81 45 4 1394

FRP

*2
1
1 -
2

. 1000 2000
.
. .
FRP .
.
FRP
. FRP
.

: .FRP

-1
.
.
.
.
.

. .
.
Chen 2006 ] [1 .
.
.
. .
.
FRP

93/11/7 94/5/25 *
 ... / 110 45 4 1394 81

] Batikha .[6 2008 . FRP

FRP . . FRP

FRP . FRP

] .[7 2009
. FRP
FRP .


] Ramadan .[8 .
2010
FRP 60 FRP 1990 .
].[9 FRP
.
FRP
FRP .
.
FRP
. ].[4-2

FRP
.
FRP
-2 .
Teng [5] Hu
ABAQUS .
500 250 FRP
0/25 . .
1:1000 2007 FRP
.
FRP
.
. .
S4R . FRP

] .[10
 ... / 111 45 4 1394 81

.
CFRP .
Riks CFRP
. 5 .
FRP
FRP
. Cohesive 240 210 370 0/3
0/3 .
.
FRP . ) (1 ) 2
Cohesive . (3 ) (1 .

. -1 CFRP

E1=235GPa E2=17GPa E3=17GPa

. 12=0.32 13=0.32 23=0.45

G12=4.5GPa G13=4.5GPa G23=2.5GPa

.


Hashin
Sikadur-30

. 30

18
] .[10
.
.

-3
.
-1-3
CFRP .

- 0/2 .
500 250
0/25 .
0/05 0/2

.
5 .
. S1 S6 .

. S0 S7 .
. -
 ... / 112 45 4 1394 81

.
) (3 ) (4 . .
0/05 44/468 0/2
12/359
. 72/21 .
- ) (2 . -
)(3 ) (1 .
.
1:1000
. .
- R/tfy
. .
-
S1 S6 ) (2
0/4 .
2 . .

250 0/25 0/1 - .
. S8 S14 1:1250
) (4 . 500 250
0/2 S15 S22
0/7 1/2 1:833
. 500 250 0/3
0/7 1/2 . .
0/05 0/2

-2 1:1000

S0 )(kN )(MPa ) (FRP

- - 44/578 - S0
%0/25 %0/25 44/468 0/005 S1
%0/26 %0/51 44/352 0/01 S2
%12/24 %12/68 38/925 0/05 S3
%24/48 %34/06 29/396 0/10 S4
%19/71 %47/05 23/603 0/15 S5
%47/64 %72/28 12/359 0/20 S6
- - - 0/24 S7
 ... / 113 45 4 1394 81

-1 -
-4
1:1000
0/1

-3-2 FRP

FRP

0/05 0/2
) (
FRP
) (5 . FRP
20
.
-2 - S1

S6

.
FRP -
. 0/05
FRP 44/47 49/87
12/14
. 0/2
FRP 12/36 25/01
102/37
. -

.
-3 -

 ... / 114 45 4 1394 81

-3 1:1250

)
S8 )(kN )(MPa (FRP

- - 30/618 - S8
% 1/20 % 1/20 30/251 0/005 S9
% 3/69 % 4/84 29/135 0/01 S10
% 16/17 % 20/23 24/425 0/05 S11
% 33/49 % 46/95 16/244 0/10 S12
% 36/90 % 66/52 10/250 0/15 S13
- - - 0/192 S14

-4 1:833
)
S16 )(kN )(MPa (FRP
- - 64/890 - S15
% 1/27 % 1/27 64/066 0/005 S16
% 2/52 % 3/76 62/450 0/01 S17
% 13/10 % 16/37 54/268 0/05 S18
% 14/25 % 28/29 46/535 0/10 S19
% 20/67 % 43/11 36/918 0/15 S20
% 25/42 % 57/57 27/535 0/20 S21
- - - 0/288 S22

-5 FRP

FRP

)(kN )(MPa
FRP FRP
44/59 S0
% 12/13 0/00
50/00 SF0
44/47 S1
% 12/14 0/005
49/87 SF1
44/35 S2
% 12/14 0/01
49/73 SF2
38/92 S3
% 28/08 0/05
49/85 SF3
29/40 S4
% 58/13 0/10
46/49 SF4
23/60 S5
% 70/34 0/15
40/20 SF5
12/36 S6
% 102/35 0/20
25/01 SF6
44/59 S0 0/00
% 12/13
50/00 SF0 0.268
 ... / 115 45 4 1394 81

) (5 .
-
0/05 0/2
) (5 ) (6 .
) (7 ) (8
5 .
FRP
.
-
FRP FRP
-6 -
) (9 .
S5 SF5
1:1000
FRP FRP
FRP
.
[11] Rotter FRP-
.

-7 )( )( FRP
) (fcr=0.605Et/R .

-8 )( )( FRP

-5 -
S1 SF1

-9 -
FRP FRP
 ... / 116 45 4 1394 81

-3-3 FRP
FRP
15 FRP
) (7 . 25 FRP
) (10 ) (11
0/05 0/2 .
) (7 .
FRP . ) (6 .

. ) (7 0/05 0/1
FRP. FRP .
20 FRP
-3-5 FRP .
FRP - 0/2
0/01 .
0/1 ) S2 S4 FRP ) 20(
SF2 (SF4 20
FRP
. -3-4 FRP

FRP

0/1 20
FRP .

-6 FRP

FRP FRP FRP FRP FRP FRP
(kN) 5cm (kN) 4cm (kN) 3cm (kN) 2cm (kN) 1cm )(kN )(MPa
49/65 50/02 49/72 49/87 50/04 44/47 0/005
49/82 50/05 50/03 49/73 49/30 44/35 0/01
49/96 50/04 49/92 49/85 49/05 38/92 0/05
47/44 48/01 44/60 46/49 47/26 29/40 0/10
30/02 28/13 24/92 25/01 23/41 12/36 0/20

-7 FRP
FRP
FRP FRP FRP
5mm
15mm 10mm FRP
)(MPa
)(kN )(kN )(kN )(kN
)(kN
26/98 36/77 45/56 50/04 29/40 0/10
 ... / 117 45 4 1394 81

) (12 ) (13
FRP

.

FRP .
.
) (
-11
0/1 ( 5
:FRP ( 10 FRP
( 15 FRP

)( )(
-10
0/1 :( FRP(
FRP

-12 S2 SF2

)(

)(

-13 S4 SF4
 ... / 118 45 4 1394 81

FRP -3-6 FRP

.
( -
FRP FRP
. .
( FRP
) ( FRP .
.
-4
-5
][1 Chen, J. F., Rotter, J. M., Teng, J. G., "A
Simple Remedy for Elephant's Foot Buckling
in Cylindrical Silos and Tanks", Advances in
Structural Engineering, 2006, 9 (3), 409-420. FRP
. . . ] [2
:
" .
(
CFRP "


.
.1391
(
. " . ][ 3


.
CFRP "
.
.1390
(
. " . ][ 4

CFRP
0/7 1/2 .
"
(
.1390
FRP
][5 Teng, J. G., Hu, Y. M., "Suppression of Local
Buckling in Steel Tubes by FRP Jacketing",
.
The 2nd International Conference on FRP
Composites in Civil Engineering, Adelaide,
.
Australia, 2004.
][6 Teng, J., Hu, Y. M., "Behaviour of FRP- ( FRP
Jacketed Circular Steel Tubes and Cylindrical
Shells under Axial Compression",
Construction and Building Materials, 2007, 20 FRP .
21, 827-838. FRP
][7 Batikha, M., Chen, J. F., Rotter, M., "Fibre
.
Reinforced Polymer Composites to Increase
the Buckling Strength of Imperfect (
Cylindrical Shells", International Conference FRP
on Structures and Granular Solids-From
Scientific Principles to Engineering .
Applications, Edinburgh, England, 1-2 July,
2008, 177-181.
... 81 1394 4 45 / 119

[8] Batikha, M., Chen, J., Rotter, J., Teng, J.,

"Strengthening Metallic Cylindrical Shells
against Elephant's Foot Buckling with FRP",
Thin-Walled Structures, 2009, 47, 1078-
1091.
[9] Al-Kashif, M. A., Ramadan, H., Rashed, A.,
Haroun, M. A., "Effect of FRP Composites
on Buckling Capacity of Anchored Steel
Tanks", Steel and Composite Structures,
2010, 10 (4), 361-371.
[10] ABAQUS, Version 6.10.1. ABAQUS/
Standard Users Manual, ABAQUS Inc., US,
2010.
[11] Rotter, J. M., "Elephants Foot Buckling in
Pressurized Cylindrical Shells", Stahlbau,
2006, 75 (9), 742-747.
JournalofCivilandEnvironmentalEngineering
Volume45,Issue4,Winter2016

EXTENDEDABSTRACTS

ElephantFootBucklingandRetrofittingofSteelThinWalled
CylindricalShellsUsingFRPCompositeMaterials

MortezaVakili,HosseinShowkati*

DepartmentofCivilEngineering,UrmiaUniversity,Urmia,Iran

Received:27January2015;Accepted:16August2015

Keywords:
Cylindrical shells, Elephant foot buckling, Strengthening, FRP

1.Introduction
Steel tanks designed in cylindrical forms are one of the shell structures which are widely used in industrial
facilities. In geometric terms, these tanks have a very small thickness compared to the other dimensions and thus are
categorized as thin walled structures which should consider the buckling failures. The buckling failure of tanks
containing liquids is possible in various modes. In this paper instability of cylindrical shell wall which occurs under
simultaneous loading of axial compression and high internal pressure, is considered. This instability generally
occurs near the base. The present study demonstrates that increasing the internal pressure and the yielding of the
wall near the base, decreases flexural stiffness and increases the local displacement of the cylindrical shells. In other
words, the resultant circumferential membrane stress increases and inelastic buckling occurs. This instability near
the base is known as Elephant Foot Buckling. This paper provides a proposal for strengthening cylindrical shells
against elephant foot buckling using FRP. Some efforts have been done on this subject [1-3]. The aim of this paper
is to investigate the strengthening of thin metallic cylindrical shells by local application of FRP to increase the
elephants foot buckling strength. Finite element analysis of shells has been used in the present paper in order to
obtain the effect of strengthening tanks using FRP on the increase of elephant foot buckling resistance. In order to
achieve the behavior of cylindrical shell when being strengthened by FRP, accurate modeling of composite fibers
and adhesive has been applied, and its appropriate dimensions and locations are presented.

2.Finiteelementmodeldefinition
The finite element software package ABAQUS [4] was employed to simulate the samples in this study. The
samples geometry has been chosen in such a way that the radius to thickness ratio of them is in a suitable scale of
the real conditions. It is assumed that the tank is completely filled with liquid and the hydrostatic load is exerted on
the wall. The thickness of the tank is considered uniform at height. The steel cylinder was modeled using element
S4R. The properties of composite and steel materials are properly defined. To evaluate the performance of
composite materials, the linear elastic behavior of undamaged materials along with Hashin theory is modeled in the
software which offers four modes of failure in the form of fibers tension, fibers compression, matrix tension and
matrix compression. Starting of the failure in this model depends on the point of material stiffness degradation. The
CFRP is modeled in the form of unidirectional fiber, in horizontal direction with the thickness of 0.2 mm.
The simply-supported boundary conditions at the lower base of the shell are considered in the analysis. The
cylindrical tank is loaded in order to achieve the combination of maximum axial and circumferential stress which
leads to the occurrence of elephant foot buckling, in the way axial force are exerted from above in the presence of
the internal hydrostatic pressure. In the samples studied in this paper, internal hydrostatic pressure loading is
increased only in the highest value. The purpose of this kind of loading is to increase circumferential stress and
study its effect on elephant foot buckling of the samples near the base. The procedure of loading in the finite

*CorrespondingAuthor
Emailaddresses:morteza866@gmail.com(MortezaVakili),h.showkati@urmia.ac.ir(HosseinShowkati)
MortezaVakiliandHosseinShowkati/J.Civ.Env.Eng.45(2016)

element analysis is divided into two steps. In the first step, through nonlinear static analysis of the loading, the
internal pressure is hydrostatically applied and then while keeping its effect, in the second step and through Riks
analysis, the axial load is exerted above the cylindrical sample. In samples which are strengthened by FRP, the
interaction between the steel and FRP is modeled considering the properties of adhesive.

3.Resultsanddiscussion
3.1.Theeffectofchangesingeometryonthecapacityofcylindricaltanks
To study the effect of changes in geometry on tanks capacity, under different internal pressures and axial forces,
some samples with various ratios of thickness to radius have been considered. The ultimate capacity of the tanks
under various internal pressures shows the significant impact of internal pressure on reducing the capacity of these
tanks due to the axial force. Internal pressure-axial force interaction diagram is provided for all samples in various
ratios of thickness to radius in Fig. 1.

0.7
1:1250
0.6 t/R=1:714
t/R=1:833 1:1000
0.5 t/R=1:1000
1:833

0.4 1:714
Fz/Fy

1:1667
0.3

0.2
t/R=1:1667
0.1
t/R=1:1250
0
0 0.2 0.4 0.6 0.8 1 1.2
PR/tfy

Fig. 1. The interaction of internal pressure-axial force in various thickness to radius ratios

3.2.TheeffectofstrengtheningFRPcompositematerialsonthecapacityofcylindricaltanks
In order to study the effects of strengthening FRP composite materials on the capacity of cylindrical tanks, 8
cylindrical tanks with the same geometric conditions are considered which are categorized as follows: 6 samples
with maximum changes of hydrostatic pressure ranging from 0.05 to 0.2 MPa and two samples under mere internal
pressure and mere axial force in the two conditions of with or without FRP application. FRP height in this condition
is modeled according to the elephant foot buckling wave length of 20 mm and FRP materials location starting from
the base. Internal pressure-axial force interaction diagram of these tanks are illustrated in Fig. 2 in the two
conditions of strengthened with and without FRP.

1.2
1:1000(withoutFRP)
1
1:1000(withFRP)

0.8
fz/fcr

0.6

0.4

0.2

0
0 0.2 0.4 0.6 0.8 1 1.2
PR/tfy

Fig. 2. The interaction of internal pressure-axial force in the samples with and without FRP
MortezaVakiliandHosseinShowkati/J.Civ.Env.Eng.45(2016)

3.3.TheeffectofFRPheightonthecapacityofstrengthenedtanks
To study the effect of FRP height on the capacity of strengthened tanks, 25 cylindrical samples are modeled. The
FRP heights of the samples vary from 10 to 50 mm starting from the base and are under the internal pressure
ranging from 0.05 to 0.2 MPa. The optimal and minimum FRP height of 20 mm which is consistent with the
elephant foot buckling wave length is advisable.

3.3.TheeffectofFRPlocationonthecapacityofstrengthenedtanks
To study the effect of FRP position on the capacity of strengthened tanks, 5 cylindrical samples under the axial
force and with the internal pressure of 0.1 MPa and a fixed height of 20 mm for FRP, are investigated. The position
of FRP composite materials have changed by upward movement starting from the base. The results of sample
analysis shows that locating FRP in the area of elephant foot buckling occurrence which is positioned near the base
can lead to the best strengthening result.

4.Conclusions
In this paper, elephant foot buckling and retrofitting of steel thin walled cylindrical shells against that, using FRP
composite materials were investigated. The present results indicate the usefulness of this method of retrofitting to
prevent the occurrence of elephant foot buckling.
- Evaluation of ultimate capacity of tanks in different internal pressures show that increasing in the internal pressure,
leading to reduction in the capacity of the tanks due to axial force.
- The results show that in higher ratios of thickness to radius in tanks, the combination of internal pressure and axial
force leads to a lower buckling resistance. In other words, buckling occurs in lower load combinations in these
tanks.
- Evaluation of ultimate capacity of strengthened tanks in different internal pressures shows the positive effect of
FRP composites at the buckling resistance. Increase in the buckling resistance at higher internal pressures is much
more.
- The optimal and minimum FRP height which is consistent with the elephant foot buckling wave length is advisable
in strengthening tanks against elephant foot buckling.
- The results of samples analysis show that locating FRP in the area of elephant foot buckling occurrence which is
positioned near the base can lead to the best strengthening result.

5.References
[1] Teng, J. G., Hu, Y. M., Suppression of Local Buckling in Steel Tubes by FRP Jacketing, The 2nd International
Conference on FRP Composites in Civil Engineering, Adelaide, Australia, 2004, pp 749-53.
[2] Teng, J., Hu, Y. M., Behaviour of FRP-Jacketed Circular Steel Tubes and Cylindrical Shells under Axial
Compression, Construction and Building Materials, 2007, 21, 827-838.
[3] Batikha, M., Chen, J. F., Rotter, M., Fibre Reinforced Polymer Composites to Increase the Buckling Strength
of Imperfect Cylindrical Shells, International Conference on Structures and Granular Solids-From Scientific
Principles to Engineering Applications, 2007, pp 177-181.
[4] ABAQUS, Version 6.10.1. ABAQUS/Standard Users Manual, ABAQUS Inc., US, 2010.