:1). The purpose of testing is to demonstrate
the functionality and survivability of the system and the components, and to obtain information with
which to validate numerical models (Cruz et al., 2008). The basic guidelines for physical testing are as
follows:
Initial testing of a system (scale = 1:1001:50)hydrodynamic performance and power absorption

efficiency, WEC farm performance, and mooring configuration;
Validation of numerical models or tools (scale = 1:331:20);
Behaviour in extreme conditions (scale = 1:201:5)nonlinear hydrodynamics;
Functionality and survivability of all components (scale = 1:51:1).
Uncertainties and Scale Effects

In any physical modelling scaling issues must be considered, particularly during small scale tests.
Hydrodynamically, wave energy converters are similar to conventional ships and offshore platforms
where Froude scaling is most appropriate. Sheng et al. (2014) discusses the basic scaling issues for
WECs. The most difficult of these involves modelling the PTO system. Sheng et al. (2014) argue that
geometrically similar modelling of the PTO system is not generally appropriate because it is usually very
difficult, if not impossible, especially at very small scales. Instead, an accurate mathematical model of the
PTO should be implemented in the physical model.
PTOs for a Wells turbine can be relatively accurately approximated using a porous membrane, or an
orifice for an impulse turbine. However, consideration should be made of the compressibility of air in the
chamber (Weber (2007), Sheng et al. (2014), Falco and Henriques (2014)). In particular, Falco and
Henriques (2014) presented similitude laws for model testing of OWCs concerning geometric,
hydrodynamic, thermodynamic and aerodynamic similarity between prototype and scaled-down physical
models, with particular emphasis on air compressibility effects and on turbine aerodynamics. It was
shown that the correct volume scale ratio for the air chamber should take into account the
thermodynamics of the compressible flow through the air turbine. A numerical example was detailed to
illustrate the importance of appropriately simulating the air compressibility effects when testing at model
scale.
4.2.1
701
Laboratory testing and validation of numerical tools
With respect to laboratory testing, some of the guidelines and procedures developed by the ITTC for
testing of ships and offshore structures are applicable. However, it should be noted that blockage and wall
effects are significantly more important for WEC experiments than they are for ship testing, and this
should be considered in post-processing and data analysis.
Many model tests have been carried out under the MariNet program. One of the aims of this program is
to characterise and verify the performance behaviour of the European test facilities. Results are intended
to be used to establish correction factors which can be applied to test results, so that engineers and
potential investors can accurately compare the performance of devices tested in different facilities.
In recent years, some well-known WEC concepts have undergone laboratory testing. The Langlee
WEC (Pecher et al., 2010) tests focussed on power generation, and the Wave Star point absorber tests
(Zurkinden et al., 2014) focussed on the comparison between the numerical and experimental results
considering the weakly nonlinear hydrodynamic effects. A recent model test on the Pelamis model was
conducted with the goal of reducing the maximum loads on the mooring lines and anchors, the maximum
excursions of the device, and the influence on the internal motions which are used for power conversion
(Casaubieilh et al., 2014). Two different types of mooring systems were used; a conventional catenary
mooring, and a semi taut-leg mooring. The elastic components in the taut-leg system reduced the
maximum forces on the mooring line by 67%. The taut-leg system also significantly reduced excursions,
while the motions required for power conversion were not affected.
More recent WEC devices have also been tested under controlled conditions in laboratories. Siegel
et al. (2012) tested a lift-based cycloidal wave energy converter, CycWEC, at a scale of 1:300 in a 2D
wave flume. The experimental results showed an energy extraction efficiency of more than 95% with
optimal parameters. Zanuttigh et al. (2013) performed tests of the DEXA device, (www.dexawave.com) a
hinged-barge-type concept, with the focus on the influence of various mooring systems on efficiency.
Rogne (2014) undertook tests of a hinged 5-body WEC in a small towing tank at Marintek, and studied
the effect of nonlinear Froude-Krylov and restoring forces on the platform motions and power absorption.
Imai et al. (2014) studied a floating OWC-type WEC, the Backward Bent Duct Buoy (BBDB),
endeavouring to understand the effects of various design parameters on the effect of power absorption
including the dimension of the duct-extended buoy, the mass of the floater and the shape of the air
chamber.
4.2.2
Field testing
A number of devices are currently being field-tested at various scales, though relatively little has been
published about these tests due to the commercially sensitive nature of the work. Kofoed et al. (2006)
reported the results of the 57m long, 27m wide and 237ton prototype of the Wave Dragon overtopping
device at 1:4.5 scale that was tested in Nissum Bredning, Denmark beginning in 2003. The preliminary
tests supported the data obtained earlier from laboratory testing and focused mainly on the hydrodynamic
performance. Experiences on operational aspects were also obtained from this test campaign, such as
regulation strategies for crest freeboard and turbines, and remote control of operation and testing. The
power performance of the prototype was investigated by Tedd and Kofoed (2009) indirectly by measuring
the overtopping flow time series. Comparison between the simulated and measured data showed that the
measurements support the use of the algorithm for generating a simulated flow.
In the Wave Power Project Lysekil, more than 10 prototypes of point absorber WEC have been
deployed and tested at the Lysekil research site located on the Swedish west coast since 2006 (Leijon
et al. (2008), Lejerskog et al. (2015)). Each WEC has an installed capacity of 10kW and consists of a
buoy at the surface, which is connected by a mooring line (steel wire) to a linear generator placed in a
capsule on the seabed. Significant research work was carried out with respect to WEC sea tests, analysis
of measurement data, comparative study of simulations and measurements, optimization of buoy
geometry for wave energy extraction (Sjkvist et al., 2014), performance of large arrays of WECs
(Engstrm et al., 2013). The developed simulation models have been extensively compared with the
measurement data and a good correlation between the simulation and the experimental results was
obtained for normal operations (Lejerskog et al., 2015). Besides the technical/functional verification and
development, the environmental, marine biological and marine ecological aspects the system have also be
investigated scientifically (LysekilProject, 2015).
Field testing of various WEC components has also been conducted. Henderson (2006) summarised the
test results of a novel hydraulic PTO system for the Pelamis concept, including a 1:7 scale model tested
702
both in a towing tank and at sea, and a full scale prototype with a 1 MW rated hydraulic system tested on
a joint test rig. The experiments indicated that the performance of the PTO was as expected, with a
combined efficiency of the primary transmission of more than 80% at full scale over a representative
range of operating conditions.
Thies et al. (2014) presented field trials from the South West Mooring Test Facility including mooring
line fatigue damage results. This facility is a large scale installation designed to investigate mooring loads
and responses in real sea conditions. It consists of a generic 3.25ton buoy, that can be moored in various
configurations, which has been installed with a three-leg catenary hybrid (rope-chain) mooring since
March 2010. It is located in a relatively sheltered site in the southwest part of Falmouth Bay (Cornwall,
UK), in a water depth of 27m (low tide mark) and tidal range of up to 5.4m. The installed buoy is
extensively instrumented to acquire data regarding the buoy position, motion response and the associated
mooring line loads, as well as the wind velocity. In addition, an Acoustic Doppler system is installed
nearby to record the environmental conditions including incident waves and tidal currents.
4.3
Rules and standards
Significant efforts are underway to establish rules and standards for WEC development. The main goal is
to assess the different technologies on a level playing field to allow for a fair and standardised
comparison. For example, most device developers rate their devices using the maximum power output
(SI-Ocean, 2012). This can lead to confusion since WECs generally differ from other renewable energy
devices in terms of the relation between maximum power output and average power output. Specifically,
under design environmental conditions, wind, tidal or solar power will produce consistent power at their
rated maximum, while for WECs the average power production will generally be much less than the
maximum rated power. Thus it can be misleading to estimate the annual power output for a WEC using
the rated power.
In order to correct this problem, several international collaborations have been initiated to establish
rules and standards for rating the power of WECs. This should lead to standardised methodologies for
assessing different devices so that fair comparisons can be made. Moreover, guidelines, rules and
standards should also provide detailed procedures and methodologies for design, testing and operation of
wave energy converters.
The International Electrotechnical Commission established the Technical Committee (TC) 114 on
Marine Energy: Wave, Tidal and other Water Current Converters in 2009, and relevant standards have
progressed through several stages. The most advanced standards in wave energy conversion are for
resource assessment and mooring system, both of which are in the final stage and the standards are
expected to be published in the first half of 2015. More tasks are being carried out by the IEC-TC-114
(2014).
The CarbonTrust (2005) commissioned DNV for a guideline on design and operation of WECs in
2005. This document included many aspects on design and operation of WECs. It should be noted that
this document should be regarded as a guideline instead of a standard, but it is hoped that it can be used as
a starting point for the future development of standards.
DNV (2008) published the principles and procedures for certification of tidal and wave energy
converters in 2008, and made two amendments in 2011 and 2012. The document is built on experience
and relevant standards in the oil and gas industry, but also includes the best practices in this particular
area. This document is supposed to cover all types of tidal and wave energy converters.
Generally, the technologies for wave energy conversion are still immature, and many devices are in
different development stages, while new ideas are continuously being proposed. To guide the rational
development of wave energy converters, some best practices have been proposed. Unlike the standards,
the best practices are meant to be guidelines on how to move WECs from the concept to the commercial
stage, and avoid the most common failure traps.
Under the Implementing Agreement for a Co-operative Programme on Ocean Energy Systems
(OES-IA) created by the International Energy Agency, Annex II, Nielsen (2010) published a
document entitled Development of Recommended Practices for Testing and Evaluating Ocean Energy
Systems, Summary Report, in which the design basis and the cost basis for wave and tidal energy
systems are discussed. Similarly, the EquiMar project has published many deliverables on their website
http://www.equimar.org/. Among the deliverables, two may be more relevant to the topic here: Best
Practice for Tank Testing of Small Marine Energy Devices and Protocols and Guidance for Device
Specification and Quantification of Performance. Other relevant information can also be found from the
703
MariNet project (http://www.fp7-marinet.eu/joint-activity_standardisation-best-practice.html) and the

EMEC website (http://www.emec.org.uk/standards/).
5.
TIDAL AND OCEAN CURRENT TURBINES
This chapter focusses upon different aspects of tidal energy conversion and ocean current energy extraction.
Since 2012, several prototype devices have been deployed in various test locations around the world, with
good levels of success. The key to further development of the industry towards a full commercialization is
the reduction in uncertainty (and therefore the reduction in cost); be this in device development itself,
installation and offshore management, resource modelling or environmental impact assessment.
5.1
Development, modelling and testing of tidal current energy converters
5.1.1
Device development
The previous ISSC reports covered the different concepts available, several of which have been
developed into functioning prototypes. Most notably Voith, Atlantis, Andritz Hammerfest, Open Hydro,
Marine Current Turbines and Alstom have all deployed successful test devices. Recent developments
include a 1MW horizontal axis system from Kawasaki designed for deployment on a gravity base
Yanamoto (2012). The turbine is set for test at EMEC, Scotland in 2016 and Japan in 2017.
A Wing-In-Ground effect Turbine (WIGT) concept has been developed (Liu, 2012a). The dual foil unit
can be arranged vertically or horizontally (Liu, 2012b), Figure 9. The WIGT has the potential to produce
1.73 times the energy of a conventional Horizontal Axis Tidal Turbine (HATT). It was also found that
WIGTs have the ability to improve power capture in shallow water flows and low speed streams.
Figure 9. A dual foil arrangement in the WIGT concept (Liu, 2012b).
Some devices have explored the concept of an underwater kite (Lazakis et al., 2013) which improved
energy capture in low-speed tidal streams.
An interesting development is increased attention regarding microgeneration systemswhether this be
for local schemes in outlying areas or for a number of smaller devices to be coupled to produce a larger
yield (lvarez et al., 2014).
Several prototype HATTs have experienced problems with structural blade failure. In order to optimise
rotor structural strength and reliability a procedure was developed by Liu and Veitch (2012) using an
example was of a 20m rotor in a flow of 10 knots. The procedure has been so developed that it may be used
for both metallic and composite blades. Composite blades have been shown to have the potential to
incorporate structured tailoring of the layup so as to result in coupled motions, for example application of a
pure bending moment would result in blade twist (Nicholls-Lee et al., 2012). Such coupled deformations can
reduce fatigue and prolong service life expectancy. Composite materials are finding favour in more sections
of tidal device design; blades have been commonly manufactured from composites for a while, however
parts of the support structure, or even the whole foundation, have the potential to gain from use of composite
materials in both ease of manufacture and structural integrity (Coppens, 2014).
5.1.2
Numerical modelling and experimental testing
Alongside physical prototyping, modelling of tidal energy converters, both numerically and
experimentally, assists design and development of devices in the industry as a whole. Takeda et al. (2013)
developed a numerical method of modelling a floating, tethered tidal generator. A joint CanadianAustralian collaboration on a series of bi-directional tidal turbine prototyping, design and optimisation
704
(Liu and Bose, 2012) resulted in a methodology for generic tidal turbine hydrodynamic prototyping and
optimisation. Further work involved a test program on seven selected HATT rotors from the previous
series, with a very detailed geometry and motion parameters (Liu et al., 2014b). The work resulted in a
key reference platform for tidal turbine design, and measured data for validation of numerical tools.
In addition to the design basis of hydrodynamic performance of the rotor, effects of marine fouling
need to be considered as these negatively impact the efficiency of all devices. Tests have been carried out
in the waters of Ikitsuki Island, Nagasaki, Japan for eight months to assess biofouling on a small array of
three bladed HATTs (Katsuyama et al., 2014). Significant fouling occurred on the uncoated devices and
was not affected by turbine rotation. Use of antifouling coatings resulted in only a thin film of fouling.
The effect of array layout is also key to optimising turbine performance. Myers and Bahaj (2012), Bahaj
and Myers (2013), Blackmore et al. (2014) discussed the concept of turbines arrangement by experimental
studies conducted using multiple actuator disks. Highly turbulent wakes were apparent indicating that
optimal lateral spacing as well as the synergistic effect of a downstream row of devices are key for array
performance. These results imply that proper design of support structures and arrays optimisation could
increase the energy yield as the site-specific turbulence characteristics are taken into consideration.
Numerical assessment of array layout, and the subsequent effect on energy yield has always been a
source of interest. This has been carried out a local scale using an actuator disc RANS model (Batten
et al., 2013), intermediate scale through adaptive mesh coastal modelling (Divett et al., 2013) and large
scale through interpretation of the blockage of a large array of devices in a channel (Vennell, 2012).
Staggered arrays were found to show increased energy capture, however increasing the number of rows in
an array past the theoretical optimum has harsh consequences due to diminishing returns. Primarily it
should be noted that maximising the power production of large arrays is not the same as maximising the
conversion efficiency of the turbines.
5.2
Environmental impact
With devices reaching the stage at which deployment into arrays is becoming likely in the near future,
concern over the environmental impact of arrays as well as individual devices is increasing. Often the use
of potential sites for marine energy extraction are hampered by ambiguities in marine legislation, and
high quality Environmental Impact Assessments (EIAs) are therefore imperative (MacLean et al., 2014).
Often physical resource assessment and environmental assessment can go hand in hand, with coupled
studies reducing time scales and costs relating to marine energy installations (Ashton et al., 2014).
Fauna of interest include fish (Viehman and Zydlewski (2014), Broadhurst et al. (2014)), marine
mammals (Worthington (2014), Willis et al. (2013)) and seabirds (Furness et al. (2012), Frid et al.
(2012)). Assessment of several impacts through a single study is desirable in order to provide a good
coverage of risk in a reasonable time frame. The FLOWBEC seabed platform combines a number of
instruments to record information at a range of physical and multitrophic levels at a resolution of several
measurements per second, for a duration of 2 weeks to capture an entire spring-neap tidal cycle
(Williamson et al., 2014). The interactions between diving seabirds, prey, flow hydrodynamics and tidal
energy extraction structures were analysed for a period of five 2-week deployments at EMEC. The results
can be used to guide marine spatial planning, device design, licensing and operation.
Other key environmental impacts that must be considered when planning a tidal scheme are those
affecting sediment transport (Frid et al., 2012), sand banks (Neill et al., 2012), water quality (Kadiri
et al., 2012), and energy removal on the ecosystem (Copping et al., 2013).
5.2.1
Marine planning
The rapid development of the tidal energy sector has led to an increased requirement for Marine Spatial
Planning (MSP) and, increasingly, this is carried out in the context of the Ecosystem Approach (EA) to
management (Alexander et al., 2012). Conflicts between the interests of tidal energy developers and
commercial and recreational users of the area must be identified, and use preferences and concerns of
stakeholders considered (Johnson et al. (2012), Alexander et al. (2013), Tweddle et al. (2014)). It is
imperative that, in order to provide balance between national and local demands, marine planning is
properly carried out, with tools being developed specifically pertaining to this (Janssen et al., 2014).
5.3
Economic feasibility
The choice of which type of electrical power generation technology to adopt is driven by a number of
factors including (Johnstone et al., 2013):
705
Cost of generated electricity;

Responsiveness of generating plant to demand;
Security of supply/resource availability;
Environmental impact;
Execution risk.
In the shorter term, in order for tidal energy to gain commercial acceptance, tidal technologies under
development need to produce electricity at a competitive price. Key to achieving this is bringing down the
cost of installation and operation and maintenance (O&M) of tidal projects. Offshore operations are costly
in relatively benign environments, but with the extreme conditions often experienced in tidal locations
vessel requirements are onerous and equipment that is suitable is often very expensive to source (Lazakis
et al. (2013), Morandeau et al. (2013), Nicholls-Lee et al. (2013)). Modelling of marine operations,
incorporating vessel and equipment costs, route analysis, weather windows etc. is imperative to enable
proper planning and risk analysis of operations (Lazakis et al. (2013), Morandeau et al. (2013)). Many
vessels currently employed in the tidal energy field come from the Oil and Gas sector, and consequently
demand high day rates and are often over-specified for the job; i.e. a large amount of unnecessary
equipment is present on an overly large deck. There is a requirement for fit-for-purpose vessel to be
developed in order to bring installation and O&M costs down, for example the HF4 developed for
operation in high currents (Nicholls-Lee et al., 2013), Figure 10, thereby bringing tidal energy closer to
economic feasibility.
Figure 10. Purpose-built vessel concept HF4 for tidal turbine installation (Nicholls-Lee et al., 2013).
6.
COMBINED USE OF OCEAN SPACE
European waters in the coming years will be subjected to a massive advance of marine energy farms/arrays,
and the development of these facilities will increase the need for marine infrastructure to support their
installation and operation. The most obvious structures include offshore wind farms, constructions for
marine aquaculture, and the exploitation of wave and tidal energy. It is therefore crucial that the monetary
cost, use of marine space and the environmental impacts of these activities remain within acceptable limits.
Hence, offshore platforms that combine multiple functions within the same infrastructure offer significant
economic and environmental benefits. The analysis, design and construction of offshore structures is one of
the most demanding set of tasks faced by the engineering profession. The offshore structures have the added
complication of being located in an aggressive ocean environment where hydrodynamic interaction effects
and dynamic responses become major considerations in their design in comparison to land-based structures.
Thus, the analysis, design and construction of offshore structures used for the combined activities in the
ocean environment are one of the most important studies that need to be performed (Clment et al. (2002),
Boer et al. (2007), Brandon et al. (2010), Menicou and Vassiliou (2010), Da Rocha et al. (2010), Villers and
Vign (2010), Buck (2011), Perez-Collazo et al. (2015)). EU-FP7 funded projects regarding multi-use
offshore platforms include TROPOS, MERMAID, H2OCEAN, ORECCA, MAREN2 and MARINA, and
are underway for the development of a multi-use platform system integrating a range of technologies.
TROPOS:
Hydrodynamic effects and the dynamic response of offshore structures play an important role in their
design. Delory et al. (2011) analysed a multi-disciplinary, multi-platform, observing system for the central-eastern Atlantic ocean using the PLOCAN observatory program. Quevedo et al. (2013) presented a
706
detailed design and analysis of the multi-use platform configuration in the scope of the FP7 TROPOS
project. The synergy and the compatibilities amongst the platforms to be used for transport, energy, aquaculture and leisure activities are studied in detail. Lu et al. (2014) analysed the environmental aspect of
designing multi-use offshore platforms in the scope of the FP7 TROPOS project.
There are four main components to the platform: Transport, Energy, Aquaculture and Leisure. These
are integrated with the aim of enhancing the potential and increasing the added value. Three different
platforms configurations are designed in order to form an initial idea of the synergies and compatibilities
among components (Quevedo et al., 2013).
MERMAID:
The EU-FP7 funded project MERMAID assessing multi-use offshore platforms is being carried out to
develop concepts for the next generation of offshore platforms which can be used for multiple purposes;
including energy extraction, aquaculture and platform related transport. The project has undertaken the
theoretical examination of new concepts, such as combining structures and building new structures on
representative sites under different conditions.
In the analysis four offshore test sites with different environmental characteristics have been carefully
selected for their specific challenges. The four sites, representing different environmental, social and
economic conditions, are located in different seas. The Baltic Sea, North Sea, Atlantic Ocean and
Mediterranean Sea are considered. The most appropriate design options for a given offshore area are
being created, and verified, as part of the MERMAID project so that the end users can use the results to
aid marine planning strategies.
H2OCEAN:
H2OCEAN is an EU-FP7 funded project assessing the development of an innovative design for an
economically and environmentally sustainable multi-use, open sea platform. The main aim is to harvest
both wind and wave power for multiple applications on-site, including the conversion of energy into
hydrogen that can be stored and shipped to shore as green energy carrier and a multi-trophic aquaculture
farm.
The unique feature of the H2OCEAN concept lies in the novel approach for the transmission of
offshore generated renewable electrical energy through hydrogen. The concept endeavours to enable
effective transport and storage of the energy, decoupling energy production and consumption, thus
avoiding the grid imbalance problem inherent to current offshore renewable energy systems. This
integrated concept can take advantage of several synergies between the activities present as part of the
platform, thereby significantly boosting the environment.
ORECCA:
The ORECCA (Offshore Renewable Energy Conversion platformsCoordination Action) Project is an
EU FP7 funded collaborative project in the offshore renewable energy sector. The principal aim is to
overcome the fragmentation of know-how available in Europe and its transfer amongst research
organizations, industry stakeholders and policy makers. The objective is to stimulate these communities
to take the necessary steps to foster the development of the offshore renewable energy sector in an
environmentally sustainable way. The project brings together a combination of experts from a wide
variety of multinational companies, research institutions, consultancies, utilities and project developers.
The focus is on European technology, with a specific emphasis on the opportunities that exist across
Europe when the three offshore renewable energy sectors within the scope are considered together.
Within the ORECCA project, the scope of the offshore renewable energy sector was confined to offshore
wind, wave energy and tidal stream. These energy sectors have been identified as those that are currently
expected to make significant contributions to the energy system in the medium to long term.
MAREN2:
MAREN2 is an inter-regional project financed by the EC and included in the 20072013 Atlantic Area
(AA) Programme. The main objective of the project is regarding the exploitation of the renewable energy
potential of the marine and coastal environment and protecting, securing and enhancing its sustainability.
MAREN2 is focused on the multipurpose platform design, energy uses with other resources, multiple use
of facilities and increase stakeholder involvement to minimise negative impact. The activities involved
include comparable data collection and processing, multi-purpose renewable energy platform
identification and assessment of the hydro-environmental impact of marine renewable energy schemes.
707
The assessment of the wave energy farm in the Portuguese maritime pilot zone and Atlantic European
coasts are analysed in Bento et al. (2014) and Guedes Soares et al. (2014b). A brief review on the feasible
business model for the multi-purpose offshore platforms for marine renewable energy and synergies with
different technologies is discussed in Karmakar and Guedes Soares (2014).
MARINA:
In the EU FP7 MARINA Platform project, a set of equitable and transparent criteria for the evaluation of
multi-purpose platforms for marine renewable energy have been established and applied to select three
combined wind and wave energy concepts, from amongst >100 proposed concepts, for detailed study by
numerical and experimental methods. These include the spar-torus-combination (STC), the
semi-submersible-flap-combination (SFC) and the OWC Array plus one wind turbine. Laboratory tests
have been carried out and used for validation of the developed numerical models; for the STC (Wan
et al., 2014) with focus on the survivability in extreme wave conditions, for the SFC Michailides et al.
(2014) and the OWC Array concept (OSullivan et al., 2013) with focus on both the functionality and
survivability. In one of the work packages, an atlas of combined wind-wave-current resources has been
created for European waters (Cradden et al., 2012) and used in combination with a GIS (Geographic
Information System) tool for site selection based on the resources with constraints such as bathymetry and
environmental sensitivity. Other aspects of combined renewable energy devices have also been studied in
detail in different work packages, such as technology risk assessment, economic feasibility assessment,
critical component engineering and grid connection and macro-system integration.
Most of these projects on combined concepts are at the level of conceptual design only. Future work in
this direction should focus on technical development and economic feasibility through concept
demonstration.
7.
CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK
In the past few years, extensive research has been conducted in the area of offshore renewable energy;
ranging from conceptual design, numerical tool development, to model-scale and full-scale physical
testing. From the industry point of view, offshore wind technology with bottom-fixed foundations is
relatively mature and has been successfully applied in the recent development of large-scale offshore
wind farms. Floating wind turbine technology has been demonstrated by several prototypes being tested
at sea and is progressing with a few small wind farms under planning. Recent experiences in industry
indicate that cost reduction is the main challenge for offshore wind development and should be the focus
of research work in the future. However, a commercial development of either wave energy or tidal and
ocean current energy has not yet taken place. Significant research efforts with clear focus and good
continuity are required to develop large-scale wave energy converters or marine current turbines.
Resource assessment is one of the important aspects for developing offshore renewable energy.
Generic assessments of wind, wave, tidal and ocean current energy are readily available. Detailed
evaluation of resources with sufficient space and time resolutions are necessary for site selection
however. This is particularly important for wave and ocean current energy assessment for areas with
complex local bathymetry and coastal geometry. Numerical models have been developed for such
assessment; however, further validation against long-term measured data is needed. From the structural
design point of view, joint environmental data (either measured or hindcast data) of metocean conditions
is needed for both ULS and FLS design of offshore wind turbines. Accurate weather forecasting systems
are particularly useful for planning transportation and installation of offshore renewable energy devices.
The recent developments in the offshore wind industry show a clear trend towards an increase in size
of individual turbines, and wind farms, in order to reduce the costs associated with installation, operation
and maintenance. Wind turbines of 5-7MW are now being installed in several wind farms and even larger
turbines (8-10MW) are under active development. This calls for a development of optimised or novel
support structures, in order to achieve economic feasibility, since these account for a significant part of
the total capital cost, especially regarding floating wind turbines.
Numerical tools for dynamic response analysis of both bottom-fixed and floating wind turbines have
been developed extensively and rapidly in recent years, by coupling the codes developed originally for
the onshore wind industry with those from the offshore oil and gas industry. Developing simplified
analysis methods for preliminary design assessment, that are both efficient and accurate, is also needed. A
code-to-code benchmark study was carried out under the IEA OC3 and OC4 activities and a code-toexperiment validation is in progress through OC5 using the model test data; however, further validation
of the numerical tools against field measurements needs to be carefully planned and thoroughly
708
performed. Due consideration should be given to uncertainties associated with the field measurements of
environmental conditions and response parameters.
Further development of experimental methods and techniques for testing of floating wind turbines is
needed. This requires associated development of laboratory infrastructure and equipment to generate
high-quality wind and wave fields simultaneously. Whilst field testing of some devices is underway, the
information obtained from the prototype tests tends to be commercially sensitive and is not often
published, therefore there is a need to develop large-scale research centres for testing floating wind
turbines at sea. This would be invaluable to both academia and industry alike to aid development of the
industry as a whole. The results gained from such centres through the testing of new wind turbine and
floater concepts could be utilised for validating numerical tools, testing advanced control systems,
applying novel condition monitoring techniques, and gaining experience on installation, operation and
maintenance.
Design rules, in particular for floating wind turbines, need further development with details on the use
of numerical methods for global and local response analyses considering both normal and fault
conditions. Methods or procedures for modelling and analysis of wind turbine faults and their effects on
the dynamic responses of offshore wind turbines need to be established, and validated, preferably against
field measurements.
Thus far, limited research has been conducted in the field of offshore wind turbine transportation,
installation, operation and maintenance. Conversely, a significant amount of offshore work (installing in
the order of 500 wind turbines per year) is currently undertaken by the offshore wind industry. The
research community may be able to assist in more efficient and economic offshore operations by
developing novel methods for installation, O&M and decommissioning which have the potential to be
carried out in more severe metocean conditions safely thereby decreasing downtime.
Developing a wave energy converter (WEC), which is commercially viable, is a challenging aim. The
recommended method of achieving this is to follow the five development stages from conceptual design
through to demonstration of multiple full-scale units, in order to overcome the technological challenges
occurring at the different stages of development. In recent years, a significant number of different WEC
concepts have been proposed. These have been analysed using numerical methods, or undergone physical
testing at model scale or full scale. One key challenge is how to integrate the broad spread of experiences
gained through the different industrial and research projects, to further advance the technology
development with such a wide range of different design options. It is recommended that this be addressed
in future research work. Survivability in extreme wave conditions, and long-term performance with
respect to fatigue, are particular challenges for structural design of WECs and also require further
investigation.
Numerical tools for the analysis of a single WEC and WEC arrays have been developed. Further
validation against model tests is required however; a benchmark study of various numerical tools for
analysis of a specific WEC type (either oscillating bodies or oscillating water column) might be beneficial
to be performed by the next ISSC committee. Code-to-experiment comparison is preferred. The focus
should be given not only to the aspects of hydrodynamics and power absorption for operational
conditions, but also to the motion and structural responses in extreme wave conditions.
Developing tidal and ocean current turbines faces a particular challenge due to the conflict between
selecting a suitable site with sufficient tidal or ocean current energy resources and finding a suitable time
window for installation of these devices. Numerical tools developed so far focused on the assessment of
power extraction efficiency and tools for structural response analysis and design need to be developed
and validated. Several devices exist and have been tested successfully as full scale prototypes; however,
the uncertainties and the costs associated with installation, O&M and decommissioning of the devices are
still high. It is recommended that future research be directed in this area, potentially furthering the
development of recent novel installation solutions, or improving the infrastructure surrounding key tidal
array locations.
Integration of offshore renewable energy solutions with other potential uses of ocean space, such as
transport, sea food production, and leisure, are currently being investigated in several EU-funded projects;
however, most of these are only at the stage of conceptual design. Future research work is needed to
address the plethora of engineering challenges associated with the development of multiuse platforms.
709
REFERENCES
ABS 2013a. Guide for Building and Classing Offshore Wind Turbine Installations, ABS #176. Houston, Texas, USA.
ABS 2013b. Guide for Building and Classing Floating Offshore Wind Turbine Installations, ABS #195. Houston,
Texas, USA.
Adam, F., Myland, T., Dahlhaus, F. & Gromann, J. 2014. GICON-TLP for wind turbinesthe path of
development. The 1st International Conference on Renewable Energies Offshore (RENEW). November 2426,
Lisbon, Portugal.
Adam, F., Steinke, C., Dahlhaus, F. & Gromann, J. 2013. GICON-TLP for wind turbinesValidation of calculated
results. Proceedings of the Twenty-third International Offshore and Polar Engineering Conference. June 30July
5, Anchorage, Alaska, USA.
Alexander, K. A., Janssen, R., Arciniegas, G., O'higgins, T. G., Eikelboom, T. & Wilding, T. A. 2012. Interactive
marine spatial planning: Siting tidal energy arrays around the Mull of Kintyre. PLoS ONE, 7 (1).
Alexander, K. A., Wilding, T. A. & Heymans, J. J. 2013. Attitudes of Scottish fishers towards marine renewable
energy. Marine Policy, 37, 239244.
lvarez, E. ., Navarro-Manso, A., Gutirrez-Trashorras, A. J., Fernndez-Francos, J. & Rico-Secades, M. 2014.
Design and feasibility study of a microgeneration system to obtain renewable energy from tidal currents. Journal
of Renewable and Sustainable Energy, 6 (3), 033109.
Arapogianni, A. & Genachte, A.-B. 2013. Deep WaterThe Next Step for Offshore Wind Energy, Technical
Report. European Wind Energy Association.
Ashkan, R., Elsaesser, B. & Dias, F. 2013. Numerical simulation of wave interaction with an oscillating wave surge
converter. Proceedings of the ASME 2013 32nd International Conference on Ocean, Offshore and Arctic
Engineering. Paper No. OMAE201310195, June 914, Nantes, France.
Ashton, I., Argall, R., Nicholls-Lee, R. & Johanning, L. 2014. Monitoring spatial variability for marine energy sites.
Proceedings of the 2nd International Conference on Environmental Interactions of Marine Renewable Energy
Technologies (EIMR2014). April 28May 2, Stornoway, Isle of Lewis, Outer Hebrides, Scotland.
Azcona, J., Bouchotrouch, F., Gonzlez, M., Garcianda, J., Munduate, X., Kelberlau, F. & Nygaard, T. A. 2014.
Aerodynamic thrust modelling in wave tank tests of offshore floating wind turbines using a ducted fan. Journal
of Physics: Conference Series, 524, 012089.
Babarit, A. 2013. On the park effect in arrays of oscillating wave energy converters. Renewable Energy, 58, 6878.
Babarit, A., Hals, J., Muliawan, M. J., Kurniawan, A., Moan, T. & Krokstad, J. 2012. Numerical benchmarking
study of a selection of wave energy converters. Renewable Energy, 41, 4463.
Bacelli, G. & Ringwood, J. 2013. Constrained control of arrays of wave energy devices. International Journal of
Marine Energy, 34, 5369.
Bachynski, E. E. & Moan, T. 2012. Design considerations for tension leg platform wind turbines. Marine
Bachynski, E. E., Young, Y. L. & Yeung, R. W. 2012. Analysis and optimization of a tethered wave energy
converter in irregular waves. Renewable Energy, 48, 133145.
Bae, Y. H. & Kim, M. H. 2014. Aero-elastic-control-floater-mooring coupled dynamic analysis of floating offshore
wind turbine in maximum operation and survival conditions. Journal of Offshore Mechanics and Arctic
Engineering, 136 (2), 020902.
Bagbanci, H., Karmakar, D. & Guedes Soares, C. 2011a. Dynamic analysis of spar-type floating offshore wind turbine.
Proceedings of the 2nd Coastal and Maritime Mediterranean Conference. November 2224, Tangier, Morocco.
Bagbanci, H., Karmakar, D. & Guedes Soares, C. 2011b. Comparative study on the coupled dynamic analysis of
spar-type and barge-type floating wind turbine. Proceedings of the 1st International Symposium on Naval
Architecture and Maritime. October 2425, Istanbul, Turkey.
Bagbanci, H., Karmakar, D. & Guedes Soares, C. 2012a. Effect of environment on the design loads on monopile
offshore wind turbine. In: Guedes Soares, C., Garbatov, Y., Sutulo, S. & Santos, T. A. (eds.) Maritime
Engineering and Technology. London: Taylor & Francis Group, 547552.
Bagbanci, H., Karmakar, D. & Guedes Soares, C. 2012b. Review of offshore floating wind turbine concepts. In:
Guedes Soares, C., Garbatov, Y., Sutulo, S. & Santos, T. A. (eds.) Maritime Engineering and Technology.
London: Taylor & Francis Group, 553562.
Bahaj, A. S. 2013. Marine current energy conversion: the dawn of a new era in electricity production. Philosophical
Transactions of the Royal Society A, 371, 20120500.
Bahaj, A. S. & Myers, L. E. 2013. Shaping array design of marine current energy converters through scaled
experimental analysis. Energy, 59, 8394.
Barahona, B., Jonkman, J., Damiani, R., Robertson, A. & Hayman, G. 2015. Verification of the new FAST v8
capabilities for the modelling of fixed-bottom offshore wind turbines. Preprint, to be presented at AIAA SciTech
2015. Kissimmee, Florida, USA.
Barker, A., Sudom, D. & Sayed, M. 2014. Conical structures in ice: The roles friction slope and shape play.
Proceedings of the Offshore Technology Conference, Arctic Technology Conference. Paper No. OTC-24566,
May 57, Houston, Texas, USA.
Batten, W. M. J., Harrison, M. E. & Bahaj, A. S. 2013. Accuracy of the actuator disc-RANS approach for predicting
the performance and wake of tidal turbines. Philosophical Transactions of the Royal Society A 371, 20120293.
Bayati, I., Belloli, M., Ferrari, D., Fossati, F. & Giberti, H. 2014. Design of a 6DoF robotic platform for wind
tunnel tests of floating wind turbines. Energy Procedia, 53, 313323.
710
Bechrakis, D. A., Deane, J. P. & Mckeogh, E. J. 2004. Wind resource assessment of an area using short term data
correlated to a long term data set. Journal of Solar Energy, 76, 725732.
Beels, C., Troch, P., De Visch, K., Kofoed, J. P. & De Backer, G. 2010. Application of the time-dependent mildslope equations for the simulation of wake effects in the lee of a farm of wave dragon wave energy converters.
Renewable Energy, 35, 16441661.
Behera, M. R. & Tkalich, P. 2014. Assessment of kinetic tidal energy resources using SELFE. The International
Journal of Ocean and Climate Systems, 5 (3), 141150.
Bekker, A. T., Sabodash, O. A. & Balakin, B. V. 2013. Numerical prediction of contact surface between hummock
and ice fields for estimation of ice loads on structure. Proceedings of the Twenty-third International Offshore and
Polar Engineering Conference. June 30July 5, Anchorage, Alaska, USA.
Bell, P. S., Lawrence, J. & Norris, J. V. 2012. Determining currents from marine radar data in an extreme current
environment at a tidal energy test site. Proceedings of the Geoscience and Remote Sensing Symposium (IGARSS),
IEEE International. July 2227, Munich, Germany.
Benassai, G., Campanile, A., Piscopo, V. & Scamardella, A. 2014. Ultimate and accidental limit state design for
mooring systems of floating offshore wind turbines. Ocean Engineering, 92, 6474.
Bento, A. R., Rusu, E., Martinho, P. & Guedes Soares , C. 2014. Assessment of the changes induced by the wave
energy farm in the nearshore wave conditions. Computers and Geosciences, 71, 5061.
Bischof, O., Wuerth, I., Tiana Alsina, J., Gutierrez, M. & Cheng, P. W. 2014. Motion effects on lidar wind
measurement data of the EOLOS buoy. The 1st International Conference on Renewable Energies Offshore
(RENEW). November 2426, Lisbon, Portugal.
Bjerks, M., Alsos, H. S. & Wsj, K. 2014. Estimates of the number of vibration cycles from frequency locked-in
ice loads. Proceedings of the ASME 2014 33rd International Conference on Ocean, Offshore and Arctic
Engineering. Paper No. OMAE201423134, June 813, San Francisco, California, USA.
Bjerks, M., Meese, A. & Alsos, H. S. 2013. Ice induced vibrations-observations of a full scale lock-in event.
Proceedings of the Twenty-third International Offshore and Polar Engineering Conference. June 30July 5,
Anchorage, Alaska, USA.
Blackmore, T., Batten, W. M. J. & Bahaj, A. S. 2014. Influence of turbulence on the wake of a marine current
turbine simulator. Proceedings of the Royal Society A, 470, 20140331.
Bland, T. 2004. MERLIN Offshore Wind Turbine Installation System. Technical Report. The Engineering Business Ltd.
Boer, W. W., Verheij, F. J., Zwemmer, D. & Das, R. 2007. The energy islandan inverse pump accumulation station.
Proceedings of the European Wind Energy Conference & Exhibition (EWEC 2007). May 710, Milan, Italy.
Borg, M., Collu, M. & Kolios, A. 2014b. Offshore floating vertical axis wind turbines, dynamics modeling state of the
art. Part II: Mooring line and structural dynamics. Renewable and Sustainable Energy Reviews, 39, 12261234.
Borg, M., Shires, A. & Collu, M. 2014a. Offshore floating vertical axis wind turbines, dynamics modelling state of
the art. Part I: Aerodynamics. Renewable and Sustainable Energy Reviews, 39, 12141225.
Bossler, A. 2013. Floating Offshore Wind Foundations: Industry Consortia and Projects in the United States,
Europe and Japan. Technical Report. Main(e) International Consulting LLC.
Bottasso, C. L., Campagnolo, F. & Petrovi, V. 2014. Wind tunnel testing of scaled wind turbine models: Beyond
aerodynamics. Journal of Wind Engineering and Industrial Aerodynamics, 127, 1128.
Brandon, Y. A., Grard, N. C., Patrick, T. K., G., L. G., Jan, W. C., Koji, O., Kazuyuki, O. & Stephen, M. M. 2010.
Deep ocean water resources in the 21st Century. Marine Technology Society, 44 (3), 8087.
Broadhurst, M., Barr, S., David, C. & Orme, L. 2014. In-situ ecological interactions with a deployed tidal energy
device: an observational pilot study. Ocean & Coastal Management, 99, 3138.
Brommundt, M., Krause, L., Merz, K. & Muskulus, M. 2012. Mooring system optimization for floating wind
turbines using frequency domain analysis. Energy Procedia, 24, 289296.
Buck, B. H. 2011. Opportunities and progress towards a new vision for a green economy in the marine realm: multiuse interaction of offshore wind farms and open ocean aquaculture. North Sea Marine Cluster Conference
(NSMC). April 7, London, UK.
Butterfield, S., Musial, W., Jonkman, J. & Sclavounos, P. 2005. Engineering challenges for floating offshore wind
turbines. Proceedings of the 2005 Copenhagen Offshore Wind Conference. October 2628, Copenhagen, Denmark.
BV 2010. Classification and Certification of Floating Offshore Wind Turbines, Guidance Note NI 572. Paris, France.
Cagninei, A., Bracco, G., Raffero, M., Colicchio, G., Fontan, S., Giorcelli, E., Mattiazzo, G., Orlando, V. & Poggi,
D. 2013. Inertial Sea Wave Energy Converter (ISWEC): scale model and wave tank test. Proceedings of the 10th
European Wave and Tidal Energy Conference. September 25, Aalborg, Denmark.
CarbonTrust 2005. Guidelines on Design and Operation of Wave Energy Converters. A Guide to Assessment and
Application of Engineering Standards and Recommended Practices for Wave Energy Conversion Devices.
Commissioned by the Carbon Trust and carried out by Det Norske Veritas. Available on http://www.glgroup.com/pdf/WECguideline_tcm4270406.pdf. Accessed December 2014.
Casaubieilh, P., Thiebaut, F., Bosma, B., Retzler, C., Shaw, M., Letertre, Y. & Sheng, W. A. 2014. Performance
improvements of mooring systems for wave energy converters. The 1st International Conference on Renewable
Energies Offshore (RENEW). November 2426, Lisbon, Portugal.
Cerveira, F., Fonseca, N. & Pascoal, R. 2013. Mooring system influence on the efficiency of wave energy
converters. International Journal of Marine Energy, 34, 6581.
ClassNK 2012. Guidelines for Offshore Floating Wind Turbine Structures, 1st Edition.
711
Clment, A., Mccullen, P., Falco, A., Fiorentino, A., Gardner, F., Hammarlund, K., Lemonis, G., Lewis, T.,
Nielsen, K., Petroncini, S., Pontes, M. T., Schild, P., Sjstrm, B. O., Srensen, H. C. & Thorpe, T. 2002. Wave
energy in Europe: current status and perspectives. Renewable and Sustainable Energy Reviews, 6 (5), 405431.
Conde, J. M. P. & Gato, L. M. C. 2008. Numerical study of the air-flow in an oscillating water column wave energy
converter. Renewable Energy, 33 (12), 26372644.
Coppens, P. 2014. Tidal energyan emerging market for composites. Reinforced Plastics, 58 (3), 2627.
Copping, A., Hanna, L., Whiting, J., Geerlofs, S., Grear, M., Blake, K., Coffey, A., Massaua, M., Brown-Saracion,
J. & Battey, H. 2013. Environmental Effects of Marine Energy Development around the World, Annex IV Final
Report, IEA OES Initiative, Technical Report, PNNL-22176. IEA.
Coulling, A. J., Goupee, A. J., Robertson, A. N., Jonkman, J. M. & Dagher, H. J. 2013. Validation of a FAST semisubmersible floating wind turbine numerical model with DeepCwind test data. Journal of Renewable and
Sustainable Energy, 5, 023116.
Cradden, L. C., Kalogeri, C., Spyrou, C., Adam, A., Stathopoulos, C., Galanis, G., Sofianos, S., Ingram, D. M.,
Kallos, G., Papapostolou, A. & Axaopoulos, P. 2012. A combined resource atlas for marine energy. Proceedings
of the 4th International Conference on Ocean Energy. October 17, Dublin, Ireland.
Crozier, R., Bailey, H., Mueller, M., Spooner, E., Mckeever, P. & Mcdonald, A. 2011. Hydrodynamic and
electromechanical simulation of a WEC with a novel non-linear PTO. Proceedings of the 9th European Wave
and Tidal Energy Conference. September 59, Southampton, UK.
Cruz, J., Rea, M., Sarmento, A., Thomas, G. & Henderson, R. 2008. Numerical and experimental modelling of
WECs. In: Cruz, J. (ed.) Ocean Wave EnergyCurrent Status and Future Perspectives. Berlin, Heidelberg:
Springer-Verlag.
Curran, R., Folley, M., Danielsson, O., Thorburn, K., Leijon, M., Taylor, J. & Cruz, J. 2008. Power take-off
systems. In: Cruz, J. (ed.) Ocean Wave EnergyCurrent Status and Future Perspectives. Berlin, Heidelberg:
Springer-Verlag.
Da Rocha, A. B., Lino, F. J., Correia, N., Matos, J. C., Marques, M. & Morais, T. 2010. Offshore renewable energy
development of ocean technology projects at Inegi. The VI Cuban Congress on Mechanical Engineering and
Metallurgy (CCIM). November 29December 3, Havana, Cuba.
Damgaard, M., Bayat, M., Andersen, L. V. & Ibsen, L. B. 2014. Assessment of the dynamic behaviour of saturated
soil subjected to cyclic loading from offshore monopile wind turbine foundations. Computers and Geotechnics,
61, 116126.
De Backer, G., Vantorre, M., Frigaard, P., Beels, C. & De Rouck, J. 2010. Bottom slamming on heaving point
absorber wave energy devices. Journal of Marine Science and Technology, 15, 119130.
De Ridder, E.-J., Otto, W., Zondervan, G.-J., Huijs, F. & Vaz, G. 2014. Development of a scaled-down floating
wind turbine for offshore basin testing. Proceedings of the ASME 2014 33rd International Conference on Ocean,
Offshore and Arctic Engineering. Paper No. OMAE201423441, June 813, San Francisco, California, USA.
De Ridder, J. E., Aalberts, P., Van Den Berg, J., Buchner, B. & Peeringa, J. 2011. The dynamic response of an
offshore wind turbine with realistic flexibility to breaking wave impact. Proceedings of the ASME 2011 30th
International Conference on Ocean, Offshore and Arctic Engineering. Paper No. OMAE201149563, June 19
24, Rotterdam, the Netherlands.
Del Jesus, F., Guanche, R. & Losada, I. J. 2014. High resolution reanalysis data and floating met-mast
measurements at deep water locations influenced by coastal topography. The 1st International Conference on
Renewable Energies Offshore (RENEW). November 2426, Lisbon, Portugal.
Delory, E., Brito, J. H. & Llnas, O. 2011. The PLOCAN Observatory: a multidisciplinary multi-platform observing
system for the Central-Eastern Atlantic Ocean. Proceedings of the IEEE OCEANS Conference. June 69,
Santander, Spain.
Divett, T., Vennell, R. & Stevens, C. 2013. Optimization of multiple turbine arrays in a channel with tidally
reversing flow by numerical modelling with adaptive mesh. Philosophical Transactions of the Royal Society A,
371, 20120251.
DNV 2008. Offshore Service Specification (OSS)-312 Certification of Tidal and Wave Energy Converters. Hvik, Norway.
DNV 2010. Recommended Practice - Modelling and Analysis of Marine Operations, RP-H103. Hvik, Norway.
DNV 2013a. Design of Offshore Wind Turbine Structures, DNV-OS-J101. Hvik, Norway.
DNV 2013b. Design of Floating Wind Turbine Structures, DNV-OS-J103. Hvik, Norway.
DOE. 2014. Available: http://energy.gov/eere/wind/offshore-wind-advanced-technology-demonstration-projects
[Accessed November 2014].
Donaire, J. M. S. 2009. Sea Transport Analysis of Upright Wind Turbines. Master Thesis, Technical University of Denmark.
Dong, W. B., Moan, T. & Gao, Z. 2012. Fatigue reliability analysis of the jacket support structure for offshore wind
turbine considering the effect of corrosion and inspection. Reliability Engineering & System Safety, 106, 1127.
Easton, M. C., Woolf, D. K. & Bowyer, P. A. 2012. The dynamics of an energetic tidal channel, the Pentland Firth,
Scotland. Continental Shelf Research, 48, 5060.
EC 2008. Communication from the Commission to the European Parliament, the Council, the European Economic
and Social Committee and the Committee of the Regions, Offshore Wind Energy: Action Needed to Deliver on
the Energy Policy Objectives for 2020 and Beyond, COM(2008)768 Final. European Commission.
Edwards, E. C., Cradden, L. C., Ingram, D. M. & Kalogeri, C. 2014a. Verification within wave resource
assessments. Part 1: Statistical analysis. International Journal of Marine Energy, 8, 5069.
Edwards, E. C., Cradden, L. C., Ingram, D. M. & Kalogeri, C. 2014b. Verification within wave resource assessments.
Part 2: Systematic trends in the fit of spectral values. International Journal of Marine Energy, 8, 7083.
712
Engstrm, J., Eriksson, M., Gteman, M., Isberg, J. & Leijon, M. 2013. Performance of large arrays of point
absorbing direct-driven wave energy converters. Journal of Applied Physics, 114, 204502.
EWEA. 2014. The European Offshore Wind IndustryKey Trends and Statistics 2013 [Online]. Available:
http://www.ewea.org/fileadmin/files/library/publications/statistics/European_offshore_statistics_2013.pdf
[Accessed November 2014].
Falco, A. F. O. 2008. Phase control through load control of oscillating-body wave energy converters with
hydraulic PTO system. Ocean Engineering, 35 (34), 358366.
Falco, A. F. O. 2010. Wave energy utilization: A review of the technologies. Renewable and Sustainable Energy
Reviews, 14 (3), 899918.
Falco, A. F. O. & Gato, L. M. C. 2012. Air turbines. In: Sayigh, A. (ed.) Comprehensive Renewable Energy.
Oxford, UK: Elsevier.
Falco, A. F. O., Gato, L. M. C. & Nunes, E. P. a. S. 2011. A new radial self-rectifying air turbine for use in OWC
wave energy converters. Proceedings of the 9th European Wave and Tidal Energy Conference. September 59,
Southampton, UK.
Falco, A. F. O. & Henriques, J. C. C. 2014. Model-prototype similarity of oscillating-water-column wave energy
converters. International Journal of Marine Energy, 6, 1834.
Falco, A. F. O., Henriques, J. C. C. & Cndido, J. J. 2012. Dynamics and optimization of the OWC spar buoy
wave energy converter. Renewable Energy, 48, 369381.
Fonseca, N. & Pessoa, J. 2013. Numerical modelling of a wave energy converter based on U-shaped interior
oscillating water column. Applied Ocean Research, 40, 6073.
FOWC 2011. Fukushima Floating Offshore Wind Farm Demonstration Project (Fukushima FORWARD).
Fukushima Offshore Wind Consortium.
FOWC 2013. Fukushima Floating Offshore Wind Farm Demonstration Project (Fukushima FORWARD)
Construction of Phase I. Fukushima Offshore Wind Consortium.
Fowler, M. J., Kimball, R. W., Thomas Iii, D. A. & Goupee, A. J. 2013. Design and testing of scale model wind
turbines for use in wind/wave basin model tests of floating offshore wind turbines. Proceedings of the ASME
2013 32nd International Conference on Ocean, Offshore and Arctic Engineering. Paper No. OMAE201310122,
June 914, Nantes, France.
Frid, C., Andonegi, E., Depestele, J., Judd, A., Rihan, D., Rogers, S. I. & Kenchington, E. 2012. The environmental
interactions of tidal and wave energy generation devices. Environmental Impact Assessment Review, 32 (1), 133139.
Furness, R. W., Wade, H. M., Robbins, A. M. C. & Masden, E. A. 2012. Assessing the sensitivity of seabird
populations to adverse effects from tidal stream turbines and wave energy devices. ICES Journal of Marine
Science, 69 (8), 14661479.
Fylling, I. & Berthelsen, P. 2011. WINDOPTAn optimization tool for floating support structures for deepwater
wind turbines. Proceedings of the ASME 2011 30th International Conference on Ocean, Offshore and Arctic
Engineering. Paper No. OMAE201149985, June 1924, Rotterdam, the Netherlands.
Gkikas, G. D. & Athanassoulis, G. A. 2014. Development of a novel nonlinear system identification scheme for the
pressure fluctuation inside an oscillating water column-wave energy converter, Part I: Theoretical background
and harmonic excitation case. Ocean Engineering, 80, 8499.
GL-NobleDenton 2013a. Guidelines for Marine Transportations, 0030/ND. Rev. 5.
GL-NobleDenton 2013b. Guidelines for Marine Lifting & Lowering Operations, 0027/ND. Rev. 10.
GL-NobleDenton 2013c. Guidelines for the Transportation & Installation of Steel Jackets, 0028/ND. Rev. 5.
GL-NobleDenton 2013d. Guidelines for Concrete Offshore Gravity Structure Construction & Installation, 0015/ND. Rev. 4.
GL 2005. Rules and Guidelines, IV Industrial Services, Part 2 Guideline for the Certification of Offshore Wind
Turbines. Hamburg, Germany.
Gomes, R. P. F., Henriques, J. C. C., Gato, L. M. C. & Falco, A. F. O. 2010. IPS two-body wave energy converter:
Acceleration tube optimization. Proceedings of the Twentieth International Offshore and Polar Engineering
Conference. June 2025, Beijing, China.
GOTO-FOWT-Website. 2014. Available: http://goto-fowt.go.jp/ [Accessed December 2014].
Goundar, J. N. & Ahmed, M. R. 2014. Marine current energy resource assessment and design of a marine current
turbine for Fiji. Renewable Energy, 65, 1422.
Goupee, A. J., Fowler, M. J., Kimball, R. W., Helder, J. & De Ridder, E.-J. 2014. Additional wind/wave basin
testing of the DeepCwind semi-submersible with a performance-matched wind turbine. Proceedings of the ASME
2014 33rd International Conference on Ocean, Offshore and Arctic Engineering. Paper No. OMAE201424172,
June 813, San Francisco, California, USA.
Graczyk, M. & Sandvik, P. C. 2012. Study of landing and lift-off operation for wind turbine components on a ship
deck. Proceedings of the ASME 2012 31st International Conference on Ocean, Offshore and Arctic Engineering.
Paper No. OMAE201284273, July 16, Rio de Janeiro, Brazil.
Guedes Soares, C., Bento, A. R., Gonalves, M., Silva, D. & Martinho, P. 2014b. Numerical evaluation of the wave
energy resource along the Atlantic European coast. Computers and Geosciences, 71, 3749.
Guedes Soares, C., Bhattachrjee, J. & Karmakar, D. 2014a. Overview and prospects for offshore wave and wind
energy. Brodogradnja, 65 (2), 91113.
Gueydon, M. & Weller, S. 2013. Study of a floating foundation for wind turbines. Journal of Offshore Mechanics
and Arctic Engineering, 135 (3), 031903.
Gunawan, B., Neary, V. S. & Colby, J. 2014. Tidal energy site resource assessment in the East River tidal strait,
near Roosevelt Island, New York. Renewable Energy, 71, 509517.
713
Gnther, H. & Behrens, A. 2011. The WAM Model Validation Document, version 4.5.3. Technical Report. Institute
of Coastal Research, Helmholtz-Zentrum Geesthacht.
Guo, Y., Keller, J. & Lacava, W. 2014. Planetary gear load sharing of wind turbine drivetrains subjected to nontorque loads. Wind Energy, DOI: 10.1002/we.1731.
Hall, M., Buckham, B. & Crawford, C. 2014b. Hydrodynamics-based floating wind turbine support platform
optimization: A basis function approach. Renewable Energy, 66, 559569.
Hall, M., Buckham, N. & Crawford, C. 2014a. Evaluating the importance of mooring line model fidelity in floating
offshore wind turbine simulations. Wind Energy, 17 (12), 18351853.
Hall, M., Moreno, J. & Thiagarajan, K. 2014c. Performance specifications for real-time hybrid testing of 1:50scale
floating wind turbine models. Proceedings of the ASME 2014 33rd International Conference on Ocean, Offshore
and Arctic Engineering. June 813, San Francisco, California, USA.
Helsen, J., Vanhollebeke, F., De Coninck, F., Vandepitte, D. & Desmet, W. 2011a. Insights in wind turbine drive
train dynamics gathered by validating advanced models on a newly developed 13.2 MW dynamically controlled
test-rig. Mechantronics, 21 (4), 737752.
Helsen, J., Vanhollebeke, F., Marrant, B., Vandepitte, D. & Desmet, W. 2011b. Multibody modelling of varying
complexity for modal behaviour analysis of wind turbine gearboxes. Renewable Energy, 36 (11), 30983113.
Henderson, R. 2006. Design, simulation, and testing of a novel hydraulic power take-off system for the Pelamis
wave energy converter. Renewable Energy, 31, 271283.
Hendrikse, H., Renting, F. W. & Metrikine, A. V. 2014. Analysis of the fatigue life of offshore wind turbine
generators under combined ice and aerodynamic loading. Proceedings of the ASME 2014 33rd International
Conference on Ocean, Offshore and Arctic Engineering. Paper No. OMAE201423884, June 813, San
Francisco, California, USA.
Hermans, A. J., Van Sabben, E. & Pinkster, J. A. 1990. A device to extract energy from water waves. Applied
Ocean Research, 12 (4), 175179.
Hilding, D., Forsberg, J. & Grtner, A. 2012. Simulation of loads from drifting ice sheets on offshore structure. The
12th LS-Dyna User Conference.
Hirdaris, S. E., Bai, W., Dessi, D., Ergin, A., Gu, X. K., Hermundstad, O. A., Huijsmans, R., Iijima, K., Nielsen, U.
D., Parunov, J., Fonseca, N., Papanikolaou, A., Argyriadis, K. & Incecik, A. 2014. Loads for use in the design of
ships and offshore structures. Ocean Engineering, 78, 131174.
Hou, J. & Shao, W. 2014. Structural design for the ice-resistant platform. Proceedings of the Twenty-fourth
International Offshore and Polar Engineering Conference. June 1520, Busan, Korea.
Hsu, W.-T., Thiagarajan, K. P., Hall, M., Macnicoll, M. & R., A. 2014. Snap loads on mooring lines of a floating
offshore wind turbine structure. Proceedings of the ASME 2014 33rd International Conference on Ocean,
Hu, Z. Z., Causon, D. M., Mingham, C. G. & Qian, L. 2011. Numerical simulation of floating bodies in extreme
free surface waves. Natural Hazards and Earth System Sciences, 11 (2), 519527.
Huijs, F., De Ridder, E.-J. & Savenije, F. 2014. Comparison of model tests and coupled simulations for a semisubmersible floating wind turbine. Proceedings of the ASME 2014 33rd International Conference on Ocean,
IEA 2013. IEA Technology RoadmapWind Energy. Paris, France: IEA.
IEA 2014a. IEA Wind 2013 Annual Report. Prepared by the Executive Committee of the Implementing Agreement
for Co-operation in the Research, Development, and Deployment of Wind Energy Systems of the International
Energy Agency.
IEA 2014b. OESOcean Energy Systems, an International Energy Agency Technology Initiative. Retrieved from
the authoritative international voice on Ocean Energy: http://www.ocean-energy-systems.org.
IEC-TC-114. 2014. Technical Committee on Marine Energy Wave, Tidal and Other Water Current Converters
[Online]. International Electrotechnical Commission. Available: http://www.iec.ch/dyn/www/f?p=103:23:0::::
FSP_ORG_ID,FSP_LANG_ID:1316,25 [Accessed December 2014].
IEC 2005. Wind TurbinePart 1: Design Requirements, IEC 614001, Third Edition. International Electrotechnical
Commission.
IEC 2009. Wind TurbinesPart 3: Design Requirements for Offshore Wind turbines, IEC 614003, First Edition.
International Electrotechnical Commission.
IEC 2012. Wind Turbines, Part 4: Standard for Design and Specification of Gearboxes, IEC 614004. International
Electrotechnical Commission.
Imai, Y., Nagata, S., Toyota, K. & Murakami, T. 2014. An experimental study on primary efficiency of a wave
energy converter Backward Bent Duct Buoy in regular wave conditions. Journal of the Japan Society of Naval
Architects and Ocean Engineers, 19, 7988.
Ishida, S., Kokubun, K. & Nimura, T. 2013. At-sea experiment of a hybrid spar type offshore wind turbine.
Proceedings of the ASME 2013 32nd International Conference on Ocean, Offshore and Arctic Engineering.
Paper No. OMAE201310655, June 914, Nantes, France.
ISO 2006. Calculation of Load Capacity of Spur and Helical Gears, Part 3: Calculation of Tooth Bending Strength,
ISO 63363. International Organization for Standardization.
ISO 2010. Petroleum and Natural Gas IndustriesArctic Offshore Structures, ISO/FDIS 19906. International
Organization for Standardization.
ISSC 2006. Report of Specialist Committee V.4Ocean Wind and Wave Energy Utilization. Proceedings of the
16th International Ship and Offshore Structures Congress. August 2025, Southampton, UK, 165211.
714
ISSC 2009. Report of Specialist Committee V.4Ocean Wind and Wave Energy Utilization. Proceedings of the
17th International Ship and Offshore Structures Congress. August 1621, Seoul, Korea, 201257.
ISSC 2012. Report of Specialist Committee V.4Offshore Renewable Energy. Proceedings of the 18th
International Ship and Offshore Structures Congress. September 1013, Rostock, Germany, 153199.
Jacobsen, P. 2011. Mapping and Assessment of the United States Ocean Wave Energy Resource. Retrieved from
http://energy.gov/sites/prod/files/2013/12/f5/mappingandassessment.pdf. Palo Alto, California, USA: Electric
Power Research Institute.
Janssen, R., Arciniegas, G. & Alexander, K. A. 2014. Decision support tools for collaborative marine spatial
planning: identifying potential sites for tidal energy devices around the Mull of Kintyre, Scotland. Journal of
Environmental Planning and Management, March 2014.
Jeans, T. L., Fagley, C., Siegel, S. G. & Seidel, J. 2013. Irregular deep ocean wave energy attenuation using a
cycloidal wave energy converter. International Journal of Marine Energy, 1, 1632.
Jeon, S. H., Cho, Y. U., Seo, M. W., Cho, J. R. & Jeong, W. B. 2013. Dynamic response of floating substructure of
spar-type offshore wind turbine with catenary mooring cables. Ocean Engineering, 72, 356364.
Jiang, Z. Y., Karimirad, M. & Moan, T. 2013. Dynamic response analysis of wind turbines under blade pitch system
fault, grid loss, and shutdown events. Wind Energy, 17 (9), 13851409.
Jimenez, B., Durante, F., Lange, B., Kreutzer, T. & Tambke, J. 2007. Offshore wind resource assessment with
WAsP and MM5: Comparative study for the German Bight. Wind Energy, 10, 121134.
Johanning, L., Smith, G. H. & Wolfram, J. 2007. Measurements of static and dynamic mooring line damping and
their importance for floating WEC devices. Ocean Engineering, 34 (1415), 19181934.
Johnson, H. K. 1998. On modelling wind-waves in shallow and fetch limited areas using the method of Holthuijsen,
Booij and Herbers. Journal of Coastal Research, 14 (3), 917932.
Johnson, K., Kerr, S. & Side, J. 2012. Accommodating wave and tidal energyControl and decision in Scotland.
Ocean & Coastal Management, 65, 2633.
Johnson, K. E. & Fleming, P. A. 2011. Development, implementation, and testing of fault detection strategies on the
National Wind Technology Center's controls advanced research turbines. Mechatronics, 21 (4), 728736.
Johnstone, C. M., Pratt, D., Clarke, J. A. & Grant, A. D. 2013. A techno-economic analysis of tidal energy
technology. Renewable Energy, 49, 101106.
Jonkman, J. & Musial, W. 2010. Offshore Code Comparison Collaboration (OC3) for IEA Task 23 Offshore Wind
Technology and Deployment. Technical Report, NREL/TP-500048191. Golden, Colorado, USA: National
Renewable Energy Laboratory (NREL).
Jonkman, J. M. & Matha, D. 2011. Dynamics of offshore floating wind turbinesAnalysis of three concepts. Wind
Energy, 14 (4), 557569.
JWPA. 2012. Available: http://jwpa.jp/pdf/5032roadmapV3.2.pdf [Accessed December 2014].
JWPA 2014. Target and Roadmap for Japanese Wind Power. The Grand Renewable Energy 2014 International
Conference and Exhibition. July 27August 1, Tokyo, Japan.
Kadiri, M., Ahamadian, R., Bockelmann-Evans, B., Rauen, W. & Falconer, R. 2012. A review of the potential
water quality impacts of tidal renewable energy systems. Renewable and Sustainable energy Reviews, 16 (1),
329341.
Kaldellis, J. K. & Zafirakis, D. P. 2012. Trends, prospects and R & D directions in wind turbine technology. In:
Sayigh, A. (ed.) Comprehensive Renewable Energy. Elsevier, 671724.
Kamizuru, Y. 2012. Efficient power take-offs for ocean energy conversion. Proceedings of the 4th International
Conference on Ocean Energy. October 1719, Dublin, Ireland.
Kamizuru, Y., Fissmann, C. & Murrenhoff, H. 2013. Hydrostatic drive trains for wave energy converters:
Simulation and experiments for efficient design. Proceedings of 10th European Wave and Tidal Energy
Conference. September 25, Aalborg, Denmark.
Karimirad, M. 2014. Design aspects. In: Karimirad, M. (ed.) Offshore Energy Structures. Springer International
Publishing, 129164.
Karimirad, M. & Moan, T. 2013. Stochastic dynamic response analysis of a tension leg spar-type offshore wind
turbine. Wind Energy, 16 (6), 953973.
Karmakar, D. & Guedes Soares, C. 2013. Reliability based design loads of an offshore semi-submersible floating
wind turbine. In: Guedes Soares, C. & Pea, F. (eds.) Developments in Maritime Transportation and
Exploitation of Sea Resources. London, UK: Taylor & Francis Group, 919926.
Karmakar, D. & Guedes Soares, C. 2014. Review on the design criteria and scope of multi-use offshore platforms. The
1st International Conference on Renewable Energies Offshore (RENEW). November 2426, Lisbon, Portugal.
Katsuyama, I., Kobayashi, S., Igawa, S., Kyozuka, Y. & Ida, M. 2014. Biofouling of model turbines for tidal
current power generation and the effect of anti-fouling paint. Sessile Organisms (The Sessile Organisms Society
of Japan), 31 (1), 15.
Kelly, S. M., Nash, J. D., Martini, K. I., Alford, M. H. & Kunze, E. 2012. The cascade of tidal energy from low to
high modes on a continental slope. Journal of Physical Oceanography, 42 (7).
Kim, G., Lee, M. E., Lee, K. S., Park, J. S., Jeong, W. M., Kang, S. K., Soh, J. G. & Kim, H. 2012a. An overview
of ocean renewable energy resources in Korea. Renewable and Sustainable Energy Reviews, 16 (4), 22782288.
Kim, Y.-K., Shin, J.-R. & Yoon, D.-Y. 2012b. A design of windmill turbine installation vessel using jack-up
system. Proceedings of the Twenty-second International Offshore and Polar Engineering Conference. June 17
22, Rhodes, Greece.
715
Kofoed, J. P., Frigaard, P., Friis-Madsen, E. & Srensen, H. C. 2006. Prototype testing of the wave energy
converter Wave Dragon. Renewable Energy, 31, 181189.
Konispoliatis, D. N. & Mavrakos, S. A. 2013a. Hydrodynamics of multiple vertical axisymmetric OWCs devices
restrained in waves. Proceedings of the ASME 2013 32nd International Conference on Ocean, Offshore and
Arctic Engineering. Paper No. 11156, June 914, Nantes, France.
Konispoliatis, D. N. & Mavrakos, S. A. 2013b. Hydrodynamics of arrays of OWCs devices consisting of
concentric cylinders restrained in waves. Proceedings of the 10th European Wave and Tidal Energy Conference
(EWTEC 2013). September 25, Aalborg, Denmark.
Konispoliatis, D. N. & Mavrakos, S. A. 2014a. Mean drift loads on arrays of free floating OWC devices consisting
of concentric cylinders. Proceedings of the 29th International Workshop on Water Waves and Floating Bodies
(IWWWFB2014). March 30April 2, Osaka, Japan.
Konispoliatis, D. N. & Mavrakos, S. A. 2014b. Hydrodynamics and power absorption characteristics of free
floating and moored arrays of OWCs devices. Proceedings of the ASME 2014 33rd International Conference on
Ocean, Offshore and Arctic Engineering. Paper No. OMAE201424493, June 813, San Francisco, California,
USA.
Korsnes, M. 2014. Chinas Offshore Wind Industry 2014An Overview of Current Status and Development. CenSES.
Kurniawan, A. & Moan, T. 2013. Optimal geometries for wave absorbers oscillating about a fixed axis. IEEE
Journal of Oceanic Engineering, 38, 117130.
Kvittem, M. I., Bachynski, E. E. & Moan, T. 2012. Effects of hydrodynamic modelling in fully coupled Simulation
of a semisubmersible wind turbine. Energy Procedia, 24, 351362.
Kvittem, M. I. & Moan, T. 2015. Frequency versus time domain fatigue analysis of a semi-submersible wind
turbine tower. Journal of Offshore Mechanics and Arctic Engineering, 137 (1), 011901.
Lazakis, I., Turan, O. & Rosendahl, T. 2013. Modelling of vessel and equipment cost for the maintenance activities
of an offshore tidal energy array. Proceedings of the 12th International Symposium on Practical Design of Ships
and Other Floating Structures (PRADS2013). October 2025, Changwon City, Korea.
Lee, C. 1995. WAMIT Theory Manual. Cambridge, Massachusetts, USA: Department of Ocean Engineering,
Massachusetts Institute of Technology.
Lee, K. H. 2005. Responses of Floating Wind Turbines to Wind and Wave Excitation. Master Thesis, Department of
Ocean Engineering, Massachusetts Institute of Technology.
Lee, Y. S., Choi, B. L., Lee, J. H., Kim, S. Y. & Han, S. 2014. Reliability-based design optimization of monopile
transition piece for offshore wind turbine system. Renewable Energy, 71, 729741.
Leijon, M., Bostrm, C., Danielsson, O., Gustafsson, S., Haikonen, K., Langhamer, O., Strmstedt, E., Stlberg, M.,
Sundberg, J., Svensson, O., Tyrberg, S. & Waters, R. 2008. Wave energy from the North Sea: Experiences from
the Lysekil research site. Surveys in Geophysics, 29 (3), 221240.
Lejerskog, E., Bostrm, C., Hai, L., Waters, R. & Leijon, M. 2015. Experimental results on power absorption from
a wave energy converter at the Lysekil wave energy research site. Renewable Energy, 77, 914.
Li, L., Gao, Z., Moan, T. & Ormberg, H. 2014. Analysis of lifting operation of a monopile for an offshore wind
turbine considering vessel shielding effects. Marine Structures, 39, 287314.
Li, Y. & Yu, Y. H. 2012. A synthesis of numerical methods for modelling wave energy converter Point absorbers.
Renewable and Sustainable Energy Reviews, 16, 43524364.
Liberti, L., Carillo, A. & Sannino, G. 2013. Wave energy resource assessment in the Mediterranean, the Italian
perspective. Renewable Energy, 50, 938949.
Link, H., Lacava, W., Van Dam, J., Mcniff, B., Sheng, S., Wallen, R., Mcdade, M., Lambert, S., Butterfield, S. &
Oyague, F. 2011. Gearbox Reliability Collaborative Project Report: Findings from Phase 1 and Phase 2,
Technical Report. Golden, Colorado, USA: National Renewable Energy Laboratory.
Liu, F., Li, H., Li, W. & Wang, B. 2014a. Experimental study of improved modal strain energy method for damage
localisation in jacket-type offshore wind turbines. Renewable Energy, 72, 174181.
Liu, P. 2012a. Oscillating Foil Turbine. PCT Patent: PCT/CA2011001224.
Liu, P. 2012b. Wing-in-ground effect oscillating foil turbine: From concept to innovation. Proceedings of the Power
and Energy Systems Conference (AsiaPES 2012). April 24, Phuket, Thailand.
Liu, P. & Bose, N. 2012. Prototyping a series of bi-directional horizontal axis tidal turbines for optimum energy
conversion. Applied Energy, 99, 5066.
Liu, P., Bose, N., Frost, R., Macfarlane, G., Lilienthal, T., Penesis, I., Windsor, F. & Thomas, G. 2014b. Model
testing of a series of bi-directional tidal turbine rotors. Energy, 67, 397410.
Liu, P. & Veitch, B. 2012. Design and optimization for strength and integrity of tidal turbine rotor blades. Energy,
46 (1), 393404.
Long, H. Y. & Moe, G. 2012. Preliminary design of bottom-fixed lattice offshore wind turbine towers in the fatigue limit
state by the frequency domain method. Journal of Offshore Mechanics and Arctic Engineering, 134 (3), 031902.
Lotsberg, I., Serednicki, A., Bertnes, H. & Lervik, A. 2012a. On the capacity of grouted connections in monopile
offshore wind turbine structures. Proceedings of the 10th International Conference on Advances in Steel
Concrete Composite and Hybrid Structures. July 24, Singapore.
Lotsberg, I., Serednicki, A., Bertnes, H. & Lervik, A. 2012b. Design of grouted connections for monopile offshore
structuresResults from two Joint Industry Projects. Stahlbau, 81 (9), 695704.
Lotsberg, I., Serednicki, A., Oerlemans, R., Bertnes, H. & Lervik, A. 2013. Capacity of cylindrical shaped grouted
connections with shear keys in offshore structures. The Structural Engineer, 91 (1), 4248.
716
Lovas, S., Mei, C. C. & Liu, Y. M. 2010. Oscillating water column at a coastal corner for wave power extraction.
Applied Ocean Research, 32, 267283.
Lu, S. Y., Yu, J. C. S., Golmen, L., Wesnigk, J., Papandroulakis, N., Anastasiadis, P., Delroy, E., Quevedo, E.,
Hernandez, J. & Llinas, O. 2014. Environmental aspects of designing multi-purpose offshore platforms in the
scope of the FP7 TROPOS project. Proceedings of the IEEE OCEANS Conference. April 710, Taipei, Taiwan,
China.
Luan, C., Gao, Z. & Moan, T. 2013. Modelling and analysis of a semi-submersible wind turbine with a central
tower with emphasis on the brace system. Proceedings of the AMSE 2013 32nd International Conference on
Ocean, Offshore and Arctic Engineering. Paper No. OMAE201310408, June 914, Nantes, France.
Luo, Y., Nader, J. R., Cooper, P. & Zhu, S. P. 2014. Nonlinear 2D analysis of the efficiency of fixed oscillating
water column wave energy converters. Renewable Energy, 64, 255265.
LysekilProject. 2015. Available: http://www.el.angstrom.uu.se/forskningsprojekt/WavePower/Lysekilsprojektet_E
.html [Accessed March 2015].
Maciel, J. G. 2012. The WindFloat project - Deep offshore windAn opportunity for Europe. The Atlantic Forum
organized by European Commission. October 2930, Brest, France.
MacLean, I. M. D., Inger, R., Benson, D., Booth, C. G., Embling, C. B., Grecian, W. J., Heymans, J. J., Plummer, K. E.,
Shackshaft, M., Sparling, C. E., Wilson, B., Wright, L. J., Bradbury, G., Christen, N., Godley, B. J., Jackson, A. C.,
Mccluskie, A., Nicholls-Lee, R. & Bearhop, S. 2014. Resolving issues with environmental impact assessment of
marine renewable energy installations. Frontier in Marine Science, DOI: 10.3389/fmars.2014.00075.
Mardfekri, M. & Gardoni, P. 2013. Probabilistic demand models and fragility estimates for offshore wind turbine
support structures. Engineering Structures, 52, 478487.
Martin, H. R., Kimball, R. W., Viselli, A. M. & Goupee, A. J. 2012. Methodology for wind/wave basin testing of
floating offshore wind turbines. Proceedings of the ASME 2012 31st International Conference on Ocean,
Offshore and Arctic Engineering. Paper No. OMAE201283627, July 16, Rio de Janeiro, Brazil.
Masciola, M. & Jonkman, J. M. 2014. Extending the capabilities of the mooring analysis program: a survey of
dynamic mooring line theories for integration into FAST. Proceedings of the ASME 2014 33rd International
Masciola, M. D., Robertson, A. N., Jonkman, J. M., Coulling, A. J. & Goupee, A. J. 2013. Assessment of the
importance of mooring dynamics on the global response of the DeepCwind floating semisubmersible offshore
wind turbine. Proceedings of the Twenty-third International Offshore and Polar Engineering Conference. June
30July 5, Anchorage, Alaska, USA.
Mavrakos, S. A. & Konispoliatis, D. N. 2012. Hydrodynamics of a free floating vertical axisymmetric oscillating
water column device. Journal of Applied Mathematics, Special Issue on Mathematical Modelling of Marine
Structures, Article ID 142850.
McGovern, D. J. & Bai, W. 2014. Experimental study of wave-driven impact of sea ice floes on a circular cylinder.
Cold Regions Science and Technology, 108, 3648.
Menicou, M. & Vassiliou, V. 2010. Prospective energy needs in Mediterranean offshore aquaculture. Renewable
and Sustainable Energy Reviews, 14 (9), 30843091.
Michailides, C., Luan, C. Y., Gao, Z. & Moan, T. 2014. Effect of flap type wave energy converters on the response
of a semi-submersible wind turbine in operational conditions. Proceedings of the ASME 2014 33rd International
MIKE-21 2008. Wave Modelling User Guide. Hrsholm, Denmark: Danish Hydraulic Institute.
Molyneux, D., Spencer, D. & Liu, L. 2013. Loads due to first year ice ridges on a vertical cylinder. Proceedings of
the ASME 2013 32nd International Conference on Ocean, Offshore and Arctic Engineering. Paper No.
OMAE201310068, June 914, Nantes, France.
Morandeau, M., Walker, R. T., Argall, R. & Nicholls-Lee, R. F. 2013. Optimisation of marine energy installation
operations. International Journal of Marine Energy, 34, 1426.
Mrk, G., Barstow, S., Mollison, D. & Cruz, J. 2008. The wave energy resources. In: Cruz, J. (ed.) Ocean Wave
EnergyCurrent Status and Future Perspectives. Berlin, Heidelberg: Springer-Verlag.
Mueller, M. A., Polinder, H. & Baker, N. 2007. Current and novel electrical generator technology for wave energy
converters. Proceedings of the International Electric Machines & Drives Conference. May 35, Antalya, Turkey.
Muliawan, M. J., Gao, Z., Moan, T. & Babarit, A. 2013. Analysis of a two-body floating wave energy converter
with particular focus on the effects of power take-off and mooring systems on energy capture. Journal of
Offshore Mechanics and Arctic Engineering, 135, 031902.
Mller, K., Sandner, F., Bredmose, H., Azcona, J., Manjock, A. & Pereira, R. 2014. Improve tank test procedures
for scaled floating offshore wind turbines. The International Wind Engineering Conference Support Structures
& Electrical Systems. September 34, Hannover, Germany.
Myers, L. E. & Bahaj, A. S. 2012. An experimental investigation simulating flow effects in first generation marine
current energy converter arrays. Renewable Energy, 37 (1), 2836.
Myhr, A., Maus, K. J. & Nygaard, T. A. 2011. Experimental and computational comparisons of the OC3HYWIND
and Tension-Leg-Buoy (TLB) floating wind turbine conceptual designs. Proceedings of the Twenty-first
International Offshore and Polar Engineering Conference. June 1924, Maui, Hawaii, USA.
717
Myhr, A. & Nygaard, T. A. 2012. Load reductions and optimizations on Tension-Leg-Buoy offshore wind turbine
platforms. Proceedings of the Twenty-second International Offshore and Polar Engineering Conference. June
1722, Rhodos, Greece.
Myrhaug, D. & Ong, M. C. 2013. Scour around vertical pile foundations for offshore wind turbines due to longcrested and short-crested nonlinear random waves. Journal of Offshore Mechanics and Arctic Engineering, 135
(1), 011103.
Nader, J. R., Zhu, S. P., Cooper, P. & Stappenbelt, B. 2012. A finite-element study of the efficiency of arrays of
oscillating water column wave energy converters. Ocean Engineering, 43, 7281.
NEDO. 2013. Offshore Wind Energy Progress, Edition II [Online]. Available: http://www.nedo.go.jp/content/
100534312.pdf [Accessed December 2014].
Negro, V., Lpez-Gutirrez, J.-S., Esteban, M. D. & Matutano, C. 2014. Uncertainties in the design of support
structures and foundations for offshore wind turbines. Renewable Energy, 63, 125132.
Neill, S. P., Jordan, J. R. & Couch, S. J. 2012. Impact of tidal energy converter (TEC) arrays on the dynamics of
headland sand banks. Renewable Energy, 37 (1).
Nejad, A. R., Gao, Z. & Moan, T. 2013. Long-term analysis of gear loads in fixed offshore wind turbines
considering ultimate operational loadings. Energy Procedia, 35, 187197.
Nicholls-Lee, R. F., Hindley, S. & Parkinson, R. P. 2013. Development of an economic and efficient installation
vessel for tidal stream energy converter arrays. Proceedings of the ASME 2013 32nd International Conference on
Ocean, Offshore and Arctic Engineering. Paper No. OMAE201310663, June 914, Nantes, France.
Nicholls-Lee, R. F., Turnock, S. R. & Boyd, S. W. 2012. Application of bend-twist coupled blades for horizontal
axis tidal turbines. Renewable Energy, 50, 541550.
Nielsen, F. G. 2013. HywindDeep offshore wind operational experience. The 10th Deep Sea Offshore Wind R & D
Conference (DeepWind 2013). January 2425, Trondheim, Norway.
Nielsen, K. 2010. Report T020.0 Development of Recommended Practices for Testing and Evaluating Ocean
Energy Systems, OES-IA Annex II Extension Summary Report. International Energy Agency.
Norouzi, M., Wells, E., Cioc, S., Nikolaidis, E. & Afjeh, A. 2013. Significance of ice impact on structural integrity
of a monopile offshore wind turbine in the Great Lakes. Proceedings of the Twenty-third International Offshore
and Polar Engineering Conference. June 30July 5, Anchorage, Alaska, USA.
NORSOK 1997. NORSOK StandardMarine Operations, J-003. Rev. 2.
Orourke, F., Boyle, F. & Reynolds, A. 2014. Irelands tidal energy resource: An assessment of a site in the Bulls
Mouth and the Shannon Estuary using measured data. Energy Conversion and Management, 87, 726734.
Osullivan, K., Murphy, J. & Osullivan, D. 2013. Power output performance and smoothing ability of an
oscillating water column array wave energy converter. Proceedings of the ASME 2013 32nd International
Conference on Ocean, Offshore and Arctic Engineering. Paper No. OMAE201311375, June 914, Nantes,
France.
OEE. 2014. Ocean Energy Europe, a Trade Association for Ocean Renewables [Online]. Available:
http://www.oceanenergy-europe.eu/index.php/en/.
stvik, I. 2012. Cost-effective foundation installation vessels for offshore wind. Conference on Offshore Wind
Operations. May 1415, Bergen, Norway.
Oyague, F., Butterfield, C. P. & Sheng, S. 2009. Gearbox Reliability Collaborative Analysis Round Robin,
Technical Report, NREL/CP-50045325. Golden, Colorado, USA: National Renewable Energy Laboratory.
Pacheco, A., Ferreira, ., Carballo, R. & Iglesias, G. 2014. Evaluation of the production of tidal stream energy in an
inlet channel by coupling field data and numerical modelling. Energy, 71, 104117.
Palodichuk, M., Polagye, B. & Thomson, J. 2013. Resource mapping at tidal energy sites. IEEE Journal of Oceanic
Engineering, 38 (3), 433446.
Paulsen, B. T., Bredmose, H., Bingham, H. B. & Schler, S. 2013. Steep wave loads from irregular waves on an
offshore wind turbine foundation: Computation and Experiment. Proceedings of the ASME 2013 32nd
International Conference on Ocean, Offshore and Arctic Engineering. Paper No. OMAE201310727, June 914,
Nantes, France.
Pecher, A., Kofoed, J. P., Espedal, J. & Hagberg, S. 2010. Results of an experimental study of the Langlee wave
energy converter. Proceedings of the Twentieth International Offshore and Polar Engineering Conference. June
2025, Beijing, China.
Peeters, J. L. M., Vandepitte, D. & Sas, P. 2006. Analysis of internal drive train dynamics in a wind turbine. Wind
Energy, 9 (12), 141161.
Perez-Collazo, C., Greaves, D. & Iglesias, G. 2015. A review of combined wave and offshore wind energy.
Perveen, R., Kishor, N. & Mohanty, S. R. 2014. Offshore wind farm development: Present status and challenges.
Philippe, M., Babarit, A. & Ferrant, P. 2013b. Modes of response of an offshore wind turbine with directional wind
and waves. Renewable Energy, 49, 151155.
Philippe, M., Courbois, A., Babarit, A., Bonnefoy, F., Rousset, J.-M. & Ferrant, P. 2013a. Comparison of
simulation and tank test results of a semi-submersible floating wind turbine under wind and wave loads.
Proceedings of the ASME 2013 32nd International Conference on Ocean, Offshore and Arctic Engineering.
Paper No. OMAE201311271, June 914, Nantes, France.
718
Polagye, B. & Thomson, J. 2013. Tidal energy resource characterization: methodology and field study in Admiralty
Inlet, Puget Sound, WA (USA). Proceedings of the Institute of Mechanical Engineers, Part A: Journal of Power
and Energy, 227 (3), 352367.
Popko, W., Heinonen, J., Hetmanczyk, S. & Vorpahl, F. 2012. State-of-the-art comparison of standards in terms of
dominant sea ice loads for offshore wind turbine support structures in the Baltic Sea. Proceedings of the Twentysecond International Offshore and Polar Engineering Conference. June 1722, Rhodes, Greece.
PrinciplePower. 2014. Available: http://www.principlepowerinc.com/products/windfloat.html [Accessed June 2014].
Qin, H. Y., Liu, M. L., Wang, Y., Zhao, J. Z. & Zeng, X. L. 2010. China: An Emerging Offshore Wind
Development Hotspot. Beijing, China: Chinese Wind Energy Association.
Quevedo, E., Cartn, M., Delory, E., Castro, A., Hernndez, J., Llnas, O., Bard, J., Lara, J., Jeffrey, H., Ingram, D.,
Papandroulakis, N., Anastasiadis, P. & Wesnigk, J. 2013. Multi-use offshore platform configurations in the
scope of the FP7 TROPOS project. Proceedings of the MTS/IEEE OCEANS Conference. June 1013, Bergen,
Norway.
Rademakers, L. W. M. M., Braam, H. & Obdam, T. S. 2011. Operation and maintenance of offshore wind energy
systems. In: Srensen, J. D. & Srensen, J. N. (eds.) Wind Energy SystemsOptimizing Design and Construction
for Safe and Reliable Operation. Cambridge, UK: Woodhead Publishing Limited, 546583.
Rashid, A. 2012. Status and potentials of tidal in-stream energy resources in the southern coasts of Iran: A case
study. Renewable and Sustainable Energy Reviews, 16 (9), 66686677.
Renzi, E., Abdolali, A., Bellotti, G. & Dias, F. 2014. Wave-power absorption from a finite array of oscillating wave
surge converters. Renewable Energy, 63, 5568.
Renzi, E. & Dias, F. 2012a. Resonant behaviour of an oscillating wave energy converter in a channel. Journal of
Fluid Mechanics, 701, 482510.
Renzi, E. & Dias, F. 2012b. Relations for a periodic array of flap-type wave energy converters. Applied Ocean
Research, 39, 3139.
Rhinefrank, K., Schacher, A., Prudell, J., Cruz, J., Stillinger, C., Naviaux, D., Brekken, T., Von Jouanne, A.,
Newborn, D., Yim, S. & Cox, D. 2013. Numerical analysis and scaled high resolution tank testing of a novel
wave energy converter. Journal of Offshore Mechanics and Arctic Engineering, 135 (4), 041901.
Roald, L., Jonkman, J. M., Robertson, A. N. & Chokani, N. 2013. The effect of second-order hydrodynamics on
floating offshore wind turbines. Energy Procedia, 35, 253264.
Robertson, A., Jonkman, J., Vorpahl, F., Yde, A., Qvist, J., Fryd, L., Buils, R., Nygaard, T. A., Popko, W., Chen, X. H.,
Azcona, J., Uzunoglu, E., Guedes Soares, C., Luan, C. Y., Huang, Y. T., Fu, P. C., Larsen, T., Nichols, J., Liu, L.,
Manolas, D., Heege, A., Vatne, S. R., Ormberg, H., Duarte, T., Godreau, C., Hansen, H. F., Nielsen, A. W., Riber, H.,
Le Cunff, C., Gurinel, M., Beyer, F., Yamaguchi, A., Jung, K. J., Shin, H., Alves, M., Shi, W. & Park, H. 2014.
Offshore Code Comparison Collaboration Continuation Within IEA Wind Task 30: Phase II Results Regarding a
Floating Semisubmersible Wind System. Proceedings of the ASME 2014 33rd International Conference on Ocean,
Offshore and Arctic Engineering. Paper No. 24040, June 813, San Francisco, California, USA.
Robertson, A. N., Jonkman, J. M., Goupee, A. J., Coulling, A. J., Prowell, I., Browning, J., Masciola, M. D. &
Molta, P. 2013. Summary of conclusions and recommendations drawn from the DeepCwind scaled floating
offshore wind system test campaign. Proceedings of the ASME 2013 32nd International Conference on Ocean,
Offshore and Arctic Engineering. Paper No. OMAE201310817, June 914, Nantes, France.
Roddier, D., Cermelli, C., Aubault, A. & Weinstein, A. 2010. WindFloat: A floating foundation for offshore wind
turbines. Journal of Renewable and Sustainable Energy, 2, 033104.
Roddier, D., Cermelli, C. & Weinstein, A. 2009. WindFloat - A floating foundation for offshore wind turbines, Part
1: Design basis and qualification process. Proceedings of the ASME 28th International Conference on Ocean,
Offshore and Arctic Engineering. Paper No. OMAE200979232, May 31June 5, Honolulu, Hawaii, USA.
Roddier, D., Peiffer, A., Aubault, A. & Weinstein, A. 2011. A generic 5 MW WindFloat for numerical tool
validation & comparison against a generic spar. Proceedings of the ASME 2011 30th International Conference
on Ocean, Offshore and Arctic Engineering. Paper No. OMAE201150278, June 1924, Rotterdam, the
Netherlands.
Rodriguez, R. Z., Alonso, P. G., Lpez, J. A., Martin, V. D., Dinoi, P., Simos, A. N. & Iglesias, A. S. 2014. Model
scale analysis of a TLP floating offshore wind turbine. Proceedings of the ASME 2014 33rd International
Rogne, . Y. 2014. Numerical and Experimental Investigation of a Hinged 5Body Wave Energy Converter. PhD
Thesis, Department of Marine Technology, Norwegian University of Science and Technology.
Ruer, J., Decrin, M. K., Tosello, A. & Colmard, C. 2009. New offshore wind turbines installation device Dubbed
Castoro Vento. Proceedings of the European Offshore Wind Conference. September 1416, Stockholm, Sweden.
Sakai, K., Kashiwagi, M. & Takaramoto, R. 2014. Wave-energy absorption by a rotating electric-power generator
set inside an asymmetric floating body. Journal of the Japan Society of Naval Architects and Ocean Engineers,
19, 7988.
Salo, O. & Syri, S. 2014. What economic support is needed for Arctic offshore wind power? Renewable and
Sustainable Energy Reviews, 31, 343352.
Salter, S. H., Taylor, J. R. M. & Caldwell, N. J. 2002. Power conversion mechanisms for wave energy. Proceedings of
the Institution of Mechanical Engineers, Part M: Journal of Engineering for the Maritime Environment, 216 (1), 1
27.
719
Salvacao, N. & Guedes Soares, C. 2014. An operational wind forecast system for the Portuguese pilot area of
Aguadoura. The 1st International Conference on Renewable Energies Offshore (RENEW). November 2426,
Lisbon, Portugal.
Sarkar, A. & Gudmestad, O. T. 2011. Installation of monopiles for offshore wind turbineby using end-caps and a
subsea holding structure. Proceedings of the ASME 2011 30th International Conference on Ocean, Offshore and
Arctic Engineering. Paper No. OMAE201149129, June 1924, Rotterdam, the Netherlands.
Sarkar, A. & Gudmestad, O. T. 2012. Study on a new methodology proposed to install a monopile. Proceedings of
the Twenty-second International Offshore and Polar Engineering Conference. June 1722, Rhodes, Greece.
Schwartz, M., Heimiller, D., Haymes, S. & Musial, W. 2010. Assessment of Offshore Wind Energy Resources for
the United States. Golden, Colorado, USA: National Renewable Energy Laboratory.
Scott, A. 2013. Technology and innovation challenges for UK Offshore Wind Energy. Offshore Wind
Operations/Science Meets Industry. September 10, Bergen, Norway.
Scruggs, J. T., Lattanzio, S. M., Taflanidis, A. A. & Cassidy, I. L. 2013. Optimal causal control of a wave energy
converter in a random sea. Applied Ocean Research, 42, 115.
Seigel, S. 2012. Atargis Energy Corporation [Online]. Available: http://www.atargis.com.
Serena, A., Molinas, M. & Cobo, I. 2012. Design of a direct drive wave energy conversion system for the Seaquest
concept. Energy Procedia, 20, 271280.
Sethuraman, L. & Venugopal, V. 2013. Hydrodynamic response of a stepped-spar floating wind turbine: Numerical
modelling and tank testing. Renewable Energy, 52, 160174.
Sheng, W. A., Alcorn, R. & Lewis, T. 2014. Physical modelling of wave energy converters. Ocean Engineering, 84,
2936.
Sheng, W. A., Alcorn, R. & Lewis, T. 2015a. On improving wave energy conversion, part I: Optimal and control
technologies. Renewable Energy, 75, 922934.
Sheng, W. A., Alcorn, R. & Lewis, T. 2015b. On improving wave energy conversion, part II: Development of
latching control technologies. Renewable Energy, 75, 935944.
Shin, H., Dam, P. T., Jung, K. J., Song, J., Rim, C. & Chung, T. 2013. Model test of new floating offshore wind
turbine platforms. International Journal of Naval Architecture and Ocean Engineering, 5, 199209.
Shirzadeh, R., Devriendt, C., Bidakhvidi, M. A. & Guillaume, P. 2013. Experimental and computational damping
estimation of an offshore wind turbine on a monopile foundation. Journal of Wind Engineering and Industrial
Aerodynamics, 120, 96106.
SI-Ocean. 2012. Ocean Energy: State of the Art. Strategic Initiative for Ocean Energy [Online]. Available: http://siocean.eu/en/upload/docs/WP3/Technology%20Status%20Report_FV.pdf [Accessed December 2014].
Sichani, M. T., Chen, J. B., Kramer, M. M. & Nielsen, S. R. K. 2014. Constrained optimal stochastic control of nonlinear wave energy point absorbers. Applied Ocean Research, 47, 255269.
Siegel, S. G., Fagley, C. & Nowlin, S. 2012. Experimental wave termination in a 2D wave tunnel using a cycloidal
wave energy converter. Applied Ocean Research, 38, 9299.
Sirnivas, S., Musial, W., Bailey, B. & Filippelli, M. 2014. Assessment of Offshore Wind System Design, Safety,
and Operation Standards. Technical Report, No. NREL/TP-500060573. Golden, Colorado, USA: National
Renewable Energy Laboratory (NREL).
Sjkvist, L., Krishna, R., Rahm, M., Castellucci, V., Hagnestl, A. & Leijon, M. 2014. On the optimization of point
absorber buoys. Journal of Marince Science and Engineering, 2, 477492.
Sjolte, J., Mclisky, S. C., Tedeschi, E. & Molinas, M. 2013. Exploring the potential for increased production from
the wave energy converter Lifesaver by reactive control. Energies, 6 (8), 37063733.
Song, H., Damiani, R., Robertson, A. & Jonkman, J. 2013. A new structural-dynamics module for offshore
multimember substructures within the wind turbine CAE tool FAST. Proceedings of the Twenty-third
International Offshore and Polar Engineering Conference. June 30July 5, Anchorage, Alaska, USA.
Srensen, J. D. 2012. Reliability-based calibration of fatigue safety factors for offshore wind turbines. International
Journal of Offshore and Polar Engineering, 22, 234241.
Spencer, P., Morrison, T. & Ausenco, C. 2014. Quantile regression as a tool for investigating local and global ice
pressures. Proceedings of the Offshore Technology Conference, Arctic Technology Conference. Paper No. OTC24550, May 57, Houston, Texas, USA.
Statoil. 2014. Available: http://www.statoil.com/en/TechnologyInnovation/NewEnergy/RenewablePowerProduction/
Offshore/Hywind/Pages/HywindPuttingWindPowerToTheTest.aspx?redirectShortUrl=http%3a%2f%2fwww.stat
oil.com%2fhywind [Accessed June 2014].
Stevens, C. L., Smith, M. J., Grant, B., Stewart, C. L. & Divett, T. 2012. Tidal energy resource complexity in a
large strait: The Karori Rip, Cook Strait. Continental Shelf Research, 33, 100109.
Strauss, D., Mirferendesk, H. & Tomlinson, R. 2007. Comparison of two wave models for Gold Coast, Australia.
Journal of Coastal Research, 50, 312316.
Suzuki, H., Shibata, H., Fujioka, H., Hirabayashi, S., Ishii, K. & Kikuchi, H. 2013. Development and verification of
analysis code UTWind for rotor-floater-mooring coupled response of a floating offshore wind turbine. Journal
of the Japan Society of Naval Architects and Ocean Engineers, 18, 8190.
SWAN. 2009. SWAN User Manual, SWAN Cycle III, version 40.72 [Online]. Available: www.swan.tudelft.nl
[Accessed December 2014].
Sweetman, B. & Wang, L. 2012. Floating offshore wind turbine dynamics: Large-angle motions in Euler-space.
Journal of Offshore Mechanics and Arctic Engineering, 134 (3), 031903.
720
Taguchi, E., Stammer, D. & Zahel, W. 2014. Inferring deep ocean tidal energy dissipation from the global highresolution data-assimilative HAMTIDE model. Journal of Geophysical Research: Oceans, 119 (7), 45734592.
Takao, M. & Setoguchi, T. 2012. Air turbines for wave energy conversion. International Journal of Rotating
Machinery, Article ID 717398.
Takeda, K., Sakata, K. & Takagi, K. 2013. Mooring simulation of a twin-type ocean current turbine. Journal of the
Japan Society of Naval Architects and Ocean Engineers, 18, 5561.
Tang, H. S., Kraatz, S., Qu, K., Chen, G. Q., Aboobaker, N. & Jiang, C. B. 2014. High-resolution survey of tidal
energy towards power generation and influence of sea-level-rise: A case study at coast of New Jersey, USA.
Taylor, R. S. & Richard, M. 2014. Development of a probabilistic ice load model based on empirical descriptions of
high pressure zone attributes. Proceedings of the ASME 2014 33rd International Conference on Ocean, Offshore
and Arctic Engineering. Paper No. OMAE201424353, June 813, San Francisco, California, USA.
Tedd, J. & Kofoed, J. P. 2009. Measurements of overtopping flow time series on the Wave Dragon wave energy
converter. Renewable Energy, 34, 711717.
Thiagarajan, K. P. & Dagher, H. J. 2014. A review of floating platform concepts for offshore wind energy
generation. Journal of Offshore Mechanics and Arctic Engineering, 136 (2), 020903.
Thies, P. R., Johanning, L., Harnois, V., Smith, H. C. M. & Parish, D. N. 2014. Mooring line fatigue damage
evaluation for floating marine energy converters: Field measurements and prediction. Renewable Energy, 63,
133144.
Thomas, G. A. & Graham, A. N. T. 2014. Design method for ISO 19906 arctic offshore structures. Proceedings of
the Offshore Technology Conference, Arctic Technology Conference. Paper No. OTC-24546, May 57, Houston,
Texas, USA.
Thomson, J., Polagye, B., Durgest, V. & Richmons, M. C. 2012. Measurements of turbulence at two tidal energy
sites in Puget Sound, WA. Oceanic Engineering, 37 (3), 363374.
Thns, S., Faber, M. H. & Rcker, W. 2012. Ultimate limit state model basis for assessment of offshore wind
energy converters. Journal of Offshore Mechanics and Arctic Engineering, 134 (3), 031904.
TOMAWAC. 2014. Available: http://actimar.free.fr/mambo/index.php?option=com_content&task=view&id=159&
Itemid=215&lang=en [Accessed December 2014].
Tweddle, J. F., Marengo, I., Gray, L., Kelly, C. & Shucksmith, R. 2014. Developing regional locational guidance
for wave and tidal energy in the Shetland Islands. Marine Policy, 50 (A), 5366.
Ulstein. 2012. F2F (Floating to Fixed)Offshore Wind Installation Concept [Online]. Available:
http://www.ulstein.com [Accessed November 2014].
Utsunomiya, T., Sato, I., Yoshida, S., Ookubo, H. & Ishida, S. 2013. Dynamic response analysis of a floating
offshore wind turbine during severe typhoon event. Proceedings of the ASME 2013 32nd International
Conference on Ocean, Offshore and Arctic Engineering. Paper No. OMAE201310618, June 914, Nantes,
France.
Vennell, R. 2012. The energetics of large tidal turbine arrays. Renewable Energy, 48, 210219.
Venugopal, V., Davey, T., Girard, F., Smith, H., Cavaleri, L., Bertotti, L. & Sclavo, M. 2011. EquiMar Project
Deliverable D2.4Wave Model Intercomparison.
Venugopal, V., Davey, T., Girard, F., Smith, H., Smith, G., Cavaleri, L., Bertotti, L. & J., L. 2010. EquiMar
ProjectDeliverable D2.3Application of Numerical Models.
Venugopal, V. & Nemalidinne, R. 2014. Marine energy resource assessment for Orkney and Pentland waters with a
coupled wave and tidal flow model. Proceedings of the ASME 2014 33rd International Conference on Ocean,
Vicente, P. C., Falco, A. F. O., Gato, L. M. C. & Justino, P. a. P. 2009. Hydrodynamics of triangular grid arrays of
floating point-absorber wave energy converters with inter-body and bottom slack-mooring connections.
Proceedings of the 8th European Wave and Tidal Energy Conference. September 710, Uppsala, Sweden.
Vicente, P. C., Falco, A. F. O. & Justino, P. a. P. 2011. Non-linear slack-mooring modelling of a floating two-body
wave energy converter. Proceedings of the 9th European Wave and Tidal Energy Conference. September 59,
Southampton, UK.
Viehman, H. A. & Zydlewski, G. B. 2014. Fish interactions with a commercial-scale tidal energy device in the
natural environment. Estuaries and Coasts, January 2014.
Villers, F. & Vign, P. 2010. Marine spatial planning and its application in the marine energy field in France.
Proceedings of the 3rd International Conference on Ocean Energy (ICOE). October 68, Bilbao, Spain.
Viselli, A. M., Goupee, A. J. & Dagher, H. J. 2014. Model test of a 1:8 scale floating wind turbine offshore in the
Gulf of Maine. Proceedings of the ASME 2014 33rd International Conference on Ocean, Offshore and Arctic
Engineering. Paper No. OMAE201423639, June 813, San Francisco, California, USA.
Vorpahl, F., Strobel, M., Jonkman, J. M., Larsen, T. J., Passon, P. & Nichols, J. 2014. Verification of aero-elastic
offshore wind turbine design codes under IEA Wind Task XXIII. Wind Energy, 17 (4), 519547.
Wan, L., Gao, Z. & Moan, T. 2014. Model test of the STC concept in survival modes. Proceedings of the ASME
2014 33rd International Conference on Ocean, Offshore and Arctic Engineering. Paper No. OMAE201423213,
June 813, San Francisco, California, USA.
Wang, K., Luan, C. Y., Moan, T. & Hansen, M. O. L. 2014. Comparative study of a FVAWT and a FHAWT with a
semi-submersible floater. Proceedings of the Twenty-fourth International Ocean and Polar Engineering
Conference. June 1520, Busan, Korea.
721
Wsj, K., Bermdez, J., Bjerks, M. & Sreide, T. 2013. A novel concept for self-installing offshore wind
turbines. Proceedings of the ASME 2013 32nd International Conference on Ocean, Offshore and Arctic
Engineering. Paper No. OMAE201311439, June 914, Nantes, France.
Weber, J. 2007. Representation of non-linear aero-thermodynamics effects during small scale physical modelling of
OWC WECs. Proceedings of the 7th European Wave and Tidal Energy Conference. September 1114, Porto,
Portugal.
Wehmeyer, C., Ferri, F., Skourup, J. & Frigaard, P. B. 2013. Experimental study of an offshore wind turbine TLP in
ULS Conditions. Proceedings of the Twenty-third International Offshore and Polar Engineering Conference.
June 30July 5, Anchorage, Alaska, USA.
Wilkinson, M. & Hendriks, B. 2011. Report on Wind Turbine Reliability Profiles, Deliverable D.1.3, ReliaWind
Project, Technical Report. Garrad Hassan.
Williamson, B., Scott, B., Waggit, J., Blondel, P., Armstrong, E., Hall, C. & Bell, P. 2014. Using the FLOWBEC
seabed frame to understand underwater interactions between diving seabirds, prey, hydrodynamics and tidal and
wave energy structures. Proceedings of the 2nd International Conference on Environmental Interactions of
Marine Renewable Energy Technologies (EIMR2014). April 28May 2, Stornoway, Isle of Lewis, Outer
Hebrides, Scotland.
Willis, M. R., Broudic, M., Haywood, C., Masters, I. & Thomas, S. 2013. Measuring underwater background noise
in high tidal flow environments. Renewable Energy, 49, 255258.
WindFlip. 2013. Available: http://www.windflip.com [Accessed April 2013].
Woebbeking, M. & Argyriadis, K. 2013. New guidelines for the certification of offshore wind turbines.
Proceedings of the Twenty-third International Offshore and Polar Engineering Conference. June 30July 5,
Anchorage, Alaska, USA.
Work, P. A., Hass, K. A., Defne, Z. & Gay, T. 2013. Tidal stream energy site assessment via three-dimensional
model and measurements. Applied Energy, 102, 510519.
Worthington, M. 2014. Acoustic Monitoring of Beluga Whale Interactions with Cook Inlet Tidal Energy Project,
ORPC Alaska. Technical Report, DOE Contract EE0002657.
Wright, B. A. 2014. Feasibility of Tidal and Ocean Current Energy in False Pass, Aleutian Islands, Alaska.
Technical Report, US DOE Office of Energy Efficiency and Renewable Energy (EERE), Weatherization and
Intergovernmental Program (EE-2K), DOE Contract number EE0005624.
Wu, M. K. 2014. Numerical analysis of docking operations between service vessels and offshore wind turbines.
Xiao, W. T., Liu, Y. M. & Yue, D. K. P. 2009. Ocean wave prediction using large-scale phase-resolved
computations. DoD High Performance Computing Modernization Program Users Group Conference. June 15
18, San Diego, California, USA.
Xing, Y., Karimirad, M. & Moan, T. 2013. Modelling and analysis of floating spar-type wind turbine drivetrain.
Wind Energy, 17 (4), 565587.
Xing, Y. & Moan, T. 2013. Multi-body modelling and analysis of a planet carrier in a wind turbine gearbox. Wind
Energy, 16 (7), 10671089.
Xu, N., Yue, Q., Qu, Y., Yuan, S. & Liu, X. 2014. Comparison and cause analysis of ice-induced structural
vibration of upward and downward cones. Proceedings of the Twenty-fourth International Offshore and Polar
Engineering Conference. June 1520, Busan, Korea.
Yanamoto, T. 2012. Development of the Kawasaki Heavy Industries tidal turbine. The 6th International Tidal
Energy Summit. November 2829, London, UK.
Yang, Z., Wang, T. & E., C. A. 2013. Modelling tidal stream energy extraction and its effects on transport processes
in a tidal channel and bay system using a three-dimensional coastal ocean model. Renewable Energy, 50, 605
613.
Yeter, B., Garbatov, Y. & Guedes Soares, C. 2014a. Fatigue damage analysis of a fixed offshore wind turbine
supporting structure. In: Guedes Soares, C. & Pea, F. (eds.) Developments in Maritime Transportation and
Exploitation of Sea Resources. London, UK: Taylor & Francis Group.
Yeter, B., Garbatov, Y. & Guedes Soares, C. 2014b. Spectral fatigue assessment of an offshore wind turbine
structure under wave and wind loading. In: Guedes Soares, C. & Pea, F. (eds.) Developments in Maritime
Transportation and Exploitation of Sea Resources. London, UK: Taylor & Francis Group.
Yeter, B., Garbatov, Y. & Guedes Soares, C. 2015a. Fatigue crack growth analysis of a plate accounting for
retardation effect. In: Guedes Soares, C. & Santos, T. A. (eds.) Maritime Technology and Engineering. London,
UK: Taylor & Francis Group.
Yeter, B., Garbatov, Y. & Guedes Soares, C. 2015b. Fatigue reliability assessment of an offshore supporting
structure. In: Guedes Soares, C. & Santos, T. A. (eds.) Maritime Technology and Engineering. London, UK:
Yu, B., Karr, D. G. & Sirnivas, S. 2014. Ice nonsimultaneous failure, bending and floe impact modelling for
simulating wind turbine dynamics using FAST. Proceedings of the ASME 2014 33rd International Conference
on Ocean, Offshore and Arctic Engineering. Paper No. OMAE201424320, June 813, San Francisco,
California, USA.
Yu, B., Karr, D. G., Song, H. & Sirnivas, S. 2013. A surface ice module for wind turbine dynamic response
simulation using FAST. Proceedings of the ASME 2013 32nd International Conference on Ocean, Offshore and
Arctic Engineering. Paper No. OMAE201311179, June 914, Nantes, France.
722
Yue, D. K. P. & Solodonz, P. J. 2008. Nonlinear wave environments for ship motion analysis. Proceedings of the
27th Symposium on Naval Hydrodynamics. October 510, Seoul, Korea.
Zanuttigh, B., Angelelli, E. & Kofoed, J. P. 2013. Effects of mooring systems on the performance of a wave
activated body energy converter. Renewable Energy, 57, 422431.
Zurkinden, A. S., Ferri, F., Beatty, S., Kofoed, J. P. & Kramer, M. M. 2014. Non-linear numerical modelling and
experimental testing of a point absorber wave energy converter. Ocean Engineering, 78, 1121.
Zvyagin, P. & Sazonov, K. 2014. Analysis and probabilistic modelling of the stationary ice loads stochastic process
with lognormal distribution. Proceedings of the ASME 2014 33rd International Conference on Ocean, Offshore
and Arctic Engineering. Paper No. OMAE201424713, June 813, San Francisco, California, USA.

7 10 SEPTEMBER 2015
CASCAIS, PORTUGAL
VOLUME 2
COMMITTEE V.5
NAVAL VESSEL DESIGN

COMMITTEE MANDATE
Concern for structural design methods for naval vessels including uncertainties in modeling techniques.
Particular attention shall be given to those aspects that characterise naval vessel design such as blast
loading, vulnerability analysis, specialised naval structures and others, as appropriate.
COMMITTEE MEMBERS
Chairman:
R.S. Dow, UK
J.Q.H. Broekhuijsen, Netherlands
S. Cannon, Australia
A. Fredriksen, Norway
J. Liu, China
N. Pegg, Canada
D. Truelock, USA
J.M. Underwood, UK
F. Viejo, Spain
A. Yasuda, Japan
KEYWORDS:
Naval ships, classification rules, design criteria, progressive collapse, lightweight materials, military load
effects, shock, blast, underwater explosions (UNDEX), whipping, naval masts, high speed naval vessels,
naval specific structural design, naval service life.
724
ISSC committee V.5: NAVAL VESSEL DESIGN
CONTENTS
1.
INTRODUCTION .................................................................................................................... 726
2.
NAVAL CLASS RULE DEVELOPMENT/PROGRESS ...................................................... 726

2.1 Introduction .................................................................................................................... 726
2.2 Military structural requirements .................................................................................... 727
2.3 Military operational safety loads ................................................................................... 728
2.4 Military performance loads ........................................................................................... 729
2.5 Concluding remarks ....................................................................................................... 729
3.
MILITARY LOADS ................................................................................................................ 730

3.1 Underwater weapon effects ........................................................................................... 730
3.1.1 Primary shock wave ......................................................................................... 730
3.1.2 Shock wave reflections and cavitation ............................................................. 730
3.1.3 Bubble dynamics and jetting ............................................................................ 731
3.1.4 Numerical modelling ........................................................................................ 731
3.2 Above water weapons effects ........................................................................................ 731
3.2.1 External blast .................................................................................................... 732
3.2.2 Internal blast...................................................................................................... 732
3.2.3 Bullets and fragments ........................................................................................... 733
3.3 Maritime improvised explosive devices........................................................................ 733
3.4 Concluding remarks ....................................................................................................... 733
4.
NAVAL SERVICE LIFE MANAGEMENT .......................................................................... 733

4.1 Introduction .................................................................................................................... 733
4.2 Ship service life in context ............................................................................................ 734
4.2.1 Australian LPA class ........................................................................................ 734
4.2.2 Australian Adelaide Class FFG-07 .................................................................. 734
4.2.3 ANZAC Class ................................................................................................... 735
4.3 Determining the remaining life of a warship ................................................................ 735
4.4 Naval structural monitoring programs .......................................................................... 737
4.5 Consequence of increasing displacement...................................................................... 738
4.6 Options for enhancing fatigue life of warships ............................................................. 738
5.
NAVAL SPECIFIC STRUCTURE DESIGN ......................................................................... 739

5.1 Structural uniqueness of naval ships ............................................................................. 739
5.2 Naval integrated permanent structures .......................................................................... 739
5.2.1 Flight decks (Vertical) ...................................................................................... 739
5.2.2 Stern ramps (Launch and Recovery Systems) ................................................ 740
5.2.3 Blast resistant structures .................................................................................. 741
5.3 Naval modular flexible structures ................................................................................. 742
5.3.1 Mission bays ..................................................................................................... 742
6.
NAVAL MAST DESIGN ........................................................................................................ 744

6.1 Introduction .................................................................................................................... 744
6.2 Types of naval masts...................................................................................................... 745
6.3 Materials (composite vs. steel vs. aluminum) ............................................................... 746
6.4 Loads .............................................................................................................................. 747
6.4.1 Weight of equipment ........................................................................................ 747
6.4.2 Environmental loadings (includes wind and seaway loads)............................ 747
6.4.3 Thermal ............................................................................................................. 747
6.4.4 Shock and Blast ................................................................................................ 747
6.5
6.6
6.7
6.8
6.9
725
6.4.5 Load combinations ........................................................................................... 747

Vibration and resonance ................................................................................................ 748
Structural analysis and design ....................................................................................... 748
Other considerations ...................................................................................................... 749
Classification society rules for mast design .................................................................. 749
Conclusions .................................................................................................................... 749
7.
PROGRESSIVE COLLAPSE ANALYSIS AND RESIDUAL

STRENGTH ASSESSMENT ................................................................................................. 750
7.1 Introduction .................................................................................................................... 750
7.2 Progressive collapse method overview ......................................................................... 750
7.3 Development of the progressive collapse method ........................................................ 751
7.4 Residual strength assessment by progressive collapse method .................................... 751
7.5 Use of FEA for progressive collapse assessment ......................................................... 752
7.6 Progressive collapse analysis within classification society rules ................................. 752
7.7 Discussion and Conclusions .......................................................................................... 753
8.
HIGH SPEED NAVAL CRAFT.............................................................................................. 754

8.1 Naval Applications ........................................................................................................ 754
8.2 Defining a high speed craft ............................................................................................ 755
8.2.1 Principles........................................................................................................... 755
8.2.2 Hull Form.......................................................................................................... 756
8.3 Defining operational limitations .................................................................................... 757
8.3.1 Operational profile ............................................................................................ 757
8.3.2 Operational envelope ........................................................................................ 757
8.4 Accelerations effects ...................................................................................................... 758
8.4.1 Slamming .......................................................................................................... 758
8.4.2 Human factors................................................................................................... 758
8.4.3 Fatigue............................................................................................................... 759
8.5
Material technologies.................................................................................................... 759
8.5.1 Steel ................................................................................................................... 759
8.5.2 Aluminium ........................................................................................................ 760
8.5.3 Fibrereinforced plastics (FRP) ......................................................................... 760
8.6 Unmanned naval high speed craft ................................................................................. 760
8.7 Classification Society Rules .......................................................................................... 760
8.8 Conclusion ..................................................................................................................... 761
9.
BENCHMARK STUDIES ....................................................................................................... 761

9.1
Whipping Response of Ship ......................................................................................... 761
9.1.1 Introduction...................................................................................................... 761
9.1.2 UNDEX Bubble Phenomena ........................................................................... 762
9.1.3 Experimental Investigations ............................................................................. 763
10. DISCUSSIONS AND CONCLUSIONS ................................................................................. 764

REFERENCES ................................................................................................................................. 766
726
INTRODUCTION
Previous ISSC Committee V.5 reports have addressed various aspects of Naval Vessel Research and Design these topics have included:
Design for Military load Effects

Vulnerability assessment
Class society rules for naval vessel design
Use of alternative matierials
Ultimate Strength
In this report the committee has concentrated on both the structural difference and role differences that
make naval vessels uniquely different form all commercial vessels.
There are fundamentally unique elements that differentiate the structural design of naval warship from a
commercial ship. First considerations that come to mind are undoubtedly the fact naval warships are characterized by large integrated guns and armor reinforced structure to engage in hostile encounters. There are
less evident unique structures that inhabit a naval warship design than her fighting attributes. To set the context and key differentiators of naval warship operations to a commercial vessel, a recurring quote from the
Naval Ship Code, ANEP 77, states:The role of Naval ships is such that the safety of the Naval ship and embarked personnel may be secondary to the safety of those under the protection of the Naval ship.
The above statement is directly contrary to the commercial ship operation mentality that holds personnel safety as paramount to all other safety. Although naval warships are operated to meet mission requirements, their design ideology is very close to that of a commercial ship, generally speaking. Much
like a container ship being built to carry containers from one port to another, the naval ship is built with a
specific purpose in mind. A common misconception is commercial ships are not designed specifically to a
mission and fit-for-purpose mindset. This methodology is only unique to naval ships. Commercial vessels
are designed in the same manner as a naval warship, with purpose and/or mission driven specifications,
and with the capability to survive a predetermined unexpected event. What drives the divergence of Naval
vs. commercial designs is the ability of a naval ship to adapt to the unexpected mission or event as well as
respond to this in a manner that may put ship and crew at risk. The container ships purpose is known, as
well as its mission, with very few unknowns to be accounted for; so, mission flexibility is normally not
needed. While the naval ships purpose is also known, her mission at any given instance is known to be
unknown, so mission flexibility is a must. This philosophical paradigm is highlighted only to drive home
the reasoning and distinct need for unique structures not common on commercial ships that enhance the
naval ships agility and support the unknown mission.
NAVAL CLASS RULE DEVELOPMENT/PROGRESS
2.1
Introduction
Classification rules are being used more and more as the technical standard for naval ships. The Class
Societies coverage of military design requirements for naval craft have been referred and discussed in
previous ISSC V5 reports:
ISSC (2006) report discussed the general approach to naval ship design philosophy
ISSC (2009) report discussed the class process for military design
ISSC (2012) report discussed the similarities and differences between commercial and naval design
Naval craft structural design has a lot in common with that of commercial craft. They all operate in the
same sea environment, and are built on the same structural principles and calculation formulas.
Classification Rules are now well accepted as a technical standard for the general strength calculations.
The question is whether Classification Rules also cover the more specific military operational and
damage scenarios for naval ships. Rules from all the Class Societies in the Naval Ship Classification
Association (NSCA) were investigated. This includes the following Class Societies:
American Bureau of Shipping (ABS)

Lloyds Register (LR)
Bureau Veritas (BV)
DNV GL: (Legacy DNV Rules (DNV) and Legacy GL Rules (GL))
RegistroItalianoNavale (RINA)
Polish Register (PRS)
Turk Loydu (TL)
727
The main focus for this report is to give a status on whether the Class Societies cover the military
aspects of naval ship structures. The different levels of detail between class societies rules have not been
evaluated in this report.
2.2
Military structural requirements
When discussing military structural loads it is necessary to consider what category the loads belong to.
For the purpose of this discussion, the two main categories have been defined as:

Military performance scenarios
Military operational safety loads are typically firing of weapons, replenishment at sea, and extreme use
such as maximum speed in high sea states. The safety type loads defines the required safety for the ship
and crew under its own normal duties. Normally the safety type loads belong to the role of the Naval Flag
Authority and Classification.
Military performance scenarios are typically related to the resistance to various enemy weapons and residual strength after such damage. An example is the resistance to underwater shock loads, internal or external blast etc. The performance type requirements for a naval ship define generally the warfighting
capabilities. The Performance type loads normally belong to the role of the Owner (Owners specification), and are normally followed up by the Owners representative.
The two main categories of loads are listed in the Tables 1 and 2. The Table 1 covers operational loads
related to normal military operations of naval vessels. The table is based on review of the different class
Societies naval class rules and some feedback from the respective Class Societies. The list of rules used
for the table is given under references. The load scenarios listed here are applicable for normal peace-time
operations (training, patrols, firing of weapons, handling of ammunition etc.) and wartime operations. The
Table 2.2 covers structural damage scenarios from hostile military weapons. These damage scenarios are
only relevant for war-type operations and terrorist attacks. The table is based on a review of the respective
class societies naval rules and some feedback from the respective Class Societies.
Table 1. Military operational hull strength aspects covered by classification rules.
Special strength
assessment
Beaching loads (When
applicable)
Dry-docking loads
Deformation of hull
Ultimate strength
Extreme strength
Fatigue check
Fixed wing aircraft deck
Helicopter deck
Aircraft elevators
Aircraft catapults
Aircraft lashing
Vertical Replenishment
Gun blast
Gun recoil
Missile blast
Launch & recovery of
mission vessel
Replenishment at Sea
Masts
Sonar dome strength
Towing of sensors
Towing (being towed)
Towing (towing others)
DNV
BV
LR
ABS
(2014) (2014) (2011) (2014)
C/CN CN
GL
(2012)
2)
2)
C
C/CN
CN
C
C
C
C/CN
C
C/CN
CN
2)
2)
2)
C
C
C
C
C/CN
C/CN
C/CN
C/CN
C
C/CN
C/CN
C
C
C
C/CN
C
C
C
C
C
C
C
C
C
C
C 1)
2)
C
C
2)
2)
2)
CN
C
C(O)
C
C/CN
C/CN
C(O) 1) CN
C
C
C 1)
C
C
C
C
C/CN
C
C(NFA)
C
C(NFA)
C
C(NFA)
C(NFA)
C(NFA)
C
RINA PRS TL
(2011) (2014) (2013)
C
C
CN
C
C
C
CN
C
C
CN
C
C
C
CN
CN
C
C
C
C
C
C
C
C
C
C
CN
C
C
C
C
C
C
C
C
C
C
728
DNV
BV
LR
ABS
(2014) (2014) (2011) (2014)
C
C
C
C
C
C
CN
CN
CN
2)
C/CN C/CN C
C
C
C
C
C
GL
(2012)
CN
C
Ammunition storage
Magazine flooding
Hull monitoring
C
In-water survey
C
Gas tight bulkheads
C
Deck loads: Armoured
tracked vehicles
C
CN
C(NFA)
Extended Lifetime
CN
(beyond 20 -25 years)
Explanations:
C : Covered by classification rules
CN : Covered by a separate class notation
(O) : Loads specified by Owner
(NFA): Loads specified by Naval Flag Authority
1) : Only hull attachments
2) : Compliance verification based on Naval Administration standards
2.3
RINA PRS
(2011) (2014)
C
C
C
CN
CN
C
C
C
C
CN
TL
(2013)
C
C
C
C
During the work of comparing the Class Societies naval rules in Table 2.1, it became clear that for most
Class Societies, the general structural calculation methods and formulas are the same for both naval and
merchant ships. In most cases when Class Societies publish a complete set of Naval Rules, the general
structural parts are copied from the commercial Rule book.
Table 2. Military hull performance scenarios covered by classification rules.
RINA
BV DNV GL
LR
ABS
C/CN(O)
2)
CN (NFA)
Underwater shock of hull
2)
structure
C
2)
C
NFA
Shock details
2)
C(O)
2)
C
CN (NFA)
Shock mounts
2)
CN(O) 1)
2)
C(NFA) CN (NFA)
Air blast, internal
2)
CN(O) 1)
2)
C(NFA) CN (NFA)
Air blast, external
2)
CN(O) 1)
2)
CN (NFA)
Fragmentation
2)
CN(O) 1)
2)
C
CN (NFA)
Structural damage
C
CN(O) 1)
2)
CN(NFA)
Small arms fire
2)
CN(O) 1)
2)
CN (NFA)
Fragmentation protection
2)
CN(O) 1)
2)
CN
CN (NFA)
Armour protection
2)
CN 1)
2)
CN (NFA)
Whipping
C
CN(O) 1)
2)
C(NFA)
Blast bulkheads
2)
CN(O) 1)
CN
C(NFA) CN (NFA)
Residual strength, structural
2)
CN
damage
C/CN
C
C
C
C
Fire protection of structure
C
C/CN
C
Residual strength after fire
C
CN
Vulnerability analysis
2)
C(O)
2)
CN (NFA)
Shock mounts of equipment
2)
Explanations:
C : Covered by classification rules
CN : Covered by a separate class notation
(O) : Loads specified by Owner
(NFA): Loads specified by Naval Flag Authority
1) :Different methods of analysis and acceptance are allowed for within the Rules
2) : Compliance verification system based on Military, Owner, Naval Administration standard
PRS
C
TL
C
C
C
C
C
C
C
C
C
C
C
CN
CN
729
The different Class Societies cover the various naval shiptypes in different ways. Some Class Societies
link the ship types to generic functions (combatant, non-combatant). Others arrange the combination of
their class requirements under specific shiptypes, such as Frigate, Corvette etc. In some cases, the ship
type is only used to indicate the type of ship, without any accompanying requirements. For more complex
naval ships, such as Logistics Support Vessels, class notations (functions) from both merchant and naval
type ships may be combined. Most Class Societies accept that the naval class notations may be combined
with some merchant ship class notation when this is relevant.
The general impression from the table is that the list of military operational loads covered by class
rules is quite comprehensive. In addition to the items in the table, there are items that are covered by
general strength criteria without being explicitly mentioned in the class rules. Considering this, it can be
said that most Class Societies has a good coverage of the typical military operational loads.
2.4
Military performance loads
The Table 2.2 indicates the different class societies approach to the military performance loads. Some
Class Societies have extensive coverage of the military damage scenarios in the class rules, and others
have little or no coverage. The different class societies handle the military type loads in a number of
different ways. There does not seem to be a common view on how this should be handled. Table 2.
identifies examples of 5 different ways of handling the military loads:
Military loads and calculation method given by the Class Rules

Military loads given by the Owner, and calculation method given by the class rules
Military loads given by the Naval Flag Authority, and method given by the class rules
Military loads and methods verified against a military standard
When Naval Ship Code (NATO 2014) is used, the Naval Flag Authority may define the military
damage scenarios, the loads, and the acceptance criteria
One Class Society, Legacy DNV, handle the military damage scenarios as a verification service against
military technical standards instead of specifying these in the classification Rules. This is a different
approach than chosen by the other Class Societies. It is a principal question whether the military
performance loads should be in the class rules at all. Military performance loads as in Table 2.2 do not
fall into a strict interpretation of class rules covering safety issues. They would normally belong in the
performance specification of the ship. It is expected that these issues will develop over time as more
experience is gained in the relation between Navy, Naval Flag Authority and Class Societies. This may be
a subject for later ISSC reports.
2.5
Concluding remarks
From the study of different classification rules it can be concluded that:
Military operational loads are well covered by Classification Rules.

Military damage scenarios are either well covered by some Class Societies, and little or no coverage
by other Class Societies.
The responsibility for input loads for military damage scenarios follow different practice from one
Class Society to another.
Certification against Naval Ship Code may overrule the class requirements on military damage
scenarios
One Class Society handles the military damage scenarios as performance requirements that are
verified against military standards instead of Class Rules requirements.
There is still a way to go for Class Societies and Navies to have a common understanding of the
responsibilities for defining the military damage scenarios and how to handle these.
Many Class Societies have put a lot of investment, competence and effort in the coverage of military
loads and damage scenarios. Also some navies have put competence and investment in this development.
It is now up to the Navies to utilize this resource so that class societies can follow up with practical
experience and further development in this area.
730
MILITARY LOADS
Naval ships can be exposed to a vast variety of weapon effects from above and below water that can affect the
survivability of naval vessels, both surface and submarines. A general overview of many different threats and
their effect on the ship structure is described in Chapter 6 of ISSC 2006 committee V.5 for Naval Ship Design
(ISSC 2006). Considerable efforts are needed in the design stage in order to accurately assess the appropriate
loads and predict the related impact. In this chapter several methods and tools will be described to simulate the
loading and structural response ranging from analytic to full finite element including fluid structural interaction. The focus is put on calculation methods presented with an experimental validation. The information given
in this report is based on what is available in the public domain. It needs to be mentioned that the greater part
of information about weapons effects and military loads remains classified within navies.
3.1
Underwater weapon effects
The physical effects from underwater weapons such as torpedoes and mines are on the load side: primary
shock wave, surface reflection of shock wave, possible bottom reflection of shock wave, bulk cavitation,
hull cavitation, bubble dynamics and jetting.
The structure responds with elastic plastic deformation, low frequency vibration and shock wave
transmission into the structure. The calculation methods for the different load effects will be described on
the basis of recent publications.
3.1.1 Primary shock wave

The main characteristics of the primary undisturbed shock wave resulting form an underwater explosion are well
described due to a combination of large scale experiments and modeling since World War II. Useful summaries
are given by Swisdak(1978) and Cole (1948) where analytical formula for the pressure time history are given dependent on charge weight and standoff distance. More recently Fleck and Deshpande (2004) extended the one
dimensional fluid-structure interaction model described by Taylor (1963) for flat monolithic plates and plane
waves to clamped sandwich beams. The structural response of the sandwich beam is split into three sequential
steps: stage I is the one-dimensional fluid-structure interaction problem during the shock wave loading event, and
results in a uniform velocity of the outer face sheet; during stage II the core crushes and the velocities of the faces
and core become equalized by momentum sharing; stage III is the retardation phase over which the beam is
brought to rest by plastic bending and stretching. The third-stage analytical procedure is used to obtain the dynamic response of a clamped sandwich beam to an imposed impulse. Performance charts for a wide range of
sandwich core topologies are constructed for both shock loading and air blast, with the monolithic beam taken as
the reference case. These performance charts are used to determine the optimal geometry to maximize blast resistance for a given mass of sandwich beam. For the case of shock loading, an order of magnitude improvement in
blast resistance is achieved by employing sandwich construction. However, in air blast, a sandwich construction
gives only a moderate gain in blast resistance compared to a monolithic construction.
3.1.2 Shock wave reflections and cavitation

The primary shock wave will reflect off the free water surface, wetted structure and the elastic or rigid sea bottom. The reflection from the sea surface and wetted structure produces a tension wave that propagates downward reducing the water pressure and cutting off the increased pressure around the structure due to the primary
shock wave. Typically, the pressure near the surface is reduced to zero and an area of bulk cavitation is formed,
shown in Figure 1. Reflection off the bottom is a compression wave that adds additional load to the structure.
The bulk cavitation is formed due to the waters inability to support a negative pressure. The water vapour cavity that is created consists of two separate boundary regions an upper and a lower boundary
which are a function of weight and depth of the detonated charge. An analytical description of the bulk
cavitation region is given by Didoszak(2004).
Figure 1. Bulk cavitation zone resulting from an underwater explosion (2004).
731
3.1.3 Bubble dynamics and jetting

As already mentioned in an under water explosion (UNDEX) loading event several phenomena occur.
The primary shock wave is followed by multiple bubble expansions and collapses while the bubble migrates to the surface, causing a whipping response of the structure. At the surface the gas bubble vents in
the air. Another feature of bubble behaviour is its attraction to rigid surfaces. A rigid surface exerts a
weak repulsion when the bubble is expanding but during its contraction there is a strong attraction. The
overall effect is that the bubble will move towards a surface such as a ships hull and remain there if contact is established, usually causing heavy damage through bubble jetting.
Riley (2010) compares various analytical models for predicting the growth and collapse of an underwater
explosion gas bubble. The different models use various underlying assumptions such as the compressibility
of the surrounding fluid and gas bubble, energy losses and the coupling of radial and migration motion. It is
demonstrated that reduction in gas bubble radius through growth and collapse cycles is underestimated by
all of the models when compared to experimental results as reported by Swift and Decius (1948).

Apart from analytical descriptions for the separate above mentioned under water explosion effects a numerical approach is also possible of course. In the full numerical approach detonation and bubble behaviour are
modelled in the fluid mesh. Cavitation is included via the fluid response. For the finite element model an ArbritrarelyLagrangian Eulerian (ALE) approach is used where the structural response is calculated using the
Lagrangian solver and the water, air and solid or gaseous phase of the high explosive with the multi material
Eulerian solver. The water and air are modelled using the Grneisen equation of state and the explosive
charge by using the Jones Wilkins Lee (JWL) equation of state. Examples of full calculations using LSDYNA and MSC.DYTRAN are given by Boyd et al. (2000), Petrusa (2004) and Ding and Bujik (2005). The
modelling effort needed depends on the aim of the calculation. A numerical model aimed to simulate all the
UNDEX phenomena should in principle cover the volume of thousands cubic meters from the detonation
point to the structure. The fluid mesh sizes should be sufficiently fine (typically cubic centimetres) to avoid
unrealistic damping and flattening of the pressure time curve or the primary shock wave. Consequently a
large number of elements is required and it becomes very time consuming to simulate the transient response
of a ship structure. To overcome this difficulty a boundary element approach is often introduced which uses
the boundary surface around the structure to take into account the fluid-structure interaction effects. This
eliminates the need to model the complete fluid domain. The most used method is the USA (Underwater
Shock Analysis) code based on the DAA (Doubly Asymptotic Approximation) technique (Shin, 2004). This
code can be used in combination with commercial FE packages such as MSC.Dytran and LS-Dyna.
An intermediate approach also exists in which the loading is estimated using analytical formulations that are placed directly onto the structural model. The calculation effort is in that case no longer
than a normal structural analysis. The accuracy is however as good as the analytical formulations used.
A description of this Simplified Interaction Tool (SIT) is given by Trouwborstand Bosman (2006).
As the cost of conducting a Ship Shock Tests are high, the tremendous advances in simulation capabilities are driving the question: Can Shock Test Trials be substituted by simulations? by Mairet al. (1997).
Figure 2 with advantages for both tests and simulations is presented together with strong arguments in favour of a combination of both approaches. Physics based UNDEX vulnerability simulations should be
employed in the acquisition cycle for:
3.2
Early design assessments for surface ships and submarines

Pre-test predictions prior to the shock test
Extrapolation of the shock test to realistic threat conditions after validation of the test
Analysis of the differences in each ship of a class after the model is validated for the shock tested first of class
Above water weapons effects
Above water weapon effects from bullets, shells unguided and guided missiles result in the following
physical loading on the structure: air blast, heat and high velocity impact form fragments and bullets.
High explosive weapons may have sufficient velocity and casing strength to penetrate ship hulls and explode internally. They may also be fused to explode before contact or on contact.
732
Figure 2. Ship Shock Tests versus simulations (Mair et al. 1997).
3.2.1 External blast

For an external blast loading event an analytical description of the pressure-time curve for the positive
pressure phase is given by the Friedlander equation (Baker, 1973). The over overpressure and impulse re
lations are given by Kingeryand Bukmash (1984) and Kinney and Graham (1985). Detonations produce
an overpressure peak and afterwards the pressure decreases and drops below the reference pressure (generally atmospheric pressure). The influence of this negative phase depends on the scaled distance.
Z = d/W1/3 (m/kg1/3)
(1)
where W is the equivalent TNT mass (kg) of the explosive charge and d (m) is the distance to he centre of
the charge.
For scaled distances larger than 50 the influence of the negative phase cannot always be neglected. The
size of the positive impulse and the negative impulse is then nearly the same. An analytical description of
the pressure time relation for the negative phase is given by Krauthammer and Altenberg (2006).
3.2.2 Internal blast

For an internal detonation of a warhead the initial spherical blast wave is reflected several times at the
boundaries of the compartment and structural elements will be loaded by various blast waves. These blast
waves are followed by a Quasi Static Pressure (QSP) build-up which is formed by the heat effects and
additional gas effect. The maximum level of QSP and the total loading time are of course governed by
venting of the explosion gases into the atmosphere or adjacent compartments (ISSC 2006). There is no
single accepted method discussed in the open literature for simulating the time evolution to QSP. Methods presented in the open literature generally define a delay time and then dump a specific amount of energy into the explosive products over a defined time period as presented by Daily (2004).
For the numerical modelling of an internal or external blast event a similar ALE approach is mostly
used as for the underwater explosions effects. Examples of this approach are given byLarcher(2007) and
Zakrissonet al. (2010). Also here examples can be found of intermediate, more computationally efficient
approaches where empirical explosive blast loads are placed directly on the air domain or structure domain. A comparison between three different approaches is given by Slavik (2010).
733
3.2.3 Bullets and fragments

Bullets and fragments can penetrate bulkheads, decks and the ships hull but in general dont influence
the failure of a structural component. They are however a danger to personnel, system components,cabling, cooling, fire main etc. Reduction of the fragmentation effects on ship systems can be
achieved by designing particular geometric arrangements and using ballistic protective materials (ISSC
2006).
3.3
Maritime improvised explosive devices
Maritime Improvised Explosive Devices (MIEDs) such as small boats filled with explosives are likely to
be a threat in future combat scenarios. Such a scenario with attacks with explosives at the water line is described by Slavik (2010) and named FLOATEX (Figure 3).
Figure 3. Floatex attack with a floating explosive (Slavik 2010).
As the explosion takes place partly in air and in water the above mentioned calculation methods for UNDEX and Blast need to be combined. Van den Heuvelet al. (2013) gives a comparison between experimental tests, full ALE Finite Elemente calculations and a simplified approach where analytical determined pressure time histories of the shock and blast loads are put directly on the structural FE model. It is
concluded that the simplified approach correctly predicts the onset of hull failure but over predicts the
panels plastic deformations. The plastic deformations are correctly predicted correctly with the ALE approach for the first two shots.
3.4
Concluding remarks
For the main military load effects analytical and numerical methods are in place. For the presented methods however limited validation material is available especially on the full ship scale. Depending on the
time available for the analysis and the needed accuracy, analytical, numerical methods or a mix of both is
being used. Where in the latter case the load is determined analytically and put onto the finite element
model of the structure. As however treats, naval ship structures and there materials continue to develop
there will always be a need for experimental research in which validation of numerical methods should
play an essential role.
NAVAL SERVICE LIFE MANAGEMENT
4.1
Introduction
A naval vessel, whether it is a surface ship or a submarine, is designed to operate for a certain amount of
time in a safe and efficient manner. In order to achieve this requirement a platform is first designed
against a set of criteria that determines sufficient safety against failure for a number of structural failure
modes. These may include buckling against extreme loads and fatigue failure for through-life cyclic
loads. Furthermore an effective through-life management system for the structure is required to ensure
protection against progressive failure modes and that the consequences of major changes to the structural
734
loads is acceptable. It is the intention of this section to review the structural aspects of naval service life
management.
4.2
Ship service life in context
A merchant ship is designed for an effective life so that it can make a return of its investment for the ship
owners. The process will often involve predicting the balance between the initial purchase prices of the
ship, the operating costs including daily running and voyage costs and cargo handling costs with the predicted income from providing the service. However the situation is a little more complex for a naval ship
as many other factors come into play.
For a warship there are a number of different measures that can be used to describe the life of type
(Koenig et al., 2008). The different types of life described include:
Expected Service Life this is determined by force level planning numbers and is generated by senior
defence planners. Within a ship class, all ships are expected to have the service life and this is used to
determine through life costs.
Actual Service Life is the life of the ship taken from initial commissioning to decommissioning and
where appropriate excludes out-of-commission periods where the ship is laid up.
Operational Service Life this life accounts for the different rate of ship aging between operational
years and years out of commission. This age will therefore assume a lower aging rate whilst out of
commission.
Assessed age is a measure that has been agreed by interested parties that accounts for the material
condition of the ship. This allows operational flexibility but means that the material condition of the
ship needs to be tracked.
Design Service Life is the service life that the ship is designed to achieve. This life, from a
structural perspective is similar to the approaches used for most merchant ships.
Economic life as discussed above is used by merchant ship owners to determine their return on
investment. Economic factors will impact a warships life however the process for defining this is not
a transparent process.
In order to explain some of these lives it is considered worthwhile to outline the service lives of a few
Royal Australian Navy (RAN) ships.
4.2.1 Australian LPA class

The Royal Australian Navy (RAN) amphibious transport ships, HMASs Manoora and Kanimbla. These
ships started out life as the USS Saginaw and Fairfax County respectively. Under the US service they
were commissioned in 1971 and decommissioned in 1994. After being sold to the RAN they were converted from the original LST format to the RAN LPA format. Each ship underwent a major upgrade with
a helicopter hangar amidships and in order to provide enough space for the forward operating helicopter
or the carriage of two army LCM8 landing craft the forward end was modified with the removal of the
34m bow ramps and horns. A 70 tonne crane was also installed. This major refit took several years and
they re-entered service 1999. During this refit period some lives were extended and others put on hold.
The actual service life is this case deviates from the operational service life. In this modification the reporting displacement of the ships remained constant (8534 tons fully loaded).
See http://en.wikipedia.org/wiki/Kanimbla_class_landing_platform_amphibious(2014b)
4.2.2 Australian Adelaide Class FFG-07

The Oliver Hazard Perry Class frigates ordered by the RAN were classed as the Adelaide Class. HMAS
Sydney (FFG-03) was the third of the class to be built in the US. At launch (1980) HMAS Sydney was
136 metres long, 13.7 metres wide and has a draught of 4.5 metres. The length was increased to 138 metres in 1987 to permit the operation of the S-70B Seahawk. The displacement at launch was 3,678 tons;
by 2009, various refits and upgrades had increased this displacement to 4,200 tons. See
http://en.wikipedia.org/wiki/HMAS_Sydney_(FFG_03)(2014a). The impact of this increased displacement on sea keeping performance and structural loads (e.g. fatigue loads) needs to be assessed in terms of
design service life given that the ship is now entering her 31st year of service (2014). This ongoing trend
to increase the actual service life of warships is becoming more prevalent given the increasing costs of
new acquisitions.
735
4.2.3 ANZAC Class

The ANZAC class frigates owned and operated by the Australian and New Zealand Navies were based on
the German MEKO 200 concept. Although purchased through a common build program each nation has
undergone an upgrade program with different requirements. Naval Technology.com reported that the
New Zealand upgrade was Designed to extend the operational life of TenixDefense systems-built TeKaha and Te Mana frigates for another 15 years, the upgrade programme seeks to replace hardware and
software of the combat management systems of the ships and the self-defence missile system. This was
completed as a result of some major relation of equipment on the quarter deck and internal structural
changes to support new bulkheads and relocated equipment. Furthermore additional solid ballast has also
been added to the ship to address some pre-refit trim issues.
In a similar fashion the Australian ANZAC frigates have or will be undergoing an Anti-Ship Missile
Defence (ASMD) system upgrade. In this upgrade the addition of an enclosed mast for the new radar system introduces complexity in terms of evaluating structural life assessments.
These case studies, although related to Australian ships are common across all navies and present the
structural engineer with a number of challenges. These are:
What design service life for a warship should be used at the initial design stage?
How should the consequences of a mid-life upgrade be taken into account at the design stage?
How do you assess the Assessed age of a current warship?
How do you determine whether the service life of a warship can be extended (Life- of- TypeExtension LOTE) beyond the initial design life?
What extra management practices can a warship manager use to enhance the life of a warship?
What is the impact of using commercial classification societies to manage warship life?
The above factors are to be considered along with the fact that the average achieved service life of
warships is increasing. Collette (2011) presented in Table 3, shown below, which indicates the
average achieved service lives of US Navy vessels.
Table 3. Average Achieved Service Life (after Collette 2011).
Ship Type
Carriers
BB-61 Class
Amphibious Vessels
Cruisers
Destroyers
Frigates
4.3
Average Service Life Years

40.4
48.8
29.2
26.3
25.4
19.8
Determining the remaining life of a warship
The approaches available to the Navies for determining the assessed structural condition of an aged warship are similar to those used by commercial agencies for merchant ships. The initial requirement is to
carry out a detailed analysis of the warship structures based on the most current survey information and to
use this to determine the feasibility of the life of type or whether a life of type extension is possible. This
survey usually includes all critical and primary structures with an analysis of the problems experienced to
date (if any) and an assessment of likely problems reoccurring into the future.
The terms critical structure, in naval case, relates to any highly stressed structure within 0.2L amidships and primary structure is that which contributes significantly to the main longitudinal strength, shear
and torsional strength and watertight integrity and compartmentation of the vessel. The surveys will report on items such as the general condition of the structures, any deformations that have occurred, fractures and any breakdowns of the protective coatings. Once the survey has been complete it is then necessary to consider the factors that will influence the hull condition in the future. This will require an
understanding of the maintenance philosophy used, the area of operation of the ship, the manner of operation (e.g. whether any speed restrictions or sea-state restrictions will be imposed upon the vessel and the
type of shipboard maintenance used.
Several approaches to automate the determination of remaining life of a warship have been
investigated by researchers in the past. The US Navy and the American Bureau of shipping embarked on
a program which resulted in the Surface Ship Life Cycle Management (SSLCM) activity (Eccles et al.,
736
2011). The main tool behind the program is to build a finite element model for the as-built configuration
of the warship and apply projected operational loadings envisioned for the ship to identify the following:
Inherent structural margins

Inherent corrosion allowances
Critical inspection plans
Plate renewal criteria
Expected fatigue life for the ship
The latter required the use of the FEA model and navy provided data which summarised the structural
loading the vessel had experienced to date so that the consumed fatigue life to date could be estimated.
Once the FEA model had been updated using the results of the hull surveys the remaining fatigue life
could be estimated for the ship. The ship is then managed via check sheets and finally a global representation of the ship is developed to record the ship history. The two key processes are shown in Figure 4 and
5 (Eccles et al., 2011).
A second approach to this problem is described by Collette (2011). His hypothesis is based on the premise that in the future will see an increasing use modular payloads and therefore the low cost ship structure will be required to endure longer lifecycles than previously expected possible up to 100 years. He
proposes extending existing semi-empirical component-based structural design rules primarily based on
safety concerns to a system performance model for ship structures.
Figure 4. Fatigue Life Assessment As Built (after Eccles et al. 2011).
Figure 5. Fatigue Life Assessment Approach Remaining Fatigue Life (after Eccles et al., 2011).
4.4
737
Naval structural monitoring programs
In order to facilitate the determination of a warships loading history a number of navies have embarked
upon structural monitoring programs (Phelps &Morris 2013). Many of the systems are again, common
with commercial systems and therefore it is not the intention to cover all platform monitoring systems but
to focus on those systems that have been developed or primarily used for warships. The overall purpose
of the hull monitoring system usually to:
Provide real time guidance of loading information and trends and to warn of extreme loading events
so that the risk of structural overload may be avoided, and to
Obtain actual in-service information of the extreme and long term fatigue loads to which the ships are
exposed so that this can be used for improvements in both design and the through-life management of
the ship structure.
The first of these requirements is to provide operator guidance to the embarked crew and relies on
techniques that are available to predict future events given the changing statistical responses obtained
and to suggest effective countermeasures before the overload event occurs. The second requirement
is to provide the navies with the loading history of the ships so that remaining fatigue lives can be
determined. The basis here is that the hull girder fatigue calculations are based upon an assumed lifetime loading profile and cross sectional properties of the hull girder, typical calculated at several sections along the hull. The well-known Palmgrem-Miner linear damage hypothesis is used with S-N
(Stress Range No of Cycles) curves for welded structures to establish the fatigue life for each hull
section.
There are an increasing number of commercial options available to provide requirements listed above.
Furthermore the classification society requirements have led to a fairly standardized system consisting of
two long base strain gauges on the weather deck, an accelerometer or pressure transducer in the bow and
a data processor and graphical user interface on the bridge (Lloyd Register, 2004). Over the years there
has been considerable development in the use of short base strain gauges (150 mm to 375 mm) that are
considerable more compact that the long base strain gauges (2 m to 0.5 m long). These shorter gauges offer much more flexibility as to where they can be located on the ship. This means they are generally more
suited to warships as they have much more complex structures and have less clear deck areas than tankers
and bulk carriers for which the original systems were developed. The placement of these gauges is however sensitive to localized stress concentrations so care must be taken as to where the gauges are placed
and how the results are interpreted; tri-axial rosettes are often required to correctly identify stresses in
areas where stress concentrations occur.
Over the recent years a number of navy vessels have been instrumented with hull monitoring systems.
Probably one of the earliest systems was the successful application of the extreme-value strain recorder
by the Royal Navy and its subsequent development into the fatigue strain recorder by the Defence Evaluation and Research Agency (DERA) during the 1980s and 1990s. A full description of the system is described by Brown (Brown, 1999) which included the use of an optical grating strain transducer to avoid
the problem of electromagnetic interference associated with foil strain gauges and filtering and algorithms
for the separation of wave induced and slam induced strains.
The Royal Norwegian Navy have installed a ship hull health monitoring system (SHHM)on their Royal
Norwegian Navy (RNoN) Oksy/Alta-class mine counter measures (MCM) vessel the HNOMS OTRA
(Torkildsenet al., 2005), shown in Figure 4.3. This vessel is a fibre reinforced plastic (FRP) surface effect
ship. The development of the system started in 1995 and included both a base system and an extended
system. The base system uses a fibre optic system and includes a total of 36 strain gauges and 8 temperature fibre optic sensors located around the amidships section, the forepeak and the wet deck between the
side hull keels. The enhanced system included measurement of the environmental parameters and global
ship responses. This included ship motion data and the use of an X-band radar to determine wave height
and direction measurements.
Further developmental systems have been trialed by the UK onboard the RV Triton (Kiddy et al.,
2002) and by the Royal Australian Navy on an Armidale Class Patrol Boat (Gardiner et al., 2009), however the sole purpose of these systems is to determine the load history for the class of ships and therefore
is applicable to both merchant and naval vessels. In all of the systems being developed it is deemed necessary that a detailed finite element analysis of the ship structure is used to determine the optimum place
to locate the strain gauges.
738
Figure 6. Overview of environmental and operating sensors on HNOMS OTRA (from Torkildsenet al., 2005).
4.5
Consequence of increasing displacement
As stated previously a number of navy ships undergo a mid-life upgrade. Whilst the primary purpose of
the upgrade is to enhance and improve many of the systems onboard the upgrade may have a consequence
on the structure. Newer systems integrated into the ship may increase the payload for the ship and may or
may not require extra structure to support the new systems. The ANZAC ASMD upgrade for the Australian Navy involved the design of a new enclosed mast structure for the phased array radar system. This enclosed mast was constructed from aluminium alloys to ensure that the final displacement was close to the
original design displacement. However in the event that there is a small increase in the final displacement
of the upgraded vessel it is important that the consequences of this increase is known.
As the original vessel was designed to a specific limiting displacement using design rules that have migrated into the Germanischer Lloyd warship rules (now DNV-GL) it was important to understand the
consequences of this small increase in displacement from a fatigue perspective. During the requirements
phase the intended operational profile was set for the ship which included typical speed profiles, areas of
operation and maximum number of days at sea. Critical areas of the hull were designed to ensure that the
proposed fatigue design life was met for the original loads. The process for this calculation was based on
one of the commercial classification society rule methodology.
The minor increase in displacement, together with the improvement in trim following the upgrade had
an influence on the global response of the ship. The change in weight distribution influenced the still water bending moment and therefore influenced the wave bending moments. The net effect of this change
was an increase in the bending moment range experienced by the ship. Once this new range was determined the classification society required a recalculation of the fatigue lives at the critical locations for the
ship in the previous condition and for any new locations that arose as a consequence of the upgrade. Following this analysis a management philosophy had to be determined for the critical joints. In some cases
the fatigue live is enhanced due to a better distribution of the loads whereas others may be more highly
stressed and therefore resulting in a lower fatigue life. In the latter case this may either result in a shorter
remaining design fatigue life for the ship or places a requirement to manage the locations via a methodical
inspection and repair strategy to ensure the required life is maintained.
4.6
Options for enhancing fatigue life of warships
Warships are typically very structurally complex and fatigue cracking, which is very dependent on
the structural detail design, can occur at virtually any time in the life of the ship. The usual treatment,
used by both commercial ship owners and navies is to weld repair the fatigue cracks at the earliest
739
opportunity. In some cases it may be required that a local design modification is made to enhance the
performance. Novel repair methods such as the use of composite patching are being increasingly researched.
The RAN has used a carbon fibre composite patch overlay on the aluminium alloy superstructure
of the FFG frigate HMAS Sydney. The patch itself was designed to reinforce the superstructure so
that a recurrence of major fatigue cracking did not occur after initial repair. The patches have endured 20 years service now since they were first installed (Grabovac & Whittaker 2009). They report firstly that the composite patch is effective in restoring the strength and function of the damaged
structure. Secondly the durability and repairability of the composite patch itself is successful in the
long term.
More recently though there has been significant interest in the use of composite patch technology as a
repair method for warships with fatigue cracks. The advantage of this methodology of managing fatigue
cracks is that there is an avoidance of any hot work (welding) and the ability to seal cracks. Qinetiq in the
UK have demonstrated the technology to repair cracks in Type 21 frigates and Type 42 destroyers (Turton et al., 2005). During these trials they report that the patches can enhance the fatigue life of a cracked
plate or restore strength of a corroded or damaged substrate. They conclude that the wider use of composite patch repair to marine vessels is held back by a lack of repair schemes approved by the classification
societies.
Other technologies which are starting to gain momentum include the commercial SPS Overlay repair
method which involves placing a new top plate separated by an injected polyurethane resin to produce a
sandwich (original plate/elastomer core/new plate). This method however appears to be most suited to
larger flat areas such as decks.
NAVAL SPECIFIC STRUCTURE DESIGN
5.1
Structural uniqueness of naval ships
In this chapter novel, in addition to traditional, naval warship structures will be discussed. If the holistic
design must account for mission adaptability, almost all of the described naval unique structures that will be
covered are in direct support of mission survivability. Whether it is mission capability, combat readiness,
endurance, or adverse environmental conditions, these structural attributes are characteristics of the ships
mission driven profile. These features aid in the adaptability as well as flexibility of the naval ships highly
volatile nature. Two categories are explored; one type is permanent in nature and integrated into the ship
during construction, and the other type facilitates modularity into the design and enables structure to be
removed, replaced, or upgraded throughout the ships life. Both types are unique and ultimately part of the
specification and holistic design of many naval warships.
5.2
Naval integrated permanent structures
A Nappi and Collette (2013) paper describes structural design as consisting of two tasks; achieving a true
fitness-for-purpose design and meeting higher design goals of minimizing weight and production cost.
Motivations for the design of specialized permanent naval warship structures inherently follow the same
methodology as the ship itself. These specialized structures are true fit-for-purpose designs. Similar to a
ships mission or defined purpose, integrated specialized structures must have an overarching function in
mind. Once this functional goal has been established, specific objectives will need to be identified before the
naval architect will be able to arrive at a unique design solution. Convergence of the design can then be
achieved.
This chapter addresses consistently emerging permanent specialized structures integrated within a naval
warship. Discussed structures include flight decks, stern ramps, and blast resistant structures, all of which
are fully integrated into the holistic design of the ship. These are not snap-on solutions, but structural
systems in which Naval Platforms are built around. In order to achieve a precise converged design, the naval
architect must also make some known sacrifices to accommodate these functional structures.
5.2.1 Flight decks (Vertical)

Flight decksin many naval platforms are considered to be the critical component to the design. The flight
decks addressed here are those which serve vertical Naval flight operations only, as shown in Figure 7.
Flight decks allow resources and critical supplies to be transited to and from the ship. These structures also
allow a naval ship to carry out missions well outside its geographic operational range if participating in
reconnaissance or surveillance operations.
740
Figure 7. French Navy AS565MB landing on the helicopter deck of the La Fayette frigate. (naval-technology.com
2014a).
Space and weight are the sacrifices made for incorporation of a flight deck. Flight decks are clear of
obstructions and geometrically sized to accommodate predetermined aircraft. This poses a challenge to
positioning of towing/mooring stations as well as weapon locations and any other types of payload. Flight
deck structures require a large array of safety equipment and machinery integrated into the design. Vertical
replenishment/recovery at sea (VERTREP/RAS) equipment and winches must be incorporated below deck
to retrieve payloads as well as aircraft while underway. Flight operations also require structural
reinforcement to the deck itself in order to account for a variety of loads including takeoff, landing, stowage,
crash, and vertical retrievals. Flight decks make upper extreme fiber of the hull girder when located within
the 0.4L of the ship. Therefore, global hull girder bending effects are considerations that must be taken into
account as flight decks are most times part of a continuous or stepped strength deck. Firefighting systems
and stations, safety nets, JP5 systems and stations, deck lights, and the flight operations station are just some
of the other items that must be designed into the ship itself when a flight deck is present; ultimately, adding
to the overall weight. The integrated flight deck is a feature that is rarely found in commercial ships and
truly unique and critical to naval ships.
5.2.2
Stern ramps (Launch and Recovery Systems)
The stern and bow ramp concept has been used in naval landing craft since WWII; personnel, craft, and
vehicles could all be tactically launched in an expeditious manner from a slow moving or stationary craft.
However, stern ramps have evolved to allow for high speed recovery and deployment of craft without
sacrificing normal vessel operations. It is becoming more common that stern ramps are being found in larger
patrolling ships because of the stealth it provides the ship when performing launch and recover operations.
The stern ramp design is an inclined ramp built at the centerline of the stern portion of the ship and used
mainly for small boat and RHIB operations. RHIBS can run-up into a ships stern ramp or be launched aft
while the vessel is making higher speeds, as shown in Figure 8. Using this technique eliminates the need for
traditional rescue boat or tender davits and cranes that ultimately can require time, manpower, and a
sacrifice of stealth.
Samras (2011) wrote an article summarizing the critical design elements to achieving an effective stern
ramp design; they are:
The incline of the ramp and its surface coatings.

Entrance configuration, such as a funnel shape, and side clearances for the boat.
The depth below the waterline of the ramp entrance, or sill.
Overhead clearance, if support bars or other structures are above the ramp. (In some cases, the ramps
are built below a flight deck, which requires strict avoidance of flight-path interference.)
Stern door or gate design and pivoting, operating, and securing mechanisms.
741
He also stated that incorporating a stern ramp into a vessel design may increase weight, sacrifice space,
as well as pose some stability challenges because of the loss of buoyancy in the aft portion of the ship.
An example of this is shown in Figure 9 for small craft.
Figure 8. USCG FRC Sentinel Class stern ramp operations (naval-technology.com, 2014b).
Figure 9. Mark V Patrol Craft Stern Ramp (U.S. Navy photo by Mass Communication Specialist 3rd Class Michelle L.
Kapica).
Stern ramps are close to or below the vessels waterline which, in shallower draft ships, makes for some
creative solutions in laying out the aft machinery and steering compartment arrangements. The ramp itself
usually consists of heavier structural material and reinforcement scantlings in order to absorb the impact of
RHIB and small craft run-up during recovery operations. A securing winch or hoist is then tethered to the
craft for securing it up into the higher elevations of the ramp structure. Dry and loaded ratings of the RHIBs
and other small craft must be taken into consideration when the ramp and LHRS are in the design phase.
Seaway accelerations should also be determined for accounting for loads imposed on the ramp when craft
are secured in the ramp bay.
5.2.3
Blast resistant structures
Structures that are designed to withstand air blast and fragmentation from explosions are often integrated
throughout the designs of naval ships. These structures are more protective in nature and not necessarily
meant to absorb the force from a direct hit from a missile or bomb. These structures are often in various
forms including bulkheads, doors, windows, hatches, as well as individual mounted shields protecting
certain vital areas. The designer most often tries to achieve a sense of balanced survivability throughout the
742
ship. This concept protects all critical systems, subsystems, and resources needed for carrying out the
mission at a predefined threat level (Walton, 2003). Preventing mass casualties and progressive collapse,
addressed by Chapter 7 in this report, are also basic functionalities and motivations for blast resistance
structures.
Most often the overarching cost of incorporation of blast structures is weight and space considerations.
Blast structures have historically been constructed of steel for warships because of steels strength and
ductility as well as overall fire resisting characteristics.
5.3
Naval modular flexible structures
Modular design concepts in naval ship platforms have long been discussed in detail as a means of
maintaining military relevance within the confines of available budgets (Doerry, 2012). If a permanent naval
structure is said to follow a fit-for-purpose design methodology, modular naval designs must subscribe to
the idea of flexible-for-purpose designs.The driving factor of such designs is undoubtedly the exponential
growth rate of technology in correlation with the consistent change of types of threats.
Figure 10. Cruiser Modular Concept (Doerry, 2013).
Today, modular technology has lacked the adoption of many of the worlds leading navies, including the US
Navy, even though one would think it to be a vital component in todays naval ship designs. Doerry (2013)
presented a paper on the Modular Adaptable Ship (MAS) to industry which he described as stemming from
the need for ships to be flexible yet robust.Their systems must be robust yet flexible to accommodate
changing and uncertain requirements.Although the modular design concept is not unique to naval ships,
they are to certain types of modular structures, as illustrated in Figure 10.
In order to highlight their unique functions, benefits, and any overall sacrifices that must be made, the
naval modular structures discussed in this section will be: mission bays, hangars, weapon modules, and
apertures.
5.3.1 Mission bays

In many ways missions bays are shepherds to modularity for naval warships. Mission bay designs are
designed to be very conducive to mission modules, and have attributes that can accommodate quick swapouts of these modules.For this discussion flight hangars are also lumped into the same ideology as mission
bays since traditional hangars are now being designed as flexible spaces to support a ships mission more
than just flight operations.Although most existing mission bay structures are not modular, new designs are
emerging with this concept. The DDG 1000 platform has a composite hangar/mission bay that is fastened
743
onto the steel hull of the ship (Figure 11). Doerry (2012) explains in his paper that in order to capture proper
specifications for mission bays during design such as the X-Craft (Sea Fighter) and LCS ships, the
following issues should be addressed up front:
Size of the Mission Bay

Relative Value of Different Mission Bay Sizes
Types of Distributed Services Needed for Mission Modules
Mission Module Distributed Service Needs for Sizing of Ships Distribution Systems
Defining a General Mission Module Interface that is Agnostic to Ship Class
Figure 11. DDG 1000 Composite Mission Bay being integrated to the hull (Bartlett & Jones, 2013).
5.3.2 Weapon modules

Over the last 20 years, the US Navy has not had a strong need for modular weapons describe Doerry (2012)
in a MAS concept paper.The weapon module concept has been largely overshadowed by the vertical launch
system (VLS) and 5 inch guns flexibility in munitions. Until recent modularity demands by the new LCS
and DDG platforms, the existing classes of warships are largely unable to retrofit weapons in a modular way
(Figure 12). Advancements in weapon technology as well as changing demands in mission needs now drive
this concept.
Figure 12. LCS Platform Weapon Module (Doerry, 2012).
5.3.3
Advanced enclosed masts/sensor (Enclosed aperture stations)
Chapter 6 within this report specifically addresses integrated enclosed mast design as well as the associated
design factors.It is important to note advancements in integration of AEM/S, Advanced Enclosed
744
Mast/Sensor, systems into naval ship topside designs have been incorporated into new naval ship
platforms.Doerry (2012) described the DDG 1000 program, Zumwalt Class Destroyer, is one recent
example of this type of modular mast design. The AEM/S systems can be upgraded and modernized as
technology advances without the necessity of large refit efforts.For her modular mast design, the DDG 1000
platform utilized composite materials that have also emerged as a possible solution to topside AEM/S and
enclosed mast applications.Barlett and Jones (2013)in ASNE 2013 presented additional examples of these
designs which can be seen in the LPD 17 AEM/S, CVN 77 mast, and other foreign Navy designs.
5.4
Conclusion
Naval warship design is fundamentally similar yet ideologically different than commercial ship design.The
uncertainty of the mission along with nobility of the role forces naval designers to consider increased
probabilities.These probabilities, such as hostile threats in conjunction with adverse environmental
conditions, are also related to higher consequences.All have to be accounted for in conjunction with the
endurance readiness of the ship.The same considerations would not normally be a concern for a commercial
ship designer.Accommodating the structural mission needs of naval warships is done through two distinct
ways;permanent integrated structures and modular flexible structures. Today, traditional naval warship
design has embraced the methodology of permanent, functional integrated structural design features; navies
are seeing a necessity to move to more flexible naval mission structure with modular capabilities.With
advancement of technology and weaponry, experts say this type of flexible and adaptable behavior will be
critical to the ever evolving mission needs.With the emergence of the next generation US Navy fleet looking
to these methods, such as the LCS class variants, time will tell whether there is a need to institutionalize the
modular adaptable ship.
NAVAL MAST DESIGN
6.1
Introduction
The 2015 edition of ISSC Naval Ship Design decided to include chapters on specific naval structures that
have not been covered in previous volumes. Naval masts are a unique structure, having many forms and
having evolved considerably over recent years to meet changing demands. This chapter presents an unclassified overview of naval mast design.
The primary purpose of a naval mast is to extend the surveillance capability of a ship. The basic design
objective of the greater the height of the mast, the further the range over-the-horizon leads to conflict
with other naval ship design objectives, namely concern for stability by creating weight high above the
center of gravity, and increasing radar cross-section and visibility from over-the-horizon opponents. Topside weight has become of even greater concern as the number of sensors to be placed on a mast continues to increase in support of surveillance and combat systems.
The issue of topside weight has been addressed by the use of light-weight materials and by having
more than one mast to hold all of the required sensors (which reduces height but not weight). Radar crosssection has been addressed through the use of stealth designs and specialized radar absorbing materials.
Naval masts must also be designed to withstand all of the same environmental and weapons loads
as a naval ship. Being more exposed, and housing sensitive equipment, makes this a special challenge
(Figure 13).
Figure 13. Issues of Concern for Naval Mast Design (Savage & Kimber, 2010).
745
Finite element analysis has become the primary means of structural analysis with challenges remaining
in modelling complex composite materials and specialized connections to the ship structure.
6.2
Types of naval masts
There are four main types of masts which have been commonly used in naval vessels, stayed and unstayed pole masts (Figure 14), tripod, lattice (Figure 15), and enclosed (Figure 16).
While lattice is probably the most common type of mast used on naval vessels, the use of enclosed
mast structures (metal or composite) is an emerging technology that is being used on recent builds by
countries such as the USA, UK and the Netherlands, and is quickly becoming the norm for naval mast
construction.
Figure 14. Typical Pole Mast Structure (www.masttechnologies.com/military, Sept 2014).
Figure 15. Cage lattice mast, USS Idaho (Photograph from the Bureau of Ships Collection in the US National
Archives) and Traditional Lattice Mast HMCS Nipigon (DND Photo), websites October 2013.
Figure 16. Typical Enclosed Mast on left (www.thalesgroup.com) and enclosed mast as deckhouse (Hackett, 2011).
746
SELEX (2013), DAMEN (2013) and THALES (2013) point to manufacturers websites for recent advances in enclosed mast design while Hackett (2011) describes the overall process of accepting composite
materials into USN ships including demonstrator enclosed composite masts. Recent advances in mast and
sensor technology have resulted in enclosed integrated masts of lower height and larger footprint (or more
than one enclosure) which really begin to behave more like a deck superstructure (deckhouse) rather than
a traditional mast (Kane et al., 2010 and Savage & Kimber, 2010).
Enclosed masts have the advantages of:
Reduced radar cross-section

Protection of sensors from the environment
Easier maintenance
Possible improved survivability from weapons loads
Use of complex embedded sensors
Reduction in electromagnetic interference between sensors
Improvement in lines of sight (Reduction in number of masts)
The disadvantages of enclosed masts can be:
Heavier topside weight

Increased surface area for wind loads and effect on air wake
Cost
Survivability from Fire
Another type of mast that is specific to Naval vessels are those that are designed to allow replenishment at sea (RAS) of other vessels. RAS masts may allow replenishment of liquid or solid stores or both,
negating the requirement for frontline warships to leave combat areas to replenish in a non-hostile port
location many miles away. Replenishment is conducted at sea, underway with both offloading and receiving vessel alongside, but at sufficient distance to avoid dangerous hydrodynamic effects that could lead to
the two vessels being drawn together. As such, loading on connecting cables between vessels can vary in
the forward and aft directions and vertically as the vessels move relative to each other, and machinery
acts to maintain tension in connecting lines. Loading on the mast from cables and machinery can therefore be significant and in unfavourable directions.
Little specific guidance is available from Classification Societies regarding the design of RAS mast
structure or its integration into the vessels hull. In fact plan approval of the mast structure may be considered an owner requirement and outside the scope of hull plan approval unless specifically requested.
However, the integration of the mast into the hull will be considered under hull plan approval and design
of the mast structure to a suitable code is advisable. Lloyds Registers Rules and Regulations for the
Classification of Naval Ships (LR 2014) provides guidance on RAS mast loads and operational considerations, though directs the designer to the requirements for lifting appliances in a marine environment
(Lloyds Register, 2013) for determination of acceptance criteria. As most Classification Societies publish
rules for the design of offshore lifting appliances, design to such rules may be considered to be an acceptable way forward when designing RAS mast structural arrangements. Such design may be considered
most complete when undertaken in conjunction with known loads from the RAS equipment manufacturer
or supplier.
6.3
Materials (composite vs. steel vs. aluminum)
Previous ISSC Naval Ship Design reports have covered materials fairly extensively. Naval masts have
been constructed of steel, aluminum and composite materials, each with its advantages and drawbacks.
Steel probably remains the most common material, even for enclosed masts. This is primarily due to
its strength, cost, ease of construction, durability and EMC characteristics. The obvious drawback is
weight.
Aluminum masts have been used to reduce top-side weight but suffer from lower strength, lower resistance to fire and higher susceptibility to fatigue failure.
747
Composites are becoming more popular, particularly for their light-weight and signature (imbedded
sensor) properties. This has come after a long research and development phase (Hackett, 2011 and Kane
et al., 2010).
6.4
Loads
As mentioned, masts must be designed to withstand all of the same loads as the naval vessel, and in many
cases these loads are exacerbated by the exposure and sometimes lighter construction of the mast.
6.4.1 Weight of equipment

As with any structure, the mast must be able to hold its own weight under inertial loads. In the case of naval vessels the inertial loads include accelerations from ship motions in addition to gravity. The overall
weight and center of mass of the mast must also be considered in ship stability.
6.4.2 Environmental loadings (includes wind and seaway loads)

Wind poses significant loads on naval masts. Design values are typically 100-120 knots coming from all
directions with overpressure, drag and vortex induced vibration having to be considered. Wind loads take
on considerably different characteristics for lattice type masts versus enclosed masts which have a much
bigger surface area. The overall force from wind should also be considered for its effect on ship stability
due to a lateral overturning moment.
Ship motions (roll, pitch and heave) produce inertial loads on the mast with calculations being commensurate with design values for the overall ship motions (such as fully operational in sea state 5 and
surviving sea state 8). Resulting forces and moments on the mast must take into consideration the center
of pitch and roll for the ship.
Another important environmental load for naval masts is additional weight, and thus inertial loads
from ice accretion. In addition to potential structural damage to the mast itself, this produces a significant
danger to ship stability and may also render the sensors on the mast non-functional.
6.4.3
Thermal
Masts can be subject to extreme thermal loads from exhaust, weapons or fire. Masts are usually located
near exhaust uptakes and hence subject to exposure from exhaust gases. The thermal effects from this can
be increased when exhaust plume cooling suppression systems are used as it concentrates the heat closer
to the deck and hence the mast structure.
Fire is a primary concern for masts constructed of composite materials and for aluminum where sufficient insulation must be used to maintain structural integrity and hence mast function. This may be especially important where the communication systems to bring aid during a fire or weapons attack may be
all, or mostly all, on the mast structure.
6.4.4 Shock and Blast

Blast loads are similar in nature to wind loads in that the structure must be designed to a specific overpressure for the defined weapon as well as considering the thermal effects of the load.
Shock loads usually are a design driver for the mast, particularly where there are large suspended
weights (sensors) on the upper portions of the structure. Shock loads are analyzed as for any other
dynamic load on a structure, using quasi-static, shock spectra or base acceleration time history, in order of complexity and accuracy. Finite element analysis is now readily used for complex sensitive
structures such as naval masts and provides effective means for dynamic analysis, either through determining natural frequencies to be used in a specified design shock spectra or a full linear or nonlinear time history analysis. It is important to remember that shock loads can come from all lateral and
vertical directions. While dynamic analysis via FEA is readily available for large problems, uncertainty still exists for proper modelling techniques for joints and foundations and for complex composite materials.
6.4.5
Load combinations
The loads described above do not happen independently so that most naval mast design guidelines specify
worst case scenarios for load combinations. While it may be reasonable to assume that weapons loading
748
will not coincide with the worst environmental loads, the highest wind loads usually coincide with the
worst pitch and roll accelerations leading to a dominant design load case.
6.5
Vibration and resonance
Masts, being constructed of lightweight thin lattice members, or panels, often have overall or local member natural frequencies in the 1-5 Hz range. This often coincides with vertical and/or lateral hull vibration
modes and with shaft rpm and propeller blade rates. This is a difficult problem to deal with and sometimes it is not possible to design a mast that falls outside of all possible excitation frequencies. In this case
avoiding hull modes and propulsion excitations at common operating speeds becomes the design objective. While FEA can predict natural frequencies, numerical results may not be accurate enough to
represent the real hull and mast values due to uncertainties in modelling mass distribution, connections,
foundations, composite materials and damping characteristics. Full scale hull and mast vibration measurements may be necessary to accurately determine frequencies and take action to sufficiently separate
excitation and mast natural frequencies and avoid resonant conditions.
Another type of vibration load concern mentioned previously is that of vortex induced vibration
(VIV). This is primarily of concern for local lattice members. If not taken into consideration, local failure
from low-cycle high-stress fatigue can happen very quickly into a structures life.
6.6
Structural analysis and design
Mast structural design follows the same process as other structures; defining loads, safety factors and
failure criteria. The finite element method, both static and dynamic, has become the standard analysis method for complex structures such as masts. Loads have already been described in Part 4. Safety factors are
specified by design rules (Navy or Class Society). The failure criteria that must be considered for masts,
as for other structures, are:
Exceeding yield strength

Fracture
Fatigue
Buckling
Excessive displacement (may affect sensor performance)
Vibration (discussed above)
Buckling analysis presents some challenges, particularly for lattice masts where member endconnections and effective buckling lengths must be considered. Lattice members may also be subject to
a combination of axial and bending loads. Finite element discretization must also properly model the
likely buckling (and vibration) failure modes. Again these are not different than what must be considered in analyzing civil lattice structures, other than needing to consider dynamic loads from ship motions and weapons loads.
Lattice structures in the civil domain, particularly hydro transmission towers and similar lattice structures are often designed via structural optimization, usually to minimize the cost. Care has to be taken
when applying this approach to naval masts, as a certain amount of structural redundancy may be desired
to withstand weapons loading. Naval structures are also usually subject to future weight growth. Masts
are no exception to this as often more and improved sensors are added through-life. Applying structural
optimization may negate redundancy and capacity for future growth.
Proper methods of modelling structural connections and foundations remain a topic of research, often with experimental programs to develop behavior properties. Construction errors such as misalignment of connections and foundations are also significant sources of uncertainty leading to premature
failure, primarily by fatigue. In most cases, the mast is modelled and analyzed as a separate entity,
usually with greater discretization than the rest of the ship. This poses a challenge in correctly modelling the boundary conditions of the foundation of the mast with the ship. Typically the primary supporting legs of a mast extend down below the weather deck, by several more decks. A common modelling
practice is to provide a fixed boundary condition to the supporting legs one deck below the weather
deck and pinned boundary conditions at the weather deck, assuming that the mast is free standing
beyond the weather deck.
Further mast structural design requirements are added by the combat system. Most significant of these
requirements is the combat element to element relative distortion allowance, which aim is to limit the
749
amount of relative angular distortion between specific pairs of combat system. The required assessment
includes global and local effects of different loading conditions (between full load to minimum operative
conditions) including environmental conditions such as wind loads as well as sea and air temperature
(both external and internal air temperature gradient affecting the ship structure). Such assessment is usually solved by means of global Finite Element analyses as illustrated in Figure 17.
Figure 17. Example of Combat System Elements distortion due to temperature gradient and wind loading (Navantia
communication).
6.7
6.8
Other considerations
Stealth: While already mentioned as a design driver, stealth is a primary concern for naval mast
design. The overall shape needs to minimize radar cross-section via flat-sided facets. Special radar
absorbing materials are also used in designing composite materials or applied over metal structure.
Protection: Shock and blast loads have been mentioned, but further consideration of fragmentation
and small arms fire may also be necessary to protect sensitive equipment inside or on the mast.
Interference between structure and sensors: While not a structural concern, overall layout of the mast
in concert with other components of the superstructure needs to be considered to avoid reducing the
effectiveness of sensors.
Height restriction for navigation under bridges: As for civil ships, height restrictions for operating in
certain ports need to be considered.
Airwake effect on flight deck ops: The mast, in addition to other components on the superstructure
contributes to the air flow over the helicopter or aircraft flight deck. The air flow is termed air wake
and can adversely affect flight operations. This is a specialized field of study usually analyzed by
computational fluid dynamics, model tests in a wind tunnel and full scale trials.
Classification society rules for mast design
The Classification Societies cover Rules for Mast design in either specific naval ship rules or in their steel
vessel rules (see Table 4). Stayed and unstayed pole masts are covered in all Rules. LR (2014) and DNV
(2012) consider enclosed, integrated masts as superstructure for design and classification purposes. ABS
treats masts with specific allowable stresses from military loads along with other specific ship structures.
None of the rules seem to specifically cover lattice masts and the inherent issues of member buckling and
VIV excitation causing fatigue.
In general, the rules cover allowable stresses, loads from ship motions and wind, vibration and ice accretion. Specific military loads are left to the additional calculations regime.
6.9
Conclusions
This chapter has provided a brief overview of naval mast design. There are few, if any comprehensive
publications covering all aspects of mast design, and a designer are left on his own to design a mast as
any other structure. Unfortunately this includes Class naval rules where various aspects of mast design are
distributed throughout the rules, and some not included at all. Given the complexity and importance of a
naval mast structure, this is not a particularly desirable state, and this committee recommends that Class
societies, or naval design authorities, produce a comprehensive, collected rule set for naval mast design.
750
PROGRESSIVE COLLAPSE ANALYSIS AND RESIDUAL STRENGTH

ASSESSMENT
7.1
Introduction
The assessment of ship hull girder ultimate strength has been considered and reported on within
previous reports of ISSC Committee III.1 Ultimate Strength and specialist committees such as ISSC
2006 Committee V.1 Collision and Grounding, ISSC 2009 Committee V.1 Damage Assessment After
Accidental Events and ISSC 2012 Committee V.5 Naval Vessels and in particular ISSC 2000
Committee VI.2 Ultimate Hull Girder Strength.All note the benefits of non-linear finite element
analysis (FEA) for the assessment of the intact and residual strength and the increasing accessibility of
the method as computing and software technology increase.However, it is also recognised that within
the design, emergency response and salvage sectors of the marine industry, FEA is not the first method
of choice due factors such as design maturity, time to complete analysis and cost constraints.Therefore,
assessment of the ultimate bending strength in the intact and damaged conditions (the latter being
referred to as residual strength assessment) is predominantly assessed using specialist software
implementing the progressive collapse method. Example software that implements this method include,
Lloyds Register Rules and Naval Ship Rules software, Nauticus produced by DNV, Maestro produced
by Maestro Marine, Paramarine produced by QinetiQ GRC, HULLST developed at Hiroshima
University (Yao, 1991), etc.
Although the base level requirements of IACS (2014) is not an ultimate bending strength assessment
but a comparison of midship section modulus against wave bending moment; in the plan approval and
Classification of ship structures, most class societies accept or request the use of approved software
implementing progressive collapse analysis for undertaking an ultimate bending strength assessment.In
addition to this, Class societies such as Lloyds Register in their Rules and Regulations for the
Classification of Naval Ships (LR, 2014) offer the hull notation ESA for compliance with the Extreme
Strength Assessment criteria and RSA for compliance with the Residual Strength Assessment criteria,
both assessed by progressive collapse method using their NSR software.In such an assessment, the
predicted ultimate bending strength is compared to the hull bending moment data from hydrostatic
analysis to ensure a suitable safety margin exists between the two curves.
This chapter is focussed on the use of the progressive collapse method for the ultimate bending
strength assessment of intact and damaged ship structures, developments in the method since its inception
and recent comparative assessments through the use of non-linear FEA.
7.2
Progressive collapse method overview
The term progressive collapse is used when collapse of a structure commences with the failure of one or a
few components which then progresses over successive components (Starossek, 2009). The use of the
term is not limited to the marine sector.However, when used in relation to the assessment of the ultimate
bending strength of ship hull girders, specifically relates to the method originally developed by Smith and
Dow (Smith, 1977 andSmith & Dow, 1981) and often referred to as the Smith method.The method
implements the concept of structural idealisation, where by combining the known response of component
parts of a system under a defined loading condition, the response of the whole system can be
determined.In applying the concept to ship hull girders, the progressive collapse method draws on
assumptions originally presented by Caldwell (1965) that plane sections remain plane under global
bending and that failure of a transversely framed, longitudinally stiffened vessel will be by interframe
collapse.That is, the failure will be of the stiffened structure between frames before failure of larger
panels or grillages spanning multiple frame boundaries.Therefore, modelling is limited to independent
sections that longitudinally span between two adjacent frames.
For progressive collapse analysis of ship hull structure,discretisation of the interframe structure is
commonly made into an assembly of plate-stiffener combination units (Paik, 2006). Under global
bending, it is assumed that each plate-stiffener element is under pure compressive or tensile loading
depending on its location above or below the neutral axis of the interframe section.The strength
characteristics of the elements are calculated and stored in the form of load vs deflection or normalised
average stress vs average strain curvesfor both tensile and compressive loading, the latter often referred to
as load shortening curves.Fundamentally, analysis is undertaken by assuming an incremental bending
curvature is applied to the hull from which the strain in each element can be calculated, and the
contribution to the strength of the section calculated from the appropriate strength data curve. The neutral
751
axis of the section is identified by adjusting its location until the sum of forces above and below equate to
zero.Final sum product of the element forces with their distance from the neutral axis allows the
calculation of the bending moment induced.The process is iterated increasing curvature of the hull,
allowing the ultimate strength to be calculated.
7.3
Development of the progressive collapse method
Initial implementation of the Smith progressive collapse method by the original authors utilised the
computer program FABSTRAN (frame and beam static and transient response analysis (nonlinear)) (Dow
& Smith 1986) to determine the load shortening curve data for use in the assessment. This approach
ensures the strength data is bespoke for the section being assessed.In the implementation of the
progressive collapse method into commercially available software, a different approach is utilised to
derive the strength data for the section being assessed. Software packages such as Lloyds Register Rules
and Naval ship rules software, Paramarine and Maestro utilise a database of generic strength curves,
usually derived through the use of non-linear FEA, covering a range of plate and column slenderness
ratios from which the required strength data can be interpolated.This approach naturally gives the analyst
less control over their model as characteristics such as initial imperfections or residual stresses cant be
controlled, and the levels included in the derivation of the generic data is often not declared. As a result
the analyst can expect to calculate different results through the use of different applications.Differences
seen in assessments between different analysts utilising different progressive collapse tools has been
demonstrated in benchmark studies undertaken by the ISSC Committee III.3, Ultimate Strength (ISSC
2006&2012)and ISSC 2000 Committee VI.2, Ultimate Hull Girder Strength.
Since its inception, development of the progressive collapse method has been in relation to the strength
data captured within the load shortening curves.These include developments in the incorporation of initial
imperfections, distortions and local dent damage, as well as modelling of heat affected zones.Recognising
the potential limitation to the method in its current form should the hulls failure mode not be by
interframe collapse, Bensonet al. (2011, 2013a) developed a method based on the Smith progressive
collapse method to include assessment of overall grillage collapse.The method utilises orthotropic plate
approach to model and assess the overall or grillage collapse of the deck or keel, allowing the progressive
collapse of the hull to be initiated by overall collapse, and not just interframe collapse, depending on the
dominant mode.Implementation of the method requires greater modelling definition across multiple frame
spaces as assessment moves towards a compartment level assessment.The method developed by Benson
has subsequently been incorporated in the MAESTRO software package.
7.4
Residual strength assessment by progressive collapse method
TheSmith progressive collapse method is currently implemented for residual strength assessment after
damage in design, in emergency response by Classification Society emergency response services, and by
commercial salvage companies for vessel salvage.Current implementation of the progressive collapse
method for residual strength assessment assumes that the failure mode in the damaged condition remains
by interframe collapse. The damage to the hull is modelled by removing stiffened panel elements from the
interframe section in advance of performing the assessment.This method of damage modelling is not
changed whether caused by collision, grounding or weapon effects.Therefore, assessment of the residual
strength is by assumption that the remaining structure will fail in the same manner as an intact vessel with
a large opening.It is understood that to provide some conservatism to the residual strength assessment,
some software applications attempt to account for the potentially reduced strength of the elements
adjoining the damage aperture by modifying the strength curves to reduce their effectiveness.However,
this approach and its validity are not well documented.
Recognising that in a damage condition the failure mode of a vessel may no longer be by interframe
collapse, Underwood (2013) developed a method to allow the assessment of the hull girder residual
strength whereby the damage aperture and resulting failure mode in the damaged condition are accounted
for.The method replaces interframe stiffened panels within the damaged region with a grillage element
spanning multiple frame spaces, which includes the overall size and an approximation of the shape of the
damage.Assessment showed that accounting for changes in failure mode away from interframe collapse
can be critical when calculating the residual strength of the hull girder.In order to be able to account for
these influences within an application suitable for use in an emergency response scenario, Underwood
incorporated the use of the response surface method Kriging to capture the strength data of a generic set
of damaged panels, with the base strength data calculated through non-linear FEA.This has the advantage
of being able to capture the influence of the damage aperture on the strength of the structure, whilst
reducing the quantity of FEA required in generating a comprehensive set of base data.
752
A number of studies have implemented non-linear FEA to investigate the influence of a damage
aperture on failure modes.Benson et al. (2012) simulated damage in a box girder caused by collision of a
larger indenter with the top or side of the box girder.Analysis was undertaken through the use dynamic
FEA simulation.Following the collision the residual bending strength of the box girder was assessed
showing that the presence of the damage hole and the residual stresses caused by the damage event can be
significant in the assessment of the residual strength of the structure.
Downes et al. (2013) investigated the use of progressive collapse analysis for the assessment of a
lightweight naval vessel featuring side shell damage extending vertically from below the shear strake
down to the double bottom and longitudinally up to two frame spaces.Comparisons of non-linear FEA
with the progressive collapse method for this damage example showed the good correlation, with the FEA
also confirming the failure mode to be by interframe collapse. However, analysis of different extents of
transverse damage within a single frame bay presented by Benson et al. (2013b) showed a tendency for
the Smith Progressive Collapse method to over predict the ultimate bending strength when compared to
non-linear FEA.
7.5
Use of FEA for progressive collapse assessment
Although this chapter has focused on the use of the simplified Smith Progressive Collapse Method, it
would seem pertinent to discuss the use of FEA for progressive collapse assessment, both in its own right
and in relation to its use in the derivation of load shortening curve data for use within applications of the
Smith method.
FEA is becoming more widely used across the marine industry as software and computing capability
increase, allowing the potential for even quite complicated analysis to be within the reach of those with
only modest resources.Studies such as that presented by Storhaug (2009) demonstrate how the failure
mechanism seen in the vessel MSC NAPOLI, could be well recreated through the use of non-linear FEA
when detailed modelling of a known vessel can be undertaken. However, this contrasts with some of the
variations in ultimate strength prediction demonstrated in the benchmark study presented by the ISSC
2012 Committee III.1 on Ultimate Strength, where following analysis by a number of analysts using a
number of FE software packages, ultimate bending strength results varied from almost exact comparisons
to differences of up to 28%.
Results of non-linear collapse analysis of stiffened steel structures at stiffened-plate, grillage and ship
scale by FEA can be significantly influenced by the modelling details and the handling of initial
imperfection sizes and shape and the handling of residual stresses.Whilst guidance on accepted
imperfection levels and shapes is historically presented by Smith et al. (1991) and more recently by the
ISSC 2009 Committee III.1 on Ultimate Strength, and methods for modelling residual stresses have been
presented (Paik, 2006), analyst interpretation on how best to apply these within a model can lead to
variations in predicted collapse strength.This may go some way to explain some of the variation seen in
the benchmark study presented in the ISSC 2012 Committee III.1, but also seen in the ISSC 2006
Committee III.1 benchmark study comparing analysis by Smith progressive collapse method.As such,
utilising FEA to predict the ultimate bending strength of a whole ship or to verify simplistic methods
should be treated with extreme care, recognising the potential variations between model and real life, or
the assumptions contained within different mathematical models.Similarly the influence of such
variations should be considered when developing basis data for use by simplistic methods such as the
Smith progressive collapse method.
7.6
Progressive collapse analysis within classification society rules
In the general application of the Smith Progressive Collapse method in Classification Rules, there is no
difference in how the method is applied whether concerning a commercial or naval vessel.However, as
has been noted earlier, Lloyds Register offer additional hull notations for vessels that comply with the
extreme strength (ESA) or residual strength (RSA) requirements. Both assessments implement the Smith
progressive collapse method through the use of the Lloyds Register Naval Ship Rules software, with
calculated ultimate strength compared to maximum wave bending moments calculated from hydrostatic
or hydrodynamic analysis. Additional levels of RSA notation can also be achieved by meeting strength
requirements for greater extents of damage or accounting for whipping effects caused by underwater
shock in the bending moment assessment.
753
Table 4. Summary of Smith Progressive Collapse Method and Non-linear FEA Capabilities.
Structural Maturity
Smith
Progressive
Collapse
Method
Can be implemented as soon as
vessel configuration and initial
midship scantlings are generated.
Modelling Time
Rapid modelling based on limited

information.
Solution Time
In the order of minutes.
Initial Imperfections
Residual Stresses
Loading
Often a feature of the software with

no user control.
Often a feature of the software with
no user control.
Ultimate bending strength
assessment.
Failure Type Assessed
Interframe collapse under global

bending.
Also overall collapse under global
bending in Maestro only.
Ability to interrogate
failure
Software may be limited to

highlighting elements that fail first.
Local failure modes may not be
possible to identify.
Damage Assessment
Damage modelled by removing

structure from model and assessing
ultimate bending strength of
remaining structure. Requires
assumption that vessel fails by
interframe collapse.
Damage Assessment
Timescales for
Emergency Response
Rapid modelling and rapid solution

timescales allow method to be used
in an emergency response scenario.
7.7
Non-linear FEA
Vessel configuration and more

advanced scantlings to be known to
allow useful model to be built.
Requires a more detailed structural
arrangement to be known to be able
to construct a useful model.
Many hours to days depending on
structural definition and mesh
refinement levels.
User defined and controlled.

Interframe and overall collapse along

with shear, torsional failure and
potentially crack propagation
depending on the user applied loading
condition.
Detailed failure analysis can be
undertaken stepping through the load
steps and refining structural details in
critical areas to improve
understanding.
Damage can be modelled by
removing structure or simulating
damage dynamically. Failure type is
not limited and will be driven by the
structural configuration in relation to
the user define load type.
Time to build or edit models and run
a solution do not currently favour the
use of non-linear FEA in emergency
response.
Discussion and Conclusions
Table 4 provides an overview of the different capabilities offered by the Smith Progressive Collapse
method and non-linear FEA.It can be seen that whilst FEA may have advantages of offering more detailed
modelling and analysis, and what may be considered a more accurate assessment of the ultimate strength of
a ships hull girder, the Smith method still has a significant role to play in both early design stages and
emergency response, provided the analyst is confident in the assumptions made in its application. Caution
must be taken when implementing the method as factors such as modelling details e.g. initial imperfections,
residual stresses, etc., and variations between the true failure mode of the vessel (interframe, overall, torsion,
shear, etc) compared to the assessed interframe failure mode could lead to incorrect prediction of the
ultimate strength of the vessel.
Whilst FEA may be the obvious solution for performing more complex bespoke analysis, time constraints
along with factors such as design maturity, analyst experience, local model detail handing can all affect the
quality and accuracy of the results obtained and their value in answering to the question being asked.
In conclusion, whilst the capability and accessibility of FEA has increased significantly since the Smith
Progressive Collapse Method was originally conceived, the method continues to hold a valuable place in the
design and analysis of ship structures.
754
HIGH SPEED NAVAL CRAFT
8.1
Naval Applications
In previous a specialist committee during the 2000 & 2003 ISSC the structural design of high speed vessels
were addressed, but not in the context of naval craft. This discussion is meant to provide an overview of the
various structural design aspects of high speed naval craft and how these types of vessels are being utilized
in government and naval applications.
There has always been the need to use high speed craft in the commercial industry.Speed is necessary to
sustain the ever increasing pace of economy.Traditional commercial use of high speed craft range from
ferries to pleasure fishing boats to the offshore industrys need for crew boats. The procurement and use of
high speed craft within navies and government agencies have become more common within the last few
decades. This chapter discusses the application of high speed craft in naval scenarios as well as what makes
this type of craft unique to the naval community. A summary of high speed craft types and their naval
applications are shown in Table 5.
Table 5. High Speed Naval Craft Types.
NAVAL APPLICATIONS
Rigid Hull
Innerbottom
(RHIB) &
Inland Craft
Craft that operate on near shore or inland

waterways. RHIBs are used as a tender or
rescue boat on naval ships. Serve short
tactical missions for special force operations
and general utility due to size, speed, and
weight.
Coastal Patrol
and Insertion
Craft
Used for coastal security or light patrol

missions involving high speed maneuvering
and pursuits as well as supporting insertion
and extraction of military/security personnel
Offshore Patrol
and Support
Craft
Littoral
Combatant
Craft
EXAMPLE
Serve medium duration missions providing

support to offshore facilities and aid in
coastal protection.
Used mainly for strategic long term

deployments with a range of missions
involving littoral waters throughout the
world. Craft are able to undergo blue water
crossings if needed.
There is a growing necessity for countries to secure their borders and coastlines as opposed to worry with
blue water issues. Special operations units as well as national security and coast guard naval forces have
become a focus of governments as more international terrorist groups mobilize and become homeland
threats. Creating small high speed craft in order to support these groups now grows in demand. The oil and
gas industry has expanded its reach to all parts of the world allowing countries to harness offshore energy
resources, while a rise in pirate presence has alarmed the world with recent successes in commercial ship
755
takeovers.These and other factors now drive the need to have craft that can quickly provide security and
access to shallower waters. A littoral craft are those that primarily operate in littoral waters offshore within
and up to ocean depths of around 60 m (Figure 18).
Figure 18. Offshore Littoral Region (Wikipedia, 2014c).
This shift in attention can be seen with the US Navy developing the recent Littoral Combat Ship (LCS)
platform variants, as well as the Joint High Speed Vessel (JHSV) platform.This naval triad consists of
monohull, catamaran, and trimaran craft all over 90 m in length and all having semi-planing hull
forms.These attributes allow the vessels to operate efficiently in shallow littoral seas, but engage in open
ocean crossings when necessary.
8.2
Defining a high speed craft
A ship or craft achieves high speeds differently depending on a number of factors including various
combinations of naval architecture principles, hull form characteristics, as well as propulsion selection.The
common phrase High Speed Craft is used in discussions of many of todays naval platforms as well as
defined quite clearly in international standards and regulatory publications. Depending on the approach; all
are correct definitions of high speed craft, yet all are slightly different in criteria.
8.2.1
Principles
Hull speed or speed-to-length ratio is a principle which defines the required length of a ships hull in order to
achieve a desired speed.Naval platforms have traditionally fallen into several groupings of speed-to-length
ratio. Figure 19 shows the placement of commercial and naval ship classes grouped according to this ratio.
Figure 19. High Speed Naval Craft According to Speed-to-Length Ratio.
756
8.2.2
Hull Form
Characteristics of the hull form and its performance while under propulsion play a significant role in
providing the conditions necessary as described in naval architecture principles for a ship to achieve high
speeds.Savitsky (2003) describes the negative resistant pressures that occur at high speeds as directly
dependent on hull form performance and characteristics.These hydrodynamic pressures increase
substantially on the submerged portion of the hull as the ships speed increases. This phenomenon has
subsequently led to a range of high speed vessel types touting various types of hull forms and appendages to
mitigate this effect.High speed naval craft are typically those that have small displacements and are hard
chined in order to achieve a planing mode under high speed.Contrary to craft, high speed naval warships are
fine hulled with rounded bilges and maintain a displacement mode under high speeds. Examples of some of
the common high speed naval craft types and hull characteristics are shown below in Figure 20.
Figure 20. Example showing some high speed naval hullforms.
8.2.3 Standards and Regulations

The specialized design of high speed craft are pushing more and more navies to leverage and reference
International Standards as well as Regulations within their specifications to incorporate safety features as
well as mature design techniques. Modern naval technical communities have much less experience with this
type of vessel than traditional ships of the blue water fleet. This also compliments the movement of many
navies to a commercialized design and building methodology by incorporating partnerships with
Classification Societies and commercial shipyards. The following are current definitions of high speed craft
in international regulations:
IMO (2000) High Speed Craft Code and ANEP 77 (Naval Ship Code) (NATO 2014):
A high-speed craft is a craft capable of a maximum speed,
3.7
with a volume of displacement, , corresponding to the design waterline (m), excluding craft the hull of
which is supported completely clear above the water surface in non-displacement mode by aerodynamic
forces generated by ground effect. (SOLAS X/1.2, HSC Code 2000 para 1.4.30).
(1)
8.3
Defining operational limitations
8.3.1
Operational profile
757
During the design process, like any vessel, a high speed naval craft must be given the beginnings of an
operational profile in order to provide the window of design criteria that will apply to an existing design or
used for a new design.The operational profile is usually much like a master use case during design and
likewise a historical archive for the craft during operation.The crewing, payload, geographic operating
locations, necessary mission endurance, speed, and sea state design criteria would all be specified under this
profile.The profile is a living set of criteria that will not be able to provide the baseline for a vessel until the
design is complete.Iterations of the criteria will usually be vetted before a final set of parameters is
completed.This profile is iterative throughout the lifecycle of the craft which can drive the basics such as the
general arrangement all the way through the maintenance considerations and lifecycle upgrades for a high
speed naval craft.
8.3.2
Operational envelope
An operational envelope is a supporting element and outcome of a crafts operational profile.The envelope
is created as a limiting set of data from the crafts final design particulars and scantling sizes.The main
purpose of this information is to expose to the master of the vessel the operating limitations as it relates to
speed and sea conditions for certain headings.A common naval term is Safe Operating Envelope (SOE), but
this can also be interpreted Structural Operating Envelope (SOE) as well.This presents an issue as the
structural limitations of a design would not be appropriate safe operating conditions for crew and
equipment.The context and analysis used to derive the envelope could be either structural in nature but
could also be based on other factors including human and payload limitations.
60.00
Slamming Unlikely
Slamming Possible
Slamming Probable
50.00
Maximum Speed (kts)
40.00
30.00
20.00
10.00
Survival Condition - Best Heading

0.00
0
2
Sea State 0 -3
6
Sea State 4
10
Sea State 5
12
14
16
18
Sea State 6
20
Significant Wave Height (ft)
Figure 21. Sample Operational Envelope for Head/Bow Seas.
Therefore, each envelope is a combination of elements folded into a graphical format and should be
supported with instructions or narratives that clearly define the context in which they were constructed. This
would prevent a structural envelope being used under the assumption that this was the safe operation
limitations of the craft.An example of a commonly used operational envelope is shown in Figure 21.
758
8.4
Accelerations effects
One of the major design restrictions in high speed naval craft is the direct correlation of accelerations with
high speed operation.Accelerations, especially in excess, cause a number of problems on craft and more so
on high speed-to-length ratio smaller vessels.Commercial type high speed craft usually avoid higher sea
states when operating at high speeds and reduce their speed to take a best course approach when seas
become rough.Larger naval ships can accommodate higher sea state operations because of their sheer size
and hull form as seen in Table 6. However, smaller naval craft inherently operate in a capacity and that
dictates the craft be operated into high sea states in conjunction with high speeds.
Table 6. ABS HSNC Design Criteria.
Operational
Service Type Condition
Naval Craft
Coastal
Naval Craft
Survival
Condition
H1/3
H1/3
4m
(13ft.)
2.5m
(8.5ft.)
6m
Vm (20ft.)
4m
Vm (13ft.)
10
knots
10
knots
Riverine
0.5m
1.25m 10
Naval Craft (1.75ft.) Vm (4ft.) knots
Vm = max speed indesign condition
8.4.1
Slamming
Part of designing a high speed craft is accounting for hull slamming, but this can become more of an issue
when trying to predict the accelerations produced by a naval craft design.Empirical methods exist for
designers to estimate equivalent static slamming loads by using accelerations derived from methods such as
model testing, theoretical calculations, or instrumented vessels and then utilizing these in sizing scantlings.
For instance, the American Bureau of Shipping minimum design criteria regarding speed and sea states for a
high speed naval craft is shown in Table 6.
In the real world, slamming loads fail to behave the same on any two naval craft. Slamming events
can occur along the entire length of the crafts hull during planing, but will also be focused within the
forward lengths of the vessels bottom and sides during semi-planing.Slamming loads are driven by
accelerations which is comprised of two major variables, wave height and speed. For design, the crafts
scantling loads should be derived at full speed under the lightest loading condition in head seas as a
worse case condition.
Slam-induced whipping should also be evaluated within the hull girder as a global load case. Once a
converged design has been achieved, an operational envelope should be developed to articulate the designs
limitations graphically to the crafts operator.
8.4.2
Human factors
Excessive accelerations at sea can cause humans to suffer head, neck, and spinal injuries that could
permanently damage ones way of life.The shock of weighted accelerations of 2.5g over the course of 1
hour could cause serious permanent damage to a human body [ASTM F1166 2013]. Furthermore, the same
type of accelerations in the 6g range for any length of time would have an even greater traumatic effect.
The fact is that high speed naval craft experience these ranges of accelerations all the time. Many
planning naval craft may see high accelerations in the hull over short time periods during high speed
missions similar to Figure 22. Shock mitigating seats are common staples in small planning patrol craft in
order to alleviate some of these injury risks.Other techniques include shock absorbing hulls as well as
reducing speeds during rough seas. Drone vessels may be an emerging technique that could prevent the need
for human occupancy on small high speed naval craft, but this concept has limited traction so far.Whatever
the preventative measure, this human factor is almost always the limiting factor in craft operation and
should be reflected in the vessels operational envelope.
759
Figure 22. ASTM F1166 (2013) Human Acceleration Tolerance with High Speed Naval Craft Operational
Accelerations.
8.4.3
Fatigue
Accelerations directly amplify fatigue in high speed naval craft when they occur in large cycles and at high
amplitudes.Since large portions of small high speed naval craft are constructed of lightweight material such
as aluminum, fatigue cracking can be a serious issue. Fatigue critical joints should be monitored on a more
frequent basis when high speed craft have experienced heavy use in high seaways where slamming and hull
whipping are present.The preventative method would depend on the material used in the craft as well as the
location and range of the stress concentrations. Fatigue should also be considered in areas subject to
vibration due to propulsion machinery as well as payload securing fittings. These areas with the use of
lightweight thin material are many times the first to show signs of fatigue and overstress in new first of class
naval craft.
8.5
Material technologies
The ISSC 2012 V.5 Naval Vessels specialist committee paper previously addressed lightweight material
selection from a design and lifecycle perspective, so material technology will be briefly touched on.Along
with hull form and speed, weight shedding is another key component to rounding out and reducing
displacement footprint in a high speed naval design. The majority focus is on structural weight since
structural scantling weights makeup approximately one third of a crafts displacement. Material selection is
critical in producing efficient and lightweight scantlings. By using high performance materials it is entirely
possible to reduce a crafts structural weight by as much as 40% as compared to traditional steel
construction techniques (Furio, 2002). Fortunately, there are many state-of-the art material options for
achieving lightweight designs. The following are some technologies that have emerged allowing craft to do
more with less (weight.)
8.5.1
Steel
In lieu of traditional steel construction, proponents of steel can still benefit from some advances in
construction and analysis techniques for design. Decisions on steel placement can be dictated by running
advanced analyses on designs prior to construction to target overdesigned and lacking areas. The maturity of
welding advancements in automation as well as a progression in protective coating effectiveness will allow
steel weights to be further reduced for high speed designs in the near future.
760
8.5.2 Aluminium
The big opportunity for aluminum use is the 1/3 density of steel that can achieve a 30% weight savings right
up front with minimal optimization.To further the ability of aluminum weight savings and efficiency is to
utilize it in extruded sandwich planking configurations.The sandwich planking construction leverages thin
sheets of series 6000 aluminum in an extruded truss-like sandwich panel that can be used in deck and
bulkhead applications. This is a further variant to the US Navys Light-Weight Metallic Sandwich (LMS)
program progressed in the late 70s for topside weight savings.LCS variants have shown significant weight
savings by incorporating aluminum sandwich extrusions into their designs as described by Furio (2005.)
8.5.3
Fibrereinforced plastics (FRP)
Composite advantages where discussed in the 2012 ISSC Naval Vessel paper as having significant
maintenance and overall weight savings costs accumulated over the life of a naval vessel.A technical thesis
by Torrez (2007) touches on the uses of composites in the DDG 1000 and LCS platforms as critical
materials in reducing topside weight.The paper goes on to list a number of advantages within naval
applications as referenced by Furio (2002) in his naval assessment of lightweight composite design.
Despite the number of advantages and studies suggesting that composites are the most effective material
in high speed craft designs, navies have still not embraced this material technology as a holistic answer to
structural weight savings. This is partly due to the lack of naval institutionalized knowledge of the material
and design techniques in conjunction to the major disadvantage revolving around high combustibility and
toxicity which composite materials exhibit in a fire or explosion. An example of this horrific scenario played
out in 2012 as a brand new Indonesian Naval fast missile patrol vessel (FMPV), KRI Klewang 625 (a
composite trimaran), burned to the waterline off of the coast of Banyuwangi, East Java while undergoing
routine maintenance.
8.6
Unmanned naval high speed craft
One novel concept becoming more of a reality is the building of unmanned high speed naval surface vessels.
While many of these are currently only smaller drone unmanned surface vessels known as USVs, some
more recent developments have shown a naval interest is maturing these craft into larger, more functional
sizes.Navies have deployed these unmanned craft mainly to hunt and neutralize subsurface mines, but some
are now being utilized to detect more than just mines below the surface. SAIC (Science Applications
International Corporation) and DARPA have recently begun construction and testing on designs for an
ACTUV (Anti-Submarine Warfare Continuous Trail Unmanned Vessel) due for delivery in 2015 (Figure
23), which will be built as a full-scale vessel for operation for months at sea without maintenance or
refueling.This vessel will use computerized logic by leveraging pre-programmed algorithms within its
operation and monitoring systems in order to make decisions depending on a number of factors.She will
also be remotely controlled much like her smaller surface and subsurface predecessors.
Figure 23. SAIC ACTUV Concept (Drones of the Navy).
8.7
Classification Society Rules
The current rules for classification of high speed naval craft from the major societies can be found in Table
7 with the applicability limitations, if any.
761
Table 7. Classification society rules applicable to high speed naval craft.

Classification
Society
ABS
Rules
ABS Rules for Building and

Classing High-Speed Naval Craft
(2014)
DNV-GL
DNV Rules for Classification of

High Speed, Light Craft and Naval
Surface Craft (2014)
LR
LR Rules and Regulations for the

Classification of Naval Ships
(2014)
8.8
Applicability
The HSC Rules or HSNC Rules is applicable to highspeed craft or high-speed naval craft constructed of steel,
aluminum, or FRP and having L V not less than 2.36
(1.30) where L is as defined in 3-1-1/3 and V is as defined
in 3-2-2/1.1.2. Applicable craft type and length are as
follows:
ApplicableLength
CraftType
Mono-hull
< 130m (427ft)
Multi-hull
< 100m (328ft)
SurfaceEffectsShip
< 90m (295ft)
Hydrofoil
< 60m (197ft)
The rules for naval surface craft are intended to define
technical safety requirements for such craft of frigate/destroyer size or smaller. The rules are limited to marine safety and its interfaces with military aspects (fit for
purpose).
The Rules are applicable to naval ships designed and constructed for the purpose of carrying and operating naval
systems.
Conclusion
The use of high speed craft by navies worldwide will continue to grow in demand as countries race to secure
coastal and littoral waters. The speed capabilities, payload agility, and range that high speed craft offer
governmental agencies and the naval force will be driving factors for the continued acquisition of new
platforms.Today, the commercial shipbuilding industry is leveraged by naval authorities for their extensive
experience in design and construction of high speed craft such as ferries and pleasure craft.Classification
societies are used as third party validators for design and construction and also in order to supplement the
knowledge gap within naval technical authorities. As state-of-the art technology in lightweight materials
continues to mature and techniques for mitigating slamming consequences progress, naval technical
communities will continue to acquire knowledge and understanding in small high speed craft design. This
will hopefully propel a new era of research and development specific to this area.In conclusion, the
projection of high speed design techniques into the traditional processes and framework of naval standards
will ultimately redefine the next generation of naval fleet.
BENCHMARK STUDIES
9.1
Whipping Response of Ship
9.1.1
Introduction
Underwater explosion (UNDEX) induced hull girder whipping, as distinct from the early time shockwave
response, involves the excitation of the ships low frequency global vertical bending modes in response to
pulsations of the explosion gas bubble. If whipping loads become too great, the hull girder may fail
globally in bending - breaking the ships back. The ability to assess whipping response of a warship
early in the design phase is necessary to ensure the global ship structure is capable of withstanding
potential weapon threats. Simplified analytically based explosion bubble models (Riley, 2010) have been
formulated for predicting, under far-field assumptions, the growth and collapse of an UNDEX gas bubble
and the associated fluid pressure and velocity field impinging on the ship. It is rare to find experimental
whipping results in the open literature which may be used to verify the application of the available bubble
models for whipping problems.
A benchmark study completed by the International Ship and Offshore Structures Congress (ISSC)
naval vessel design committee aims to compare different numerical codes for UNDEX whipping analysis.
The Naval Design committee were given access to a unique set of experiments that involved a segmented
scale model of a warship subjected to various UNDEX events. This section of the report compares the
762
whipping data obtained from the experimental tests with numerical predictions obtained using a number
of different approaches.
9.1.2
UNDEX Bubble Phenomena
The physical phenomena associated with an UNDEX event is comprehensively described by Cole (1948).
Following passage of a detonation wave through the submerged explosive, thereby converting the solid
explosive to an approximately spherical bubble of gaseous detonation products, a high intensity,
spherically spreading shockwave is propagated into the surrounding water. Immediately following
emission of the shockwave, the pressure in the gas bubble, although considerably decreased from the
initial detonation pressure, is still much higher than the local ambient pressure (i.e. hydrostatic +
atmospheric pressure). The bubble therefore expands rapidly to reduce the internal gas pressure. Bubble
expansion and the associated outward flow of displaced water continues until the bubble reaches a point
of maximum radius and minimum pressure. Due to the inertia of the displaced water, the minimum
bubble pressure falls below the local ambient pressure. Thus after reaching a maximum radius, the bubble
begins to contract. The inward motion continues at an increasing rate until the compressibility of the gas
acts as a check against further contraction and causes a rapid reversal of the motion. The inertia of the
water, together with the elastic properties of the compressible gas, provide the conditions for an
oscillating system and the bubble undergoes repeated cycles of expansion and contraction. Of the total
chemical energy released from the explosive, approximately 50% is associated with the shockwave with
the remaining 50% available for the later bubble oscillation phase.
Figure 24 shows a schematic of the far-field pressure-time history experienced at point in the water
following an underwater explosion. The peak of each bubble pulse corresponds to a bubble minima and
rapid flow reversal. The relatively larger duration of under pressure between peaks corresponds with the
bubble expansion phase. Compared to the initial shockwave, the bubble pulses are of less magnitude and
have longer rise and decay times. The peak pressure of the first bubble pulse is typically 10-20% of the
shockwave peak pressure. For subsequent bubble oscillations, a significant decrease is observed in the
peak bubble pulse pressure, the maximum bubble radius and the bubble oscillation period. The reduced
peak pressure however is accompanied by broadening of the pulse such that the impulse is less affected,
at least for the first 2-3 cycles.
Figure 24. Schematic of a far field pressure-time history of an underwater explosion and the relative gas bubble size
and shape.
The decrease in bubble pulse pressure with successive cycles is caused by energy losses which occur
primarily around the time of bubble minima. One obvious source of energy lose is the radiated energy of
the bubble pulse itself which is not returned to the bubble. However, other more complex energy
dissipation mechanisms are known to occur during the bubble collapse phase which may have a dominant
effect. Although the details of such mechanisms are not completely understood (Cole, 1948 and Hicks,
1972& 1986), they are known to be dependant on bubble size. In the case of relatively large bubbles for
which the difference in hydrostatic pressure between the top and bottom of the bubble is significant with
respect to the average hydrostatic pressure at the bubble centre, the bubble remains approximately
spherical during expansion but distorts into a toroidal shape around the time of the bubble minima,
recovering the spherical form upon re-expansion. The toroidal bubble generates a dissipative flow of high
velocity water directed through the centre of the toroid, referred to as a water jet. In the case of smaller
bubbles, particularly at larger depths, bubble distortion is much reduced. For such bubbles however,
evidence as been observed for the occurrence of so-called Taylor instabilities of the bubble surface during
the bubble collapse phase. Such instabilities allow thin jets of water to penetrate into the bubble, thus
763
cooling the bubble and reducing its pressure. Analytical bubble models necessarily adopt a simplified
approach to account for complex energy loss mechanisms during bubble oscillation.
Buoyancy force (displacement volume of the bubble) causes the pulsating bubble to migrate upwards
toward the free-surface. The vertical migration velocity of the bubble is much greater when the bubble
volume is near minima than during the expansion phase. This can be understood by noting that although
buoyancy forces are smallest at bubble minima, the inertia of the surrounding water (added mass effect) is
much reduced in comparison to the bubble maxima. The proximity of the free-surface has a marked effect
on the bubble oscillation period, maximum bubble radius, migration rate and, therefore, the number and
magnitude of the bubble pulses impinging on the vessel. Larger bubbles relatively close to the surface are
characterised by both high migration rates and high multi-cycle energy dissipation due to the
aforementioned bubble distortion.
9.1.3 Experimental Investigations

The ISSC Naval Vessel Design committee has been given access to a unique set of experimental results
for a scaled warship model subjected to various UNDEX events. The ship model, Figure 25, consists of a
segmented beam type arrangement with 8 external steel hull segments and 7 internal steel cylindrical
tubes along the length. The ends of each tubular member are welded to transverse bulkheads with
supporting longitudinal brackets. These connections provide a flexible joint between the tubular members
which together form a multi-segment beam backbone for the model. The external structure of the model
consists of eight hull segments which are welded to the bulkheads and connected to each other by means
of soft rubber transverse strips to provide a watertight joint. The inclusion of the soft rubber strips ensures
that the external hull sections contribute negligibly to the global bending stiffness of the model, the later
stiffness being provided almost entirely by the internal tubular backbone. The external hull segments
contribute some structural mass to the model, however a much larger added mass of external fluid is
determined by the hull section geometry.
Figure 25. Segmented scaled warship model used in UNDEX whipping experiments.
The main particulars of the scaled ship model are given in Table 8 includes measured values of the first
(lowest) two vertical bending mode natural frequencies of the model structure in water (wet frequencies)
and the damping ratio of the first bending mode. The first and second modes correspond to the familiar 2node and 3-node modes respectively for a free-free Euler beam. The materials properties of the model are
given in Table 8. Details of the moments of inertia and segment masses are given in Table 9.
Table 8. Key parameters of the scaled ship model
Main Parameters
Experimental Model
Length overall L(m)
7.95
Breadth B(m)
0.89
Draught Dt(m)
0.23
Structural Mass M(kg)
906
1st bending mode frequency- wet (Hz)
10.7
2nd bending mode frequency - wet (Hz)
22.9
Damping ratio of 1st bending mode
0.3%
764

Table 9 Material properties of the model.
Property
Steel
Sealing Rubber
Young modulus (MPa)
2.11E5
784
Poisson ratio
0.3
0.47
Yield stress (MPa)
235
Ultimate tensile stress (MPa)
375
29
Elongation ( %)
26
900
Density (kg/m3)
7850
950
The scaled ship model was subjected to three UNDEX events as described in Table 10. The charge depth
is defined as the depth of the charge below the water surface, is the angle from the base of the keel to
the charge location, R is the distance from the charge to the closest point on the keel and the bubble
frequency ratio is defined as the ratio of the bubble oscillation frequency to the wet natural frequency of
the first vertical bending mode of the model.The longitudinal location of the charge was at amidships
(L/2) for Cases 1 and 2 and at quarter of the length from the bow (L/4) for Case 3.
Table 10. Moments of Inertia and segment mass for the model.
Cylinders No.
1
2
3
4
5
6
7
Inertia moment (m4)

1.967E-05
2.975E-05
4.016E-05
4.078E-05
3.543E-05
2.885E-05
1.876E-05
Segments No.
1
2
3
4
5
6
7
8
Segment mass (kg)

155
102
99
116
97
86
83
168
Total = 906 kg
Table 11. Experimental UNDEX events.

ChargeCase
Mass.
No.
W(kg)
Charge
Long.
Position
Charge
Depth.
D(m)
Attack
Angle. R(m)
()
R to
BubbleBubble.
Freq.
Radius
Ratio
Ratio
1
2
3
L/2
L/2
L/4
1.77
1.77
1.77
46.7
90
48
1.041
0.763
0.961
0.05
0.02
0.04
2.12
1.54
2.07
3.74
3.71
3.96
Further information and results from this benchmark study are presented in ANNEX 1.
10
DISCUSSIONS AND CONCLUSIONS
In this report the committee have tried to highlight differences between military and commercial
structural design and also to concentrate on naval specific structural requirements.
From the study of different classification rules it can be concluded that:
Military operational loads are well covered by Classification Rules.

Military damage scenarios are either well covered by some Class Societies, and little or no coverage
by other Class Societies.
The responsibility for input loads for military damage scenarios follow different practice from one
Class Society to another.
Certification against Naval Ship Code may overrule the class requirements on military damage
scenarios
765
One Class Society handles the military damage scenarios as performance requirements that are
verified against military standards instead of Class Rules requirements.
There is still a way to go for Class Societies and Navies to have a common understanding of the
responsibilities for defining the military damage scenarios and how to handle these.
Many Class Societies have put a lot of investment, competence and effort in the coverage of military
loads and damage scenarios. Also some navies have put competence and investment in this development.
It is now up to the Navies to utilize this resource so that class societies can follow up with practical
experience and further development in this area.
For the main military load effects analytical and numerical methods are in place. For the presented
methods however limited validation material is available especially on the full ship scale. Depending on
the time available for the analysis and the needed accuracy, analytical, numerical methods or a mix of
both is being used. Where in the latter case the load is determined analytically and put onto the finite element model of the structure. As however treats, naval ship structures and there materials continue to develop there will always be a need for experimental research in which validation of numerical methods
should play an essential role.
Naval warship design is fundamentally similar yetideologically different than commercial ship design.
The uncertainty of the mission along with nobility of the role forces naval designers to consider increased
probabilities. These probabilities, such as hostile threats in conjunction with adverse environmental
conditions, are also related to higher consequences.All have to be accounted for in conjunction with the
endurance readiness of the ship.The same considerations would not normally be a concern for a commercial
ship designer. Accommodating the structural mission needs of naval warships is done through two distinct
ways;permanent integrated structures and modular flexible structures. Today, traditional naval warship
design has embraced the methodology of permanent, functional integrated structural design features; navies
are seeing a necessity to move to more flexible naval mission structure with modular capabilities.With
advancement of technology and weaponry, experts say this type of flexible and adaptable behavior will be
critical to the ever evolving mission needs.With the emergence of the next generation US Navy fleet looking
to these methods, such as the LCS class variants, time will tell whether there is a need to institutionalize the
modular adaptable ship.
In Chapter 6 a brief overview has been provided of naval mast design. There are few, if any
comprehensive publications covering all aspects of mast design, and a designer are left on his own to
design a mast as any other structure. Unfortunately this includes Class naval rules where various aspects
of mast design are distributed throughout the rules, and some not included at all. Given the complexity
and importance of a naval mast structure, this is not a particularly desirable state, and this committee
recommends that Class societies, or naval design authorities, produce a comprehensive, collected rule set
for naval mast design.
Chapter 7 discusses progressive collapse techniques and Table 7.1 provides an overview of the different
capabilities offered by the Smith Progressive Collapse method and non-linear FEA.It can be seen that whilst
FEA may have advantages of offering more detailed modelling and analysis, and what may be considered a
more accurate assessment of the ultimate strength of a ships hull girder, the Smith method still has a
significant role to play in both early design stages and emergency response, provided the analyst is confident
in the assumptions made in its application. Caution must be taken when implementing the method as factors
such as modelling details e.g. initial imperfections, residual stresses, etc., and variations between the true
failure mode of the vessel (interframe, overall, torsion, shear, etc) compared to the assessed interframe
failure mode could lead to incorrect prediction of the ultimate strength of the vessel.
Whilst FEA may be the obvious solution for performing more complex bespoke analysis, time constraints
along with factors such as design maturity, analyst experience, local model detail handing can all affect the
quality and accuracy of the results obtained and their value in answering to the question being asked.
In conclusion, whilst the capability and accessibility of FEA has increased significantly since the Smith
Progressive Collapse Method was originally conceived, the method continues to hold a valuable place in the
design and analysis of ship structures.
The use of high speed craft by navies worldwide will continue to grow in demand as countries race to
secure coastal and littoral waters. The speed capabilities, payload agility, and range that high speed craft
offer governmental agencies and the naval force will be driving factors for the continued acquisition of new
platforms. Today, the commercial shipbuilding industry is leveraged by naval authorities for their extensive
experience in design and construction of high speed craft such as ferries and pleasure craft. Classification
societies are used as third party validators for design and construction and also in order to supplement the
knowledge gap within naval technical authorities. As state-of-the art technology in lightweight materials
766
continues to mature and techniques for mitigating slamming consequences progress, naval technical
communities will continue to acquire knowledge and understanding in small high speed craft design. This
will hopefully propel a new era of research and development specific to this area.In conclusion, the
projection of high speed design techniques into the traditional processes and framework of naval standards
will ultimately redefine the next generation of naval fleet.
Chapter 9 gives an overview of benchmark problems carried out by the committee. The reults of which
are presented in the Annex to the report.
REFERENCES
ABS 2014. Rules for Building and Classing High Speed Naval Craft 2014 Part III. ABS: Houston, TX. 2014.
ASTM, F 1166-07, 2013. Standard Practice for Human Engineering Design for Marine Systems, Equipment, and
Facilities.
Baker, W.E. 1973. Explosions in the Air, University of Texas Pr., Austin.
Bartlett, S. & Jones, Brian. 2013. Composite Ship Structures, Naval Ship Warfare Center Carderock Structures
and Composites Division Code 65, ASNE Day.
Benson, S. 2011. Progressive collapse assessment of lightweight ship structures.PhD Thesis, Newcastle University
School of Marine Science & Technology, UK.
Benson, S., AbuBakar, A. & Dow, R.S. 2012. A comparison of computational methods to predict the progressive
collapse behaviour of a damaged box girder.Engineering Structures 48: pp266-280.
Benson, S., Downes, J. & Dow, R.S. 2013a. Compartment level progressive collapse analysis of lightweight ship
structures.Marine Structures 31: pp. 4462.
Benson, S., Syrigou, M., Dow, R.S. 2013b. Longitudinal strength assessment of damaged box girders.Proceeding of
the 6th International Conference on Collision and Grounding of Ships and Offshore Structures: pp. 305314.
Boyd, R., Royles, R. & El-Deeb, K.M.M. 2000.Simulation and Validation of UNDEX Phenomena Relating to
Axisymmetric Structures. 6th International LS-Dyna Users Conference Simulation
Brown, J.C. A wave induced fatigue strain recorder for surface ships. Society of Naval Architects and Marine
Engineers. Proceedings of the Annual meeting 30th September-1st October 1999.
Bureau Veritas (BV) 2011. Rules for Classification of Naval Ships
Caldwell, J.B. 1965. Ultimate longitudinal strength.Trans.RINA 1965: pp. 411430.
Cole, R.H. 1948. Underwater Explosions. Princeton University Press, NJ, USA
Collette, M. 2011. Hull Structure as a system: Supporting Lifecycle Analysis. American Society of Naval Engineers.
No.3: pp. 4555.
Daily, L.D., 2004. Simulating Afterburn with LLNL Hydrocodes, Lawrence Livermore National Laboratory,
UCRL-TR-206313
DAMEN 2013. Royal Netherlands Navy OPV Holland DAMEN Schelde Naval Integrated Mast Module (IMM)
URL: www.damennaval.com/nl/news, October 2013.
Det Norske Veritas (DNV) 2012. Pt3 Ch3, Section 4, Masts and Rigging.
Det Norske Veritas (DNV) 2014. Rules for Classification of ships.
Didoszak, J.M., 2004. Parametric studies of DDG-81 Ship Shock Trial Simulations, Thesis Naval Postgraduate
School, Monterey CA
Ding, P., Buijk, A. 2005. Simulation of Underwater Explosion using MSC.Dytran, MSC software Corporation.
Doerry, N. H. 2012. Institutionalizing Modular Adaptable Ship Technologies, SNAME Paper No. 047-2012.
Doerry, N. H. 2013.Modular Adaptable Ship Design Implementation, Naval Sea Systems Command Technology
Group SEA 05TD, ASNE Day.
Dow, R.S. & Smith, C.S. 1986. FABSTRAN: A computer program for frame and beam static and transient response
analysis (nonlinear). Admiralty Research Establishment, Dunfermline. Report No. ARE-TR-86205.
Dow, R.S., Hugill, R.C., Clark, J.D., Smith, C.S. 1981. Evaluation of ultimate ship hull strength.Extreme Loads
Response Symposium: pp. 133148.
Downes, J., Ham, W.L., Dow, R.S. 2013. Assessment of the residual strength of a lightweight naval
vessel.Proceedings of the 12th International Symposium of Practical Design of Ships and Other Floating
Structures (PRADS 2013): pp. 11891196.
Drones of the Navy, ACTUV, DARPA/SAIC. URL: http://www.navaldrones.com/ACTUV.html
Eccles, T.J., Ashe, G. & Albrecht, S. 2011. The Achieving Service Life Program. American Society of Naval
Engineers No.3: pp. 103112.
Fleck, N.A, Deshpande, V.S., 2004. The resistance of clamped sandwich beams to shock loading. Journal of
Applied Mechanics. ASME 71(3): pp. 386401.
Furio, A.J. 2002.Loads, Materials, and Structures, NAVSEA Caderock.
Furio, A.J. 2005.Lighweight Metal Sandwich Structures (LMS) Technology, Littoral Combat Ship (LCS) Technical
Information Paper (TIP), August 2005.
Gardiner, C.P., Vincent, P., Wilson, A., Ellery, D. & Armstrong, A. 2009. Hull Structural monitoring of the
Armidale class Patrol Boat. RINA Innovation in high Speed Marine Vessels. 28-29th January 2009.
Germanisher Lloyds (GL) 2012. Rules for Classification and Construction Naval Ship Technology
767
Grabovac, I. & Whittaker, D. 2009. Application of bonded composites in the repair of ship structures A 15-year
service experience. Composites. Part A 40: pp. 13811398.
Hackett, J.P. 2011. Composites Road to the FleetA Collaborative Success Story.Special Report 306. Naval
Engineering In The 21st Century, The Science And Technology Foundation For Future Naval Fleets. URL:
onlinepubs.trb.org/onlinepubs/nec/61810Hackett.pdf
Heuvel van den, W., Vaders, J.A.A. &Trouwborst, W. 2013. Maritime improvised explosive devices modelling and
large scale trials. Imarest Engine As A Weapon Symposium
Hicks, A.N. 1972. The Theory of Explosion Induced Ship Whipping Motions. NCRE Report R579
Hicks, A.N. 1986. Explosion Induced Hull Whipping. Advances in Marine Structures. Proceedings of an
International Conference held at the Admiralty Research Establishment, Dunfermline, Scotland. Editors C.S.
Smith and J.D. Clarke. Elsevier Applied Science.
IACS 2014.Requirements concerning strength of ships.International Association of Classification Societies.
IMO International Code of Safety for High Speed Craft 2000. 2008 Edition As Amended.
ISSC 2000. Committee VI.2 Ultimate Hull Girder Strength
ISSC 2006. Committee V.5 Naval Ship Design
Kane, C.E., Fixter, G. &Pogson, G. 2010. The Advanced Technology Mast For HMS Ark Royal. 17th International
conference on Composite materials (ICCM 17).
Kiddy, J.S., Baldwin, C.S., Salter, T., & Chen, P.C. 2002.Structural Load monitoring of the RV TRITON using fibre
optic sensors.Proc SPIE 4698, Smart Sensors and Materials 2002. Industrial and Commercial Applications of
Smart Structures Technologies: pp. 462.
Kingery, C.N. &Bukmash, G. 1984 .Airblast Parameters from TNT Spherical Air Burst and Hemishperical Surface
Burst.Defence Technical Information Center.Ballistic Research Laboratory.Aberdeen Proving Ground. Maryland.
Kinney, G.F., Graham, K.J. 1985. Explosive Shocks in Air. Springer. Berlin.
Koenig, P., Nalchaajian, D. & Hootman, J. 2008.Ship Service Life and Naval Force Structure. Paper accepted for
ASNE symposium Engineering the Total Ship. Tysons Corner, Va: pp. 2325.
Krauthammer, T., Altenberg, A. 2006. Negative phase blast effects on glass panels. International Journal of Impact
Engineering. 24 (1): pp. 271277.
Larcher, M. 2007. Simulation of the Effects of an Air Blast Wave.JRC Technical Notes.
Lloyds Register (LR) 2004. Ship right Ship Event analysis, May 2004.
Lloyds Register (LR) 2013. Code for Lifting Appliances in a Marine Environment. August 2013.
Lloyds Register (LR) 2014. Rules and Regulations for the Classification of Naval Ships
Mair, H.U., Reese, R.M. &Hartsough, K. 1997.Simulated Ship Shock Tests/Trials. URL: www.ida.org/LFTE
simulation/documents/SSSTT.htm
NATO Standard 2014. ANEP-77, Naval Ship Code, Edition E, Version 1
Naval-technology.com, 2014a. URL:http://www.naval-technology.com/projects/as565-panther/images/4s-
as565mb-helicopter.jpg
Naval-technology.com, 2014b. URL: http://www.naval-technology.com/projects/sentinel-class/images/3s-
sentinel-class.jpg
Paik, J.K. & Thayamballi, A.K. 2006.Ultimate limit state design of steel-plated structures.John Wiley & Sons, Ltd.
Petrusa, D.C. 2004. Evaluation and analysis of DDG-81 simulated athwartship shock response. Thesis Naval
Postgraduate School Monterey CA
Phelps, B. & Morris, B. 2013.Review of Hull Structure Monitoring Systems for Navy Ships.DSTO-TR-2818.
Polish Register (PRS) 2014. Correspondence
Registro Italiano Navale (RINA) 2011. Rules for Classification of Naval Ships
Riley, M. 2010. Analytical Solutions for Predicting Underwater Explosion Gas Bubble Behaviour.Defence R&D
Canada Atlantic. Technical Memorandum 2010-237
Samras, R. 2011. Stern Ramps: The Next Innovation. The TCG Pulse.
Savage, C.L. & Kimber, A. 2010. When Is A Mast A Mast? The Impact Of Integrated Masts On Platform Design
And Classification. RINA Warship 2010: Advanced Technologies in Naval Design and Construction, June 2010.
www.bmtdsl.co.uk/.../BMTDSL-When-is-a-Mast-a-Mast-Confpaper-RINA
Savitsky, D. 2003. On the Subject of High-Speed Monohulls.SNAME.Greek Section, Athens, Greece.
SELEX ES Unimast 2013. URL: www.selex-es.com. Accessed: October 2013.
Shin, Y.S. 2004.Ship Shock modeling and simulation for far-fiels underwater explosions.Computers and Structures.
Slavik, T.P. 2010. A coupling of empirical explosive blast loads to ALE air domains in LS-DYNA. IOP Materials
Sience and Engineering.
Smith, C.S. & Dow, R.S. 1981. Residual Strength of Damaged Steel Ships and Offshore Structures.Journal of
Constructional Steel Research. Vol 1 No.4: pp. 215.
Smith, C.S. 1977. Influence of local compressive failure on ultimate longitudinal strength of a ships hull.
Proceedings of the International Symposium on Practical Design of Shipbuilding (PRADS): pp. 7379.
768
Smith, C.S., Anderson, N., Chapman. J,C., Davidson, P.C. & Dowling, P.J. 1991. Strength of stiffened plating under
combined compression and lateral pressure.Transactions of the Royal Institution of Naval Architects: pp. 131
147.
Starossek, U. 2009. Progressive collapse of structures. Thomas Telford Limited.
Storhaug, G. 2009. The 4400TEU Container Vessel MSC NAPOLI Broke Its Back, But Did Whipping Contribute?
Hydroelasticity 2009 Proceedings, Hydroelasticity in Marine Technolog: pp. 233243.
Swisdak, M. M. 1978. Explosion Effects and PropertiesPart II: Explosion Effects in Water. Technical
Report.Naval Surface Weapons Center. Dahlgren, VA.
Taylor, G. I. 1963. The Pressure and Impulse of Submarine Explosion Waves on Plates.The Scientific Papers of G.
I. Taylor. Vol III, Cambridge Univ. Press, Cambridge, UK: pp. 287303.
THALES 2013. The Thales Integrated Mast Family. URL: www.thalesgroup.com/en/netherlands/defence.
Accessed: October 2013.
Torkildsen, H.E., Grvlen, .,Skaugen, A., Wang, G., Egil Jensen, A., Pran, K. &Sagvolden, G. 2005. Development
and Applications of Full-Scale Ship Hull Health Monitoring Systems for the Royal Norwegian Navy. Meeting
Proceedings RTO-MP-AVT-124, Paper 22. Recent Developments in Non-Intrusive Measurement Technology for
Military Application on Model- and Full-Scale Vehicles: pp. 114. Neuilly-sur-Seine, France: RTO.
Torrez, J.B. 2007. Light-Weight Materials Selection for High-Speed Naval Craft.Masters Thesis. Massachusetts
Institute of Technology.
Trouwborst, W. & Bosman, T.N. 2006. A Simplified Fluid-Structure Interaction Model For Underwater Shock
Loading Of Naval Vessels. European Survivability Workshop.
Turk Loydu (TL) 2013. Rules for the classification of naval ships.
Turton, T.J., Dazel-Job, J. & Livingstone, F. 2005. Oil Platforms, destroyers and frigates case studies of QinetiQs
marine composite patch repairs. Composites. Part A (36): pp. 10661072.
Underwood, J.M. 2013.Strength Assessment of Damaged Steel Ship Structures.Engineering Doctorate (EngD)
Thesis, University of Southampton Fluid Structure Interactions Research Group.
Walton, B. 2003. Designing Blast Hardened Structures for Military and Civilian Use. AMPTIAC.Vol. 6(4).
Wikipedia, 2014a.URL: http://en.wikipedia.org/wiki/HMAS_Sydney_(FFG_03) Accessed 2nd January 2014
Wikipedia, 2014b.URL: http://en.wikipedia.org/wiki/Kanimbla_class_landing_platform_amphibious Accessed 2nd
January 2014.
Wikipedia, 2014c. URL: http://upload.wikimedia.org/wikipedia/commons/2/28/Littoral_Zones.jpg
Yao, T. &Nikolov, P.I. 1991. Progressive collapse analysis of a ships hull under longitudinal bending. Journal of
The Society of Naval Architects of Japan. Vol 170: pp. 449461.
Zakrisson, B., Wikman, B & Haggblad, H. 2010, Numerical Simulations of Blast Loads and Structural Deformation
from Near-field explosions in Air.

710 SEPTEMBER 2015
CASCAIS, PORTUGAL
VOLUME 2
COMMITTEE V.6
ARCTIC TECHNOLOGY
COMMITTEE MANDATE
Concern for development of technology of particular relevance for the safety of ships and offshore
structures in Arctic regions and ice-covered waters. This includes the assessment of methods for calculating
loads from sea ice and icebergs, and mitigation of their effects. On this basis, principles and methods for the
safety design of ships and fixed and floating structures shall be considered. Recommendations shall also be
made regarding priorities for research programmes and efficient implementation of new knowledge and
tools.
COMMITTEE MEMBERS
Chairman:
S. Ehlers, Norway
F. Cheng, UK
W. Kuehnlein, Germany
I. Jordaan, Canada
P. Kujala, Finland
Y. Luo, China
K. Riska, France
J. Sirkar, USA
Y.T. Oh, Korea
K. Terai, Japan
J. Valkonen, Norway
F. Ralph, Canada
KEYWORDS
Ship and offshore structures in ice, rule-based and first principal-based ice loads, mission-based design
770
ISSC committee V.6: ARCTIC TECHNOLOGY
CONTENTS
1.
INTRODUCTION
1.1 Limitations
771
772
2.
PRESENT DESIGN METHODS

2.1 Ships
2.1.1 Rules
2.2 Offshore structures
2.2.1 Rules
2.3 Validation methods
772
772
773
776
780
783
784
788
3.
CASE 1: SHIP TRANSPORTATION IN ARCTIC WATERSTHE NSR
790
4.
CASE 2: FLOATING OFFSHORE STRUCTURES IN ARCTIC WATERS
793
5.
FUTURE PERSPECTIVES AND CHALLENGES

5.1 Numerical simulations
5.2 Ice induced fatigue
795
797
799
6.
SUMMARY AND RECOMMENDATIONS
801
ACKNOWLEDGMENTS
802
REFERENCES
802

1.
771
INTRODUCTION
Oil and gas, mining, transport, fishing and tourism activities are continuously increasing in arctic regions
and in need of reliable structural designs. The arctic region in this report is understood as a sea area with
seasonal or year round, partial or full, ice coverage and cold climate, see Figure 1.
Figure 1. Illustration of t he Arctic region (Illustration: Bjarne Stenberg, Copyright: NTNU).
The presented report is focussing on the safety of vessels in ice and offshore structures, i.e. ice forces and
stresses, but in general the following issues should also considered for arctic operations: Environment,
Social aspects (offshore), Human factors, Operational aspects including ice management, Engineering,
Installation and Decommissioning.
To obtain reliable structural designs, present design methods benefit from the vast experience of ships
operating in first year ice. The elements of interest in these design methods, and consequently in this report,
are the design ice-induced pressures and their occurrence as well as the corresponding minimum scantlings
determination and their applicability to new operational areas. However, the current design methods are not
necessarily transparent by means of design pressure and scantlings determination using intrinsic design
criteria. This concept is presented in Figure 2, which indicates the dependency of the design criterion and
pressure on the resulting scantlings, respectively plate thickness. For the design of offshore structures for
arctic regions the design pressures and their occurrence are supposed to be site specific, however, they are
primarily for iceberg encounters, apart from Prirazlomnoye and the Sakhalin structures, since practically no
offshore structure is deployed today at a location dominated by multi-year ice.
Figure 2. Present design method variations to obtain scantlings based on design pressures and design criteria.
This report seeks to at first discuss present design methods followed by first principle-based pressure and
occurrence determination methods to obtain corresponding scantlings. Consequently, these first principlebased methods may contribute to the design of structures beyond the current minimum requirements as
well as for new operational areas in the arctic region and to the further development of the present design
methods. To demonstrate these first principle-based methods this report includes two case studies, a
772
transit operation along the Northern Sea Route (NSR) and a floating offshore structure located in icebergprone waters. In conclusion, we will motivate for a more mission-based design methodology in addition
to the present design methods using the presented first principle-based methods.
1.1
Limitations
The influence of the following aspects on the overall design of ship and offshore structures are not addressed in this report: still water and wave loads, specific metocean and environmental conditions, coldclimate in general, extended periods without daylight, remoteness and the lack of infrastructure including
communication. Polar lows, which can cause unidirectional icing on exterior surfaces and thus stability
problems are not addressed, because they are winter phenomena while shipping takes place mostly in the
summer season. Further, hull girder strength issues caused by icing on the deck relevant for loading or
unloading operations at port are not considered, in spite of the lack of a load margin relevant for such
cases besides the wave load margin. The overall design of ships and offshore structures must however
consider all these phenomena from an operational perspective. The latter is addressed in the conclusive
chapter on future perspectives and challenges.
2.
PRESENT DESIGN METHODS
This section presents at first the current regulatory design methods to be used to identify design ice loads
and corresponding scantlings. Further, first principles-based methods to obtain actual pressures and exposures to be used for design scantling determination are presented as possible way forward.
2.1
Ships
Larger ships are entering ice-covered waters even though present design methods are based on the experience
and ice load measurements from small ships. Furthermore, large ships are stiffened primarily in longitudinal
direction, for which there are no observations or measurements as to how longitudinally framed structures
experience ice loading on large ships. The ships operating in Arctic Sea and particularly along the Northern
Sea Route (NSR) consist of conventional ice strengthened ships with ice breaking or bulbous bow. Most
transits along the NSR are destinational traffic where hydro carbons and minerals are shipped out of the Arctic. Hydrocarbons and minerals are transported year around mainly by Sovcomflot from Varandei and
Prirazlomnoye in Pechora Sea and by Norilsk Nickel from Dudinka along the Yenisei River. This transport is
done by modern double acting ships. Communities along the Russian Arctic coast and Siberian rivers are
supplied during the summer months by traditional ice going ships. Ice class IA (FSICR) or its IACS equivalent PC 7 or RS Arc 4 is the most common ice class among the ships sailing through the NSR in 2013
(CHNL, 2013). The second most common class is ARC 5 (IACS PC5 or IA Super). Most modern ships have
higher ice class up to Arc 6 or IACS equivalent PC 4 and they are equipped with azimuthing propulsion and
operate as a double acting ships (DAS) (Juurmaa et al., 2001). Norilsk Nickel and Sovcomflot (Niini et al.,
2007) operate these ships. This concept is also selected for the new ice classed LNG Carriers (LNGC) for the
Yamal LNG terminal. The Yamal LNGC is membrane type LNG carrier with a capacity of 170k m3 in 4
cargo tanks, being the largest in arctic region to date, see Figure 3. The underlying design aspects consider
the following: operate in the Kara Sea with independent operation (Arc 7 requirement), RS Winterization
(50C) and open water performance of 19.5-knot design speed. Thus, the DAS-concept with 3 PODs was
chosen to allow for moderate open water performance as well as a continuous speed in 2.1 m of level ice.
Von Bock und Polach et al. (2014) discusses the conflict of designing a vessel for both, open water and ice,
and and introduces the Combined Ship Merit Factor (CSMF) as a part of a design methodology operating in
ice and open water in detail. In the astern mode, the Azipods are working in a pulling mode, which together
with the specifically designed aft hull shape, do not require an ice knife to protect the struts. Further, the
LNGCs have a reinforce bottom above the rule requirements due to the shallow water depth in the Ob
bay, which also imposes a challenge on the draft limitations due to the additional structural weight increase. The ship side structure, also including a double hull for the engine room section, was reinforced
according to ARC7 with steel grades complying to an external air temperature of 50C.
Figure 3. Yamal LNGC (DSME, 2014).
773
The icebreaker development has also taken steps ahead. AkerArctic (2013) has developed a concept of
an oblique icebreaker that is breaking wide channel sideways. In Russia nuclear powered icebreakers are
under construction to replace aging fleet of existing icebreakers. Construction of a nuclear icebreaker LK60 has been started and two nuclear icebreakers are ordered in Russia (Barentsobserver, 2013).
However, most of the structural design follows a conventional approach to define ice strengthening as
an add-on to open water scantling. Therefore, Pedersen et al. (2014) presented an integrated approach to
identify the compliance to a target ice class by means of optimization, see Figure 4, and concluded that
that an optimum structural arrangement can differ for different ice classes and thus the inclusion of ice
strengthening as an add-on to an existing open water design may not result in the lightest or most economic structure.
Figure 4. The optimisation-based progression of weight for Arc4 to Arc7 for an LNG tanker (Pedersen et al., 2014).
2.1.1
Rules
An international legislative framework regulates the design and operation of arctic ships. The main bodies
of this framework are the United Nations Convention on the Laws of the Seas (UNCLOS), the International Maritime Organization (IMO), the maritime states, Recognized Organizations (ROs), and the International Association of Classification Societies (IACS) (DNV, 2012). In addition, there is the International Labour Organization (ILO) by the United Nations (UN), whose aim is to promote rights at work,
encourage decent employment opportunities, enhance social protection and strengthen dialogue on workrelated issues (ILO, 2014). The enforcement of mandatory requirements of the IMO conventions depends
upon the individual IMO members, which include most maritime states. The member state acts both as
flag states and as port states. A flag state has the authority and responsibility to enforce regulations over
vessels registered under its flag. Since all ships have to meet the international requirements set by the
IMO, flag states need to integrate their own statutory requirements with the requirements set by IMO.
When a Government accepts an IMO Convention it agrees to make it a part of its own national law and
to enforce it just like any other law (IMO, 2014b). As a result, any IMO member (maritime state) has the
authority to carry out so-called Port State Controls (PSC) to ensure that the condition and equipment of
ships visiting their ports comply with IMO standards (IMO, 2014c). This complex framework regulates
the design and operation of ships in general.
The Finnish-Swedish Ice Class Rules (FSICR) can be considered as the industry standard for ships operating in first year ice. The FSICR have been developed as part of the Finnish-Swedish winter navigation
system, because during an average winter every Finnish port and every Swedish port north of Stockholm
is ice bound. Thus, economical feasibility of the ship-based import and export depends on safe and efficient winter navigation through ice-covered waters for these countries in the Baltic Sea. However, it is
also this economical feasibility which conflicts with the ship design criteria as the current rules develop
along a thin line of acceptable ice induced damages, see Hnninen (2004), and acceptable CAPEX,
VOYEX and OPEX-additions for the vessels ice capabilities in view of changing environmental conditions and freight rates. Additionally, the cost of the escorting icebreaker fleet of each country is covered
by the vessels harbour fees, which decrease with increasing ice class. Hence, these fees may also motivate the choice of ice class. A detailed description on the evolution of the FSICR is given in Riska and
Kmrinen (2012). It must be noted, that even though these rules appear to be made for level ice
774
breaking, they in fact design the ship for brash ice conditions, e.g. a broken channel in level ice, with the
design point being the impact of the ship with the channel edge. Therefore, a simplified rectangular pressure patch has to be applied to the structure, where its height relates to the level ice thickness at the channel edge. Thus, a higher ice class (IA Super, hi= 1m) calculation requires a higher loading height compared to a lower ice class (IC, hi= 0.4m) for the same nominal pressure. The latter does also not impose
any speed restrictions, i.e. velocity dependency is not considered. However, the magnitude of the nominal
ice pressure is a constant value of 5.6 MPa for all ice classes, thus resulting in different design forces for
each ice class due to the different load heights to be used. Furthermore, the dependency of the ships propulsion power and displacement on the ice pressure is considered through a single factor, which accounts
for the possibility of the vessel to collide with ice at a high speed and is based on ice damage statistics.
Ice load measurements of two icebreakers operating in the Baltic Sea unveiled a maximum level of 2
MN/m, which appears to confirm the maximum rule height of 0.35 m at the nominal ice pressure and is
thus used as an upper limit value for the scantling design. Since the FSICR have been released in 1971,
the next revision occurred in 1985 as a result of ice load measurements, see for example Kujala and
Vuorio (1986), showing that the load height compatible with the measured ice load and structural response is much less than the ice thickness. As a result, the load height was adjusted to be a function of the
ice class, while the overall ice pressure remained equal for all ice classes. This development is shown in
Figure 5 together with the line-like contact.
Figure 5. The FSICR load height developments (Riska and Kmrinen, 2012).
The International Maritime Organization (IMO) Polar Code is a ship focused code with specific provisions for ship structure, subdivision, stability, equipment carriage (i.e. life-saving, navigation, and communications), crew training, and environmental protection for ships in the Arctic (N of 60oN) and Antarctic (S of 60oS). Consistent with other IMO codes (i.e. ISPS Code), the Polar Code contains both mandatory and non-mandatory provisions. The Polar Code provisions are additions to the IMO Conventions
(Safety of Life at Sea (SOLAS), Prevention of Pollution from Ships (MARPOL), and Standards for Training, Certification, and Watch-keeping (STCW) that will take effect through amendments to these instruments. The Polar Code requirements will apply only to ships to which the SOLAS and MARPOL conventions apply. In general, these are large commercial internationally trading cargo ships, tankers, and passenger ships. Key provisions of the Code include: Mandatory requirements for Certification, risk assessments, and voyage planning; ice strengthening, cold temperature protection, tank protection, and ice navigation training where appropriate; and additional restrictions on the discharges of wastes, including zero
discharge of oily mixtures. The Maritime Safety Committee of the IMO adopted the Polar Code in November 2014. The Marine Environment Protection Committee is expected to adopt the Polar Code in May
2015. The Polar Code is then expected to enter into force in January 2017.
In August 2006, the International Association of Classification Societies (IACS) released the Unified
Requirements (UR) for Polar Ships, which standardized global ice classification specifications. A general description of seven polar classes is given therein in terms of nominal ice conditions where the highest class is approximately equal to the highest Russian Register of Shipping ice class and the lowest to
FSICR IA Super. The intent of the technical requirements is to assist owners in matching the requirements for the ship with its intended voyage or service. For each polar class, factors are provided to scale
design forces according to the expected severity of ice loading conditions. While polar class ship will
experience a complex mix of ship-ice interactions during its operational life, the glancing shoulder bow
impact scenario has been selected as the basis for the scantling requirements. The scenario includes initial
ship/ice edge contact over a small area, with growing contact area until the entire structural grillage is
loaded. The design ice load is characterized as an average pressure uniformly distributed over a
775
rectangular patch load of height and width. Since concentrated pressures exist within the load patch, peak
pressure factors are assigned to account for pressure concentration on localized structural members.
The new ice class by the Russian Maritime Register of Shipping were published in 2010 (RMRS,
2010). The new ice classes are named as ARC4 to ARC 9. ARC 4 and ARC 5 result in similar scantlings
as the Finnish Swedish ice class IA and IA Super and ARC 9 is the highest ice class enabling navigation
in all ice conditions. The new rules define the link between ice class requirements and the ice conditions
in winter/spring or summer/autumn for the different Russian Arctic seas; see also Figure 27. In addition,
the type of navigation has also an effect on the ice lass needed and is thus divided into the following
items: independent navigation, icebreaker escorted navigation, extreme navigation (average periodicity
once in ten years), hard, medium and easy navigation (average periodicity once in three years). The scantling requirements are mainly based on the ships displacement and shape of the hull.
The Lloyds Register (LR) rules concern the structural integrity of the vessel in the form of ice class rules,
but generally, the equipment and operation for operation at low temperatures are the prerogative of the owners specification and behest. They include the Canadian Arctic Shipping Pollution Prevention Regulations,
Finnish and Swedish Ice Class Rules and the Russian Maritime Register of Shipping. LRs rule philosophy is
to align with these regulations supplemented by LRs service experiences and technology development in the
form of additional notations and guidance. For ships operate in first year ice condition, Lloyds Register rules
are aligned with the FSICR. For ships, which transit waters in which multi-year ice is present, LR rules incorporated the IACS UR for Polar Ships. A notable feature of the LR rules is that a wider range of shell, including bottom shell, is required to be ice-strengthened. This might stem from the consideration that ice is
pushed passing the bottom of a ship as the ship advances in more open water with swells. In addition, LR
released the new procedure, ShipRight Fatigue Direct Assessment (FDA) ICE, in July 2011, which assesses
the potential for structural fatigue in vessels designed to trade in ice-covered waters and thereby helps to reduce the risk of fatigue damage in the hull structures of their ice-strengthened vessels.
DNV GL has three categories of their Rules for Classification of Ships for operations in ice. These
range from the lowest ice classes intended for very light ice conditions to the highest ice classes covering
year-round operation in all ice-covered waters. The basic ice classes (ICE-C and ICE-E) are intended for
service in waters with light ice conditions. This category typically applies for vessels, which are not designed for continuous operation in ice, but need ice class for commercial or operational reasons. The
structural reinforcements for these classes are generally limited to the bow area. The structural arrangements are often similar to open water ships, but plating is thicker and intermediate ice frames are placed
between the main frames in the waterline at the bow area. The second category sets requirements for service in the northern Baltic Sea in the winter or areas with similar ice conditions to the FSICR. The majority of commercial ice-strengthened vessels fall in this category. Additional guidelines for the propeller
blade design using direct calculations and ballast system design are provided. The third category is for
IACS Polar Class vessels. The requirements are in general equivalent to the IACS Unified Requirements
for Polar Ships (URI1 to URI3). Ships designed for ice breaking, e.g. PC-1PC-6, may be given the additional Icebreaker notation. This notation sets additional requirements for the propulsion machinery system
and bow form. Further, requirements for the steel are provided and for winterization of ships to ensure
compliance to cold climate. Three different levels of winterization are defined in the rules covering operations in light winter to extreme Arctic conditions. Additional notations for redundancy, noise and vibration and reduction of the risk to the environment are provided.
The ClassNK Ice Class Rules (CNKICR) are based on the FSICR for single-year ice and the IACS UR
for polar ships. Different to FSICR, CNKICR also considers the vessels bottom area for ramming loads.
Further, the adopted load scenarios are divided into a local load scenario and a global load scenario. In the
local load scenario, glancing impact between sea ice and side shell is considered. In the global load scenario, ramming is considered in order to check the longitudinal strength of ships. The design ice loads and
the corresponding requirements for shell plating and for longitudinal strength are excerpted from the
IACS UR for polar ships.
The Unites States Cost Card (USCG) policies design USCG icebreakers to follow the US Navy design
methods with additional considerations of ice loadings (based on Polar Star/Polar Sea and other icebreaker ice load measurements) rather than any class society rules apart from the IACS UR for polar ships
for minimum scantlings design and Popov (Popov et al., 1967) based ice impact simulations, see Yu et al.
(2007). The procedure relies on the application of environmental and operating loads with analysis of the
structural response, as opposed to scantling determination by empirical formula, see Daidola and
Sheinberg (1988). As an example of USCG framing design POLAR STAR and MACKINAW have a
haunched transverse/vertical framing with varying section area properties to minimums at mid span
through the ice belt. The contractor-designed HEALY ice frames incorporate essentially straight nonvarying sections with bracketed end connections. For the POLAR STAR/POLAR SEA, HEALY and
MACKINAW designs with the specified ice loads, the calculation requirement was for an elastic analysis
776
with allowable stress levels up to yield. Linear elastic FEA was performed to validate the stress levels
under ice loads. Further, the hull girder bending load cases associated with ice operations include the case
where the bow has risen up on an ice edge or ridge with an associated concentrated vertical load at the
bow and the waterline at the stem at the ice knife or stopper.
The Ministry of Transportation of Russian Federation regulates sailing along the Northern Sea Route
(NSR) (NSR, 2013) in a legislative framework that came in force in January of 2013. Management of the
navigation of ships in the water areas of the NSR is conducted by the Northern Sea Route Administration
Office (NSR, 2014) established as a federal government institution, which operates under the Ministry of
Transportation of Russian FederationAgency for River and Sea Transport. Application of permit for
sailing the NSR (no fee is paid for the permit) is to be submitted to the NSR Administration and is evaluated on the basis of data provided by the ship owner, which include the following: information about ship
and voyage, classification and measurement certificates, documents certifying availability of the insurance, other special certification if needed. In addition, to the structural compliance of the vessel in line
with the NSR rules, additional requirement concerns the ice pilot presence on-board during a vessel sailing the NSR. In case that a permit for sailing is granted, the ship owner is informed of considerations
regarding the ice conditions along the route and is advised under which circumstances sailing is allowed
and whether or not icebreaking assistance is needed. Ice conditions are described as: heavy (20% more
severe ice conditions than multi-year average for certain part of the route), medium (multi-year average
ice conditions for certain part of the route) or low (20% less severe ice conditions than multi-year average
for certain part of the route). Ice conditions are evaluated by the Arctic and Antarctic Research Institute
(AARI, 2014) and are available at the NSR Administration website. Rules are clearly defining the operational window for allowed sailing, which is dependent on the vessels ice class and ice conditions along
the route. Vessels ice class in terms of hull strengthening, machinery system and winterization is assessed according to rules set by Russian Maritime Register of Shipping (RMRS, 2014). Icebreaking assistance is usually provided by Rosatomflot (ROSA, 2014) and the fee in USD per ton is directly negotiated
with Rosatomflot representatives.
2.1.2
First principles
The basic concept to determine scantlings based on first principles is presented in Figure 2, if a design
pressure and occurrence is defined for the operational area in question to obtain design scantlings using a
corresponding design criterion, i.e. yield.
For the FSICR this concept may be shown according to Riska and Kmrinen (2012) with the example load measurements and Gumbel-I-based estimations predicting the most probable load for a given
time period as shown in Figure 6. The next step to determine the scantlings is the selection of the design
point as a function of a limit state and the frequency with which this limit state is reached for the operational area in question. If this limit is denoted as w, eventually representing yielding, and the structural
detail as d, than the limit state can be presented as
q = f ( w, d )
Following the most probable load and its return period, q = g (T ) , and the design scantlings according
to the ice loading statistic the design equation for the structural detail in question can be obtained as
d = h ( w,T ) .
Figure 6. Example of measured and estimated load versus return period for a bow frame (IA Super rule value is 790
kN/m) (Riska and Kmrinen, 2012).
777
The design point of FSICR is reaching yield at least once per winter. Therefore, some ice damage is likely
during lifetime. Kujala and Ehlers (2014) investigate the possibility to align eventual structural damages
following ice loading requiring repair in a risk-based assessment that considers foreseeable repair intervals of the vessel. As a result, they identify the required increase in design load and the corresponding
savings for one example vessel; see Figure 7.
Figure 7. Net Present Value for different structural alternatives, repair intervals and corresponding design load levels
(Kujala and Ehlers, 2014).
Ehlers et al. (2014a) presented a possible utilization of line-like contact as presented in Figure 5 for structural compliance calculations by means of a simplified, non-uniform, pressure patch (SNPP). They compare their simulations to medium-scale measurements with level ice at low velocities, where the line like
contact phenomena occurred as well. Generally, the consideration of a line-like contact is important, because local pressures of up to 60 MPa (Frederking and Sudom, 2008) have been observed. By doing so,
the influence of the temperature and impact velocity on the local contact mechanics must be considered.
Sub-zero temperature compliance of the base material and welded structures is still scarce in the current regulatory ship-specific frameworks and is typically concerned with the fracture toughness alone,
given as a Charpy requirement. Generally, classification societies suggest the material grade for ship
structures based on temperature, plate thickness and importance of structural members according to IACS
unified requirements. However, various steel grades emerged that can comply with cold climate, i.e. increase in yield strength, in Lders-plateau length and in ultimate strength and similar fracture strain as
found for ambient temperature, see Ehlers and stby (2012). Furthermore, they showed that even the
crashworthiness of structures under sub-zero temperatures can be improved significantly using such arctic
materials, while Ehlers et al. (2013) identified the possibility to even increase the global hull girder ultimate strength under arctic and non-arctic conditions with optimized structures using arctic materials.
In case of the IACS rules the choice a glancing shoulder bow impact scenario does not suffice, because
it represents a collision avoidance manoeuvre only. Head on collisions are necessary, particularly for the
highest ice classes, and should be considered, i.e. if a bulk carrier or icebreaker needs to transect a large
flow, it must ram head on until it breaks through. The designer or a carrier may choose to not design his
vessel for head on ramming, and accept delay, but the requirements should allow for such choice and
account for the indented operational scenario. Thus the question arises if a plastic response formulation
for plate and stiffeners for a glancing blow represents the most conservative interaction scenario already,
given that a certain level of permanent set in the plating and framing is expected, see also Figure 2. In
addition, global hull-girder loading due to ramming operation is specified. To investigate this further, the
probabilistic methods identified by, e.g. Ralph and Jordaan (2013) can be used to interpret the deterministic treatment of ship design in the IACS Polar Code. One finding is that CAC4 vessels may carry out direct rams of a limited number; possibly the glancing blow scenario covers these rams. This finding resulted from a revision of the Canadian Arctic Shipping Pollution Prevention Regulations (ASPPR) in
1989 as in Carter et al. (1992). In the latter work, a probabilistic basis for the ASPPR revisions was provided. The revisions involved four arctic classes: CAC1, CAC2, CAC3, and CAC4. Probabilistic support
was provided by consideration of the number of rams into ice per year, which varied from a few to thousands, depending on the class. This is essentially a measure of exposure. Furthermore, the analysis of
Kigoriak results supported the local pressure formulation in the standard and use of the equation
=1.25a 0.7 ; see also Figure 8. As for IACS, and in spite of the vessel experiencing a complex mix of
ship-ice interactions during its operational life, the glancing impact scenario has been selected in ASPPR
778
as the basis for the scantling requirements, due to the availability of a relatively mature model describing
this type of interaction. In this approach, repeated rams are not explicitly addressed. Thus, it is assumed
that what is meant is that the single ship/ice interaction event is the worst of a set corresponding to the
arctic class. Furthermore, it was observed during the ASPPR review process that multiyear ice does not
come in a set of thicknesses, it varies in thickness and ridges can be randomly placed in a floe and have
random thicknesses too. It was suggested in Carter et al. (1996) that exposure might be better measured
by number and severity of significant interactions. An example of how to assess exposure is presented in
the chapter 3 of this report. Figure 9 shows that global area for ship impacts are consistent and possibly
predated with those for other structures.
A rational approach to design of arctic ships based on probabilistic methods is detailed in Ralph and
Jordaan (2013) including global impact forces and local panel design. Using such probabilistic approach,
extreme design events can be identified by combining annual, seasonal and regional variability in environmental conditions with model uncertainty, and integrating these directly in the design methodology.
Extreme design loads are estimated based on the annual number of significant interaction events (i.e.
multi-year ice interaction), and the design strategytarget exceedance criteria established based on general
public safety. The approach also provides a comprehensive basis for the selection of an appropriate ice
class given certain operational requirements (e.g. an icebreaker for facility protection, or suitability of a
tanker, bulk carrier or cruise liner, having minimal ice class, to operate through a particular season). Otherwise, design for extremes is largely based on observational experience and judgment from one or more
experts and such experience only reflects a relatively short span of natural occurrence. Neither is it appropriate to arbitrarily establish the most extreme condition imaginable.
Figure 8. Design pressure area curve (Jordaan et al., 2005).

A = Cx2
A = Cx
Figure 9. Global areas for ship-ice interactions (derived from Riska (1995)).
For design, concern is for the safety of personnel, the structure and the environment, specifying design
loads that are appropriate. A design strategy references a specified load level from a distribution of annual
maximums that corresponds to some annual target exceedance probability pe. The influence of the number
of interactions in a year (or period) in developing an extreme value design load distribution from a parent
distribution is illustrated in Figure 10. With increasing interactions per year the design load distribution
will shift to the right of the parent distribution; with multiple years between interactions, the design distribution shifts to the left with a probability spike at zero load (Jordaan, 2005). For design, one strategy may
reflect a 1 in 100 year extreme level ice event (1% annual exceedance probability). The corresponding
structure may be designed elastically with some plasticity depending on whether local damage (e.g. a
dent) is tolerated without loss of vessel integrityfailure mode is important. Another strategy may reflect
779
a 1 in 10,000 year abnormal level ice event corresponding to a full plastic ultimate limit state (ULS) design. A load exceeding this level may cause substantial damage crippling the vessel and require immediate repairs but not catastrophic loss of the vessel. The strategy may also consider risk mitigation through
detection and avoidance. In no case should loss of life or environmental damage occur.
Using this design approach, one can quantitatively consider different classes of vessels based on number of annual encounters. The classifications used in the ASPPR include CAC1 through CAC4. Here,
maximum force would be estimated for the highest class (CAC1) and then factored for reduced classes
(i.e. 1.0, 0.8, 0.6, 0.4). The linear trend in loads with a step change in arctic class corresponding to an
exponential change in exposure (e.g. 10,000, 1000, 100, 10 impacts per year for vessel classes ranging
from CAC1 through CAC4 respectively (see Carter et al. (1996)) is observed. Based on interviews with
Arctic Captains, the assignment of annual number of rams (i.e. 10,000s, 1000s, 100s, 10s) for the categorical CAC1 through CAC4 vessel types in Table 1 was considered reasonable. Impact simulations
based on characterization of environmental conditions, including concentration and size of MY floes in
addition to number of transits per year also verified that this assignment of exposure is reasonable (Carter
et al., 1996). As a reference point, the MV Arctic was classified as a CAC4 equivalent and encountering
on average 10 to 15 MY impacts per year. The corresponding impact force is on the order of 35 MN.
Figure 10. Illustration of design point and corresponding annual exceedance probability pe including influence of rare
(i.e. years between interactions) and frequent (i.e. many interaction per year) events in developing extreme
value distributions from the parent per event load distribution (Jordaan (2005), Ralph and Jordaan (2013)).
Using the parent distribution of measured MV Arctic vertical bow forces, an extremal analysis was
conducted using an exponential fit to the tail of the measured forces (Ralph and Jordaan, 2013). Forces
for different classes based on expected number of class rams (or significant interaction events) is
illustrated in Table 1. Reasonable consistency is demonstrated between class factors in the Canadian
ASPPR (1995) rules and estimated design forces (based on expected number of rams) normalized to
highest class CAC1. This demonstrates that correlating class factors and force with the number of annual
interactions is quite reasonable. Otherwise, decisions concerning class are prone to subjectivity, relying
on expert judgment.
Table 1. Illustration of ASPPR class factors, estimated force and normalized force
based on annual number of rams (Ralph and Jordaan, 2013).
Class CAC class Number of significant Force at 10-2 annual
Force normalized
interactions per year exceedance probability to CAC1
CAC1 1.0
10.000
54.3
1
CAC2 0.8
1.000
46.0
0.85
CAC3 0.6
100
38.8
0.71
CAC4 0.4
10
31.5
0.51
1
24
0.44
Further analyses considered vertical concentric bow impact forces, estimated using the IACS UR for polar ships and comparing with estimates using the rational probabilistic approach discussed using measured
full scale MV Arctic data for the parent force distribution (see Table 2 and 3). The analysis illustrates how
measured forces and expected exposure can be directly used for design and classification, as well as calibration. Forces are presented for design strategy and annual exceedance probabilities corresponding to
780
both 10-2 (1/100 years) and 10-4 (1/10,000 years). As noted earlier, a 10-2 design strategy may correspond
to elastic-limited plastic design where an occasional minor dent may be tolerable without loss of structural integrity. A 10-4 design strategy may correspond to a fully plastic ultimate limit state (ULS) design.
The preliminary results highlight some interesting points regarding exposure and classification. The
PC3 force aligns with on average 1 ram per year at a 10-2 annual load level. The PC2 force aligns with
10,000 rams per year and a 10-2 annual load level. A rather broad range of exposure exists between PC2
and PC3. It is possible that an owner may wish to build a vessel that is more capable than a one (1) ram
per year level but less than 10,000 ram per year level. The PC1 force is considerably higher than the rational based load for 10,000 annual rams and a 10-4 exceedance level, which seems quite conservative.
Guidance on the design strategy used for a 10-4 load level (i.e. ULS) compared with a 10-2 level should be
considered.
Table 2. MV Arctic particulars (Ralph and Jordaan, 2013).
Length bp
199 m
Breadth
23 m
Draft
11 m
Displacement
39.000 t
Stem ange
30
Bow opening angle

33
Table 3. Comparison of IACS Polar Class design forces for MV Arctic with rational based forces based on annual
number of rams for 10-2 and 10-4 design strategies (Ralph and Jordaan, 2013).
IACS
Polar Class
PC1
PC2
vertical bow
force [MN]
82
56
Rational design approach

Number of
max bow force*
rams/yr
at 10-2 [MN]
10.000
1.000
100
10**
1
54
46
39
31**
24
max bow force*

at 10-4 [MN]
70
62
55
46
40
normalized 10-2
max bow force
1.00
0.85
0.71
0.57
0.44
PC3
25
PC4
16
*Max. vertical bow force estimated with extremal analysis and parent distribution from MV Arctic trials
** MV Arctic is a CAC4 equivalent encountering on the order of 10 rams per year
2.2
Offshore structures
Up to now, the Arctic fields developed are in water depths up to 125m and have used the Gravity Based
Structures and detachable FPSOs, see Table 4, besides other systems such as jacket platforms and islands
used in shallower water as shown in Figure 11. Current developments of various arctic stakeholders include drill ships for arctic regions, especially for the Kara, Beafourt and Chucki Sea. The design basis
currently stems from ice going vessels with dedicated ice breaking lines in the bow area, which are
upgraded with platform and drilling components for harsh environments. Some existing offshore structures involved in production in presence of icebergs include the following gravity based structures (GBS):
Steel Drilling Caisson (SDC), Molikpaq (now Sakhalin), Concrete Island Drilling System (CIDS), Hibernia GBS, Prirazlomnoye Platform and the Lun-A & PA-B Platform. Floating production structures include the Terra Nova and White Rose facilities involved in oil production on the Grand Banks. Probabilistic design methods have been used for some of these structures (for example in the Sakhalin area) and in
particular those on the Grand Banks.
Table 4. Summary of floating platforms in Arctic waters.
Location
Stationkeeping method
Beaufort Sea
Disconnectable mooring lines
(Canmar drill ships)
through the waterline
Beaufort Sea
Disconnectable submerged
(Kulluk)
mooring system
Grand Banks
(Terra Nova)
turret mooring system
Grand Banks
(White Rose)
turret mooring system
Pechora Sea
Loading tower
Sakhalin
Dynamic Positioning
Arctic Basin
Dynamic Positioning
Type of operation
Drilling
Drilling
Hydrocarbon
production
Hydrocarbon
production
Offloading
Diving
Core drilling
Length of operation
Long term 1976 to late
1980s
Long term 1983 to 1993
Long term 2002 to
Long term 2005 to
Long term 2000 to

Short term 1999
Short term 2004
781
Esso Kugmallit Island

Others:
Kashagan Oil Field, Northstar Island
Comments:
Non Retained Island (NRI)
Sandbag or Rock Retained Gravel
Island
a. artificial islands
c. caisson retained islands
e. Gravity Based Structures (GBS)
b. ice islands
d. floating structures
f. Pile Based Structures
Figure 11. Offshore structures for arctic regions.
Mooring system of floating structure operated in Arctic waters shown in Table 4 may encounter extreme
ice loads including iceberg encounters to which the mooring lines have to comply. In case the capacity of
the mooring system is below the incoming load, disconnect must take place as a mitigation measure. The
latter may further require that the turret be submerged to a safe depth below the iceberg. The disconnectable mooring system of the Terra Nova, the first FPSO site in the Grand Banks, uses a Thruster Assisted
Position Mooring System (TAPMS) consisting of 9 mooring chains attached to the base of the turret
which hold the FPSO in position together with five 5MW thrusters, (2 forward, 3 aft), which allow the
FPSO to control heading and position in severe environmental conditions.
Aggarwal and D'Souza (2011) indicated that the strongest chain available with minimum breaking load
of 31MN, with 24 mooring lines would provide maximum capacity of approximately 77MN and a maximum offset of 33m. In comparison, the detachable 3x3 chain mooring system found at the Terra Nova
FPSO has a capacity of about 20MN. Furthermore, the mooring system may consist of other materials
such as ultra-high strength molecular fibre rope with even higher capacities. The maximum ice load estimated for existing fixed platforms in shallow waters ranges from 500MN to 1,000MN depending on their
locations and the size of the multi-year ice or icebergs to be encountered. Thus, mooring systems may not
be adequate for all locations, but may be quite suitable in certain locations depending on the ice conditions present.
John Fitzpatrick of CJK Engineering Ltd. of Calgary has developed a concept for these conditions in
the arctic. This is shown in Figure 12. Tension Leg Platforms (TLPs) are for the most part compliant
with respect to horizontal wave loading. However the concept illustrated is non-compliant for both wave
and ice loading conditions. For extreme ice loads it becomes compliant to avoid catastrophic over load. In
the compliant condition, lateral movement of the structure absorbs energy. Up to 10,000 MJ of energy can
be safely absorbed without damage or tilting. This amount of energy is equivalent to a 1500m diameter by
50-meter thick ice island floe impacting with a speed of 1 knot.
782
Figure 12. Arctic TLP Concept by CJK Engineering Ltd.
Dynamic Positioning (DP) based stationkeeping in ice is an integrated operation between a DP vessel (the
protected vessel) and an ice management system consisting of one or several icebreakers, an ice monitoring system, and ice risk management Van der Nat (2012), see also Figure 13. However, traditional DP
control systems have not been designed to handle large and rapidly varying sea-ice loads, and DP stationkeeping has been done in manual joystick mode. Furthermore, a stationary vessel lacks the momentum
available from forward movement of a transiting vessel to overcome the ice load maxima. Additionally,
ice-monitoring systems are not custom-designed for online surveillance of sea-ice. Ice monitoring region
is vast, satellite images are infrequent, sensors are unreliable and have too low resolution to give sufficient details, and sensor carriers are manned (Aggarwal and D'Souza, 2011). In addition, ice breaking has
been developed based on transit and shipping, but ice breaking for ice management needs icebreaker vessels with different properties, typically higher capability, and a paradigm shift in how the icebreaking
operation is carried out (Liferov et al., 2011).
The current challenges for successful offshore DP operation in ice range from sea-ice load detection,
vessel design, hazard identification and mitigation strategies. Sea-ice loads on the DP vessel are large,
rapidly varying, and there is a lack of models to accurately estimate and predict the loads based on feasible measurements. DP vessels must typically be designed for optimal operation both in ice (Liferov et al.,
2011) (with sufficient DP-Ice capability) and in open water (transit and open water DP capability. Keeping headingbow against drifting ice is required. When the ice-drift direction changes, a massive increase
in ice loads and potentially dangerous situations may occur. This is especially critical for a DP drillship
that is connected with riser and the allowable positional offset is small. Reliable hazard
detection, identification, tracking, and forecasting of icebergs and severe ice features are not sufficiently
developed. Online monitoring of necessary ice properties and ice dynamics for safe situational awareness
and estimation/prediction of load evolution on DP vessel including the integration of an ice surveillance
network consisting of many sensors and sensor platforms, such as on-board sensors, underwater stations
and vehicles, aerial sensor platforms, and satellites (Aggarwal and D'Souza, 2011). In addition, the optimal operation of the icebreaker fleet to minimize loads on protected vessel is required. Therefore the efficiency of ice breaking must be modelled with sufficient fidelity to be able to plan and predict icebreaking
patterns. Present sea-ice parameters (sensors, sensor platforms, data processing, decision support tools,
etc.) for online calibration of models are required. This shall lead to an efficient deployment and the icebreakers for an optimal ice-breaking pattern to guarantee that the maximum loads on the protected vessel
are below acceptable limit within specified T-time. Therefore, online measurements and analysis of the
resulting efficiency of the ice management (e.g. providing real-time data on floe size distribution in the
ice management area) are needed. As a result, this shall ensure sufficient safety of integrated operation
between DP system and ice management system. Limiting conditions (design loads for DP vessel, ice
management capabilities, etc.) must be identified.
The current trends to meet these goals are the introduction of new sensors and ice surveillance decision
support systems (Loenhout, 2012), new icebreakers are built and physical ice management experience is
growing, sophisticated numerical simulators capable of reasonable simulation of loads and motions on
floating structures in sea-ice, are emerging (Bauduin, 2011), and DP control systems are becoming better
to handle sea-ice loads (Baan, 2011). The latter includes the following 4 steps: 1) Retuning existing DP
783
controllers to more aggressively estimate and reactively compensate ice loads; 2) Development of improved ice characteristics models and corresponding redesign of DP software to reactively handle ice
loads based on existing sensors; 3) Utilization of new sensors to directly estimate ice forces and reactively
compensate by feed forward (as for wind forces); and 4) Development of a sophisticated online ice monitoring system that is capable of locally predicting the future evolution of sea-ice loads on DP vessel, for
proactive compensation of sea-ice loads. This gives the DP system time to build up sufficient Thruster
forces to compensate incoming load variations before they hit the vessel. Furthermore, guidelines for
integrated stationkeeping operations in ice regarding surveillance zones, alert levels, T-time, repair time,
emergency procedures, etc., under development.
Figure 13. Offshore Dynamic Positioning operations in ice (Illustration: Bjarne Stenberg, Copyright: NTNU).
2.2.1
Rules
The main set of rules is the new ISO 19906 Petroleum and natural gas industriesArctic offshore structures, International Organization for Standardization, 2010. In brief, there are three Exposure Levels
(L1, L2, and L3) that depend on factors such as whether the platform is manned, whether it is planned to
evacuate, and consequence categories that relate to the potential risk to life and the environment and to
possible economic loss. The standard was calibrated to target safety levels, see Table 5, using a special
calibration exercise.
Table 5. Target safety levels in ISO 19906 for the three Exposure Levels.
Exposure Level
L1
L2
L3
Reliability target expressed as

annual failure probability
1.0 x 10-5
1.0 x 10-4
1.0 x 10-3
Both extreme-level (EL) and abnormal-level (AL) events are to be considered, with associated ice events
(ELIE and ALIE respectively). These events or actions are specified at annual exceedance probabilities of
10-2 and 10-4 respectively.
It is to be noted that Exposure in the ISO context above refers to the exposure with regard to safety, a
classification system used to define the requirements for a structure based on consideration of life-safety
and of environmental and economic consequences of failure. This related to the exposure with regard to
human life, for example. This ISO usage is different from the other connotation often used of the exposure of a structure to environmental hazards, for example the number and duration of interactions with ice
features. This latter interpretation is used below.
There are some deterministic provisions in ISO (19906), mainly aimed at providing estimates for the
100-year loads. The calibration work was very similar to that which had been carried out for CSA S471,
784
Canadian Standards Association CAN/CSA-S.471-04, General Requirements, Design Criteria, the Environment, and Loads which is the first of five CSA Standards that formed the Codes for the Design,
Construction, and Installation of Fixed Offshore Structures. The Standard set safety objectives for the
code as a whole. The Standard distinguished between Safety Class 1 and 2, with Safety Class 1 being
defined as a structure in which failure would result in great risk to life or a high potential for environmental damage, for the loading condition under consideration. This standard introduced 10-2 and 10-4
loading events in the context of frequent and rare occurrences. This standard is being superseded by the
ISO but was the precursor of risk-based offshore standards with consideration of ice loading.
The ISO 19906 and CSA S471-04 standards are the first standards to base the ice loads or actions on
full scale data obtained by the monitoring of offshore structures in areas such as the Beaufort Sea, Cook
Inlet, the Baltic Sea, the Gulf of Bohai and to a limited extent the Sea of Okhotsk. Thus the loads obtained
from the methods of these two standards tend to differ from those obtained by standards based on small
scale data (Masterson and Tibbo, 2011).
Furthermore ISO 19906 defines criteria regarding material properties under Arctic conditions as a general statement without concrete requirements. The offshore industry in the North Sea has used 14C as
the lowest operational temperature and little experience below this temperature is available.
A brief summary of different codes including International (ISO), American (API), Canadian (CSA),
Russian (SNIP and VSN), Norwegian (NPD and NORSOK), as well as Class Society (DNV and Lloyds)
is given in Annex 1 of this report. The CSA and ISO Standards have many similarities with care taken in
writing the ISO standard to consider methods used in the SNiP code.
An example code comparison using illustrative calculations according to Masterson and Tibbo (2011)
is now provided. The comparison is made for deterministic ice loads or actions calculated using methods
contained in ISO 19906 as well as those contained in API (American Petroleum Institute) RP2N (1995),
CSA (Canadian Standards Association) S471-04 (2004), and Russias SNiP 2.06.04-82* (1996). Scenarios considered are given in Table 6. Resultant loads are illustrated in Table 7. Specific details for the calculations can be found in Masterson and Tibbo (2011). It can be seen that the deterministic loads are quite
comparable. API loads for Scenario 3 do not consider aspect ratio and reduction in pressure on wider
structures as well as reduction in pressure for increasing ice thickness. SNiP load for Scenario 3 is considerably larger due to high-specified global pressure for MY ice.
2.2.2
First principles
Methods to determine design loads include probabilistic approaches, which offer a rational method to
obtain realistic designs. The analyses are rarely tractable in closed formalthough parts may beso that
Monte Carlo methods are generally used. Probabilistic methods are advocated in the ISO 19906 Standard,
which was calibrated to a set of safety targets. Software to obtain EL and AL is exemplified by
C-COREs Iceberg Load Software (ILS) and Sea Ice Load Software (SILS) (C-CORE, 2012). Figure 14
shows a general flow chart for a probabilistic analysis.
Table 6. Example offshore code comparison according to Masterson and Tibbo (2011).
Scenario
1
2
3
4
Description
A level sheet of ice of 1.2 m thickness interacting with a fixed offshore structure in mid-winter.
Assume that the ice is moving at a rate of 0.2 m/s, and the structure is vertical-sided with a width of
100 m and in deep water.
A first-year ridge of total thickness 10 m interacting with the same structure as Scenario 1.
Assume a keel-to-sail ratio of 4.4, a consolidated layer thickness of 1.5 m, and a width of 23m. Assume that the ridge is imbedded in an ice sheet with the same characteristics of Scenario 1.
Multi-year FloeA large drifting multi-year ice floe, approximately 1 km in diameter impacts the
structure at an impact speed of 0.5 m/s. The floe has thickness of 6 m, and an average temperature
of 5C. The floe has some roughness but no significant ridges.
A level first-year ice sheet of thickness 1.5 m surrounds the structure for a distance of 50 km. The ice is
level with no appreciable ridges or roughness. A wind gradually increases from 0 m/s to 25 m/s over a
period of 12 hours. Ice velocity increases over the same period and reaches a maximum value of 0.05
m/s. Assume that there is no ad freeze at the beginning of the event. A conical-shaped structure with a
45 slope lies offshore in Arctic waters. The width of the structure is 50 m at the waterline. Assume that
it is a perfect cone and that it has a low friction coating.
Table 7. Summary of ice loads in MN according to Masterson and Tibbo (2011).

Scenario
1
2
3
ISO (ELIE 10-2)

156
201
597
API
180
244
900
CSA
170
223
540
SNiP
199
299 to 398
1166
785
Design for ice loads requires consideration of the following.

1. Global loads, representing the total force applied to the structure, and
2. Local loads, representing the force (or pressure) on particular areas of structural importance such as
the plate between frames or other assemblies of structural elements that are important in design.
It is important to have clear definitions for these two areas associated with the two design load situations.
The global interaction area (also termed the nominal interaction area) is the area determined by the projection of the structure onto the original shape of the ice feature, without any reduction of the area for
spalls and fractures that take place during the interaction (see Figure 15). The global interaction area can
be determined from the shape of the iceberg and the shape of the structure. Within this area, there will be
areas that carry little or no pressure, as well as the high-pressure regions. When analysing data to formulate relationships for global load estimation, the analysis must also be done in terms of the global interaction area as just defined.
Figure 14. General flow chart for probabilistic analysis (DNV_Canada, 1986).
For design purposes, one needs to consider the local design area, which is the area of part of the structure,
for example the plate between frames or a panel that is under consideration in design. This is a fixed area
on the structure. The area might traverse through an ice feature during an interaction. Figure 20 shows the
concept. One can think of the global area as being fixed to the ice, but within the structure face, and the
local area as being fixed to the structure.
Figure 15. Global and local area definitions for iceberg impacts (Jordaan et al., 2005).
Figure 16 illustrates the Monte Carlo approach to analysis for iceberg impacts. The determination of iceberg design loads requires the following inputs for the simulations:
size, velocity and shape of the icebergs;

concurrent sea state and associated hydrodynamic effects on iceberg motion;
eccentricity of loading to account for oblique impacts;
786

global ice pressures developed on the basis of pressure-area relationships (derived from analysis of
ship rams into hard, multiyear ice); and
local ice pressures associated with the design loads (a function of the duration of individual impact
events and the frequency with which they occur).
The full analysis is quite complex and only a flavour of the methodology can be given here.
Inputs and models for the software including environmental and iceberg population data are summarized. The inputs required for the model are presented in Table 8, while Table 9 describes the various submodels contained. Ice environment factors considered for this example include the areal density of icebergs and the distributions of iceberg size, iceberg drift speed and sea state. The areal densities of icebergs have been calculated using data obtained from the International Ice Patrol (IIP) for the years 1984
to present. The iceberg drift speed model is based data collected during the mid 1980s along with more
recent data collected by PAL Environmental Services.
Figure 16. Methodology for iceberg impacts (Stuckey et al., 2008).
Consideration of iceberg detection, physical management and disconnection (when applicable) can be a
basis for revision and possible reduction of ice loads; see Figure 17. Detection capabilities are based on
an interpretation of enhanced marine radar capabilities in the vicinity of the structure. In practice, other
methods such as aerial reconnaissance, HF radar, and satellite-based radar and ship observations will be
used. Figure 18 shows a typical input. Physical management capability is based on past towing experience
on the Grand Banks and offshore Labrador.
The iceberg impact loads include consideration of wave-induced iceberg motions, mass, shape, eccentricity and strength. A random pressure-area relationship is an option for the global pressures and results
can be compared with results using a constant ice pressure. Local pressures are based on a detailed analysis of ice pressure distributions from ship impacts.
A basic parameter guiding several aspects of the analysis is the iceberg length, and several other parameters are linked to this. A large amount of analysis has been made for each of the required parameters; details
are not given here. Figure 19 gives an illustration of one of the inputs into analysis of iceberg shape.
Figure 20 illustrates global load estimates at 10-4 exceedance probabilities for design accounting for iceberg encounter rate; note that the exceedance probability on the vertical axis pertains to the parent distribution, i.e. without consideration of encounter rate. Figure 21 shows a typical result for local pressure.
Table 8. Outline of inputs and probability distributions.
Quantity
Iceberg drift velocity, VD

Sea state, HS
Iceberg waterline length, L
Draft, D
Mass, M
Added mass, Ma
Iceberg shape
Contact point
Areal density, r
Basis
Gamma distribution ( ,s )
Empirical CDF based on data
Iceberg measurements, 2 combined (weighted) exponential distributions
Empirical relationship with associated uncertainty
Empirical relationship with associated uncertainty
Empirical relationship based on iceberg length, width, draft, and water
depth
Based on CA = modified lognormal distribution with (,s ) and
DA = f(CA)
Based on iceberg profiles and structural configuration
Mean value based on iceberg sightings
An outline of progress for the design of structures in the presence of ice will be given now. Figure 1 illustrates the areas in the northern hemisphere where such calculations might be made. We shall first discuss
fixed offshore structures. There are various structure types that can be considered, e.g. Figure 22. A flow
chart for Monte Carlo analysis in the case of interaction of a structure with vertical walls with multiyear
787
ice is given in Figure 23. The area definitions for vertical structures, Figure 24, are an extension of those
previously given for iceberg interactions as shown in Figure 15.
In C-COREs SILS, consideration is also given to sloping structures in Table 10. There is considerable
uncertainty surrounding loads on sloping structures as a result of lack of full-scale measurements. Analyses of FPSOs or FPUs in sea ice have received little attention, but have been carried out for concepts in
the Barents Sea. In practice this will need attention to ice management, which will certainly be required in
heavy ice conditions. Detection and disconnection are again aspects that need careful study.
Table 9. Models in the software.
Model
Basis
Encounter rate
f = p ( L + WS ) V where WS is the structure width
Area-penetration
A = CA DA
Global pressure
P is constant or random given as
Local pressure
DP is a normal distribution
Impact duration; Cumulative distribution given as
Eccentricity and resulting

iceberg rotation at impact
Impact velocity
Ice management
Impact force
P = CP A DP where CP is a lognormal distribution and
Fz ( z ) = exp exp ( z / a ) where z is the pressure, m is the exposure,

a = 1.25a 0.7 where a is area
Profiled icebergs, Inertial parameters, G factor calculated for each orientaitation, Independent of the other parameters (dependent on contact depth)
VD, L, HS and iceberg RAOs
Detection performance (based on L, HS, range R), Towing performance (based on L, HS,
time t)
Kinetic energy model based on models above
Icebergs Potentially Impacting Installation
Detect
Cant Detect
Impact
Manage
OK
Cant Manage
Disconnect
OK
Cant Disconnect
Impact
Figure 17. Event tree for analysis of effect of iceberg

management and FPSO disconnection on the probability
of an iceberg impacting a structure (Jordaan et al., 2014).
Figure 18. Hibernia, X band detection of a 50m Iceberg,

Single Scan (Jordaan et al., 2014).
Probability of Exceedence
0.1
Influence of
Encounter Rate
1 impact / 100 yrs

0.01
1 impact / 10 yrs
0.001
1 impact / 1 yr
0.0001
L 100 yr
L 10 yr
L 1 yr
Iceberg Design Load (10-4 exceedence)
Figure 19. Typical 3-D shape constructed from Figure 20. Illustration of influence of encounter rate on
contour data (to 1 m resolution) based on Dobrocky iceberg design loads at 10-4 exceedance probability. The
data set (Jordaan et al., 2014).
probability of exceedance on vertical axis refers to that of
the parent distribution, without considering the encounter
rate (Jordaan et al., 2014).
788
14
0.1 interactions per year
12
Pressure (MPa)
10
10- 4
8
10- 3
8.63(area)- 0.7
5.76(area)- 0.7
10- 2
2.88(area)- 0.7
2
0
0
Area (m )
Figure 21. Illustration of local pressure result

(Jordaan et al., 2014).
Figure 22. Some possible structure types.
Figure 23. Flow chart for probabilistic analysis of Figure 24. Area definitions for interaction with
multiyear ice interactions with structures with vertical faces (C-CORE, 2012).
vertical faces (Fuglem, 1998).
Table 10. Assumptions regarding general structure shape for ice load models available in SILS.
Ice conditions
Level ice and

consolidated layer
of first-year ridges
Structure
face
Vertical
Sloped
Unconsolidated layer
of first-year ridges
Multi-year ridges
Vertical and
sloped
Vertical
Sloped
2.3
General structure shape
Crushing models consider ice thickness and contact width, the structure shape
in plan is not explicitly taken into account except for development of contact
width.
Croasdale model is a 2D model with a 3D correction to account for limited
structure width. Strictly speaking the model is applicable for level ice moving perpendicular to a flat face.
Nevel model is applicable to conical structures
Combined Croasdale/Dolgopolov model considers only the thickness of

layer and the width of the structure.
Crushing models consider ice thickness and contact width, the structure shape
in plan is not explicitly taken into account except for development of contact
width.
Nevel and Wang models consider and idealized conical structure where the
ice rides up the front centerline of the structure.
Validation methods
Validation methods are of crucial importance for all analytical and numerical calculations conducted.
This is especially true for the assessment of ice-induced loads due to the complex nature of the ship-ice
interaction and due to the large variation in ice conditions and ice properties found in natural sea ice.
Therefore, this section presents recent examples on validation methods including full-scale and modelscale measurements and observations. In addition, Annex 2 of this report seeks to summaries large- and
full-scale experiments to obtain ice loads carried out to date.
Full-scale measurements are the most reliable method to validate theoretical models. Since the early
1970s large- and full-scale measurements have been carried out in various ice-covered waters of the
world, see also Annex 2. Often strain gauges are installed on hull structures and then by finite element
modelling or by physical calibration (see e.g. Suominen (2013)) the correspondence between the measured strains and ice-induced loads can be evaluated. Generally, the stochastic nature of ice-induced loads
789
causes one of the main challenges when analysing such measured ice loads. Further, the need to establish
a link between ice conditions, ship parameters, operation situations and these ice-induced loads exists.
Long-term data obtained over a number of years can provide a link between the stochastic nature and the
prevailing ice conditions (see e.g. Kujala (1994), Taylor et al. (2010), Jordaan et al. (2005)). However,
long-term measurements, i.e. over several winters, have been performed only in Baltic and Antarctic waters. Ice loads from various measurement campaigns have been collected as 10-minute maximum load
values in the EU SAFEICE Ice Load Database (Kujala et al., 2007). Local ice loads from impacts with ice
features have been studied using motion reference units that record the ships motions with 6 degrees of
freedom. Measurements have been performed on CCGS Terry Fox (see e.g. Ritch (1994), in Russia
(Krupina et al., 2009), Norway (Nyseth, 2006) and in the Baltic (Valkonen, 2013). Further, global loads
on ships from ice impacts are important for the design of the hull girder strength and for the design of the
liquid cargo or LNG storages for ice impact loads. Local loads are measured with a strain gauge panel on
board Canadian, US and Swedish icebreakers; see e.g. Ritch (1994). Long lasting fibre optic strain gauges
has proven to be reliable instrumentation on KV Svalbard and SA Agulhas II (Mejlnder-Larsen and
Nyseth (2007), Suominen (2013)). Concerning larger vessels, only one large shuttle tanker operating between Murmansk and the Varandei terminal in Russia has been instrumented with local ice load measurement systems (Choi et al., 2009). Iyerusalimskiy et al. (2011) published interim results from these
measurements.
Another challenge is the need to define the prevailing ice conditions. The ice conditions are typically
observed by visual observations and only lately new approaches are developed to measure the prevailing
ice conditions using e.g. Electromagnetic (EM) measurements and stereo cameras (Suominen et al.,
2014). Unfortunately, no sound approach has emerged yet to link the stochastic ice induced loads with
prevailing ice conditions in spite of extensive efforts done by analysing full scale data (see e.g. Suominen
(2014)).
Ice damage statistics are another important measure to validate the design load level, i.e. if damages
are too frequent the design load level may be increased or decreased if no damages occur. Unfortunately,
such damage statistics are mostly not publically available. In the Baltic Sea, two damage measurement
campaigns were carried out to validate the design load level found in the FSICR, see Kujala (1991) and
Hnninen (2004).
The analysis Automatic Identification System (AIS) data also allows for the evaluation of the ships
progression behaviour in various ice conditions. Kujala et al. (2014) studied the probability of ships to get
stuck in moving ice, which can cause compressive loads on the hull and subsequently damage. Figure 10
shows the speed profile of a bulk carrier (21355 DWT, ice class 1AS, engine power 9720 kW) navigating
in March 2011 in the Bay of Bothnia. The ship sailed 94 nm in 14 h due to the severe ice conditions,
causing her to ram ice features multiple timed and forcing her to idle in ice for three hours. The total recorded data contains significant variability in environmental conditions influencing the ships speed, as
depicted in Figure 25. Therefore, Kujala et al. (2014) utilized Bayesian-based statistical models to identify the probability of a ship to get stuck in ice to be 0.03.
Model-scale testing is used to evaluate the ship performance in various ice conditions. Latest research results indicate, however, that the model scale tests may also be used to investigate ice-induced
loads (Kujala and Arughadhoss, 2012). Therefore, I-Scan 210 tactile sensors were used to measure the
pressures induced by ice. The total dimension of the sensor sheet is 238 mm by 238 mm and the area of
one sensing element is 5.4 mm by 5.4 mm, corresponding however to very few measurement point
through the ice thickness. The sensing elements were arranged in 44 by 44 grids with a sample rate of
50 Hz. In addition to these measurements, the total load on the same area was measured by installing
force transducers on the model hull (Suominen, 2013). Thus, the spatial and temporal distribution of
the local pressure can be captured. However, great care is needed to scale this type of data to full-scale.
However, straightforward Newtonian scaling suggests a good comparison to full-scale data, see Figure
26, in spite of the underlying assumptions and uncertainties in this comparison. The daily maxima of
the measured ice loads between the years 1979 to 1985 form the basis of the statistical analysis of the
long-term data (Kujala and Vuorio, 1986). The maximum ice thickness in the Bay of Bothnia during
1985 varied from 1.07 to 1.10 m with a higher probability of ice ridges or rafts. Whereas the maximum
ice thickness during the ice tank measurements for MT Uikku was only between 0.63 and 0.79 m of
level ice. This explains why the measured load values of IB Sisu are higher than the MT Uikku load
values. In addition, it has to be noted that the measuring time in an ice tank is only a few hours. The
load level on-board MS Arcturus is somewhat higher than on-board MT Uikku. This is understandable
as the measuring period of MS Arcturus in full-scale was longer than the measuring period used for MT
Uikku in the model-scale. In addition, the full-scale ice conditions can naturally vary significantly
unlike the model ice conditions.
790
Figure 25. Time series of a bulk carrier navigating in varying ice conditions in the Bay of Bothnia on March 2011
(Kujala et al., 2014).
Figure 26. Comparison of model-scale test with MT Uikku with the full-scale data obtained on-board IB Sisu and
MS Arcturus. Maximum ice thickness in model-scale corresponded to 0.8 m at a speed of 8 kn (Kujala
and Arughadhoss, 2012).
The mechanical properties and the differences between model ice and full-scale ice was investigated by
Von Bock und Polach and Ehlers (2014) based on a new approach to simulate numerically the failure of
model-ice as a result of numerous model-ice property tests (Von Bock und Polach et al. (2013), Von Bock
und Polach and Ehlers (2013)). They conclude that the number of unknown phenomenon in the internal
mechanical processes of model-scale ice adds to the uncertainty for the scaling process from model-scale
to full-scale and suggest various aspects to be investigated prior to general conclusions.
3.
CASE 1: SHIP TRANSPORTATION IN ARCTIC WATERSTHE NSR
In this chapter, we seek to present the applicability of first principle-based methods to assess ice loads
relevant for the design of a ship operating along the NSR in compliance and comparison to the above
mentioned design methods as an alternative to be used for ships operating along the NSR.
Draught limitations along the NSR, e.g. in Sannikov Strait at New Siberian Islands used to be around
13 m, but latest measurements indicate that up to 17 m are feasible. The administration of the NSR defines three different ice condition categories (easy, medium, heavy) and ships are allowed sail independently or with icebreaker depending on their ice class. A description of ice conditions along the NSR is
given by Ragner (2000). Figure 27 shows areas along the NSR where ice can typically be observed during
the summer sailing season. Naturally, variations between different years exist and the shown ice massifs
might appear in different compositions each year. Currently, first year ice is the predominant ice type
along the NSR, while the southwest Kara Sea is known to have heavy ice conditions with ridged ice. A
description of historical ice conditions in the Arctic and along the NSR is given e.g. by Romanov (1995).
In the summer season (AugustOctober) most of the NSR has been ice-free in resent years, at least for a
short period of time. Ice typically remains the longest in the Vilkitsky Strait and in the Laptev Sea. The
summer sailing season is characterised by various amounts of melting ice along the route. The extent of
the Arctic summer sea ice has been decreasing in the last decades (Stroeve et al., 2012) and also the local
791
volume of Arctic sea ice has decreased (Laxon et al., 2013). Predictions of the future ice extent show even
further decrease (Overland and Wang, 2007). Besides ice, fog is typical phenomena in the summer season
in the Arctic and forces ships to reduce their speed due to the low visibility where also possible ice floes
are harder to spot. Further, fog affects also eventual search and rescue operations significantly. In the later
part of the sailing season the nights are getting longer in the Arctic and darkness along with snowfall reduce visibility at the same time the sea is freezing up.
Figure 27. General locations of summer ice massifs along the NSR. Source: Ragner (2000), Published with
permission from the original author Claes Lykke Ragner, FNI.
The example calculation procedure consists of 5 steps as indicated in Figure 28 and is derived from Ralph
and Jordaan (2013). They linked the number of expected interaction per year with a parent distribution to
estimate the extreme load for certain probability level. This methodology is further based on Jordaan
(2005). The procedure requires the definition of waypoints along the NSR route, see Figure 29, and the
corresponding ice conditions shown in Figure 30. The expected interaction must be assessed to further
utilize a parent distribution corresponding to the ice conditions to obtain the extreme load prediction.
STEP 1:
Route
defini on
STEP 2:
Ice condi on
along the route
STEP 3:
Expected rams
per year
STEP 4:
Selec on of
parent
distribu on
Figure 28. Steps of a probabilistic ice load assessment (derived from Ralph and Jordaan (2013)).
Figure 29. The NSR sailing route for this case study (Tns et al., 2015a).
STEP 5:
Load
predic on
792
Figure 30. Example ice concentration and ice type along the route for a week in February 2009 (Tns et al., 2015b).
The example ice conditions presented in Figure 29 are based on satellite images provided by the U.S National Ice service, which relevant weekly and bi-weekly data since 1978. In this example, weekly data ice
utilized consisting of concentration for the two most dominating ice types. For this case study, weekly
data for one year is utilized for the defined route as shown exemplary in Figure 12. Therefrom the distance travelled in specific conditions can be assessed where different ice types are assigned according to
their share in concentration. For this example, it is further assumed that the vessel is sailing for 5 round
trips along the route in one year.
The USCGC Polar Sea full-scale measurements performed in the Beaufort Sea in 1982 are used to describe vessel interactions in multi-year ice. The total number of 167 impacts was reordered in multi year
ice during the reference voyage with 3.2 events per hour. For the thick first-year ice full-scale measurements performed in North Chukchi are considered. This dataset consist of ice impact measurements combined with first-year and multi-year ice. In total 513 impacts were recorded with a frequency of 3.6 events
per hour, mainly in medium and thick first-year ice of 0.9 m to 1.8 m. Furthermore, the pressure-area
relationship based on the Polar Sea measurements in the North Bering Sea was used for the reference
loading in thin and medium first-year ice. The ice thickness during the recorded 8.2 impacts per hour was
approximately 0.15 m to 1.2 m. Consequently, these recorded impacts per hour together with the maximum USCGC Polar Sea speed in corresponding ice conditions results in the total number of expected
impacts for this case study, see Table 11. The illustrative calculations are based on the assumption that the
ship operates along the route all year round or 8760 hours. Floe size and ice concentration could influence
exposure, which is however not taken into account in this case study. This follows the statement by Ralph
and Jordaan (2013), that the global and local ship ice interaction process causes the loads to the hull and
vessel to which it has to comply. Thereby, the design relevant load is typically caused by the most severe
ice action encountered, which may be in order of severity: brash ice, single year ice, multi year ice and
ridges.
To predict the 100-year extreme ice load based on the obtained ice conditions along the NSR parent
distributions following full-scale measurements are utilized from Figure 8. Based on Ralph and Jordaan
(2013) the design load for a 100-year return period can be estimated as:
z100 = x0 + ln ( ln Fz ( ze )) + ln = x0 + [ 4.6 + ln ]
where represents the actual panel hits and can be calculates as = v r t / tk where the average impact
duration is, t, is assumed to be 2s, the proportion of true hits is set to 0.5, the exposure parameter x0 is
considered 0 and tk is the reference impact duration based on the parent distribution. The resulting extreme load prediction is presented in Figure 31.
793
Table 11. Number of expected impacts for the case study.

Ice type
Operational speed [kn]
Duration [hours/year]
Open water
Medium first year ice
Thick first year ice
Multi-year ice
18
9
3
2
2964
1887
2969
1213
Impacts
[events/year]
16000
10000
4000
Figure 31. Predicted extreme load for the specified route (Tns et al., 2015b).
4.
CASE 2: FLOATING OFFSHORE STRUCTURES IN ARCTIC WATERS
In this chapter, we seek to present, as an example, the applicability of first principle-based methods to
obtain ice loads relevant for the design of a floating offshore structure in iceberg prone waters.
Icebergs occur in many areas of the arctic and subarctic, for example West Greenland, east of Baffin
Island and Labrador, on the Grand Banks, southeast Greenland, in the area neighbouring Svalbard, in the
Barents Sea, and many other areas in the Russian arctic, see also Figure 1. Determination of iceberg loads
for design of offshore facilities for exploration and especially for production is an important engineering
task.
Engineering design aims at an appropriate balance between safety and economy. The use of probabilistic methods offers a solution that assists in obtaining such a balance. The specification of iceberg loads is
guided by the (ISO, 19906) International Standard. The methodology used in the present study results in
load-exceedance curves, which can be used to determine design loads at a desired annual exceedance
probability. In ISO 19906 the Extreme Level Ice Event (ELIE) and the Abnormal Level Ice Event (ALIE)
for the design of an offshore platform are defined at annual exceedance probabilities of 102 and 104 respectively for L1 exposure as defined earlier in this report.
Ice loads have been modelled using Monte Carlo methods, which take into account the underlying
probabilistic distributions of the areal density of the ice feature (for example the number of icebergs per
10,000 km2), the size and mass of the features, their added mass, their velocity, eccentricity of the collision, forces from surrounding pack ice, compliance of the structure, and the strength of the ice. Previous
work focussed on the Grand Banks, where two floating production platforms are now operating, the Terra
Nova and Sea Rose FPSOs. Probabilistic methods were developed for these developments. Extension of
the methodology for use in other areas is of considerable interest. An example is suggested in Figure 32
The moderating influence of the North Atlantic Current results in conditions north of Norway and in the
region of Svalbard that are similar in many respects to those offshore Newfoundland.
794
The floating vessel considered in the analysis is illustrated in Figure 33. The determination of iceberg
design loads requires the following inputs for the simulations:
areal density of icebergs

size, velocity and shape of the icebergs;
concurrent sea state and associated hydrodynamic effects on iceberg motion;
eccentricity of loading to account for oblique impacts;
global ice pressures developed on the basis of pressure-area relationships (derived from analysis of
ship rams into hard, multiyear ice, or other relationships such as constant-pressure); and
local ice pressures associated with the design loads (also derived from analysis of ship rams into
hard, multiyear ice, and a function of the duration of individual impact events and the frequency
with which they occur).
In addition to these, detection and management of icebergs, and possible disconnection of the floating
unit are modelled.
Models have been developed for iceberg management based on Canadian experience on the Grand
Banks. Three key elements of iceberg management include detection, towing and disconnection. Detection performance is based on special ice radar that is designed for small target detection in high seas.
The system processes multiple scans to minimize clutter and false targets. Performance is based on the
probability of detection (POD) of icebergs given iceberg size, sea state and range from the platform. Figure 18 illustrates a typical input.
Without any management, the encounter frequency is about 5 10-3 per annumless than the value at
which extreme-level design should be considered, but certainly greater than the value used in the abnormal-level case. Figure 34 shows a typical result of calculation of the mechanics of interaction. The dynamic analysis results in much reduced loads as compared to the values based on quasi-static analysis; for
example, the quasi-static 10-3 and 10-4 exceedance loads of 10 and 150 MN reduce to about 5 and 40
MN respectively (without management). Iceberg management reduces the value at the 10-4 annual exceedance level to 26 MN.
A comprehensive methodology has been developed for obtaining design loads due to iceberg impacts, which was illustrated here and can be found in the previous chapter of this report, the original
area of interest being the Grand Banks. In the present study, the methodology was applied to a region
with much reduced areal densities of icebergs, and a greater proportion of bergy bits. The ExtremeLevel loads were found to be zero, but significant loads and local pressures have been found at the
Abnormal-Level (104 annual exceedance probability). Arrival rates, global forces including the effects
of dynamics of iceberg and vessel, mooring loads and local loads have been determined using the
methodology.
Figure 32. Some geographical areas with approximate areal densities of icebergs (excluding bergy bits) per 104 km2
indicated (Jordaan et al., 2014).
795
Figure 33. Schematic of generic floating vessel used in study (Jordaan et al., 2014).
Figure 34. Results for floater inertial response including mooring stiffness. Mass = 1 106, V = 0.30 m/s (black
lineiceberg, and gray line mooring and floater) (Jordaan et al., 2014).
5.
FUTURE PERSPECTIVES AND CHALLENGES
New projects in the arctic region need to consider operational and technical aspects as well as social and
physiological aspects. Thus, prior to such project, these aspects must be assessed to ensure agreement of
all relevant stakeholders from day one. Therefore, the non-exclusive items presented in Table 12 shall be
considered.
Table 11 clearly indicates that each project in the arctic region must be approached with a tailor-made
solution, because it is not possible to copy one project from one location to another location, even if the
defining parameters appear to be rather equal. However, lessons learned from existing projects are invaluable to any new tailor-made project. When addressing items 1 to 5 from Table 11, no rules or proven
concepts are available and thus they have to be by an integrated consisting of at least psychologists, sociologists, operators, mariners, engineers and managers. The current report only contributes to item 6 and 7,
yet these important challenges are crucial to the overall success or projects in the arctic region.
Table 12. Non-exclusive items to be considered for new projects in arctic regions.
Item
1
2
3
4
5
6
7
Description
Involvement of 1st nations, their fears how they can be accounted for
Social and physical environment existing and/or to be created to ensure that people can work there efficiently and satisfied
Environmental and environment protection aspects to be considered and maintained
Identification of the projects main purpose and how it will be achieved
Operational aspects to be considered and to be achieved
Technical challenges of the project and the availability of adequate technology
Assessment of the project risk and evaluation of the risk mitigation measures
796
During the last decades great efforts have been put into the human safety and the protection of the environment wherefore both risk and reliability analysis are important decision support means. The development of structural reliability analysis in general started over 35 years ago as a new discipline in engineering, after probabilistic theory linked reliability to rules. Standardized methods, guidelines, and related
software tools nowadays support structural reliability analysis. The basis for the methods and terminology
may be found in (ISO, 19906). However, since about 1980, risk analysis has been mandatory within the
offshore industry to identify risks, implement risk reducing measures, and to alert operators to the risks
connected with their activities. Integration of risk and reliability analysis into the design process leads to
risk-based design and Bainbridge et al. (2004), Moan et al. (2006) place this into the context of ship design. Furthermore, Jordaan et al. (1987) and Ralph and Jordaan (2013) developed an approach for a probabilistic design criterion for arctic shipping.
Reliability for ice-going vessels concerns the assessment of uncertainties related to ice-induced service
loading, i.e. structural reliability analysis (see e.g. Kujala (1991), Kaldasaun and Kujala (2011)). The
assessment of risk in ice refers to the consequences of accidental ice-induced loading, which leads to a
risk-based methodology. Therein, all analysis shall be performed in a consistent way, by explicitly aiming
at a common and transparent general criteria and analysis methods. This approach is in line with Goal
Based Standards (IMO, 2014a).
Although the risk-based approaches are already established quite well in the maritime industry and can
be found for passenger ships, and some examples of cargo ships, e.g. MSC 76/INF.15, MSC 82/23/3,
there is a difference between risk assessment in the oil and gas industry and in the maritime industry. In
the oil and gas industry each facility is assessed separately while in the maritime industry it primarily is
done industry wide to establish rules and regulations. However, the approval of a ship can only be done
for agreed operational and ice conditions, thus the imminent possibility that the ship operates outside
these agreed conditions, due to the lack of knowledge of arctic wide conditions, may result in unwanted
consequences.
A way forward is a mission-based treatment of the design relevant features and their identification to
ensure safe arctic operations and transport.
The mission-based context is introduced here to pinpoint the need to consider site and route specific
conditions in the design philosophy and widens the risk-based design scope from structural to operational
aspects. Mission-based design shall include: definition of hazard scenarios, their occurrence probability
and consequences, see Figure 35. For arctic operations, the definition of all three elements is challenging.
The definition of hazard scenarios has to include the possible variation in ice conditions and operation
principles in the arctic region. For shipping we can have independent navigation or navigation with
icebreaker assistance e.g. in level ice, ice floes and ridged ice with various amount of first year and multiyear ice features. In addition, the possibility of moving ice has to be included as well as dynamic seasonal
variations in ice conditions and long-term ice condition changes due to climate change.
Altimetry
Modeling
Observations
Predictions for the sea ice cover

Concentration, Thickness, Movement...
Structure-ice interaction
Risk control by design
Ice load models

Structural limit states
First principle-based design methods
Occurrence probability
Figure 35. Mission-based design with risk control by design using first principle methods (Ehlers et al., 2014b).
797
The first task for mission-based is to define the most relevant ice condition in a format suitable to design for the operational scenario in question. Thereafter first principle tools shall be used to evaluate the
ice induced loads and their occurrence probabilities. Based on this, the occurrence probabilities of various
extreme and accidental limit states can be determined by applying structural analysis methodologies. The
possible consequences of each limit state, including the environmental effects, shall than be evaluated so
as the conceptual design in question. Figure 36 summarizes this process. Therein, the mission-based
treatment can be achieved through a consistent link between these elements and the analysis of their effects on the conceptual design phase.
Figure 36. The basic elements and the process needed for a mission-based design approach (Ehlers et al., 2014b).
The mission-based design of ships sailing in ice-covered waters can come possible when the uncertainties
related to the service, accidental actions and the ship operations are assessed. Ideally, we seek to identify
the exact safety margin using the accurately assessed uncertainty related to the ice-induced load. As a
result, there would be no need to speculate if a design load suffices, since the compliance to the safety
level for a given environmental condition would be clearly identified. Thus the risk level for different
designs must be quantified. In fact, risk shall be used to measure the safety performance. Therefore, with
having safety measurable, it is possible to effectively optimize the ship design by introducing risk minimization as a new objective along standard objectives. In other words, one additional constraint enters the
design optimization as follows:
Rdesign Racceptable
where Rdesign is the risk of the considered ship or system and Racceptable the acceptable risk. Furthermore, the
trend is to apply the ALARP (As Low As Reasonably Possible) principle, which relates the acceptable
probability of fatality, pollution and economic loss as a function of the magnitude of the consequences to
the costs of reducing the risk. However, in order to assess the risk from accidental ice-induced loading,
the consequences, i.e. the structural response and strength, needs to be obtained. The evaluation of the
consequences of alternative accident scenarios has to cover the process from accident occurrence to its
consequences, to ecosystem response and recovery including related uncertainties. Assessing the consequences to a vast number of possible accidental scenarios requires efficient computational methodologies
to optimize the structural layout to minimize this risk. However, in order to minimize the risk of a ship
subjected to accidental ice-induced loading, conflicts are often faced. For example, being incapable to
create a lightweight, safe and inexpensive ship might lead to conflicting objectives. Therefore multiobjective optimization is needed to foster proper treatment of global objective to minimize the risk, i.e.
maximize safety.
Concerning ship structures, multi-objective optimisation of ship structures can be found e.g. in Shi
(1992) and Parsons and Singer (2000) while Ehlers (2010) and Ehlers (2012) presented structural optimisation procedures for accidental loading suitable for computational risk-assessment.
5.1
Numerical simulations
Present design methods use empirical data to assess design pressures as a result of structure-iceinteraction. Therefore, this chapter presents recent developments in numerical simulations towards the
798
development of theoretical and first principle-based design pressure calculations. The primary challenges in structure-ice-interaction simulations concern the ability to simulate the phenomena found
therein, i.e. the damage and fracture process and the formation of high pressure zones as shown in Figure 15. Thus, it crucial to consider the basic behaviour of ice adequately, i.e. its visco-elastic behaviour,
creep and damage, rather than to violate it by using inappropriate commercially available analogies in
material modelling.
The predictions of ice failure, i.e. fracture; under various loading rates at temperatures around its
melting point represent a challenge when modelled. Thus, local ice failure includes transitions between
different phases. Impact induced ice failure typically involves compaction around the centre of the impact location, possible closed micro-cracks in the confided area and flaking of the free edges as a result
of tensile cracks. Depending on the impact speed ice behaves ductile with visco-elastic deformations
and damage during low loading rates and brittle during high loading rates and generally it is temperature dependent. Experimental data available concerning the microstructural behaviour of ice (creep,
constant-deformation-rate) can be found in Melanson et al. (1999), Meglis et al. (1999) and compressive behaviour of ice is available by Jordaan et al. (1997). Therein, pressure within high-pressure zones
of up to 70MPa-100MPa have been found as well as the fact that recrystallization to very small grains
and softening under pressure leads to extrusion resulting in crushed ice (layer formation). In addition,
this recrystallization without fracture can occur besides spall damage. Moore et al. (2013) presented
damage process and fracture process based on stress history for each element in the form of softening.
Furthermore, the multiple-scales involved in any structure ice interaction process shown in Figure 37
cause various numerical challenges including local contact issues and general scale effects. Thus the
scaling laws must cover a wide range of scale all the way from atomistic Korlie (2007) up to the km
range. Therein, tension fracture scaling appears to apply in the 0.1100 m range (Mulmule and
Dempsey, 1998). In the case of compressive failure size effects are significant according to Blanchet
(1998). In smaller laboratory scale the importance of inhomogeneities and polycrystallinity have been
established to play important roles, but no approaches on how to transfer this to larger scales exists.
Furthermore, the size effect at smaller scales seems to be clearly linked to the combination of several
possible failure modes (tension or shear fracture, shear localization, evolution of damage, etc.). Consequently, Jordaan (2009) presents scale effects indicating that local contact phenomena can changes the
local ice behaviour up to a million times.
Another topic is the different behaviour of fresh and sea water ice. There have been developed constitutive models for fresh water ice; however, these do not apply to saline water ice due to differences in microstructure (Schapery, 1997). Specific physically based models, applying dislocation and grain-boundary
relaxation, have also been developed to describe inelastic deformation in sea water ice (Cole et al., 1998).
Furthermore, it has been shown that strain-rate and time effects are important in fracture of ice, and constitutive models addressing high strain rate have been developed (Sain and Narasimhan, 2011).
Scale: 1x1 nmi
Ice mechanics
Bending
failure
Ice
velocity
Ice-Structure
interaction
domain
Steel mechanics
Large scale
Ship velocity
Basic ice properties
Intermediate scale
Grain structure
Fracture /
chipping
Inelastic deformation
Stress-state
Strain rate
Temperature
Crushing
Shear
failure
Fracture energy
Void growth
Pre-existing
damage
Materials scale
Structural
modeling scale
Crack
Plasticity theories
Cohesive zone
properties
XFEM
Ductile or brittle?
Fracture toughness
Void growth
Detailed plasticity behaviour
Weld properties
Figure 37. Example range of scales involved in structure ice interaction processes (Ehlers et al., 2014b).
Jordaan (1986) analysed floating masses, such as multi-year floes and their interaction with offshore
structures and discusses the need for efficient numerical methods to produce quantitative results in the
799
future. In general, ice may be discretized by finite elements and failure treated by element erosion or
deletion. If the mesh is fine, then erosion may approximate initial failure in the continuum representation of the ice well (Yu et al., 2007). Explicit modelling of the anisotropy and damage, e.g. distributed
ice grains, air and water, has been carried out by Von Bock und Polach and Ehlers (2013) for model
scale ice, which is however different from natural ice as discussed in Von Bock und Polach and Ehlers
(2014).
Besides ALE discretization approaches, which proved effective for relevant short cracks being in
plain strain condition according to Schapery (1997), while the simulation of rate dependent long cracks
still presents a challenge. For qualitative analysis of brittle fragmentation Polojrvi and Tuhkuri (2008)
presented a BEM-based specialized simulation tool, which is suitable for 2D beam-based models with
cohesive cracks for rubble formation, but not for local ice mechanics. Rubble formation was simulated
using 2D combined FEMDEM by Paavilainen et al. (2011), which concerns the rubble build up and
internal contact. Furthermore, Ranta et al. (2014) concluded that material parameters for ice are not
sufficient to determine the load level of ice interaction against an inclined wall. Consequently, both
numerical models and experiments are needed to understand the underlying processes, see Polojrvi
and Tuhkuri (2008). One rapidly developing tool to analyse these processes is the integrated finite discrete (DEM) and finite-element method (FEM), in which both the ice breaking process and loose ice
pieces can be modelled, see Paavilainen et al. (2009). Heinonen (2004) and Liferov (2005) present
pseudo-discrete models in the form of continuum models for ice rubble formation. Lau et al. (2011)
modelled free-floating level ice with three-dimensional plate bending elements using a DEM-Code for
ice-related problems. The general shortcoming of these approaches is their utilization of discrete energies for the involved damage modes, which are not known for the discretization scale, hence the simulations can only be calibrated using experiments and thus their general applicability may be achieved.
Sawamura et al. (2008) study the dynamic bending behaviour of a floating ice-sheet subjected to the
dynamic force by using the explicit FE-method. Lately, Su et al. (2011) applied a numerical method to
simulate a ship moving forward in uniform or randomly varying ice conditions. Furthermore, Su et al.
(2011) investigated both global and local ice loads on ship hulls by simulating a full-scale icebreaking
trial considering the interdependence between the ice load and the ships motion as well as the 3D-rigid
body equations of surge, sway and yaw. As a result, their simulations comply with the overall resistance of the trial in various ice conditions. However, their model does not consider localized ice failure
directly, assumes a certain ice fracture pattern and due to its calibration based on large-scale measurements it may not be applicable to new ships in general. In conclusion, the above-mentioned simulations
may contribute to a better understanding of the processes involved in ship-ice interaction. A step ahead
would be to consider localized pressures and the response of the ship structure, specifically the steel
and weldments under sub-zero temperature.
In general, failure of steels, either due to plastic collapse, ductile fracture or brittle fracture, has been
studied heavily in the offshore, ship, and automotive industries. However, how these phenomena change
and interact at sub-zero temperatures is still subject to large uncertainties. There is a common belief that
although the yield stress and tensile strength increases with decreasing temperature the fracture strain
decreases leading to poorer impact resistance at lower temperatures. However, recent results, combing
experimental tensile data and numerical modelling, strongly indicate that this is not always the case
(Ehlers and stby, 2012). Furthermore, beside the general tensile and fracture properties of steels as a
function of temperature, the performance of weldments plays an important role. Laser-based welding is
much used in the ship industry, however little systematic work has been done regarding the sub-zero temperatures behaviour. Also the possible transition from ductile to brittle behaviour in steels with reducing
temperature has not been studied with regards to ship-ice interaction. This presents the backdrop for the
need to further investigate and obtain understanding of the performance of steel and associated weldments, and underlines that the linking of this new physical understanding with the general structural response of ships will be essential to achieve in the future.
5.2
Ice induced fatigue
Fatigue damage is a common problem for structures subjected to cyclic loading, such as ships and offshore structures exposed to wave and wind loads. Ice-going ships experience additional loads due to
crushing of ice and ramming of ice floes. According to Zhang et al. (2011) fatigue cracks were found in
both ice classed and non-ice classed ships after an averaged time of 13.0 years and 12.7 years, respectively. This indicates that structural design according to fatigue limit state (FLS) is equally important for
both ship types. Moreover, most cracks were found in the forward region of ice-going ships, while in nonice classes ships cracks are more often found in the mid region of a ship, see Figure 38.
800
Figure 38. Damage distribution of side shell structures of ice-going and non ice-going vessels (Zhang et al., 2011).
Fatigue assessment of structural details of ships or offshore structures is usually performed by means of a
local stress approach. Herein, the local stress acting at a position at which fatigue crack growth might
initiate is used to estimate the life by inserting it into a SN design curve of, i.e., a classification society.
However, the difficulty lies with the definition of an appropriate load scenario, especially since ice conditions vary significantly. For this purpose an example FLS assessment study for an ice-going ship will now
be presented.
The reference ship is MT Uikku is a double hull ice breaking tanker, which is currently transporting oil from an oil terminal in the Ob Bay to a FSO in the Kola Bay near Murmansk, with an
approximate distance of 2170 km (Bambulyak and Frantzen, 2009). In order to find a load spectra for
this route a Monte Carlo based method suggest by Zhang et al. (2011) may be used to estimate the
frequency of impacts due to crushing of level ice per travelled meter. The local stress acting at a
critical weld detail in the ice belt of the ship can then be found by applying classic beam theory assuming the ice load to be acting at the mid span of the frame. Based on the so obtained local stress it
is possible to estimate the lifetime of the transverse stiffener in the forward region of MT Uikku by
means of linear damage calculation. Using, i.e., FSICR to obtain the stiffener scantlings this procedure results in an expected fatigue damage of 0.0241, which corresponds to a lifetime estimate of
41.6 years. However, we must keep in mind, that this estimate is only based on the damage accumulated in ice-covered areas. According to Zhang et al. (2011) it is feasible to assume that a ship will
experience half the damage in open water and half the damage due to ice impacts. A reasonable estimate of the lifetime would therefore be equal to 20.8 years, which is nearly half the time MT Uikku
is already in service. However, this is example is just an estimate using the minimum scantling requirements from FSICR. Given a 7% larger section modulus is installed in the reference ship, the ice
load based lifetime estimate increases to 54.1 years.
For narrow fixed structure, the dynamic ice-structure interaction known as frequency lock-in can occur
when level ice acts continuously on a vertical structure at a moderate ice speed. This phenomenon may
lead to fatigue. Based on field experience, structures with an Eigen frequency in the range of 0.4 Hz to 10
Hz have experienced self-excited vibrations, if their total structural damping was lower than about 3%.
Figure 39 illustrates the three primary modes of interaction in terms of ice force, F(t), and the corresponding displacement, u(t), as measured in full-scale structures in the Bohai Sea. These measurements suggest
that a conical waterline geometry of the structure reduces the magnitude of ice-induced vibrations.
a) Intermittent ice crushing
b) Frequency lock-in
c) Continuous brittle crushing
Figure 39. Modes of time-varying action due to ice crushing and corresponding dynamic component of structure
response (ISO, 19906).
6.
801
SUMMARY AND RECOMMENDATIONS
This report presented current design methods for ships and offshore structures followed by first principlebased pressure and occurrence determination methods to obtain design ice loads. Further, it was shown
that these first principle-based methods can contribute to the design of structures beyond the current
minimum requirements as well as for new operational areas in the arctic region and thus to the further
development of the present design methods.
The second chapter presented the current developments for ships and offshore structures as well as
the underlying rules and regulations. For ships, the essential rules are: the FSICR for first year ice, the
IACS polar classes and the RMRS covering also multi-year ice, besides the IMO Polar Code. The corresponding design ice loads are based on experience from specific types and sized vessels operating in
corresponding ice conditions. Offshore rules are based on target safety levels which than translate into
design ice loads. Their underlying design load, load combination and structural design considerations
and assumptions are presented in depth in the Annex of this report. A rational way forward and a possibility to analyse existing design ice pressures is presented based on probabilistic methods. Here extreme design events are identified as well as the model uncertainty. Additionally, different limit states
for ships and offshore rules are provided. Ship rules typically use some kind of elastic limit state and
allow for a certain amount of plastic deformation, offshore limit states comprise elastic, plastic, ultimate and serviceability, fatigue and accidental limit states, also corresponding to ELIE and ALIE and
may also use Load and Resistance Factor Design and probabilistic analysis. The performance based
design approach of ISO19903 is big step forward. Target reliability levels are established based on
consequence of damage including risk of loss of life and environmental damage. Subsequent designs
must satisfy the established targets directly. The approach considers both Extreme Level (10-2 annual
exceedance probability) and Abnormal Level (10-4 annual exceedance probability) ice loads events. For
structural design, these events would correspond with elasti limited plastic designs and ultimate limit
state plastic designs respectively. This robust approach to design gives the designer a clear understanding of the performance of the system for certain loading conditions. Mode of failure should also be
considered.
The design load obtained with the probabilistic approach presented in the NSR case study confirms that
multi-year ice results in the design driving load even if it is only encountered at a fraction of the route.
Furthermore, the comparison to IACS polar code confirms the general suitability of the approach to define ice loads on a route, respectively mission, basis. In other words, the presented approach allows for
the identification of the design driving conditions, which is also presented for the floating offshore structure case.
In conclusion, the report motivates for a more mission-based design methodology in addition to the
present design methods using the presented first principle-based methods. The contents of the missionbased design are presented and an outlook to possible future contents of such design approach for arctic
ships and offshore structures is presented. In addition, the current developments in numerical methods are
presented, which may be used to understand underlying phenomena better and to simulate processes.
Finally, an example on how to assess fatigue damage for structural details was presented.
In view of future research needs, it can be noted that more and more ships are going to operate in the
marginal ice zone or close to the ice edge. In those areas, small ice floes can be accelerated by waves and
reach substantial impacts speeds when the ship is advancing with relatively high speeds. The resulting
damage of such collision event can be extensive and similar approaches as for bergy bits may be applied
(Fuglem, 1998). Thus, it is advisable to consider ice in the presence of waves as a possible design scenario for the bow structures. Furthermore, investigation for stern ice loads on vessels, especially DAS, as
well as load in compressive ice should be investigated. As a result of the high lateral thrust from azimuthing propulsion systems, ice loads on the aft shoulder might be larger than suggested by the present
design methods. Valkonen et al. (2007) observed lateral movement of the ship in model scale. However,
the stern of a DAS may experience similar loading as a regular ice strengthened bow in forward motion.
Nevertheless, there is no publically available information on the ice loads on the aft of a DAS. Further,
the mid-ship section can experience significant ice loads when the ship gets stuck or operates in a compressive ice field, see, e.g., Suominen and Kujala (2012). These aspects, among others, generally indicate
the need to investigate the influence from the type of operations in ice on the design considering also
human factors. In view of offshore operations, ice management and the loads expected from unmanaged,
managed and over-managed ice should be investigated as well as disconnection operation and resulting
consequences. A clear definition of extreme load events and specific links between ice conditions and ice
induced load levels that could, i.e., lead to a transparent link between measured ice pressures and ship
design ice pressures would be very practical. The latter may also help to develop more reliable stochastic
models for ice-induced loads in various ice conditions and for selected routes. As for numerical simulations of ice, reliable numerical ice pressure simulations are needed to evaluate structures under target ice
802
conditions and to study load carrying mechanism under spatially and temporally varying loads. For the
evaluation of structures, the base material and weld compliance to arctic conditions, i.e., in terms of ductile to brittle transition behaviour should be in investigated.
ACKNOWLEDGMENTS
The committee would like to thank Roger Skjetne, NTNU, for his input to the DP section, Moritz Braun,
TUHH, for his input to the ice-induced fatigue chapter and Tnis Tns, NTNU, for his input to the NSR
case study.
REFERENCES
Aari. 2014. Arctic and Antarctic Research Institute [Online]. http://www.aari.nw.ru.
Aggarwal, R. & D'souza, R. 2011. Deepwater Arctic-Technical Challenges and Solutions. Arctic Technology
Conference. Houston, Texas, USA.
Akerarctic 2013. Aker Arctic brings heavy duty Oblique Ice Breaker to market. http://www.akerarctic. fi/news.htm:
Aker Arctic.
Baan, D. 2011. ASSEMBLY OF TURRET AND DISCONNECTABLE BUOY.
Bainbridge, J., Christensen, H., Hensel, W., Sames, P. C., Skjong, R., Sobrino, M. P., Strang, T. & Vassalos, D.
Design, Operation and Regulation for SafetySAFEDOR. PRADS2004, 2004 Travemnde, Germany.
Bambulyak, A. & Frantzen, B. 2009. Oil transport from the russian part of the barents region, status per January 2009.
Barentsobserver. 2013. Available: http://www.barentsobserver.com/en/arctic/three-new-nuclear-icebreakers-pipeline-14-11.
Bauduin. 2011. DISCONNECTABLE RISER-MOORING SYSTEM.
Bergstrm, M., Ehlers, S. & Erikstad, S. O. A framework for risk-based design of arctic ships. Submitted to
OMAE2015, 2015.
Blanchet, D. 1998. Ice loads from first year ice ridges and rubble fields. Canadian Journal of Civil Engineering, 25,
206219.
C-Core 2012. Sea Ice Loads Software (SILS)Version 1.6Theory Manual. C-CORE.
Carter, J. E., Daley, C., Fuglem, M., Jordaan, I. J., Keinonen, A., Revill, C., Butler, T., Muggeridge, M. & Zou, B.
1996. Maximum Bow Force for Arctic Shipping Pollution Prevention RegulationsPhase II. In: Memorial
University of Newfoundland, O. E. R. C. (ed.) Canadian Coast Guard Arctic Ship Safety.
Carter, J. E., Frederking, R., Jordaan, I. J., Milne, W. J., Nessim, M. A. & Brown, P. W. 1992. Review and
Verification of Proposals for the Revision of the Arctic Shipping Pollution Prevention Regulations. Memorial
University of Newfoundland Ocean Engineering Research Centre.
Chnl. 2013. Available: http://www.arctic-lio.com/node/207.
Choi, J., Park, G., Kim, Y., Jang, K., Park, S., Ha, M., Yu, H., Iyerusalimskiy, A. & John, J. S. 2009. Ice load
monitoring system for large Arctic shuttle tanker RINA. Royal Institution of Naval Architects International
ConferenceInternational Conference on Ship and Offshore Technology: Ice Class VesselsPapers.
Cole, D. M., Johnson, R. A. & Durell, G. D. 1998. The cyclic loading and creep response of alligned first-year sea
ice. Journal of Geophysical research, 103, 2175121758.
Daidola, J. C. & Sheinberg, R. Procedure for the Structural Design of Icebreakers and Other Ships Navigating in Ice.
1988. SNAME.
Dnv 2012. What is on the regulatory agenda? Recently adopted and coming rules and regulations from IMO, IACS
and DNV. http://www.dnv.ee/binaries/Estonia_presentation1_Regulatory_agenda_tcm172-531755.pdf.
Dnv_Canada 1986. Probability-based design criteria for ice loading on fixed offshore structures in the Beaufort Sea.
Final report, Vol. 1Study Description. Calgary, Alberta, Canada.
Dsme. 2014. Available: http://www.dsme.co.kr/.
Ehlers, S. 2010. A procedure to optimize ship side structures for crashworthiness. Proceedings of the Institution of
Mechanical Engineers, Part M: Journal of Engineering for the Maritime Environment, 224, 111.
Ehlers, S. 2012. High strength steel in crashworthy ship structures. Journal of Ship production and Design, 28, 19.
Ehlers, S., Benson, S. & Misirlis, K. Ultimate strength of an intact and damaged LNG vessel subjected to sub-zero
temperature. Collision and Grounding of Ships and Offshore Structures, 2013 Trondheim. CRC Press, 289296.
Ehlers, S., Erceg, B., Jordaan, I. & Taylor, R. Structural analysis under ice loads for ships operating in arctic waters.
MARTECH2013, 2014a.
Ehlers, S., Kujala, P., Veitch, B., Khan, F. & Vanhatalo, J. Scenario based risk management for arctic shipping and
operations. Proceedings of the ASME 2014, 33rd International Conference on Ocean, Offshore and Arctic
Engineering OMAE2014, 2014b San Francisco, California, USA. ASME.
Ehlers, S. & stby, E. 2012. Increased crashworthiness due to arctic conditionsThe influence of sub-zero
temperature. Marine Structures, 28, 86100.
Frederking, R. & Sudom, D. Local Ice Pressure Distributions during 1990 Hobson's Choice Ice Island Multi-Year
Ice Indentation Tests. Proceedings 19th IAHR Symposium on Ice, 2008 Vancouver, B.C., Canada. 815827.
Fuglem, M. Design Load Calculations for Iceberg impacts. Proceedings of the Eighth International Offshore and
Polar Engineering Conference, 1998 Montreal, Canada.
Hnninen, S. 2004. Incidents and Accidents in Winter Navigation in the Baltic Sea, Winter 20022003. Winter
Navigation Research Board.
803
Heinonen, J. 2004. Constitutive modeling of ice rubble in first-year ridge keel. Doctoral degree, TKK.
Ilo. 2014. Mission and objectives [Online]. Available: http://www.ilo.org/global/about-the-ilo/mission-andobjectives/lang--en/index.htm.
Imo 2014a. Goal based New Ship Construction Standards.
Imo 2014b. Implementation, Control and Coordination.
Imo 2014c. Port State Control.
Iso 19906. Petroleum and natural gas industriesArctic offshore structures. ISO.
Iyerusalimskiy, A., Choi, J., Park, G., Kim, Y., Yu, H. & St.John, J. The interim results of the long-term ice loads
monitoring on the large arctic tanker. Proceedings of the 21st International Conference on Port and Ocean
Engineering under Arctic Conditions, POAC'11, 2011.
Jordaan, I. 2005. Decisions under uncertainty. Probabilistic Analysis for Engineering Decisions, Cambridge,
Cambridge University Press.
Jordaan, I. Ice crushing, damaged layers, and pressure-area relationships. POAC2009, 2009.
Jordaan, I., Li, C., Sudom, D., Stuckey, P. & Ralph, F. Principles for local and global ice design using pressure-area
relationships. POAC2005, 2005 Potsdam, N.Y., 375385.
Jordaan, I., Nessim, M., Ghoneim, G. & Murray, M. A Rational Approach to the Development of Probabilistic
Design Criteria for Arctic Shipping. Proceedings of the 6th International Offshore Mechanics and Arctic
Engineering Symposium, 1987.
Jordaan, I., Stuckey, P., Liferov, P. & Ralph, F. Global and local iceberg loads for an arctic floater. OTC paper
24628. Arctic Technology Conference, 2014 Houston, Texas, USA.
Jordaan, I. J. Numerical and Finite Element Techniques in Calculation of Ice-Structure Interaction. IAHR Ice
Symposium, 1986 Iowa, USA.
Jordaan, I. J., Matskevitch, D. M. & Meglis, I. Disintegration of ice under fast compressive loading. Proceedings of
the Symposium on Inelasticity and Damage in Solids subject to Microstructural Change
Extended version in the International Journal of Fracture 1999, 97(14), 279300, 1997 Memorial University of
Newfoundland. 211231.
Juurmaa, K., Mattsson, T. & Wilkman, G. The Development of the New Double Acting Ships for Ice Operation.
Proceedings of the 16th International Conference on Port and Ocean Engineering under Arctic Conditions,
POAC01, 2001. 719726.
Kaldasaun, J. & Kujala, P. Risk-Based Approach for Structural Design of Ice-Strengthened Vessels Navigating in
the Baltic Sea. Proceedings of the 21st International Conference on Port and Ocean Engineering under Arctic
Conditions, 2011 Montral, Canada.
Korlie, M. S. 2007. 3D simulation of cracks and fractures in a molecular solid under stress and compression.
Computers and Mathematics, 54, 638650.
Krupina, N. A., Likhomanov, V. A., Chernov, A. V. & Gudoshnikov, Y. P. Full-scale ice impact study of icebreaker
Kapitan Nikolaev. General description. Proceedings of the International Offshore and Polar Engineering
Conference, 2009. 614620.
Kujala, P. 1991. Safety of ice-strengthened ships in the Baltic Sea. London: Royal Institution of Naval Architects.
Kujala, P. 1994. On the statistics of ice loads on ship hull in the Baltic. Dissertation, Helsinki University of
Technology.
Kujala, P. & Arughadhoss, S. 2012. Statistical analysis of ice crushing process. Cold Regions Science and
Technology, 70, 111.
Kujala, P. & Ehlers, S. A risk-based evaluation ice-strengthened hull structures. ICETECH2014, 2014.
Kujala, P., Montewka, J. & Filipovic', A. 2014. Risk management for Arctic shipping. SORTA. Baska, Croatia.
Kujala, P., Valkonen, J. & Suominen, M. Maximum ice induced loads on ships in the Baltic Sea. The 10th
International Symposium on Practical Design of Ships and Other Floating Structures, PRADS, 2007 Houston.
Kujala, P. & Vuorio, J. 1986. Results and statistical analysis of ice load measurements onboard icebreaker Sisu in
winters 1979 to 1985. Winter Navigation Research Board.
Lau, M., Lawrence, K. P. & Rothenburg 2011. Discrete element analysis of ice loads on ships and structures. Ships
and Offshore structures, 6, 311221.
Laxon, S. W., Giles, K. A., Ridout, A. L., Wingham, D. J., Willatt, R., Cullen, R., Kwok, R., Schweiger, A., Zhang,
J., Haas, C., Hendricks, S., Krishfield, R., Kurtz, N., Farrell, S. & Davidson, M. 2013. CryoSat-2 estimates of
Arctic sea ice thickness and volume. Geophysical Research Letters, 40, 16.
Liferov, P. 2005. Ice rubble behaviour and strength: Part II. Modelling. Cold Regions Science and Technology, 41,
153163.
Liferov, P., Le Marechal, G. & Albertini, M. M. Approach to Performance, Operability and Risk Assessment of the
Shtokman Floating Platform in Ice. Arctic Technology Conference, 2011.
Loenhout, V. 2012. MOORING SYSTEM.
Masterson, D. & Tibbo, S. Comparison of Ice Load Calculations using ISO 19906, CSA, API and SNIP.
Proceedings of the 21st International Conference on Port and Ocean Engineering under Arctic Conditions, 2011
Montreal, Canada.
Meglis, I. L., Melanson, P. M. & Jordaan, I. J. 1999. Microstructural change in ice: II. Creep behavior under triaxial
stress conditions. Journal of Glaciology, 45, 438448.
Mejlnder-Larsen, M. & Nyseth, H. 2007. Ice load monitoring RINA, Royal Institution of Naval Architects.
International Conference - Design and Construction of Vessels Operating in Low Temperature Environments
Papers.
804
Melanson, P., Meglis, I., Jordaan, I. J. & Stone, B. S. 1999. Microstructural change in ice: I. Constant deformationrate tests under triaxial stress conditions. Journal of Glaciology, 45, 417437.
Moan, T., Das, P., Friis-Hansen, P., Gu, X., Hovem, L., Parmentier, G., Shigemi, T. & Spencer, J. 2006. Reliability
based structural design and code development. 16th Int. Ship and Offshore Structures Congress. Southampton
University Press.
Moore, P. F., Jordaan, I. J. & Taylor, R. S. Explicit finite element analysis of compressive ice failure using damage
mechanics. Proceedings of the 22nd International Conference on Port and Ocean Engineering under Arctic
Conditions, 2013 Espoo, Finnland.
Mulmule, S. V. & Dempsey, J. P. 1998. A viscoelastic fictitious model for fracture of sea ice. Mechanics of timedependent materials, 1, 331356.
Niini, M., Kaganov, S. & Tustin, R. D. 2007. Development of Arctic double acting shuttle tankers for the
Prirazlomnoye project. TSCF 2007 Shipbuilders Meeting.
Nsr. 2013. NSR rulesRules of navigation in the Northern Sea Route water area [Online]. Available:
http://www.mintrans.ru/documents/detail.php?ELEMENT_ID=19481, English translation available at:
http://www.arctic-lio.com/docs/nsr/legislation/20130425185806en-Rules_unof.pdf.
Nsr. 2014. Northern Sea Route Administration Office [Online]. Available: http://asmp.morflot.ru/en/celi_funktsii.
Nyseth, H. 2006. Strain measurements on board KV Svalbard with respect to ice loading. Hvik, Norway: Det
Norske Veritas.
Overland, J. E. & Wang, M. 2007. Future regional Arctic Sea ice declines. Geophys. Res. Lett., 34.
Paavilainen, J., Tuhkuri, J. & Polojrvi, A. 2009. 2D combined finitediscrete element method to model multifracture of beam structures. Engineering Computations, 26, 578598.
Paavilainen, J., Tuhkuri, J. & Polojrvi, A. 2011. 2D numerical simulations of ice rubble formation process against
an inclined structure. Cold Regions Science and Technology, 68, 2034.
Parsons, M. G. & Singer, D. A Fuzzy Logic Agent for Design Team Communications and Negotiations.
COMPIT2000, 2000 Potsdam, Germany. 14.
Pedersen, R. A., Molnes, D. A., Stokkeland, L. S. & Ehlers, S. Structural optimisation for ice-strengthened vessels.
MARTECH2014, 2014.
Polojrvi, A. & Tuhkuri, J. 2008. 3D discrete numerical modelling of ridge keel punch through tests. Cold Regions
Popov, Faddeyev, Kheysin & Yakoviev 1967. Strength of Ships Sailing in Ice. US Army Technical Translation.
Ragner, C. 2000. Northern Sea Route Cargo Flows and InfrastructurePresent State and Future Potential. Lysaker,
Norway: FNI.
Ralph, F. & Jordaan, I. Probabilistic Methodology for Design of Arctic Ships. Proceedings of the 32nd International
Conference on Ocean, Offshore and Arctic Engineering (OMAE), 2013 Nantes, France.
Ranta, J., Tuhkuri, J., Paavilainen, J. & Polojrvi, A. Statistical reconstruction of peak ice load data based on 2D
combined finite-discrete element method simulations of ice interactions against inclined wall. Proceedings of the
22nd IAHR International Symposium on Ice, 2014 Singapore.
Riska, K. 1995. Models of ice-structure contact for engineering applications. Mechanics of Geomaterial Interfaces,
77103.
Riska, K. & Kmrinen, J. 2012. A Review of Ice Loading and the Evolution of the Finnish-Swedish Ice Class
Rules. SNAME Transactions.
Ritch, R. S. J., J. 1994. Ice load impact measurements on the CCGS Louis S. St. Laurent during the 1994 Arctic
Ocean crossingA trip report and data summary. Report to Fleet Systems, Canadian Coast Guard.
Rmrs 2010. Rules for the Classification and Construction of Sea-going ships.
Rmrs 2014. Russian Maritime Register of ShippingRules for the Classification Surveys of Ships in Service.
Romanov, I. P. 1995. Atlas of Ice and Snow of the Arctic Basin and Siberian Shelf Seas.
Rosa 2014. Rosatomflot.
Sain, T. & Narasimhan, R. 2011. Constitutive modeling of ice in the high strain rate regime. International Journal of
Solids and Structures, 48, 817827.
Sawamura, J., Riska, K. & Moan, T. Finite Element Analysis of Fluid-Ice Interaction during Ice Bending. 19th
IAHR International Symposium on Ice, 2008 Vancouver, B.C., Canada. 12.
Schapery, R. A. 1997. Linear elastic and viscoelastic deformation behaviour of ice. Journal of Cold Regions
Shi, W. B. 1992. In-Service Assesment of Ship Strctures: Effecto of General Corrosion on Ultimate Strength. RINA
Spring Meeting.
Stroeve, J. C., Kattsov, V., Barrett, A., Serreze, M., Pavlova, T., Holland, M. & Meier, W. N. 2012. Trends in Arctic
sea ice extent from CMIP5, CMIP3 and observations. Geophysical Research Letters, 39, 28.
Stuckey, P., Ralph, F. & Jordan, I. Iceberg Design Load Methodology. ICETECH08, 2008 Banff, Canada.
Su, B., Riska, K. & Moan, T. 2011. Numerical study of ice-induced loads on ship hulls. Marine Structures, 121.
Suominen, M., Karhunen, J., Bekker, P., Kujala, P., Elo, M., Von Bock, R.F., Polach, R., Endlund, H. & Saarinen,
S. Full-scale measurements on board PSRV S.A. Agulhas II in the Baltic Sea. Proceedings of the 22nd
International Conference on Port and Ocean Engineering under Arctic Conditions (POAC), 2013 Espoo,
Finnland.
Suominen, M. & Kujala, P. Ice Model Tests in Compressive Ice. Proceedings of the 21st IAHR International
Symposium on Ice, 2012 Dalian, China.
805
Suominen, M., Kulovesi, J., Lensu, M., Lehtiranta, J. & Kujala, P. A Comparison of Shipborne Methods for Ice
Thickness Determination. 22nd IAHR International Symposium on Ice, 2014 Singapore.
Suominen, M. K., P. 2014. Variation in short-term ice-induced load amplitudes on a ships hull and related
probability distributions. Cold Regions Science and Technology, 106, 131140.
Taylor, R. S., Jordaan, I. J., Li, C. & Sudom, D. 2010. Local design pressures for structures in ice: Analysis of fullscale data. Journal of Offshore Mechanics and Arctic Engineering, 132, 17.
Tns, T., Freeman, R., Ehlers, S. & Jordaan, I. Probabilistic design load method for the Northern Sea Route.
submitted to OMAE2015, 2015a.
Tns, T., Ralph, F., Ehlers, S. & Jordaan, I. Probabilistic Design Load Method for the Northern Sea Route.
submitted to OMAE2015, 2015b St. Johns, Canada.
Valkonen, J. Uncertainty of a methodology to estimate global ship loads during interaction events with ice features.
Proceedings of the 22th International Conference on Port and Ocean Engineering under Arctic Conditions,
POAC'13, 2013.
Valkonen, J., Izumiyama, K. & Kujala, P. Measuring of ice induced pressures and loads on ships in model scale.
Proceedings of the 10th International Symposium on Practical Design of Ships and Other Floating Structures,
PRADS2007, 2007. 12061213.
Van Der Nat, R. R. 2012. Developing a turret mooring system for arctic FPSO units. Arctic Technology Conference.
Houston, Texas, USA.
Von Bock Und Polach, R. F. & Ehlers, S. 2013. Model Ice as Material B: Numerical Modeling. Cold Regions
Science and Technology, 94.
Von Bock Und Polach, R. F. & Ehlers, S. On the scalability of model-scale experiments. Proceedings of the 33rd
International Conference on Ocean, Offshore and Arctic Engineering. OMAE 2014, 2014 San Francisco, CA, USA.
Von Bock Und Polach, R. F., Ehlers, S. & Erikstad, S. O. 2014. A Decision-based Design Approach for Ships
Operating in Open Water and Ice. Journal of Ship Production and Design, 30, 111.
Von Bock Und Polach, R. F., Ehlers, S. & Kujala, P. 2013. Model Ice as Material A: Experiments. Cold region
Science and Technology, 94.
Yu, B., Wu, W., Xu, N., Yue, Q. & Liu, S. Numerical Simulation of Dynamic Ice Forces on Conical Structures.
Proceedings of the 19th International Conference on Port and Ocean Engineering Under Arctic Conditions, 2007
Dalian, China. 277285.
Zhang, S., Bridges, R. & Tong, J. Fatigue design assessment of ship structures induced by ice loadingan
introduction to the shipright fda ice procedure. Proceedings International Offshore and Polar Engineering
Conference, ISOPE, 2011 Maui, USA.

710 SEPTEMBER 2015
CASCAIS, PORTUGAL
VOLUME 2
COMMITTEE V.6
ARCTIC TECHNOLOGY
Annex
COMMITTEE MANDATE
Concern for development of technology of particular relevance for the safety of ships and offshore
structures in Arctic regions and ice-covered waters. This includes the assessment of methods for
calculating loads from sea ice and icebergs, and mitigation of their effects. On this basis, principles
and methods for the safety design of ships and fixed and floating structures shall be considered.
Recommendations shall also be made regarding priorities for research programmes and efficient
implementation of new knowledge and tools.
COMMITTEE MEMBERS
Chairman:
S. Ehlers, Norway
F. Cheng, UK
I. Jordaan, Canada
W. Kuehnlein, Germany
P. Kujala, Finland
Y. Luo, China
F. Ralph, Canada
K. Riska, France
J. Sirkar, USA
Y.T. Oh, Korea
K. Terai, Japan
J. Valkonen, Norway
F. Ralph, Canada
KEYWORDS
Ship and offshore structures in ice, rule-based and first principal-based ice loads, mission-based
design
808
CONTENTS
1.
BRIEF OFFSHORE STRUCTURES CODE SUMMARIES
809
2.
FULL SCALE ICE LOAD MEASUREMENT CAMPAIGNS
813
3.
REFERENCES
816
1.
809
BRIEF OFFSHORE STRUCTURES CODE SUMMARIES
This section summaries the different offshore codes including International (ISO), American
(API), Canadian (CSA), Russian (SNIP and VSN), Norwegian (NPD and NORSOK), as well as
Class Society (DNV and LR) in a Table 1 to 14 (see also Allan et al. 2000). The CSA and ISO
Standards have many similarities with care taken in writing the ISO standard to consider methods
used in the SNiP code.
Table 1. CSA S471-04.
Design loads:
Performance based reliability approach to design as with ISO. Two key safety classes (1 and 2)
considered based on risk of loss of life and environmental damage. Annual reliability targets set
in each case
Probabilistic and deterministic methods.
Offshore environment data has considerable detail.
Loads based on full scale data and actions.
Full range of ice models and ice types considered.
Structural design:
Structural design based on Limit State methodload factors and material resistance factors used
to achieve safety targets.
Two limit states include Ultimate Limit State (two safety Classes depending on risk to personnel
and environment) and Serviceability Limit State (interruption to operations).
Loads and Load Combination:

Load categories include: dead (GD), deformation (GR), operational (Q), and Accidental (A) and
Environmental (Ef & Er for freqwind/wavesand rareiceberg/earthquake) loads.
Load combinations are illustrated in Table 10.
Table 2. ISO19906.
Design loads:
Performance based approach to designoverall safety targets set and guidance to achieve targets.
Target reliability levels reflect consequence of failure through loss of life, and severity of environmental loading considering Extreme Level and Abnormal Level Ice load Events (ELIE and ALIE).
Both probabilistic and deterministic calculations are considered.
Offshore environment has considerable detail.
Loads are based on full scale data and actions from monitoring of offshore structures not small scale
dataBeaufort Sea, Cook Inlet, Baltic Sea, Gulf of Bohai, Sea of Okhotsk.
Full range of modern ice load models.
Full range of ice conditions considered (FY level, FY ridges, MY floes, MY ridges, icebergs).
Structural design:
Structural design based on Limit State methodload factors and material resistance factors used to
achieve safety targets.
Limit states correspond to ELIE and ALIE allowing robust design.
limit states considered include Ultimate Limit State (plastic design without loss of life or environmental damage) and Serviceability Limit State (elastic design with limited plasticity where operations
may be interrupted), Fatigue Limit State and Accidental Limit State.

Where data are available, joint probability distributions of the principal (i.e. ELIE and ALIE) and relevant companion actions should be used to determine magnitude of combined EL or AL environmental
actions.
In absence of joint probability data, companion EL and AL factors are recommended depending on
whether companion actions are stochastically dependant or independent of the principal action
Principal and companion factors are given in Table 8.
Action factors and combinations are listed in Table 9. Load categories include: permanent action dead
(G1) and, deformation (G2), variable action long duration (Q1) and short duration (Q2), Environmental
action Extreme Level (EL) and Abnormal Level (AL) and Accidental (A).
810

Table 3. API RP 2N.
Design loads:
Probabilistic and deterministic methods.
Do not explicitly define safety classes or reliability targets although annual probability of failure for
structure designed to API RP 2A is approximately 1.5 x 10-5 (Nevel, 1997).
Consideration given to personnel safety compliance with existing regulations and pollution prevention
Offshore environment has considerable detail.
Loads based on full scale data and actions.
Full range of ice models and ice types considered.
Doesnt consider the effect of aspect ratio on global pressures on wide structures.
Doesnt account for reduced pressure for increased thickness.
No provision for pack ice driving force.
Structural design:
Structural design has option for using Load and Resistance Factor Design (LRFD) based on API RP
2A Working Stress Design (WSD)calibration based on component design practices in the GoM;
waves being the predominant environment hazard.
Two specified load cases for ice loads: 1) frequent events; and 2) infrequent events.

Load categories include: Gravity dead (permanent D1 and semi-permanent D2 changing from one ops
mode to another), Gravity live (consumables, L1 and short duration, L2 loads e.g. crane loads), Wind,
Wave and Current (extreme We, operating Wo, inertial Dn), ice operational (IOP), ice frequent (IF), Ice
rare (IR), Earthquate (E) and Fabrication.
Combinations are illustrated in Table 11.
Table 4. SNiP & VSN.

Design loads:
Offshore is extension of methods for bridge and port infrastructure design in rivers.
Offshore environment not specifically addressed.
Only two types of ice loads considered: level ice and ridges.
Loads based on small scale tests and factored to obtain global ice pressures.
Provision of loads use semi-probabilistic methods.
Return period for extreme environmental loads is not specified.
Target level for structural reliability is not clear.
No provision for driving forces which may limit ultimate loads on facility (i.e. loads may be
quite conservative).
Structural design:
Structural design considers two limit states: whether the limit state leads to 1) cessation in
operationsLoss of platform and operation due to foundation sliding, or structural collapse, or
2) no cessationno loss of normal operations due to loss of local strength.

Constant permanent loads (e.g. self-weight, and soil pressure), and Temporary loads a) Long
Term operational, b) Short Term Environmental, and c) Special (extreme environment, earthquake or explosive)
Combinations are illustrated in Table 12.
Table 5. NPD.
Design loads:
Do not distinctively specify safety classes. Requirements are stipulated to maintain and further develop an
adequate level of safety for people, environment, assets, and financial interests. Three reliability levels
representing degree of exposure include Low (10-3), Medium (10-5), and High (10-7).
Structural design:
Four categories of limit states: Serviceability Limit State, Fatigue Limit State, Ultimate Limit State, Limit
State for Progressive Collapse.

Loads classified as permanent loads, variable functional loads, environmental loads, and accidental loads.
Combinations are summarized in Table 13.
811
Table 6. NORSOK.
Design loads:
Standards include as far as possible provision of teh NPD.
Loads defined and classified according to ISO 13819-1.
Limit states or Load and Resistance Factor Design (LRFD).
Structural design:
Principles referenced according to ISO 13819-1.
May consider reliability design method provided it can be documented that the method is theoretically justified and provision is made for adequate safety in typical known cases. For Norwegian petroleum activities, the decision to use reliability based methods for design sits with the
NPD.

Loads and load effects are referenced according to NPD guidelines, DNV Class. Note No. 30.5
and API RP 2N.
Combinations are summarized in Table 13 and 14.
Table 7. DNV.
Design loads:
Probabilistic and deterministic considerations.
Safety classes are not specified. Rules do state that structures are to be designed to maintain acceptable
safety for personnel and environment.
Minimum target reliability levels are to be established based on calibration against well-established
cased know to have adequate safety.
Structural design:
Structural design methods include: Partial Coefficient Method, Allowable Stress Method, Reliability/Probabilistic Analysis, design by testing or observation of performance.
Loads and load effects are referenced according to NPD guidelines.
Table 8. ISO Combination factors for Companion EL environmental processes.

Principal Action
Factor for EL companion environmental action
Companion action stochastically

dependent of principal action
Companion action stochastically

independent of principal action
EL Action
0.9
0.6
AL Action
0.5
0.4
Table 9. ISO Exposire levels L1 and L2ULS and ALS action factors and action combinations.
Action Combina- Limite State Action Factors
tion
Permanent
Variable
Q1
Enviornmental
Accidental
EL
AL
G1
G2
1 Gravity and
deformationshort
and long
duration
1.3a or 0.9
1.5a
1.5a
0.7
2 Extreme
environ-
1.1a or 0.9b
1.1a or 0.8b
1.35
(L1)cd
Q2
Ultimate Limite State
812
mental
0.9
(L2)cde
1.1a or 0.9b
3 Damaged
conditionf
1.1a or 0.8b
Abnormal (accidental) Limite State

4 Abnormal
environmental
1c
2 Accidental
a. For the gravity and deformation action combination, a partial action factor of 1,20 may be used for permanent hydrostatic pressure and for physically limited variable actions.
b. The lower partial action factor applies for permanent G1 actions when the permanent action resists overturning or uplift, and for variable Q1 actions when there is a reversal of variable action effects.
c. The representative value of the environmental action to which the action factor applies shall be calculated
as described in 7.2.3.
d. The limit state partial action factor for the extreme-level seismic action (ELE) shall be as defined in ISO
19902 for steel structures and shall be as defined in ISO 19903 for concrete structures. The partial action
factor for the ELE shall be 1,0 for all other structures.
e. For L2 exposure level, all action factors for all action combinations are as for L1 except for the EL factor
in the extreme environmental action combination.
f. Damaged condition or progressive collapse limit states are defined generally for structures in ISO 19900
and for particular types of structures in ISO 19902, ISO 19903 and ISO 19904-1.
Table 10. CSAS471 Load Combinations.

Loads:
Permanent
Dead
Deform
Live
Environmental
Freq.
Rare
Accidental
Safety
Class 1
ULS1
ULS2
ULS3
ULS4
1.25
1.05/0.9
1.05/0.9
1.05/0.9
1
1
1
1
0.7
1
1
1
1.35
-
1
-
-1
1
Safety
Class 2
ULS5
ULS6
1.05/0.9
1.05/0.9
1
1
1
1
0.9
-
Section Strength (concrete)
ULS7
1.05/0.9
1.35
Fatigue (steel)
ULS8
Fatigue (concrete)
ULS9
0.7
0.7/1
Limite State/ Load

Temporary Temporary
Combination
Constant long term short term
Special T1
T2
T2
Main Comb. 1
0.9
Servicability limited state
Table 11. SNiP Load Combinations.
0.95
Main Comb. 2
Main Comb. 3
0.8 0.6
Special Comb.
0.95
0.8
Table 12. NPD Load Combinations.

Limite State
Permanent
Variable
functional loads
Environmental
loads
Deformation
loads
Accidental
loads
Ultimate a
1.3
1.3
0.7
813
Ultimate b
1.3
Serviceability
Fatigue
Progressive Collapse
Table 13. NORSOK Load Combinationsin addition to NPD (m- mean water level;
m*mean water level including storm flood).
Limite State
Wind
Ice Snow Earthquake
Sea Level
Ultimate
-2
Waves Current
10
10-1
10-1
-
-2
10
10-1
10-1
-
-1
10
10-2
10-1
-
10-2
-
10-2
-
10-2
10-2
10-2
m
m
m
Progressive
10-4
10-2
10-1
-
10-2
10-4
10-1
-
10-1
10-1
10-4
-
10-4
-
10-4
m*
m*
m*
m
m
Table 14. API Load Factors and Combinations.

Limite State/
Load Combination
Case 1
Operating Wave
Case 2
Operating Ice
Case 3
Extreme Wind, Wave,
and Current
Case 4
Design ice load,
frequent events
Case 5
Design ice load,
rare events
Case 6
Earthquake load
2.
D1
D2
L1
L2
Dn
We
IOP
IF
IR
1.3
1.3
1.5
1.5
1.5 1.2
W0
1.3
1.3
1.5
1.5
1.5 -
1.2 -
1.1
0.9
1.1
0.9
1.1
0.8
1.7 -
1.35 -
1.1
0.9
1.1
0.9
1.1
0.8
1.7 -
1.35 -
1.1
0.9
1.1
0.9
1.1
0.8
1.7 -
1.1
0.9
1.1
0.9
1.1
0.8
FULL SCALE ICE LOAD MEASUREMENT CAMPAIGNS
This section seeks to summarize the full scale measurement campaignes carried out world wide in
the past in Table 15 to 18 without any claim to be complete.
Table 15. Baltic Sea.
Project
Possesion of data Sea Area
Content of the data
IB URHO, 1976
VTT
Stresses during ice trials
IB SISU, 1978-85 VTT, AALTO
Baltic Sea
Bay of Bothnia Stresses, loads and pressures during service; bow and bow
shoulder areas
References: Kujala and Vuorio (1986)

MT IGRIM, 1978 VTT
References: Korri and Varsta (1979)
Bay of Bothnia Stresses and ice pressures on ship bow and bow shoulder
during ice trials
814
MS KEMIRA
AALTO, VTT
Baltic Sea
and1985-91
References: Kujala (1989), Muhonen (1992)
Stresses and loads during service; bow, mid

aftship frame
MT KASHIRA
VTT, AALTO
Gulf of Finland Stresses and loads during service; bow areas
1984-90
References: Kujala (1994), Lensu and Hnninen (2003)
MS ARCTURUS VTT, AALTO

Gulf of Finland Stresses and loads during service; bow areas
1983-88
References: Kujala (1994), Lensu and Hnninen (2003)
MT UIKKU 2001, AALTO, Arcdev Gulf of Finland Stresses and load during service and ice trials; mid, aft,
2003
continuous time series
References: Lensu (2002), Kujala et al. (2009)
NEUWERK
1999
HSVA, FHK
Gulf of Bothnia Stress and ice loads during

ice trials and transits; aft shoulder area
IB OTSO, 2005
AALTO
Bay of Bothnia Ice loads during service; bow,

continuous time histories
SA AGULHAS II AALTO
Bay of Bothnia Stresses and ice load on ship bow, bow shoulder
and stern shoulder during ice trials
References: Suominen et al. (2013)
Table 16. Arctic Sea (1/2).

Project
Possesion of data Sea Area
WERDERTOR
loads measurements,
1977
RV AKADEMIK
FEODOROV, 1991, 1994
GL
Spitsbergen
Stresses and ice loads during

ice trials; bow ice shoulder area
VTT, AARI
Russian Arctic
Ice loads on ship bow and midbody
CANMAR
KIGORIAK, 1979
Reference: Dome (1982)
MV ARCTIC
1984, 1981
VTT, AALTO
Canadian Arctic
Ramming trials where local

bow loads were measured
CCG, VTT
Canadian Arctic
Total ice load in ramming and

local forces on ship bow
MT KASHIRA
198490
VTT, AALTO
Russian Arctic
Stresses and loads during

service; bow areas
USCGC Polar Sea

NRC, USCGC
1982, autumn
Reference: Daley et al. (1990)
Beaufort Sea
Bow, 167 events
USCGC Polar Sea

NRC, USCGC
1983, March
South Bering
Sea
Bow, 172 events
USCGC Polar Sea

1983, March-April
South Bering
Sea, North Bering Sea,
South Chukchi Bering
Sea, North Chukchi
Bow 1225 events
USCGC Polar Sea

NRC, USCGC
1985 September and
October
Reference: Minnick et al. (1990)
Beaufort Sea
40 Global ram events

with MY ice
USCGC Polar Sea

NRC, USCGC
1986, March
Bering Sea
Bow, 653 events
NRC, USCGC
Reference: Daley et al.

(1990)
Content of the data
Bow
815
POLARSTERN
1984
GL
Labrador
Stress and ice loads during ice

trials and transits; bow shoulder
area, nozzle ring; superstructure
POLARSTERN
1984
HSVA
Labrador
Loads on ca. 1m2 pannel in bow

area during ice trials and transits, load statistics
POLARSTERN
1985
GL
Spitzbergen

trials and transits; bow shoulder
area, nozzle ring; superstructure
POLARSTERN
1985
HSVA
Spitzbergen
Loads on ca. 1m2 pannel in bow

area during ice trials and transits, load statistics
MUDZUG (Thyssen/Waas) GL
strain and acceleration, 1987
Spitzbergen

trials and transits; bow area
MUDZUG (Thyssen/Waas) HSVA

strain and acceleration, 1987
Spitzbergen
Loads on ca. 0.25m2 pannel in

bow area during ice trials and
transits, load statistics
KAPITAN SOROKIN (T/W)

strain and acceleration
measurements, 1991
Kara Sea, Yenessey

Esturary

trials and transits; bow area
GL
Table 17. Arctic Sea (2/2).

Possesion of data
Project
Louis S. St.
NRC
Laurent, 1980
Reference: Glen et al. (1981)
Sea Area
Content of the data
Labrador Sea,
Davis Straight, Baffin Bay
Louis S. St.
Laurent, 1994
Polar transit
NRC
References
Bow; 30 subpanels, 1730 events. Shoulder;

12 subpanels, 1289 events Bottom; 11
sub-panels over area, 48 events
References: Ritch and St.John (1994), Frederking and Collins (2005)
Bow and bow girder

North Pole
786 events
Voyage
References: St. John and Minnick (1993), Frederking and Collins (2005)
Oden, 1991
NRC
Bow, 1172 events and

North Pole
side, 1135 events
Voyage
References: St. John and Minnick (1993), Frederking and Collins (2005)
Oden, 1996
NRC
MT UIKKU, 1998
AALTO, Arcdev
Russian Arctic
Stresses and load during service and ice

trials; bow, mid, aft
Reference: Kotisalo and Kujala (1999)

IB Teshio
1998, February
NMRI, JCG
Sea of Okhotsk
Stress and ice load measurements at 5

locations along the hull
IB Teshio
1999, February
NMRI, JCG
Sea of Okhotsk
Stress and ice load measurements at 5

locations along the hull
Baffin Bay
Davis Strait
Loads during ice trials; bow shoulder 1800

events; transom, 1660 events
Arctic
Stress and ice loads during service

bottom area
USCGC HEALY
AALTO
2000
Reference: Hnninen et al. (2001)
POLARSTERN
2002 ff
Laeisz, FHB
Large tanker 2010

Reference: Iyerusalimskiy et al. (2011)
Russian Arctic
KV Svalbard
DNV, NTNU
Spitsbergen
Reference: Mejlnder-Larsen and Nyseth (2007)
816
Table 18. Antarctic Sea.

Project
Possesion of
data
Sea Area
Content of the data
RV AKADEMIK
FEODOROV, 1991, 1994
VTT, AARI
Antarctica
Ice loads on ship bow and midbody
POLARSTERN
2002 ff
Laeisz, FHB
Antarctica
Stress and ice loads during service

bottom area
SA AGULHAS II
2013, 2014
AALTO
Antarctica
Stresses and ice load on ship bow,

bow shoulder and stren shoulder
during ice trials
References
USCGC Polar Sea 1984

Antarctica Bow, 310 events
References: Daley et al. (1990), Frederking and Collins (2005)
3.
REFERENCES
Daley, C., St.John, J. W., Brown, H. & Glen, I. 1990. Ice forces and Ship Response to IceConsolidation Report.
Report for Ship Structures Committee.
Dome 1982. Full Scale Measurements of the Ice Impact Loads and Response of the Canmar Kigoriak. August and
October, 1981. Prepared by Dome Petroleum Limited.
Frederking, R. & Collins, A. 2005. NRC Catalogue of Local Ice Pressures (CLIP). Canadian Hydraulics Centre
Technical Report.
Glen, I., Blount, H., Comfort, G. & Tam, G. 1981. Measurement of Ice Impact Pressures and Loads Onboard CCGS
Louis S. St. Laurent during a 1980 Fall Arctic Probe. Report for Canadian Coast Guard by ARCTEC Canada Ltd.
Hnninen, S., Lensu, M. & Riska, K. 2001. Analysis of the Ice Load Measurements During USCGC Healy Ice
Trials. Spring 2000. Helsinki University of Technology, Espoo, Ship Laboratory.
Iyerusalimskiy, A., Choi, J., Park, G., Kim, Y., Yu, H. & St. John, J. The interim results of the long-term ice loads
monitoring on the large arctic tanker. Proceedings of the 21st International Conference on Port and Ocean
Engineering under Arctic Conditions, POAC'11, 2011.
Korri, P. & Varsta, P. 1979. On the ice trial of a 14500 DWT tanker on the Gulf of Bothnia.
Kotisalo, K. & Kujala, P. Ice load measurements onboard MT Uikku during the Arcdev voyage. POAC99, 1999
Espoo, Finland. 974987.
Kujala, P. 1989. Results of Long-Term Ice Load Measurements on Board Chemical Tanker Kemira in the Baltic Sea
during the Winter 1985 to 1988. Winter Navigation Research Board.
Kujala, P. 1994. On the Statistics of Ice Loads on Ship Hull in the Baltic. Dissertation. Dissertation, Helsinki
University of Technology.
Kujala, P., Suominen, M. & Riska, K. Statistics of ice load measured on MT Uikku in the Baltic Sea. POAC2009,
2009 Luulaja, Finnland.
Kujala, P. & Vuorio, J. 1986. Results and statistical analysis of ice load measurements onboard icebreaker Sisu in
winters 1979 to 1985. Winter Navigation Research Board.
Lensu, M. 2002. Short Term Prediction of Ice Loads Experienced by Ice Going Ships. Espoo, Finnland: Helsinki
University of Technology, Ship Laboratory.
Lensu, M. & Hnninen, S. Short Term Monitoring of Ice Loads Experienced by Ships. 17th International
Conference on Port and Ocean Engineering under Arctic Conditions, POAC03, 2003 Trondheim, Norway. 535
544.
Mejlnder-Larsen, M. & Nyseth, H. 2007. Ice load monitoring RINA, Royal Institution of Naval Architects.
International ConferenceDesign and Construction of Vessels Operating in Low Temperature Environments
Papers.
Minnick, P., St.John, J., Cowper, B. & Edgecombe, M. 1990. Global Ice forces and Ship Response to Ice. Report for
Ship Structures Committee.
Muhonen, A. 1992. Ice Load Measurements on Board MS Kemira. Winter 1991. Espoo, Finnland: Helsinki
University of Technology, Ship Laboratory.
Ritch, R. & St.John, J. 1994. Ice load impact measurements on the CCGS Louis S. St. Laurent during the 1994
Arctic Ocean crossingA trip report and data summary. Report to Fleet Systems, Canadian Coast Guard.
St. John, J. & Minnick, P. 1993. Swedish icebreaker Oden ice impact load measurements during International Arctic
Ocean Expedition 1991; Instrumentation and measurement summary. STC Technical Report 2682 to U.S Coast
Guard Headquarters.
Suominen, M., Karhunen, J., Bekker, P., Kujala, P., Elo, M., Von Bock Und Polach, R. F., Endlund, H. & Saarinen,
S. Full-scale measurements on board PSRV S.A. Agulhas II in the Baltic Sea. Proceedings of the 22nd
International Conference on Port and Ocean Engineering under Arctic Conditions (POAC), 2013 Espoo,
Finnland.

7 10 SEPTEMBER 2015
CASCAIS, PORTUGAL
VOLUME 2
COMMITTEE V.7
STRUCTURAL LONGEVITY
COMMITTEE MANDATE
Concern for the structural longevity of ship, offshore and other marine structures. This shall include
diagnosis and prognosis of structural health, prevention of structural failures such as corrosion and fatigue,
and structural rehabilitation. Attention should be given to ongoing lifetime extension of existing structures.
Focus shall be on methodologies for translating monitoring data into operational advice and lifecycle
management. The research and development in passive, latent and active systems including their sensors and
actuators should be addressed. Further, self-healing and smart materials should be addressed.
COMMITTEE MEMBERS
Chairman:
P. Hess, USA (Chair)

S. Aksu, Australia
J.I.R. Blake, UK
D. Boote, Italy
P. Caridis, Greece
A. Egorov, Ukraine
A. Fjeldstad, Norway
M. Hoogeland, Netherlands
H. Murayama, Japan
M. Rye Anderson, Denmark
M. Tammer, Netherlands
KEYWORDS
Ship structures, fatigue, service life, corrosion, structural health monitoring, structural longevity, structural
inspection, structural repair, structural maintenance, structural damage detection, fatigue life, crack detection,
structural lifecycle assessment, structural lifecycle management, structural sensing, acoustic emission
818
ISSC committee V.7: STRUCTURAL LONGEVITY
CONTENTS
1.
INTRODUCTION
1.1
Background & Mandate
1.2
Relationship with other ISSC Committees
820
820
820
2.
LIFECYCLE ASSESSMENT & MANAGEMENT FOR STRUCTURAL LONGEVITY

2.1
Introduction
2.2
The Need for Lifecycle Assessment and Management
2.3
Conclusions
821
821
821
823
3.
CURRENT PRACTICE
3.1
Introduction
3.2
The Role of Regulators and Classification Societies
3.3
Classification Rules and Guidance
3.4
3.4.1
International trading vessels
3.4.2
High-speed Craft (HSC)
3.4.3
Vessels operating in inland waterways
3.5
Offshore structures
3.5.1
Offshore drilling units
3.5.2
Floating Production Storage and Offloading (FPSO) units
3.5.3
Fixed production platforms
3.6
Naval vessels
3.7
Conclusions
823
823
823
824
825
825
826
826
826
826
827
827
827
828
4.
PREDICTION OF LONGEVITY
4.1
Introduction
4.2
4.2.1
Prediction of corrosion
4.2.2
Fatigue strength prediction
4.2.3
Buckling prediction
4.3
4.4
Conclusions
828
828
828
829
829
830
830
830
5.
PREVENTION & REPAIR OF STRUCTURAL FAILURES

5.1
Introduction
5.2
Prevention of failure - Design stage
5.2.1
5.2.2
Material selection
5.2.3
Structural design
5.3
Prevention of failure - Operation
5.3.1
Maintenance & inspection
5.3.2
Repair and rehabilitation
5.4
Conclusions and Recommendations
831
831
831
831
832
832
833
833
834
836
6.
INSPECTION METHODS & TECHNIQUES

6.1
Introduction
6.2
6.3
6.4
Limitations
6.5
836
836
837
837
838
839
7.
SENSING TECHNOLOGIES
7.1
Introduction
7.2
Passive systems
839
839
840
7.3
7.4
7.5
7.6
7.2.1
Strain
7.2.2
Acoustic Emission
7.2.3
Vibrations
7.2.4
Crack
7.2.5
Corrosion
7.2.6
Acceleration
7.2.7
Metocean information
Active system
7.3.1
Impedance-based methods
7.3.2
Lamb wave-propagation methods
Maturity of Structural Hull Monitoring Systems
8.
METHODOLOGIES FOR USING INSPECTION & SENSED DATA

8.1
Introduction
8.2
Operational Advice
8.2.1
Identifying loading to stay within safe operating envelope
8.2.2
Quantifying operational loading and changes
8.3
Lifecycle Management Advice
8.3.1
Condition Based Maintenance (CBM)
8.3.2
Reliability Centered Maintenance
8.3.3
Reliability Based Inspections
8.4
8.5
Discussion
8.6
Conclusions
9.
LIFE TIME EXTENSION, COMPARISON OUTSIDE & WITHIN THE

MARITIME I NDUSTRY
9.1
Introduction
9.2
Lifetime Extension of Existing Structures
9.3
Other industries
9.4
Differences in approaches for ships, offshore structures, and other
marine structures (ranging from navy to renewable energies)
9.5
Conclusions
819
840
840
841
841
841
841
842
842
842
843
844
844
844
845
845
846
846
848
848
850
850
850
851
851
851
852
852
852
854
855
855
10. CONCLUSIONS & RECOMMENDATIONS

10.1 Conclusions
10.2 Recommendations
856
856
856
REFERENCES
857
820
1.
INTRODUCTION
1.1
Background & Mandate
Ships, offshore and other marine platform structures are designed with assumptions regarding construction methods, loading, environment, design criteria, material performance, operation, maintenance, and
service life. However, if any design assumption proves incorrect then this could result in a risk and cost to
the owner and operator. The actual condition, use, and performance of the platform structure changes
over time, which requires updated maintenance requirements for scheduling and budgeting, decisions on
limiting or expanding the operational use, and predicting remaining useful service life.
There are great challenges in ensuring todays complex ships, offshore, and other marine structures
have an affordable and adequate service life, which, ideally, should not be limited by structural considerations such as deterioration from corrosion and fatigue cracking. Developing technology for diagnosis and
prognosis of structural health enhances prediction and planning of future structural maintenance costs.
Classification societies are providing guidance and additional class notation for the installation of onboard structural monitoring systems, and research continues into means of translating the data collected
from those systems into operational advice and lifecycle management. Making new designs more resilient
by going beyond the safety-based requirements specified by the cognizant authority such as a classification society involves a greater initial cost, which can be justified by incorporating lifecycle maintenance
considerations into the initial design cycle. Allowance for condition-based maintenance strategies might
be made in the design process to reduce conservatism in design assumptions and support a more sophisticated and economical lifecycle management scheme.
1.2
Relationship with other ISSC Committees
The following committees have areas in their mandates that overlap the work of this committee and were
therefore not specifically addressed by this committee.
Committee III.2 Fatigue and Fracture

Committee V.3 Material and Fabrication Technology the impact of corrosion protection techniques on structural performance.
Past ISSC Committees addressed areas contained within the work of this committee.
Committee I.2, Loads, of ISSC 2012 procedures for stress monitoring and for operational guidance
Committee II.1, Quasi-Static Response, of ISSC 2012 strain gauges embedded in a composite,
reliability-based structural assessment, and reliability-based inspection, maintenance, and repair, including the use of real-time condition monitoring systems
Committee II.2, Dynamic Response, of ISSC 2012 monitoring of both ship structures and offshore
structures
Committee IV.2, Design Methods of ISSC 2012 means to effectively use all of the data accumulated over the lifetime of a ship for more cost-effective management
Committee V.2, Natural Gas Storage and Transportation, of ISSC 2012 monitoring of hull structures and containment tanks
Committee V.3, Materials and Fabrication Technology, of ISSC 2012 structural health monitoring
including a study of the time-varying reliability of ship structure due to corrosion
Committee V.5, Naval Vessels, of ISSC 2012 monitoring of structure
Committee V.6, Arctic Technology, of ISSC 2012 requirements by ABS for ice load monitoring
(ABS, 2011)
Given this interest in structural longevity, this specialized committee, Structural Longevity, was
formed for ISSC 2015. Because the technology is currently being applied to all marine structures and
has the potential for improvement in the management of those structures, the mandate for this committee includes all ship, offshore, and other marine structures. As the primary causes of structural deterioration are fatigue and corrosion, with the two mechanisms sometimes assisting each other, the diagnosis and prognosis of structural health concentrates on these mechanisms. Assessment of structural
health and structural rehabilitation is essential for the continued safety of a marine structure throughout
its life, but is also important for the decision-making process for extending the service life of an existing structure and these topics have been dealt with in this report. As more ships and other marine struc-
821
tures are receiving more instrumentation to monitor the structure, including strain gauging, motion
detectors, and fatigue and corrosion sensors, the data from these monitoring systems needs to be taken
to full advantage. Therefore, this committee has had a focus on the methodology for translating that
data into operational advice as well as lifecycle management of structures. Improvements in these
monitoring systems, including passive, latent and active systems and their sensors and actuators were
also addressed.
2.
LIFECYCLE ASSESSMENT & MANAGEMENT FOR STRUCTURAL

LONGEVITY
2.1
Introduction
This chapter will examine the need to assess the structural lifetime of vessels and floating offshore structures and the subsequent management for structural longevity.
Through-life assessment and appreciation of the full lifecycle costs are needed for the asset owner to
make decisions on continuing operation and managing longevity. This requires accurate monitoring technologies, effective inspection methodologies, predictive tools for damage evolution and preventive methods (maintenance, repair and rehabilitation).
Structural assessment following a change in conditions, such as structural deterioration from corrosion
or fatigue, structural damage, a change in loading, or an extension of the design working life qualifies an
assets (vessel or platform) instantaneous capability to function as intended by examining its residual
capacity to withstand loads (Rucker et al., 2006). Design code updates or operator-led safety measures
may also initiate structural assessment (Dier, 2004). Lifetime assessment monitors structural health and
extrapolates, or predicts, the expected structural life of the asset allowing informed but complex decisions
by the owner/operator on the assets future. Nappi and Collette (2013) describe the complexity of the
lifecycle performance predictions, and argue for a greater investment in improving design frameworks
and structural assessment tools for new structures.
Structural codes that have been developed for new designs are often not appropriate for structural assessment since there are significant differences between design and assessment processes. There is a tendency to develop conservative, redundant designs to account for the uncertainties that arise from the prediction of load and resistance parameters of a new structure. Hence, whilst a conservative design does not
result in a significant increase in structural cost, a conservative assessment on the other hand may result in
unnecessary and costly repairs or replacement (Rucker et al., 2006). Kwon and Frangopol (2012a) reason
that in order to account for uncertainty, a probabilistic methodology for lifetime structural performance
assessment and management has to be developed in a rational way. In fact, Hess (2003) used the framework for structural reliability analysis coupled with performance goals of operational availability, capability, and dependability (Ao, Co, Do) metrics to link lifecycle guidance to structural assessment. Stacey
et al. (2008) discuss the identification of uncertainty in structural performance as a major issue for any
life extended facility, including the loss of corporate knowledge, increasing relevance of small defects,
and structural condition and associated responses. It is therefore necessary to quantify residual structural
capacity, but its definition at any instant in time is uncertain, so any decision on life extension and structural management comes with a risk from the costs associated with action or inaction. Recently, Stambaugh et al. (2014) presented a framework for evaluating a function of risk against total ownership costs
and tested the framework against the lifecycle management of a US Coast Guard vessel. A Monte Carlo
simulation is used to compute the time-varying structural fatigue reliability, which is then used to evaluate the effect of alternative design and maintenance strategies on total ownership cost.
With a large number of existing offshore installations reaching or close to reaching their nominal installation life, ageing and life extension (ALE) has been of particular interest. The practice of operation
beyond design life is becoming more and more common because of obvious economic drivers such as
lack of funds for decommissioning the original asset and installation of a replacement (Stacey et al.,
2008). The development and approaches to life extension of assets for the marine industry has been driven by regulatory bodies such as the UKs Health and Safety Executive Key Program 4 (KP4). This
program examines the extent to which asset integrity risks associated with ALE are being managed effectively by asset owners/duty holders and how good practice is being promoted and developed.
2.2
The Need for Lifecycle Assessment and Management
As structures age, operational and organizational changes in strategy influence the way in which the
structure should be managed. To provide the owners information to make a decision on the future of their
asset such as scrapping the asset and building a new asset, the lifetime assessment is informed by
Roberton, (2014):
822

Economic: with or without intervention (based on condition), what is the cost of operating a vessel
for X number of years?
Structural: by assessing load input and structural response within material limits what's the life of the
vessel?
Maintenance: inspection techniques and condition based monitoring defining appropriate service
scheduling and subsequent life extension after maintenance.
Systems: system-focused life prediction: holistic approach using the combined influence of component reliability on the system's life prediction.
Resilience: the capacity of the structure to withstand off-design variations and flexibility in operation role.
With ageing structures, the economic benefit of continuing to operate the asset may not balance with
incumbent costs associated with maintenance, repair, or rehabilitation. The question remains: with or
without intervention (based on condition), what is the cost of operating a vessel for a defined number of
years? Extending the life of an asset is a complicated decision process which hinges on the lifecycle costs.
The process of ageing isnt readily defined by the length of service the asset has already seen: poorly
designed and maintained assets will fail faster than older, more conservatively designed assets with effective inspection and maintenance regimes. What is known about the condition, how to reduce or recognize
the onset of degradation, how that degradation evolves, and what to do to mitigate the degradation is
more important than the physical age. Economically a decision then centers on the possible actions of
continued service, re-rating (for example, vessels looking at costly steel repair to make good minimum
thickness limits following corrosion could operate under a limited service, reassessed by Class, affording
a more generous corrosion allowance), repair, or scrapping in favor of a new build (HSE, 2014).
The attractions for life extension seem obvious: for example, Jean (2008) writes to modernize a [US Navy]
surface combatant costs a fifth of what it takes to build a new ship. A new destroyer costs about $1 billion.
The price to upgrade a destroyer is about $180 million... However, at what point does it become unfeasible to
life extend? How are costs modeled? What is the total cost of ownership and how transparent is this?
Business economists Ntuen & Moore (1986) define lifecycle costs as capital acquisition costs, operating costs including maintenance, and end of life disposal costs. The authors include a classification of
approaches to availability-based lifecycle cost analysis that covers established techniques in optimization,
queuing, and simulation as applied to design, unavailability, and operation (availability) costs (earliest
reference from 1946 up to 1985). More recently, well cited work by Umeda et al. (2000) explores lifecycle simulations (LCS) incorporating the aforementioned techniques. Physical deterioration of a product
can be represented by a stochastic element in the simulation (Komoto et al., 2011). Well established methods and tools exist therefore to aid the decision process but the complexity of the asset itself as a system, the number of stakeholders, the supply chain, the length of service, diversity of the physical operating environment, and a fickle economic climate lead to an almost intractable simulation of every scenario.
Of the whole lifecycle cost, through the stages of requirements definition, design, production, operation, and disposal, it is expected that around 70% is consumed during the operational phase (BMT, 2006)
and yet two-thirds of the budget has been fixed by the time that the requirements definition has been
completed (Fabrycky, 2003). Decisions made in the early phases of a systems lifecycle therefore have
significant cost impact during the operational phase of a system (Iyer, 1999, Swift and Brown, 2003). To
avoid budget versus performance problems its imperative to pay attention to lifecycle costs at the earliest
of the design stage. Turan et al. (2009) present a design-oriented optimization study as an approach for
analyzing lifecycle costs for a commercial vessel but again discuss this process limitation in the breadth
of the available cost information. Methods such as System Dynamics or Discrete Event Simulations following Brailsford and Hilton (2001) have been used by the UKs Royal National Lifeboat Institution
(RNLI) in determining appropriate maintenance schedules and costs, at both boat and fleet level.
Maintenance activities are a significant portion of the operational costs. In the UK manufacturing industry, the impact of maintenance is between 12 and 23% (Cross, 1988) of the operating costs but for bulk carriers, average maintenance activities account for 40% of the operating costs (Alhouli et al., 2010). A logical
goal therefore is for management to balance the lifecycle costs with system effectiveness or availability. As a
significant amount of the annual operational costs are attributed to maintenance costs, an effective and efficient maintenance policy is preferable to investment in redundant capacity or reactive maintenance efforts.
It is unsustainable under increasing pressures and greater financial constraints to be investing in assets
that cannot be flexible in their operational capability or designed to be easily maintained. Page (2002)
shows that driving operating costs down is challenging, but for new assets theres an opportunity to
appreciate the impact of design subtleties on lifecycle costs, allowing for greater flexibility into operational capability and extend the assets useful life. Gratsos et al (2009) also reflect upon the impact of
823
design decisions on the lifecycle costs of an asset, balancing steel corrosion allowances for more robust
ship structures with lower maintenance burdens against through-life carbon foot-printing.
Stambaugh and Barry (2014) cite lifecycle cost and total ownership cost modeling undertaken by the US
Navy, US Coast Guard and the US Government Accountability Office. In the UK, a similar convergence has
been observed where the Royal National Lifeboat Institution (RNLI) has restructured the management of
their mostly composite, 25-m or less boats lifecycle, bringing their build in-house, adopting lean manufacturing, incorporating condition based monitoring of their vessels, and actively managing for future obsolescence of components. This has resulted in a strategy for existing vessels to have their life extended by 25
years and new builds being designed for a 50-year service life (Walshe et al., 2011). Currently there is a
program of research to investigate the residual fatigue life remaining for their structures by continual strain
gauge monitoring to better inform the maintenance scheduling and through-life costing (Roberton, 2014).
An otherwise general absence of a holistic methodology within the maritime industry was noted by
Nappi and Collette (2013) who report on the limited investigations of time-based reliability of structures
and the integration of this information into design and maintenance scheduling. Collette with coresearchers have subsequently worked on probabilistic models including Bayesian networks which aim to
predict future structural capacity from structural health monitoring (SHM) data (Groden and Collette,
2013) and optimize the maintenance/inspection intervals considering costs imposed by maintenance, corrosion, fatigue damage and possible life extension (Temple and Collette, 2013). Frangopol and Soliman
(2013) have used genetic algorithms to search for optimal through life decisions in the large problem
space of component, system and network management levels simultaneously balancing financial, safety
and environmental constraints.
2.3
Conclusions
In comparison to the aerospace, chemical processing, and nuclear power industries, the maritime industry
is not as prolific in its uptake of lifecycle assessment and management. The organizers of the Condition
Based Maintenance conference series, the Institute of Marine Engineering, Science and Technology, report that the industry is aware of the maturity of technology and methods supporting through life asset
management but requires an organizational culture change for their general adoption. It is clear that the
movement is towards integrated approaches and with goal based standards firmly rooted within the International Maritime Organization as the future for maritime design and operations, holistic approaches factoring in uncertainty, likelihood, reliability, consequence and risk can better inform the owner/operator of
the assets health and provide a route for managing that health efficiently and effectively.
3.
CURRENT PRACTICE
3.1
Introduction
Current practice is dominated by owners and operators complying with the rules and other guidance of
regulators and classification societies. This chapter briefly reviews that guidance and indicates how it is
currently used for maintaining structural longevity primarily until the next required inspection in a cycle.
3.2
The Role of Regulators and Classification Societies
Classification societies can aid the unraveling of direct and indirect cost implications of life extension or
scrapping. Det Norske Veritas (DNV) report that for some owners balancing income against maintenance
and running costs is an art (DNV, 2006). But class societies can help with the engineering science to prepare, for example, different steel and coating deterioration predictions for different maintenance scenarios
based upon current vessel condition. An appropriate maintenance scheme is then developed with the
owners/operators input. In fact Paik and Melchers (2014) report that much of the shipbuilding industry
is starting to follow the systematic risk-based approach seen in the offshore industry and it is more common that owners and operators are requesting vessel-specific inspection and maintenance guidance. As
an example of where this can benefit and improve the transparency of some of the lifecycle costs, DNV
(2010b) report that coating costs are a major factor in the dry-docking budget but only the very recent
vessels are likely to have a 3D computer model leaving the yard little option but to overestimate the
amount of blasting and coating required.
Ships and platforms are mandatorily inspected and surveyed under the requirements set out by IACS,
the IMO, flag and port states, but the industry has also established recommended practices or standards to
better manage its vessels (Paik and Melchers, 2014). MARPOLs Condition Assessment Scheme (CAS)
exists for oil tankers but in response to broader asset owner pressure on driving down lifecycle costs and
subsequently the life extension of their assets, class societies are actively responding to industry needs for
managing all ageing structures. Many have introduced condition assessment programs aimed at hull con-
824
dition and renovation and the American Bureau of Shipping, Lloyds Register, and Bureau Veritas also
have wider schemes looking at machinery and electrical systems, cargo, and ballast (Caridis, 2009).
Increased stakeholder involvement in the application of technology and research to lifecycle assessment
and management is evidenced by a number of international projects in research and evaluation. Stakeholders
in this process include owners, operators, shipyards, and classification societies. There are many examples in
the European Community funded through the Framework Programmes that importantly look at the holistic
problem of lifetime assessment and management (e.g. Condition Assessment Scheme for Ship Hull Maintenance, FLAGSHIP, HULLMON+, Inspection Capabilities for Enhanced Ship Safety (INCASS), OPTIMISE) and it is clear that the culture for lifecycle assessment and management for structural longevity is
becoming not only increasingly recognized in its need but also in its adoption.
3.3
Classification Rules and Guidance
The vast majority of commercial ships and other mobile offshore units are built to and surveyed for compliance with the standards established by classification societies. The purpose of a classification society is
to provide statutory services and assistance to the maritime and offshore industry with regard to maritime
safety and pollution prevention, and allow for insurance of ship and cargo to be obtained. Classification
Societies aim to achieve this objective through the development and application of their own rules and by
verifying compliance with international and national statutory regulations on behalf of flag administrations (IACS, 2014a).
Table 1. Primary Classification Society References.
Author
American Bureau of
Shipping (ABS)
Bureau Veritas (BV)
Det Norske Veritas

(DNV)*
Korean Register of
Shipping (KR)
Lloyds Register (LR)
Reference
Guide for Hull Inspection and Maintenance Program (2013)
Guide for Nondestructive Inspection of Hull Welds (2014b)
Guidance Notes on the Inspection, Maintenance and Application of Marine Coating
Systems (2007a)
Guide for Surveys using Risk-Based Inspection for the Offshore Industry (2003b)
Rules for Survey After Construction (2014c)
Guide for Hull Condition Monitoring Systems (2003a)
NR 467 Rules for the Classification of Steel Ships (2015)
NI 456 Harmonized condition assessment programme (1999)
NI 531 Guidelines for the application of the IMO performance standard for protective
coatings (2007)
NI 422 Type Approval of Non Destructive Testing Equipment Dedicated to Underwater Inspection of Offshore Structures (1998)
NI 567 Risk Based Verification of Floating Offshore Units (2010)
DNV Rules for Classification of Ships: part 7 Ships in Operation, Chapter 1, Survey
Requirements (2013)
DNV-RP-G101 Risk Based Inspection of Offshore Topsides Static Mechanical
Equipment (2010a)
DNV-RP-G103 Non-Intrusive Inspection (2011)
DNV Class Note 7 Non-destructive testing (2012b)
DNV Class Note 10.2 Guidance for Condition Monitoring (2008)
DNV Guideline 10 Guide for Ultrasonic Thickness Measurements of Ships Classed
with DNV (2009)
Rules for the Classification of Steel Ships (2014)
NSRC/14 Rules and Regulations for Classification of Naval Ships (2014b)

MTSTCG/14 Tank Coatings Condition Guide (2014c)
Nippon Kaiji Kyokai
Rules for the Survey and Construction of Steel Ships (2014b)
(ClassNK)
Rules for Hull Monitoring Systems (2014a)
Germanischer Lloyd*
Rules for Classification and Construction. Ship Technology Part 0 Classification and
Surveys (2015)
*Det Norske Veritas and Germanischer Lloyd have recently merged but at the time of this writing their respective
rules have not been harmonized.
A group comprising the largest classification societies are members of the International Association of Classification Societies (IACS), founded in 1968. With the aim of harmonizing the class rules of the respective
classification societies, Common Structural Rules (CSR) for tankers and bulk carriers (IACS, 2014b) have
been developed by the IACS members. Through the IACS membership the classification societies have undertaken to implement these rules to their own class rules. Today it is estimated that more than 90% of the
worlds cargo carrying tonnage is covered by the classification design, construction, and through-life com-
825
pliance rules and standards set by the member societies of IACS. IACS is a non-governmental organization,
but it also plays a role within the International Maritime Organization (IMO), for which IACS provides
technical support and guidance and develops unified interpretations of the international statutory regulations
developed by the member states of the IMO. The link between the international maritime regulations, developed by the IMO and the classification rule requirements for a ship is established though the International
Convention for the Safety of Life at Sea (SOLAS)(IACS, 2011). Classification societies are required to implement requirements of IMO and/or Flag states to their rules.
Vessels not covered by the IACS Common Structural Rules include container ships, passenger ships,
high-speed craft, and river and river-sea vessels. These are covered by specific rules developed by the
individual classification societies. As a consequence of not being harmonized the rules may have less
common features. International trading vessels are still subject to IMO regulations.
In addition to specifying rules and requirements to design and operation of vessels, classification societies also provide guidance and practical information on classification of ships and other objects in varying
ways. Classification societies provide guidance on a wide range various topics ranging from design (fatigue, strength, buckling etc.) to operation (repair, non-destructive testing, monitoring, etc). Table 1 lists
some of the primary references to some of the recognized classification societies.
3.4
3.4.1
International trading vessels
Commercial shipping vessels are required to be designed according to class rule requirements as a minimum. Although a specific design life is not stated, a design life of 20 years is expected through adherence
to class rules, yet there is limited emphasis on the total lifetime in the class rules because the class system
for ships in operation is based on renewal programs carried out each fifth year. The vessel is approved for
continued operation provided that all requirements are met.
The class rules state that a vessel shall be subjected to annual, intermediate, and renewal surveys. The
purpose of an annual survey is to confirm that the general condition of the hull is maintained at a satisfactory level. The Intermediate Survey is to be held at or between either the 2nd or 3rd Annual Survey.
Those items that are additional to the requirements of the Annual Surveys may be surveyed either at or,
between the 2nd and 3rd Annual Survey. In addition to the requirements of the annual survey, the intermediate survey contains requirements for extended overall and close-up surveys including thickness measurements of cargo and ballast tanks. The renewal surveys are carried out in five-year intervals with the
purpose of establishing the condition of the structure to confirm that the structural integrity is satisfactory
in accordance with the classification requirements, and will remain fit for its intended purpose for another
five-year period, subject to proper maintenance and operation of the ship. The renewal survey normally
covers overall and close-up examination, thickness measurements, and testing, and is aimed at detecting
fractures, buckling, corrosion, and other types of structural deterioration (IMO, 2004). A vessel can operate as long as requirements for class renewal are satisfied. Hence, provided that the technical condition is
sufficient, then, in principal, there is no limitation for how long the vessel can operate. Vessels are however subject to more comprehensive surveys as the age increases.
The Harmonised System of Survey and Certification (HSSC) adopted by IMO entered into force in
2000 to simplify the survey and certification process. The HSSC seeks to standardize the period of validity and the intervals between surveys for the nine main convention certificates to a maximum period of
validly for all certificates to five years. In practice, many administrations and classification societies already operated a form of harmonized survey and certification, but now all vessels under the jurisdiction
of IMO are subject to a unified survey regime.
Since compliance with the class rules is required for ships to remain in class, the class survey system is an
important starting point for how structural longevity is managed by the owner. There is however considerable variation among the owners on how this is handled. Some do not consider the longevity beyond the fiveyear class perspective. This is regarded as being sufficient as the satisfactory technical condition of the vessel is ensured through maintenance and repairs required during periodical surveys. A short time horizon may
also fit well with the economical perspective that is limited by the contracts they work under. Costly maintenance and repairs that are not required may therefore be postponed, but from a long term structural integrity point of view these actions would be reasonable to undertake. Alternately, other vessel owners may have
implemented a quality and safety philosophy that extends beyond the time horizon of renewal surveys.
Besides the mandatory inspection regime specified by class and additional measures taken by the owner, other requirements may be invoked by relevant national regulations, port authorities etc.
826
3.4.2
High-speed Craft (HSC)
Design and operation of high-speed craft is governed by the International Code of Safety for High-Speed
Craft (HSC code) adopted by IMO in 1994 and updated in 2000 (IMO, 2000). In 1994, IMO adopted a
new SOLAS Chapter X Safety measures for high-speed craft, which makes the HSC Code mandatory
for high-speed craft built on or after 1 January 1996 (IMO, 2014). With the development of many new
types of HSC in the 1980s and 1990s, IMO found it necessary to adopt new international regulations
dealing with the special needs of this type of vessel. The regulations take into account that a high-speed
craft is of a light displacement compared with a conventional ship, and consequently the regulations allow
for use of non-conventional shipbuilding materials, provided that a safety standard at least equivalent to
conventional ships is achieved (Hoppe, 2005).
Due to the special characteristics of high-speed craft all major classification societies have developed
specific rules for this vessel type where the HSC code requirements are implemented. The code does not
provide specific requirements related to materials, structural strength, cyclic loads, and design criteria.
The requirements implemented by the respective classification societies may therefore vary for these
topics.
Renewal surveys in five year intervals are specified by the HSC code. Moreover, the HSC also applies
for high-speed craft meaning that classed vessels are subject to annual, intermediate, and renewal surveys.
ABS Rules for Survey After Construction (2014c) requires annual drydocking and survey of passenger
vessels and high-speed craft.
3.4.3
Vessels operating in inland waterways
Inland vessels operating within one geographical zone (e.g. USA, Europe, or Russia) are not subject to
IMO regulations. These are subject to either national requirements or regional requirements such as e.g.
the European Code for Inland Waterways (CEVNI) issued by United Nations Economic Commission for
Europe (UNECE). Similar to international trading vessels, inland operating vessels are required to be
classed by a recognized classification society. Besides classifications societies operating globally, inland
operating vessels also receive class from local actors such as The Shipping Register in Ukraine (RU) and
Russian River Register (RRR). Classification rules of global actors like GL, BV, ABS, and Lloyds Register have implemented a survey regime that is identical or very similar to that of IACS. Annual, intermediate, and renewal surveys as described for international trading vessels are therefore often practiced for
vessels operating in inland waterways unless regulations enforced by the authorities indicate otherwise.
3.5
Offshore structures
There is a large variety in offshore units both in terms of functionality and design. This ranges from drilling units that operate on a global scale at multiple locations to floating or fixed units that are designed for
operation at one specific field their entire operational life. How the structural longevity aspect is handled
varies significantly and a description of the practice of the most common offshore units will be given in
the following.
3.5.1
Offshore drilling units
Typical mobile offshore drilling units are semisubmersibles, drillships, and jack-ups. The code for Mobile
Offshore Drilling Units (MODU) (MODU, 2001) has been set to be the governing code by the IMO. Offshore drilling units are, according to the MODU code, required to be classed, and are hence subject to a
five year renewal survey regime similar to that of commercial trading vessels with annual, intermediate,
and renewal surveys.
Class rules require offshore drilling units to be designed for at least twenty years of operation. The unit
is however allowed to operate further provided that it undergoes more comprehensive surveys. The use of
Risk Based Inspection (RBI) planning has by some classification societies been accepted as an alternative
to the extended surveys specified in the class rules. Inspection plans based on RBI assessments are often
less comprehensive than the standard extended class surveys for ageing rigs, and is a service that increasingly is being used by rig owners.
Operation of offshore drilling units is very cost intensive, thus the industry is very focused on having
the unit in operation as much as possible. Operation offshore is therefore planned for long periods before
it is scheduled for dry docking at a yard for maintenance and inspection, when the necessary repairs, often
required by the classification society, are made in order to be able to operate the next five years without
interruption. The time horizon seldom extends beyond the five year cycle, and repairs not required by the
classification society are often postponed until the next renewal survey.
3.5.2
827
Floating Production Storage and Offloading (FPSO) units
An FPSO is a floating vessel used for the processing of hydrocarbons and for storage of oil, receiving
hydrocarbons produced from nearby platforms or subsea templates, processing them, and storing oil until
they can be offloaded. FPSOs can be a conversion of an oil tanker or can be a vessel built specially for the
application. It is the owner's decision whether vessel is classed or not. It is quite common that construction or conversion is carried out under a class regime. However, some choose to take the FPSO out of
class afterwards when it enters operation.
The FPSO is often planned to operate continuously at the field. A situation where the unit must enter a
shipyard for repairs or other modifications would be disastrous from an economic point of view. The incentive for having a structural integrity perspective that covers the entire time in operation is thus strong.
This is especially evident in harsh environments like the North Sea where structures are prone to fatigue.
Inspections and repairs are often part of a long-term strategy to ensure the structural integrity of the unit
during operation.
3.5.3
Fixed production platforms
Fixed production platforms, mostly jacket structures but also concrete deepwater (or Condeep) platforms, are designed to operate at a specific location in a predefined period. Governing standards for design and operation are specified by the national authority responsible for the petroleum activities of the
Coastal state.
Operators of fixed platforms have often a long term perspective when planning for safe operation during the lifetime of the field, with additional measures which go beyond the requirements of the governing
design standards commonly implemented. Typically, more strict safety factors are used in fatigue analysis, even for details where this is not required, and additional corrosion tolerances are implemented in the
design. Inspection and maintenance is also often part of long term plans. If a situation should occur where
a critical damage is found, e.g. fatigue crack, it may take significant time before a repair can be carried
out. This will depend on how long it takes to mobilize and plan for the repair. Furthermore, difficult accessibility and harsh weather conditions, such as during winter season, can lead to a repair being be postponed until the season is over. This represents a significant risk for the operator compared to mobile offshore units or commercial trading vessels that can be transported to a yard for repair in a relatively short
time.
Many platforms set into production during the seventies and eighties were typically designed for an
operational life of twenty years but improved subsea and drilling technology has permitted operators to
take more resources from the reservoirs, leading to longer periods of operation. In order to justify operation beyond original design, comprehensive lifetime extension projects that document that the structure is
fit for extended operation and specify areas that need reinforcement have been quite common the last
couple of decades. Furthermore, through complete structural analyses with updated metocean data and
state of the art computational tools inspection plans of structural details that do not fulfill requirements
related to fatigue or corrosion tolerances are established. The NORSOK standard N-006 (NORSOK,
2009) was developed and published in 2009 to provide guidance to life extension assessments at the
Norwegian Continental Shelf.
Extending the lifetime is a relatively costly process that may include comprehensive inspection and
maintenance plans in the additional years of operation. There is now a tendency to design structures for a
structural life beyond that of the planned operation with a structural design that is more flexible with regard to possible extension of field production.
3.6
Naval vessels
Larger navies have from a historical perspective had extensive national standards covering all aspects of
shipbuilding and operation. Smaller navies on the other hand have not necessarily had the resources or
competence to develop a complete set of standards, and have built vessels based on standards they have
accessed through cooperation with other navies. This practice has gradually changed since the Cold War
ended, as naval budgets have decreased, making it a challenge to maintain and develop internal standards.
Smaller navies are especially vulnerable for this development as it may take several decades between
when a naval vessel is placed in service and when it is replaced. This is challenging because the experience and knowledge gained from earlier projects may no longer be available.
Over the past couple of decades there has been an increasing tendency that navies choose to use the
services of classification societies. In a traditional naval regime, the roles and responsibilities have been
more unclear, and it could be difficult to distinguish the different roles, as many of these were undertaken
by different units within the navy (Fredriksen, 2012). Navies that choose to class naval vessels also take
828
the benefit from both civilian and naval shipbuilding knowledge. As navies are increasingly using class
services, classification societies have begun developing rules adapted for naval vessels. Some examples
are; DNV rules for High Speed, Light Craft and Naval Surface Craft (DNV, 2005), ABS International
Naval Ships (ABS, 2014a) (ABS, 2010), Lloyds Register Rules and Regulations for the Classification of
Naval Ships (Lloyds Register, 2014a), GL Rules and Guidelines III, Naval Ship Technology
(Germanischer Lloyd, 2012), and BV Rules for the Classification of Naval Ships (Bureau Veritas, 2014).
Reference is made to ISSC Committee V.5 Naval Vessels for discussion and comparison of naval rules of
classification societies.
Naval vessels that are classed undergo the same survey regime as classed civilian vessels, and the navies have the same obligations as owners of civilian vessels to comply with maintenance and repairs required by classification society. Naval vessels outside class undergo a maintenance and inspection regime
specified by the standard used by the individual navy. Instead of involving a third party (class), the
process is of execution and control managed by the navy. It is still common, especially among the larger
navies, to have their vessels outside the class regime.
3.7
Conclusions
All vessels operating internationally are subject to IMO regulations. These regulations are enforced
though the rules of classification societies which are required to implement relevant IMO requirements.
Today the class survey system is nearly identical for vessels operating internationally regardless of vessel
type and classification society. Vessels not bound by IMO regulations also tend to be working under a
similar regime because these are classed by the same classification societies. The survey system is the
main basis for how structural longevity is handled by vessel owners. This may be due to both economical
and competence reasons. However, the most important reason is believed to be that safe operation during
the renewal cycle of five years is ensured by following the class requirements, though owners may have a
longer time horizon.
4.
PREDICTION OF LONGEVITY
4.1
Introduction
Failures in marine structures are prevented from occurring by implementing measures throughout their
lifetime and beyond, from the concept design stage to their final disposition. Failure prevention is of fundamental importance in the drafting of codes and regulations related to ship construction. In fact, it could
be said to be their raison dtre. The complete cycle includes feedback from ship operation, the
processing of information and the concomitant use of research results in the drafting of codes and regulations. Failure prevention measures are implemented at the design stage, the fabrication stage, and
throughout the lifetime of the structure. In order to be able to introduce failure prevention measures it is
necessary to have an accurate picture of the behavior of a structure in its operating environment. In this
way it is possible to predict the effects that individual factors that influence the behavior of a structure
have, so that the required longevity is achieved. The prediction of response is therefore a necessary step
before the development of rules and regulations, and is therefore carried out on a continuous basis. It may
also be carried out at the design stage of an individual structure, especially if it is a novel design. During
the operating life of a structure it may also be necessary in certain cases to predict the remaining lifetime,
or longevity, of a structure. In this, different requirements lead to different approaches being followed in
the maritime and the offshore industries.
4.2
In the maritime industry, in the overwhelming majority of cases no specific calculation is made of the
remaining lifetime of the structure. This is for historical reasons and also because of the economic environment within which merchant ships operate. The actual lifetime of a ship thus does not depend exclusively on technical factors and so a calculation on this basis is not of much practical value. This is
discussed in the next paragraph.
With regard to economic factors, it is expected that the investment in a newbuilding will be
amortized well before the expected lifetime of the structure is reached. It is market conditions and
the condition of the structure that determine whether a ship will continue trading or not. When
freight rates are high, overage ships find employment so that necessary repairs are carried out and
these ships continue trading. When freight rates are low, only newer ships find employment, so
that older vessels are condemned to the scrap yards. The fluctuation in scrap prices around the
world is thus market-dependent and is recorded and reported daily in the shipping press. Other data
829
reported include the average age at scrapping for each vessel type by organizations such as Clarksons shipbrokers.
A classic, extreme example of a modern fleet being condemned to the scrapyard is that of oil tankers in
the oil crisis of the early 1970s. New VLCCs that could not find employment sailed directly to the breakers from construction yards. More recently, during the 20042008 period when freight rates were exceptionally high, a large number of overage vessels were employed, thereby leading to a significant increase
in the average age at scrapping. A further, slightly different example is that of the Russian inland waterways vessels. In this case, older vessels cannot always be scrapped because of limited private investments
in the river fleet and slow fleet renewal. These river ships do not operate for four to five months each year
due to ice conditions and closed locks on rivers. So charterers employ older vessels in spite of the risks
involved to their cargo and to the vessels themselves.
For these reasons, the actual determination of the longevity of the hull structure on the basis of technical factors is not always an overriding priority. Classification society rules and statutory regulations do not
include any limitations on the operating lives of ships, and limit themselves to ensuring that the condition
is satisfactory for the period between successive Special Surveys as described in Chapter 3.
Having said that, in addition to market conditions, technical factors also play an important role in the
lifetime, or longevity, of a ship. The most important of these factors is fatigue, and because operating a
ship with unreliable structure can result in damaged cargo, underwriters are reluctant to or refuse to insure
cargoes carried in overage ships. As a result, these ships no longer find employment, are fraught with
damages that are costly to repair, and so are scrapped.
An example of a use of fatigue analysis for planning life extension of ship structure is provided by
Braidwood (2013). A single-hull tanker that had been converted to a floating storage and offloading platform was examined in 2007 for a 6-year extension of service life. In addition to an extensive structural
survey, a detailed fatigue analysis of the unit was conducted that considered the previous 20 year of service and the projected six additional years of service. As a result of that analysis, modifications were
made to the structure, which when surveyed in 2013 was found to be in good condition. The structural
surveys and fatigue analyses were repeated in 2013 to develop the measures needed to ensure an additional eight years of service without drydocking.
In the sections that follow the technical factors that affect structural longevity will be considered individually. These are corrosion, fatigue and buckling.
4.2.1
Prediction of corrosion
In the case of surface ships, the statistical approach is used to assess the current condition of the hull
structure vis--vis corrosion and also to predict the condition at some future point in time. This is
based on the rate of corrosion concept, which is in effect the diminution of thickness that takes place
during a specific period. The extensive data that have been obtained have shown that it depends on the
type of structural member, the type of compartment, the arrangement of compartments as well as the
type of ship. In order to avoid any loss of strength that may jeopardize structural integrity, classification societies have introduced the net thickness and the corrosion allowance concepts, both of which
are expressed in absolute terms. The net thickness required by classification societies is the original
thickness minus a corrosion allowance added to the calculated thickness during design and is thus the
thickness required to ensure sufficient strength throughout the lifetime of the ship. It therefore
represents the minimum value permitted under any circumstances. Rates of corrosion for different ship
types have been obtained by IACS and other international organizations. Since ocean-going ships undergo Special Surveys at 5-year intervals, it is necessary to allow for the expected loss of thickness that
will occur over this period. The total expected diminution is added to the net thickness and thus the
total member thickness is obtained.
The methods used to ensure structural integrity during a ships lifetime are permanent measures (such
as the use of protective coatings and cathodic protection) and occasional measures, which relate to inspection, maintenance and repairs, including replacement of corroded members. Corrosion prevention is discussed further in Chapter 5.
4.2.2
Fatigue strength prediction
Fatigue design has been introduced relatively recently and has been undergoing significant developments
in recent years. ISSC has maintained Technical Committee III.2 on Fatigue and Fracture, whose past and
newest reports should be considered an important resource in understanding this technical area. The rules
of different societies give different predictions of fatigue life partly because of the sensitivity of results to
the underlying assumptions made. Broadly speaking, fatigue design may be carried out at different levels
of increasing complexity and sophistication. The basic approach involves following rule guidelines on
830
local detail design. Elementary fatigue life calculations can be carried out using methods such as Miners
rule whereas the more sophisticated approaches involve finite element modeling of the hull girder with
particular attention to structural details. In general the remaining fatigue life of a vessel that has been in
operation for some years is not usually calculated, although such information, as discussed above and in
section 8.1.2 below, can be of benefit in forecasting maintenance costs.
4.2.3
Buckling prediction
The question of strength (yielding, buckling) has been addressed using the principles of mechanics
since the development of large tankers during the 1970s. The problems that arose at that time made it
clear that it was necessary to abandon empirical formulations and make use of engineering principles
to determine strength. Later on, when high tensile steel was introduced, the problem of fatigue came to
the fore and fatigue design methods started to be introduced. The most recent development in structural
design concerns the determination of the ultimate strength of the hull girder in vertical bending. The
recently introduced International Association of Classification Societies Harmonized Common Structural Rules for oil tankers and bulk carriers include a step-by-step procedure to calculate ultimate
strength that is based on beam theory, but which allows for progressive collapse of the hull girder. It is
important to realize that failure does not occur instantaneously throughout the section but in a progressive manner. As the hull girder bending load increases, individual stiffened plate elements fail incrementally and thus have reduced effectiveness in contributing to the overall hull girder resistance to the
bending loads. The method can also be applied to predict the ultimate strength throughout the lifetime
of a ship, provided data concerning scantlings are maintained and updated. It can also be used to determine residual strength following accidents such as grounding and thus enables classification societies to provide advice as to the handling of the vessel and the necessary repairs. The ISSC Technical
Committee III.1 on Ultimate Strength should be referenced for greater insight into the methods appropriate for capturing this failure mechanism.
4.3
When discussing the prediction of longevity of a structure including the fatigue life, it should be recognized that the nature of fatigue is cumulative and local. Hence, the only thing that can meaningfully be
stated is that the fatigue life at a specific location in a structure has been consumed when exposed to a
given load history. Local fatigue failure will in a redundant structure generally not pose an immediate
danger to its survival, Vugts (2013). Hence, the longevity of a structure is not necessarily equal to the
minimum fatigue life of a given component.
The design life (expected longevity) is normally defined by the platform operator/owner at the design
stage. A general approach is then to design the structure so that it can sustain all design loads for the service, ultimate, accidental and fatigue limit states for the given design life, which includes that all hotspots
have a fatigue life which is larger than the target fatigue life at the given location. The safety against fatigue put into the design depends on the subsequent inspection philosophy defined by the operator and the
minimum safety defined by the design codes such as the guidelines in ISO 19902 (Din, 2007) or API and
DNV-GL rules. In general the approaches are well established. However, a survey presented by May
(2014) showed a request for more guidance by the codes for managing short fatigue lives and handling
corrosion and material degradation.
4.4
Conclusions
Prediction of structural longevity requires the estimation of the time-varying probability of failure from
fatigue, buckling, or corrosion, but examples of the systematic application of the principles required to do
this are lacking. Failures in marine structures are prevented from occurring by implementing measures
throughout their lifetime and beyond, from the concept design stage to their final disposition. The ISSC
Technical Committees explore the key elements essential to prediction of structural behaviors that affect
longevity, including fatigue failure, buckling and degradation such as corrosion that might lead to failure.
Prevention of failure is of fundamental importance in the drafting and implementation of codes and regulations related to ship construction. Failure prevention measures are implemented at the design stage, the
fabrication stage, and throughout the lifetime of the structure, based upon an understanding of failure
mechanisms and modes and their prediction.
5.
PREVENTION & REPAIR OF STRUCTURAL FAILURES
5.1
Introduction
831
This chapter presents a general overview of the latest methods applied in the offshore and naval societies
for repair, and even more importantly, the prevention of structural failures with the aim of providing input
to the operators for safe operation of their assets. Prevention of structural failure is a manifold task, which
starts at the design stage and continues during operation of the asset by proper inspection and maintenance. Naturally, the owners of both ship and offshore structures aim to avoid failures. But, the approaches
differ for the two communities throughout the life-time of the structures.
5.2
Prevention of failure - Design stage
The design phase is the optimal time to incorporate measures for minimizing the risk of failure. Hence, it
is important that the right basis and tools are available for design considering the full service life of the
structure. In the following sections some of the key issues for the prevention of structural failure in the
design phase are discussed.
5.2.1
Experience from the shipping industry including mobile offshore units shows that corrosion accounts for
more than half of the total damage and that damage from corrosion builds up with the age of the vessel,
(Ayyub et al., 2000). Hence, in a corrosive environment to which most ship and offshore structures are
exposed it is of major importance that the risk of corrosion is dealt with throughout the lifetime of the
structure. At the design stage, corrosion allowance combined with coating can be included in certain areas
such as the splash zones for offshore platforms and tank areas for ships. Proper coating and/or cathodic
systems are normally selected for the protection of the rest of structure.
The correct coating system should be applied at the fabrication yards, as the cost of repair can be relatively expensive, e.g. a new build FPSO can be delivered with coating cost in the order of $10 per coated
m2 while the cost for offshore repair is in the range of $500 to $3,000 per coated m2 depending on the
cause, extent, and location of failure (Kattan et al., 2013). Kattan et al. presented some of the aspects
which should be focused on in the selection and application of coating systems. These include: surface
preparation, application of coating, correct specification accounting for condition under which the coating
is applied, and the design of the structure. Kattan et al. state that, as a rule of thumb, coating failure on
edges are 7 times more likely than on a flat plate. Hence, from a coating point of view, simple designs
with few sharp edges and corners are very beneficial.
An important input for the design of coating systems is the expected coating breakdown process and
timeline. Material has been published in the past concerning average annual corrosion rates of ship and
offshore structures, which should be accounted for in assessing the structural longevity of these assets.
However, those average corrosion rates need to be considered in view of the increased ability of coating
systems to completely protect structure for long periods. In one study by Slebodnick et al. (2013), highsolids coatings applied to the ballast tanks of a US Navy ship were observed to be in excellent condition
after 15 years of service, with strong indications that they would remain so for the projected 20-year service life of the coatings. A new coating system in the ballast tank may effectively increase the lifetime
when the quality is better than standard (De Baere et al., 2013a, De Baere et al., 2013b). In these economic evaluations, coating is seen as superior lifetime extension option over added steel thickness or steel
with reduced corrosion. Improving the impact resistance will enhance the effect further.
Taking the development of coatings even further, it is very desirable to develop coatings that can provide corrosion protection but also include possible early detection of corrosion. A tremendous amount of
research focuses on smart coatings that can sense corrosion and respond. An overview of recent work
within this area is presented by Wheat (2012). In the development of smart coatings some of the main
challenges are that indicators must be incorporated without reducing the protectiveness of the coating and
that indicators must be active/sensitive over a prolonged period of time. The ultimate goal is to have coatings that can provide prolonged corrosion protection and at the same time monitor damage and detect
and/or repair damaged coatings.
Despite the efforts mentioned above, it is not is not always economically possible to fully avoid corrosion. Hence it is necessary at the design stage to consider inspectability and repairability. For ships, special problems arise in relation to the particular conditions in various compartments such as ballast spaces,
cargo tanks and holds. In order to develop a comprehensive and effective anti-corrosion plan, it is necessary to consider the whole lifecycle of a ship and to simulate the conditions that develop in various types
of compartments under the environmental, cargo and hull structure conditions that prevail.
832
5.2.2
Material selection
The selection of material is of importance when failure due to brittle fracture is to be avoided. Within the
offshore industry, normally relatively ductile materials are used in cyclically loaded structures with strict
requirements on the test temperature relative to the operation temperature. This ensures ductile redundant
structures. Additional safety can be built into the designs by the selection of SUF materials (Surface layer
with Ultra-Fine grained microstructure). These materials, which introduce possible crack arrest by the
material itself, are now commonly applied in the building of large container ships. Hence, effective guidelines/procedures for the determination of crack arrestability for SUF materials are needed. Ishikawa et al.
(2012) worked on a test method based on small scale specimens for the determination of crack arrestability (Kca). The method revealed good results when applied for EH47-class steel.
An alternative or a supplement to the use of corrosion protective coatings is the use of Corrosion Resistant Steels (CRS). Such steels have been developed for many purposes including protection of ballast
tanks for cargo oil tankers, as described in Yamaguchi and Terashima (2011). Focus has been on steels
which: a) have chemical compositions and mechanical properties that fulfill the existing requirements
from the classification societies; b) have elements added to the steels for corrosion protection that are to
within 1% of the full batch in order to ensure sufficient weldability; and c) do not initiate problematic
galvanic corrosion with conventional steel applied in the cargo oil tankers. Pitting corrosion is one of the
major forms of corrosion found in the crude oil tanks. Ito et al. (2012) developed methods for laboratory
testing of pitting corrosion and conducted on-board evaluation of Corrosion Resistant Steel applied for
bottom plates in VLCCs. Comparing conventional coated bottom plates with bottom plates constructed
from CRS, it was found that pitting corrosion is significantly reduced using CRS and the overall corrosion rate was lower than when using conventional coated steel.
5.2.3
Structural design
The key to avoiding failures is a proper design which considers the full lifecycle of the structures, i.e.
service, fatigue and ultimate limit states and in some cases also abnormal events. Additional safety put
into the structure at the design stage can come manifold back at a later stage, i.e. if you think safety is
expensive, try an accident.
One of the main causes of failures is fatigue induced cracks. The stresses that lead to fatigue failures in
the structure are cyclic and normally initiate at singular points of the structure, i.e. at stress concentrations. As a consequence, the immediate measures that can be taken against fatigue cracks are either to
reduce the cyclic stress loading or to release the stress concentrations. Generally, this can be obtained by
smooth transitions and limited offsets in the design. As an example for a ship structure, this can be accomplished in the following ways (see also Figure 1.):
Increase the dimensions in order to reduce the nominal stresses
Use symmetrical stiffener cross-sections (e.g. T-sections) to avoid the development of bending stresses
due to torsion that arises in asymmetrical sections
Use larger transitional sections and limitations to geometrical discontinuities in order to reduce stress
concentrations
Add supporting brackets on the opposite side of existing tripping brackets. Their use can bring about a
reduction in stresses of up to 35 per cent
Replace the existing bracket with one whose toe has been suitably shaped and add counter-supporting
brackets. This can bring about reductions in the local stresses of up to 65 per cent
The grinding of the weld at the bracket toes also contributes to reductions in the stress concentrations
and can increase the useful life of the structure by up to 100 per cent.
In recent years extensive catalogues of design improvements in order to reduce fatigue failures have been
made available for standard details. Such guidelines are included in documents issued by the Ship Structure Committee, Lloyds Register of Shipping, and the Tanker Structures Co-operative Forum for bulk
carriers and tankers, see e.g. SSC-405 (Glenn, 1999).
Finally it should be noted that experience shows that gross errors are more likely for new types of
structures or details than if well proven designs and fabrication methodologies are used (Lotsberg and
Sigurdsson, 2014), i.e. conservative designs are less likely to fail. Hence, special focus shall be made
when nonstandard approaches are applied such as excessive use of high strength steel.
833
Avoidance of discontinuity with bracket and

properly shaped toe
Addition of supporting bracket to a tripping bracket
Use of a doubler plate with shaped edges and

increased area
Use of a shaped stiffener and supporting brackets
Figure 1. Methods of improving the structure at a local level.
5.3
Prevention of failure - Operation
5.3.1
Maintenance & inspection
Inspections play a major role in the prevention of structural failures. Ocean going merchant ships undergo
regular inspections including special surveys every 5th year following an inspection program. Repairs and
maintenance are carried out including reestablishment of coating and cathodic protections as required
following guidelines such as those produced by IACS. 3D software can be used for the storing of information and assist in optimization of the inspection plans, see e.g. Witte et al. (2011).
With a focus on ship structures, the subject of corrosion is reviewed in the MegaRust symposia held
annually by the American Society of Naval Engineers, see e.g. Gonzlez Nez et al. (2013) who investigated detection of corrosion using acoustic emissions, where the emissions occur as a result of stresscorrosion cracking in the corroded areas. The technique was demonstrated in a number of situations, including laboratory experiments and in a railroad tank car. Underwater inspection to detect corrosion
through thickness readings using remotely operated vehicles was described by LeHardy and Walters
(2013). In one test case on a naval vessel, internal inspectors identified five regions of reduced hull thickness, but subsequent inspection with an underwater apparatus found sixteen regions of reduced thickness,
and a final internal inspection verified 10 of those 16 regions, with the remaining six regions not readily
assessable internally. The cost of inspections can be relatively high due to the cost of preparation for inspection. Nelson et al. (2012) presented some experience on the use of a corrosion sensor-based tank
monitoring system combined with an Insertable Stalk Inspection System (ISIS) as a means to reduce requirements for manned entry into tanks and voids during their inspections. These systems are so promising that they are now implemented in several US Navy vessels including new ship constructions.
For offshore structures, the level of inspection and maintenance can be determined at the design stage.
The requirement for inspection (inspection intervals) will depend on the fatigue resistance of the platform
and the safety put into the design. Often, a Risk Based Inspection (RBI) plan will be established and updated as the conditions of the platforms change due to modifications, unplanned events or extension of
the service life. The RBI is a cost efficient way to establish an inspection strategy that legislative and
operator requirements for safety and environmental risk are fulfilled and documented. The RBI approach
is a condition-based approach by which the inspection effort is adapted to the condition of the item and
prioritized in accordance with the importance of the individual items. The objective to the RBI methodology is to give feedback on where to inspect, what to inspect, how to inspect and when to inspect. Inspections, inspection techniques and structural integrity management systems are further discussed in chapters
6 to 8.
834
5.3.2
Repair and rehabilitation
During operation many ships and offshore structures will experience the need for strengthening, repair, or
modification to ensure safe operation. Typically, damage will be detected during inspection of the structures. When a structural failure is detected, a decision must be made as to the most cost effective repair.
This includes an assessment stage for the structure before deciding which, if any, Structural Modification
and Repair (SMR) techniques are to be used. An overview of possible repair methods is outlined in the
following starting with methods for offshore structure. Many of the techniques for repair and rehabilitation apply to both offshore and ship structures.
5.3.2.1 Fixed offshore structures

For fixed offshore installations, the initiator of SMR can be divided into the following categories: 1)
change of platform operation, 2) lifetime extension 3) code updates or 4) damage including vessel collision, dropped objects, fatigue, corrosion, fabrication faults or unintended operations. A large number of
SMR techniques exist. The most common industry adopted methods in the offshore industry are welded,
clamped, and grouted repairs/modifications. A very thorough discussion on SMR techniques can be found
in MSL (Dier, 2004) including a table with pros and cons for the different techniques, as reproduced in
Table 2. An evaluation of requirements in terms of offshore equipment and timescale, costs, and loading
penalties is included in MSL (Dier, 2004).
From the work presented in MSL (Dier, 2004), it can be concluded that grouted clamps offer both
technically and economically attractive solutions for SMR work. Consequently, grouted repairs are used
widely to provide additional strength, prevent propagation of dents or buckling, strengthening of legs, and
clamped repairs for load transfer. Grouted connections have the major advantage that these can be utilized
to accommodate geometrical imperfections and installation tolerances. With focus on structural integrity
control of ageing structures, the current developments of grouted connections is evaluated by Samarakoon
et al. (2013).
Table 2. Pros and cons for different SMR techniques, reproduction from MSL (Dier, 2004).
Technique
Defect
Fatigue
crack
Non-fatigue
crack
Dent
Corrosion
Inadequate static
strength
Member Joint
yes(1)
yes(1)
yes(1)
yes(1)
no
no
yes
yes(3)
yes(3)
Dry welding yes(1)
yes
yes(3)
yes(3)
Wet welding no(2)
no
no
no
Toe grinding no
Remedial
yes
yes(1)
no
no
no
no
grinding
Hammer peenno
no
no
no
no
no
ing
Stressed meyes
yes
no
yes
yes
no
chanical clamp
Unstressed
yes
yes
yes
yes
yes
yes
grouted clamp
Stressed
yes
yes
yes
yes
yes
yes
grouted clamp
Neoprene-lined
no
yes
no
yes
no
no
clamp
Grout-filling of
no
no
yes
yes
yes
yes(4)
members
Grout filling of
no
no
yes
no
yes
yes(4)
joints
yes
no
no
no
no
no
Bolting
Member reyes(5)
yes(5)
yes(5)
no
no
yes(5)
moval
yes
yes
yes
yes
yes
yes
Composites
(1) Usually in conjunction with additional strengthening measures
(2) Except to apply weld beads in unstressed grouted connection/clamp repairs
(3) To apply patch plates
(4) Applicability depends on type and sense of loading
(5) If member is redundant (otherwise replace it)
Inadequate fatigue
strength
High loads Fabr. Fault
yes
no
yes
no
yes
no
no
no
yes
no
yes
yes
yes
yes
yes
yes
no
no
yes
(4)
no
yes(4)
no
no
no
yes(5)
yes(5)
yes
yes
835
5.3.2.2 Repair methods - ships

For ships the repair methods also vary widely. They can range from temporary cold patches to stop leaks
to complete re-design of the structural detail and replacement of steel nearby the detail. Welding cracks is
a popular repair, but the detail frequently fails again within a short time. Drilling the ends of the cracks is
another frequently used temporary repair measure that is used until the ship is dry-docked.
For a fatigue crack in a particular structural detail, there are several methods to repair it. The expected
repair life and repair cost of each repair method varies. Ship owners usually choose the most cost- and
time-effective method. A robust, but extremely expensive repair method may not be the best alternative.
A less robust and cheaper repair may not be favored either, because later the repair may fail again. It will
cost even more money to repair the detail again and again. Selecting a repair alternative requires a large
measure of judgment and engineering insight. The general strategies for crack repair of critical structural
details can be classified as follows:
Re-weld the cracks to the original construction: Gouging and re-welding is an easy and common way
of repair. However, the repaired weld can/will create new crack potentials and thus fail again in a
shorter time interval.
Re-weld the cracks plus post weld improvement: This repair is basically the same as the previous
one, but the life extension effect of post weld improvement can be significant.
Replace the cracked plate: The existing plate is replaced with an identical new plate. However, if the
loading history and dimensions are identical to those of the failed plate, its fatigue life should be
about the same as the failed time of the crack.
Modify design by adding a bracket, stiffener, lug, or other plate: The more robust way of repair is to
modify the local geometry to reduce the stress concentration (see Figure 1). While adding a detail
component and not cropping a large section, this repair may be one of the best. It can reduce the
stress concentration and therefore increase the repair life significantly.
Change configuration by applying a soft toe, increasing radius, trimming face plate, enlarging drain
holes, etc. This is another way to modify the local geometry to reduce the stress concentration. If a
longer life continuance is expected for the ship, a more robust repair such as this should be considered.
Enhance scantlings in size or thickness: Increasing the size of a detail like a bracket is good. If discontinuity in thickness is introduced this should be carefully located outside the high stress area.
It is difficult to define which repair method is most reliable and cost effective for a particular crack.
The selection of a particular repair alternative depends on the location of the crack and the expected life
of the ship. Comprehensive lists of recommendations are nowadays available for most types of ship structures, see e.g. IACS (2007b), IACS (2005) and IACS (2007a) which cover double hull oil tankers, container ships and bulk carriers respectively.
5.3.2.3 Bonded patch repairs

A relatively new attractive efficient method for repair of corroded or cracked metal structures is bonded
patch repairs. This type of repair is based on restoring the strength of the component by adding a bonded
patch to the damage area. As an example, several ships of the US Navy CG47 Class have experienced
sensitization of the 5456-H116 aluminum plate in their superstructures, which leads to stress corrosion
cracking that cannot be repaired by welding because the stresses caused by shrinking associated with
welding results in cracking of adjacent plate. Initial design and testing of a glass reinforced plastic (GRP)
composite patch repair system for sensitized aluminum plate was performed, which was followed by
shipboard repair with a system that was simple to install with minimal investment in specialized equipment, time, and technical expertise, with more than 120 m2 (1,300 ft2) of composite laminate laid down
on the upper levels of aluminum superstructures of ten ships (Noland et al., 2013). In-service composite
patch performance and results of reinforced aluminum center-cracked tension specimens testing to date
provide confidence and data supporting the long term use of composite patches. The results of some of
this testing and analysis were presented by Hart et al. (2014).
A European project on composite reinforcement of ship structures was the CO-PATCH project which
involved 15 organizations representing eight countries. The primary goal of the project was to define and
demonstrate composite patches as an effective repair or reinforcement method for ships and bridges with
defects that can be environmentally stable and permanent. Areas investigated included the business and
regulatory requirements and application cases for composite patch repairs and development of guidelines
for the analysis and design of composite patch reinforcement and repair.
836
Another European project was BONDSHIP (Weitzenbck and McGeorge, 2005) which focused on the
introduction of adhesive bonding into shipbuilding as an industrial process for joining of lightweight and
dissimilar materials and structures. The project resulted in guidelines for design, modeling, testing, production, fire protection, inspection, and repair of bonded joints; acceptance tests and criteria; test and
inspection methods for bonded joints; material data; documented application cases and joint designs; and
documented production and assembly procedures and practical experience and skills from using adhesives in a shipyard. The project resulted in about twelve technical papers published by project participants, but the culminating document was the project guideline for the design, fabrication, operation and
qualification of bonded repair of steel structures issued by DNVGL, Det Norske Veritas AS (2012a). This
guideline also outlines some of the challenges with the use of bonded repairs, e.g. the provision of correct
surface preparation for the patch repair, the fact that the repair makes inspection very difficult, and the
insufficient long-term experience with patch repairs. Hence, these repairs are recommended to be limited
to noncritical components or on critical components with noncritical defects (Echtermeyer et al., 2014).
The patch repairs can also be used to postpone critical repairs until the next planned maintenance.
5.4
Prevention of structural failures is a manifold task, which starts at the design stage and continues during
operation of the asset by proper inspection and maintenance. Some of the key issues and new developments are listed below:
The design phase is the optimal time to incorporate measures that ensure the desired longevity of the
structure. Additional cost spent at the design/construction stage can come manifold back, as the cost
of repair can be very high compared to original fabrication cost
Use of Corrosion Resistant Steels (CRSs) which can be applied directly using existing codes looks
promising. Full scale testing shows that, relative to conventional coated steel, pitting corrosion is significantly reduced using CRS and the overall corrosion rate was lower
A tremendous amount of research is focusing on smart coatings that can sense corrosion and respond.
The ultimate goal is to have coatings that can provide prolonged corrosion protection and at the same
time monitor damage and detect and/or repair damaged coatings.
Cold patch repairs are becoming an attractive alternative to welding based repairs A guideline has
been established (DNV-RP-C301). Due to limited lifetime experience, the technique is recommended
to be applied for noncritical components or on critical components with noncritical defects. Research
is recommended to extend the lifetime of these repairs so that they may become permanent.
6.
INSPECTION METHODS & TECHNIQUES
6.1
Introduction
This chapter describes current practices and trends in structural inspection approaches and execution to
evaluate the state of the structure from inception to end of life, including deployed methodologies and
techniques, and their inherent capabilities and limitations.
From the perspective of classification, legal requirements and proper asset integrity management, structural inspections are a necessity to safeguard structural longevity. In general, the outcome of periodical
and event-driven asset inspections provide input for the determination of the components (compiled)
probability of failure, which is combined with the consequences of failure to provide a risk profile and
future inspection scheme to prevent incidents, maintain a specific safety level and to enhance design and
operational practices (such as future inspections) through feedback. Hence, in essence, current practices
(still) consist of the a-priori determination of technical and organizational measures to ensure future economic system effectiveness and safety. Measure optimization is generally done by posterior analysis on
correlation and causality of usage, external influences, and costs to improve the knowledge on physical
system degradation, predict the future behaviour, and further refine the measures accordingly (Tammer
and Kaminski, 2013). By linking this understanding of degradation propagation with the classification of
the inherent risks of this process and the consequences of failure, a more specific inspection plan can be
made as an alternative for prescriptive practices - which could be unsuitable for a specific asset design
and/or operational context (over- or under stringent). In essence, this is the foundation on which practices
such as Risk Based Inspection are based (Tammer and Kaminski, 2013).
The highly stochastic nature of the aging process has provided a multitude of models and inherent uncertainties that emphasize a probabilistic foundation. However, despite considerable developments in both
structural reliability theory and computational methods, the probabilistic approach has gained little
ground on the deterministic practices. The lack of acceptance of probabilistic methods for the assessment
837
of aging may be assigned to the complexity and computational effort concerned with the approach, and
the long absence of research into practical applications. As a consequence, most operators of ship and
offshore structures base decisions regarding inspection, repair and maintenance efforts primarily on empirical procedures, ergo, inspection. Hence, structural inspection practices are deployed as the key instrument to assess the actual asset integrity by identification and mitigation of system anomalies and unanticipated defects to ensure structural longevity and an adequate level of safety to comply with statuary
rules and company guidelines (Berg et al., 2014). With this in mind, the general perspective of inspections is logically based on empirical findings.
6.2
At this time, close visual inspection is deployed as the de-facto and dominant standard for initial inspection, and is the first step used to detect damage before a more refined assessment takes place with a suitable testing methodology. In general, the criterion and scope includes:
Overall high-level inspection on component level

Close-up visual inspections of selected and prescribed details
Suspect areas examination
Critical area inspection including fatigue hotspots
Corrosion measures inspection (coating condition and anode inspection)
The execution is guided by asset-specific checklists, which are constructed as an execution guideline and
used to collect textual descriptions and empirical evidence (such as photographic material) of the asset state.
The inspector(s) assess the condition based on prescriptive instructions from these checklists, additional
(class) guidelines and experiences gained combined with their judgement and experience. The recording of
the findings differs per situation (and industry), but is often qualified by terms as good, fair or poor for
general cases and in more refined and quantifiable parameters in the case of critical details or the extent of
the damage found. Note that classification societies often require an approval of service suppliers for performing surveys, NDT/NDE tasks etc. In addition, Kalghatgi et al. (2009) presented a universally adaptable
Hull Inspection and Maintenance Program (HIMP) devised by the American Bureau of Shipping (ABS),
which aims to improve the structural and maintenance efficiency through the implementation of zone-based
assessment and grading scheme based on six inspections criterions: the state of the coating, general corrosion, pitting and grooving, deformation, fatigue and cracks and general housekeeping and cleanliness.
6.3
Table 3 refers to the current dominant inspection methods categorized in visual, acoustic and magnetic
techniques.
Table 3. Primary Inspection Techniques.
(Enhanced) Visual and Radiographic

Strain
Surface strain (material deformation) gauges are bonded to reference points on the evaluated
Monitoring
structure to measure loads through elongation between these points. Traditionally, relatively low
sensitivity and drift properties limited the operational monitoring time. However, more advanced
and durable applications emerge, such as deployed in the Monitas II project (Kaminski and
Aalberts, 2010) and through embedded (in-situ) gauges, which are often used in composite materials like carbon-reinforced structures (Okuhara and Matsubara, 2005).
Thermography Constitutes the thermal mapping of a material surface to investigate its facial and/or its subsurface
status that may affect its performance. This is done using photonic or bolometric arrays sensitive
in the (113mm) band of the electromagnetic spectrum (Maldague, 2001). It has effectively been
used to detect latent corrosion in different steel structures (Morel et al., 2001).
Liquid (Dye) Is employed for the detection of open-to-surface discontinuities in a non-porous material and
widely used for the testing of non-magnetic materials. In this method a liquid penetrant (either
Penetrant
visible dye penetrant or fluorescent variant) is applied to the surface. The penetrant remains in the
discontinuity and indicates the presence as well as the location, size and nature of the discontinuity
(IAEA, 2000), e.g. lack of fusion or porosity in welding and also for leak testing applications.
X-Ray
X-Ray radiography (sometimes abbreviated XR) is the most commonly used radiographic NDTRadiography method. X-Rays are used to take a shadowgraph image of the sample, and shades in the shadowgraph
show the attenuation to the signal while it has passed through the corresponding spot in the sample. In
the X-Ray backscatter method (XB) both the X-Ray source and the detector are on the same side of the
sample (Vaara and Leinonen, 2012). The method is very suitable and widely used to detect cracks and
voids in steel structures and welding and also to detect delamination and porosity in composites.
Acoustic Techniques
838
The basic principle of ultrasonic testing is that an ultrasonic sound wave (0.5 - 20 MHz) is introduced in the concerned material and when a discontinuity is present in the wave path, part of the
sound energy will be reflected back from the flaw surface. The location of the flaw can be easily
determined since the direction of the wave and the signal travel time are known. Sometimes other
flaw properties like its size and orientation can be retrieved from the reflected signal. Advantages
consist of the sensitivity to both surface and subsurface discontinuities, the depth of penetration,
and high accuracy in determining location and size of flaws whilst providing instantaneous results.
The primary drawback consists of the necessity of a coupling medium to promote the transfer of
sound energy into the test specimen (van der Horst et al., 2013).
Guided Wave Ultrasound (GWUT) employs mechanical stress waves that propagate along an
elongated structure, guided by the dimensional boundaries of the structure. This permits the ultrasonic waves to travel a long distance with minimum loss of energy and has the potential to achieve
very long range inspection by a single scan at a single sensor location (an entire hull plate) compared to the point-by-point scanning of conventional ultrasonic inspection methods. Due to the
low wave frequency (10-100kHz), the operational principles of GWUT are fundamentally different from conventional. Despite seemingly superior performance, there are a few known disadvantages/issues, of which are a high dependency on the operator skills, difficulty in detecting very
small pitting defects, and the lack of sound procedural protocols. Finally, the method is ineffective
in thick material structures due to the very low frequency wave.
Phased Array Ultrasound (PAUT) is very effective for wall thickness measurements combined
with corrosion testing as the phased arrays in a variety of frequencies allows for more specific
steering of the measurements in many angles and depths.
One of the most reliable ways to detect the early stages of stress corrosion cracking and is commonly
Acoustic
deployed for the inspection of storage tanks, pipelines and offshore structures. However, widespread
Emission
adaptation/application in the shipping industry still has not been seen, although several on board trials
have been conducted in the recent years by ABS (2007b) and MISTRAS (2009), Tscheliesnig (2006)
and as described by Baran et al. (2012) on fatigue estimation and corrosion assessment of ships.
Magnetic Techniques
Eddy Current Is widely used to detect surface flaws, to measure thin walls from one surface only and to measure
thin coatings in electrically conductive materials. Electric currents are produced in the structure by
bringing it close to an alternating current carrying coil. The alternating magnetic field of the coil is
modified by the magnetic fields of the eddy currents. This modification, which depends on the condition of the part near to the coil, is then shown as a meter reading or cathode ray tube presentation.
Besides surface probes, internal probes can be used for the in-service testing of e.g. heat exchanger
tubes and encircling probes for the testing of rods and tubes during manufacturing (IAEA, 2000).
Magnetic
This method is capable of detecting open-to-surface and just below the surface flaws. In this
Particle
method the test specimen is first magnetized, either by using a permanent or an electromagnet, or
by passing electric current through or around the specimen. The magnetic field thus introduced
Inspection
into the specimen is composed of magnetic lines of force. Whenever there is a flaw, which inter(MPI)
rupts the flow of magnetic lines of force, some of these lines must exit and re-enter the specimen.
These points of exit and re-entry form opposite magnetic poles. Whenever minute magnetic particles are sprinkled onto the surface of such a specimen, these particles are attracted by these magnetic poles to create a visual indication approximating the size and shape of the flaw (IAEA,
2000). A major disadvantage of the MPI-method is that it does not allow a quantitative measurement of the leakage field in the vicinity of a flaw.
Magnetic
The magnetic flux leakage (MFL) method that is derived from the MPI method resolves this probFlux Leakage lem by using a magnetometer instead of magnetic particles to detect the perturbed flux lines (Jiles,
1990). With the magnetometer, which is most often a Hall probe, it is even possible to measure the
(MFL)
field components in three directions, perpendicular and parallel to the flaw and normal to the surface (van der Horst et al., 2013).
Alternating
Is a non-contact electromagnetic technique used for identifying, localising and sizing surface deCurrent Field fects on metallic surfaces without the need for further evaluation by techniques, such as ultrasonic.
Measurement It is based on the quantification of magnetic field disturbances, which arise when an electric cur(ACFM)
rent is disturbed by the presence of a surface breaking crack or a defect. The ACFM is a relatively
recent technology widely used for structural weld inspection of offshore structures; however several initiatives are being pursued to incorporate ACFM in modern ship hull inspection schemes.
The technique has the ability to detect defects through several millimetres of surface coatings on
both (non)ferritic metals and up to 25mm with a high degree of accuracy.
Ultrasonic
Inspection
6.4
Limitations
The outcome of the inspections and hence quality of the data and information gained is limited by the
Probability of Detection (PoD), which is influenced by the deployed methodology and technology, circumstances during inspection execution, human factors, and the inspectability of details due to limitations
such as accessibility or technological constraints. There are several references, which denote specific
839
PoD-curves for different inspection methods and -scenarios such as the HSE Offshore Technology Report 2000/018 PoD/PoS curves for non-destructive examination (HSE, 2002) and the ICON database
(HSE, 1996) or other references and experiences gained. The work by Demsetz and Cabrera (1999) for
the report SSC-408 Detection probability assessment for visual inspection of ships shows the factors and limitations of human inspection and the PoD (Tammer and Kaminski, 2013). In addition, it is
important to note that the PoD-curves are often based on technical constraints and it should be considered
that:
Inspections are performed by qualified and trained inspectors. Note that the framework for both competency/qualification and execution is often limited
If qualified, intrinsic human limitations and external factors have a significant influence
Inspection outcomes often have an inherent subjective element of subjectivity due to interpretation
Non-ideal conditions due to confined spaces, accessibility, weather etc. also pose significant influences on the inspection outcomes.
A review from the SSC-355 (Shinozuka, 1990) showed that there is very little information available to
assess the reliability of inspection techniques, as a performed case review delivered only one (1) reliable
PoD-curve.
6.5
There are many options available to the inspection practitioner, each with its own detectable phenomena,
probability of detection and cost of employment. Guidance on inspection techniques trade-space is provided in this chapter, but it is important to emphasize the significance of the inspection paradigm. Inspections are carried out as a result of the intersection between the economic principle of reasonableness and
the fact that unnecessary, disruptive and costly inspection and maintenance could result in unintended and
expensive downtime, subsequent damage and inherent safety, health and environmental risks. Hence,
best-practices should be deployed to approximate the optimum of efforts to limit these risks and safeguard
structural longevity.
7.
SENSING TECHNOLOGIES
7.1
Introduction
Sensing systems provide primary information for the purpose of assessing the state of structural health.
Systems can directly measure strain, acceleration, temperature, acoustic emission, pressure and so on, and
indirectly measure such things as the location and extent of some kind of damage. Progress in electronics,
optronics, mechatronics, material sciences, and processing technology and fusion of them (e.g. microelectromechanical systems, MEMS) has been continuously developing various aspects of sensors. A measurand is detected by sensors in the sensing system. The varieties of technologies for detecting a measurand
have been expanded and many sensor options are available for a sensing system design. For instance,
strain can be measured by not only conventional electrical-resistance strain gauges but also PZT (lead
zirconium titanate ceramics) and PVDF (polyvinylidene fluoride) sensors, fiber-optic sensors, and so on.
Sensors can be multiplexed with wireless sensor networks and distributed fiber-optic sensor technologies
to increase measurement area/density or multifunctional capability.
If the sensing system is monitoring changes in physical parameters of a structure only by sensors that
the structure is equipped with, using structural responses and behaviors resulting from operation, we can
call this action passive sensing (Balageas et al., 2006). Most field-deployed structural health monitoring
strategies for civil engineering structures are using a passive sensing system (Farrar and Worden, 2012).
On the other hand, the active sensing approach is similar with that taken by non-destructive evaluation
(NDE) methodologies, since this approach is concerned with directly assessing the state of structure by
trying to detect the presence and extent of structural damage as a result of signals initiated in the structure
by the sensing system (Giurgiutiu, 2007). In the active sensing system the structure is equipped with both
actuators such as piezoelectric materials, including PZT, which generate perturbations in the structure,
and then PZT or similar patches, which are used as dynamic strain gauge sensors.
In a conventional non-destructive inspection, as described in Chapter 6 (NDI) a portion of the sensing
system, such as sensor probes or interrogator, has to be brought and set in the structure and therefore can
be applied only to periodic inspection and not to real-time monitoring, which requires small, nonintrusive, low power, and robust systems.
840
7.2
Passive systems
Passive sensing systems are mainly concerned with identifying the operational and environmental loads
acting on a structure and the structural responses caused by these loads based on measurement data
(Adams, 2007), inferring the state of structural health.
7.2.1
Strain
Strain measurement, which often holds a prominent position for structural health monitoring systems,
determines the stress and loading regime in part of a structure (Atkins, 2009) using point sensors, longgauge sensors, and distributed sensors (Glisic and Inaudi, 2008).
Conventional bondable or weldable foil resistance strain gauges are most famous. Short base strain
gauges (SBSGs) have a shorter gauge length and are generally more suitable for naval ships, which have
much more complex structures and have less clear deck areas than tankers and bulk carriers (Phelps and
Morris, 2013). Among fiber-optic sensors fiber Bragg gratings (FBGs) (Rao, 1997) and extrinsic FabryPerot interferometers (EFPIs) (Rao, 2006) are used as SBSGs in a manner similar to the foil strain gauges.
For dynamic strain measurements PZT and PVDF bonded on surfaces are employed (Sirohi and Chopra,
2000). Fiber-optic and piezoelectric strain sensors can form strain gauge rosettes to measure individual
components of strain tensors (Nichols et al., 2008, Zhao et al., 2011). Fiber-optic strain sensors survive in
harsh environment and provide high sensitivity measurements (Rao, 2006, Martinez et al., 2011).
Long-base strain gauges (LBSGs), such as linear variable differential transformers (LVDTs), which
measure the relative displacement between two points based on the inductive principle are designed for
uniaxial strain measurement and are relatively insensitive to local stress concentrations. LBSGs are generally suitable for weather deck applications on large ships, such as tankers and bulk carriers where space
in more readily available (Phelps and Morris, 2013). LBSGs based on different types of fiber-optic sensors are also available (Li and Wu, 2007, Rodrigues et al., 2013).
Fiber-optic sensors offer new and unique sensing topologies, including in-line multiplexing (quasidistributed) and fully distributed sensing, offering novel monitoring opportunities (Murayama et al.,
2012). Brillouin based sensing systems can measure strain at an arbitrary position along an optical fiber
(Bao and Chen, 2012, Galindez-Jamioy and Lpez-Higuera, 2012), and the spatial resolution, which used
to be more than one meter but has been recently enhanced to cm-order or sub-cm-order (Song et al., 2006,
Zhang and Wu, 2012). Strain sensing systems using optical frequency domain reflectometry (OFDR)
have achieved the sub-mm-order resolution, so strain distributions with steep stress gradients can be
measured accurately (Duncan et al., 2007, Murayama et al., 2013).
Full scale measurements using point or long-gauge strain sensors provide real time quantitative data
about stresses, fatigue and extreme loading in commercial ships (Barhoumi and Storhaug, 2014, Li et al.,
2014), naval vessels (Wang et al., 2001, Okasha et al., 2011), sailing boats (Micron, 2013), and offshore
and marine structures (Bang et al., 2011, Inaudi, 2011). Although there are a number of examples of full
scale measurements by in-line multiplexing and fully distributed sensing for civil infrastructures (Wan et
al., 2013, Ye et al., 2014), those for ship structures are almost unexplored and expected fields (Murayama
et al., 2003, Ivce et al., 2009).
7.2.2
Acoustic Emission
Acoustic emission (AE) monitoring is the process of monitoring a structure for the release of transient
elastic waves that can result from yielding, fracture, debonding, or corrosion. A system with AE sensors
can effectively monitor for crack initiation and growth and corrosion in metal structures as well as delamination and debonding in composite ones. AE frequencies used to detect fatigue cracks in marine steel
structures are usually between 50 and 300 kHz. Due to attenuation of the AE signal, this method is generally suitable for local monitoring over several meters in the structure (Atkins, 2009).
Most AE sensors are of the resonance type. Finding the right frequency range for a specific application
has to consider factors such as material, specimen size, and background noise. AE sensors with lower
resonant frequency can extend the sensor spacing (Vallen, 2012).
Generally, PZT is used as the detection element in AE sensors. Although the resonance type of piezoelectric transducers have high sensitivity to signals within the bandwidth of interest, they have to be located in close proximity of the interrogation equipment if a preamplifier is not used. They also tend not to
be used in harsh environments, in the presence of corrosive agents, under exposure to high temperatures,
or in hazardous and explosive areas. In contrast, fiber-optic sensors dont suffer from the drawbacks that
limit the use of conventional electrical-based sensors. The most common configurations in fiber-optic
acoustic sensors are the Mach-Zehnder interferometers. Michelson interferometers, EFPI, and FBG are
841
also used for AE detection (Burns, 2012). An ultra-wide frequency interrogation system allows a distributed array of in-line FBGs to act as strain, vibration and AE sensors (Mendoza et al., 2013).
AE has been commonly used in onshore storage tanks, pipelines, and offshore structures to detect
cracks and corrosion but the widespread adaptation and application in the shipping industry still has not
been seen, although several on board trials have been conducted in recent years. A recent pilot research
program initiated by the Alaska Tanker Company, ABS and MISTRAS group ventures into evaluating
commercial application of AE technologies in the marine industry. The sea trial was conducted to detect
crack initiation on a double hull TAPS trade tanker (ABS, 2007b, MISTRAS, 2009). Tscheliesnig (2006)
notes a European Commission (EC) funded research project was conducted by a consortium of R&D
organizations to assess the feasibility of adapting AE to detect corrosion on ship hulls at the start of 2002.
The main focus was to establish the feasibility of both a permanent on-board AE monitoring system and a
discontinuous spot testing system. Tscheliesnig noted that environmental noise could be a vulnerability
with a permanent on-board AE monitoring system. Baran et al. (2012) describe a recent development in
AE technology for fatigue estimation and corrosion assessment of ships. This initiative was focused on
preventative maintenance through reliable, cost effective NDT methods. On-board trials were conducted
on the ICARUS III oil tanker and a database of structural response measurements was established. The
ability to localize significant corrosion and extreme fatigue areas utilizing AE was also discussed.
7.2.3
Vibrations
Detection of defects in structure can also be made using the vibration of the structure when excited by
natural sources of ship operation. Sabra and Houston (2011) describe a system using Diffuse Vibration
Interferometry (DVI), in which the cross-correlation of vibration from separated sensors in a network is
analyzed to form a coherent waveform of local structural response. The method was used to analyze strain
gauge data from an instrumented high-speed aluminum vessel, but there were no known defects, so the
trial was inconclusive, although DVI was successful in laboratory tests of small structures with known
defects.
7.2.4
Crack
Crack propagation on a component is remotely determined by repeated or continuous measurements with

gauges consisting of conductor tracks connected in parallel, which will tear if the crack extends under the
crack propagation gauge, or electrically separated resistances, in which individual circuits are interrupted
as the crack extends. Changes in resistances of such gauges can be measured using a conventional resistance meter or a strain gauge amplifier. A wireless crack propagation monitoring system can deliver timely data without the need for constant visual inspections and equipment downtime (Gahbauer, 2012).
Some indirect techniques can serve to monitor various parameters such as hydrogen, corrosion potential, pH, conductivity, dissolved oxygen on-line in real-time.
7.2.5
Corrosion
The marine environment is generally the most corrosive of naturally occurring environments. The hull
being constantly exposed to the corrosive seawater environment experiences general corrosion that reduces the plate thickness uniformly but it is also likely to experience pitting, galvanic corrosion, and other
types. Corrosion can interfere with the operation of ship and impose increased loading stress, accelerate
deterioration of structure, and increase the hydrodynamic drag (Babu et al., 2014). Corrosion monitoring
techniques may be broadly classified into direct corrosion measurement techniques, which measure parameters directly affected by the corrosion processes, and indirect corrosion measurement techniques,
which provide data on parameters that either affect, or are affected by the corrosivity of the environment
or by the products of the corrosion process (Roberge, 2007).
7.2.6
Acceleration
Accelerometers are used for motion monitoring in hull monitoring systems and vibration monitoring in
condition monitoring of rotating machinery. Servo accelerometers are designed with low frequency sensitivity (DC-300 Hz), whereas piezoelectric sensors have higher frequency sensitivities (several Hz to several dozen kHz). Most MEMS accelerometers measure static as well as transient accelerations (DC to
1kHz). A number of FBG-based accelerometers are recently commercially available and they show different characteristics in terms of sensitivity, frequency range and maximum allowable acceleration
(Baldwin et al., 2005).
842
7.2.7
Metocean information
The environment is an important element to measure and quantify to understand the loading on and response of the ship or offshore structure. Environmental measurement is typically undertaken via in situ
wave-measurement buoys, air-borne or satellite devices, or mounted sensors on the ship or offshore structure itself. Wind speed is handled by use of an anemometer, with a growing range of technologies being
employed beyond the original form of small cups driven by wind to rotate around a central axis. X-band
wave radars (such as WaMoS by OceanWaves or Wavex by MIROS) have proven increasingly popular to map the water surface and determine wave heights, direction and periods, typically in the frequency
domain.
Local wave behavior close to the hull may be measured with other devices such as the TSK Shipborn
Wave Height Meter which measures wave elevation beneath the sensor. Further exploration and description of environmental sensors and their capabilities is the domain of the ISSC Committee I.1, Environment, and the reader is directed there to read past reports as well as the newest report.
7.3
Active systems
Excitation sources such as strain, acceleration, and AE, caused by the ambient load and environment
as an excitation are often not stationary. This means that robustness and long-term stability in measured data and normalization procedures are required to determine if the change in the structural response is the result of damage or deterioration as opposed to changing operational and environmental
conditions.
As an alternative, the active sensing system can be designed to provide a local excitation tailored to the
damage detection process. The use of a known and repeatable input makes it easier to process the response signals for damage detection in a designated area. For instance, by exciting the structure in an
ultrasonic frequency range, the sensing system can focus on monitoring changes of structural properties
with minimum interference from operational and environmental variability, which tend to be lowfrequency in nature.
Examples of the active sensing system for damage detection include impedance-based methods (Park et
al., 2003) and Lamb wave-propagation methods (Raghavan and Cesnik, 2007). Active sensing systems
are concerned with detecting the presence and extent of structural damage, while passive systems are
concerned with measuring various structural responses as well as loads and environments.
7.3.1
Impedance-based methods
The electromechanical impedance (EMI) technique uses swept-sine, high-frequency excitations, typically
higher than 30 KHz, applied through surface-bonded piezoelectric patches to identify changes in the
structures mechanical impedance, which can be considered an indication of a change in the mechanical
impedance of the host structure due to damage such as the formation of a crack or the loosening of a
bolted connection (Farrar and Worden, 2012).
In order to ensure high sensitivity to damage, the electrical impedance is measured at high frequencies
in the range of 30 400 kHz. A frequency range higher than 200 kHz is found to be favorable in localizing the sensing, while a frequency range lower than 70 kHz covers a larger sensing area. It has been estimated that the sensing area of a single patch can vary anywhere from 0.4 m (sensing radius) on composite
structures to 2 m on simple metal beams (Park et al., 2003).
The impedance method usually utilizes PZT sensors/actuators as well as macro-fiber composite
(MFC) to acquire dynamic responses of a structure. Typical SHM applications based on EMI generally use commercial impedance analyzers or alternative systems based on frequency response function
(FRF) or other methods. Such analyzers are often bulky, expensive and not suited for permanent
placement on a structure. As a promising alternative, a new microcontrolled SHM system based on
the EMI principle has been proposed to be small and low-cost (Park et al., 2007, Cortez et al., 2013,
Grisso, 2013).
Using the impedance method, damage and deterioration, such as cracking, bolt loosening, composite
delamination, and corrosion have been successfully detected on a variety of structures from simple beams
and plates to bridge truss structures and composite patch repairs. Developed prototypes are still only in
the laboratory verification stage (Park et al., 2007). Grisso (2013) provides details of development of a
structural health monitoring prototype for aluminum ship structures using an impedance-based method in
a highly complex ship sub-structure in the laboratory environment. Encouraged by the results of the laboratory tests, the initial investigations into miniaturized impedance hardware that could be used in actual
naval settings show great promise.
843
In order to handle real-life applications, extensive efforts are being devoted to study the issues that include developing miniaturized and portable impedance measurement equipment, packaging of the sensors
to facilitate installation, integrating this method with other NDE techniques, such as wave propagation
methods, acoustic emissions and ultrasonics, and investigating the possibility of wireless communication
between the sensor and the signal processing equipment (Park and Inman, 2007).
7.3.2
Lamb wave-propagation methods
The term guided-wave (GW) refers to elastic waves that propagate along a path defined by the structures
boundaries (Worden, 2001). GWs can travel at large distances in structures with only little energy loss, so
they enable the SHM of large areas from a single location. Among GWs, Lamb waves are usually analyzed in the SHM field. Lamb waves are guided between two parallel free surfaces, such as the upper and
lower surfaces of a plate. The symmetric Lamb waves (S0, S1, S2, ) resemble the axial waves, whereas
the antisymmetric Lamb waves (A0, A1, A2, ) resemble the flexural waves. Lamb waves are highly dispersive, and their speed depends on the product between frequency and the plate thickness (Giurgiutiu,
2007).
To be able to distinguish between damage and structural features, prior information forming a baseline signal is required about the intact structure. To allow the receiving sensors to record the response
with a minimal interference, it is essential to choose a driving frequency (Raghavan and Cesnik, 2007).
With the appropriate drive signal chosen, the approaches to SHM involve some form of either a pitchcatch actuation-sensing scheme or a pulse-echo scheme. In the pitch-catch scheme a Lamb wave is
launched from an actuator and received by a sensor at another location. From various characteristics of
the received signal, such as delay in time of transit, amplitude, frequency content, etc., information
about the damage can be obtained. With the pulse-echo approach the signal is generated at one location
and the reflections of these waves off the free surfaces such as crack boundaries are measured at this
same location. Defects can be located using the wave speed. In either approach, damage-sensitive features are extracted from the signal using some signal-processing algorithm, and a pattern recognition
technique is required to classify the damage and estimate its severity. The choice of the threshold to
decide whether damage is present in the structure or not is usually application-dependent (Giurgiutiu,
2007, Raghavan and Cesnik, 2007).
GW testing can offer an effective method to estimate the location, severity, and type of damage and it
is a well-established practice in the NDE/NDT industry. The most commonly used transducers are angle
piezoelectric wedge transducers, comb transducers and electromagnetic acoustic transducers (EMATs).
Other options, such as Hertzian contact transducers and lasers, have been explored in recent years. However, while these types of transducers function well for maintenance checks when the structure is offline
for service, they are not compact enough to be permanently onboard the structure during its operation.
PZT is one of the most popular choices in the field of GW SHM (Raghavan and Cesnik, 2007). In order
to overcome the disadvantage of PZT in terms of brittleness and flexibility, different types of piezocomposite transducers, such as the active fiber composite (AFC) (Bent and Hagood, 1997) and the macro
fiber composite (MFC) (Williams et al., 2002), have been investigated. Fiber-optic sensors, which have
flexibility, immunity to electromagnetic interference, and corrosive resistance can be also used as sensor
elements (Li et al., 2010).
Variations of Lamb wave propagation reflect some changes in effective thickness and material properties caused by such structural flaws as fatigue cracks, corrosion and disbonding. Yeo and Fromme (2006)
and Fromme (2008) investigated application of guided ultrasonic wave inspection for detecting corrosion
in ship hull structures. A stiffened aluminum plate fatigue specimen which was constructed with typical
naval ship details was instrumented with PZT to allow for GW SHM implementation during fatigue testing and structural changes due to fatigue were identified using Lamb waves prior to any visually observed
cracking (Grisso et al., 2011). Corrosion damage in a thin aluminum plate was detected and monitored by
the pitch-catch approach and a tomographic algorithm using damage index information produced by a
signal processing method (Grisso et al., 2011). Delamination or disbonding in composites are also detected by the pitch-catch approach (Balasubramaniam, 2014, Ricci et al., 2014). Packaging of the transducers, as well as ensuring reliable mechanical and electrical connections for them, is an important element in in situ SHM applications. The packaging design should account for the demands of harsh environments, load conditions and cycling fatigue experienced by the structure. Some systems including sensors, hardware, and damage detection algorithms have been developed (Alcaide et al., 2012, Bergman et
al., 2014). Several projects have been carried out to validate the system for aircraft structures in flight
tests (Dragan et al., 2013, Eckstein et al., 2013), but more full field testing of the system need to be addressed.
844
7.4
In order to identify data features that allow one to correctly distinguish between the undamaged and damaged structure, the acquisition of necessary and sufficient data is needed as well as the normalization and
cleansing process (Farrar and Worden, 2007).
The data acquisition portion of the SHM process involves selecting the excitation methods, the sensor
types, number and locations, and the data acquisition/storage/transmittal hardware. Economic considerations and the interval at which the data should be collected will play a major role in making these decisions. The data normalization process is separating changes in sensor reading caused by damage from
those caused by varying operational and environmental conditions. One of the most common procedures
is to normalize the measured responses by the measured inputs. Since strain sensors show cross-sensitive
to temperature, temperature compensation is required as the normalization to obtain reliable strain or
stress information. The data cleansing process is choosing data selectively to pass on to or reject from the
feature selection process. If a sensor is loosely mounted and that fact is found, the data from this particular sensor may be selectively deleted. Filtering and resampling can also be thought of as data cleansing
procedures.
7.5
A sensor network integrates sensors, data processors, network controllers, and data links. The capability
of a given sensor network can far exceed the sum of the capability of individual components (Balageas
et al., 2006). For example, in the case of bridge or a ship hull regarded as a beam, point strain sensors
provide just local strains and stresses individually, whereas the sensor network dealing with perfectlysynchronized strain data from point or distributed strain sensors along the beam structure can provide not
only local strains from each components but also vertical displacements or deformation modes through
double integration of curvature (Rodrigues et al., 2013).
In wired sensor networks data acquisition, storage, and analysis all occur at the one central location.
Such systems have the advantage that recordings from multiple channels are more easily timesynchronized, which is important when damage sensitive features are based on relative information between sensors. In a fiber-optic sensor network one can reduce costs for the measurement system by using
sensor multiplexing technique (Perez-Herrera and Lopez-Amo, 2013). The Norwegian Defense Research
Establishment (FFI) developed an extended ship hull monitoring system (ESHHM) including a fiber-optic
sensor network which had 36 strain sensors and 6 temperature sensors based on FBG. They installed it on
a mine countermeasure vessel and measured the extreme sagging/hogging moment induced by slamming
(Torkildsen et al., 2005). A significant limitation of such systems, however, is the deployment, since over
75% of the installation time is attributable to the laying of wires in a structure with installation costs
representing up to 25% of the system total cost for larger-scale structures, such as long-span bridges
(Lynch et al., 2004). Furthermore, wires installed in a combat vessel are vulnerable to detriments such as
heat, moisture, and toxic chemicals common in harsh military operational environments (Swartz et al.,
2012).
The integration of wireless communication technologies into SHM methods has been widely investigated in order to overcome the limitation of wired sensing networks (Farrar et al., 2006). The efficacy of
the wireless hull monitoring system in a realistic marine environment was demonstrated in a high-speed
littoral combat vessel (FSF-1 Sea Fighter). A multi-tiered, hybrid wired/wireless network with 3 wireless
sub-networks was installed and tested for strain sensing as well as modal analysis using accelerometers
(Swartz et al., 2012). A four AE channels platform with a self-powered wireless node has been developed
(Godnez-Azcuaga et al., 2011). Wireless sensor networks using active sensing approaches have been
studied for simple structures, such as an aluminum beam or sandwich plate (Drager et al., 2013, Cortez
et al., 2014). In a European project, a complete embedded sensor system (FBG sensor and interrogator)
for composite structures has been developed so that any fragile external connection between the sensor
fiber and the interrogator is avoided. The measured data is transmitted to an external read-out unit through
the wireless channel, and the external unit, in addition, provides the wireless power supply to the embedded interrogator (SmartFiber, 2014).
7.6
Maturity of Structural Hull Monitoring Systems
The level of maturity of different SHM systems differs significantly for each type and area of application.
Table 4 presents maturity of the SHM systems against the area of application.
845
Table 4. The Level of Maturity of SHM Systems Against the Area of Application.
Type of SHM or sensors
Application
Area
Goals
Level of Maturity
Sensor technology is mature; however it is

Slam avoid- Detecting a slam before it not easy to detect a slam before it occurs.
Hence, slam detection algorithms are not
occurs
ance
mature yet.
Quantifying whipping and
Accelerometers, motion
Whipping and springing responses, their Sensors and technology are mature. Howevsensor units to provide
contribution to fatigue and er, analysis may require effort by specialists.
rigid body ship motions, Springing
design loads.
strain gauges
Determining short and
Sensors and technology are mature. ComMotion Based long term loads and
sensor units to provide
mercial off the shelf tools are available.
Monitoring avoidance of excessive
rigid body ship motions
motions.
Voyage optimisation for
Sensors and technology are mature. ComWeather
fuel efficiency and imsensor units to provide
Routing
proved ship safety
rigid body ship motions
Determination of bending
Motion sensor units to
moments and hydrodyprovide rigid body ship
Sensors and technology are mature. ComLoads moninamic loads for short term
motions, load cells, strain
toring
(voyage) and long term
gauges, wave height sen(design) analysis
sors
Motion sensor units to
Determination of speed,
provide rigid body ship
Fatigue moni- seaway and stress ranges Sensors and technology are mature. Howevmotions, strain gauges,
over a long term monitor- er, analysis may require effort by specialists.
toring
wave height sensors,
ing
weather hindcast data
Technology has been demonstrated in a
Detecting crack before
Crack deteclaboratory or a small scale experiment.
visually seen or before
Acoustic Emission
tion
Widespread application is not available yet.
they occur.
Technology has been demonstrated in a
Detecting cracks before
laboratory or on a small scale experiment.
Impedance based detection Crack detecvisually seen or before
Application of it onboard a ship is yet to be
tion
methods
they occur
successfully demonstrated.
Detecting corrosion under
Small scale experiments have been shown.
Corrosion
insulation or in areas difWidespread application is not available yet.
detection
ficult to inspect
Detecting failure in ma- It is still in research phase in laboratory
Failure
Fibre optic sensors
rine composite structures environments.
detection
sensor units
8.
METHODOLOGIES FOR USING INSPECTION & SENSED DATA
8.1
Introduction
Effective inspection, Structural Health Monitoring (SHM) and operational guidance techniques allow
the detecting, locating, and quantifying of potential damage situations in order to ensure the proper
operational performance of the existing ship and offshore structures. It also ensures the ability to minimize maintenance and lifecycle cost by taking correctional measures in a timely manner. The current
SHM systems have contributed to a significant improvement in the operational safety of ships, and will
continue to play a critical role in ensuring structural and operational performance of commercial ships
and naval vessels, including lightweight high performance ships. In the case of high performance ship
design, construction includes novel lightweight structure which warrants the utilization of composite
materials, aluminum alloys, or high strength steel with innovative application techniques to maximize
the vessel performance (Sabra and Huston, 2011). At the same time, high performance vessels are subjected to significant wave induced impacts and slamming while operating in heavy/extreme environments which contribute to accumulation of fatigue damage. Thus, the development and application of
new and innovative SHM systems are necessary for assessing the structural reliability of such high
performance vessels.
846
8.2
Operational Advice
The implementation of SHM on vessels may lead to potentially effective and efficient ways to:
Provide a real-time feedback system equipped with real time loading information and trends, and
inform of the possibility of extreme loading events so that the risk of structural overload/ failure may
be avoided, and
Obtain actual in-service information of the extreme and long term fatigue loads to which the vessels
are exposed, so that this information can be utilized to improve both the design and lifecycle management of ship and offshore structures (Phelps and Morris, 2013).
Operational decision-making also plays a key role in ensuring the safety of the vessel and the crew at
sea. Most operational decisions require the assessment of both the ship conditions and environmental
conditions by the ship operator. Operational risks governed by the seaway conditions are made at all stages of decision-making. Pre-operation decisions include considerations taken before the voyage. In term of
passenger ships, the question might be whether to sail or to stay at quay. Preventative decision making
becomes more important during the operation in order to avoid situations that might be uncomfortable or
hazardous. It includes weather routing, and could also be a question of extra lashing or ballasting before
entering heavy seas.
Within the last decade the on-board and post-voyage analysis of ship structures have become a common theme for the maritime industry. The analysis and recommendations developed through the implementation has improved both the design and operational aspects within the shipping industry. The combinational diagnostic and prognostic approach through SHM can significantly improve the vessel operation
capabilities and lead to a significantly more optimized design methodology with a quantifiable increase in
structural reliability.
8.2.1
Identifying loading to stay within safe operating envelope
8.2.1.1 Motion Based Monitoring

The roll, pitch, and heave response of a ship to a sea state are generally of the most concern from either a
synchronous motion aspect or extreme motion aspect and whose limitations will bring operational benefits. Ship motion monitoring systems are primarily installed to warn the operating personnel that the vessel motions are approaching intensities where selected motion threshold limits are likely occurrence to be
exceeded. Ship motions themselves are often not a problem for the vessels integrity, but they are often
the reason that a Master reacts to deteriorating environmental conditions. However, the limiting motions
will sometimes be established on the basis of computed structural limits.
Topics to be discussed next such as slam avoidance, whipping and springing, weather routing are all
related to motion based monitoring.
8.2.1.2 Slam Avoidance

Slam avoidance systems can predict the possibility of a vessel experiencing wave slams that may lead to
either localized or hull girder structural damage. Structural damage warning levels on displays and alarm
levels are set taking into account the approved scantlings and their conditions of approval. The criteria to
judge the occurrence of slams is derived from calculations, model tests or full-scale trial results, which
should be updated based on the actual experience. Methods of identifying slam impacts include:
Accelerometers and strain gauges: The recognition of a decaying vibratory shape on the acceleration or
strain (converted to stress) signals (Magoga et al., 2014) at the frequency of the 2-node mode of vibration
of the vessel which can be determined using spectral analysis or other techniques. The severity of the
impact is indicated by the amplitude of the vibration. For the applications that have utilized multiple accelerometers to identify the presence of slamming through the detection of vibratory bow acceleration, the
placement of accelerometers as close as possible to the extreme ends of the ship is noted to be particularly
important for this system to maximize the two-node vibration detection (Moe et al., 2005).
Pressure Transducers: Pressure sensors at the bow of the vessel initiate a warning when a zero pressure
state is recorded as a result of bow emergence. The severity of impact can be indicated by the re-entry
pressure (Drummen et al., 2014). Also the strain gauge data from the bow panels may be used to calculate
the mean pressures across the bow panels due to slamming.
Wave-induced hull girder vibrations are normally described by the terms springing and whipping.
The resonant vibrations induced are termed springing, whilst transient vibrations which increase rapidly
due to wave loads is termed whipping. The normal cause of whipping is attributed to the impact of loads
847
generated from bottom slamming, bow flare slamming, or stern slamming. Full scale measurements on
board a number of vessel types indicate the effect of vibrations on the structural fatigue damage is comparable to the conventional wave loading effects. The whipping contributes to extreme loading scenarios
which may also exceed the classification society rule wave bending moment (Moe et al., 2005, Storhaug,
2012).
Barhoumi and Storhaug (2014) extensively studied the whipping and springing responses on a large
8,600TEU container vessel equipped with a hull monitoring system, based on fiber-optic sensors. The
study indicated that the higher fatigue rates are limited in occurrence but contribute significantly to fatigue damage.
Sheinberg et al. (2011) investigated the effects of wave impact and whipping response on fatigue life
and ultimate strength of a semi-displacement patrol boat based on model tests conducted on a segmented
structural model. Stambaugh et al. (2014) strongly suggested onboard structural hull monitoring systems
to provide life fatigue damage accumulation and guidance to the operator for heavy weather avoidance
since fatigue damage is proportional to the third power of stress range.
Developments in detection of slamming events and understanding their contribution to localized
stresses and fatigue life of the hull structure have significantly improved ship design and maintenance
through post voyage analysis. New methods are being proposed to implement onboard monitoring and
analysis of slam induced fatigue. At the same time, extensive data collection systems are installed for
monitoring of engine and hull performance and for voyage performance evaluation, etc. Such systems are
expanded to encompass procedures for stress monitoring and for decision support. Nielsen et al. (2011)
outlines a conceptual calculation procedure for fatigue damage rate prediction in hull girders taking into
account whipping stresses, proposing a method to automatically estimate hypothetical changes in ship
course and speed in a wave environment to assess the accumulated fatigue damage in the hull girder.
8.2.1.3 Weather Routing

Weather related incidents have been a significant cause of maritime casualties/incidents all throughout the
past. Therefore it is critical that effective weather routing technologies and systems are developed, in
order to mitigate the effect of weather routing induced structural stresses. The main objective of weather
routing systems for high-speed craft is to reduce the risk of crew injuries and to avoid significant hull
damage. Weather routing systems provide decision support for the navigator regarding optimum speed
and course based on limit values for relevant ship response (Rathje, 2009). Table 5 highlights several
commercially available weather routing systems and services.
Table 5. Commercially Available Weather Routing Systems and Services (list is not exhaustive).
System
Navi Sailor 4000
ECDIS: Premium+
Vendor
TRANSAS
Headquarters
St Petersburg,
Russia
Bon Voyage System
AWT Worldwide
Headquarters
California, USA
Jeppesen Vessel
Voyage
Optimisation
Solution
A Boeing
Company
Headquarters
System Details
AIS, Dual ARPA and Target Management
Provide information on the tides and currents
Compliance approved by
1. US Coast Guard Certificate of Approval
2. Russian Maritime Register of Shipping
3. DNV EC
4. China Classification Society Certification
Severe motion alerts
Graphical depiction of Weather, Route and Currents
High resolution data availability
Claims to be able to provide 16 day forecast and four time a
day
Automatic generation of a full range of optimized routes
Optimize to minimum fuel speed plan for required arrival time
VVOS optimal speed management
Simulation tools facilitate analysis of any route using highresolution forecast weather to weigh trade-offs among ETA,
fuel consumption, ship motions, hull stresses, and weather and
sea conditions.
One of the leading voyage optimization tools offered by Jeppesen (2014) provides commercial benefits
by enabling every vessel in the fleet to attain higher efficiency, as well as improved operating margins
and operating risk mitigation. The Vessel Voyage Optimization Solution (VVOS) utilizes advanced
routing algorithms to accurately optimize each route for on-time arrival while minimizing fuel consumption. The guidance system recommends speed and heading changes to manage ship motions and help
848
minimize heavy weather damage. Krata and Szlapczynska (2012) explored the possibility of adapting prevoyage Multi-criteria Evolutionary Weather Routing Algorithm (MEWRA). The authors calculated a
safety index for dangerous dynamic phenomena (such as reduction of intact ship stability when riding a
wave crest amidships or parametric roll phenomena) and suggested a similar safety index could potentially be applied regarding the structural condition of the vessel. Szlapczynska (2013) provided possible
practical applications of MEWRA. Mao et al (2012) presented a ship fatigue routing procedure, using a
simple spectral method developed by the authors, and the analysis of long-term fatigue assessment of a
typical container vessel. The potential benefits of using the proposed ship fatigue routing procedure was
demonstrated by a case study of a 2,800TEU container ship and the authors stipulated that the fatigue life
of the vessel can be extended by at least 50% by choosing more well-suited ship routings.
8.2.2
Quantifying operational loading and changes
Researchers in China have made significant progress in various aspects of SHM for quantifying and
tracking operational loads and responses. Ou and Li (2010) reported that almost 100 bridges, offshore
platforms, tall buildings, spatial structures, underground infrastructures, and pavement have been implemented with SHM systems in China in an attempt to quantify loadings and the associated responses.
The VALID Joint Industry Project (JIP) resulted in SHM systems being deployed in new USCG ships
and vessels. The project incorporated numerical analysis of the USCG National Security Cutter, segmented model tests and full scale sea trials to find insight into actual fatigue life vs. as-designed
(Stambaugh et al., 2014).
Murawski et al. (2012) provide examples of marine safety improvements by the application of SHM
systems. Li et al (2006) describe a methodology for the health monitoring of composite marine joint
structures based on strain measurements under operational loading using embedded fiber Bragg grating
sensors, which enables the sensor measurements to be used for damage detection without reference to a
high-fidelity numerical model of the structure. The technique is shown to provide successful damage diagnoses with an acceptable level of accuracy.
Effective damage detection utilizing smart sensor technologies offer new opportunities and possibilities
for structural health prognostics, with the results that include actual damage used in the prognostic models
for accurate, real-time prediction of future trends. A prognostic model can be continually updated over
time as the damage develops (Herszberg et al., 2007).
The (remaining) fatigue life estimation has been the topic of many research papers recently. New onboard hull monitoring systems are expanded to encompass procedures for stress monitoring and for decision support, where critical wave-induced ship responses and fatigue damage accumulation can be accurately estimated for hypothetical changes in ship course and speed in the automatically estimated wave
environment. Nielsen et al. (2011) outline a conceptual calculation procedure for fatigue damage rate
prediction in hull girders taking into account whipping stresses and a fast and reliable stress calculation
procedure for the derivation of spectral properties at specific structural details for given environmental
data, as the final step in developing a decision support system for accumulated fatigue damage. Considering only the contributions from vertical bending moments, Jensen et al. (2008) have described a simplified calculation procedure for the long term probability distribution of the combined wave and whipping
induced stresses and Neilson et al. (2011) highlights that reasonable agreement with full-scale measurements were found, provided the pertinent parameters related to the estimation of the impulsive slamming
load are chosen with care. Sheinberg et al. (2011) and Stambaugh et al. (2014) investigated the fatigue
life of US Coast Guard patrol boats; Storhaug (2012) studied the effect of extreme responses to fatigue
life of a container vessel; Temple and Collette (2013) discussed fatigue of naval vessels; Moe et al.
(2005) conducted full scale measurements on a bulk carrier, and Kwon and Frangopol (2012a) investigated the fatigue life of aluminum ships.
Wagstaff et al. (2013) presented a case study where corrosion under insulation could be detected using
the Asset Integrity Management (AIM) process working with measured data. An example test result is
illustrated for sensing the crevice corrosion of nickel-aluminum bronze using a boron-doped diamond
sensor array.
8.3
Lifecycle Management Advice
In order to ensure the safety of ships and prevent catastrophic effects of maritime accidents, reliable efforts are required through inspection, maintenance and repair to ensure the safety of the ship and its crew.
As the current shipping fleets age, and due to the growing budgetary restrictions governments and operators are forced to extend the service life of marine vessels, including naval vessels, long past their acceptable operational lifecycles.
849
The European Unions Condition Assessment Scheme for Ship Hull Maintenance (Cabos et al., 2008)
defined a standard ship electronic model into which the measurements of on-board thickness measurements, cracks and coating condition would be recorded. A computerized analysis of the ship electronic
model (with measurements recorded inside) was developed to enable the triggering of automatic alarms if
the condition of the vessel is degrading dangerously.
The approaches available to the navies for determining the assessed structural condition of an aged
warship are similar to those used by commercial agencies for merchant ships. The US is very active in
driving holistic treatments to manage and ensure structural longevity. Hecht and An (2004) sponsored by
the US Naval Sea Systems Command showed how non-destructive evaluation methods can be justified in
their uptake by integrating structural component level strength probabilities with system level descriptions and economic evaluations. The US Navy and the American Bureau of Shipping (ABS) embarked on
a program which resulted in the Surface Ship Life Cycle Management (SSLCM) activity (Eccles et al.,
2010). The main tool behind the program is to build a finite element model for the as-built configuration
of the warship and apply projected operational loadings envisioned for the ship to identify the inherent
structural margins, inherent corrosion allowances, critical inspection plans, plate renewal criteria and
expected fatigue life for the ship.
The Naval Sea Systems Command and ABS are partnered in the Achieving Service Life Program
(ASLP) for the US Navy. ASLP provides a structured, third-party periodic condition assessment process
for selected systems and structures on a number of US surface combatants. ABS extended the ASLP to
the Service Life Evaluation Program (SLEP) so that it can be employed by other navies (Anderson et al.,
2013). SLEP has three principal components: survey reporting, strength analysis, and fatigue analysis. In
a SLEP program the assessment cycle is defined with Baseline Service Life Assessment, Continual Service Life Assessment, and two Intermediate Service Life Assessments in between the two. For strength
and fatigue analyses, and remaining fatigue life estimation, SLEP provides specific methods to apply with
risk based assessment criteria and recommended fatigue inspection intervals. For more in-depth discussion on the topic of lifecycle management of naval ships, the readers are encouraged to refer to ISSC
2015 Committee V.5 Naval Ships, report.
General maintenance schedules for the large majority of marine vessels are largely based on their
availability, usually calling for repairs on fixed intervals that are not chosen with the specifics of the
ships structure in mind. It is desirable to optimize the vessel lifecycle management/ maintenance systems
based on the specific design of physical structure of the ship to improve safety and lifecycle management
(Collette, 2011). An extensive study by Temple and Collette (2013) proposes a framework to determine
the maintenance cycles for a naval vessel to be specific to the vessel itself, and take into account the potential for an extension to the original service life in order to minimize the lifetime maintenance cost for
the ship in the face of operational uncertainty. Lee et al. (2006) presented the evolution of inspection and
maintenance strategies as shown in Figure 2.
Figure 2. Evolution of Inspection and Maintenance Strategies (Lee, 2006).
Time-based maintenance where maintenance activities are carried out periodically is the traditional
maintenance technique. It assumes that the failure behavior (characteristic) of the equipment is predictable based on hazards or failure rate trends, known as bathtub curves, as shown in Figure 3. As can be seen
from the figure, failure rate trends can be divided into three phases: burn-in, useful life, and wear-out. It is
expected that a high failure rate will be experienced at the start of the operational life of a complex structure. During the burn-in phase, the structure or equipment experiences decreasing failure rates early in its
850
lifecycle, followed by a near constant failure rate (useful life). At the end of their lifecycles (wear-out),
structure or equipment experience increasing failure rates (Ahmad and Kamaruddin, 2012).
Figure 3. Typical bathtub curve.
8.3.1
Condition Based Maintenance (CBM)
Condition based maintenance (CBM), also known as predictive maintenance, is the modern and popular
maintenance technique which, according to Jardine and Tsang (2013), recommends maintenance actions
(decisions) based on the information collected through condition monitoring (CM) process. CM, performed either periodically or continuously enables the collection of the condition data (information) of
the equipment or the structure or increases knowledge of the failure causes and effects and the deterioration patterns of equipment or structure. Applications to vibration monitoring, sound and acoustic monitoring, and lubricant monitoring using CBM are provided by Ahmad and Kamaruddin (2012).
CBM has become an integral part of many vessel monitoring sectors in the marine industry. Gudze and
Melchers (2007) discussed the necessity for the CBM requirements (usually performed on board) for
Australian Navy ships. Usually the major issues faced with condition based maintenance arise from the
poor structural design or operational tactics of the vessel.
8.3.2
Reliability Centered Maintenance
Reliability Centered Maintenance is used to develop scheduled maintenance plans that will provide an
acceptable level of operability, with an acceptable level of risk in an efficient and cost-effective manner.
Camara and Cyrino (2012) presented reliability applications in design and maintenance planning of ships
subjected to fatigue and corrosion. Based on a target reliability index, a comparative study was carried out
considering fatigue of a critical joint based on S-N and fracture mechanics approaches. The fracture mechanics-based model presented slightly shorter inspection periods (510%) than that of the model based
on a S-N curve. Camara and Cyrino (2012) reasoned this difference resulted from the crack initiation
period not being taken into account in the model of fracture mechanics.
8.3.3
Reliability Based Inspections
Reliability based inspection relies on gaining experience from the initial and subsequent inspections. In
this way, degradation patterns for probabilistic models can be constructed. These models are able to produce estimations and predictions about asset degradation and structural integrity at a specific time in the
future. At the same time, the models are linked with the classification of the risks of this process and the
consequences of failure to determine a more specific assessment and risk ranking. This practice is referred
to as Risk Based Inspection (RBI) (Tammer and Kaminski, 2013). Tammer and Kaminski (2013) argue
that the RBI methodology is a more comprehensive successor of Reliability Centered Maintenance methodology. The fundamental difference between the two methods is that reliability centered approach
strives towards minimal loss of functional integrity of (mostly) dynamic equipment, while RBI deploys a
directed inspection effort to the most critical components. It is important to recognize that seeking ways
to relax inspection practices is not the goal of establishing an RBI program (ABS, 2003b).
There has been a body of research found in the literature discussing the applications of RBIs to ship
and offshore structures (Lee et al., 2006, Bharadwaj and Wintle, 2011, Ku et al., 2012). Using the minimization of expected fatigue damage detection delay as the objective function, optimum inspection planning of ship hull structures subjected to fatigue was determined by Kim and Frangopol (2011). The study
851
highlighted that an increase in the number of inspections and/or inspection quality can lead to reduction
of the expected damage detection delay. Therefore, cost-effective inspection planning is important based
on the Pareto solution front of the optimization problem. Kwon and Frangopol (2012a) applied the
probabilistic approach to estimate fatigue reliability of aluminum ship structures and thus suggested lifetime optimum inspection/repair. System reliability of a very large crude carrier structure under corrosion
and fatigue was investigated by Kwon and Frangopol (2012b). Perisic and Tygesen (2014) investigated
two cost effective load monitoring methods for fatigue life estimation of offshore platforms.
8.4
Lifecycle experience is an important factor in the updating and improvement of design codes. Update
effectiveness is highly dependent upon an understanding of the operation and failure of earlier designs.
After a full scale experimental and numerical study, design pressures for fatigue of bilge keels of a
class of naval vessels were found incompatible with those provided by the classification rules (Flockhart
et al., 2010), which led to relevant class societies updating their rule set.
Another example is the United States Coast Guards (USCG) project to assess fatigue design approaches for its new National Security Cutters (NSCs), known as the Fatigue Life Assessment Project (FLAP)
(Stambaugh et al., 2014). To support the FLAP program, VALID Joint Industry Project (JIP) was formed
by MARIN with multiple partners. The aim of the VALID project is to further improve the confidence in
determining the wave loading leading to fatigue damage on a naval frigate type hull form and structure.
This was achieved through the use of numerical predictive model for fatigue predictions (Hageman et al.,
2014) and a model test program supported by dedicated full-scale trials (Drummen et al., 2014).
The key findings from the FLAP and VALID JIP projects relevant to the USCG that may have implications to design of future vessels are (Stambaugh et al., 2014):
8.5
Fatigue life predictions using the traditional approach under-predict wave-induced fatigue loading by
approximately 7%. Improvements are recommended to the traditional approach to fatigue damage
from impact loading and whipping response.
Uncertainties in the SFA process have been quantified on a sufficient level required for use in time
varying fatigue reliability based assessments and sustainment evaluations.
Conservative fatigue life estimates based on the standard S-N design curves are useful in fatigue design; however, forecasts of fatigue life must consider the large uncertainties in the operational environment, influence of the operator and the wide scatter of S-N data.
Discussion
Despite the above mentioned wide-spread application and use of Structural Health Monitoring data, it is
important to recognize that there is still a considerable amount of effort required in the storage and handling of data. The analysis of such data presents significant challenges in the sense it requires steps such
as filtering to remove noise, separation of high and low frequency responses, sampling techniques, statistical analysis. A typical SHM system will record in excess of hundred channels of data with sampling
rates in the kHz range (Kiddy et al., 2002, Grassman and Hildstrom, 2003). For example, a sampling rate
of 2 kHz is typically used if slamming associated data is required (Grassman and Hildstrom, 2003,
Drummen et al., 2014). Drummen et al. (2014) reported that SHM data collected onboard NSC USCGC
BERTHOLF is at a rate of 600 gigabytes per month and collected for a period of a full year.
LHostis et al. (2013) discussed the development of a fully automated advisory monitoring system for
the fatigue of FPSO hulls as part of MONITAS II JIP. MONITAS I and II JIPS have demonstrated the
need for detailed integration of the hull designer into the SHM design (Kaminski and Aalberts, 2010,
L'Hostis et al., 2010). Understanding the FPSO hull design assumptions and embedding them into the
data analysis tool is the only way to provide tools for fatigue management. It is argued that failing to do
so during the project design phase (newbuilding or conversion) will lead into gigabytes of measured data
that will probably remain untouched by and therefore useless to the operator (LHostis et al., 2013).
8.6
Conclusions
As was shown in Chapter 7, there are varying levels of maturity of different structural hull monitoring
systems, varying from the conceptual design stage to proof-of-concept laboratory tests to wide-spread
industrial applications. Use of the collected data is still limited but a wide range of efforts have been undertaken to develop and implement systems that take advantage of monitored data to guide operation and
lifecycle management of the structure.
852
9.
LIFE TIME EXTENSION, COMPARISON OUTSIDE & WITHIN THE

MARITIME INDUSTRY
9.1
Introduction
This chapter describes the issues not specifically considered in the preceding sections. It covers the areas
related to lifetime extension of existing structures, not meaning the repair actions. The second section
covers the lessons that can be learnt from the other industries and the last section is devoted to the differences that exist between the approaches for ships, offshore structures and other marine applications, ranging from navy to renewable energies.
9.2
Lifetime Extension of Existing Structures
Lifetime extension is required when vessels or offshore structures need to remain longer in service than
considered in their initial design. For this reason, HSE UK (Stacey et al., 2008, Stacey, 2011, Stacey and
Sharp, 2011) and Norways PSA (Ersdal et al., 2011) started projects on the lifetime extension need for
offshore units. Stacey describes the issues that are important, such as improved monitoring and maintenance and better (internal) guidelines on how to treat ageing units. Ersdal points out that maintenance may
be lacking, but especially the recording of it. When a mismatch exists between the expected conditions
and the actual condition, risk for accident escalation is increased. Operators are to be prepared better for
the potential risks, but in the meantime maintenance needs to be brought to a higher level, especially to
cover increased needs of older installations.
For offshore installations, Risk Based Inspections are used successfully by different companies, e.g.
Petronas (Nichols et al., 2008). The RBI methods are often used for jacket structures. Deep water access
for inspection is costly and has safety risks. Available measurements and inspection data are combined
with analyses to predict the condition for similar structural details. These methods are gaining attention
from the floating platforms as well (Landet et al., 2011). In this paper 4 FPSOs are analysed according the
RBI methods outlined in NORSOK N-006. As the idea behind may be rather straight forward, the actual
implementation on ship shaped structures is very complex.
For vessels, lifetime extension is more an evaluation of whether the vessel might be operated after the
next survey or not. Caridis (2014) compiled a comprehensive book on the various options, issues and
solutions available today.
The appropriate acceptability criteria of the structure and maintenance strategy need to be made clear
in order to facilitate life extension analysis. Various papers have been published to show the effect of
defects (either corrosion or cracks) on the structural integrity (Gao et al., 2012, Tran Nguyen et al., 2012,
Tran Nguyen et al., 2013). It is noted that the rules and criteria may have changed over the years of service, giving additional challenges. For jackets, it has been shown that an update of the rules may increase
the projected lifetime (Jia, 2008).
The objective of a re-assessment and life extension of an existing structure is to establish that the structure is fit for its intended purpose over the residual service life, and that the consequences in terms of risk
to human life, to the environment and to the assets, in the event of structural failure, are acceptable from
both a reliability and an economical point of view.
The most common reason for re-assessment is uncertainty about the safety of the structure and the assessment process is focused towards reassuring that the structure has adequate safety. The interpretation
of adequate may be discussed, since it can be understood as simply fulfilling given code requirement or
as determining the consequences and associated probabilities for different failure events and evaluate
these against acceptance criteria. Other reasons for re-assessment may be doubts about the durability, the
serviceability or the functionality of the structure.
The assessment process will typically include or be based on:
documentation of as-is condition,
planned changes and modifications of the facility,
calibration of analysis models to measurements of behavior if such measurements have been performed,
history of degradation and incidents,
prediction of future degradation and potential incidents based on history,
the effect of degradation on future performance of the structure,
documentation of technical and operational integrity,
planned mitigations,
plan or strategy for maintenance.
853
The assessment for life extension should conclude with a safe life extension period with respect to
technical and operational integrity of the facility. The assessment should also identify the circumstances
that will limit the life of the facility without major repairs or modifications, and specify criteria defining
safe operation (e.g. permissible cracks lengths, permissible corrosion or remaining thickness, remaining
anodes, degrading of paint protection, subsidence), including appropriate safety factors.
An example of the assessment process is illustrated in the flow sheet shown in Figure 4. (NORSOK,
2009). The flow sheet may be followed for assessment of all groups of limit states (ULS, SLS, ALS and
FLS). Structural analyses for assessment may be performed provided that a sufficient data basis for performing reliable analysis is available. Then the safety level of the structure can be assessed and it can be
decided if mitigation is required. An analysis, in this case, is an engineering process that can imply assessment based on simple hand calculation or more refined structural analysis.
Figure 4. Flow sheet of the assessment process (NORSOK, 2009)
A plan for mitigation as indicated in the flow sheet in Figure 4involves:
A plan for the mitigation itself

Documentation
Plan for maintenance after mitigation
Examples of circumstances or initiators that may trigger a need reassessment are as follows:
Changes from the original design or previous assessment basis, including
o modification to the facilities such that the magnitude or disposition of the permanent, variable or
environmental actions on a structure are more onerous,
o more onerous environmental conditions and/or criteria,
o more onerous component or foundation resistance data and/or criteria,
854

physical changes to the structure's design basis, e.g. excessive scour or subsidence, or relocation
of mobile offshore units to a new location and,
o Inadequate deck height, such that waves associated with previous or new criteria will impact the
deck, and provided such action was not previously considered.
Damage or deterioration of a primary structural component: minor structural damage can be assessed
by appropriate local analysis without performing a full assessment. However, cumulative effects of
multiple damages should be documented and included in a full assessment, where appropriate.
Exceedance of design service life, if either:
o the remaining fatigue life (including design fatigue factors) is less than the required extended
service life, or
o Degradation of the structure due to corrosion beyond any design allowances has occurred, or is
likely to occur within the required extended service life.
o
Existing design documentation can be applied as basis for the assessment if inspection of the structure
shows that time-dependent degradation (i.e. fatigue and corrosion) has not become significant and that
there have been no changes to the design criteria (any changes to the original design basis are assessment
initiators, see a) above). This requires that in-service inspection has been performed to document a proper
safety level.
A structure which has been totally decommissioned (e.g. an unmanned facility with inactive flow lines
and all wells plugged and abandoned), or a structure in the process of being decommissioned (e.g. wells
being plugged and abandoned) generally does not need to be subjected to the assessment process unless
its failure has consequences for nearby facilities.
9.3
Other industries
Leadership in the technology of structural longevity has come from the aerospace community for decades.
For example, the US Air Force has developed, beginning in 1958, a methodology for using structural
health monitoring of aircraft through the Aircraft Structural Integrity Program, which is intended to
achieve structural integrity in USAF aircraft while managing cost and schedule risks through a series of
disciplined, time-phased tasks, (US DOD, 2005). The program uses a damage-tolerant, fail-safe philosophy that incorporates a surveillance program that is based on allowable damage limits and damage growth
rates for all observed defects in aircraft structure. This philosophy requires detailed finite element analysis
to reveal probable locations of fatigue crack initiation, analysis of detected cracks to assess anticipated
growth rates, and updating the analyses based on observed crack growth rates and changes in aircraft
operations. The program incorporates risk analysis to assist in developing inspection times, evaluate options such as repair versus modification, and to operate aircraft beyond the design service life. Extensive
research has been conducted into using instrumentation to monitor cracks. Also Paul Hoffman (2009)
describes naval aircraft where fatigue loading is closely monitored in conjunction with probabilistic assessments of strength and failure detection. As a naval aircraft is relatively small and relatively well monitored such an approach is feasible.
The structural longevity of civil engineering structures, particularly bridges has become increasingly
important. For example, over 20 percent of railway bridges across Europe are metallic, 70 percent of them
are over 50 years old, and 30 percent of them are over 100 years old. As a result of this, multiple communities have developed with regular meetings to share information. For example, the International Association for Life Cycle Civil Engineering (IALCCE) started in 2006 and meets bi-annually and is supported by the journal Structure and Infrastructure Engineering: Maintenance, Management, Life-cycle
Design and Performance (Editor-in-Chief: Dr Dan Frangopol, published by Taylor and Francis) in which
appropriate civil structural research is reported, but with some coverage of other platforms such as ships
and offshore structures.
Reinforcement of degraded civil structures with composites is being done to compensate for loss due to
corrosion, provide strengthening to reduce fatigue crack initiation, or to increase the load-bearing capacity of the structure. A consortium of fifteen organizations from eight European countries was formed in
January 2007 to explore composite patch repair for marine and civil engineering infrastructure applications (CO-PATCH, 2013).
Monitoring of bridge structures is being done to assess the load profile, determine stresses in the structures. Technology development is being pursued to detect cracks in bridge structures, particularly in areas
that are difficult to inspect. An example of this is provided by Inaudi (2009) who reported on the use of
fiber-optic sensors to monitor bridges.
9.4
855
Differences in approaches for ships, offshore structures, and other marine structures
(ranging from navy to renewable energies)
Ships may be active in international trade, but also operate in offshore support activities. These units typically have regular inspections, have recorded routes and face limited exposure as they can be in harbor,
sailing in benign waters and use weather routing. They also may be exposed to extreme conditions, with
loading closer to their design values, including polar conditions. The condition of the structure is heavily
dependent upon the care the owner gives to the unit. Defects are typically repaired in controlled conditions.
Offshore structures, considered to be stationary off the coast, may experience fiercer loads, especially
in the more harsh areas. Rough conditions may refer to extreme wind, high waves and low temperatures.
Maintenance and repair are harder to execute, due to the operational limitations. Challenges include the
ability to execute hot work as well as limitations on the amount of people you can take on board to do the
work. The activities associated with inspections start with preparation such as staging and cleaning. The
inspection itself may require different techniques to be applied (MPI, CVI, X-ray, eddy current), while
working at heights or deep under water. This means the variety of workers and effective working time is
limited by the safety precautions and inaccessibility. Together, this means that the drive to limit inspections and maximize the use of it is mainly driven from offshore.
Other marine structures are those that are none of the above categories such as naval vessels and the
renewable energy installations. The maintenance and loads are different from any commercial standard,
although there is a trend to class naval vessel under commercial class societies. Further, due to budgetary
constraints, there is an increasing need to extend the life of navy vessels, far beyond the lifetime as originally planned. Load measurements and defect analyses are popular ways to justify lifetime extension.
Renewable energy installations are very new and range from wind turbine generators installed offshore
to tidal energy converters and current driven turbines. For tidal energy converters, the current must be
high to be effective. In such case the maintenance and inspections are very difficult to carry out. This
topic should surely come back in the next ISSC.
These approaches result in areas where differences are found:
9.5
Inspection scheme: The rules prescribe inspection schemes and intervals. For ships (DNV, 2014),
there are, for instance, intervals of 2.5, 5 and 15 years. Each interval comes with a prescribed scope
of survey. For offshore units, continuous survey can be opted for (ABS, 2003b). In such case the inspections can be staggered and done at convenient times. For offshore units, the underwater inspections are required to avoid dry docking. For ships this is getting more popular as well and it is included in the vessel rules (ABS, 2014d), however it is not an operational requirement.
Repair requirements: Ships are required to have anomalies repaired and can face lay-up to enforce
repairs. For offshore units, anomalies can also result in conditions of class. Unsatisfactory resolution
ultimate may lead to cease of operation and abandonment. Class and owner will work together to
avoid this (Newport et al., 2004).
Reported anomalies: Ships may face different kind of defects, ranging from corrosion, to fatigue,
buckling to impact damage. As a sailing unit, the variation in loads is large. For offshore units, loads
are more concentrated. It means that waves may hit the unit form a predominant direction or at a predominant waterline level. The extremes may be larger, resulting in green water as a destructive load.
Ships however, are more prone to slamming or impacts with other vessels. Several monitoring
projects have been launched to get a better view on the loads experienced by offshore units, vessel
and naval platforms.
Conclusions
When it comes to extending the lifetime of an existing ship or offshore structure, improved monitoring
and maintenance methods receive greater attention. Risk-based inspections and other assessments of
structural integrity are carried out, including fatigue analysis and assessment of corrosion and other structural damage. A plan is formed to mitigate the effects of the damage, including structural modifications
and repairs.
Other industries and organizations, such as the US Air Force, have more organized and documents
programs to ensure structural longevity. The structural longevity of civil engineering structures, particularly bridges has been subjected to more analysis, and measures such as reinforcement with composites
has been done to some of those structures. New offshore structures for energy collection, such as wind
turbines and tidal energy converters should receive more attention in assessment of structural longevity.
856
10.
CONCLUSIONS & RECOMMENDATIONS
10.1
Conclusions
The purpose of this newly formed committee was to develop an understanding of structural longevity and
the factors that shape this topic, both in importance and how it is managed or maintained. Chapters 2, 3,
and 9 describe the importance of structural longevity from different perspectives. Chapters 4, 6, and 7
describe the information gathered and knowledge needed in order to support an understanding of structural longevity. The use of this information for operational and maintenance activities is covered in Chapters
5 and 8.
From this work, it is clear that there is a growing concern for the structural longevity of ship, offshore
and other marine structures, with systemized methods developed or under development to provide the
owners with information to make a decision on the future of their assets, the lifetime assessment that is
balanced by economic, structural, maintenance, systems and resilience considerations. However, the development and approaches to life extension of assets for the marine industry has been driven by regulatory bodies, with little reporting in the literature by ship and offshore owners and operators of current or
planned practice. Significant work has been done on lifecycle fatigue analysis, but there is little indication
that the results of such studies have been integrated into lifecycle management plans or structural health
monitoring systems other than identifying problem areas for inspections. For many owners, the concept of
structural longevity is limited to following the requirements of classification societies and only performing the structural repairs and modifications that are necessary to last until the next five-year inspection.
Specific conclusions are as follows:
In comparison to the aerospace, chemical processing, and nuclear power industries, the maritime industry is not as prolific in its uptake of lifecycle assessment and management, and may require organizational culture change for their general adoption.
The goal-based standards of the International Maritime Organization, which apply to all international-trading vessels, are enforced though the rules of classification societies, resulting in a class survey
system that is nearly identical for all vessels, and is the main basis for how structural longevity is
handled by vessel owners today.
The design phase is the optimal time to incorporate measures that ensure the desired longevity of a
structure, with the additional cost spent at the design/construction stage much less than the cost of
repair.
Bonded repairs are becoming an attractive alternative to welding based repairs. Although currently
considered to be temporary repairs, ongoing research is aimed at improving the methods of fabrication and materials so that these can become permanent repairs.
Strategies for inspection of structure must consider factors such as probability of detection, cost of
inspection, risks of structural failure if damage is not found, and avoiding costly inspection and
maintenance that could result in unintended and expensive downtime.
A variety of sensing technologies, both active and passive, for the detection of structural damage are
becoming available, including sensor networks that can integrate various measurands of interest and
collect them effectively, with data-feature extraction and statistical pattern recognition.
Structural health monitoring systems on ships and offshore structures have the potential to provide
real-time guidance of structural loading trends and advise on the possibility of an extreme loading
condition, including slam avoidance. These systems can also obtain real-time operational data on
long-term fatigue loads experienced by the ship structure, so that this information can be used to estimate future fatigue life.
Many operators have now implemented a structural integrity management system in order to ensure
safe operation of the asset throughout the service life of the structure, beginning at the design phase
with a risk-based inspection plan that will be updated as the conditions of the asset changes due to
modifications, unplanned events, or extension of the service life...
It is very desirable to develop coatings that can provide corrosion protection but also include possible
early detection of corrosion through the use of smart coatings that can sense corrosion and respond.
10.2
Recommendations
This committee report describes elements important to structural longevity, building upon past ISSC
committee efforts. The future for this committee and topic area rests on the ability to capture and integrate
the aspects that will grow in importance for the prediction and assurance of structural longevity such as
857
uncertainty quantification and management, on-board monitoring and inspection, effective use of collected data, and integration of knowledge gained by collected data in asset management and operation.
On-board hull instrumentation systems as required by regulatory bodies and classification societies
should be developed and deployed as lifecycle structural management tools that consider the holistic
aspects of application of a life extension assessment on an asset.
Structural health monitoring systems have been actively studied, but laboratory proven SHM techniques need to be expanded to address real structures, with more full-field testing of the system to
develop their full maturity. Further development of on-board instrumentation to detect structural
damage should be encouraged, including trial installations of operating ships and offshore structures
to prove out the technology.
Past and future operational environment and operating conditions needs to be known as well as the
fatigue parameters of the materials and an appropriate fatigue model for use in calculating the useful
fatigue life of the asset.
Further research is needed to develop optimum inspection strategies that will safeguard structural
longevity but minimize cost and disruption of service.
Smart coatings that can sense corrosion and respond could be further developed to provide prolonged
corrosion protection and at the same time monitor damage and detect and/or repair damaged coatings.
Repair of structure with composite patch repairs need further exploration and evaluation aimed at
developing systems for permanent repairs.
Surface layer with Ultra-Fine grained microstructure (SUF) materials could provide for crack arrest
and should be explored.
New offshore installations are being rapidly developed in the renewable energy sector and their longterm lifecycle needs should be considered as an extension of ship and traditional offshore structures,
specifically wind turbines, tidal energy converters, and current-driven turbines.
Ship breaking and platform removal are costly enterprises that influence decisions on structural longevity and should be explored further
REFERENCES
ABS 2003a. Guide for Hull Condition Monitoring Systems. Houston, TX: American Bureau of Shipping (ABS).
ABS 2003b. Guide for Surveys Using Risk-Based Inspection for the Offshore Industry. Houston, TX: American
Bureau of Shipping (ABS).
ABS 2007a. Guidance notes on the Inspection, Maintenance and Application of Marine Coating Systems. Houston,
TX: American Bureau of Shipping (ABS).
ABS 2007b. TAPS Tankers Monitored for Corrosion and Cracks: First Commercial Testing Using Acoustic
Emissions in a Marine Environment. Activities, 1213.
ABS 2010. Guide for Building and Classing Naval Vessels (NVR). Houston, TX: American Bureau of Shipping
(ABS).
ABS 2011. Guide for Ice Load Monitoring Systems. Houston, TX: American Bureau of Shipping (ABS).
ABS 2013. Guide for Hull Inspection & Maintenance Program. Houston, TX: American Bureau of Shipping (ABS).
ABS 2014a. Guide for Building and Classing International Naval Ships. Houston, TX: American Bureau of
Shipping (ABS).
ABS 2014b. Guide for Nondestructive Inspection of Hull Welds. Houston, TX: American Bureau of Shipping
(ABS).
ABS 2014c. Rules for Survey After Construction. Houston, TX: ABS.
ABS 2014d. Spectral-based Fatigue Analysis for Floating Production, Storage and Offloading (FPSO) Installations.
Houston, TX: ABS.
Adams, D. 2007. Health monitoring of structural materials and components: methods with applications, Hoboken,
NJ, USA, John Wiley & Sons.
Ahmad, R. & Kamaruddin, S. 2012. An overview of time-based and condition-based maintenance in industrial
application. Computers & Industrial Engineering, 63, 135149.
Alcaide, A., Barrera, E., Ruiz, M. & Aranguren, G. Damage detection on Aerospace structures using PAMELA
SHM System. 2012 2012. 965973.
Alhouli, Y., Ling, D., Kirkham, R. & Elhag, T. Performance Measurement of Marine Vessel Maintenance
Operations: A Case Study of Kuwaiti Shipping Companies. 5th BEAN Annual Conference, 2010/01/01/ 2010.
Liverpool, UK.
Anderson, K., Boulares, J. & Ashe, G. ABS Service Life Evaluation Program and Navy Life Cycle Maintenance.
2013 2013. Pune, India.
Atkins, L. 2009. Structural Integrity Monitoring: Review and Appraisal of Current Technologies for Offshore
Applications. Norwich, UK: Health and Safety Executive.
858
Ayyub, B. M., Akpan, U. O., De Souza, G. F., Koto, T. S. & Luo, X. 2000. Risk-based life cycle management of
ship structures. Ship Structure Committee Report No SSC-SR, 416.
Babu, P. K. S., Mathiazhagan, A. & Nandakumar, C. G. 2014. Corrosion Health Monitoring System for Steel Ship
Structures. International Journal of Environmental Science and Development, 5, 491495.
Balageas, D., Fritzen, C.-P. & Gemes, A. 2006. Structural health monitoring, Wiley Online Library.
Balasubramaniam, K. 2014. Lambwavebased structural health monitoring technique for inaccessible regions in
complex composite structures. Structural Control and Health Monitoring, 21, 817832.
Baldwin, C., Niemczuk, J., Kiddy, J. & Salter, T. Review of fiber optic accelerometers. 2005 2005.
Bang, H.-j., Janga, M.-s., Kima, S.-w. & Shina, H.-k. Structural health monitoring of offshore jacket structure using
fiber Bragg grating sensor array. EWEA, 2011 2011. Brussels, Belgium.
Bao, X. & Chen, L. 2012. Recent progress in distributed fiber optic sensors. Sensors, 12, 86018639.
Baran, I., Nowak, M., Jagenbrein, A. & Buglacki, H. Acoustic Emission Monitoring of Structural Elements of a
Ship for Detection of Fatigue and Corrosion Damages. 2012 2012.
Barhoumi, M. & Storhaug, G. 2014. Assessment of whipping and springing on a large container vessel.
International Journal of Naval Architecture and Ocean Engineering, 6, 442458.
Bent, A. A. & Hagood, N. W. 1997. Piezoelectric fiber composites with interdigitated electrodes. Journal of
Intelligent Material Systems and Structures, 8, 903919.
Berg, D. v. d., Tammer, M. D. & Kaminski, M. L. Updating Fatigue Reliability of Uninspectable Joints using
Structurally Correlated Inspection Data. 2014 2014. Busan, Korea: International Society of Offshore and Polar
Engineers.
Bergman, J. D., Lee, S. J., Chung, H. & Li, I. Real-Time Active Pipeline Integrity Detection (RAPID) System for
Corrosion Detection and Quantification. 2014 2014.
Bharadwaj, U. & Wintle, J. 2011. Risk-Based Optimization of Inspection Planning in Ships. Journal of Ship
Production and Design, 27, 111117.
BMT 2006. White Paper: The Procurement of Naval and Government Ships. Bath: BMT Defence Services.
Braidwood, I. T. 2013. Life Extensions of a 100,000 Deadweight Floating Storage Unit for Crude Oil. ASNE Fleet
Maintenance and Modernization Symposium. San Diego, California.
Brailsford, S. C. & Hilton, N. A. 2001. A comparison of discrete event simulation and system dynamics for
modelling health care systems. In: Riley, J. (ed.) Planning for the Future: Health Service Quality and Emergency
Accessibility. Operational Research Applied to Health Services (ORAHS). Glasgow Caledonian University.
Bureau Veritas 1998. Type Approval of Non Destructive Testing Equipment Dedicated to Underwater Inspection
France: Bureau Veritas (BV).
Bureau Veritas 1999. Harmonized condition assessment programme (NI 456). France: Bureau Veritas (BV).
Bureau Veritas 2007. Guidelines for the application of the IMO performance standard for protective coatings (NI
531). France: Bureau Veritas (BV).
Bureau Veritas 2010. Risk based verification of floating offshore units (NI 567). France: Bureau Veritas (BV).
Bureau Veritas 2014. Rules for the Classification of Naval Ships. France: Bureau Veritas (BV).
Bureau Veritas 2015. Rules for the Classification of Steel Ships (NR 467). France: Bureau Veritas (BV).
Burns, J. M. 2012. Development and characterisation of a fibre-optic acoustic emission sensor. University of
Birmingham.
Cabos, C., Jaramillo, D., Stadie-Frohbs, G., Renard, P., Ventura, M. & Dumas, B. Condition assessment scheme
for ship hull maintenance. 2008 2008.
Cmara, M. C. & Cyrino, J. C. R. Structural Reliability Applications in Design and Maintenance Planning of Ships
Subjected to Fatigue and Corrosion. 2012 2012. American Society of Mechanical Engineers, 503514.
Caridis, P. A. 2009. Inspection, repair and maintenance of ship structures, Witherby Seamanship International.
ClassNk 2014a. Rules for Hull Monitoring Systems. Japan: Nippon Kaiji Kyokai (ClassNK).
ClassNk 2014b. Rules for the Survey and Construction of Steel Ships. Japan: Nippon Kaiji Kyokai (ClassNK).
CO-PATCH. 2013. Composite Patch Repair for Marine and Civil Engineering Infrastructure Applications
[Online]. Available: www.co-patch.com.
Collette, M. 2011. Hull Structures as a System: Supporting Lifecycle Analysis: Hull Structures as a System. Naval
Engineers Journal, 123, 4555.
Cortez, N. E., Jozu Filho, V. & Baptista, F. G. 2013. A new microcontrolled structural health monitoring system
based on the electromechanical impedance principle. Structural Health Monitoring, 12, 1422.
Cortez, N. E., Vieira Filho, J. & Baptista, F. G. 2014. Design and implementation of wireless sensor networks for
impedance-based structural health monitoring using ZigBee and Global System for Mobile Communications.
Journal of Intelligent Material Systems and Structures.
Cross, M. 1988. Raising the Value of Maintenance in the Corporate Environment. Management Research News, 11,
8-11.
De Baere, K., Verstraelen, H., Rigo, P., Van Passel, S., Lenaerts, S. & Potters, G. 2013a. Reducing the cost of
ballast tank corrosion: an economic modeling approach. Marine Structures, 32, 136152.
De Baere, K., Verstraelen, H., Rigo, P., Van Passel, S., Lenaerts, S. & Potters, G. 2013b. Study on alternative
approaches to corrosion protection of ballast tanks using an economic model. Marine Structures, 32, 117.
Demsetz, L. & Cabrera, J. 1999. Detection Probability Assessment of Visual Inspection of Ships. Washington D.C.:
Ship Structure Committee.
859
Dier, A. F. 2004. Assessment of repair techniques for ageing or damaged structures. Surrey: MSL Engineering
Limited.
Din, E. 2007. 19902: 2008-07: Petroleum and natural gas industriesFixed steel offshore structures (ISO 19902:
2007). English version EN ISO, 19902.
DNV 2005. Rules for Classification of High Speed and Light Craft and Naval Surface Craft. Hvik, Norway: Det
Norske Veritas Classification.
DNV 2008. DNV Class Note 10.2 Guidance for Condition Monitoring. Hvik, Norway: Det Norske Veritas AS.
DNV 2009. DNV Guideline 10 Guide for Ultrasonic Thickness Measurements of Ships Classed with DNV. Hvik,
Norway: Det Norske Veritas AS.
DNV 2010a. DNV-RP-G101 Risk Based Inspection of Offshore Topsides Static Mechanical Equipment. Hvik,
Norway: Det Norske Veritas AS.
DNV 2010b. Hull Integrity Advisory Services Case Studies. Hvik, Norway: Det Norske Veritas AS.
DNV 2011. DNV-RP-G103 Non-Intrusive Inspection. Hvik, Norway: Det Norske Veritas AS.
DNV 2012a. Design, Fabrication, Operation and Qualification of Bonded Repair of Steel Structures. Hvik,
Norway: Der Norske Veritas As.
DNV 2012b. DNV Class Note 7 Non-destructive testing. Hvik, Norway: Det Norske Veritas AS.
DNV 2013. Determination of Structural Capacity by Non-linear FE analysis Methods. Hvik, Norway: Der Norske
Veritas AS.
DNV 2014. Fatigue Assessment of Ship Structures (Classification No. 30.7). Hvik, Norway: Der Norske Veritas
AS.
Dragan, K., Dziendzikowski, M., Kurnyta, A., Leski, A. & Uhl, T. Active structural integrity monitoring of the
aircraft based on the PZT sensor network the SYMOST project. 2013 2013. 10151022.
Drummen, I., Schiere, M., Dallinga, R., Thornhill, E. & Stambaugh, K. Full Scale Trials, Monitoring and Model
Testing Conducted to Assess the Structural Fatigue Life of a New US Coast Guard Cutter. 2014 2014. Linthicum
Heights, Maryland.
Duncan, R. G., Soller, B. J., Gifford, D. K., Kreger, S. T., Seeley, R. J., Sang, A. K., Wolfe, M. S. & Froggatt, M.
E. OFDR-based distributed sensing and fault detection for single-and multi-mode avionics fiber-optics. 2007
2007.
Drager, C., Heinzelmann, A. & Riederer, D. 2013. A wireless sensor system for structural health monitoring with
guided ultrasonic waves and piezoelectric transducers. Structure and Infrastructure Engineering, 9, 11771186.
Eccles, T. J., Ashe, G. & Albrecht, S. 2010. The achieving service life program. Naval Engineers Journal, 122,
103112.
Echtermeyer, A. T., McGeorge, D., Grave, J. H. L. & Weitzenbck, J. 2014. Bonded patch repairs for metallic
structuresA new recommended practice. Journal of Reinforced Plastics and Composites, 33, 579585.
Eckstein, B., Bach, M., Bockenheimer, C., Cheung, C., Chung, H., Zhang, D. & Li, F. Large Scale Monitoring of
CFRP Structures by Acousto-UltrasonicsA Flight Test Experience, 2013. 2013 2013. 528535.
Ersdal, G., Hrnlund, E. & Spilde, H. Experience from Norwegian Programme on Ageing and Life Extension. 2011
2011. American Society of Mechanical Engineers, 517522.
Fabrycky, W. J. 2003. Evaluation in Systems Engineering. UNESCO Encyclopedia of Life Support Systems. Article
E1.28.04.05 ed. Paris: Developed under the Auspices of the UNESCO, EOLSS Publishers Co. Ltd.
Farrar, C. R., Park, G., Allen, D. W. & Todd, M. D. 2006. Sensor network paradigms for structural health
monitoring. Structural Control and Health Monitoring, 13, 210225.
Farrar, C. R. & Worden, K. 2007. An introduction to structural health monitoring. Philosophical Transactions of the
Royal Society A: Mathematical, Physical and Engineering Sciences, 365, 303315.
Farrar, C. R. & Worden, K. 2012. Structural health monitoring: a machine learning perspective, John Wiley &
Sons.
Flockhart, C., Phelps, B. & Cannon, S. Monitoring of plate structures using guided ultrasonic waves. AMEC 2010,
2010/12/06/8 2010. Singapore.
Frangopol, D. M. & Soliman, M. 2013. Application of Genetic Algorithms to the Life-cycle Management
Optimization of Civil and Marine Infrastructure Systems. In: Tsompanakis, Y., Ivnyi, P. & Topping, B. H. V.
(eds.) Civil and Structural Engineering Computational Methods. Stirlingshire, UK: Saxe-Coburg Publications,
117128.
Fredriksen, A. 2012. 10 Years of Naval Ship Classification The Start of a New Direction. Defence Technology
Asia (DTA) 2012. Singapore.
Fromme, P. Monitoring of plate structures using guided ultrasonic waves. 2008 2008. AIP Publishing, 7885.
Gahbauer, S. 2012. CBM Goes Wireless. Plant West.
Galindez-Jamioy, C. A. & Lpez-Higuera, J. M. 2012. Brillouin distributed fiber sensors: an overview and
applications. Journal of Sensors, 2012.
Gao, D.-w., Shi, G.-j. & Wang, D.-y. 2012. Residual ultimate strength of hull structures with crack and corrosion
damage. Engineering Failure Analysis, 25, 316328.
Germanischer Lloyd 2012. Rules for Classification and Construction III Naval Ship Technology. Hamburg,
Germany: Germanischer Lloyd.
Germanischer Lloyd 2015. Rules for Classification and Construction. Ship Technology Part 0 Classification and
Surveys. Hamburg, Germany: Germanischer Lloyd.
Giurgiutiu, V. 2007. Structural health monitoring: with piezoelectric wafer active sensors, Academic Press.
860
Glenn, I. F. 1999. Fatigue Resistant Detail Design Guide for Ship Structures. Ship Structure Committee.
Glisic, B. & Inaudi, D. 2008. Fibre optic methods for structural health monitoring, John Wiley & Sons.
Godnez-Azcuaga, V. F., Inman, D. J., Ziehl, P. H., Giurgiutiu, V. & Nanni, A. Recent advances in the development
of a self-powered wireless sensor network for structural health prognosis. 2011 2011. International Society for
Optics and Photonics, 798325-798325-7.
Gonzlez Nez, M. A., Proust, A., Lenain, J.-C. & Godinez, V. 2013. Active Corrosion Monitoring - Successful
Case Studies. Mega Rust 2013: Naval Corrosion Conference. Newport News, Virginia.
Grassman, J. M. & Hildstrom, G. A. 2003. Structural Trials of the RV Tritona Status Update and Quick-Look
Report [Online]. Available: http://hildstrom.com/publications/pub-StructuralTrialsOfTheRVTriton-Feb-10-2003Final.pdf.
Gratsos, G. A., Psaraftis, H. N. & Zachariadis, P. Life Cycle Cost of Maintaining the Effectiveness of a Ship's
Structure and Environmental Impact of Ship Design Parameters: An Update. 2009 2009. Athens, Greece.
Grisso, B. L. 2013. Development of a structural health monitoring prototype for ship structures. Ship Structure
Committee.
Grisso, B. L., Park, G., Salvino, L. W. & Farrar, C. R. 2011. Structural Damage Identification in Stiffened Plate
Fatigue Specimens Using Piezoelectric Active Sensing. Proceedings of the 8th International Workshop on
Structural Health Monitoring, 16831690.
Groden, M. & Collette, M. 2013. Bayesian Updating of Marine Structural Reliability Models Based on In-Service
Measurements. 2013 SNAME Annual Meeting. Bellevue, WA.
Gudze, M. T. & Melchers, R. E. 2007. Prediction of corrosion in ballast tanks of naval ships using operational
characteristics. Corrosion Control, 7, 2528.
Hageman, R., Drummen, I., Stambaugh, K., Dupau, T., Herel, N., Derbanne, Q., Schiere, M., Shin, Y. & Kim, P.
Structural Fatigue Loading Predictions and Comparisons with Test Data for a New Class of US Coast Guard
Cutters. 2014 2014. Linthicum Heights, Maryland.
Hart, D., Udinski, E. P. & Hayden, M. J. Fatigue Performance of Composite Patch Repaired Aluminum Plates. 2014
2014. Linthicum Heights, Maryland.
Hecht, M. & An, X. A stochastic model for determining inspection intervals for large marine vessels. 2004 2004.
IEEE, 559564.
Herszberg, I., Bannister, M. K., Li, H. C. H., Thomson, R. S. & White, C. Structural health monitoring for advanced
composite structures. 2007 2007. Kyoto, Japan, 113.
Hess III, P. E. 2003. Reliability-based operational performance metrics for ship structures. Bethesda, MD: Naval
Surface Warfare Center, Carderock Division.
Hoffman, P. C. 2009. Fleet management issues and technology needs. International Journal of Fatigue, 31,
16311637.
Hoppe, H. International regulations for high-speed craft an overview. 2005 2005. St. Petersburg, Russia.
HSE 1996. Intercalibration of offshore NDT (ICON). Health and Safety Executive (HSE).
HSE 2002. Offshore Technology Report 2000/018 PoD/PoS curves for Non-Destructive Examination. Norwich:
Health And Safety Executive.
HSE.
2014.
Ageing
&
Life
Extension
of
Offshore
Installations
[Online].
Available:
http://www.hse.gov.uk/offshore/ageing.htm.
IACS 2005. Container ships - Guidelines for Surveys, Assessment and Repair of Hull Structures. London:
International Association of Classification Societies.
IACS 2007a. Bulk Carriers - Guidelines for Surveys, Assessment and Repair of Hull Structure. London:
International Association of Classification Societies.
IACS 2007b. Double Hull Oil Tankers - Guidelines for Surveys, Assessment and Repair of Hull Structures.
London: International Association of Classification Societies.
IACS 2011. Classification Societies - Their Key Role. London: International Association of Classification Societies
(IACS).
IACS 2014a. Classification Societies - What, Why and How? London: International Association of Classification
Societies.
IACS 2014b. Common Structural Rules. International Association of Classification Societies.
IAEA 2000. Training Guidelines in Non-destructive Testing Techniques: Liquid Penetrant and Magnetic Particle
Testing at Level 2. Vienna, Austria: International Atomic Energy Agency.
IMO 2000. International Code of Safety for High Speed Craft. International Maritime Organization (IMO).
IMO 2004. Hull and Structural Surveys - Model Course 3.07, London, International Maritime Organization.
IMO.
2014.
High-Speed
Craft
[Online].
Available:
http://www.imo.org/OurWork/Safety/Regulations/Pages/HSC.aspx.
Inaudi, D. Overview of 40 Bridge Structural Health Monitoring Projects. International Bridge Conference,
2009/06/14/17 2009. Pittsburg, PA, 343350.
Inaudi, D. Long-gauge strain sensors for underwater and deep-water applications. 2011 2011. International Society
for Optics and Photonics, 77535R-77535R-4.
Ishikawa, T., Inoue, T., Funatsu, Y. & Otani, J. Evaluation Method for Crack Arrestability of Steel Plates Using
Small-Scale Fracture Test Results. 2012 2012. International Society of Offshore and Polar Engineers.
861
Ito, M., Kaneko, M., Nishimura, S. & Sato, H. Development of Corrosion Resistant Steel for Bottom Plates of
Crude Oil Tankers and Onboard Evaluation Results. 2012 2012. Brazil: American Society of Mechanical
Engineers, 223228.
Ivce, R., Jurdana, I. & Mrak, Z. Longitudinal ship's hull strength monitoring with optical fiber sensors. 2009 2009.
IEEE, 167170.
Iyer, P. 1999. The effect of maintenance policy on system maintenance and system life-cycle cost [electronic
resource]. Virginia Tech.
Jardine, A. K. S. & Tsang, A. H. C. 2013. Maintenance, replacement, and reliability: theory and applications, Boca
Raton, Taylor & Francis Group, LLC.
Jean, G. 2008. New Ships are Breaking The Bank So the Navy is Fixing its Old Ones. National Defense.
Jensen, J. J., Pedersen, P. T., Shi, B., Wang, S., Petricic, M. & Mansour, A. E. Wave induced extreme hull girder
loads on containerships. 2008 2008. Houston, Texas, 128152.
Jeppesen.
2014.
The
Vessel
Voyage
Optimisation
Solution
(VVOS)
[Online].
Available:
http://ww1.jeppesen.com/marine/commercial/vvos/commercial-marine-product.jsp.
Jia, J. 2008. An efficient nonlinear dynamic approach for calculating wave induced fatigue damage of offshore
structures and its industrial applications for lifetime extension. Applied Ocean Research, 30, 189198.
Jiles, D. C. 1990. Review of magnetic methods for nondestructive evaluation (Part 2). NDT International, 23,
8392.
Kalghatgi, S. G., Serratella, C. & Hagan, J. B. 2009. Hull inspection and maintenance systems. American Bureau of
Shipping (ABS) Technical Papers.
Kaminski, M. L. & Aalberts, P. SS: FPSOs and Floating Production Systems: Implementation of the Monitas
system for FPSOs. 2010 2010. Offshore Technology Conference.
Kattan, M. R., Speed, L. A. & Broderick, D. Minimising the risk of coating failures. 2013 2013. Houston, Texas.
Kiddy, J. S., Baldwin, C. S., Salter, T. & Chen, P. C. Structural load monitoring of the RV Triton using fiber optic
sensors. 2002 2002. International Society for Optics and Photonics, 462472.
Kim, S. & Frangopol, D. M. 2011. Optimum inspection planning for minimizing fatigue damage detection delay of
ship hull structures. International Journal of Fatigue, 33, 448459.
Komoto, H., Tomiyama, T., Silvester, S. & Brezet, H. 2011. Analyzing supply chain robustness for OEMs from a
life cycle perspective using life cycle simulation. International Journal of Production Economics, 134, 447457.
KR 2014. Rules for the Classification of Steel Ships. Korea: Korean Register of Shipping.
Krata, P. & Szlapczynska, J. 2012. Weather Hazard Avoidance in Modeling Safety of Motor-driven Ship for
Multicriteria Weather Routing. Methods and Algorithms in Navigation: Marine Navigation and Safety of Sea
Transportation, 6, 7178.
Ku, A., Nietmann, B., Krzonkalla, V., Wang, X., Chen, J. Y. & Howser, R. Structural Reliability Applications in
Risk-Based Inspection Plans for Semi-Submersible Floating Structures. 2012 2012. Perth, Australia.
Kwon, K. & Frangopol, D. M. 2012a. Fatigue Life Assessment and Lifetime Management of Aluminum Ships
Using Life-Cycle Optimization. Journal of Ship Research, 56, 91105.
Kwon, K. & Frangopol, D. M. 2012b. System Reliability of Ship Hull Structures Under Corrosion and Fatigue.
Journal of Ship Research, 56, 234251.
L'Hostis, D., Kaminski, M. L. & Aalberts, P. Overview of the Monitas JIP. 2010 2010. 36.
LHostis, D., van der Cammen, J., Hageman, R. & Aalberts, P. Overview of the Monitas II Project. 2013 2013.
International Society of Offshore and Polar Engineers.
Landet, E., Oma, N., Ersdal, G., Sigurdsson, G. & Sorensen, T. Assessment of Ageing Structures: Case Studies.
30th International Conference on Ocean, Offshore and Arctic Engineering, 2011 2011. Rotterdam, The
Netherlands, 487498.
Lee, A. K., Serratella, C., Wang, G., Basu, R. & Spong, R. Flexible approaches to risk-based inspection of FPSOs.
2006 2006. Offshore Technology Conference.
LeHardy, P. & Walters, J. Electronic Underwater Thickness Inspections on Navy Ships: Using Technology to
Achieve Precise, Measurable, and Repeatable Results. 2013 2013. Newport News, Virginia: American Society of
Naval Engineers.
Li, F., Murayama, H., Kageyama, K., Meng, G., Ohsawa, I. & Shirai, T. 2010. A fiber optic doppler sensor and its
application in debonding detection for composite structures. Sensors, 10, 59755993.
Li, H. C. H., Herszberg, I., Davis, C. E., Mouritz, A. P. & Galea, S. C. 2006. Health monitoring of marine
composite structural joints using fibre optic sensors. Composite Structures, 75, 321327.
Li, S. & Wu, Z. 2007. Development of distributed long-gage fiber optic sensing system for structural health
monitoring. Structural Health Monitoring, 6, 133-143.
Li, Z., Mao, W., Ringsberg, J. W., Johnson, E. & Storhaug, G. 2014. A comparative study of fatigue assessments of
container ship structures using various direct calculation approaches. Ocean Engineering, 82, 6574.
Lloyds Register 2014a. Rules and Regulations for Classification of Naval Ships. England: Lloyd's Register.
Lloyds Register 2014b. Rules and Regulations for the Classification of Naval Ships.
Lloyds Register 2014c. Tank Coatings Condition Guide. England: Lloyd's Register.
Lotsberg, I. & Sigurdsson, G. A New Recommended Practice for Inspection Planning of Fatigue Cracks in Offshore
Structures Based on Probabilistic Methods. 2014 2014. American Society of Mechanical Engineers,
V005T03A005-V005T03A005.
862
Lynch, J. P., Law, K. H., Kiremidjian, A. S., Carryer, E., Farrar, C. R., Sohn, H., Allen, D. W., Nadler, B. & Wait,
J. R. 2004. Design and performance validation of a wireless sensing unit for structural monitoring applications.
Structural Engineering and Mechanics, 17, 393408.
Magoga, T. F., Aksu, S., Cannon, S., Ojeda, R. & Thomas, G. Indentification of slam events experienced by a highspeed craft. 2014 2014. Glasgow, UK, 113.
Maldague, X. 2001. Theory and practice of infrared technology for nondestructive testing, New York, Wiley
Interscience Publication.
Mao, W., Ringsberg, J. & Rychlik, I. What is the potential of using ship fatigue routing in terms of fatigue life
extension? , 2012 2012. 681687.
Martinez, C., Zheng, Y., Easton, D., Farinholt, K. & Park, G. 2011. Strain sensors for high field pulse magnets.
Structural Dynamics, Volume 3. Springer, 15371551.
May, P. 2014. Gaps and Issues Regarding Codes and Standards for SIM. Structural Integrity Management
Conference. Aberdeen, UK.
Mendoza, E., Prohaska, J., Kempen, C., Esterkin, Y., Sun, S. & Krishnaswamy, S. Distributed Fiber Optic Acoustic
Emission Sensor (FAESense) System for Condition Based Maintenance of Advanced Structures. 2013 2013.
Optical Society of America, 375382.
Micron, O. 2013. Case Study Team Alinghi. Georgia.
MISTRAS 2009. An Authorative Newsletter on Acoustic Emission. PAC News.
MODU, I. M. O. 2001. The Code for the Construction and Equipment of Mobile Offshore Drilling Units.
International Maritime Organization (IMO).
Moe, E., Holtsmark, G. & Storhaug, G. Full scale measurements of the wave induced hull girder vibrations of an
ore carrier trading in the North Atlantic. 2005 2005.
Morel, G., Wyckhuse, A. & Maldague, X. 2001. Active pulsed infrared thermography for concrete bridge deck
inspection. Journal of the Canadian Institute for NDE [CINDE] 4th ed.
Murawski, L., Opoka, S., Majewska, K., Mieloszyk, M., Ostachowicz, W. & Weintrit, A. 2012. Investigations of
Marine Safety Improvements by Structural Health Monitoring Systems. TransNav, the International Journal on
Marine Navigation and Safety of Sea Transportation, 6.
Murayama, H., Kageyama, K., Naruse, H., Shimada, A. & Uzawa, K. 2003. Application of fiber-optic distributed
sensors to health monitoring for full-scale composite structures. Journal of Intelligent Material Systems and
Murayama, H., Tachibana, K., Hirano, Y., Igawa, H., Kageyama, K., Uzawa, K. & Nakamura, T. Distributed strain
and load monitoring of 6 m composite wing structure by FBG arrays and long-length FBGs. 2012 2012.
International Society for Optics and Photonics, 84212D-84212D-4.
Murayama, H., Wada, D. & Igawa, H. 2013. Structural health monitoring by using fiber-optic distributed strain
sensors with high spatial resolution. Photonic Sensors, 3, 355376.
Nappi Sr, N. & Collette, M. 2013. Structural Design of Naval Vessels: Recent Developments and Emerging
Challenges.
Nelson, B. N., Slebodnick, P., Wegand, J., Lysogorski, D. & Lemieux, E. J. 2012. Corrosion Sensors and ISIS: A
Condition-Based Approach to the Inspection and Preservation of Tanks and Voids on US Navy Ships. Naval
Engineers Journal, 124, 115129.
Newport, A., Basu, R. & Peden, A. Structural Modifications to the FPSO Kuito Cargo Tanks. OMAE Specialty
Symposium on FPSO Integrity, 2004 2004. Houston, TX, 279286.
Nichols, J. M., Seaver, M., Trickey, S. T., Scandell, K. & Salvino, L. W. 2008. Finer-Optic Strain Monitoring on a
Navy Cruiser. DTIC Document.
Nielsen, U. D., Jensen, J. J., Pedersen, P. T. & Ito, Y. 2011. Onboard monitoring of fatigue damage rates in the hull
girder. Marine Structures, 24, 182206.
Noland, J. M., Hart, D. C. & Udinski, E. P. Initiatives in Bonded Ship Structural Repairs. 2013 2013. Arlington,
Virginia, USA: American Society of Naval Engineers.
NORSOK 2009. Assessment of structural integrity for existing offshore load-bearing structures. Norway: Standards
Norway.
Ntuen, C. A. & Moore, T. L. 1986. Approaches to life cycle cost analysis with system availability constraintsa
review. Microelectronics Reliability, 26, 341354.
Okasha, N. M., Frangopol, D. M., Saydam, D. & Salvino, L. W. 2011. Reliability analysis and damage detection in
high-speed naval craft based on structural health monitoring data. Structural Health Monitoring, 10, 361379.
Okuhara, Y. & Matsubara, H. 2005. Memorizing maximum strain in carbon-fiber-reinforced plastic composites by
measuring electrical resistance under pre-tensile stress. Composites science and technology, 65, 21482155.
Ou, J. & Li, H. 2010. Structural health monitoring in mainland China: review and future trends. Structural Health
Monitoring, 9, 219231.
Page, J. 2002. Flexibility in Early Stage Design of US Navy Ships: An Analysis of Options. MSc Thesis,
Massachusetts Institute of Technology.
Paik, J. K. & Melchers, R. E. 2014. Condition assessment of aged structures, Elsevier.
Park, G. & Inman, D. J. 2007. Structural health monitoring using piezoelectric impedance measurements.
Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 365,
373392.
863
Park, G., Sohn, H., Farrar, C. R. & Inman, D. J. 2003. Overview of piezoelectric impedance-based health
monitoring and path forward. Shock and Vibration Digest, 35, 451464.
Park, S., Grisso, B. L., Inman, D. J. & Yun, C.-B. 2007. MFC-based structural health monitoring using a
miniaturized impedance measuring chip for corrosion detection. Research in Nondestructive Evaluation, 18,
139150.
Perez-Herrera, R. A. & Lopez-Amo, M. 2013. Fiber optic sensor networks. Optical Fiber Technology, 19, 689699.
Perii, N. & Tygesen, U. T. Cost-Effective Load Monitoring Methods for Fatigue Life Estimation of Offshore
Platform. 2014 2014. American Society of Mechanical Engineers, V04BT02A005-V04BT02A005.
Phelps, B. & Morris, B. 2013. Review of Hull Structural Monitoring Systems for Navy Ships. DTIC Document.
Raghavan, A. & Cesnik, C. E. S. 2007. Review of guided-wave structural health monitoring. Shock and Vibration
Digest, 39, 91116.
Rao, Y.-J. 1997. In-fibre Bragg grating sensors. Measurement science and technology, 8.
Rao, Y.-J. 2006. Recent progress in fiber-optic extrinsic FabryPerot interferometric sensors. Optical Fiber
Technology, 12, 227237.
Rathje, H. Shipboard Weather Routing Operational Benefits. 2009 2009. Hamburg, Germany: Germanischer
Llyod AG.
Ricci, F., Mal, A. K., Monaco, E., Maio, L., Boffa, N. D., Di Palma, M. & Lecce, L. Guided Waves in Layered
Plate with Delaminations. 2014 2014.
Roberge, P. R. 2007. Corrosion inspection and monitoring, John Wiley & Sons.
Roberton, D. 2014. Residual Life Assessment of Composite Structures: With Application to All Weather Lifeboats.
EngD Thesis, University of Southampton.
Rodrigues, C., Inaudi, D. & Glii, B. 2013. Longgauge fibre optic sensors: performance comparison and
applications. International Journal of Lifecycle Performance Engineering, 1, 209233.
Rucker, W., Hille, F. & Rohrmann, R. 2006. F08a Guideline for the Assessment of Existing Structures. EU FP5
Network - Structural Assessment, Monitoring & Control (SAMCO) Final Report.
Sabra, K. G. & Huston, S. 2011. Passive structural health monitoring of a high-speed naval ship from ambient
vibrations. The Journal of the Acoustical Society of America, 129, 29912999.
Samarakoon, S., Ratnayake, R. M. C. & Siriwardane, S. Structural Integrity Control of Ageing Offshore Structures:
Repairing and Strengthening With Grouted Connections. 2013 2013. France: American Society of Mechanical
Engineers, V003T03A032-V003T03A032.
Sheinberg, R., Cleary, C., Stambaugh, K. & Storhaug, G. Investigation of Wave Impact and Whipping Response on
the Fatigue Life and Ultimate Strength of a Semi-Displacement Patrol Boat. 2011/09// 2011. Honolulu, Hawaii.
Shinozuka, M. 1990. Relation of inspection findings to fatigue reliability. DTIC Document.
Sirohi, J. & Chopra, I. 2000. Fundamental understanding of piezoelectric strain sensors. Journal of Intelligent
Material Systems and Structures, 11, 246257.
Slebodnick, P., Martin, J. R., Wegand, J., Stanke, I., Lemieux, E. J., Nelson, B. N., Kuljian, G., Zuskin, D., Ingle,
M., Steele, R. & Cuzzocrea, J. L. 2013. A 13 Year Assessment of the Condition of Seawater Ballast Tanks of the
USS ESSEX (LHD 2) preserved with High Solids Coatings. ASNE Fleet Maintenance and Modernization
Symposium. San Diego, California.
SmartFiber 2014. Miniaturized structural monitoring system with autonomous readout microtechnology and fiber
sensor network: D8.4 Final Technology white paper.
Song, K.-Y., He, Z. & Hotate, K. Distributed Strain Measurement with Millimeter-Order Spatial Resolution Based
on Brillouin Optical Correlation Domain Analysis and Beat Lock-in Detection Scheme. 2006 2006. Optical
Society of America.
Stacey, A. KP4: Ageing and Life Extension Inspection Programme for Offshore Installations. 2011 2011. American
Society of Mechanical Engineers, 33-48.
Stacey, A., Birkinshaw, M. & Sharp, J. V. Life Extension Issues for Ageing Offshore Installations. the 27th
International Conference on Offshore Mechanics and Arctic Engineering, 2008 2008. Estoril, Portugal:
American Society of Mechanical Engineers, 199215.
Stacey, A. & Sharp, J. V. Ageing and Life Extension Considerations in the Integrity Management of Fixed and
Mobile Offshore Installations. 2011 2011. American Society of Mechanical Engineers, 4965.
Stambaugh, K. & Barry, C. 2014. Naval Ship Structure Service Life Considerations. Naval Engineers Journal, 126,
103117.
Stambaugh, K., Drummen, I., Cleary, C., Sheinberg, R. & Kaminski, M. L. Structural Fatigue Life Assessment and
Sustainment Implications for a new class of US Coast Guard Cutters. 2014 2014. Linthicum Heights, Maryland.
Storhaug, G. The Effect of Heading on Springing and Whipping Induced Fatigue Damage Measured on Container
Vessels. 6th International Conference on Hydroelasticity, 2012/09/20/ 2012. Tokyo, Japan.
Swartz, R. A., Zimmerman, A. T., Lynch, J. P., Rosario, J., Brady, T., Salvino, L. & Law, K. H. 2012. Hybrid
wireless hull monitoring system for naval combat vessels. Structure and Infrastructure Engineering, 8, 621638.
Swift, K. G. & Brown, N. J. 2003. Implementation strategies for design for manufacture methodologies.
Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, 217,
827833.
Szlapczynska, J. 2013. Multicriteria Evolutionary Weather Routing Algorithm in Practice. TransNav: International
Journal on Marine Navigation and Safety of Sea Transportation, 7, 6165.
864
Tammer, M. & Kaminski, M. L. Fatigue Oriented Risk Based Inspection and Structural Health Monitoring of
FPSOs. Twenty-third International Offshore and Polar Engineering, 2013 2013. Anchorage, Alaska:
International Society of Offshore and Polar Engineers, 438449.
Temple, D. W. & Collette, M. Optimum Lifetime Maintenance Schedule for Naval Vessels Subjected to Fatigue
and Corrosion. 2013 2013. Changwon City, Korea.
Torkildsen, H. E., Grovlen, A., Skaugen, A., Wang, G., Jensen, A. E., Pran, K. & Sagvolden, G. 2005.
Development and applications of full-scale ship hull health monitoring systems for the Royal Norwegian Navy.
DTIC Document.
Tran Nguyen, K., Garbatov, Y. & Guedes Soares, C. 2012. Fatigue damage assessment of corroded oil tanker
details based on global and local stress approaches. International Journal of Fatigue, 43, 197206.
Tran Nguyen, K., Garbatov, Y. & Guedes Soares, C. 2013. Spectral fatigue damage assessment of tanker deck
structural detail subjected to time-dependent corrosion. International Journal of Fatigue, 48, 147155.
Tscheliesnig, P. Acoustic Emission Testing (AT) for the detection of corrosion attack on ships (oil tankers).
ECDNT Conference, 2006/09/25/29 2006. Berlin, 18.
Turan, O., ler, A., Lazakis, I., Rigo, P. & Caprace, J.-D. 2009. Maintenance/repair and production-oriented life
cycle cost/earning model for ship structural optimisation during conceptual design stage. Ships and Offshore
Umeda, Y., Nonomura, A. & Tomiyama, T. 2000. Study on life-cycle design for the post mass production
paradigm. AI EDAM, 14, 149161.
US DOD 2005. MIL-STD-1530C - Department of Defense Standard Practice: Aircraft Structural Integrity Program
(ASIP).
Vaara, P. & Leinonen, J. 2012. Technology Survey on NDT of Carbon-fiber Composites. Kemi-Tornio University
of Applied Sciences.
Vallen, S. 2012. Acoustic Emission Sensors Specification.
van der Horst, M. P., Kaminski, M. L. & Puik, E. Methods for Sensing and Monitoring Fatigue Cracks and Their
Applicability for Marine Structures. 2013 2013. Anchorage, Alaska: International Society of Offshore and Polar
Engineers, 445462.
Vugts, J. H. 2013. Handbook of bottom founded offshore structures - Part 1. General Features of Offshore
Structures and Theoretical Background, Eburon.
Wagstaff, S., Horton, J. & Kwong, A. Asset Integrity Management & Corrosion Assesment.pdf. 2013/10/07/9 2013.
Australia: Frazer-Nash Consultancy.
Walshe, J. H., Watson, D. A., Tate, A. J., Austen, S. J., Nguyen, D. K., Blake, T. D. & Blake, J. I. R. Integrated and
Concurrent, Engineering and Cost Modelling, for Through-Life Asset Management within the RNLI. 2011 2011.
Trieste, Italy: The Royal Institution of Naval Architects.
Wan, C., Hong, W., Liu, J., Wu, Z., Xu, Z. & Li, S. 2013. Bridge Assessment and Health Monitoring with
Distributed Long-Gauge FBG Sensors. International Journal of Distributed Sensor Networks, 2013.
Wang, G., Pran, K., Sagvolden, G., Havsgrd, G. B., Jensen, A. E., Johnson, G. A. & Vohra, S. T. 2001. Ship hull
structure monitoring using fibre optic sensors. Smart Materials and Structures, 10.
Weitzenbck, J. R. & McGeorge, D. 2005. BONDSHIP project guidelines. Det norske Veritas.
Wheat, H. G. Smart Coatings for Corrosion Detection-A Review of Recent Advances. 2012 2012. Rhodes, Greece:
International Society of Offshore and Polar Engineers, 360-363.
Williams, R. B., Grimsley, B. W., Inman, D. J. & Wilkie, W. K. Manufacturing and mechanics-based
characterization of macro fiber composite actuators. 2002 2002. American Society of Mechanical Engineers,
7989.
Witte, C. C., Ribeiro, D. M. & Grossman, B. Avoiding Disasters: Evolution in Integrity and Maintenance
Management. 2011 2011. American Society of Mechanical Engineers, 712.
Worden, K. 2001. Rayleigh and Lamb WavesBasic Principles. Strain, 37, 167172.
Yamaguchi, Y. & Terashima, S. Development of Guidelines on Corrosion Resistant Steels for Cargo Oil Tanks.
2011 2011. American Society of Mechanical Engineers, 333339.
Ye, X. W., Su, Y. H. & Han, J. P. 2014. Structural Health Monitoring of Civil Infrastructure Using Optical Fiber
Sensing Technology: A Comprehensive Review. The Scientific World Journal, 2014.
Yeo, F. & Fromme, P. Guided Ultrasonic Wave Inspection of Corrosion at Ship Hull Structures. American Institute
of Physics, 2006 2006. Brunswick, Maine.
Zhang, H. & Wu, Z. 2012. Performance Evaluation of PPP-BOTDA-Based Distributed Optical Fiber Sensors.
International Journal of Distributed Sensor Networks, 2012.
Zhao, P., Pisani, D. & Lynch, C. S. 2011. Piezoelectric strain sensor/actuator rosettes. Smart Materials and
Structures, 20.

710 SEPTEMBER 2015
CASCAIS, PORTUGAL
VOLUME 2
COMMITTEE V.8
RISERS AND PIPELINES

COMMITTEE MANDATE
Concern for the structural failure modes of risers and pipelines. Consideration shall be given to the dynamic
response of risers under environmental conditions as well as pipe-soil interaction. Aspects related to the
installation methods shall be considered. Attention is recommended for aspects related to maintenance,
inspection and repair, especially in deepwater conditions.
COMMITTEE MEMBERS
Chairman:
H. Suzuki, Japan (Chair)

S. Chai, Australia
Y. Chatzigeorgiou, Greece
H. Howells, UK
G. Kuiper, Netherlands
K. Kavanagh, Ireland
Y. M. Low, Singapore
C. Morooka, Brazil
S. Saevik, Norway
J. K. Seo, Korea
N. Sodahl, Norway
G. Stadie-Frohboes, Germany
L. Sun, China
J. Wu, USA
Y. Watanabe, Japan (Secretary)
KEYWORDS
Flexible risers, hybrid towers, pipeline, dynamic response, VIV, soil-pipeline interaction, failure mode,
installation, inspection and repair.
866
ISSC committee V.8: RISERS AND PIPELINES
CONTENTS
1.
INTRODUCTION .......................................................................................................................................... 867
2.
NEW DESIGN CONCEPTS ......................................................................................................................... 867

2.1 Latest design practice of flexible risers .............................................................................................. 867
2.1.1 Present application envelope ............................................................................................... 867
2.1.2 Deep water............................................................................................................................ 868
2.1.3 Shallow water ....................................................................................................................... 868
2.1.4 Singing risers ........................................................................................................................ 868
2.1.5 Hybrid towers ....................................................................................................................... 869
2.2 Latest design practice of pipeline ....................................................................................................... 870
3.
DYNAMIC RESPONSE INVESTIGATION REVIEW .............................................................................. 871

3.1 Riser .................................................................................................................................................... 871
3.1.1 Wave load induced dynamic response ................................................................................ 871
3.1.2 VIV ....................................................................................................................................... 873
3.2 Free span VIV of pipeline .................................................................................................................. 878
3.2.1 Assessment ........................................................................................................................... 878
3.2.2 Mitigation ............................................................................................................................. 879
4.
SOIL-PIPELINE INTERACTION ................................................................................................................ 879

4.1 Introduction ........................................................................................................................................ 879
4.2 Soil behavior near pipelines ............................................................................................................... 880
4.3 Pipeline as-laid embedment and riser touchdown ............................................................................. 880
4.4 Lateral pipe-soil interaction................................................................................................................ 881
4.5 Axial pipe-soil interaction .................................................................................................................. 882
4.6 Pipeline stability during trenching and backfilling ............................................................................ 882
4.7 Pipeline stability during sediment transport and liquefaction ........................................................... 883
5.
FAILURE MODES OF RISERS AND PIPELINES .................................................................................... 884

5.1 Steel riser and pipelines ...................................................................................................................... 884
5.1.1 Buckling (buckle propagation), collapse and fatigue failure .............................................. 884
5.1.2 Corrosion .............................................................................................................................. 885
5.1.3 Crack .................................................................................................................................... 886
5.1.4 Erosion ................................................................................................................................. 886
5.2 Flexible pipes ...................................................................................................................................... 886
5.2.1 Failure modes ....................................................................................................................... 886
5.2.2 Design analysis..................................................................................................................... 886
5.2.3 Monitoring............................................................................................................................ 887
6.
INSTALLATION ........................................................................................................................................... 888

6.1 Risers .................................................................................................................................................. 888
6.2 Pipelines .............................................................................................................................................. 888
7.
INSPECTION AND REPAIR ....................................................................................................................... 889

7.1 Risers .................................................................................................................................................. 889
7.2 Pipelines .............................................................................................................................................. 891
7.2.1 Maintenance ......................................................................................................................... 891
7.2.2 Inspection ............................................................................................................................. 891
7.2.3 Repair ................................................................................................................................... 892
8.
CONCLUSIONS ............................................................................................................................................ 893
REFERENCES ....................................................................................................................................................... 895
1.
867
INTRODUCTION
The immediately previous report regarding risers and pipelines is ISSC2000 Specialist Committee V.5
Structural Design of Pipeline, Riser and Subsea Systems. It represented the sequel to the report of
Committee V.7 Structural Design of Pipeline Systems of the 1997 1SSC. The mandate of ISSC2000
report is as follows.
Concern for the development of appropriate principles for life cycledesign of offshore pipeline,
riser and subsea systems designed for the conveyance of fluids, including load and dynamic response.
Consideration shall be given to inspection and monitoring procedures, and to the influence of operational
decisions on safety.
Both reports dealt with the issues related to the structural design of pipeline, riser and subsea systems
designed for the conveyance of fluids. There were two main factors driving pipeline and riser technologies
during the preparation of ISSC2000 report. The first was the requirement of longer but reliable cost-reduced
pipelines connecting wells and the production facilities. The second factor was the inexorable increase in
water depths between 1700m and 2800m associated with the exploitation of offshore oil and gas. The
contents of the report consist of load and load effect, capacity, installation, design criteria and inspection,
repair and maintenance. The topics and brief outputs of the report are summarized below:
Relatively matured subsea pipeline and riser technology with further work required for reeled steel
catenary risers, deep-water flexible pipes, steel tube umbilicals, composite tubes and large bore coiled
tubing for pipeline applications.
Importance of accurate prediction of hydrodynamic loads with the aid of rapidly advancing Computational Fluid Dynamics (CFD).
Attractive steel catenary risers for deep water applications with additional research for the in-service
fatigue life of the welded joints.
Maturity of global riser analysis except pipe-sea bed interaction.
Necessity of challenge toVoltex Induced Vibration (VIV) especially at high Reynolds numbers and in
deep water applications.
It has been fifteen years since the ISSC2000 report was published. However, the increase in water
depths is still continuing and is an important factor. This increase in water depths can be located in the
construction records of drilling ships. Associated with the exploitation of offshore oil and gas, Discoverer
Enterprise was constructed in 1999 whose maximum water depth is 3,048m and Discoverer Clear Leader
was constructed in 2009 whose maximum water depth is 3,658m. Furthermore, carbon fiber reinforced
plastic (CFRP) risers for 40005000 m water depths are under development in the field of scientific ocean
drilling to exploit the mantle (Watanabe et al. 2007). This ISSC2015 report is based on the research
literature of the last ten years and describes new design concepts, dynamic response with emphasis on
vortex-induced vibration, soil-pipe interaction, failure modes, installation and maintenance, inspection
and repair regarding risers and pipelines.
Technical developments in this broad technical area since ISSC2000 are focused and reported from the
research point of view. As a result, contents are more or less selective although the all technical fields
were tried to be covered equally as much as possible.
2.
NEW DESIGN CONCEPTS
2.1
Latest design practice of flexible risers
2.1.1
Present application envelope
The unbonded flexible riser is a key enabler for floating production of hydrocarbons thanks to its ability
to absorb floater motions as well as direct wave and current loading by geometrical deflection of the
configuration without compromising the structural integrity of the internal cross-sectional components.
More than 3500 flexible pipes have been deployed over the last 40 years. This includes about 1900
flexible risers with an accumulated field experience of about 22000 years. The vast majority of flexible
pipes are in the range of 412 inner diameter with a few low pressure pipes up to 20. Most pipes are
designed for a pressure of 30005000 psi, although some small diameter pipes have a design pressure up
to 15000 psi. The design pressure time inner diameter is widely referred to as the performance capacity of
flexible pipes. Current limit for the performance capacity is about 7000080000 psi-inch.
868
The operating temperature is below 60C for half of the pipes, 5% of the pipes operate in a temperature
range of 120130C. At present, flexible risers are operating down to a water depth of 1800 m. However,
roughly half of the flexible pipes are installed at a water depth less than 500m (Sparks 2003, Kenny 2010,
4subsea 2013).
2.1.2
Deep water
Weight reduction is a key design driver for deep-water application of flexible risers. This is twofold - to
reduce hang-off loads as well as to improve fatigue performance as friction stresses increase with the tension
in the pipe. New materials (e.g. carbon fiber) as an alternative to steel are introduced for tensile as well as
pressure armors. Special challenges are related to deepwater sour service application (e.g. pre-salt petroleum
production) limiting the use of high strength steel due to its limited ability to withstand the aggressive
annulus corrosion environment (Paumier & Mesnage 2011, Lambert et al. 2012). A reactive polymer layer
has been proposed (Epsztein et al. 2011) to eliminate H2S in annulus and thereby allowing usage of high
strength sweet service steel for applications with high H2S content.
Segmented flexible risers are proposed to optimize deepwater flexible riser configurations. The idea is to
apply different cross-section designs along the riser optimized for the main design driver at each water depth.
This involves use of lightweight crosssections with high axial capacity in the upper part combined with
crosssections with high collapse resistance for the lower part of the riser (Lambert et al. 2012, Rytter
et al. 2002, Paumier & Mesnage 2011).
Carbon fibre tensile armours (CFA) consisting of unidirectional carbon fibres embedded in an epoxy
matrix has been proposed for weight reduction as well as for improving fatigue performance as CFA is less
prone to an aggressive annulus environment. As the CFA concept has limitations with regard to the
compressive capacity, this can either be mitigated by riser segmentation (Lambert et al. 2012) or by
combining steel and CFA by means of new profile shapes (Rytter et al. 2002) .
For deepwater risers, concerns are the dynamic axial stresses and fatigue resulting from the combined
action of floater motion and riser mass. Since these stresses are transferred directly into the end-fitting, stress
concentrations inside standard design end-fitting designs may cause fatigue failure (De Sousa et al. 2013). A
new end fitting design has been proposed to solve this problem (Campello et al. 2012).
Another important effect is the torsion buckling phenomena that may occur due to the combined action of
reversed end cap and cyclic loading either during installation or operation. This has traditionally been
handled by DIP testing (Secher et al. 2011). However, a way forward with regard to faster technology
qualification would be development of simplified material models that can be integrated into the global
dynamic analysis. These models need to be capable of describing the coupled axial and torsion cyclic
responses and may be obtained from either structure testing (stergaard et al. 2011) or numerical analysis
(Svik 2014).
2.1.3
Shallow water
Design of flexible riser configurations for shallow water in combination with harsh environment is
challenging. This is because it is difficult to obtain the required capability to absorb floater motions using
traditional flexible riser configurations such as lazy wave and pliant wave. Another severe shallow water
constraint is to avoid collision with adjacent structures. Use of multiwave configurations in combinations
with clump weights and tethers has been proposed to achieve required static and dynamic behavior for
shallow water applications, e.g. (Hanonge & Luppi 2010).
2.1.4
Singing risers
Singing risers resulting from high frequent vortices created in the corrugated profiles of flexible pipes during
gas flow have been in focus in recent years. The problem is attributed to flow-induced pressure pulsations
from the flexible risers shackle-type carcass, which has a corrugated profile. When gas passes through the
flexible riser, vortex shedding occurs at each of the internal corrugations, generating pressure pulsations. The
pulsations are generated due to vortex shedding and shear layer instabilities at the corrugations. If a coupling
occurs with an acoustic field, a feedback mechanism occurs where the acoustical field is amplified by the
shear layer instabilities and the acoustical field itself magnifies the layer instabilities (Kristiansen & Wiik
2007). The result is a high amplitude tonal noise, which is called the singing behaviour of corrugated risers.
No direct coupling to a mechanical eigenmode of the carcass is required to start this singing, although the
acoustic-mechanic coupling can excite mechanical resonance modes. Due to the generated pulsations
mechanical vibrations occur on the associated topside and subsea piping. The resulting piping vibrations
could lead to fatigue failure, particularly in welded connections to the main piping. For at least one platform,
869
this resulted in leakages in small bore side branches. Significant efforts have been made by the industry to
mitigate the problem by means of new pipe concepts including new carcass designs (Belfroid et al. 2011).
2.1.5
Hybrid towers
(1) Overall layout of hybrid towers

Free standing hybrid riser technology has been introduced as a generic riser solution for development of
deepwater hydrocarbon recourses (Fernandes 2003, Alliot & Legras 2005, Auperin et al. 2005, Pereira
et al. 2005, Pereira, Morooka et al. 2006, Mello, Lacerda et al. 2011, Gueveneux & Le Buhan 2014,
Luppi et al. 2014). The main components of hybrid riser systems are:
A lower vertical riser section tensioned by a subsurface buoyancy module at upper end.
Flexible jumpers to connect the upper end of the vertical riser section to the floater.
The main idea behind this solution is to obtain a cost-effective generic deepwater riser solution to the
extent possible built on qualified riser technology. The lower vertical riser section has close similarities to
conventional tension leg platform/spar riser solutions. The vertical section spans the main part of the
water column. This allows for application of conventional flexible riser technology for the jumpers. The
jumpers provide the required flexibility to absorb floater motions due to combined action of wind, waves
and current. This arrangement is introduced to de-couple to the extent possible the floater motions from
the vertical riser section.
The main drivers behind the development of hybrid riser systems system for deepwater floating
production are (Luppi et al. 2014) related to small hang-off loads, riser disconnection in harsh
environments, flow-assurance by insulation/active heating/gas lift in the vertcal riser section, small
footprint, limited seabed interaction and schedule flexibility (tower pre-installation).
Free standing hybrid risers are commonly divided into two main categories depending on the crosssectional layout of the vertical riser section:
Bundle Hybrid Risers (BHR) where the crosssection consists of multiple riser lines bundeled together
in an integrated complex crosssection. Each riser line may be designed as single riser pipe or concentric
pipe in pipe. Service lines and umbilicals may in addition be included in the crosssection. All lines may
be enclosed in common insulation material to provide flow assurance and buoyancy during both tow-out
and operating phases.
Single Hybrid Risers (SHR) where the crosssection consists of a single riser line. The riser line may be
designed as single riser pipe or concentric pipe in pipe. In this case, the carrier pipe is the riser pipe itself.
(2) Buoyancy tank and interfaces
The buoyancy tank is a vital component of the hybrid riser system. The design is governed by redundant
buoyancy and riser strength requirements which means that one or several redundant compartments are
water-filled during normal operation. The redundant compartments allow for failure of one or several
compartments due to e.g. dropped objects without compromising the overall riser system integrity.
(3) Seafloor interface
The main requirements to the seafloor termination of the vertical riser are to provide angular flexiblity,
vertical anchoring and minimize the load transferred to the seafloor flow-lines. The load carrying element
(carrier pipe) of the vertical riser is anchored to the foundation by means of a flex-joint, taper-joint or
articulated pin-joint to minimize the bending of the vertical riser. The other riser lines are connected to
the on-bottom flow-lines by spools. The required holding capacity of the foundation is achieved by means
of piles, gravity base or a suction anchor. Combined pile/gravity foundation is also applicable.
(4) Installation
Installation is a crucial issue for cost-effective application of hybrid riser technology. The most
challenging part is installation of the vertical riser and buoyancy tank. Possible installation scenarios
proposed are offshore assembly of riser joints using offshore welding or threaded connectors and onshore
assembly of the complete column and buoyancy tank. The entire structure is subsequently transported by
tow-out to the field and installed by upending.
Offshore assembly is only applicable to SHR systems while tow-out/up-ending primarily is intended
for installation of BHR systems but also applicable to SHR systems. Tow-out and up-ending has been
introduced as a cost-effective installation method for BHR system. This allows for onshore fabrication
and assembly of the riser column which is a great advantage for complex bundle cross-sections. The
870
entire riser column is subsequently transported to the field by tow-out. The buoyancy tank arrangement
may be installed on the riser column onshore prior to tow-out or connected offshore prior to up-ending.
Installation by tow-out requires that the riser column is nearly neutrally buoyant.
2.2
Latest design practice of pipeline
The offshore and subsea industries recently experienced a technical revolution in the design process.
Advanced methods and analysis tools allow a more sophisticated approach to design, which takes advantage
of modern materials and revised design codes supporting the limit state design concepts and reliability
methods. The new approach is called design through analysis (STA), where the finite element method
(FEM) is used to simulate the global behavior of pipelines as well as the local structural strength. The twostep process is used in a complementary way to determine the governing limit states and optimize a
particular design.
A pipeline is a primarily horizontal pipe lying on, near or beneath the seabed, normally used for the
transportation of hydrocarbon products between offshore production facilities or between a platform and a
shore facility (ABS 2006). Subsea pipelines are used for a number of purposes in the development of subsea
hydrocarbon resources. A pipeline system can be a single-pipe, pipe-in-pipe, or bundled system. The design
process for each type of line in general terms is the same, and it is this general design approach and new
design concepts that are discussed in this chapter.
The general design of pipeline is performed in three stages: conceptual design, front end engineering
design (FEED), and detail design. The objective and scope of each of these design stages varies, depending
on the operator and the size of the project. The design approach for a pipeline is to determine, based on given
operating parameters, the optimum pipeline size parameters. These parameters include pipeline internal
diameter, pipeline wall thickness, and grade of pipeline material, type of coating-corrosion and weight, and
coating wall thickness.
Three factors have a major influence on the final compressive strength of the pipeline: quality of plate
feedstock, optimization of compression and expansion during pipe forming, and light heat treatment. By
focusing on these factors together with improving the ovality of the final pipe, it is possible to obtain a
collapse resistance comparable to that of seamless pipes.
A novel pipeline concept developed by Venas (2012), X-Stream, aims to solve the collapse challenge by
limiting and controlling the external over-pressure. In a typical scenario, the pipeline is installed partially
water-filled, thus becoming pressurized at large water depths. Then, to ensure that the internal pressure does
not drop below a certain limit during the operational phase when it is filled with gas, it is equipped with a socalled inverse high integrity pressure protection systems (HIPPS). Studies undertaken during the
development of X-Stream show that the weight increase due to flooding is more or less balanced by the
reduction in steel weight. Some practical aspects need to be studied, such as how to install large valves in
ultra-deepwater. Another aspect includes repair procedures and equipment, even though that should not be
much different from normal ultra deepwater pipelines. There are also some optimizations to be performed
with respect to pressure loss during operation and equalization of the pressure during shutdown. However,
the potential benefits of the X-Stream concept to gas export and trunk lines at ultra deepwaters are quite
significant, such as:
Reduced steel quantity and associated costs.

Use of standard pipe dimensions, even for ultra deepwater and large diameters, reduces line pipe costs.
No need for buckle arrestors and for reserve tension capacity in case of accidental flooding.
When a pipeline is subject to high pressure and high temperature, its ends expanded longitudinally and
exerts large forces and bending moments onto adjacent tied-in structures connected to it. The tied-in
structures must be designed to withstand these expansions and forces. Dumping rocks along the pipeline or a
giant spool installed at the pipeline end have traditionally been costly alternatives.
SliPIPE is a new concept developed to deal with pipeline expansion which works to reduce the end force
expansion exerted at the tie-in by absorbing the end expansion through sliding within itself and
simultaneously reducing or eliminating the effective axial compressive force in the pipeline (Yew 2013).
The concept consists of an outer pipe connected alongside a pressure chamber and an inner pipe that can
slide inside them. Seals are placed at the contacts between the pressure chamber and the inner pipe. The
inner pipe slides in or out of the outer pipes in response to an axial stress that can either be more or less than
a certain value. This value is predetermined in the design and causes an axial tension in the pipe wall to
develop, which opposes the effective axial compressive force component arising from the inner fluid
871
pressure. The axial tensile pipe wall force is produced by letting fluid pressure in. Between the outer
pipe/pressure chamber and the inner pipe of the SliPIPE concept are two main seals, a partition wall seal, an
environmental seal and a scraper seal. The seals are made of materials that allow them to function at high
temperatures and pressures.
New concepts for submarine pipelines and risers have been proposed recently in order to achieve flow
assurance in deepwater environments. It is the case of both pipe-in-pipe (PIP) and sandwich pipe (SP). The
PIP product consists of the production pipeline being sleeved into an outer pipe with the annulus being
maintained dry and filled with a high performance insulation material configured to meet the particular
project thermal requirements. The outer pipe is designed to withstand both the hydrostatic pressure dictated
by the project water depth and the installation methodology. The inner pipe can be located within the outer
pipe by the use of centralizers clamped at discrete intervals along the inner pipeline, or as for high
performance PIP system, without centralizers. The inner and outer pipe sizings are designed as a single
system which is to be installed by the preferred installation method for each projectreeling, towing, S-lay or
J-lay. The selection of the appropriate installation method depends on the characteristics of the field
development and the dimensions of the PIP systems.
In the case of sandwich pipe (Castello & Estefen 2008), the annular layer characteristics differ from PIP
by satisfying simultaneously mechanical and thermal requirements. Therefore, greater structural strength
combined with adequate flow assurance can be obtained. Sandwich structures are a particular kind of
composite characterized by the combination of different materials bonded together, contributing with their
single properties to the global structural performance. Usually, the sandwich structure is divided in three
layers: two external thin and stiff, and a central thick flexible core. The external layers are bonded to the core
to allow the load transfer between the components. Numerical and experimental studies have been carried
out to obtain data about the mechanical behavior of this kind of structure not very well understood so far, as
done by Borselino et al. (2004) and Sokolinsky et al. (2002). Sandwich structures, i.e. light and stiff panels,
have been employed in the naval industry mainly, searching the advantages associated with weight
reduction, fuel economy, stability during navigation and corrosion resistance, as mentioned by Mouring
(1999). It should be considered that various application of a sandwich pipes with experimental tests and
numerical models for verifying the influence of the inter-layer adhesion on the ultimate strength under
external pressure and longitudinal bending of new sandwich pipes.
3.
DYNAMIC RESPONSE INVESTIGATION REVIEW
3.1
Riser
3.1.1
Wave load induced dynamic response
Dynamic response of risers induced by wave loads is a relatively mature topic, thus this section focusses on
wave loads in conjunction with other considerations.
The simultaneous effect of local wind and swell conditions result in bimodal seas, which are observed in
many locations such as Brazil and West Africa. Bimodal sea states lead to bimodal riser responses, which
complicate the analysis since the regular wave approach cannot be directly applied. Tan et al. (2012)
compared three methods for fatigue analysis of flexible risers subjected to bimodal seas: (1) irregular wave
analysis using bimodal wave spectrum; (2) treating the swell and wind seas as separate cases and summing
the damage; (3) regular wave analysis, using some equivalent wave height and period. It is found that the
method (1) underestimates the fatigue damage compared to the method (2), while the method (3) over
predicts the damage. Francis (2011) compared seven different methods for combining bimodal wave fatigue
damage, as outlined in DNV-RP-F204 and RealLife JIP. These are the main conclusions: (i) The direct
summation of fatigue damage due to swell and wind seas is generally more accurate than the DNV
simplified method: (ii) Regular wave analysis is not always conservative compared to irregular wave
analysis.
Using a unidirectional wave spectrum generally leads to overestimation of the mooring and riser dynamic
response. Thus, proper consideration of the wave directionality and spreading is necessary for accurate
assessment of the system response. Caire & Schiller (2013) investigated the effect of wave spreading on the
mooring and riser systems of a spread-moored floating production, storage and offloading (FPSO) facility.
The wave environment is represented as the sum of a unidirectional swell wave and multidirectional windgenerated waves with a cos-2s spreading function (s is the spreading parameter). The case studies suggested
872
that first, the top tension of moorings are more influenced by the wave spreading compared to risers; and
second, less spreading results in larger top tensions for both mooring and risers.
To ensure safe design of riser systems, it is crucial to include all relevant dynamic loads including slug
flow, which can be combined with wave loads to exert a significant effect on the riser response. International
standards e.g. ISO 13628-2 recommend that slug flow should be considered but provide no detailed
guidance on how to incorporate slugging effects into riser analysis. Gundersen et al. (2012) developed a
method to determine the remaining fatigue life of a riser subjected to combined excitation of slow flow,
vessel motions and irregular wave loads. The slug loading parameters (i.e. density, length, velocity and
period of slug) were obtained iteratively by calibrating riser responses with field observations. This study
showed that slugging can reduce the riser fatigue life significantly (~50%). Ortega et al. (2013) investigated
the effect of irregular slug flow and regular wave loads on riser dynamic response by coupling two in-house
codes. One code is responsible for the global dynamic analysis, while the other code simulates the internal
slug flow using a finite volume method. Information is exchanged between the two coupled programs, which
are run simultaneously.
Predicting the extreme response and fatigue life of risers under wave loads is of significant practical
interest. These tasks are also academically challenging. Due to the irregular nature of waves, probabilistic
approaches are required to evaluate the dynamic response, and the stochastic simulations can be
computationally demanding. Another complicated aspect is the non-linear and non-Gaussian characteristics
of the wave loads and dynamic response. In addition, the long term variation of the sea state should be taken
into consideration. Since the same scatter diagram is often used for both the fatigue and ultimate limit states
of risers, an important question is whether there are overlapping sea states when evaluating the long-term
extreme response and fatigue damage. In a case study by Martins et al. (2009), it is shown that there is no
overlap between important regions for fatigue and ultimate limit states. The ultimate limit state is governed
by extreme sea states, while the fatigue limit state involves moderate sea states.
Focusing on extreme response prediction, Passano & Larsen (2007) proposed an efficient method for
predicting the short-term extreme response of a steel catenary riser (SCR) at the touchdown area. This
method is based on using a small number of simulations to identify clear trends between the prescribed axial
velocity at the top end and the riser responses (effective tension and bending moment) near the touchdown
point. Time histories for the axial velocity at the top can be generated very quickly using transfer functions
from the wave spectrum in conjunction with the fast fourier transform (FFT) technique. The effectiveness of
this approach is demonstrated by comparison with longer simulations. In a sequel, Passano & Maincon
(2011) proposed a fast approach for estimating the long-term extreme response of an SCR. This approach
preselects the relevant sea states (23 in this study) that contribute most to the long-term extreme response.
The approach compared favorably to analysis of the full scatter diagram. In addition, for each sea state
considered, the relevant time intervals are identified, allowing for the simulation of relatively short time
series to explore low probability events. The extreme response can also be estimated using a short-term
approach by searching the 100-year HS Tp contour for the most onerous response. Irregular time domain
analyses are performed for discrete HS Tp combinations along the contour line. This is the basis for
establishing the extreme response distribution for a given duration, typically three hours. The extreme load
effect is estimated as a percentile in the extreme value distribution in each sea state. Calibration against longterm response extreme value estimation suggests that a 90% percentile is applicable for flexible risers in
harsh environment (Baarholm & Haver 2010). The appropriate percentile level will however depend on the
riser system as well as geographical location and should be established by calibration against long-term
response statistics for applications with no prior experience.
For long-term fatigue assessment under wave loads, one of the main challenges is the high computational
cost of simulating all the sea states in the scatter diagram. A common approach used in industry, and is also
recommended in design codes such as DNV-RP-F204, is to lump several sea states into a manageable
number of blocks. The problem is that this relatively crude approach inevitably leads to errors, and there are
no guidelines on effective blocking strategies. To reduce the computational cost, Sheehan et al. (2005)
proposed to replace irregular wave analysis with regular waves of different wave heights and periods. The
Longuet-Higgins distribution is applied to extract the regular waves from the stochastic wave scatter
diagrams. Since the Longuet-Higgins distribution tends to overestimate the probability of waves occurring at
long periods, waves with periods exceeding a cut-off value are disregarded. To avoid the need to select a cutoff frequency, De Sousa et al. (2013) suggested to select the regular waves by generating time series of the
wave elevation from the wave spectra. Low & Cheung (2012) developed an asymptotic approximation
approach for fast evaluation of the long-term fatigue damage. The asymptotic approximation is based on a
second-order Taylor expansion of the logarithm of the integrand about the maximum point. The integrand
873
refers to the product of the fatigue damage response function and the joint probability density of the
significant wave height and peak period. The asymptotic approximation is found to yield reasonable
estimates of the long-term fatigue damage compared to the consideration of the full scatter diagram.
Perdrizet & Averbuch (2008) outlined an efficient method to evaluate the nonlinear extreme response of a
riser. The first order reliability method (FORM) is used to solve the short-term extreme, while the response
surface method (RSM) is used to obtain the long-term extreme response.
The excitation force from irregular waves, in conjunction with vortex-induced vibrations and wakeinduced oscillations, may cause collisions between risers in close proximity, resulting in damage. He & Low
(2010) developed an approach for predicting the probability of riser collision, taking into consideration the
uncertainty from irregular waves. The approach is based on the crossing rates of a parameter that
characterizes the critical gap between two risers. In a sequel, He & Low (2013) extended this approach to
consider various parametric uncertainties, in particular the current, drag coefficient, vessel RAO and riser
mass.
3.1.2
VIV
Hydrodynamic current vortex-induced vibration (VIV) among other environmental loads, such as waves and
those coming from the motions of the floating production unit dynamics are the other constraints that limit
viability of risers and systems. (Chaplin et al. 2005)
In this scenario, deepwater risers and novel systems have been proposed and applied last years. However,
many uncertainties still remain on the design of those systems, particularly related to the VIV. Procedures
established by classification societies and commercial software to estimate service life due to VIV are
available (Vandiver & Li 1999, Larsen et al. 2000). Those procedures are developed, in general, based on a
specific background scenario of vertical pipes or horizontal ones, although they take data from real scale
measurements as the basis. Otherwise, design simulations are based on restricted data from the VIV
phenomena observed in laboratory scale.
VIV is a complicated structure-fluid coupled response. In the early days of the study, understanding of the
phenomenon and derivation of non-dimensional parameter were investigated based on simple model, for
example spring supported cylinder in a uniform flow. In parallel to the VIV study, efforts were made to
develop dynamic analysis codes of a riser under current and wave load mainly for development and design
purpose of deepwater riser and SCR.
Development of VIV analysis code has soon started and some codes which use the empirical VIV force
model have been developed. Now they are the standard analysis codes used in industry, but improvement
and development are continuing.
The study today might be categorized into two categories a) one is research into physical phenomena
oriented study, more or less scientific, and b) another is a design oriented study. Design of riser which goes
into deeper water presents pressing design issues and efforts to develop more advanced analysis codes are
demanded.
(1) Physical phenomenon oriented scientific research
Most of the VIV research is related to fundamental studies. A large number of numerical studies are
available in the literature, sometimes with analytical evaluations as can be observed in Williamson &
Govardhan (2008). Those investigations are oriented to understand the physical phenomenon which the VIV
involves, and they are conducted by laboratory experiments and/or numerical simulations by use of the
computational fluid dynamics (CFD). Computational work can be based on commercial computer codes, and
sometimes is based on in-house developing programs. Navier Stokes equations are used to model the flow
around long pipe, and in general, viscous effects are represented by turbulence models by Reynold Averaged
Numerical Simulation (RANS), Large Eddy Simulation (LES) which takes into account the turbulence scale,
and Direct Numerical Simulations (DNS) in which a fine mesh is usually required.
Experiments in a water channel facility were conducted with stationary and yawed cylinders to investigate
effects of the inclination in the force coefficients and in the vortex-shedding frequency. Force coefficients
and Strouhal number were evaluated by assuming the Independence Principle by Franzini et al. (2014).
Irregular vortex shedding regime was observed when the cylinder is inclined with 45 degrees. And, they
observed lift force spectra as well as the mean drag coefficient different when the cylinder is yawed with
upstream or downstream orientation, respectively, due to asymmetrical end conditions performed by the
cylinder. Gonalves et al. (2013) investigated the VIV on a very short length cylinders with two degrees of
freedom. Experiments were conducted with Reynolds range from 6000 up to 70000. Three small mass ratios
(m*= 1.00, 2.62 and 4.36) and small aspect ratios (0.3<L/D<2.0) were considered. The author reported the
874
maximum amplitude of motion in cross-flow direction (A/D=1.5) and the response amplitude for in-line
direction was A/D=0.4.
Sanaati & Kato (2013) conducted an experimental study to investigate the effects of pre-tension and axial
stiffness on the VIV of a horizontally mounted flexible cylinder. Tests covered a subcritical Reynolds
number range from 1000 up to 16,000. Results showed that high pre-tension can raise the lift coefficient.
Also observed was that the lock-in bandwidth of amplitude response narrowed with increase in pre-tension.
They also concluded that the higher the applied tensions are, the lower the vibration amplitude will be, and it
could significantly raise the hydrodynamic lift force coefficient but higher applied tensions generate
narrower lock-in bandwidths. In a further research, Sanaati & Kato (2014) investigated effects of proximity
interference on the hydro-elastic responses of two pre-tensioned flexible cylinders in side-by-side vertical
arrangement. They observed for larger separation distances between cylinders that no synchronization is
observed for very small amplitudes of vibration, and the mean drag and fluctuating components of the drag
and lift forces of both cylinders showed quite different behaviors.
In a series of publications, R. Bourguet et al. (2013a, 2013b, 2012) uses DNS to simulate and analyze VIV
of a long flexible tensioned cylinder in shear flow for a Reynolds numbers ranging from 110 to 1100. The
cylinder is allowed to vibrate in cross-flow (CF) and in line (IL) directions, and transition to turbulence in
the wake is taken. Lock-in drivers to the VIV of the tensioned cylinder are investigated. Standing and
travelling waves pattern in both direction as well as synchronization of the vortex shedding with the crossflow vibration in the high oncoming velocity region, over at least 30% of the cylinder length were observed.
On the other hand, in multi-frequency response, synchronization of vortex shedding exhibits in all vibration
frequencies is observed. Finally, different behavior for spanwise patterns of the forces and added mass
coefficients are observed within the lock-in versus the non-lock-in region.
An innovative approach for the VIV calculated from numerical simulation of the flow around a solid
cylinder with movement was shown in Hirabayashi & Suzuki (2013). Numerical simulation is shown for a
cylinder with movement by the Lattice Boltzmann method, and a good result was obtained for the drag force
in comparison with the literature. In a further study, Miyamura et al. (2014) obtained good result for a
surface piercing floating structure when the same procedure was applied and movement of the free surface
was introduced to approach to the vortex-induced motion. It was concluded that the lattice Boltzman method
can also represent the motion of water surface motion accurately than conventional methods. The new
approach demonstrates good potential for applying to VIV in long free surface cylinders such as marine
risers.
Based on a series of previous studies, Pontaza & Menon (2013) predicted VIV response of a pipeline
span. A Fluid Structure Interaction (FSI) model is described by coupling a three-dimensional viscous
incompressible NavierStokes solver with a beam finite element solver in time domain. EulerBernoulli
beam element and instantaneous flow-induced forces are taken and solution is reached through a finite
element basis functions in space and an unconditionally stable Newmark-type discretization scheme in time
domain.
A selective review of recent research on vortex-induced vibrations of isolated circular cylinders and the
flow and vibration of circular cylinders in a tandem arrangement has been presented by Bearman (2011).The
influence of Reynolds number on the response of isolated cylinders is presented and recent development
using forced vibration is discussed. Huang & Larsen (2011) made a numerical simulation on vortex-induced
vibration of an elastically mounted circular cylinder with two degrees-of-freedom Two-dimensional
numerical simulation using incompressible flow with RANS was carried out to obtain forces and responses
associated with vortex induced vibration of an elastically mounted circular cylinder with two degrees-offreedom. Vandiver (2012) defined a dimensionless damping parameter, c* = 2 / U 2 , for cylinders
experiencing flow-induced vibrations and characterized VIV for all reduced velocities in the lock-in region.
Generalized the c * parameter to characterize the response of flexible cylinder in sheared and uniform flow.
Data base available from three independent outsources are used to illustrate the applications of the c *
parameter.
The hysteresis effect on the VIV on an elastically mounted rigid circular cylinder has been investigated by
Wanderley et al. (2012). Numerical solution of the two-dimensional Reynolds averaged Navier-Stokes
equation was obtained for this purpose.
Numerical investigation of two-degree-of-freedom VIV of a circular cylinder in oscillatory flow was
shown by Zhao et al. (2013). RANS equations were used and simulations are carried out for the Keulegan
Carpenter (KC) numbers of 10, 20 and 40 and reduced velocities up to 30 with the Reynolds number ranging
from 308 to 9240. From results, it was observed that the amplitude of the primary frequency is larger than
875
other frequency components at KC=10. For KC=20 and 40 most vibrations presented multiple frequencies.
Maximum amplitude occurs for reduced velocities between in-phase regime and anti-phase regime, in both
the x- and the y-directions. They also combined steady and oscillatory flows, respectively, to simulate by
RANS the VIV in a circular cylinder with constant KC=10 and flow ratio ranging 0 to 1. In a combined flow
domain, they observed a wider lock-in region compared with that one for a pure oscillatory flow. At the flow
ratio equal to 0.8 the maximum amplitude in the cross-flow direction was reached (A/D=1.5), and the widest
lock-in regime occurs at flow ratio ranging 0.4 to 0.6, about twice in the pure steady or pure oscillatory flow
case.
Kang et al (2013) compared experiments, one with two cylinders in tandem arrangement, and the other
with a single cylinder, both exposed to an incident fluid flow. Cylinders have low mass and damping ratio.
Wake induced vibration (WIV) is found, and it was observed in the near wake interference, both are exposed
to VIV, but the amplitudes of upstream and downstream cylinders, respectively, both are less than that
observed for the single cylinder, for cross-flow and in-line directions.
Fu et al. (2013) verified VIV in a flexible cylinder through tests performed with a harmonically forced
oscillation, for combinations of amplitude and period. VIV responses in CF are investigated using modal
decomposition and wavelet transformation. The results show that no VIV oscillatory flow is quite different
from the steady flow. New features denominated as intermittent VIV, amplitude modulation and the mode
transition was observed. Development process of the VIV, including its Building-up, Lock-in, and Dieout in oscillatory flow are defined and analyzed.
Carmo et al. (2013) performed numerical simulations in two and three-dimensional flow, around two
circular cylinders with tandem arrangements. The Reynolds number was varied from 100 to 645, and results
confirmed that the presence of the wake upstream of the cylinder lead to higher amplitudes of vibration
when compared to VIV.
Fundamental experimental research (Assi 2012) for curved cylinder in convex and concave configurations
to investigate the two-degrees-of-freedom VIV response with low mass-damping ratio was conducted.
Experiment Reynolds number ranged from 750 up to 15,000. Results for those curved cylinders showed
lower vibration amplitudes when compared with a straight one. On the other hand, concave cylinder
configuration showed higher amplitudes of vibration than the convex one. Assi et al. (2013) investigated by
experiment, into pair of cylinders in tandem, how the cylinder responds to excitation caused by wakeinduced vibrations (WIV) due to the interaction with vortices coming from the upstream cylinder,
characterizing the amplitude and frequency of response.
Srinil et al. (2013) presented an experimental and numerical investigation of a two-degree-of-freedom
VIV of a flexibly mounted circular cylinder with variable in-line-to-cross-flow natural frequency ratio. Tests
were carried out in a towing tank and subjected to a uniform steady flow in a subcritical Reynolds number
range from 2,000 to 50,000. The numerical prediction was based on Duffin-Van der Pol (structure-wake)
oscillators. The maximum achieved cross-flow and in-line amplitudes was 1.251.6 and 0.50.7,
respectively, depending on the level and combination of the x-y structural damping ratios.
Rahmanian et al. (2012) investigated VIV of two side-by-side cylinders of different diameters in steady
flow with the diameter ratio of cylinders of 0.1, and Reynolds number of 5000 considering the larger
cylinder diameter. Two dimensional Reynolds-averaged Navier Stokes equations are solved by FEM, by
using the Arbitrary Lagrangian Eulerian scheme with a SST k- turbulence model. Numerical method was
validated with experimental results. It was verified that collision of cylinders is dependent on the difference
of the natural frequencies of each cylinder.
(2) Riser VIV and computer code
Laboratory experiments as well as in situ monitoring of risers are essential to develop and to have a reliable
riser model for designing purpose, as well as, for understanding of the riser VIV phenomena in sufficient
manner to build a calibrated theoretical and numerical model. In this manner, few works have been
published in the literature (Vandiver & Li 1999, Larsen et al. 2000, Nozawa et al. 2010). Recently, great
effort is being conducted to describe as much as possible the VIV in risers, to develop reliable computational
tools for designing riser for the VIV. They ally theoretical analysis, reduced scale laboratory model tests,
development of numerical procedures and computational fluid dynamics (CFD).
A truncated steel catenary riser (SCR) model was experimentally tested in the ocean basin (Wang et al.
2014) by oscillating the top end of the model to simulate the heave and surge vessel motion in order to
investigate the vortex-induced vibration (VIV) features. Out-of-plane VIV responses were generally
analyzed revealing that although the root mean square (RMS) strain distributed rather a broad band, the
displacement was quite consistent within a narrowband from 0.2D to 0.3D, and the touch down point (TDP)
876
area was found not to be the place suffering the maximum out-of-plane VIV response due to near wall
effects.
CFD simulations of VIVs and wake-induced vibrations (WIVs) for two vertical risers in tandem and sideby-side arrangements were performed in Chen et al. (2013). Both risers have the same outer diameter of
0.016 m, and the same total length of 1.5 m (L/D = 93.75). Numerical simulations used the unsteady RANS
method. The riser inline and cross-flow motion responses were calculated using a tensioned beam motion
equation.
A numerical simulation procedure based on a semi-empirical VIV model to predict dynamic response of a
pipeline with free span and risers was introduced by Tsukada & Morooka (2013). Computations were
carried out in time domain using finite element method, and the VIV forces are calculated based on
hydrodynamic coefficients such as the added mass, lift and drag coefficients. Furthermore, the laboratory
experiment in a towing tank with reduced scale model of a steel catenary riser (SCR) was described in
Morooka and Tsukada (2013). The main objective of the experiment was to obtain an improved
understanding of the overall dynamic behavior due to VIV in a catenary-shaped riser. A large scale model
(1:250) was considered, with very low mass and stiffness for the riser model with the Reynolds number
ranging from 400 to 600.
Pereira et al. (2013) conducted a reduced scale model experiment of a catenary riser, and the vortex selfinduced vibrations (VSIVs) were observed experimentally. The study verified the nonlinear dynamic
behavior of risers through experimentally validated analytical and numerical models.
Cheng & Vandiver (2012) presented a theoretical formulation of the dynamic stiffness method and
combined the dynamic stiffness method with the WKB theory to investigate riser dynamic analysis.
Resvanis et al. (2012) explores the dependence of the Reynolds number on VIV of flexible risers.
Emphasis is placed on trends existing between the Strouhal number and the Reynolds number, and between
the dimensionless amplitude (A/D) and Reynolds number. Data are derived from recent experiments
conducted in a towing tank that used flexible cylinders of three different diameters, in sheared and uniform
currents.
Shi et al. (2012) examined the characteristics of the VIV response of a long flexible cylinder (model riser)
placed in uniform or sheared currents by employing real scale measurement of strains and accelerations
available from the Norwegian Deepwater Programme (NDP) experiments on an instrumented riser. The
measured data was used to describe the complexities inherent in riser motions accompanying VIV. They
discussed how such data sets can be used effectively in predicting fatigue damage rates.
(3) Suppression devices and fatigue
When it is not possible to avoid the VIV phenomena, suppression devices are needed at least partially along
the all riser length. Many kinds of VIV suppressors have been reported in the literature discussing the
effectiveness of each solution presented.
Allen & Liaps (2014) combined results from two testing programs on long cylinders towed at high
Reynolds numbers to assess the performance of helical strakes with differing conditions along the cylinder
length. The results show that the coverage length, density, and location of the helical strakes have a
substantial effect on both the local and global response of the tubular.
Tests conducted at the MARINTEK basin by Shell Company is described by Resvanis et al. (2014). The
tests involved towing densely instrumented flexible cylinders at Reynolds numbers up to 220,000, and
experimental results were described to show the effectiveness of different amounts of strake coverage and to
explore the influence of simulated marine growth. All results are compared with the bare cylinder cases
which will be used as a reference to determine how effective the strakes are in suppressing VIV and how this
effectiveness can be affected by marine growth.
Rao et al. (2013) explored the competition of the VIV excitation in smooth and segments with fairings in a
flexible pipe. An eight meters long tube model (divided into three sections) was used by evaluating
amplitude and frequency of the response, in Reynolds number around 37000. In the study, the flexible tube
was divided into three sections with and without fairings, to understand what portion of fairing or strake
coverage may be lost or damaged, before the operator takes corrective measures. Furthermore, in Rao et al.
(2014), a novel method to identify the power-in regions of long flexible cylinders subjected to VIV is
presented. For pipes with partial coverage of suppression devices in uniform flow, the bare region is
expected to be a power-in region, and the section with suppression devices, a power-out one. Experiments
are used to benchmark the proposed power-in zone identification method, and the method is used to identify
the power-in zones on a bare cylinder in a sheared flow. The occurrence of secondary power-in regions that
may exist is explored when suppression devices are placed in the primary power-in zone.
877
Experiments were carried out in Cicolin (2014), with rigid circular cylinders fitted with three different
types of permeable meshes, to investigate their effectiveness in the VIV suppression. Measurements of the
dynamic response are presented for cylinders with low mass and damping which are free to respond in the
cross-flow direction. Results for two meshes made of ropes and cylindrical tubes are compared with the VIV
response of a bare cylinder and that of a known suppressor called the ventilated trousers (VTs).
A drilling riser VIV test has been conducted by Lie et al. (2013) to verify the possible VIV suppression to
improve operability of retrievable riser systems with auxiliary lines by adding riser fins. The riser model was
elastically mounted and towed over a reduced velocity range around 410 in two different Reynolds ranges
75.000 to 192 000 (subcritical regime), and 347.000 to 553.000 (critical regime). The results showed, in
general, the slick joint riser configuration with kill and choke lines increased the displacements compared to
displacements of the bare riser model.
King et al. (2013) described a recent large-scale experimental test conducted with a new VIV suppression
device for cylindrical structures exposed to external fluid flow. The suppressor VT is a loose sleeve in the
form of a flexible liquid light with full bobbins in a particular arrangement. It is unidirectional, robust and
made with material compatible with the offshore environment.
An extensive program to study hydrodynamic models of a riser related to VIV from the database obtained
by the Winter test of the Shell Oil Company was conducted (Lie et al. 2012). A new test rig was built and
three different riser model prototypes, each 38 meters long, were towed horizontally at different advancing
speeds, simulating uniform and linear varying current sliced flow. Bending strains and accelerations were
measured for the in-line (IL) and cross-flow (CF) direction along the riser model.
Saint-Marcoux & Blevins (2012) present results for two sets of hydrodynamic tests at MARIN, in a tube
fitted with helical 16D and 12D polyethylene boards, respectively. The first set of tests established the
suppression efficiency for intact tablets configuration (no deformation), and the second set established the
performance of the same board after passing a test roll in which fins and body boards suffered from
permanent deformation due to a high contact load simulated in a sting roller during an S-lay installation.
Korkischko & Meneghini (2012) present experimental results for cylinder with the other two small
rotating cylinders located along the main cylinder length which injects momentum in the flow boundary
layer delaying the separation of flow around the cylinder surface. Consequently, the belt becomes narrower
and the oscillating transverse velocity is reduced resulting in a free zone which prevents recirculation vortex
formation. The main advantage of the moving surface boundary-layer control (MSBC) is the possibility of
combining the suppression of vortex-induced vibrations (VIVs) and drag reduction vibrations. Experiments
were performed on an installation of the water circulation channel with circular cylinders mounted on a
lower base cushion air bearing with a degree of freedom in the flow channel transverse direction. The mass
ratio is 1.8.
A dimensionless damping parameter is proposed in Vandiver (2012) which can be used to characterize all
VIVs in a low speed lock-in range. The simple product of the maximum amplitude and damping are shown
equal to the lift coefficient, thus providing a simple method for collection of data from response
measurements. Mass damping parameters are not well adapted to the organization of the response of flexible
cylinders in CFs or cylinders equipped with vortices suppressor or fairings. The dimensionless damping
parameter is well suited to use in constant flow or the cylinder partially covered with vortex suppression
fairings or data from three independent sources are used to illustrate the applications of the dimensionless
damping parameter. It is shown that the method of modal analysis can be used to generalize the application
of this parameter in flexible riser.
(4) VIV and fatigue
Besides the use of the suppression devices in the presence of the VIV, risers are exposed to fatigue. Then,
fatigue life and reliability associated are essential to be verified in riser design. Several investigations have
been conducted regarding the VIV fatigue and published in the literature in the past (Iranpour & Taheri
2006, Modarres-Sadeghi et al. 2010).
In a recent work of Passano et al. (2014), re-examination of the VIV fatigue is conducted with the basis of
the Norweigian Deepwater Program through laboratory tests with a 38 m length riser pipe model. The
observed response behavior is compared with the predicted VIV response and fatigue analysis result.
Empirical coefficients from the experiment used to estimate riser behavior through a numerical simulation
program such as VIVANA presented improved agreement with regard to amplitude and frequency of the
response.
Comparative investigation is conducted in Jhingran et al. (2012) for the ability of helical straps and
fairings, respectively, effectively suppress the VIV. Database from different test programs with a suppressor
878
are classified, and performances observed for each suppressor device has been evaluated. Effects of the main
parameters are analyzed, among those Reynolds number, surface roughness, interference, and coverage
density.
Fontaine et al. (2011) demonstrated the calibration methodology to derive consistent values for the factor
of safety (FoS). Evaluation used the most commonly used VIV prediction software by industry, and it was
based on medium scale VIV available data. Results emphasize the need for a coherent approach to estimate
the FoS, and software improvements on monitoring and measurement to reduce risk of failure if the
influence of such improvements on the FoS is not quantified.
(5) VIV considerations for flexible risers
For dynamic risers, the governing load contributions result from wave and currents loads and associated
floater motions. In addition VIV may also be a concern, especially for deepwater risers where high tensions
and external pressure may limit the amount of hysteresis damping. It has not been possible to find references
to documentation of field observations of VIV response of flexible risers in public domain. However,
laboratory tests show that umbilicals with comparable structural damping may experience significant VIV
response (Lie et al, 2007). As of todays practices, wave and VIV loadings are treated by separate models
e.g. (Zhang & Tan, 2011). Whereas, the nature of wave responses includes large response amplitudes that
requires a non-linear response model, empirical models such as Shear7 and Vivana assume a linear
frequency domain model and small response amplitudes. In order to describe the extreme responses, the
effect of VIV is included in the time domain model in terms of selecting the drag coefficient such that drag
amplification from VIV is included. For fatigue calculations, the fatigue damage is obtained from separate
models where the total damage is calculated as the sum of the wave and the VIV contributions. Significant
uncertainties are present when applying these models with regard to the assumption of time and mode
sharing as well as damping. For flexible pipes, the hysteresis damping will influence the dynamic response,
and this is particularly the case for VIV calculations. In a time domain procedure this can be handled directly
by applying an elastic-plastic material model for the pipe in bending (Smith et al. 2007). For VIV frequency
domain response models, the equivalent damping coefficient related to each mode can be handled by
iteration according to the procedure proposed by Sdahl et al. (2011). The moment curvature diagram
needed for such procedures can be obtained from full-scale measurements or from FE models. Research is
ongoing with regard to developing VIV time domain models that allows non-linear structure response effects
to be included directly (Thorsen et al. 2014).
3.2
Free span VIV of pipeline
Vortex induced vibration (VIV) can lead to fatigue as well as ultimate stress-based failures at unsupported
pipeline sections. Unsupported sections may exist directly after installation due to the seabed profile and/or
develop with time due to soil movements. Two different types of VIVs (In-line and cross-flow) are mainly
considered as shown in Figure 1.
In-line vortex-induced vibrations can lead to fatigue damage due to the high number of cycles. Cross-flow
vibrations may also lead to fatigues damage. However, high stresses can also be induced due to the high
vibration amplitudes which can be created by this type of vibration.
3.2.1
Assessment
The design against free span failure is an initial part of the design of an offshore pipeline. Most of the
pipelines are designed using DNV RP F 105. This code provides easy, usable screening criteria for measured
free span and comprehensive analysis methods like the force model and the response model.
Currently research is carried out regarding realistic soil parameter for the analyses (Yaghoobi et al. 2012),
in which simplified approaches are proposed and the approaches are compared (Bai & Duffy 2014, dos
Santos et al. 2014).
Additional codes are e.g. ABS Guidelines. Comparisons between the codes have been carried out
(Bakhtiary et al. 2007). Besides this design codes, local requirements concerning e.g. the maximum
allowable height in order to avoid hooking exists. Currently some joint industry projects (JIPs) are
planed/running, e.g. for the Free Spans in Trenches (DNVGL), Understanding of VIV (BV), Free Spans
(ITF).
Additional research is currently carried out concerning assessment of free spans in operation (Raposo
et al. 2014) and combined effect like walking, lateral buckling and free spanning (Duan et al. 2014).
3.2.2
879
Mitigation
Regular surveys are normally performed, depending on the seabed activity, measuring the length and height
of free span. If the length or height is above the acceptable criteria as described above, mitigation measures
are taken. The mitigation of free span VIV is mainly done by seabed preparation, trenching, stones/gravel,
installation of mitigation devices, see e.g. Wang et al. (2011) for details and mitigation assessment methods.
In-Line:
Forces:
constant dynamic pressure
Karman vortex street
alternating dynamic pressure
alternating resulting force

flow separation
Forces:
Cross-Flow:
flow separation
variable resulting force
Karman vortex street

alternating dynamic pressure
variable transverse force

constant resulting force
Figure 1. In-line and Cross-flow VIV.
4.
SOIL-PIPELINE INTERACTION
4.1
Introduction
Pipelines are the arteries of offshore hydrocarbon development. They transport the hydrocarbon products
and other fluids between wells and in-field processing facilities and also to shore. As the offshore pipelines are laid along the seabed, the pipeline-soil interaction and, in particular, the geotechnical issues must
be captured adequately in the design process. Pipeline geotechnics is an emerging specialty that involves
applications of geotechnical theory and practice unique to the construction of underwater pipelines. Existing pipeline design codes that are known to offshore industry, include API1111, DNV-OS-F101 and ABS
Subsea Pipeline Guide.
For on-bottom pipelines, the geotechnical design inputs primarily relate to the interaction forces between the pipeline and the seabed in the vertical, axial and lateral directions. Numerical methods used to
design against upheaval buckling require not only the ultimate capacity but also the displacement
response to mobilize this capacity. Much less interest has been shown on uplift displacements to reach
failure but Finch et al. (2000) proposed guidelines based on their experimental program.
Pipelines are laid on the seabed and may or may not be buried. When a pipeline is not buried, the interaction between the pipeline and the seabed results into many aspects of the pipeline design issues. One of
the most difficult aspects in pipeline design, which is an increasing challenge as higher operating temperatures and pressure in deepwater applications, is the management of thermal- and pressure-induced
loading. Large pressure and temperature changes during operating cycles generate high axial compressive
stresses leading to buckling. Buckling is precipitated by axial stresses on the pipelines while axial and
lateral resistance to pipeline movement comes from the soil. Uncontrolled buckling can cause serious
damage to the pipeline. In shallow water conditions, pipelines are often laid in pre-dug trenches and buried to prevent bucking. However, this approach is not feasible in deep water. A cost-effective solution is
needed to facilitate controlled buckling at prescribed locations along the pipeline, thereby relieving the
axial stresses. Another major challenge for pipeline designers is progressive axial movement of the pipeline or pipeline walking, which is significantly influenced by lateral buckles along the pipeline. Pipeline
walking is a phenomenon in which start-up/shutdown cycles cause a ratcheting response in the pipe axial
displacement. Both controlled lateral buckling and pipeline walking are very sensitive to soil-pipeline
interaction and there remains significant uncertainty associated with the characterization of the pipe-soil
880
interaction forces in design. If the pipeline must be buried for stability or to avoid fishing gear, the shielding of the pipeline via the construction of a trench usually requires proper geotechnical design.
This document first introduced the typical soil behavior near the pipeline in Section 4.2. Then the stateof-art review of geotechnics for subsea pipeline is provided in two design scenarios. One is high operating temperatures and pressure pipe in deepwater application, where the management of the thermal- and
pressure-induced expansion is a critical design issue. These include assessment of pipeline embedment
(Section 4.3), lateral pipe-soil interaction (Section 4.4) and axial pipe-soil interaction (Section 4.5). Another design scenario is the on-bottom pipeline stability under extreme hydrodynamic loading from
storm-induced currents and waves in shallow water. Geotechnics of pipeline trench construction by
ploughing and jetting (Section 4.6) is reviewed as well as sediment transport and the liquefaction around
the pipeline (Section 4.7).
4.2
Soil behavior near pipelines
The in situ soil shear strength profile is important for estimating the initial penetration of the pipeline and the
soil resistance to the lateral displacement, both of which may involve cycles of remoulding and
reconsolidation of soil. Flow-round penetrometers, such as T-bar or ball are superior to cone since they allow
cyclic penetration and extraction tests to be undertaken. Laboratory testing of reconstituted material
recovered from the seabed, is used to assess interface friction angles and remoulded strengths at the very low
effective stress levels relevant for pipeline design. After comparison of tilt-table, shear box and ring shear
test results on high plasticity clays from offshore West Africa for normal effective stress levels ranging from
2kPa to 300kPa, White & Randolph (2007) indicates a broad trend of residual friction coefficient. The
simple expression given by Bolton (1986) to link stress level and peak friction angle in sands captured the
same trend.
The pipeline geotechnical design requires determination of the geological history of the area, the seafloor
soil characteristics from laboratory and in situ tests, and appropriate analysis to ensure that the pipelines can
safely resist the operational and environmental forces. The seabed deposits are often normally consolidated
and extremely soft and compressible near the seafloor. Some also exhibit unusual behavioral characteristics,
e.g. calcareous sands or Arctic silts, and there is only limited experience with pipeline design in such soils.
Current design approaches are only good if soil follows a certain pattern of behavior, e.g. they do not lose
strength suddenly. Most pipeline laterally buckling/axial walking experimental studies were conducted in the
centrifuge using kaolin, West African clays or silicone sand in the University of Western Australia and the
large-scale pipe-soil test facility at the Norwegian Geotechical Institute (Langford et al. 2007). When
calcareous sand was first encountered in offshore, people did not understand how it behaves, and this led to a
lot of problems e.g. piles disappearing into the ground, loss of capacity etc. Also, the problems of pipeline
behaviors under static and cyclic loading, although long recognized, continue to engage the attention of
several research groups and to defy simple solutions.
4.3
Pipeline as-laid embedment and riser touchdown
Accurate prediction of the as-laid embedment is crucial to the assessment of pipe-soil resistance, pipeline
lateral buckling formation due to initial load, the buckling response, end expansion and walking behavior
due to system shutdown and restart cycles. The static pipeline embedment in fine-grained sediments has
been well investigated/understood using plastic limit analysis solutions and finite element analysis (Murff
et al. 1989, Aubeny et al. 2005, Randolph & White 2008a, Merifield et al. 2008). However, the observed aslaid embedment in soft soil is much greater than the penetration of the pipe into soil under its submerged
weight even when the stress concentration created within touchdown zone by the catenary shape is
considered (Lund 2000, Cheuk et al. 2008, Westgate et al. 2009).
The additional major embedment is due to dynamic lay effects which arise from oscillation of the
suspended section of pipe, originating from the lay-vessel motion and hydrodynamic loading. The
mechanisms that govern the dynamic embedment at the touchdown zone are not easy to quantify. The
vertical pipe motions will raise the vertical pipe contact force associated with longitudinal translation of
touchdown point accompanied by changes in pipe tension. The dynamic pipe moments soften and remold
the surrounding soil, leading to the reduction in soil strength and bearing capacity. The horizontal motions
also reduce the vertical bearing capacity due to the combined V-H loading imposed on the seabed (Merifield
et al. 2008, Randolph & White 2008b, Cheuk et al. 2008). Moreover, a narrow trench will be created due to
horizontal motions. As a result, in practice, a dynamic embedment factor fdyn is commonly applied to the
calculated embedment using plastic limit analysis solutions to account for dynamic effects. Previous studies
881
have reported that the fdyn factor is in the range of 210 (Lund 2000, Bruton et al. 2008). As a result, the
prediction of as-laid pipe embedment can be associated with significant uncertainty. In order to reduce this
uncertainty in design, the fully remoulded strength coupled with the static overstress is recommended
(Westgate et al. 2010, SAFEBUCK Design Guideline 2011). It is also attractive to set the geotechnical
components of pipeline analysis within a full probabilistic framework. This is consistent with the
probabilistic structural reliability analyses that are increasingly performed during the assessment of pipeline
on-bottom stability or lateral buckling (SAFEBUCK Design Guideline has adopt this novel approach).
4.4
Lateral pipe-soil interaction
Subsea pipeline behavior in deepwater has been studied over the past ten years in attempts to improve
understanding of lateral buckling and pipeline walking. One such key research program is the SAFEBUCK Joint Industry Project (JIP) which was initiated in 2002. The SAFEBUCK JIP has been supported
by a number of major oil and gas organizations worldwide. This program has led to new design guidelines for on-bottom lateral buckling covering single pipe and pipe-in-pipe systems (SAFEBUCK Design
Guideline 2011). A significant part of SAFEBUCK involved research into pipe-soil interaction during
lateral buckling, which is the largest uncertainty in design. This has led to a number of key insights into
the fundamental pipe-soil behavior and the development of simple models to simulate this behavior
(Cheuk et al. 2007, Bruton et al. 2008, Dingle et al. 2008, White & Cathie 2011).
The general form of the lateral response during the first load includes a breakout peak followed by a
residual or plateau resistance. In clay soils, the breakout peak can arise from the loss of suction and
therefore tensile resistance at the rear face of the pipe. After breakout, two characteristic types of lateral response (light pipe and heavy pipe) can occur, depending on the ratio of pipeline weight to
the seabed strength. For the clay soils, this ratio could be defined as V/suD; where V is the flowline
weight, su is the soil undrained shear strength and D is the pipe diameter. Observing in centrifuge
model tests conducted for the SAFEBUCK JIP (Bruton et al., 2008), values of V/suD<1.5 give a light
pipe response, characterized by the pipe rising during the initial breakout. Values of V/suD>2 give a
heavy pipe response, characterized by the pipe diving during the initial breakout. For light pipe, as
the pipe moves upwards the soil surface, the lateral resistance usually reduces from the breakout value
to a steady residual value.
Large amplitude cyclic model tests were carried out to investigate pipe-soil interaction during pipeline
operating cycles (White & Cheuk 2008, Bruton et al. 2008). It is found that surface soil, swept ahead of
the pipe on each cycle, builds up into soil berms at the extremes of the pipe displacement. The residual
resistance during the first load controls the lateral displacement at which the first buckle stabilizes, defining the initial shape of the lateral buckle and the peak bending stress in pipe. While in heavy pipe case,
the pipeline moves downwards after the initial breakout resistance is mobilized. This downward movement, coupled with the growth of a soil berm ahead of the pipe, leads to a steady increase in the lateral
resistance. This hardening form of response could increase the loads in the buckle while simultaneously
inhibiting lateral displacement. These soil berms offer significant resistance to pipe movement and define
the shape of the buckle in operation. Same as lateral response in the first load, two characteristic types of
lateral response (light pipe and heavy pipe) can occur during start-up/ shutdown cycle. The light
pipe showed a similar lateral response as its first load response. The residual friction factor remained
approximately constant throughout each cycle until the static soil berms were reached. The heavy pipe
showed a contrasting lateral response. During the first lateral sweep of 2-D amplitude the pipe embedded
deeper than 1-D. This increasing pipe embedment depth caused a sharp increase in the lateral resistance
and no steady residual value was reached.
For design purposes, the variation in lateral resistance can be estimated from a geotechnical analysis
that considers in detail the mechanisms described above. Then, to incorporate the results into the pipeline
structural analysis, they can be converted into simple relationships. Currently, the simple Coulomb frictional model based on total stress framework was adopted in design calculation. In latest SAFEBUCK
GEO JIP (started in 2010), the lateral pipe-soil response has been addressed in design by a new forceresultant plasticity model which captures results from modeling and testing of lateral pipe-soil inter
action and is designed to be used with standard software packages (Martin 2013). However, the soilpipeline contact is still modeled in total stress terms and pipeline-soil interface is assumed to be
completely cohesive and is defined by a stress-independent and time-independent interface resistance.
This clearly does not address the heaping and remoulding of the soil around and beneath the pipeline as
well as the issue of generation and dissipation of excess pore pressure and its effect on the stiffness and
strength of the soil over time.
882
4.5
Axial pipe-soil interaction
Axial response can vary significantly for different types of soil, weights of pipe, coating roughness and
pipe velocity (in clay soils). Recent model tests indicate that axial friction is strongly influenced by pore
pressure, consolidation and the level of drainage during sliding (White et al. 2011). In general, pipe displacement generates excess pore pressure (undrained condition). The excess pore pressure (normally positive) tends to increase with increasing pipe velocity, giving a low residual axial resistance. If pipe displacement is very slow such that excess pore pressure is dissipated as fast as it is generated, this fully
drained condition will give a gradual increase to a significant high level of residual resistance. These
undrained (fast) and drained (slow) responses generally represent the lowest and highest values of axial
resistance. In typical field condition, the response is likely to lie between the undrained and fully drained
conditions. Currently the important transition from undrained to drained condition is not well understood
and conventional design calculations do not provide adequate predictions of the observed behavior.
A model for axial pipe-soil interaction using the critical state theory has been proposed based on current axial test database (Hill et al. 2012, White et al. 2012). The key feature in this proposed model is the
consideration of excess pore pressure which significantly affects the axial resistance. However, this model
is still in a descriptive stage; it is therefore not really useful from the design calculation purposes. It is
also largely limited to undrained condition. Although a simpler model for the axial resistance prediction
has been proposed (Randolph et al. 2012), it is based on a number of simplified assumptions such as 1-D
consolidation, assumed depth of pore pressure influence and elastic consolidation. This simple model is
unlikely to be applicable to more complex situations that involve buckling and axial displacement. Moreover, the current database is limited to a few high quality tests in very soft, high plasticity West African
clays using some selective representative pipes (mainly 8 and 12 pipes). Influence of pipe diameter,
permeable coating, soil properties and cyclic or repeated loading remains largely unknown. The currently
limited axial test database also does not yet allow full validation of any proposed model.
4.6
Pipeline stability during trenching and backfilling
A pipeline trench may be carried out before the laying process or afterwards. Pipeline trenching was historically performed using jetting, but mechanical ploughing or cutting has become an increasingly common approach. Different trenching techniques are suited to different ground conditions. Recently a unique
concept of the Arctic subsea bucket ladder trencher was developed which is suitable for Arctic conditions.
Ploughing, cutting and jetting are often used in combination on a single machine. In shallow water,
dredging technology may be used to create a trench. Studies were conducted to assess the ploughing resistance for sleds and ploughs. In dilatant sands, the ploughing force can be significantly increased if
undrained or partially drained conditions prevail (Reece and Frinstead 1986). The mechanics of pipeline
plough performance are described by Palmer et al. (1979) and recent studies have examined the performance of ploughs in uniform sand and in sand waves (Palmer 1999).
Jet trenching techniques use high-pressure jets of water from nozzles suspended beneath seabed level
to cut, erode and fluidize soil. Then pipeline is lowed into the area cut or fluidized by the jets. For sandy
soil, the trench walls are not stable and pipe lowing is dependent on the pipeline being heavier than the
fluidized soil. Moreover, there must be sufficient weight margin to ensure that the lowing of the pipeline
is not affected by the turbulence and upward flow of water behind the trencher. White & Cathie (2011)
suggested a minimum specific gravity (or unit weight) is believed to be approximately 1.8. However, this
is without conclusive experimental support. A mathematical model of the jet trenching process in sand
has been developed by Berghe et al. (2008) and Peng & Capart (2008). This model is calibrated by a series of 1g model tests. Both the results from model tests and the mathematical model indicate the dependence of the progressive rates of jetting power and sand density, and capture the main phenomena of sidewall collapse and overspill. For clays, the soil is cut and broken up by jets. After that the essential mechanisms are similar as in sand except the clay is not fully disaggregated. To minimize the risk of floatation
in clays, Powell et al. (2002) suggest that the pipeline specific gravity should be greater than the specific
gravity of a liquefied clay at the onset of effective stress. However, Cathie et al. (2005) believes that the
problem of pipeline lowing and stability in clays is mainly dependent on the amount of slurry left in the
base of the trench. The suggestion by Powell et al. (2002) may not be useful as the design criterion as the
pipeline specific gravity is normally greater than the specific gravity of most muds.
Large plough which is towed by a vessel is also used to create a pipeline trench. A typical pipeline
plough comprises skids ahead of and behind a heavy ploughshare. For most ploughing cases, pipeline
stability is not an issue since the pipeline is lowered into the cut trench as the plough moves forward.
883
However, Cathie et al. (1996, 1998) found backfill ploughing can lead to pipeline uplift. As suggested by
Cathie et al. (1998) and Powell et al. (2002), there are several factors that combine to lead to the uplift.
Flow of soil down the slope of the trench is an important cause of uplift since high transient uplift forces
are generated when soil impacts the pipeline. A backfill plough creates turbulence and in-line water flow,
and liquefaction of backfill can also cause uplift of a lightweight pipeline. Based on their model tests,
Powell et al. (2002) found uplift does not occur if the pipe is sufficiently heavy. The suggested minimum
specific gravity (or unit weight) is approximate equal to 1.8. Accurate prediction of pipeline plough performance is also critical for offshore design. The variation of tow force with plough weight and trench
depth is reasonably well-known (Reece & Grinsted 1986). Since the plough velocity affects the tow force,
the effect of plough velocity has been considered in the current prediction models (e.g. Cathie & Wintgens 2001). In the earlier prediction models, only basic theory and a limited data set of previous trenching
were utilized for the calibration. In order to improve the understanding of plough resistance, further experimental and numerical studies have been carried out (Bransby et al. 2005, He et al. 2005, Lauder et al.
2008, Lauder 2010, Peng & Bransby 2011). The results show that the plough resistance is believed to
have two components, which are static and dynamic components. This is broadly confirmed by the mode
of Cathie & Wintgens (2001). The static component is approximated well in existing solutions, but the
passive pressure coefficient and dynamic component may need further improvement.
Current researches on pipeline stability during trenching and backfilling are mainly on the 2-D aspects
of the problem and only on soil mechanics. However, uplift limitation and propagation during trenching
and backfilling is a 3-D problem and involves pipeline structural response, soil transport and deposition,
and consolidation. All of these aspects need to be investigated integrally in the future.
4.7
Pipeline stability during sediment transport and liquefaction
In the shallow waters offshore Australia, the on bottom pipeline stability under hydrodynamic loading
from storm-induced currents and waves is a significant design issue. The mobility of the seabed may appear in the form of sediment transport and liquefaction which is not considered in current design
methods.
In order to improve design reliability through better underlying science, the physical modeling of
wave-induced liquefaction and sediment transport has been carried out in the University of Western Australia (Henning et al. 2013). From their experimental results, it is found that different soils may become
mobile due to sediment transport and/or liquefaction. Specifically, the selected North West Shell (NWS)
sandy SILT and very silty SAND will both liquefy and move due to sediment transport forces in sufficiently large cyclonic events (i.e. 100 year return periods). In contrast, the laboratory SAND and the silty
SAND will only succumb to sediment transport because these sediments dissipate excess pore pressures
quickly and avoid full liquefaction. The weakest storm event showed no sediment transport and only full
liquefaction for the sandy SILT. The laboratory observations of sediment transport show that these NWS
sediments, particularly those with a significant fraction of fines, have a different mobility to that predicted
using conventional methods calibrated to clean siliceous sands.
Henning et al. (2013) also suggested that the simple pore pressure model from Verruijt [4] can illustrate
liquefaction trends very clearly, but shows several disadvantages which include an inflexibility that hampers accurate calibration of the model parameters and, hence, an inaccurate development of the excess
pore pressure rate. Other models should be investigated to model liquefaction.
The investigation of wave-induced liquefaction and sediment transport is still ongoing, the future research will focus on better defining and quantifying the erosion mechanisms of sediments, particularly
those which exhibit behavior described as cohesive, as well as clarifying the nature of the frictional
seabed shear stress required to drive erosion (Henning et al. 2013). A better understanding will allow for
a more accurate prediction of the threshold shear stress and erosion. This is a particularly urgent need in
regions such as the NWS, where there is no distinction between clean and cohesive sediments, but
instead a continuous spectrum from coarse to fine soils, often poorly sorted with widely mixed particle
sizes.
A joint industry project (JIP) is now being proposed which will aim at identifying state-of-the-art
design methodologies for mechanically stabilizing pipelines and thereby controlling movement by
Geomarine and Crondall Energy. The aim of the JIP is to streamline the selection and optimization of
secondary pipeline stabilization schemes by providing design guidance and a framework for comparative assessments of different schemes. These schemes include the application of rock dumping, concrete mattress, block placement, suction anchors, small diameter driven piles, drilled and grouted
micro-piles, etc.
884
5.
FAILURE MODES OF RISERS AND PIPELINES
5.1
Steel riser and pipelines
Since the 1970s, offshore oil and gas development has gradually proceeded from shallow-water installations
up to around 400 m to the ultra deep waters around 3,000 m that represent the maximum today. The main
design challenge for development beyond 3,000 m is related to the high external pressure that may cause
collapse of the pipeline. From depths of 900 m onwards, external over-pressure is normally the most critical
failure mode for pipelines. The risk of collapse is typically most critical during installation when the pipe is
empty and external over-pressure is at its maximum. In addition, the pipe will be exposed to large bending
deformation in the sag bend during installation that may trigger collapse, and collapse may be relevant for
operational pipelines subject to significant corrosion. This potential failure mode is dealt with by increasing
the pipe wall thickness. But at ultra deepwater depths, this may require a very thick walled pipe that becomes
costly, difficult to manufacture, and hard to install due to its weight (Lee et al. 2014). Currently, there is a
practical limit on wall thickness that limits the maximum water depth for 42-in. pipes to around 2,000 m.
However, for a 24-in. pipe this limit is approximately doubled to 4,000 m based on developed drilling and
production capacity. Those factors will be a major influence on the final compressive strength of the pipeline.
Current deepwater gas pipelines have thick walls and, due to quality and safety requirements, the number
of pipe mills capable of producing the pipe is limited. When installing pipelines, the heavy weights are
difficult to handle and the thick walls are challenging to weld. The number of pipe-laying vessels for
deepwater pipelines is limited, too. New offshore oil and gas fields are being developed in deeper and deeper
waters and export solutions for the gas are critical. New exploration activities are also heading for ultra deep
waters. According to new exploration, transporting oil and gas from high-pressure and high-temperature
reservoirs through pipelines is also a major challenge. A pipeline laid on or buried in the seabed responds to
high pressure and high temperature by expanding, resulting in axial displacement (also known as end
expansion), lateral buckling, upheaval buckling, or a combination of these. Such pipeline movements can
cause failures and are critical to the integrity of a pipeline.
The number of potential failure modes for a multilayer structure such as a flexible pipe is high. However,
the number of different failure modes experienced in operation is more limited. API (2007), PARLOC
(2001) and Cosham & Hopkins (2004) list and describe all of the most probable failure modes and defects
for a pipe. The use of this reference is highly recommended to get a complete overview. The potential causes
and effects of damage during installation phase of the pipelines are summarized in this section.
5.1.1
Buckling (buckle propagation), collapse and fatigue failure
Lateral (horizontal) and upheaval (vertical) buckling are possible in subsea pipelines. Lateral buckling is
usually associated with unburied pipelines at areas where the alignment is curved, whereas upheaval
buckling will be on buried pipelines at over bends, where the profile is convex upward.
The key integrity issues are structural over-stress and fatigue failure, short and long term. Offshore
pipelines are required to operate at even higher temperatures and pressures (DNV-RP-C203 2005, DNV-RPF108 2006). The resulting high axial stress in the pipe wall may lead to unexpected buckling, which may
have serious consequences for the integrity of the pipeline if this is not taken into account during the design
phase. Unexpected lateral buckling has been observed in several operating pipeline systems.
The offshore industry lacks a complete understanding of lateral buckling, and efficient tools for simulating
buckling behavior early in the design phase would make a valuable contribution to our knowledge (Chiodo
& Ruggieri 2009, Gong et al. 2013, Netto et al. 2005).
Uncontrolled global buckling can cause excessive plastic deformation of the pipeline, which could lead to
localized buckling collapse or cyclic fatigue failure during operation due to multiple heat-up and cool-down
cycles, if it is not properly managed. The most relevant failure modes of pipeline lateral buckling may
include loss of containment, as result of fracture failure on the tensile side of the cross section, due to
excessive utilization and low cycle fatigue under cyclic thermal loads. When a pipeline on the seabed is
heated, it tends to expand, and the expansion is resisted by the friction generated by the seabed. When the
pipeline is cooled, it contracts, but the seabed friction resists the moving pipelines return to the original
position. If heat-up/cool-down cycles involve significant thermal gradients, then axial ratching of the
pipeline can occur, with displacement toward the cold end. Over a number of cycles, this movement can lead
to very large global axial movement with an associated overload of the spool piece or jumper at the pipeline
end. This cumulative axial movement is called pipeline walking (Bai & Bai. 2014). The phenomenon of
lateral buckling and walking of pipeline has been widely investigated over the past decades.
885
Deepwater pipelines are normally subjected to external pressure and bending, and they are designed to
prevent buckling and collapse failures. But a pipeline that is locally damaged may collapse and, if the
hydrostatic pressure is high enough, the collapse may propagate along the pipeline. The collapse propagation
pressure is the lowest pressure value that can sustain the collapse propagation. Since the external collapse
propagation pressure is quite low in comparison with the external collapse pressure, it is necessary to install
buckle arrestors, at intervals along the pipeline, with the purpose of limiting the extent of damage to the
pipeline by arresting the collapse propagation (Netto & Estefen, 1996).
Buckle arrestors are devices that locally increase the bending stiffness of the pipe in the circumferential
direction and therefore they provide an obstacle in the path of the propagating buckle. There are many
different types of arrestors, but all of them typically take the form of thick-walled rings. The external
pressure necessary for propagating the collapse pressure through the buckle arrestors is the collapse crossover pressure. Previous studies presented finite element models that simulated the collapse and post-collapse
behavior of steel pipes under external pressure and bending. Those finite element models were used to
analyze the effect of different imperfections on the collapse and collapse propagation pressures of the steel
pipes (Toscano et. al. 2008, Toscano et al. 2002, 2003, 2004, Ramasamy & Ya 2014, Gong & Li 2015).
Many experimental results are available in the literature for the cross-over of integral buckle arrestors
under external pressure (Gong et al. 2012, Lee & Kyriakides 2004, Olso & Kyriakides 2003, Lee et al. 2008,
Estefen 1999, Netto & Kyriakides 2000a, 2000b).
In deepwater applications the carrier pipe must be designed to resist collapse due to the ambient external
pressure while the design of the inner pipe is primarily based on the pressure of the hydrocarbons it carries.
The collapse of relatively thick pipes under external pressure has been studied extensively and is quite well
understood. Moreover, the buckle propagation phenomenon in pipe-in-pipe systems under external pressure
was studied in combination with hyperbaric chamber test, uniform ring collapse model, and numerical
simulation (Kyriakides 2002, Kyriakides & Vogler 2002, Kyriakides & Netto 2002) which put forth an
empirical formula of the buckle propagation pressure for two-pipe systems through the fit of limited test data.
Lourenco et al. (2008) applied a non-linear three dimensional finite element model to conduct an extensive
parametric dependence analysis of the quasi-static buckle propagation in sandwich pipes, and further
investigated the actual contribution of the core material to the propagation pressure.
5.1.2
Corrosion
The corrosion phenomenon in the oil and gas pipeline system is a serious problem in the petroleum industry
today. Previous reports (PARLOC 2001) have shown that failure mechanisms for offshore pipelines are
strongly linked to damage caused by corrosion and external loads as shown in Table 1. Corrosion problems
may occur in numerous subsystems within the offshore oil and gas production system, including the gas and
oil pipelines. It is recognized as one of the most important degradation factors of pipeline metallic material
and a great concern in maintaining pipeline integrity. Also, corrosion tolerance must be carefully considered
in the design of a pipeline (Bai & Bai 2005, Bai & Bai 2014, Klever 1995, Mohd & Paik 2013).
Previous studies have assessed the importance of corrosion damage evaluation for numerous structures,
including gas pipelines and offshore structures, and assessed their mathematical models (Gabor & Lazlo
2012, Kim et al. 2013, Kyriakides & Corona 2007, Mohd et al. 2013, Paik &Thayamballi 2007, Teixeira
et al. 2008). Also, there are lots of methods (codes) in the engineering practice to determine the burst
pressure of corroded pipes depending on the loadings and the scopes of the pipelines (Adib-Ramezani et al.
2006). These semi-empirical methods based on measurement data (ASME B31G, Modified ASME B31G,
DNV RP101) consider only the length and depth dimensions of the simple 2-D geometrical shapes which
are used to approximate the real corrosion failures.
Table 1. Allocation of failure mechanisms for offshore pipelines.

Failure mechanism
Distribution
Corrosion
36%
Material
13%
External loads causing damage
38%
Construction damage
2%
Other
11%
886
5.1.3
Crack
Stress corrosion cracking can be a serious threat to the integrity of natural gas and petroleum pipelines
(Palmer& King 2004, Bai & Bai 2005, Oh et al. 2007). The pipeline industry responded to this threat by
performing a comprehensive research program to determine the causes of the failures and investigate various
techniques (Nyhus et al. 2005) for preventing future failures. A relatively concise list of discoveries has had
a measurable impact on mitigation of the stress corrosion cracking threat. Specially, hydrogen sulphide is a
highly toxic and corrosive gas, having effects at ppm concentrations. It is soluble in hydrocarbons and water
and will partition between these depending on local pressure, temperature, and Palmer & King (2004)
categorized into three cracking forms, namely, sulphide stress cracking (SSC), hydrogen-induced cracking
(HIC) and stress-oriented hydrogen-induced cracking (SOHIC).
Internal and external corrosion mechanisms such as sulphide or chloride cracking need to be considered.
Nickel content in the range 1-9% is perceived to reduce the resistance of the material to sulphide stress
corrosion cracking. It is introduced as the addition of alloying elements alters the weld ability, sometimes
adversely, the additional costs associated with special weld procedures should also be taken into account in
the cost-benefit analysis of pipe material selection (Palmer & King 2004).
5.1.4
Erosion
Erosion has been long recognized as a potential source of problems in the hydrocarbon production system.
Many dangerous elbow failures due to erosion have occurred on production platforms, drilling units, and
other subsea equipment in the past decades. While the subject of erosion around subsea pipelines has been
studied (Chen et al. 2004), a review shows a number of areas where relatively little existing published work
has been presented, including the variation in seabed shear stresses around the shoulders of subsea pipeline
spans.
There are two primary mechanisms of erosion. The first is erosion caused by direct impingement.
Normally, the most severe erosion occurs at fittings that redirect the flow such as at elbows and tees. The
particles in the fluid can possess sufficient momentum to traverse the fluid streamlines and impinge the pipe
wall. The other mechanism is erosion caused by random impingement. This type of erosion occurs in the
straight sections of pipe even though there is no mean velocity component directing flow toward the wall.
Venkatech (1986) provides a good overview of erosion damage in oil wells.
Shear stress amplification was derived from experimental erosion rates using the approach of backcalculating an averaged value based on an assumed amplification zone. The validity and accuracy of
published amplification factors is limited to the present level of knowledge on the topic of sediment transport.
Shen et al. (2015) presents a comparison studied with physical test interpretation.
5.2
Flexible pipes
5.2.1
Failure modes
Flexible risers including end fittings have a complex mechanical behavior and are exposed to environmental
loading from waves and floater motions, external hydrostatic pressure as well as pressure and temperature
loading from the conveyed bore fluids. Identification of all relevant failure modes and loading mechanisms
is essential to ensure the structural integrity of flexible risers. Design criteria for all known failure modes are
given in (API 2008) which is the main design code for flexible risers. However, some concerns have been
raised on the time lag between determination of root cause of new failure modes and revision of the standard
(4subsea 2013). The most frequent failures for flexible pipes are related to outer sheath damage, carcass
failure and ageing/abrasion/wear (Sparks 2003, Kenny 2010, 4subsea 2013). A major incidence rate of 1.5%
per riser per operational year is reported to PSA Norway (4subsea 2013). A major incidence is defined as a
high risk of injury or pollution mainly based on judgments by the operator.
5.2.2
Design analysis
The offshore field development trend in terms of deeper waters, increased reservoir pressures and
temperatures as well as corrosive well stream conditions (H2S and CO2) in combination with the composite
nature of flexible pipes that includes materials with different as well as time dependent mechanical
properties, give significant challenges with regard to documenting sufficient service life. In addition for
existing risers, where corrosion of the tensile armor due to increasing H2S levels outside the original design
envelope or seawater ingress from damaged outer sheaths may give a significant reduction of remaining
service life for many risers (4subsea 2013). For deep water applications, collapse due to external pressure or
887
caterpillar forces during installation as well as tensile armor buckling due to the reversed end-cap effect
represent additional challenges to be addressed. From a design point of view, major challenges are related to
the long term change in mechanical properties including creep of the plastic materials and reduced long term
fatigue performance of the steel materials. In addition comes the long term buckling performance due to
combined cyclic load action. Whereas ultimate strength and fatigue performance issues of intact pipes away
from the end fittings may be dealt with by simplified local models, the structural behavior inside end fittings,
exposure to caterpillar forces and damaged pipe scenarios may require full finite element (FE) analyses
where all relevant structural details are captured. General FE software has therefore been increasingly used
in recent years to address these issues. De Sousa et al. (2012) developed an ANSYS model that was used to
correlate the tensile armor birdcaging failure to test data where good correlation was found. De Sousa et al.
(2012) and Ji (2014) developed models for the corrosion fatigue scenario assuming a set of broken tensile
wires to investigate the reduction in fatigue life. Svik (2014) developed a model addressing the lateral
transverse buckling scenario during cyclic loads. Good correlation was found with test data (stergaard
2012) and a design criteria based on cyclic loading first yield was proposed.
In recent years several new failure modes have been experienced such as fatigue failure inside the end
fitting for deepwater risers (Campello et al. 2012) and carcass failure due to collapse or tear out (Farnes
et al. 2013). In both cases, FE models have been developed to document the performance or improving the
design. For the carcass tear out case a viscoelastic model was applied describing thermal shrinkage and stress
relaxation effects in the PolyVinylidene DiFluoride (PVDF) layers exposed to the transient operation load
condition (Kristensen 2014). This points towards improved qualification procedures by means thorough
failure mode identification, use of advanced finite element analyses in combination with material testing to
avoid such failures in future applications.
For standard static stress analyses, i.e. load cases where the helical layers can be assumed to carry the
imposed load by membrane action only consistent with the API approach of mean stresses (API 2008) there
are already well established procedures used by the industry. However, facing the challenges from corrosion
fatigue in the tensile armor, extra care needs to be taken when evaluating residual life since eventual
fractured wires may cause local bending in the pressure armor leading to immediate failure. This requires FE
models that adequately describe the bending behavior of the pressure spiral, i.e. the mean stress approach is
no longer valid.
There are also well-established models for fatigue stress analysis of the tensile armor. This includes
analytical models (Skeie et al. 2012), tailor made FE models (Svik 2011) and application of general FE
codes e.g. (Perdrizet 2011). Both the FE model approaches are capable of describing the boundary condition
represented by the end fitting; however, the general FE approach e.g.(Perdrizet 2011) still requires long
computing times that are not suitable for standard engineering calculations. Experience has shown that as
long as the bend curvature load is introduced more than half tensile armor pitch length away from the end
fitting, the boundary condition effect is normally not an issue for the moderate curvatures governing fatigue
analyses. Significant efforts have also been made in recent years to provide input to the industry in terms of
tensile armor SN-data for fatigue analysis in corrosive environments (Berge et al. 2009). This work is
ongoing.
However, for large curvatures experienced either during extreme operation or installation conditions, both
the end fitting and curvature gradient effects may introduce behaviors that require FE modeling. This is
particularly the case during installation spooling operations where torsion may be introduced due to
changing the direction of curvature. If the torsion causes locking between the tensile armor layers, the nonlinear moment capacity of the pipe will increase and may ultimately cause birdcaging or transverse buckling
failures in the tensile armor. To evaluate such scenarios, FE analysis is required using material models that
allow updates of the layer locking mechanism.
5.2.3
Monitoring
In MARINTEK (2014), the flexible riser monitoring methods are grouped into; (a) commonly used methods
such as hydro test and offshore leak tests according to (API 2008), internal pressure and temperature
monitoring and bore fluid characteristics; (b) widely used methods including external visual inspection, inline coupon monitoring, annulus (volume test, ventilation test, gas sampling) monitoring; (c) systems that are
in some use including annulus vent, internal gauging, radiography and curvature monitoring; (d) systems
that are in limited use or under development such as fibre optic monitoring, eddy current methods, laser leak
testing, annulus temperature monitoring, ultrasonic inspection, X-ray tomography, torsion monitoring,
acoustic emission, nonintrusive stress (magnetic) monitoring.
888
Based on the operation experiences revealing changes in the bore chemical composition during riser life
and the complexities of failure modes encountered e.g. (Farnes et al. 2013), residual life evaluations will
totally rely on how accurate load (environmental, pressure, temperature) and chemical condition histories
that can be used as the basis for the evaluations. This is particularly the case for the corrosion fatigue issue,
where the assumption with regard to annulus conditions totally determines remaining service life. All means
of describing these conditions in detail contributes to reduce the uncertainties in residual life evaluations.
However, considering the potential dramatic consequences of failure and the inherent uncertainties, systems
that can predict wire failures are still needed. The development of fibre-optic systems e.g. (Weppenaar 2014)
and nonintrusive stress monitoring systems that can sense single wire failures are promising technologies to
meet these challenges.
6.
INSTALLATION
6.1
Risers
In the past few decades, offshore engineering projects dedicated a great effort to develop innovative
conceptual installation procedures for risers. This effort was not only for the riser itself onsite installation,
but also for the transportation procedure of riser system and components from the land to offshore. Several
engineering as well as scientific research investigation have been developed for this purpose.
Experience gained on installing subsea facilities including riser systems, in general, has been reported.
Ultra deepwater field development pushed the oil industry to pursue frontiers of the technology to develop
innovative solutions for riser system concept and installation, observing risks involved and cost reduction
pressures. Engineering work as well as laboratory investigations evaluated feasibility of those operations
facing challenges of the ultra deepwater depth beyond 3000 m. Improvements on the knowledge of
installation procedures for new systems such as the Submerged Buoy for Supporting Risers (BSR) have been
reported. Numerical and experimental investigations were conducted regarding system behavior during
installation, involving several kinds of support vessels. A full-scale test has also been conducted to verify
feasibility of the operation.
This section briefly pointed out some research works in the recent years that contributed to improve
understanding of the riser installation process, related to system behavior to environmental loads by means
of experimental, analytical or numerical analysis approaches.
In Franciss et al. (2011) and de Araujo et al. (2011), model test results are described to validate installation
procedures to anticipate problems during on-site operation, and also discuss test of pull-in operations for two
flexible risers after the actual buoy is installed. Design of monitoring systems to verify all forces and
displacements during the installation is described. Results demonstrate technology development for
petroleum production in ultra deepwater depths. Furthermore, in Saito et al. (2011) innovative fixed
submerged supports of flexible risers for lazy-S configuration is presented. Installation procedures as well as
discussion to reduce both construction and installation costs are discussed. Tolls for inspection and reliability
of the structure are compared with other floating buoy solutions for risers.
Application of the nonconventional pendulous installation method (PIM) deepwater installation
technology is introduced in Wang et al. (2013), and it describes how to install a 195-ton manifold in a water
depth of 1,500 m. The analysis methodology and numerical findings of the PIM installation are presented for
lifting off deck, overboarding, lowering through splash zone and water column and landing into the target
box on the seabed.
Installation phases and analyses performed for the offshore operation during the transport and installation
of three subsea templates at the North Sea is described in Mouhandiz and Troost (2013). A full-scale
monitoring program was performed during the installation, and accuracy of the theoretical models have been
observed.
J. L. Legras and B. Pillet (2013) presented the tethered catenary riser (TCR), a new riser concept for field
development in deep and ultradeep waters. Steel catenary risers (SCRs) are supported by a subsurface buoy
which is tethered down to seabed and anchored. Flexible jumpers are used to connect the buoy with the
floating production unit (FPU). Installation studies showed its feasibility, since this operation could be
undertaken following a similar other procedure already in use. The system is an option for use in deepwater
developments all over the world.
6.2
Pipelines
For pipelines the installation is typically the most critical phase in their lifetime. High stresses and fatigue
loads may occur. Therefore, special pipeline installation vessels have been developed since the beginning of
889
the 70s as shown in Table 2. These vessels receive the pipeline sections with a typical length of 12.4 m,
which are welded together in the firing line onboard the vessels.
Currently pipelines are installed in water depths up to 2,900m. For deeper water applications flexible
pipelines are often used instead of rigid ones. Additional challenges arise during installation due to the
combination of the installation loads together with the attached equipment, e.g. Pipeline End Terminations
(PLETs).
Among different pipeline installation methods like reeling, S-Lay, J-Lay, the reeling method is common.
The new Ceona Amazon has a unique configured G-Lay system (under patent application). Ceona (2014)
and other new concepts of installation are coming up (Perez et al. 2014).
Before the installation, comprehensive installation analyses are carried out. Software tools based on the
finite element method like OFFPIPE and Orcaflex are specialized for this purpose. Nevertheless, commercial
general purpose software like ABAQUS, ANSYS etc. can be used (Dawood & Kenny 2013, Marchionni
et al. 2011). Currently research work is carried out with respect to material behavior effects of reeling
installation (Shitamoto et al. 2014, Sriskandarajah et al. 2013) and reeling simulation (Karjadi et al. 2013).
Static analyses as well as dynamic analyses are carried out. The static results are typically assessed based on
stresses and local buckling criteria e.g. according to DNV-OS-F101. Dynamic analyses cover fatigue and
sometimes crack growth analyses.
The dynamic analysis is used to determine e.g. allowable weather windows for the installation and the
allowable holding times in one position. Currently dynamic amplification factors, side current, are
investigated in bigger JIP projects in cooperation between operator, installation and classification companies,
e.g. JIPs around installation analysis launched by DNV GL (2014).
Table 2. Pipeline installation vessel (length over 190 m).
Name
Solitaire
Castoro Otto
Saipem 7000
Audacia
Castorone
Ceona Amazon
Pieter Schelte
Owner /
Built in
Operator
Allseas
1972
Saipem
1976
Saipem
1985
Allseas
2005
Saipem
2012
Ceona
2014/5
Allseas
2014
7.
INSPECTION AND REPAIR
7.1
Risers
Method
S
S&J
J
S (Bow)
S&J
G
S
Diameter
[in]
260
460
432
260
60
624
668
Water depth Length

[m]
[m]
2,775
300
191
>2,000
198
1,050
225
3,000
330
3,000
199
> 2,775
382
Risers are generally designed to remain in service for the full field life of 20 or more years without little if
any need for maintenance or repair. Integrity management programs that determine maintenance and
inspection requirements are developed during detailed design, based on a risk assessment to identify
degradation threats and methods of mitigation (Cook et al. 2006). The operating parameters that are
monitored, methods of inspection and associated frequency of inspection are then defined according to the
expected severity of operating conditions and response.
As a minimum, visual inspection is conducted regularly, typically every year. Subsea components are
inspected by ROV and surface mounted equipment including flow control valves and trees and tensioning
systems by maintenance/inspection personnel. In the presence of aggressive production or export fluids,
corrosion coupons may be used or pigging of the riser conducted as an extension of pipeline inspection.
Findings from visual inspection have shown various forms of degradation including cathodic protection
system failure through anode depletion and excessive marine growth which can be particularly detrimental
to response when it occurs over strakes used for suppression of vortex induced vibration (Vadel 2013).
These issues have been resolved through the provision of anode sleds for cathodic protection replenishment
and jet washing to remove marine growth.
890
Fatigue is a key design driver for riser systems. Uncertainties exist in the prediction of wave, current and
vessel motion induced fatigue for which monitoring systems have been implemented. Examples of
successfully implemented systems include a stand-alone (offline) motion measurement system for top
tensioned risers (Thethi et al. 2005) and an on-line strain and motion measurement system for steel catenary
risers (SCRs) (Constantinides et al. 2011). A longer term program involving various deployments of
standalone motion monitoring systems on drilling risers for VIV measurement and analysis calibration has
also been conducted (Tognarelli et al. 2012).
Due to the differences in configuration and service requirements of different riser types, the methods of
inspection, maintenance and repair adopted can vary considerably from one riser system to the next. The
challenges and issues specific to different riser types are discussed below.
Fixed Platform RisersRisers providing well access on shallow water production platforms can be subject
to high levels of corrosion near and above the mean sea level due to wetting and drying action and the
effects of condensation. Many such risers have been in service for 20 years or more and economics are
driving a need to extend service lives. Inspection is generally focused on measurement of conductor
corrosion, conducted using ultrasonic testing (UT) probes or various forms of eddy current devices (Reber
2012). Measurement of surface casing corrosion is also being conducted less widely (Munns et al. 2007).
Methods of maintaining or improving integrity to provide extended service lives include the use of grouting
the conductor-casing annulus and use of welded sleeves or clamps to effectively replace the corroded
conductor (Ramasamy et al. 2014). For prevention of casing and conductor corrosion in the annulus, topping
up the internal fluid with rape seed oil has been adopted (Munns et al. 2007).
Deepwater Dry Tree RisersVertical or (near) vertical top tensioned risers used on spar and tension leg
production platforms provide direct well access and in some cases import or export of production fluids. The
tension setting on these devices is critical to satisfactory riser response and is monitored by way of tensioner
cylinder pressures, or load rings in the case of air-can tensioners on spar platforms. Provision is generally
made for swapping out tensioner cylinders in the case of cylinder seal or rod damage. Uncertainties in
current profiles that generate vortex induced motion of the vessel or vortex-induced vibration from direct
action on the risers may warrant monitoring of dynamic riser response to calibrate analysis predictions
(Thethi et al, 2005).
Steel Catenary Risers (SCRs)The focal point of SCR inspection includes wall loss, fatigue damage
and excessive trenching near the touchdown zone on the seabed and integrity of VIV suppression
devices such as strakes or fairings. Failure of the flex-joints that connect the riser to the vessel have
occurred (BSEE 2008) (J P Kenny 2007) resulting in the need for repair. This requires removal of the
fixed piping and simply replacing the upper body (Selden 2009) or replacement of the complete flexjoint that involves a reversal of the final stage of the installation operation. High fatigue damage can be
incurred in SCRs in a localized region around the touchdown point on the seabed. In the presence of
aggressive production fluids, wall thickness measurements from this region, obtained from intelligent
pigging, may be needed to enable qualification of long-term integrity (Urthaler et al. 2013). A means
of managing SCR TDP fatigue has been proposed that involves movement of the platform position by
adjustment of mooring lines, changing the location of fatigue damage concentration along the length of
the riser in the process.
Hybrid RisersFree standing hybrid risers that consist of a vertical steel riser pipe (or bundle) supported
by a buoyancy tank with a flexible riser connection(s) to the vessel are used in West Africa, Brazil and the
Gulf of Mexico. A comprehensive monitoring system has been implemented to confirm riser response
(Zimmerman 2009), but typical monitoring and inspection activities are focused on the integrity of the
buoyancy tank, which is critical to successful long-term performance. This is achieved by use of tension
monitoring devices mounted on the riser or by taking measurements from the tank using remotely operated
vehicles (ROVs) mounted flooded member detection devices. In the event of buoyancy tank compartment
failure, integrity can be restored through the displacement of the ballast water in the reserve compartments in
the base of the buoyancy tank using nitrogen.
Flexible RisersExternal visual inspection is conducted for flexible risers in the same way as for metallic
risers, with the added requirements that supporting structures and ancillary items such as buoyancy must be
addressed and greater attention must be paid to the integrity of the surface condition of the external sheath.
Minor surface damage may be reparable, subject to agreement between the vendor and operator. Numerous
methods of monitoring and inspection are being developed for flexible risers and pipelines. These include
annulus leak detection, fiber optics, included during manufacture, for measurement of curvature and
temperature and various armor wire corrosion and breakage measurement systems such as the magnetic
anisotropy and permeability system (MAPS), eddy current, radiographic, ultrasonic and acoustic emission
891
(Boschee 2012). A key area of potential failure for flexible risers is at the support near the surface, where
fatigue loads and potential for damage tend to be greatest (Seaflex 2007). While it is generally considered
that measurement of annulus condition to identify the presence and nature of fluids that have permeated
through the layers provides a vital guide for inferring the fatigue performance of the armor wires, it has been
found that this assumption and the assumed fatigue resistance of the armor wires may be overly conservative
(Boschee 2012). In view of these uncertainties, amongst others, a combination of inspection methods tends
to be recommended.
7.2
Pipelines
Pipeline systems shall be designed and operated safely, with respect to humans, the environment and the
economy, to maximize the life cycle value. The process is a continuous process applied throughout design,
construction, installation, operation and decommissioning phase to ensure that the system is operated safely
(Bai & Bai 2005). It is better to understand the typical characteristics of pipelines for maintenance, and
repair:
Water depths are beyond diver limits and all activity is remote; wall thicknesses are typically high
(material, welding, buckling).
Operating pressures are typically very high or very low and ambient external pressures are high,
commonly similar to internal operational pressures (coating and insulation degradation).
High levels of insulation are commonly required (insulation degradation).
Waters are typically cold approx. 4C 6C (flow assurance, materials).
Pipelines tend not to be protected by a concrete coating (damage) and geo-hazards can be significant
(spanning, buckling, damage, bend stability, turbitity and debris flows).
Slugging within produced fluids is common (spanning, fatigue).
Greater tolerances (survey inaccuracy, installation accuracy).
Metocean and environmental conditions tend to be benign (stability).
Seabed mobility is less dominant (scour, spanning).
Corrosion coatings tend to be of very high quality (corrosion, damage).
7.2.1
Maintenance
Normally subsea facilities, including the pipeline system, shall possess sufficient reliability to ensure
availability throughout the field life. Subsea pipeline that is susceptible to failure should be designed to
minimize the effort/costs required for replacement of the failed assembly (Bai & Bai 2005). Currently
maintenance methods are categorized into preventive maintenance, routine maintenance, and corrective
maintenance. Because of the high cost and potential delays associated with intervention, preventive
maintenance should be eliminated at the design stage, wherever possible. Routine maintenance tasks are
required where the elimination of specific intervention is uneconomically or technically problematic.
Normally such maintenance would be undertaken during repair activity, or combined with planned
inspection campaigns (ABS 2014). Intervention to rectify breakdown or degradation (Corrective
Maintenance) is referred to as Repair.
7.2.2
Inspection
Pipeline inspection is a part of pipeline integrity management for keeping the pipeline in good condition.
The rules governing inspection are the pipeline safety regulations (DNV-RP-F116 2010, PETRONAS 2011).
Inspection campaigns are an integral part of the IMR strategy, the purpose of the inspections being to
monitor pipeline system integrity over time and to monitor the impact of the subsea and production
environments on the pipeline (Anderson 2005). Understanding and confirming design assumptions, routine
inspections may indicate a requirement for more specific investigations involving detailed or specialist
techniques. The normal physical inspection tasks undertaken on the deepwater pipelines can be split into
locations internal and external to the pipeline. Internal and external locations are typically periodically
inspected by pigging and ROV/AUV methods respectively.
(1) Deepwater pig inspection
Pig inspection of offshore pipelines tends to look for internal problems (Bai & Bai 2005). Generally running
pigs in offshore pipelines is very similar to running in onshore lines, after the wall thickness and higher
pressures are taken into consideration. The most favored inspection methods are either ultrasonic or
magnetic flux inspection. Magnetic flux is limited by magnet strength, i.e. get enough magnetism in the wall
892
of the pipe to enable good results to be obtained. Ultrasonic can inspect very thick wall pipe, but ultrasonic
has to be run in a liquid medium. The main difference between offshore and onshore is the length of run
between pig traps, as offshore pipelines tend not to have intermediate compression stations with
conveniently located pig traps. The pig must not get stuck in the pipeline as retrieving it will be much more
expensive than from an onshore pipeline. The pig must stay alive and recording data (battery duration may
be an issue).
(2) Deepwater ROV inspection
Traditionally, external inspection of deepwater pipelines is performed using work ROVs deployed from DP
ROV support vessels (McStay et al. 2005). These vessels are expensive, and they may not be available when
they are needed most. In deep waters, ROVs become heavy to handle from these vessels, because of long
umbilicals, thus becoming prone to breakdowns. ROV inspections of long transmission lines can be very
slow and may take many months to complete end-to-end. Weather downtime is also an issue for ROV
support vessels when working in harsh and hostile environments.
(3) AUV based inspection
AUV-based inspection in deepwater fields may provide dramatic improvements in cost, performance, safety
and reliability.
Large DPII vessels with high-end ROV spreads would no longer be required for simple inspection.
AUVs have demonstrated solid performance requiring simple autonomy for missions such as bathymetric
survey and high resolution sonar imaging.
AUVs can be deployed from small utility vessels, and are capable of operations in higher seas without the
operational limitations and equipment hazards imposed by umbilical and tether management systems.
Reduction in equipment complexity, vessel size and crew size would also result in improved safety,
reliability and lower environmental impact.
In the future AUVs would become field resident, residing in the subsea field for periods of months. The
end state of Vessel Independent Operations will achieve further reductions in cost while improving
performance and safety.
(4) Optimization of Routine/Scheduled Inspection
An optimum IMR plan aims to strike an appropriate balance between the following objectives:
Maximizing the availability of the pipeline system during its operating life by maintaining and preserving
its integrity, thus maximizing revenue.
Minimizing inspection, intervention and rectification measures through the life of the pipeline system,
thus minimizing through-life IMR related costs.
Reducing to as low as is reasonably practicable all risks to people, the environment and assets, in
accordance with legislative, societal and business requirements, thus minimizing the costs of failures.
Permanent monitoring methods also exist and are becoming more commonplace. The designers have
probably planned for the worst case, but if things are not that bad and/or the operational approach
changes this can result in very different results to those planned and designed. The requirement for and
frequency of inspection will most commonly be determined using risk-based techniques (Bai & Bai
2014, Seo et al. 2015).
7.2.3
Repair
Damage to a subsea pipeline can be repaired in different ways, depending on the water depth and the type
and extent of the damage. This section describes the various types of conventional methods currently
available for repairing a damaged subsea pipeline (Manelli & Radicioni 1994). The minimum functional
requirements identified for an emergency repair system listed in several operable and capable items:
Operable at water depths up to deepest water of the pipeline, on pipe size (internal diameter) of pipelines,
with steel wall thickness up to maximum and relevant coatings, on soft seabed soils (soft calcareous clay and
silt), on seabed slopes, and capable of providing a repair capability extending from minor dents to
replacement of multiple pipe joints.
While not mandatory, it is advantageous if the system(s) and equipment also exhibit the following
characteristics: Modular and/or lightweight, minimum number of components, incur minimal shut down
and/or reduction of operation, minimum CAPEX investment. An overall pipeline repair system to install a
893
clamp or spool requires an extensive array of equipment to conduct a repair operation. The repair systems
generally perform tasks from the following list:
Metrology of the pipeline damage and repair site

Isolation of the damaged section of pipe with internal plugs if required
Soil excavation
Pipeline lifting, locally at the repair site or completely to the surface
Pipe coating removal
Pipe cutting
Removal of damaged section
Pipe end surface preparation
Metrology of the pipeline for clamp and spool piece preparation
Transport and positioning of clamps, spool pieces and connectors
Closing and sealing clamps and connectors
Testing the repair
Lower the pipeline to the seabed
Removal of repair system equipment.
Damage scenarios during installations and operation pose differing levels of risk. The most significant
potential damage scenarios during the installation phase are dry and wet buckles. The technology and
methodologies required for rectification of installation phase damage (i.e. buckles) are a direct extension of
techniques used for similar events in shallow water, and currently exists with installation contractors and
specialist equipment suppliers.
Several potential damage scenarios exist during the operational phase. The most significant are those
where a damaged section of pipeline needs to be reinforced, replaced or cleared of a hydrate blockage.
Where a replacement pipeline section is required, the length could vary significantly depending on the nature
of the event causing the damage (a few meters to several kilometers in the event of a geohazard (i.e. slope
instability). There is a wide range of qualified or nearly qualified equipment for the subsea repair, both
currently available and under continual development. The equipment exists both as individual components
(equipment, tools and fittings) and full systems. Some repair systems are owned and operated on a club
basis, by a group or consortia of pipeline operators. The clubs at present operate in specific geographical
locations. The need to access the pipeline at both ends for the purpose of recommissioning (i.e. flooding,
cleaning, dewatering, etc.), is inherent in many of the repair scenarios. Access facilities and the provision of
adequate space for equipment (particularly dewatering) are significant.
8.
CONCLUSIONS
Technology challenges for flexible risers are indispensable for floating oil and gas production systems.
Several hybrid riser concepts have been introduced, combining proven rigid and flexible riser technologies.
Several installations have been performed in benign environments whereas the major design challenge in
harsh environments is related to fatigue during installation.
Pipeline technology has experienced revolutionary advances in the design process through analysis.
Several new pipeline concepts have been introduced with using numerical analysis, analysis tools, modern
materials and revised design codes. Further challenging pipeline concepts for ultra deep waters will be
expected with using them.
Evaluation of fatigue and extreme response is crucially important to ensure safety and serviceability of
risers. Although the dynamic response is a rather matured topic, breakthrough developments are rarely
seen, but efforts are continued to develop the analysis method and improve accuracy of analysis. Dynamic
response under combined action of wind, wave, current and other relevant dynamic loads such as slug
flow is an example of effort.
An important challenge in this area of research is non-linearity. Non-linearity makes perspective of
phenomena worse and interpretation more difficult. A typical example is VIV. This is partly the reason
why the large efforts are made to the scientific investigation of the phenomenon. On the other hand
development of analysis codes used in design are continued in responding to the needs of design.
Development of VIV suppression devices are continued as before.
894
In the past decades the computer has shown dramatic developments in performance and capacity, but high
computational load is still a main challenge for numerical simulation of hydrodynamic phenomena and
hydroelastic response.
The statement on soil-pipe interaction in ISSC2000 was limited to the pipeline scoring. Since then, a large
progress has been accomplished on the soil-pipe interaction, and issues such as pipeline as-laid embedment, pipe-soil interaction and pipeline stability have been stated in this report. Lateral buckling and pipeline walking are the main themes of subsea pipeline behavior in deepwater. SAFEBUCK Joint Industry
Project (JIP) which was initiated in 2002 has led to new design guidelines for on-bottom lateral buckling
and development of simple models to simulate this behavior. However, challenges are needed to develop
better models because the simple models have been proposed under some assumptions.
A model for axial soil-pipe interaction using the critical state theory and a simpler model for the axial
resistance prediction have been proposed, which are based on a number of simplified assumptions. The
development of more sophisticated models than these simple models is requested to be applicable to more
complex situations that involve buckling and axial displacement.
Pipeline trenching was historically performed using jetting, but mechanical ploughing or cutting has become an increasingly common approach. Current researches on pipeline stability during trenching and backfilling are mainly on 2-D aspects of the problem and only on soil mechanics. However, uplift propagation
during trenching and backfilling is a 3-D problem and involves pipeline structural response, soil transport
and deposition, and consolidation. All of these aspects need to be investigated integrally in the future.
The physical modeling of wave-induced liquefaction and sediment transport has been carried out in the
University of Western Australia, but the investigation of wave-induced liquefaction and sediment transport is still ongoing. The future research will focus on better defining and quantifying the erosion mechanisms of sediments, particularly those which exhibit behavior described as cohesive, as well as clarifying the nature of the frictional seabed shear stress required to drive erosion.
A new joint industry project (JIP) is now being proposed which will aim at identifying state-of-the-art
design methodologies for mechanically stabilizing pipelines and thereby controlling movement.
The number of potential failure modes for steel pipes is smaller than that for flexible pipes. Most probable
failure modes for steel pipes are buckling, collapse, fatigue, corrosion, stress corrosion cracking and erosion.
Regarding the buckling of steel pipes, FEM has been widely used and will be more important in the future.
Identification of all relevant failure modes and loading mechanisms is essential to ensure the structural
integrity of flexible risers. The most frequent failures for flexible pipes are related to outer sheath damage,
carcass failure and aging/abrasion/wear. Flexible risers in deep waters are exposed to high pressure and
temperature and corrosive environments. Design criteria for all known failure modes are given in (API 2008)
which is the main design code for flexible risers. However, new failures modes and loading mechanisms
have to be identified with experiments and analyses, with the development of fibre-optic monitoring systems.
With the increase of water depths, failures modes and loading mechanisms would become more complex
and the FE analysis models considering the geometrical, material and boundary non-linearities would be
indispensable.
Ultra deepwater field development pushed the oil industry to pursue frontiers of the technology to develop
innovative installation procedures for riser systems, observing risk involved and cost reduction pressures.
The experimental, analytical and numerically analytical tools seem to be getting matured and their utilization
has been enabling to develop innovative conceptual installation procedures for riser systems including the
transportation procedure of riser system and components from the land to offshore. What is expected in the
years ahead is to develop more reliable and cost-effective installation procedures for ultra deepwater riser
systems and improve their accuracy based on the comparison of the procedures with the data obtained from
the monitoring program for the in situ riser system installation.
Pipeline installation gives the most critical time to its strength. Accordingly, comprehensive installation
analyses are carried out before the installation using FEM-based software tools. One of recent JIPs is
concerned with dynamic installation analysis to investigate dynamic amplifications factors, side current etc.
and is ongoing. An important challenge is to give new pipe lay barges multipurpose, multiwater depth
abilities.
As evidenced by the recent research literature, these fifteen years have given a great progress to the field of
maintenance, inspection and repair of risers and the technologies for them are almost mature. Integrity
895
management programs that determine maintenance and inspection requirements for a riser life of 20 or more
years are developed during detailed design, based on a risk assessment to identify degradation threats and
methods of mitigation. The operating parameters that are monitored, methods of inspection and associated
frequency of inspection are then defined according to the expected severity of operating conditions and
response for respective riser systems. Fatigue is a key design driver for riser systems and deployments of
monitoring systems for the prediction of wave-, current- and vessel motion-induced fatigue are successful in
recent years. However, more efforts are desired to improve such monitoring systems especially effective for
VIV-induced fatigue prediction. Maintenance, inspection and repair of deepwater and ultra deepwater
pipelines are challenging. Inside inspection of deepwater pipelines is executed with pigs. However, if the
inspection run between pig traps becomes even longer, then that cost and reliability will be future issues to
address. Their external inspection is performed with ROVs and AUVs. In deeper waters, AUVs provide
dramatic improvements in cost, performance, safety and reliability compared to ROVs. More sophistication
of AUV inspection will be a future issue.
REFERENCES
4subsea. 2013. PSA-NorwayUn-bonded Flexible Risers - Recent Field Experience and Actions for Increased Robustness.
ABS. 2014. Guide for building and classing subsea pipeline systems. Houston: American Bureau of Shipping.
Adib-Ramezani, H., Jeong, J. & Pluvinage, G. 2006. Structural integrity evaluation of X52 gas pipes subjected to
external corrosion defects using the SINTAP procedure. International Journal of Pressure Vessel and Pipeline
83:42032.
Allen, D.W. & Liapis, S. 2014. The Effect of Coverage Length and Density on the Performance of Helical Strakes.
Paper No. 23107, Proc. OMAE, June 2014.
Alliot, V. and Legras, J. L. 2005. Lessons Learned From The Evolution And Development of Multiple-Lines Hybrid
Riser Towers For Deep Water Production Applications. Proc. OTC, May 2005.
Anserson, C. 2005. Pipeline inspection. Aberdeen: HydroFest. (Available at: www.ths.org.uk)
API. 2008. API 17J Specification for unbonded flexible pipe.
API. 2007. Specification for line pipe. Washington: American Petroleum Institute.
ASME B31G. 2009. Manual for determining the remaining strength of corroded pipelines. A supplement to ASME
B31G Code for Pressure Piping. New York:USA.
Assi, G.R.S., Srinil, N., Freire, C.M. & Korkischko, I. 2012. Experimental Investigation of the Vortex-Induced
Vibration of a Curved Cylinder. Paper No. 83016, Proc. OMAE, June 2012.
Assi, G.R.S., Bearman, P.W., Carmo, B.S., Meneghini, J.R., Sherwin, S.J. & Willden, R.H.J. 2013. The Role of Wake
Stifness on the Wake-induced Vibration of the Downstream Cylinder of a Tandem Pair. Journal of Fluid Mechanics.
718. 210245. Cambridge.
Aubeny, C. P., Shi, H. & Murff, J.D. 2005. Collapse Loads for a Cylinder Embedded in Trench in Cohesive Soil. Paper
No. 19493, Proc. OTC, May 2005.
Auperin, M., et al. 2005. Lessons Learned From Development And Installation of Injection Single Hybrid Risers
Application to Production, Test Pipe-in-Pipe SHRs. Proc. OTC, May 2005.
Baarholm, G. S. & S. Haver 2010. Application of environmental contour lines on a flexible riser. Paper No. 20860,
Proc. OMAE, June 2010.
Bai, Y. & Bai, Q. 2005. Subsea pipelines and risers. MA: Elsevier Science.
Bai, Y. & Bai, Q. 2014. Subsea pipelines integrity and risk management. MA: Elsevier Science.
Bai, Q. & Duffy, B. 2014. Analytical Approach for Mechanical Responses of Spanning Pipelines. Paper No. 24524,
Bakhtiary et al. 2007; Analysis of Offshore Pipeline Allowable Free Span Length; International Journal of Civil
Engineerng. Vol. 5, No. 1
Bearman, P.W. 2011. Circular Cylinder Wakes and Vortex-Induced Vibrations, Journal of Fluids and Structures.
27 (56). 648658. Elsevier.
Belfroid, S. P. C., et al. 2011. Application of a new carcass design for prevention of singing behaviour in flexible
risers. American Society of Mechanical Engineers, Pressure Vessels and Piping Division (Publication) PVP.
Berge, S., et al. 2009. Environmental effects on fatigue strength of armour wire for flexible risers. Paper No. 80262,
Borsellino, C., Calabrese, L., Valenza, A. 2004. Experimental and Numerical Evaluation of Sandwich Composite
Structures. Composites Science and Technology. 64(1011).1709. Elsevier Ltd.
Boschee, P. 2012. Best Practices for Flexible Pipe Integrity Evolve. SPE Oil and Gas Facilities Magazine, February
2012.
Bourguet, R. Lucor, D. & Triantafyllou, M.S. 2012. Mono-and Multi-Frequency Vortex-Induced Vibrations of a Long
Tensioned Beam in Shear Flow. Journal of Fluids and Structures. 32. 52-64.
Bourguet, R. Karniadakis, G.E. & Triantafyllou, M.S. 2013a. Multi-Frequency Vortex-Induced Vibrations of a Long
Tensioned Beam in Linear and Exponential Shear Flows. Journal of Fluids and Structures, Vol. 41, pp. 3342.
Bourguet, R. Karniadakis, G.E. & Triantafyllou, M.S. 2013b. Distributed Lock-In Drives Broadband Vortex-induced
Vibrations of a Long Flexible Cylinder in Shear Flow. Journal of Fluid Mechanics. 717. 361375.
896
Bransby, M.F., Yun, G.J., Morrow, D.R., & Brunning, P. 2005. The performance of pipeline ploughs in layered soils.
Frontiers in Offshore Geotechnics I.
Bruton, D.A.S., White, D.J., Marr, M. & Cheuk, C.Y. 2008. Pipe-Soil Interaction during Lateral Buckling and Pipeline
Walking: the SAFEBUCK JIP. Paper No. 19589, Proc. OTC, May 2008.
Caire M. & Schiller R.V. 2013. The effect of wave energy spreading on the mooring and riser system response of a
spread-moored FPSO. Paper No. 10675, Proc. OMAE, June 2013.
Campello, G., et al. 2012. A novel concept of flexible pipe end fitting: Tensile armor foldless assembly. Paper No.
83511, Proc. OMAE, June 2012.
Chaplin, J.R., Bearman, P.W., Cheng, Y., Fontaine, E., Graham, J.M.R., Herfjord, K., Huera Huarte, F.J., Isherwood,
M., Lambrakos, K., Larsen, C.M., Meneghini, J.R., Moe, G., Pattenden, R.J., Triantafyllou, M.S. & Wilden, R.H.J.
2005. Blind Predictions of Laboratory Measurements of Vortex-induced Vibrations of a Tension Riser, Journal of
Fluids and Structures (21): 2540. Elsevier.
Carmo, B.S., Assi, G.R.S. & Meneghini, J.R. 2013. Computational Simulation of the Flow-Induced Vibration of a
Circular Cylinder Subjected to Wake Interference. Journal of Fluids and Structures. 41, 99108. Elsevier.
Castello, X. & Estefen, S.F. 2008. Sandwich Pipes for Ultra Deepwater Applications. Paper No. 19704, Proc. OTC,
May 2008.
Cathie, D.N., Machin, J. & Overy, R.F. 1996. Engineering appraisal of pipeline floatation during backfilling. Paper No.
8136, Proc. OTC, May 1996.
Cathie, D.N., Barras, S. & Machin, J. 1998. Backfilling pipelines: state of the art. 21st offshore pipeline technology
conference, Oslo.
Cathie, D.N., & Wintgens, J.F. 2001. Pipeline trenching using plows: performance and geotechnical hazards. Paper No.
13145, Proc. OTC, May 2001.
Cathie, D.N., Jaeck, C., Ballard, J.-C. & Wintgens, J.F. 2005. Pipeline geotechnicsstate-of-art. Frontiers in Offshore
Geotechnics I. 95113.
Chen, E., Mclaury, B.S. & Shirazi, S.A. 2004. Application and Experimental validation of a computational fluid
dynamics (CFD)-based erosion prediction model in elbows and pluged tees. Computers & Fluids. 33: 1251
1272.
Chen, H.C., Chen, C.R. & Huang, K. 2013. CFD Simulation of Vortex-induced and Wake-induced Vibrations of Dual
Vertical Risers. Proc. ISOPE, June-July 2013.
Cheuk, C.Y., White, D.J. & Bolton, M.D. 2007. Large-scale modelling of soil-pipe interaction during large amplitude
cyclic movements of partially embedded pipelines. Canadian Geotechnical Journal, Vol. 44, 977996.
Cheuk, C.Y. & White, D.J. 2008. Centrifuge modeling of pipe penetration due to dynamic lay effects. Paper No.
Cheuk, C.Y. & White, D.J. 2011. Modelling the Dynamic Embedment of Seabed Pipelines. Geotechnique 61(1): 39
57.
Chiodo, M.S.G. & Ruggieri, C. 2009. Failure Assessments of Corroded Pipelines with Axial Defects using Stressbased Criteria: Numerical Studies and Verification Analyses. Int. J. of Pressure Vessels and Piping. 86(2
3):164176.
Cicolin, M.M. & Assi, G.R.S. 2014. Response of a Circular Cylinder Fitted With Permeable Meshes Acting as VIV
Suppressors. Paper No. 24709, Proc. OMAE, June 2014.
Constantinides, Y., Chen, J., Tran, L., Enuganti, P. & Campbell, M. 2011. Comparison of SCR Field Response with
Analytical Predictions. Paper No. 50150, Proc. OMAE, July 2011.
Cook, H., Dopjera, D, Thethi, R & Williams, L. 2006. Riser Integrity Management for Deepwater Developments.
Paper No. 17891, Proc. OTC, May 2006.
Cosham, A. & Hopkins, P. 2004. An Overview of the Pipeline Defect Assessment Manual (PDAM). 4th International Pipeline Technology Conference, May 913, Ostende: Belgium.
Dawood and Kenny 2013. Finite element method simulating the pipeline mechanical response during reel-lay simulation. Paper No. 10555, Proc. OMAE, June 2013.
de Araujo, J.B., de Almeida, J.C.L. & Fernandes, A.C. 2011. Installation of a Submerged Buoy for Supporting Risers (BSR) System in Campos Basin Site. Paper No. 50167, Proc. OMAE, June 2011.
de Sousa F.J.M., Rebello A.M. & Sagrilo L.V.S., 2013. An alternative methodology to evaluate the fatigue life of risers
based on regular waves. Paper No. 10914, Proc. OMAE, June 2013.
de Sousa, J. R. M., et al. 2013. A FE model to predict the stress concentration factors in the tensile armor wires of
flexible pipes inside end fittings. Paper No. 10995, Proc. OMAE, June 2013.
de Sousa, J. R. M., et al. 2012. Fatigue analysis of a 6" flexible pipe with broken tensile armor wires. Paper No.
83532, Proc. OMAE, July 2012.
de Sousa, J. R. M., et al. 2012. An Experimental and numerical study on the axial compression response of flexible
pipes. Journal of Offshore Mechanics and Arctic Engineering 134(3).
Dingle, H.R.C, White, D.J. & Gudin, C. 2008. Mechanisms of pipeline embedment and lateral breakout in soft clay.
Canadian Geotechnical Journal 45(5): 636-652.
DNV-OS-F101. 2007. Submarine pipeline system. DNV.
DNV-RP-C203. 2005. Recommended practice for fatigue design of offshore steel structures. DNV.
DNV-RP-F101, 2004. Recommended Practice-Corroded Pipelines. DNV.
DNV-RP-F108. 2006. Fracture control for pipeline installation methods introducting cyclic plastic strain. DNV.
DNV-RP-F116. 2010. Intergrity management of submarine pipeline. DNV.
897
dos Santos, M., et al. 2014. A Comparitive Study of a Free Span Pipeline Through Numerical Simulations. Paper
No. 24069, Proc. OMAE, June 2014.
Duan, G., et al. 2014, Finite Element Analysis of global walking with spanning and lateral buckling. Paper No.
el Mouhandiz, A. & Troost, S. 2013. Subsea Templates Installation at North Sea using HLV THIALF. Proc.
ISOPE, June-July 2013.
Epsztein, T., et al. 2011. Towards an H2S free environment in flexible pipe annulus thanks to a new Anti-H2S
polymer layer. Paper No. 49664, Proc. OMAE, June 2011.
Estefen, S.F. 1999. Collapse Behaviour of Intact and Damaged Deepwater Pipelines and the Influence of the Reeling Method of Installation. Journal of Constructional Steel Research. 50(2): 99114.
Farnes, K. A., et al. 2013. Carcass failures in multilayer PVDF risers. Paper No. 10210, Proc. OMAE, June 2013.
Fernandes, A. C. J. P., Vardaro, E., Franciss, R. & Almeida, J.C. 2003. Sub-surface buoy hybrid riser system for
deepwater application. Paper No. 37412, Proc. OMAE, June 2003.
Fontaine, E., Marcollo, H., Vandiver, K., Triantafyllou, M., Larsen, C. Tognarelli, M., Constantinides, Y. & Oakley, O. 2011. Reliability Based Factors of Safety for VIV Fatigue Using NDP Riser High Mode VIV Tests. Paper No.49820, Proc. OMAE, June 2011.
Francis, K. 2011. Fatigue analysis of flexible risers due to bimodal waves. Proc. ISOPE, June 2011.
Franciss, R., Gonzales, E.C., de Almeida, J.C.L., de Arajo, J.B. & Fernandes, A.C. 2011. Alternative Method of
Buoy Supporting Riser (BSR) Installation. Paper No. 49862, Proc. OMAE, June 2011.
Franzini, G.R., Gonalves, R.T., Meneghini, J.R. & Fujarra, A.L.C. 2014. Experimental Investigation into the Flow
around a Stationary and Yawed Cylinder under Asymmetrical End Conditions. Proc. ISOPE, June 2014.
Fu, B. & Batte, A. 1999. An overview of advanced methods for the assessment of corrosion in linepipe. Offshore
Technology Report-OTO 1999 051.
Fu, S., Wang, J., Baarholm, R. Wu, J. & Larsen, C.M. 2013. VIV of Flexible Cylinder in Oscillatory Flow. Paper No.
Gabor, F. & Laszlo, V. 2012. The effect of the width to length ratios of corrosion defects on the burst pressures of
transmission pipelines. Engineering Failure Analysis 21: 2130.
Gong, S.F. & Li, G. 2015. Buckle Propagation of Pipe-in-pipe Systems under External Pressure. Engineering Structures. 84: 207222.
Gong, S.F., Sun, B., Bao, S. & Bai, Y. 2012. Buckle Propagation of Offshore Pipelines under External Pressure.
Marine Structures. 29(1): 115130.
Gong, S.F., Ni, X.Y., Bao, S. & Bai, Y. 2013. Asymmetric Collapse of Offshore Pipelines Under External Pressure.
Ship and Offshore Structures. 8(2): 176188.
Gueveneux, H. & Le Buhan, P. 2014. Hybrid riser tower design: Evolution, operational efficiency and compliance
to ultra deep offshore challenge. Paper No. 23882, Proc. OMAE, June 2014.
Gundersen P., Doynov K. & Andersen T. 2012. Methodology for determining remnant fatigue life of flexible risers
subjected to slugging and irregular waves. Paper No. 83412, Proc. OMAE, June 2012.
Hagatun, K. S., O.; Lie, H. 2010. Hybrid Riser Solutions for Harsh Environments. DOT 2010.
Hanonge, D. & Luppi A. 2010. Challenges of flexible riser systems in shallow waters. Paper No. 20578, Proc.
OTC, May 2010.
He, J., Vasblom, W.J., & Miedema, S.A. 2005. FEM analyses of cutting anisotropic densely compacted and saturated
sand. Western Dredging Association Conference, New Orleans, USA.
He J.W. & Low Y.M. 2010. Probabilistic assessment of the clashing between flexible marine risers. Paper No. 20046,
He J.W. & Low Y.M. 2013. Predicting the probability of riser collision under stochastic excitation and multiple
uncertainties. Paper No. 10381, Proc. OMAE, June 2013.
Henning, M., Scott, D. & White, D. 2013. Free field sediment mobility on Austrians north west shelf. Paper No.
Hill, A.J., White, D.J., Bruton, D.A.S., Langford, T., Meyer, V., Jewell, R. & Ballard, J-C. 2012. A new framework for
axial pipe-soil resistance, illustrated by a range of marine clay datasets. Proc. International Conference on Offshore
site investigation and geotechnics. SUT, London.
Hirabayashi, S. & Suzuki, H. 2013. Numerical Study on Vortex Induced Motion of Floating Body by Lattice
Boltzmann Method. Paper No. 11045, Proc. OMAE, June 2013.
Huang, Z. & Larsen, C.M. 2011. Numerical Simulation on Vortex-Induced Vibration of an Elastically Mounted
Circular Cylinder with Two Degrees-of-Freedom. Paper No. 49573, Proc. OMAE, June 2011.
Iranpour, M. & Taheri, R. 2006. The State-of-the-Art Review of Risers VIV Fatigue. Proc. OMAE, June 2006.
Jhingran, V., Zhang, H., Lie, H., Braaten, H. & Vandiver, J.K. 2012. Buoyancy Spacing Implications for Fatigue
Damage Due to Vortex-Induced Vibrations on a Steel Lazy Wave Riser (SLWR). Paper No. 23672, Proc. OTC,
May 2012.
Ji, G. L., B.J., Svik, S., Klbo, F., Axelsson, G. & Fergestad, D. 2014. Integrity assessment of damaged flexible pipe
cross-sections. Paper No. 24155, Proc. OMAE, June 2014.
J P Kenny BSEE 2007. Deepwater Riser Design, Fatigue Life and Standards Study. Project. Report for MMS, Project
No. 572.
Kang, Z., Liu, W. & Qin, W. 2013. Vortex-Induced Vibration Experiment Research of Two Cylinders in Tandem
Arrangement. Paper No. 10471, Proc. OMAE, June 2013.
898
Karjadi et al. 2013. Development on Aegir Reeling Pipeline Analyses Simulation by Test Validation. Paper No.
Kenny, M. 2010. Oil & Gas UK - State of the art report on flexible pipe integrity. Aberdeen, MCS Kenny.
Kiefner, J. & Vieth, P. 1989. A modified criterion for evaluating the strenght of corroded pipe. Final report for PR
3-805 project to the Pipeline Supervisory Committee of the American Gas Association, Battelle. Ohio: UK.
Kim, D.W., Mohd, H.M., Lee, B.J., Kim, D.K., Seo, J.K., Kim, B.J. & Paik, J.K. 2013. Investigation on the burst
strength capacity of aging subsea gas pipeline. Paper No. 10267, Proc. OMAE, June 2013.
King, R., Brown, A., Braaten, H., Russo, M., Baarholm, R. & Lie, H. 2013. Suppressing Full-Scale Riser VIV with the
VT Suppressor. Paper No. 11642, Proc. OMAE, June 2013.
Klever, F.J., Stewart, G., & Valk, C.A.C. 1995. New developments in burst strength predictions for locally corrodes
pipelines. Proc. OMAE, June 1995, 5:161173.
Korkischko, I. & Meneghini, R.R. 2012. Suppression of Vortex-Induced Vibration Using Moving Surface BoundaryLayer Control. Journal of Fluids and Structures. 34. 259270. Springer.
Kristensen, C. E. M., J., Skeie, G., Skjerve, H. & Sdahl, N. 2014. Carcass tear out load model for multi-layer pressure sheath risers. Paper No. 24129, Proc. OMAE, June 2014.
Kristiansen, U. R. & G. A. Wiik 2007. Experiments on sound generation in corrugated pipes with flow. Journal of
the Acoustical Society of America 121(3): 13371344.
Kyriakides, S. 2002. Buckle Propagation in Pipe-in-pipe Systems.: Part I. Experiments. International Journal of
Solids and Structures. 39(2): 351366.
Kyriakides, S. & Corona, E. 2007. Mechanics of offshore pipelines-Volume1: buckling and collapse, 1st Ed. MA:
Elsevier Science.
Kyriakides, S. & Netto, T.A. 2004. On the Dynamic Propagation and Arrest of Buckles in Pipe-in-pipe Systems. Int
J Solids Struct. 41 (20): 54635482.
Kyriakides, S. & Vogler, T.J. 2002. Buckle Propagation in Pipe-in-pipe Systems. Part II: Analysis. Int. J. Solids
Struct. 39 (2): 367392.
Lambert, A., et al. 2012. Qualification of unbonded dynamic riser with carbon fiber composite armours. Paper No.
Langford, T.E., Dyvik,R. & Cleave, R. 2007. Offshore pipeline and riser geotechnical model testing: practice and
interpretation. Paper No. 29458, Proc. OMAE, June 2007.
Larsen, C.M., Vikestad, K., Yttervik, R. & Passano, E. 2000. VIVANA Theory Manual. MARINTEK. Trondheim,
Norway.
Lauder, K., Bransby, F., Brown, M., Cathie, D., Morgan, N., Pyrah, J. & Steward, J. 2008. Experimental testing of the
performance of pipeline ploughs. Proc. ISOPE, July 2008, 212217.
Lauder, K. 2010. Perdition pipeline plough performance by scale model testing. Presentation to CTC, University of
Dundee.
Lee, L.H., Kyriakides, S., Netto, T.A. 2008. Integral Buckle Arrestors for Offshore Pipelines: Enhanced Design
Criteria. International Journal of Mechanical Sciences. 50(6): 10581064.
Lee, L.H. & Kyriakides, S. 2004. On the Arresting Efficiency of Slip-on Buckle Arrestors for Offshore Pipelines.
International Journal of Mechanical Sciences. 46(7): 10351055.
Lee, S.E., Paik, J.K., Ha, Y.C., Kim, B.J., & Seo, J.K. 2014. An Efficient Design Methodology for Subsea Manifold
Piping Systems based on Parametric Studies. Ocean Engineering. 84:273282.
Legras, J. & Pillet, B. 2013. Tethered Catenary Riser: A New Deepwater Concept. Paper No. 11102, Proc. OMAE,
June 2013.
Lie, H., et al. 2007. Free-span VIV testing of full-scale umbilical. Proc. ISOPE, July 2007.
Lie, H., Braaten, H. and Kristiansen, T. & Nielsen, F.G. 2007. Free-Span VIV Testing of Full-Scale Umbilical.
Proc. ISOPE, July 2007.
Lie, H., Braaten, H., Jhingran, V.G., Sequeiros, O. & Vandiver, J.K. 2012. Comprehensive Riser VIV Model Tests in
Uniform and Sheared Flow. Paper No. 84055, Proc. OMAE, July 2012.
Lie, H., Braaten, H., Szwalek, J., Russo, M. & Baarholm, R. 2013. Drilling Riser VIV Tests with Prototype Reynolds
Numbers. Paper No. 11643, Proc. OMAE, June 2013.
Lourenco, M.I., Pasqualino, I.P. & Palermo, T.D.S. 2008. Core Material Performance on the Propagation Pressure
of Sandwich Pipes. Paper No. 57507, Proc. OMAE, June 2008.
Low Y.M. & Cheung S.H. 2012. Evaluating the long-term fatigue response of mooring lines using an asymptotic
approximation. Paper No. 83612, Proc. OMAE, July 2012.
Lund, K.M. 2000. Effect of Increase in Pipeline Soil Penetration from Installation. Paper No. PIPE-5047, Proc. OMAE,
February 2000.
Luppi, A., et al. 2014. Deep water hybrid riser systems. Proc. OTC, May 2014.
Manelli, G., & Radicioni, A. 1994. Deepwater Pipeline Repair Technology: A Gernal Overview. Proc. OMAE,
1994.
Marchionni et al. 2011. Offshore Pipeline Installation - 3-Dimensional Finite Element Modelling. Paper No. 49832,
Marintek. 2014. Handbook on Design and Operation of Flexible Pipes, Marintek Report, MT 2014-F-001.
Martin, C.M. 2013. SAFEBUCK GEO Joint Industry Project-Task 4: User Element Development. University of
Oxford (Confidential to JIP Participants)
Martins M.A.L., Dria A.S. & Silverira E.S.S. 2009. Long-term fatigue and extreme design of steel risers. Paper No.
79272, Proc. OMAE, May-June 2009.
899
Maschner, E.A. & Abdalla, B. 2012. On pipeline lateral buckling: lessons learned and current design challenges. Paper
No. 83413, Proc. OMAE, July 2012.
McStay, D., McIlroy, J., Forte, A., Lunney, F., Greenway, T., Thabeth, K. & Dean, G. 2005. A new tool for the
rapid remote detection of leaks from subsea pipelines during remotely operated vehicle inspections. Journal of
Optics A: Pure and Applied Optics, 7(6): S346-S351.
Mello, V. C., et al. 2011. Roncador P-52 Hybrid Riser Review: Key Concepts, Construction and Installation Challenges. Proc. OTC, May 2011.
Merifield, R.S., White, D.J. & Randolph, M.F. 2008. The Ultimate Undrained Resistance of Partially Embedded
Pipelines. Geotechnique 58(6): 461470.
Miyamura, A., Hirayabashi, S. & Suzuki, H. 2014. Numerical Simulation of Vortex-Induced Motion with Free
Surface by Lattice Boltzmann Method. Paper No. 24302, Proc. OMAE, June 2014.
Modarres-Sadeghi, Y., Bourguet, R., Triantafyllou, M.S., Tognarelli, M. & Beynet, P. 2010. Re-Evaluation of VIV
Riser Fatigue Damage. Paper No. 21043, Proc. OMAE, June 2010.
Mohd, M.H., Kim, D.K., Kim, D.W., & Paik, J.K. 2014. A time-variant corrosion wastage model for subsea gas
pipelines. Ships and offshore structures. 9(2):161176.
Mohd, M.H. & Paik, J.K. 2013. Investigation of the Corrosion Progress Characteristics of Offshore Subsea Oil Well
Tubes. Corrosion Science. 67:130141.
Morooka, C.K. & Tsukada, R.I. 2013. Experiments with a Steel Catenary Riser Model in a Towing Tank, Applied
Ocean Research. 43. 244255. Elsevier
Mouring, S.E. 1999. Buckling and Postbuckling of Composite Ship Panels Stiffened with Preform Frames. Journal
of Ocean Engineering. 26.793803.
Munns, I., Crouzen, P., Rea, T, Hunt, R. & Reid, D. 2007. Inspection of Offshore Platform Well Surface Casings
and Conductors. SPE Applied Technology Workshop on Well Integity. Barcelona, November 2007.
Murff, J.D., Wagner, D.A. & Randolph, M.F. 1989. Pipeline Penetration in Cohesive Soil. Geotechnique 39(2): 213
229.
Netto, T.A. & Estefen, S.F. 1996. Buckle Arrestors for Deepwater Pipelines. International Journal of Marine Structures. 9: 873883.
Netto, T.A., Ferraz, U.S., & Estefen, S.F. 2005. The Effect of Corrosion Defects on the Burst Pressure of Pipelines.
J. of Constructional Steel Research, 61(8): 11851204.
Netto, T.A. & Kyriakides, S. 2000a. Dynamic Performance of Integral Buckle Arrestors for Offshore Pipelines. Part
I: Experiments. International Journal of Mechanical Sciences. 42:14051423.
Netto, T.A. & Kyriakides, S. 2000b. Dynamic Performance of Integral Buckle Arrestors for Offshore Pipelines. Part
II: Analysis. International Journal of Mechanical Sciences. 42:14251452.
Nozawa, T., Suzuki, H., Ohta, M. 2010. Numerical Prediction and Suppression of VIV of Deepwater Riser. Paper
Nyhus, B., Ostby, E., Knagenhjelm, H.O., Balck, S. & Rostadsand, P.A. 2005. Fracture control offshore piplines:
experimental studies on the effect of crack depth and asymmetrical geometries on the ductile tearing resistance.
Oh, C.K., Kim, Y.J., Baek, J.H., Kim, Y.P., & Kim, W.S. 2007. Ductile Failure Analysis of API X65 Pipes with
Notch-type Defects using a Local Fracture Criterion. Int. J. of Pressure Vessels and Piping. 84(8): 512525.
Olso, E. & Kyriakides, S. 2003. Internal Ring Buckle Arrestors for Pipe-in-pipe Systems. International Journal of
Non-Linear Mechanics. 38(2): 267284.
Ortega, A., et al. 2012. On the dynamic response of flexible risers caused by internal slug flow. Paper No. 83316,
Ortega A., Rivera A. & Larsen C.M. 2013. Flexible riser response induced by combined slug flow and wave loads.
Paik, J.K. & Thayamballi, A.K. 2007. Ship-shaped offshore installations design, building and operation. Cambridge: Cambridge University Press.
Palmer, A.C.,Kenny, J.P., Perera, M.R. & Reece, A.R. 1979. Design and operation of an underwater pipeline trenching
plough. Gotechnique 29(3):305322.
Palmer, A.C. 1999. Speed effects in cutting and ploughing. Gotechnique 49(3):285294.
Palmer, A.C. & King, R.A. 2004. Subsea pipeline engineering. PennWell Corporation. Oklahoma: USA.
PARLOC. 2001. The update of loss of containment data for offshore pipelines. Mott MacDonald Ltd..
Pasqualino, I.P. & Estefen, S.F. 2001. A Nonlinear Analysis of the Buckle Propagation Problem in Deepwater Pipelines. International Journal of Solids and Structures. 38(4647): 84818502.
Passano E. & Larsen C.M. 2007. Estimating distributions for extreme response of a catenary riser. Paper No. 29547,
Passano E. & Maincon P. 2011. Estimating long-term distributions of extreme response of a catenary riser. Paper No.
Passano, E., Larsen, C.M., Wu, J. 2014. On the Prediction of Fatigue Damage from VIV. Paper No. 24217, Proc.
OMAE, June 2014.
Paumier, L. & Mesnage O. 2011. Psi armour wire for high collapse performance of flexible pipe. Paper No. 49257,
Peng, A.T.H. & Capart, H. 2008. Underwater sand bed erosion and internal jump formation by travelling plane jets.
Journal of fluid mechanics, 595:1-43.
900
Peng, W. & Bransby, M.F. 2011. Numerical modelling of soil around offshore pipeline plough shares. Frontiers in
Offshore Geotechnics II. 877882.
Perdrizet, T. L., J.M., Barbin, N., Le-Corre, V., Charliac, C. & Estrier, P. 2011. Stresses in armour layers of flexible
pipes: comparison of Abaqus models. 2011 SIMULIA Customer Conference. Barcelona, Spain.
Perdrizet T. & Averbuch D. 2008. Long term failure probability calculation method applied to riser design. Paper No.
Pereira, F.R., Gonalves, R.T., Pesce, C.P., Fujarra, A.L.C., Franzini, G.R. & Mendes, P. 2013. A Model Scale
Experimental Investigation on Vortex-Self Induced Vibrations (VSIV) of Catenary Risers. Paper No. 10447,
Pereira, P. S. D., et al. 2005. Experimental study on a self standing hybrid riser system throughout tests on a deepsea model basin. Proc. OMAE, June 2005.
Pereira, P. S. D., et al. 2006. Self standing hybrid riser system in waves and current. Proc. OMAE, June 2006.
Perez et al. 2014, Moonpool Effect Assessment During Structure Installation. Paper No. 23373, Proc. OMAE, June
2014.
PETRONAS. 2011. Pipeline inspection report. Petroliam Nasional Berhad (Petronas). Kuala Lumpur: Malaysia.
Pontaza, J.P. & Menon, R.G. 2013. Prediction of Vortex-induced Vibration Response of a Pipeline Span by Coupling a Viscous Flow Solver and a Beam Finite Element Solver. Journal of Offshore Mechanics and Arctic Engineering, 135 (3). ASME.
Powell, T.R., Fisher, R., Phillips, R. & Jee, T. 2002. Reducing backfilling risks. Proc. International Conference on
Offshore site investigation and geotechnicsDiversity and sustainability. SUT.
Rahmanian, M., Cheng, L., Zhao, M. & Zhou, T. 2012. Vortex-Induced Vibration of Two Side-by-Side Circular
Cylinders of Different Diameters in Close Proximity in Steady Flow. Paper No. 83223, Proc. OMAE, June 2012.
Ramasamy, R. & Ya, T.M.Y.S.T. 2014. Nonlinear Finite Element Analysis of Collapse and Post-collapse Behaviour
in Dented Submarine Pipelines. Applied Ocean Research. 46: 116123.
Ramasamy, R., Al Jaberi, M. & Al Junaibi, H. 2014. Ageing Offshore Well Structural Integrity Modelling, Assessment
and Rehabilitation. Paper No. SPE-171741, Proc. ADIPEC, October 2014.
Randolph, M & White, D.J. 2008a. Pipeline Embedment in Deep Water: Quantification and Processes. Paper No.
19128, Proc. OTC, May 2008.
Randolph, M & White, D.J. 2008b. Upper Bound Yield Envelopes for Pipeline at Shallow Embedment in Clay.
Geotechnique 58(4): 297-301.
Randolph, M. F., White, D. J. & Yan, Y. 2012. Modelling the axial soil resistance on deep-water pipelines.
Geotechnique 62, No. 9, 837846.
Rao, Z., Vandiver, J.K., Jhingran, V. & Sequeiros, O. 2012. The Effect of Exposure Length on Vortex Induced
Vibration of Flexible Cylinders. Paper No. 83273, Proc. OMAE, July 2012.
Raposo, C., et al. 2014. On the Challenges With Pipeline Free Spans in Operational Phase. Paper No. 23403, Proc.
OMAE, June 2014.
Reber, K & Boenisch, A. 2012. Advanced Electromagnetic Technology and Tools for Subsea Inspection. SubseaUK,
Subsea Pipelines Integrity Management Conference, London, December 2012,
Reece, A.R. & Grinsted, T.W. 1986. Soil mechanics of submarine ploughs. Paper No. 5341, Proc. OTC, May 1986.
Resvanis, T.L., Rao, Z. & Vandiver, J.K. 2014. Effects of Strake Coverage and Marine Growth on Flexible Cylinder VIV. Paper No. 24462, Proc. OMAE, June 2014.
Rytter, J., et al. 2002. A Novel Compression Armour Concept for Unbonded Flexible Pipes. Proc. OTC, May 2002.
SAFEBUCK III Safe Design of Pipelines with Lateral Buckling Design Guideline 30 August 2011.
Saint-Marcoux, J.F. & Blevins, R.D. 2012. Accounting for VIV in Wake Induced Motion of Risers In Tandem At
High Reynolds Number. Paper No. 22951, Proc. OTC, May 2012.
Saito, D., Cordeiro, L.R.S., Castro, F.A., Medeiros Jr., C.J. & Ribeiro, G.D. 2011. Innovative Supports of Risers
and Corresponding Installation Procedures. Paper No. 50011, Proc. OMAE, June 2011.
Sanaati, B. & Kato, N. 2013. Vortex-induced Vibration (VIV) Dynamics of a Tensioned Flexible Cylinder Subjected to
Uniform Cross-Flow. Journal of Marine Science and Technology. 18 (2). 247-261. Springer
Sanaati, B. & Kato, N. 2014. A Study on the Proximity Interference and Synchronization between Two Side-BySide Flexible Cylinders. Ocean Engineering. 85. 6579. Elsevier
Seaflex. 2007. Flexible PipesFailure Modes, Inspection, Testing and Monitoring. Project Report for Petroleum Safety
Authority, Norway, December 2007.
Secher, P., et al. 2011. Lateral buckling of armor wires in flexible pipes: Reaching 3000m water depth. Paper No.
Selden, R. A. 2009. Innovative Solution for Emergency Repair of a Deepwater Riser (Avoiding Flooding and ReCommissioning, Minimising Lost Production) A Case Study of Independence Hub Flex-Joint Repair. Paper No.
20154, Proc. OTC, 2009.
Seo, J.K., Cui, Y., Mohd, M.H., Ha, Y.C., Kim, B.J., & Paik, J.K. 2015. A risk-based inspection planning method
for corroded subsea pipelines. Ocean Enigneering (in press).
Sheehan J.M., Grealish F.W. & Smith R.J. 2005. Characterization of the wave environment in the fatigue analysis of
flexible risers. Proc. OMAE, June 2005.
Shen, W., Griffiths, T., Zan, Z., & Leggoe, J. 2015. Shear stess amplification around subsea pipelines: Part 3, 3D
study of spanning pipelines. Proceedings of the 7th International Conference on Scour and Erosion. London:
901
Shitamoto et al. 2014. Effect of Reeling Simulation on the Mechanical Properties of New Duplex Stainless Steel for
Line Pipe. Paper No. 23694, Proc. OMAE, June 2014.
Skeie, G., et al. 2012. Efficient Fatigue Analysis of Helix Elemenets in Umbilicals and Flexible Risers: Theory and
Applications. J. of Applied Mathematics(246812).
Smith, R., et al. 2007. Computational tool for the dynamic analysis of flexible risers incorporating bending hysteresis. Paper No. 29276, Proc. OMAE, June 2007.
Sparks, C. M., A., Rishoj, N., Deserts, L., Prinet, D., Hansen, N.O., Sele, A., Grealish, F., Riber, H., Boulliat, V.,
Chomard, A. & Cunff, C.L. 2003. RISER TECHNOLOGY STATE OF THE ART REPORT.
Srinil, N., Zangneh, H. & Day, A. 2013. Experimental Investigation of Two-Degree-Of-Freedom VIV of Circular
Cylinder with Low Equivalent Mass and Variable Natural Frequency Ratio. Paper No. 10239, Proc. OMAE, June
2013.
Sriskandarajah et al. 2013. Effect of Reeling Installation on Welding Residual Stress in Pipeline Girth Welds. Paper
Svik, S. 2011. Theoretical and experimental studies of stresses in flexible pipes. Computers and Structures 89:
22732291.
Svik, S. J., G. 2014. Differential equation for evaluating transverse buckling behaviour of tensile armour wires.
Sdahl, N., Steinkjer, O, Gjolmesli, E. & Hansen-Zahl, K. 2011. Consistent VIV Fatigue Analysis Methodology of
Umbilicals. Paper No. 49459, Proc. OMAE, June 2011.
Sokolinsky, V.S., Frostig, Y. & Nutt, S.R. 2002. Special Behavior of Unidirectional Sandwich Panels with Transversely Flexible Core under Statical Loading. International Journal of Non-Linear Mechanics.37. 869895.
Tan Z.M., Hou Y.C. & Zhang John, 2012. Irregular wave simulation and fatigue damage evaluation of a flexible
riser subjected to bi-modal sea states. Paper No. 83386, Proc. OMAE, July 2012.
Teixeira, A.P., Soares, C.S., Netto, T.A. & Estefen, S.F. 2008. Reliability of Pipelines with Corrosion Defects. International Journal of Pressure Vessels and Piping. 85(4): 228237.
Thethi, R., Howells, H., Natarajan, S. & Bridge, C. 2005. A Fatigue Monitoring Strategy & Implementation on a
Deepwater Top Tensioned Riser. Paper No. 17248, Proc. OTC, May 2005.
Thorsen, M.J. Svik, S. & Larsen, C.M. 2014. A Simplified Method For Time Domain Simulation of Cross-Flow
Vortex-Induced Vibrations. Journal of Fluids and Structures. 49 (Aug 2014): 135148. Elsevier.
Tognarelli, M., Taggart, S. & Campbell, M. 2008. Actual VIV Fatigue Response of Full Scale Drilling Risers: With
and Without Suppression Devices. Paper No. 57046, Proc OMAE, June 2008.
Toscano, R.G., Amenta, P.M., Dvorkin, E.N. 2002. Enhancement of the Collapse Resistance of Tubular Products
for Deep-water Pipeline Applications. Proceedings 25th Offshore Pipeline Technology Conference, Amsterdam,
Netherlands.
Toscano, R.G, Mantovano, L.O., Amenta, P.M., Charreau, R.F., Johnson, D.H., Assanelli, A.P., Dvorkin, E.N.
2008. Collapse Arrestors for Deepwater Pipelines. Cross-over mechanisms, Computers & Structures. 86 (78):
728743.
Toscano, R.G., Mantovano, L., Dvorkin, E.N. 2004. On the Numerical Calculation of Collapse and Collapse Propagation Pressure of Steel Deep Water Pipelines under External Pressure and Bending: Experimental Verification
of the Finite Element Results. Proceedings 4th International Conference on Pipeline Technology. Belgium.
Toscano, R.G., Timms, C., Dvorkin, E.N., DeGeer, D. 2003. Determination of the Collapse and Propagation Pressure of Ultra-deepwater Pipelines. Proc. OMAE, June 2003.
Tsukada, R.I. & Morooka, C.K. 2013. A Numerical Simulation Procedure for Vortex-induced Vibration of Pipelines with Free Span. Paper No. 10964, Proc. OMAE, June 2013.
Urthaler, Y., Cerkovnik, M., Yin, F., Saldana D. & Gordon R. 2013. An Investigation of the Tolerance of Riser Fatigue
to Corrosion Pitting. Paper No. 11015, Proc. OMAE, France, June 2013.
Vadel D. 2013. Interpretation of Subsea Inspection Data. Decom World. Subsea Integrity Management Summit,
Houston, September 2013.
Vanden Berghe, J.-F., Capart, H. & Su, J.C.C. 2008. Jet induced trenching operations: mechanisms involved. Paper No.
19441, Proc. OTC, May 2008.
Vandiver, J.K. & Li, L. 1999. SHEAR7 Program Theory Manual. Department of Ocean Engineering, MIT. Boston,
USA.
Vandiver, J.K. 2012. Damping Parameters for Flow-Induced Vibration. Journal of Fluids and Structures. 35. 105
119. Springer.
Venas, A. 2012. Taking deepwater pipelines to the X-Stream, DNV has a safe and cost-effective concept in the
pipline. Pipeline update 1-2012: DNV-GL publication.
Venkatesh, E.S. 1986. Erosion damage in oil and gas wells. SPE paper 15183, Rocky Mountain Regional Meeting of
the Society of Petroleum Engineers, Billings, MT.
Wanderley, J.B.V., Sphaier, S.H. & Levi, C. 2012. Two-Dimensional Numerical Investigation of the Hysteresis
Effect on Vortex Induced Vibration on an Elastically Mounted Rigid Cylinder. Journal of Offshore Mechanics
and Arctic Engineering. 134(2), 021801. ASME
Wang, A.M., Zhu, S., Zhu, X., Xu, J., He, M. & Zhang, C. 2013. Pendulous Installation Method and its Installation
Analysis for a Deepwater Manifold in South China Sea. Proc. ISOPE, June-July 2013.
Wang, J., Fu, S. & Baarholm, R. 2014. Vortex-Induced Vibration of Steel Catenary Riser under Vessel Motion.
Wang et al. 2011. Advanced FEA for Span Mitigation Analysis. Paper No. 50239, Proc. OMAE, June 2011.
902
Wang, D., White, D.J. & Randolph, M.F. 2010. Large-deformation finite element analysis of pipe penetration and
large-amplitude lateral displacement. Canadian Geotechnical Journal, Vol. 47, 842856.
Watanabe, Y., Suzuki, H. & Tamura, K. 2007. Optimization of a Fitting Structure of CFRP Riser Pipe for Deep Sea
Drilling. Paper No. JSN-464, Proc. ISOPE, July 2007.
Weppenaar, N. A., B. 2014. Investigation of tensile armor wire breaks in flexible risers and a method for detection.
Westgate, Z.J., White, D.J. & Randolph, M.F., 2009. Video Observations of Dynamic Embedment during Pipe laying
on Soft Clay. Paper No. 79814, Proc. OMAE, May-June 2009.
Westgate, Z.J., White, D.J., Randolph, M.F. & Brunning, P. 2010. Pipeline laying and embedment in soft fine-grained
soils: field observations and numerical simulations. Proc. OTC, May 2010.
White, D.J. & Dingle, H.R.C. 2011. The mechanism of steady friction between seabed pipelines and clay soils.
Geotechnique, Vol. 61, No. 12, 10351041,
White, D. J., Ganesan, S. A., Bolton, M. D., Bruton, D.A.S., Ballard, J-C. & Langford, T. 2011. SAFEBUCK JIPObservations of axial pipe-soil interaction from testing on soft nature clays. Proc. OTC, May 2011.
White, D.J. & Cathie, D.N. 2011. Geotechnics for subsea pipelines. Frontiers in Offshore Geotechnics II. 87113.
White, D.J., Campbell, M.E., Boylan, N.P. & Bransby, M.F. 2012. A new framework for axial pipe-soil resistance,
illustrated by shear box tests on carbonate soils. Proc. International Conference on Offshore site investigation and
geotechnics. SUT, London.
Williamson, C.H.K. & Govardhan, R. 2008 A Brief Review of Recent Results in Vortex-Induced Vibrations. Journal of Wind Engineering and Industrial Aerodynamics. 96. 713735. Elsevier.
Yeganeh et al. 2012. Determining Natural Frequency of Free Spanning Offshore Pipelines by Considering the Seabed Soil Characteristics. Journal of the Persian Gulf (Marine Science), Vol. 3, No. 8
Yew, C.C. 2013. SliPIPE-new pipeline expansion concept. Pipeline update No.1-2013: DNV-GL publication.
Zhang, Y. & Tan, Z. 2011. An Investigation of Fatigue Damage due to VIV in Flexible Riser with Different Methodologies. Paper No. 49036, Proc. OMAE, June 2011.
Zhao, M., Tong, R. & Cheng, L. 2012. Numerical Simulation of Two-Degree-Of-Freedom Vortex-Induced Vibration of a Circular Cylinder between Two Lateral Plane Walls in Steady Currents, Journal of Fluids Engineering.
134(10), 104501. ASME.
Zimmermann, C., Edwards, R., Talmont, P., Duley, D., de la Cruz, D. & Maroju, S. 2009. Recent Experience with a
Comprehensive Riser Tower Monitoring System. Proc. DOT, New Orleans, February 2009.
stergaard, N. H. 2012. On Lateral Buckling of Armouring Wires in Flexible Pipes. Aalborg, Denmark, Aalborg
University.
stergaard, N. H. et al. 2011. On lateral buckling failure of armour wires. Paper No. 49358, Proc. OMAE, June
2011.
OTC= The Offshore Technology Conference.
OMAE= The International Conference on Offshore Mechanics and Arctic Engineering.(2007)
The ASME International Conference on Offshore Mechanics and Arctic Engineering.(2008)
The ASME International Conference on Ocean, Offshore and Arctic Engineering.(2009)
ISOPE= The International Society of Offshore and Polar Engineers Conference.
904
Radon, M. 415
Rahman, T. 591
Ralph, F. 769, 1\07
Rcmcs, H. 619
Ren, Huilong 669
Ringsberg, J.W. 141
Riska, K. 769, 807
Rizzo, C. 351
Rizzuto, E. 415
Roland, F. 619
Romanoff, J. 351
Rrup, J. 351
Rychlik, 1. 1
Rye Anderson, M. 817
Saevik, S. 865
Samanta, A. 351
Schipperen, l. 619
Schreier, S. 591
Scku1ski, Z. 459
Seo, J.K. 865
Sharma, P. 591
Sheng, Wanan 669
Shin, Hyun-Kyoung 669
Sicchicro, M. 459
Sida:ri, M. 141
Sirkar, J. 769, 807
Author index
Skjong, R. 415
S1 ttc, .lohan 669
Smith, M. 279
Sodahl, N. 865
Song, K.H. 73
Stadic-Frohbocs, G. 865
Sun, L. 865
Suzuki, H. 865
Tabri, K. 519
Taczala, M. 141
Takaoka, Yukichi 669
Tammer, M. 81 7
Tang, W.Y. 519
Tao, Longbin 591
Teixeira, A. 415
Tcmarc1, P. 73
Terai, K. 769, 807
Theodoulidis, A. 351
Toyoda, M. 459
Tmelock, D. 723
Underwood, .I.M. 723
Va1koncn, .1. 769, 807
Valle, 0. 591
van der Sluijs, B.R. 141
Vaz, M.A. 209

Ventura, M. 459
Vhanmane, S. 279
Vi~jo, F. 723
Vredeveldt, A. 209
W gtcr, .1. 519
Wan, Z. 141
Wang, D. 279
Wang, F. 351
Wang, G. 351
Wang, S. 73
Watanabe, Y. 865
Wu, J. 865
Xiong, Y. 209
Yamada, Y. 519
Yasuda, A. 723
Yoshikawa, T. 279
Zakky, A. 591
Zamarin, A. 141
Zanic, V. 459
Zhan, D. 209
Zhan, Z. 591
Zhang, S. 279

Ship and Offshore Structures Congress PDF

Transféré par

Informations du document

Description originale:

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Ship and Offshore Structures Congress PDF

Transféré par

Droits d'auteur :

Formats disponibles

PROCEEDINGS OF THE 19TH INTERNATIONAL SHIP AND OFFSHORE

Proceedings of the 19th International

C. Guedes Soares & Y. Garbatov

ISBN set: 978-1-138-02895-1 (2 volumes hardback and CDROM)

4.1.4 Wind and waves

Report of Committee I.2: Loads

Report of Committee II.1: Quasi-static response

Risk-based inspection, maintenance and repair

Report of Committee II.2: Dynamic response

2.7.2 New sensors technology and application

Report of Committee III.1: Ultimate strength

Steel plate and stiffened plates

Report of Committee III.2: Fatigue and fracture

4.2.2 Strain concepts

Report of Committee IV.1: Design principles and criteria

Report of Committee IV.2: Design methods

Safety level of ship structures in ALS

7.4.6 Methods of controlling numerical instability for beam element model

Report of Committee V.2: Natural gas storage and transportation

Report of Committee V.3: Materials and fabrication technology

5.2.2 Zinc-rich paints

Report of Committee V.4: Offshore renewable energy

Report of Committee V.5: Naval vessel design

Ship service life in context

Report of Committee V.6: Arctic technology

Report of Committee V.6: Arctic technology annex

Brief offshore structures code summaries

Report of Committee V.7: Structural longevity

Report of Committee V.8: Risers and pipelines

Failure modes of risers and pipelines

Carlos Guedes Soares

Cascais, September, 2015

19th INTERNATIONAL SHIP AND

ISSC committee I.1: ENVIRONMENT

ENVIRONMENTAL DATA ........................................................................................................ 5

SPECIAL TOPICS ...................................................................................................................... 40

ISSC committee I.1: ENVIRONMENT

5.2.2 Numerical and Analytical Method ...................................................................... 43

DESIGN AND OPERATIONAL ENVIRONMENT ................................................................ 47

ISSC committee I.1: ENVIRONMENT

ISSC committee I.1: ENVIRONMENT

ISSC committee I.1: ENVIRONMENT

Locally sensed wind measurements

World Data Center, Silver Spring, Maryland, United States

ISSC committee I.1: ENVIRONMENT

Remotely sensed wind measurements

ISSC committee I.1: ENVIRONMENT

Numerical modelling to complement measured data

ISSC committee I.1: ENVIRONMENT

Locally sensed wave measurements

ISSC committee I.1: ENVIRONMENT

ISSC committee I.1: ENVIRONMENT

ISSC committee I.1: ENVIRONMENT

Remotely sensed wave measurements

ISSC committee I.1: ENVIRONMENT

Numerical modelling to complement measured data

ISSC committee I.1: ENVIRONMENT

Wave description from measured ship motions

In-situ current measurements

ISSC committee I.1: ENVIRONMENT

Remotely sensed current measurements

Numerical modelling to complement measured data