CPD Lecture Unit 5

NAMP
PIECE
Outline
1.1 Introduction and definition of Process integration. 1.2 Overview of Process Integration tools
Module 8: introduction to process integration
NAMP
PIECE
1.1 Introduction and definition of Process integration.
NAMP
PIECE
A Very Brief History of Process Integration
Linnhoff started the area of pinch (bottleneck identification) at UMIST in the 60s, focusing on the area of Heat Integration UMIST Dept of Process Integration was created in 1984, shortly after the consulting firm Linnhoff-March Inc. was formed

PI is not really easy to define
NAMP
PIECE
Definition of process integration
The International Energy Agency (IEA) definition of process integration
"Systematic and General Methods for Designing Integrated Production Systems, ranging from Individual Processes to Total Sites, with special emphasis on the Efficient Use of Energy and reducing Environmental Effects"
From an Expert Meeting in Berlin, October 1993

Module 8: introduction to process integration 4
NAMP
PIECE
Definition of process integration Later, this definition was somewhat broadened and more explicitly stated in the description of its role in the technical sector by this Implementing Agreement: "Process Integration is the common term used for the application of methodologies developed for System-oriented and Integrated approaches to industrial process plant design for both new and retrofit applications. Such methodologies can be mathematical, thermodynamic and economic models, methods and techniques. Examples of these methods include: Artificial Intelligence (AI), Hierarchical Analysis, Pinch Analysis and Mathematical Programming. Process Integration refers to Optimal Design; examples of aspects are: capital investment,energy efficiency, emissions, operability, flexibility, controllability, safety and yields. Process Integration also refers to some aspects of operation and maintenance". Later, based on input from the Swiss National Team, we have found that Sustainable Development should be included in our definition of Process Integration.
rson, International Energy Agency (IEA) Implementing Agreement, A worldwide catalogue on Proc 2001).
NAMP
PIECE
El-Halwagi, M. M., Pollution Prevention through Process Integration: Systema . Academic Press, 1997.

A Chemical Process is an integrated system of interconnected units and streams, and it should be treated as such. Process Integration is a holistic approach to process design, retrofitting, and operation which emphasizes the unity of the process. In light of the strong interaction among process units, streams, and objectives, process integration offers a unique framework for fundamentally understanding the global insights of the process, methodically determining its attainable performance targets, and systematically making decisions leading to the realization of these targets. There are three key components in any comprehensive process integration methodology: synthesis, analysis, and optimization. Module 8: introduction to process integration
NAMP
PIECE
Nick Hallale, Aspentech CEP July 2001 Burning Bright Trends in Process Integration

Process Integration is more than just pinch technology and heat exchanger networks. Today, it has far wider scope and touches every area of process design. Switched-on industries are making more money from their raw materials and capital assets while becoming cleaner and more sustainable

NAMP
PIECE
So What Happened?
In addition to thermodynamics (the foundation of pinch), other techniques are being drawn upon for holistic analysis, in particular:
Process Process Process Process Process Process
modeling statistics optimization economics control design
NAMP
PIECE
Modern Process Integration context
Process integration is primarily regarded as process design (both new and retrofits design), but also involve planning and operation. The methods and systems are applied to continuous, semi-batch, and batch process.

Business objectives currently driving the development of PI:

a) Emphasis is on retrofit projects in the new economy driven by Return on Capital Employed (ROCE) b)PI is Finding value in data quality c) Corporations wish to make more knowledgeable decisions: 1.For operations, introduction to process integration 2.During the design process.
Module 8:
NAMP
PIECE
Modern Process Integration context
Possible Objectives: Lower capital cost design, for the same design objective Incremental production increase, from the same asset base Marginally-reduced unit production costs Better energy/environmental performance, without compromising competitive position Re d ucing COSTS POLLUTION EN ERGY
Incr e a sing TH ROUGH P UT YIELD PROFIT

10
NAMP
PIECE
Modern Process Integration context Among the design activities that these systems and methods address today are: Process Modeling and Simulation, and Validations of the results in order to have information accurate and reliable of the process. Minimize Total Annual Cost by optimal Trade-off between Energy, Equipment and Raw Material Within this trade-off: minimize Energy, improve Raw Material usage and minimize Capital Cost Increase Production Volume by Debottlenecking Reduce Operating Problems by correct (rather than maximum) use of Process Integration Increase Plant Controllability and Flexibility Minimize undesirable Emissions Add to the joint Efforts in the Process Industries and Society for a Sustainable Development.
11
NAMP
PIECE
Summary of Process Integration elements

Im p r ovin g ove r a ll p la n t fa cilit ie s e n e r g y e f f icie n cy a n d p r od u ct ivit y r e q u ir e s a m u lt i-p r on g e d a n a lysis in volvin g a va r ie t y of t e ch n ica l sk ills and e xp e r t ise , in clu d in g : Kn ow le d g e of b ot h con ve n t ion a l in d u st r y p r a ct ice a n d st a t e -oft h e -a r t t e ch n olog ie s a va ila b le com m e r cia lly Fa m ilia r it y in d u st r y issu e s t rends w it h and
Pr oce ss D a t a
PI syst e m s & Tools
Pr oce ss k now le d g e
M e t h od ology f or d e t e r m in in g cor r e ct m a r g in a l cost s. Pr oce d u r e s a n d t ools f or En e r g y, W a t e r, a n d raw m a t e r ia l Module 8: se r va t ion a u d it s Con introduction to process integration
12
NAMP
PIECE
In conclusion, process integration has evolved from Heat recovery methodology in the 80s to become what a number of leading industrial companies and research groups in the 20th century regarding the holistic analysis of processes, involving the following elements:

Process data lots of it Systems and tools typically computer-oriented Process engineering principles - in-depth process sector knowledge Targeting - Identification of ideal unit constraints for the overall process
NAMP
PIECE
Outline
1 1 Introduction andd definition o f P ro ce ss 1. . 1 I tro d u cti n a n d e fi i o n of Process n o n ti integration. i te g ra ti n . n o
1..2 Overview of Process Integration tools . 1 2 Overview of Process Integration tools
1..3 An around--the--world tour of PI practitioners 1 3 An around the world tour of PI practitioners focuses of expertise. focuses of expertise
14
NAMP
PIECE
1.2 Overview of Process Integration Tools
15
NAMP
PIECE

Bu sin e ss M od e l An d Su p p ly Ch a in M od e lin g . Re a l Tim e
Op t im iza t ion
Pin ch An a lysis za t ion by M a t h e m a t ica l m m in g
St och a st ic Se a r ch M e t h od s
Pr oce ss Sim u la t ion
Life Cycle An a lysis a t a - D r ive n Pr oce ss M od e lin g
St eady st at e Dynam ic
e g r a t e Pr oce ss D e sig n a n d n t r ol
Pr oce ss Dat a
16
NAMP
PIECE

Click here
Bu sin e ss M od e Click here l

Su p p ly Ch a in M anagm ent .
Click here
Re a l Tim e Op t im iza t ion
Pin ch An a lysis Click here za t ion by M a t h e m a t ica l m m in g

Click here Click here
St och a st ic Se a r ch M e t h od s
Click here
Pr oce ss Sim u la t ion
Life Cycle An a lysis Click here a t a - D r ive n Pr oce ss M od e lin g

Click here
St eady st at e Dynam ic
Click here
e g r a t e Pr oce ss D e sig n a n d n t r ol
Pr oce ss Dat a
17
NAMP
PIECE
Process Simulation
18
NAMP
PIECE
Process Simulation
Process modeling
W h a t is a m od e l? A model is an abstraction of a process operation used to build, change, improve, control, and answer questions about that process
Process modeling is an activity using models to solve problems in the areas of the process design, control, optimization, hazards analysis, operation training, risk assessment, and software engineering for computer aided engineering environments.
19
NAMP
PIECE
Process Simulation
Tools of process modeling Process Modeling
Syst em Theory
Physics and Chem ist ry
Applicat io n
Com put e s Science
St at ist ic s
Num eric al Met hods
Process m odeling is an underst anding of t he process phenom ena an d t ransform ing t his underst anding int o a m odel.
20
NAMP
PIECE
Process Simulation

What is a model used for?
Nilsson (1995) presents a generalized model, which, as depicted in the figure below, can be used for different basic problem formulations: Simulation, Identification, estimation and design.
Input I MODEL
Output O
If the model is known, we have two uses for our model: Direct: Input is applied on the model, output is studied (Simulation) Inverse: Output is applied on the model, Input is studied
NAMP
PIECE
Process Simulation
If both Input and Output are Known, we have three formulations (Juha Yaako, 1998):
Identification: We can find the structure and parameters in the model. Estimation: If the internal structure of model is known, we can find the internal states in model. Design: If the structure and internal states of model are known, we can study the parameters in model.
22
NAMP
PIECE
Process Simulation

Demands set to models:

Accuracy Requirements placed on quantitative and qualitative models. Validity Consideration of the model constraints. A typical model process is non-linear, nevertheless, nonlinear models are linearized when possible, because they are easier to use and guarantee global solutions. Complexity Models can be simple (usually macroscopic) or detailed (usually microscopic). The detail level of the phenomena should be considered. Computational The models should currently regard computational orientation. Robustness .
23
NAMP
PIECE
Process Simulation The figure below shows a comparison of input and output for a process and its model. Note that always n > m and k > t.
Input
X1, ..., Xn
Out put PROCES S

Y1, ..., Yk
A model does not include every n>m, and k>t.

All models are wrong, Some models are useful George Box, PhD University of Wisconsin
Input MODEL
X1, ..., Xm
Out put
Y1, ..., Yt
In the process industry we find, two levels of models; Plant models, and models of unit operations such as reactor, columns, pumps, heat exchangers, tanks, etc.
NAMP
PIECE
Process Simulation
Types of models:
Intuitive: the immediate understanding of something without conscious reasoning or study. This are seldom used. Verbal: If an intuitive model can be expressed in words, it becomes a verbal model. First step of model development. Causal: as the name implies, these model are about the causal relations of the processes. Qualitative: These models are a step up in model sophistication from causal models. Quantitative: Mathematical models are an example of quantitative models. These models can be used for (nearly) every application in process engineering. The problem is that these models are not documented or can be too costly to construct when there is not enough knowledge (physical and chemical phenomena are poorly understood). Sometimes the application encountered does not require such model sophistication.
From first Principles

From Stochastic knowledge

25
NAMP
PIECE
Process Simulation
Simulation: what if experimentation with a model
Simulation involves performing a series of experiments with a process model. Input Out put Steady State MODEL Snapshot
X1, ..., Xm Y1, ..., Yt
Algebraic equations
Input
X(t)1, ..., X(t)m
MODEL (t)
Output
Dynamic
Y(t)1, ..., Y(t)t
Movie (time functions) Time is an explicit variable differential equations Certain phenomena require dynamic simulation (e.g. control strategies, real time descition). 26
NAMP
PIECE
Process Simulation
Illustration:
Dynamic simulation of a storage tank
m1
Staedy state simulation of a storage tank

m1
t = time
Simulation unit
Hi-Limit M=constant Level M=f(t) Lo-Limit
m2
m2(t)
0=In - Out + Production - Consumption
Acumulation = In - Out + Production - Consumption
0 = m1 m2 + 0 0
m2 m2 t Module 8: introduction to process integration
dM = m1 m2 ( t ) + 0 0 dt
t
27
NAMP
PIECE
Process Simulation
The steady-state simulation does not solve time-dependent equations. The Subroutines simulate the steady-state operation of the process units ( operation subroutines) and estimate the sizes and cost the process units ( cost subroutines). A simulation flowsheet, on the other hand, is a collection of simulation units(e.g., reactor, distillation columns, splitter, mixer, etc.), to represent computer programs (subroutines) to simulate the process units and areas to represent the flow of information among the simulation units represented by arrows.
28
NAMP
PIECE
Process Simulation
To convert from a process flowsheet to a simulation flowsheet, one replaces the process unit with simulation units (Models). For each simulation unit, one assigns a subroutine (or block) to solve its equations. Each of the simulators has a extensive list of subroutines to model and solve the equations for many process units.
The Dynamic simulation enables the process engineer to study the dynamic response of potential process design or the existent Process to typical disturbances and changes in operating conditions, as well as, strategies for the start up and shut down of the potential process design or existing process.
29
NAMP
PIECE
Process Simulation
Differences between Steady State and Dynamic Simulation
Steady-State Simulation
Dynamic Simulation
Snapshot of a unit operation or plant Mimic of plant operation Balance at equilibrium condition Equilibrium results for all unit operations Equipment sizes in general not needed Amount of information required: small to medium Time dependent results It doesnt assume equilibrium conditions for all units Equipment sizes needed Amount of information required: medium to large
30
NAMP
PIECE
Process Simulation
Solution Strategies
The Sequent ial Modular St rat egy flowsheet broken int o unit operat ions (m odules) each m odule is calculat ed in sequence problem s wit h recycle loops The Sim ult aneous Modular St rat egy develops a linear m odel for each unit m odules wit h local recycle are solved sim ult aneously flowsheet m odules are solved sequent ially The Sim ult aneous Equat ion-solving St rat egy describe ent ire flow sheet w it h a set of equat ions all equat ions are sort ed and solved t oget her hard t o solve very large equat ions syst em s
31
NAMP
PIECE
Process Simulation
Why steady-state simulation is important: Better understanding of the process Consistent set of typical plant/facility data Objective comparative evaluation of options for Return On Investment (ROI) etc. Identification of bottlenecks, instabilities etc. Perform many experiments cheaply once the model is built Avoid implementing ineffective solutions
32
NAMP
PIECE
Process Simulation
Why dynamic simulation is important:
Online system
OPTIMIZATION of plant operations ADVANCEMENT OF PLANT OPERATIONS/ OPERATIONAL SUPPORT / OPTIMIZATION Predictive simulation Optimal conditions EDUCATION, TRAINING CONTROL SYSTEM Operation training simulator DCS control logic Plant diagnosis system PROCESS DESIGN / ANALYSIS Examination of operations Control strategies Advanced control systems Batch scheduling
33
Quasi-online system
Off-line system
NAMP
PIECE
Challenges of simulation
Simulation is not the highest priority in the plant facilities

Production or quality issues take precedence Hard to get plant facilities resources for simulation
Up front time required before results are available

Model must be calibrated, and results validated, before they can be trusted At odds with quarterly balance sheet culture May need to structure project to get some results out early
34
NAMP
PIECE
Data Reconciliation
35
NAMP
PIECE
Data Reconciliation Typical Objectives of Data Treatment. Provide reliable information and knowledge of complete data for validation of process simulation and analysis Yield monitoring and accounting Plant facilities management and decision-making Optimization and control Perform instrument maintenance Instrument monitoring Malfunction detection calibration Detect operating problems Process leaks or product loss Estimate unmeasured values Reduce random and gross errors in measurements Detect steady states
NAMP
PIECE
Data Reconciliation
Busi nes s man age men t

NO IT AM RO FN I
Data treatment is critical for

Process simulation Control and optimization Management planning
37
NAMP
PIECE
Data Reconciliation
Overview
Management planning Production
Manual data On-line data Lab data Data Treatment
Plant shutdown Equipment performance Modeling and Simulation Optimization Instrumentation design
Instrument maintenance
38
NAMP
PIECE
Data Reconciliation
Typical Problems With Process Measurements

Measurements inherently corrupted by errors: measurement faults errors during processing and transmission of the measured signal Random errors Caused by random or temporal events Inconsistency (Gross) errors Caused by nonrandom events: instrument miscalibration or malfunction, process leaks Non-measurements Sampling restriction, measuring technique, instrument failure
39
NAMP
PIECE
Data Reconciliation
Random errors
Features High frequency Unrepeatable: neither magnitude nor sign can be predicted with certitude Sources Power supply fluctuation Signal conversion noise Changes in ambient condition
Module : introduction to process integration
40
NAMP
PIECE
Data Reconciliation
Inconsistency (Gross error)

Features Low frequency Predictable: certain sign and magnitude Sources Caused by nonrandom events Instrument related
Miscalibration or malfunction Wear or corrosion of the sensors
Process related
Process leaks Solid deposits
41
NAMP
PIECE
Data Reconciliation
Illustration Of Random & Gross Errors: Ran do m erro rs
bn or ma lity
na
Gr oss err or
Reliable value
t
NAMP
PIECE
Data Reconciliation

Solutions To Problems
Random errors: Data processing Based on successive measurement of each individual variable: Temporal redundancy Traditional filtering techniques Wavelet Transform techniques Inconsistency: Data reconciliation Based on plant structure: Spatial redundancy Subject to conservation laws Unmeasured data Data reconciliation
43
NAMP
PIECE
Data Reconciliation M e a sur e m e n t Pr ob le m H a n d lin g :
F
Pro ces sin g ran do m err ors
t
44
R ec on cil in g Gr os s er ro rs
NAMP
PIECE
Data Reconciliation
Data Treatment Typical Strategy
1. Establish Plant facilities operating regimes 2. Data processing Remove random noise Detect and correct abnormalities 3. Steady state detection Identify steady-state duration Select data set 4. Data reconciliation Detect gross errors Correct inconsistencies Calculate unmeasured parameters
45
NAMP
PIECE
Data Reconciliation METHODOLOGY EMPLOYED
Process data Data processing Steady state detection Variables classification Gross error detection Data reconciliation Applications
46
NAMP
PIECE
Data Reconciliation
What is data reconciliation?

Data reconciliation is the validation of process data using knowledge of plant structure and the plant measurement system
1 + 1 = 3 !!!
THERMODYNAMIC THERMODYNAMIC PROPERTIES

PROPERTIES
STATISTICAL PRINCIPLES
STATISTICAL PRINCIPLES
ACCURATE and RELIABLE INFORMATION
1.3 + 1.3 = 2.6 ACCURATE and RELIABLE INFORMATION

NAMP
PIECE
Data Reconciliation
Objectives of Data Reconciliation Optimally adjust measured values within given process constraints mass, heat, component balances Improve consistency of data to calibrate and validate process simulation Estimate unmeasured process values Obtain values not practical to measure directly Substitute calculated values for failed instrument
48
NAMP
PIECE
Data Reconciliation

Possible Benefits:
More accurate and reliable simulation results More reliable data for process analysis and decision making by mill manager Instrument maintenance and loss detection:
e.g. US$3.5MM annually in a refinery by decreasing loss by 0.5% of 100K BPD
Improve measurement layout Decrease number of routine analysis Improve advanced process control Clear picture of plant operating condition Early detections of problems Quality at process level Work Closer to specifications.
NAMP
PIECE
Data Reconciliation
Data Reconciliation Problem of Process Under Different Status

Steady-state data reconciliation based on steady-state model Using spatial redundancy Dynamic data reconciliation based on dynamic models Using both spatial & temporal redundancy
50
NAMP
PIECE
Data reconciliation (DR)
DR Problem Of Process Under Different Status (Contd.)

General expression of conservation law: input- output + generation- consumptionaccumulation= 0 Steady state case: no accumulation of any measurement Constraints are expressed algebraically Dynamic process: Accumulation cannot be neglected Constraints are differential equations
51
NAMP
PIECE
Data Reconciliation
Data Reconciliation of Different Constraints

Linear data reconciliation Only mass balance is considered flows are reconciled Bilinear data reconciliation Component balance imposed as well as energy balance flows & composition measurements are reconciled Nonlinear data reconciliation Mass/energy/component balances are included Flow rate, composition, temperature or pressure measurements are reconciled
52
NAMP
PIECE
Data Reconciliation
Measurement Errors? Unclosed Balances? Unidentified Losses? Efficiency? Performance?
Gross Error Detection Closed Balances Identified Losses Monitored Efficiency Quantified Performance
D ATA RECON CILIATION

NAMP
PIECE
Pinch Analysis.
54
NAMP
PIECE
Pinch Analysis
W ha t is Pinch An a lysis?
The prime objective of Pinch Analysis is to achieve financial savings in the process industries by optimizing the ways in which process utilities (particularly energy, mass, water, and hydrogen), are applied for a wide variety of purposes.
The Heat Recovery Pinch(Thermal Pinch Analysis now) was discovered indepently by Hohmann (71), Umeda et al. (78-79) and Linnhoff et al. (78-79).
Pinch Analysis does this by making an inventory of all producers and consumers of these utilities and then systematically designing an optimal scheme of utility exchange between these producers and consumers. Energy, Mass, and water re-use are at the heart of Pinch Analysis activities.
With the application of Pinch Analysis, savings can be achieved in both capital investment and operating cost. Emissions can be minimized and throughput maximized.
55
NAMP
PIECE
Pinch Analysis
FEATURES
The Pinch analysis is a t echnique t o design: Recovery Networks (Heat and Mass) Utility Networks (so called Total site Analysis) The basis of Pinch Analysis: The use of thermodynamic principles (first and second law). The use heuristics (insight), about design and economy. The Pinch Analysis makes extensive use of various graphical representations
56
NAMP
PIECE
Pinch Analysis
The Pinch Analysis provides insight s about t he process. In Pinch analysis, t he design engineering cont rols t he design procedure ( int eract ive m et hod ) . The pinch Analysis int egrat e s econom ic param et ers
57
NAMP
PIECE
Pinch Analysis The Four phases of pinch analysis in t he design of recovery process: Process Sim ulat ion Dat a Ext ract ion Target ing
Where an initial design is simplified and improved Which involves collecting data for the process and the
Which establishes figures for the best performance in v
Where an initial Heat Exchanger Network is established by heuristics tools all
Design
Opt im izat ion
58
NAMP
PIECE
Pinch Analysis
Heat Exchanger Network (HEN) HEN design is the classical domain of Pinch Analysis. By making proper use of temperature driving forces available between process steams, the optimum heat exchanger network can be designed, taking into account constraints of equipment location, materials of construction, safety, control, and operating flexibility. This then sets the hot and cold utility demand profile of the plant. When used correctly, Pinch Analysis yields optimum HEN designs that one would have been unlikely to obtain by experience and intuition alone.
59
NAMP
PIECE
Pinch Analysis
Combined Heat and Power (CHP) CHP is the terminology used to describe plant energy utilities, boilers, steam turbines, gas turbines, heat pumps, etc. Traditionally, these have been referred to as "plant utilities", without distinguishing them from other plant utilities such as cooling water and wastewater treatment. The CHP system supplies the hot utility and power requirements of the process. Pinch Analysis offers a convenient way to guarantee the optimum design, which can include the use of cogeneration or threegeneration (use of hot utility to produce cold utility and power for things like refrigeration).
60
NAMP
PIECE
Pinch Analysis

Possible Benefits:
One of the main advantages of Pinch Analysis over conventional design methods is the ability to set a target energy consumption for an individual process or for an entire production site before to design the processes. The energy target is the minimum theoretical energy demand for the plant or site. Pinch Analysis will therefore quickly identify where energy savings are likely to be found. Reduction of emissions Pinch Analysis enable to the engineer with tool to find the best way to change the process, if the process let it.
61
NAMP
PIECE
Pinch Analysis
In addition, Pinch Analysis allow you to: Update or Development of Process Flow Diagrams Identify the bottleneck in the process Departmental Simulations Full Plant Facilities Simulation Determine Minimal Heating (Steam) and Cooling Requirements Determine Cogeneration and Three-generation Opportunities Determine Projects with Cost Estimates to Achieve Energy Savings Evaluation of New Equipment Configurations for the Most Economical Installation Pinch Replaces the Old Energy Studies with a Live Study that Can Be Easily Updated Using Simulation
62
NAMP
PIECE
Optimization by Mathematical Programming
63
NAMP
PIECE
Optimization by Mathematical Programming: introduction
A Mathematical Model of a system is a set of mathematical relationships (e.g., equalities, inequalities, logical conditions) which represent an abstraction of the real world system under consideration. A Mathematical Model can be developed using:
Fundamental approaches Accepted theories of sciences are used to derive the equations (e.g., Thermodynamics Laws). Empirical Methods Input-output data are employed in tandem with statistical analysis principles so as to generate empirical or Black box models. Methods Based on analogy Analogy is employed in determining the essential features of the system of interest by studying a similar, well understood system.
NAMP
PIECE
Optimization by Mathematical Programming: introduction A mathematical Model of a system consists of four key elements:
1. Variables The variables can take different values and their specifications define different states of the systems.
1. Continuous, 2. Integer, 3. Mixed set of continuous and integer.
2. Parameters The parameters are fixed to one or multiple specific values, and each fixation defines a different model. 3. Constraints the constraints are fixed quantities by the model statement 4. Mathematical Relationships The mathematical model relations can be classified as:
1. Equalities usually composed of mass balance, energy balance, equilibrium relations, physical property calculations, and engineering design relations which describe the physical phenomena of the system. 2. Inequalities consist of allowable operating regimes, specifications on qualities, feasibility of heat and mass transfer, performance requirements, and bound on availabilities and demands. 3. Logical conditions provide the connection between the continuous and integer variables.
The mathematicalto process integration algebraic, differential, or a mixed set Module 8: introduction relations can be 65
NAMP
PIECE
What is Optimization?
A optimization problem is a mathematical model which in addition to the before mentioned elements contains one or more performance criteria. The performance criteria is denoted as an objective function. It can be minimization of cost, the maximization or profit or yield of a process for instance. If we have multiple performance criteria then the problem is classified as multi-objective optimization problem.
A well defined opt im izat ion problem feat ures a num ber of variables great er t han t he num ber of equalit y const raint s, which im plies t hat t here exist degrees of freedom upon w hich we opt im ize.
NAMP
PIECE
The typical mathematical model structure for an optimiztion problem takes the following form:

min f ( x, y )
x, y
s.t. h( x, y ) = 0 g ( x, y ) 0 x X n y Y integer
Where x is a vect or of n cont inuous variables, y is a vect or of int eger variables, h(x,y)= 0 are m equalit y const raint s, g(x,y) 0 are p inequalit y const raint s, and f(x,y) is t he object ive funct ion.
NAMP
PIECE
Classes of Optimization Problems (OP)
If the objective function and constraints are linear without the use of integer variables, then OP becomes a linear programming (LP) problem. If there exist nonlinear terms in the objective function and/or constraints without the use of integer varialbes, the OP becomes a nonlinear programming (NLP) problem. If integer variables are used, they participate linearly and separtly from the continuous variables, and the objective function and constraints are linear, then OP becomes a mixed-integer linear programming (MILP) problem. If integer variables are used, and there exist nonlinear terms in the objective function and/or constraints, then the OP becomes a mixedinteger nonlinear programming (MINLP) problem. Whenever possible, linear programs (LP or MILP) are used because they guarantee global solutions.
68
NAMP
PIECE
Optimization by Mathematical Programming Applications:

Process Synthesis
Heat Exchanger Networks Distillation Sequencing Mass Exchanger Networks Reactor-based Systems Utility Systems Total Process Systems
Design, Scheduling, and Planning of Process

Design and Retrofit of Multiproduct Plants Design and Scheduling of Multiproduct Plants
Interaction of Design and Control Molecular Product Design Facility Location and allocation Facility Planning and Scheduling Topology of Transport Networks
NAMP
PIECE
Stochastic Search Methods
Module : introduction to process integration
70
NAMP
PIECE
Why stochastic Search Methods
All of the model formulations that you have encountered thus far in the Optimization have assumed that the data for the given problem are known accurately. However, for many actual problems, the problem data cannot be known accurately for a variety of reasons. The first reason is due to simple measurement error. The second and more fundamental reason is that some data represent information about the future (e.g., product demand or price for a future time period) and simply cannot be known with certainty.
Module introduction to process integration :
71
NAMP
PIECE
There are probabilistic algorithms, such as:

Simulated annealing (SA) Genetic Algorithms (GAs) Tabu search
These are suitable for problems that deal with uncertainty. These computer algorithms or procedure models do not guarantee global optimally but are successful and widely known to come very close to the global optimal solution (if not to the global optimal). GA has the capability of collectively searching for multiple optimal solutions for the same best cost.
Such information could be very useful to a designer, because one configuration could be much easier to build than another.
SA takes one solution and efficiently moves it around in the search space, avoiding local optima.
Module introduction to process integration : 72
NAMP
PIECE
What is GAs?
GAs simulate the survival of the fittest among individuals over consecutive generation for solving a problem. Each individual represents a point in a search space and a possible solution. The individuals in the population are then made to go through a process of evolution. GAs are based on an analogy with the genetic structure and behaviour of chromosomes within a population of individuals using the following foundations:
Individuals in a population compete for resources and mates. Those individuals most successful in each 'competition' will produce more offspring than those individuals that perform poorly. Genes from good individuals propagate throughout the population so that two good parents will sometimes produce offspring that are better than either parent. Thus each successive generation will become more suited to their environment. Module introduction to process integration : 73
NAMP
PIECE

A population of individuals is maintained within search space
for a GA, each representing a possible solution to a given problem. Each individual is coded as a finite length vector of components, or variables, in terms of some alphabet, usually the binary alphabet {0,1}.
The chrom osom e (solut ion) is com posed of several genes (variables). A f it n e ss scor e (t he best object ive funt ion) is assigned t o each solut ion represent ing t he abilit ies of an individual t o com pet e . The individual wit h t he opt im al (or generally near opt im al) fit ness score is sought . The GA aim s t o use select ive breeding of t he solut ions t o produce offspring bet t er t han t he parent s by com bining inform at ion from t he chrom osom es.
74
NAMP
PIECE
Stochastic Search Methods The general genetic algorithm solution is found by: 1. [ St a rt ] Generat e random populat ion of n chrom osom es
(suitable solutions for the problem) 2.[Fitness] Evaluate the fitness f(x) (objective function) of each chromosome x in the population. 5.[New population] Create a new population by repeating following steps until the new populationis complete
1.
to their fitness (the better fitness, the bigger chance to be selected) 2.[Crossover] With a crossover probability cross over the parents to form a new offspring (children). If no crossover was performed, offspring is an exact copy of parents.. 3.[Mutation] With a mutation probability mutate new offspring at each locus (position in chromosome). 4.[Accepting] Place new offspring in a new population 4.
[Selection] Select two parent chromosomes from a population according
7.[Replace] Use new generated population for a further run of algorithm 4. 8.[Test] If the end condition is satisfied, stop, and return the best solution in current population 5. 9.[Loop] Go to step 2
75
NAMP
PIECE

Encoding of a Chromosome The chromosome should in some way contain information about the solution which it represents. The most used way of encoding is a binary string. The chromosome then could look like this:
Each chrom osom e has one binary st ring. Each bit in t his st ring can represent som e charact erist ic of t he solut ion. Or t he whole st ring can represent a num ber Of course, t here are m any ot her w ays of encoding. This depends m ainly on t he solved problem . For exam ple, one can encode direct ly int eger or real num bers. Som et im es it is also useful t o encode som e perm ut at ions.
NAMP
PIECE

Crossover After we have decided what encoding we will use, we can make a step to crossover. Crossover selects genes from parent chromosomes and creates a new offspring. The simplest way how to do this is to choose randomly some crossover point and everything before this point copy from a first parent and then everything after a crossover point copy from the second parent. Crossover can then look like this ( | is the crossover point):
There are ot her ways how t o m ake crossover s, and we can choose m ult iple crossover point s. Crossover s can be rat her com plicat ed and vary depend ing on t he encoding of chrom osom e. Specific crossover s m ade for a specific problem can im prove perform ance of t he genet ic algorit hm .
77
NAMP
PIECE
Mutation After a crossover is performed, mutation takes place. This is to prevent the falling of all solutions in the population into a local optimum. Mutation changes the new offspring randomly. For binary encoding we can switch a few randomly chosen bits from 1 to 0 or from 0 to 1. Mutation can then be shown as:
The m ut at ion depends on t he encoding as w ell as t he crossover. For exam ple when we are encoding perm ut at ions, m ut at ion could be exchanging t w o genes.
NAMP
PIECE
GAs Characteristics:
A GA makes no assumptions about the function to be optimized (Levine, 1997) and thus can also be used for nonconvex objective functions A GA optimizes the tradeoff between exporting new points in the search space and exploiting the information discovered thus far A GA operates on several solutions simultaneously, gathering information from current search points and using it to direct subsequent searches which makes a GA less susceptible to the problems of local optima and noise A GA only uses the objective function or fitness information, instead of using derivatives or other auxiliary knowledge, as are needed by traditional optimization methods.
NAMP
PIECE
Stochastic Search Methods GA Solution Procedure

Start
Initial Population
1st Generation
Get Objective Function Value for Whole Population (Internal optimization) Yes
Nth Generation
Optimum ? No
Sto p
Generate New Population GA parameters GA strategies

(N+1)th Generation
80
NAMP
PIECE
SA and GA comparation: In theory and Practice
81
NAMP
PIECE
Life Cycle Analysis.
82
NAMP
PIECE
Life Cycle Analysis

What is Life Cycle Analysis?

Technique for assessing the environmental aspects and potential impacts associated with a product by: An inventory of relevant inputs and outputs of a system Evaluating the potential environmental impacts associated with those inputs and outputs Interpreting the results of the inventory and impact phases in relation to the objectives of the study heading Evaluation of some aspects of a product system through all stages of its life cycle
83
NAMP
PIECE
Life Cycle Analysis
Why LCA is important:

Tool for improvement of environmental performance Systematic way of managing an organizations environmental affairs Way to address immediate and long-term impacts of products, services and processes on the environment Focus on continual improvement of the system
84
NAMP
PIECE
Life Cycle Analysis
LCA m e t h od olog y:
LIFE-CYCLE ASSESSMENT
Goal and Scope definitio n Inventor y analysis Interpretation
DIRECT APPLICATIONS
Product development and improvement Strategic planification Public policy Marketing Etc.
Impact assessment
OTHER ASPECTS
Technical Economic Market Social etc. 85
NAMP
PIECE
Life Cycle Analysis
Goal and scope definitions goal , scope functional unit

efficiency durability performance quality standard
system boundaries ,
NAMP
PIECE
Life Cycle Analysis
Inventory analysis data collection , refining system boundaries calculation no formal description, software validation of data relating data to the specific system allocation 87 Module 8: introduction to process integration
NAMP
PIECE
Life Cycle Analysis
Impact assessment category definition classification characterization weighting

NAMP
PIECE
Life Cycle Analysis
Interpretation/improvement assessment identification of significant environmental issues evaluation , , conclusions and recommendations
89
NAMP
PIECE
Life Cycle Analysis
Possible Benefits:
Improvements in overall environmental performance and compliance Provides a framework for using pollution prevention practices to meet LCA objectives Increased efficiency and potential cost savings when managing environmental obligations Promotes predictability and consistency in managing environmental obligations More effective measurement of scarce environmental
90
NAMP
PIECE
Data-Driven Process Modeling
91
NAMP
PIECE
Data-Driven Process Modelling
Pr oce ss Int e g r a t ion Ch a lle ng e : Make sense of m asses of dat a

Drowning in data!
Many organisat ions t oday are faced wit h t he sam e challenge: TOO MUCH DATA It is t he last it em t hat is of int erest t o us as chem ical engineers
92
NAMP
PIECE
Data-Rich but Knowledge-Poor Far too much data for a human brain Limited to looking at one or two variables at a time:
12
10 8 6 4 2 0 1 2 3 4 5 6 7
Brain
Big Problem: Interesting, useful patterns and relationships not intuitively obvious lie hidden inside enormous, unwieldy databases
NAMP
PIECE
OUTSIDE
IN
Empirical Model
This approach uses the plant process data directly, to establish mathematic correlations. Unlike the theoretical models, empirical models do NOT take the process fundamentals into account. They only use pure mathematical and statistical techniques. Multi-Variable Analysis (MVA) is one such method, because it reveals patterns and correlations independently of any pre-conceived notions. Obviously this approach is very sensitive to Garbage-in, garbage-out which is why validation of the model is so important. Module 8: introduction to process integration 94
NAMP
PIECE
With MVA you move
From Data to Information. From Information to Knowledge. Knowledge From Action. to
95
NAMP
PIECE
Data-Driven Process Modelling W h a t is M VA? M u lt i- V a r ia t e A n a lysis ( > 5 va r ia b le s)

MVA uses ALL available dat a t o capt ure t he m ost inform at ion possible Principle: boil dow n hundreds of variables down t o a mere handful
MVA
96
NAMP
PIECE
MVA Example: Apples and Oranges

Measurable differences
Colour, shape, firmness, reflectivity, Skin: smoothness, thickness, morphology, Juice: water content, pH, composition, Seeds: colour, weight, size distribution, et cetera
+1
However, always only one latent attribute
-1
Apple or orange?
97
NAMP
PIECE
How MVA Works:

Tmt 1 1 1 2 2 2 3 3 3 4 4 4 5 5 5 6 6 6 X1 -1 -1 -1 -1 -1 -1 -1 -1 -1 0 0 0 0 0 0 0 0 0 X4 -1 -1 -1 0 0 0 1 1 1 X5 -1 -1 -1 1 1 1 0 0 0 1 1 1 0 0 0 -1 -1 -1 Rep 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 Y avec 2.51 2.36 2.45 2.63 2.55 2.65 2.45 2.6 Y sans 2.74 3.22 2.56 3.23 2.47 2.31 2.67 2.45 2.98 3.22 2.57 2.63 3.16 3.32 3.26 3.1 2.97 2.92
Statistical Model
. . .. . .. . . . . .
(internal to software)
Raw Data: impossible to interpret

2.53 3.02 2.7 -1 -1 -1 0 0 0 1 1 1 2.97 2.89 2.56 2.52 2.44 2.22 2.27
trends X
trends X
trends X
700 colum ns 9,000 rows

2-D Visual Outputs

98
NAMP
PIECE
Effect of Outliers on MVA
OUTLINER
1 component
What about an extreme outlier?
99
NAMP
PIECE
Effect of Outliers on MVA
1 component
Linear regression by Least squares !
New (wrong) component!

Extreme outliers very detrimental to MVA
100
Real component has become mere noise

NAMP
PIECE

Benefits: Explore Inter-Relationships

Create and Learn by modelling
What-if Exercises
Low-cost investigation of options
Soft Sensor (Inferential Control)

for parameters we cant measure directly
Feed-Forward (Model-Based) Control
101
NAMP
PIECE
Integrate Process Design and Control
102
NAMP
PIECE
Control Objectives:
Product specifications variability should be kept to a minimum --> process variability (To Control Product quality). Safety issues(separate equipments), energy costs, environmental concerns have increased complexity and sensitivity of processes
Plants become highly integrated in terms of mass and energy and therefore, process dynamics are often difficult to control. The Control is permanently necessary to do for allowing the process to operate in the best conditions. Module 8: introduction to process integration 103
NAMP
PIECE
CONTROLLABILITY
it is a property of a process that accounts for the ease with which a continuous plant can be held at a specified operating policy, despite external disturbances (resiliency) and uncertainties (flexibility) and regardless of the control system imposed on such a plant. Sources
Process Variability MIN

DESIGN
CONTROL
Changes in Process
-Dynamics -Tunings - Control configurations

104
Steady State & Dynamic Simulations

NAMP
PIECE
Integrate Process Design & Control
Fundamentals:
Input Variables PROCESS RESILIENCY
Control Loop
Disturbances sen sor
Process Internal interactions

Input Variables (Manipulated) PROCESS FLEXIBILITY Output Variables (controlled and Measured)
105
Uncertainties
NAMP
PIECE
Integrate Process Design and Control e.g. Controllability analysis for control structures design
Water, F1 Pulp, F2
C C
F C
C, F
Interac tions
INPUTS (process variables or disturbances) EFFECTS OUTPUTS (Best Selection by Controllability analysis)
106
NAMP
PIECE
Integrate Process Design and Control Why Cont rollabilit y is im port ant :
The process will be more capable to move smoothly around the possible operating edge Stability and better performance of control loops and structures System relatively insensitive to perturbations Efficient management of interacting networks
Flexibility
Improvement of current dynamics
107
NAMP
PIECE
Production rate (time) Product quality, and Energy economy.
The Top level of the process control, Strategic control level is thus concerned with achieving the appropriate values principally of:
108
NAMP
PIECE
Real Time Optimizations (RTO)
109
NAMP
PIECE
Real Time Optimizations
The Process Industries are increasingly compelled to operate profitably in very dynamic and global market. The increasing competition in the international area and stringent product requirements mean decreasing profit margins unless plant operations are optimized dynamically to adopt to the changing market conditions and to reduce the operating cost. Hence, the importance of real-time or on-line optimization of an entire plant is rapidly increasing.
110
NAMP
PIECE
Real Time Optimizations

What is RTO? Real-time Optimization is a model-based steady-state technology that determines the economically optimal operating policy for a process in the near term The system optimizes a process simulation and not the process directly Performance measured in terms of economic benefit Is an active field of research:
Model accuracy, error transmission, performance evaluation
111
NAMP
PIECE
RTO Schematically Reconciliation And gross Error Detection Updating Process Model (Steady State Dynamic Simulation)
Product Specification Product Specification
Business Objectives; Business Objectives;
Economic Data; Economic Data;
Steady State Detection
Optimization (Objectives Functions)
Cost, Process, Environmental, Product Data

Plant Facility
112
NAMP
PIECE
Direct Search Method Schematically
Dynamic Simulation (Model)
SETPOINTS (DOFs)
Selected Ouputs
RTO Algorithm (Objective Fct, Constraints)
113
NAMP
PIECE
Business Model And Supply Chain Modeling
114
NAMP
PIECE
Business Model And Supply Chain Modeling
Cost, Process, Cost, Process, Environmental & Environmental & Product Outcomes Product Outcomes
Click here
Process Process Design Design Analysis Analysis And and Synthesis Synthesis
Click Here
Integrated Business Integrated Business & & Click Here Process Model Process Model
Process Process Operation Operation Analysis Analysis and and Optimizati Optimizat
Optimization
Click here
Cost, Process, Cost, Process, Environmental & Environmental & Product Product Data Data
115
NAMP
PIECE
Cost, Process, Environmental & Product Data Integrated Business & Process Model
Reconciled P&E Data
mean all the data are consistent together throughout all the plant facilities Data Reconciliation
Data Validation & Cost, Process, Environmental P&E Data

Processed
and Product Data Processing

Accounting Data
Process (P) & Environmental (E) Data
Data Reconciliation
Reconciliation
Product Data Market Data
model is built it can be used to validate and reconcile data Plant Facilities
NAMP
PIECE
Integrated Business and Process Model with the classification, recording, allocation, and summarization forAccounting Data of the purpose
Process Data Environmental Data Product Data Market Data
d an C) (S ls ain ode Ch C M ly pp v. S Su En d an C) (S ls ain ode Ch C M ly pp v. S Su E n
Ac Mo coun de tin l g
Click here
Integrated Business
Co st
Ac Mo coun de tin l g
Co st
and
st Process Process 1Model Principles
Processed P&E data
Data Data Driven Driven Model Model Click here s s
Simulation Models Models
117
NAMP
PIECE
Supply Chain and Environmental Supply Chain

Supply Chain (SC) is a network of organizations that are involved, through upstream and downstream linkages, in the different processes and activities that produce value in the form of products and services in the hands of the ultimate customer W (Waste)
Recycling W W W W
Distribution
Distribution
Consumer Consumer
W
Supply
Manufacturing Manufacturing Supply

Remanufacturing Reuse W
Retail Retail
W Collection
Environmental Supply Chain (ESC) holds all the elements a traditional supply chain has but is extended to a semi-closed loop in order to also account for the environmental impact of the supply chain and recycling, re-use and collection of used material (Beamon 1999)
NAMP
PIECE
Supply Chain and Environmental Supply Chain
The objective of the SC and ESC models are:

To integrate inter-organizational units along a SC and coordinate materials, information and financial flows in order to fulfill customer demands with the aim of improving SC profitability and responsiveness To gain insight in the total environmental impact of the production process (from supplier to customer and back to the facility by recycling) and all the products that are manufactured. (closely linked to LCA)
119
NAMP
PIECE
Process Design Analysis and Synthesis

Process Design Analysis Design Objectives
Process simulation Data Reconciliation MVA using relational database Pinch analysis LCA SC and ESC model analysis Controllability Analysis Optimization ( Deterministic and/or Stochastic)
Process
Process Integration Design Analysis and Tools
Process Design Analysis and Synthesis Synthesis Loop
Capital Effectiveness Analysis
120
Integrated Business & Process Model
NAMP
PIECE
Process Operation Analysis and Optimization
Integrated Business & Process Model
Detailed Process Investigation to Validate Recommendations
Process Process Operation Integration Process Design Analysis Analysis and Tools Optimization Loop
and Optimization
Data reconciliation for instrument validation Dynamic simulation Process control strategies MVA (Soft sensor dev.) Real-time optimization Optimizated supply chain Model
Objective Function for Process Optimization

121
NAMP
PIECE
Outline
1 1 Introduction andd definition o f P ro ce ss 1. . 1 I tro d u cti n a n d e fi i o n of Process n o n ti integration. i te g ra ti n . n o
1.2 Overview of Process Integration tools 1 . 2 O ve rvi w o f P ro ce ss I te g ra ti n to o l e n o s
1.3 An around-the-world tour of PI practitioners 1.3 An around-the-world tour of PI practitioners focuses of expertise focuses of expertise
122
NAMP
PIECE
1.3 An around-the-world tour of PI practitioners focuses of expertise (May 2003).
123
NAMP
PIECE
Around the World tour of PI practitioners focuses of experience
Courtesy mainly of the www to capture the flavor of the evolution of Process Integration PI is relatively new: Researchers build on their strengths Many of the ground-breaking techniques are coming from universities When techniques become practical, the private sector generally capitalizes and techniques advance more rapidly
124
NAMP
PIECE

Carnegie Mellon University, Department of Chemical Engineering, Pittsburgh, USA Major Contact: Professor Ignacio E. Grossmann, head of department Web: http://www.cheme.cmu.edu/research/capd/

Research Area: Recognized as one of the major research groups in the area of Computer Aided Process Design. In Process Integration, the group is recognized for its work in Mathematical Programming, Optimization, Reactor Systems, Separation Systems (especially Distillation), Heat Exchanger Networks, Operability and the synthesis of Operating Procedures.

Current research in Process Integration includes: 1) Insights to Aid and Automate Synthesis (Invention) 2) Structural Optimization of Process Flowsheets 3) Synthesis of Reactor Systems and Separation Systems 4) Synthesis of Heat Exchanger Networks 5) Global Optimization techniques relevant to Process Integration 6) Integrated Design and Scheduling of Batch plants 7) Supply chain dynamics and optimization

Consortium: "Center for Advanced Process Decision-making" with 20 members (2001) including operating companies, engineering & contracting companies, consulting companies and software vendors. The consortium was founded 1986.
125
NAMP
PIECE

Imperial College, Centre for Process Systems Engineering, London, UK Major Contact:Prof. Efstratios N Pistikopoulos Web: http://www.ps.ic.ac.uk/ and http://www.psenterprise.com Research Area: Recognized as the largest research group in the area of Process Systems Engineering (PSE), which includes Synthesis/Design, Operations, Control and Modeling. The group is recognized as a world-wide center of excellence in Process Modeling, Numerical Techniques/Optimization and Integrated Process Design (includes simultaneous consideration of Process Integration and Control). The Centre is also an important contributor in the area of Integration and Operation of Batch Processes. Current research in Process Integration includes: 1) Integrated Batch Processing 2) Design and Management of Integrated Supply Chain Processes 3) Uncertainty and Operability in Process Design 4) Formulation of Mathematical Programming Models to address problems in Process Synthesis and Integration

Consortium: "Process Systems Engineering" with 17 members (2003) including operating, engineering & contracting companies, software vendors. introduction to process integration Module 8: 126
NAMP
PIECE

UMIST, Department of Process Integration, Manchester, UK Major Contact: Professor Robin Smith, head of department Web: http://www.cpi.umist.ac.uk/ Research Area: Recognized as the pioneering and major research group in the area of Pinch Analysis. Previous research includes targets and design methods for Heat Exchanger Networks (grassroots and retrofits), Heat and Power systems, Heat driven Separation Systems, Flexibility, Total Sites, Pressure Drop considerations, Batch Process Integration, Water and Waste Minimization and Distributed Effluent Treatment.

Current research is organized in three major areas: 1) Efficient Use of Raw Materials (including Water) 2) Energy Efficiency 3) Emissions Reduction 4) Eefficient use of capital.

Consortium: "Process Integration Research Consortium" with 27 members (2003) including operating companies, engineering & contracting companies, consulting companies and software vendors. The consortium was founded in 1984 by six multinational companies.
127
NAMP
PIECE

Chalmers Univ. of Technol., Department of Heat and Power, Gothenburg, Sweden Major Contact: Professor Thore Berntsson, head of department Web: http://www.hpt.chalmers.se/ Research Area: Methodology development and applied research based on Pinch Technology. Emphasis on new Retrofit methods including realistic treatment of geographical distances, pressure drops, varying fixed costs, etc. Important new Concepts include the Cost Matrix for Retrofit Screening and new Grand Composite type Thermodynamic Diagrams for Heat and Power applications (including Gas Turbines and Heat Pumps). Research towards pulp and paper with focus on energy and environment. Research areas are: 1) Retrofit Design of Heat Exchanger Networks 2) Process Integration of Heat Pumps in Grassroots and Retrofits 3) Gas Turbine based CHP plants in Retrofit Situations 4) Applied research in Pulp and Paper industry, such as black liquor gasification, closing the bleaching plant, etc. 5) Environmental aspects of Process Integration, especially greenhouse gas emissions)

Industry: Close co-operation with some of the major pulp and paper industry Module 8: introduction to process integration groups, including training courses, consulting, etc. 128
NAMP
PIECE

cole Polytechnique de Montral, Chemical engineering Department, Quebec, Canada Major Contact: Dr. Paul Stuart, Chair holder Web: http://www.pulp-paper.ca Research Area: the application of Process Integration in the pulp and paper industry, with emphasis on pollution prevention techniques and profitability analysis, the Efficiency use of energy and Raw Materials (including Water), process control, and plant sustainability. Research areas are:: 1) process simulation, 2) Data reconciliation, 3) Process Control, 4) Networks Analysis HEN and MEN, 5) Environmental technologies (e.g., LCA), 6) Business Model. 7) Data Driving Modeling.

Consortium: "Process Integration Research Consortium" with 13 members (2003) including operating companies, engineering & contracting companies, consulting companies and software vendors in pulp and paper industry.
129
NAMP
PIECE

Universitat Politcnica de Catalunya, Chemical Engng. Department, Barcelona, Spain Major Contact: Professor Luis Puigjaner, Director LCMA Web: http://tqg.upc.es/ Research Area: Pioneering work on Computer Aided Process Operations. Within Process Integration, the group is recognized for its contributions in Time-Dependent Processes, such as Combined Heat and Power, Combined Energy-Waste and Waste Minimization, Integrated Process Monitoring, Diagnosis and Control and finally Process Uncertainty. Current research in the area of Process Integration includes: 1) Evolutionary Modeling and Optimization 2) Multi-objective Optimization in time-dependent systems 3) Combined Energy and Water Use Minimization 4) Integration of Thermally Coupled Distillation Columns 5) Hot-gas Recovery and Cleaning Systems Consortium: "Manufacturing Reference Centre" with 12 members (1966) including Conselleria d'Indstria and associated operating companies, engineering and contracting companies, consultants and software vendors.
130
NAMP
PIECE

Texas A&M University, Chemical Engineering Department, Texas, USA Major Contact: Professor Mahmoud M. El-Halwagi Web: http://process-integration.tamu.edu/ and http://www-che.tamu.edu/cpipe/ Research Area: Recognized as a leading research group in the areas of Mass Integration and Pollution Prevention through Process Integration.
Research areas are:
1) Global allocation of Mass and Energy 2) Synthesis of Waste Allocation and Species Interception Networks 3) Physical and Reactive Mass Pinch Analysis 4) Synthesis of Heat-Induced Networks 5) Design of Membrane-Hybrid Systems 6) Design of Environmentally acceptable Reactions 7) Integration of Reaction and Separation Systems 8) Flexibility and Scheduling Systems 9) Simultaneous Design and Control 10) Global Optimization via Interval Analysis
131
NAMP
PIECE

University of Guanajuato, Faculty of Chemistry, Guanajuato, Mxico Major contact: Dr. Martin-Picon-Nunez, Director Web: http://www.ugto.mx Research Area: Hosts the only course Masters Program in process integration in North America, they are developing in the next areas Analysis of Processes, Power Systems, and to develop of technology benign Environmental.

Research areas are: 1) Synthesis of Processes; Modeling, Simulation, Control and Optimization of Processes; New Processes and Materials. 2) Recovery systems of Heat; Renewable sources of Energy; Thermodynamic Optimization. 3) Contaminated Atmosphere rehabilitation; Treatment of Effluents; Environmental Processes.
132
NAMP
PIECE

University of the Witwatersrand, Process & Materials Eng., Johannesburg, South Africa Major Contact: Professor David Glasser, AECI Professor Web: http://www.wits.ac.za/fac/engineering/procmat/homepage.html Research Area: Recognized as the major research group in the development of the Attainable Region (AR) method for Reactor and Process Synthesis. The Attainable Region concept has been expanded to systems where mass transfer, heat transfer and separation take place. In its generalized form (reaction, mixing, separation, heat transfer and mass transfer), the Attainable Region concept provides a Synthesis tool that will provide targets for "optimal" designs against which more practical solutions can be judged. Research areas are: 1) Systems involving Reaction, Mixing and Separation (e.g. Reactive Distillation) 2) Non-isothermal Chemical Reactor Systems 3) Optimization of Dynamic Systems

Clients: they have founded your own consultancy enterprise the name Wits Enterprise.
133
NAMP
PIECE
Linnhoff March Ltd., Northwich, Cheshire, UK Web: http://www.linnhoffmarch.com/ List of Services in the area of Process Integration: Linnhoff March is the pioneering company of Pinch Technology and has built a reputation for being the "Pinch Company", encompassing: Project execution and consulting Software development and support Training assistance PI Technologies: Pinch Technology (Analysis and HEN DesignTotal Site Analysis) Water Pinch for Wastewater minimization Combined Thermal and Hydraulic Analysis of Distillation Columns PI Software: Extensively proven state-of-the-art industry including iency and profitability in the hydrocarbon processingsoftware worldwide. KBC analyses SuperTarget, PinchExpress, WaterTarget and Steam97. and offers Implementation Services to assist clients in realising measurable financial im Typical Projects: 1200 assignments over 18 years - or over 50 studies per year in PI, making them the unquestionable world leader (27th February 2002)Was acquired last year by KBC process technology
134
NAMP
PIECE

American Process Inc., Atlanta, USA. Web:http://www.americanprocess.com

List of Services in the area of Process Integration:
We are the premier consulting engineering specialists dedicated to the pulp and paper industry. Prom. energy and water reduction to planning new power islands. American Process can provide solutions through practical experience, process integration, troubleshooting, and project implementation.

Founded in 1994, with offices in Atlanta, GA, Athens, Greece, and ClujNapoca, Romania, American Process is the premier specialist firm dedicated to reducing energy, water, and other operating costs for the pulp and paper industry.

Energy Targeting Using Pinch Analysis, PARIS (Decision-Making Tool for Optimizing Pulp and Paper Mill Operations)
Production Analysis for Rate and Inventories Strategies.
Module 8: introduction to process Simulation modeling,
integration
135
NAMP
PIECE
Around the World tour of PI practitioners focuses of experience Process Systems Enterprise Ltd., london, UK. Web: http://www.psenterprise.com

List of Services in the area of Process Integration:
Process Systems Enterprise Limited (PSE) is a provider of advanced model-based technology and services to the process industries. These technologies address pressing needs in fastgrowing engineering and automation market segments of the chemicals, petrochemicals, oil & gas, pulp & paper, power, fine chemicals, food, pharmaceuticals and biotech industries. gPROMS, for general PROcess Modelling System
Steady-state and dynamic process simulation, optimization (MINLP) and parameter estimation software, packaged for different users.
Model Enterprise - Supply chain modeling and execution environment. Module 8: introduction to process integration Model Care - Business model 136
NAMP
PIECE
.........and Many Many ot hers

In st it u t ion bo Akadem i Universit y Auburn Universit y Technical Univ. of Budapest Lehrstuhi fr Technische Chemie A Universty of Edinburgh INPT-ENSIGC, Chemical Engng. Lab. Swiss Federal Inst. of Technology University of Lige University of Maribor M a jor Cont a ct Professor Tapio West erlund Professor Christopher Roberts Professor Zsolt Fonyo Prof. Dr. A. Behr Professor Jack W. Ponton Professor Xavier Joulia Professor Daniel Favrat Professor Boris Kalitventzeff Professor Peter Glavic W eb http://www.abo.fi/fak/ktf/at/ http://www.eng.auburn.edu/departm http://www.bme.hu/en/organization/ ent/che/ faculties/chemical/ http://www.chemietechnik.unidortmund.de/tca/ http://www.chemeng.ed.ac.uk/ecpss e/ http://excalibur.univinpt.fr/~lgc/elgcpa6.html http://leniwww.epfl.ch/ http://www.uni-mb.si/ http://www.ulg.ac.be/lassc/
137
NAMP
PIECE

In st it u t ion M a jor Cont a ct Massachuset t s Inst it ut e of Professor George Norw. Univ. of Sci. and Professor Sigurd Skogest ad Technology, St ephanopoulosodoulos A. Princet on Universit y Professor Christ Technol. Floudas G.V. Rex Reklait is Purdue Universit y Professor Universit y of Massachuset t s Universit y College Universit y of Adelaide Indian Inst it ut e of Technology Chem ical Process Professor J M. Douglas . Dr. David Bogle Dr. B.K. O'Neill Dr. Uday V. Shenoy Professor I. Vasalos W eb http://web.mit.edu/cheme/index.htm http://kikp.chembio.ntnu.no/researc l http://titan.princeton.edu/ h/PROST/ http://che.www.ecn.purdue.edu/ http://www.ecs.umass.edu/che/ http://www.chemeng.ucl.ac.uk/ http://www.chemeng.adelaide.edu.a u/ http://www.che.iitb.ernet.in/ http://www.cperi.forth.gr
Engineering Research Inst it ut e
138
NAMP
PIECE

In st it u t ion Technical University of TU of Hamburg-Harburg, Denmark Helsinki University of Technology, Instituto Superior Tcnico, Lappeenranta University of Technol. Murdoch University University of Pennsylvania University of Porto M a jor Cont a ct Professor Bjrn Qvale Professor Gnter Gruhn Professor Carl-Johan Fogelholm, head of laboratory Professor Clemente Pedro Nunes Professor Lars Nystroem Professor Peter Lee Professor Warren D. Seider Professor Manuel A.N. Coelho W eb http://www.et.dtu.dk/ http://www.tu-harburg.de/vt3/ http://www.hut.fi/Units/Mechanic/ http://dequim.ist.utl.pt/english/ http://www.lut.fi/kete/laboratories/P rocess_Engineering/mainpage.htm http://wwweng.murdoch.edu.au/eng index.html http://www.seas.upenn.edu/cheme/c hehome.html http://www.up.pt/ http://www.ufrj.br/home.php
Universidade Federal do Rio Professor Eduardo Mach de Janeiro. Queiroz
139
NAMP
PIECE

In st it u t ion University of Queensland Technion-Israel Institute of University of Ulster Technology COMPANIES M a jor Cont a ct Professor Ian Cameron Professor Daniel R. Lewin Professor J.T. McMullan W eb http://www.cheque.uq.edu.au/ http://www.technion.ac.il/technion/ http://www.ulst.ac.uk/faculty/scienc chem-eng/index_explorer.htm e/energy/index.html
http://www.combination.com Advanced Process Combinatorics (APC) Aspen Technology Inc. (AspenTech)http://www.aspentech.com and http://www.hyprotech.com http://www.ipa-scotland.org.uk/members/nel.htm National Engineering Laboratory (NEL) http://www.quantisci.co.uk/ QuantiSci Limited
...
...
140
NAMP
PIECE
End of Tier 1
At the moment we are assuming that you have done all the reading, this is the end of Tier 1. We do not have doubt that much of this information seems fuzzy, but we are only trying to set all the pieces in the Process Integration scope.

Before to pass to tier 2 lefts to answer a short Quiz
141
NAMP
PIECE
QUIZ
142

CPD Lecture Unit 5

Transféré par

Informations du document

Description originale:

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

CPD Lecture Unit 5

Transféré par

Droits d'auteur :

Formats disponibles

NAMP

Module 8: introduction to process integration

1.1 Introduction and definition of Process integration.

Module 8: introduction to process integration

A Very Brief History of Process Integration

PI is not really easy to define

Module 8: introduction to process integration

Definition of process integration

The International Energy Agency (IEA) definition of process integration

From an Expert Meeting in Berlin, October 1993

Definition of process integration

Definition of process integration

Module 8: introduction to process integration

Definition of process integration

Process Process Process Process Process Process

modeling statistics optimization economics control design

Module 8: introduction to process integration

Modern Process Integration context

Business objectives currently driving the development of PI:

Modern Process Integration context

Incr e a sing TH ROUGH P UT YIELD PROFIT

Module 8: introduction to process integration

Summary of Process Integration elements

PI syst e m s & Tools

Definition of process integration

1 1 Introduction andd definition o f P ro ce ss 1. . 1 I tro d u cti n a n d e fi i o n of Process n o n ti integration. i te g ra ti n . n o

1..2 Overview of Process Integration tools . 1 2 Overview of Process Integration tools

Module 8: introduction to process integration

1.2 Overview of Process Integration Tools

Module 8: introduction to process integration

1.2 Overview of Process Integration Tools

Pin ch An a lysis za t ion by M a t h e m a t ica l m m in g

Pr oce ss Sim u la t ion

Life Cycle An a lysis a t a - D r ive n Pr oce ss M od e lin g

1.2 Overview of Process Integration Tools

Bu sin e ss M od e Click here l

Re a l Tim e Op t im iza t ion

Pin ch An a lysis Click here za t ion by M a t h e m a t ica l m m in g

Pr oce ss Sim u la t ion

Life Cycle An a lysis Click here a t a - D r ive n Pr oce ss M od e lin g

Module 8: introduction to process integration

Module 8: introduction to process integration

Tools of process modeling Process Modeling

Physics and Chem ist ry

Com put e s Science

Num eric al Met hods

Module 8: introduction to process integration

What is a model used for?

Module 8: introduction to process integration

Demands set to models:

Module 8: introduction to process integration

Out put PROCES S

A model does not include every n>m, and k>t.

From first Principles

From Stochastic knowledge

Simulation: what if experimentation with a model

Y(t)1, ..., Y(t)t

Module 8: introduction to process integration

Staedy state simulation of a storage tank

Hi-Limit M=constant Level M=f(t) Lo-Limit

0=In - Out + Production - Consumption

Acumulation = In - Out + Production - Consumption

Module 8: introduction to process integration