International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149

Vol 7 No 1 2021 [Link]

Design Implementation Based on Wind Girders and Finite

Element Analysis Practices towards External Floating Roof
Storage Tank

* Corresponding author: *Adel Khalleefah Hamad Darmeesh1 , Amragah Geith

Bobaker Ibraheem2
*Head of the Department of Engineering Sciences
College of Engineering1
Faculty Member at Department of Materials and Minerals Engineering,
Ajdabiya University1 , Ajdabiya University2

Abstract
Above ground steel welded external floating roof storage tanks are thin-walled structure
widely used for as a storage to various petrochemical products. These tank are subjected to
deferent loading conditions such as bending moment, external and internal pressure and
wind loading. Due to its severity and occurrence wind loading is the most tank buckling
factor. The main objective of this paper was design tank wind girders according to API 650
standard calculations. At that point, to analyze various tank geometries stability against wind
loading. These geometries includes tank without wind girder, tank with one wind girder and
tank with two wind girders. Furthermore, the effect of the hydrostatic pressure to resist wind
loading was also considered. The analysis used by finite element analysis ANSYS v -15.0
software commercial package. The results show that stored liquid hydrostatic pressure
increases tank resistance to buckling due wind loading. It is also indicated that tank wind
girders improve tank stability against wind loading, particularly the upper wind girder. Then
and there, results suggests that finite element analysis ANSYS v-15.0 software commercial
package can be used as a prediction tool to investigate tank buckling against wind loading
and effect of stored liquid hydrostatic pressure.

Keywords: Storage tank, External floating roof tank, Tank Buckling, Wind Loading, Wind
Girder.

1. Introduction
Storage tanks have been broadly used to store various liquid products in the petrochemical
industry since the discovery of hydrocarbons in 1859. Storage tanks vary significantly, in the
type and size based on the products to be stored and the volume involved. There are a wide
variety of storage tanks; they can be erected above ground, in ground and below ground. In
shape, they can be in vertical cylindrical, horizontal cylindrical, spherical or rectangular form,
on the other hand, vertical cylindrical are the most used one (Kuan, 2009). Moreover,
external floating roof steel welded vertical cylindrical aboveground storage tank is the most
widely used storage tanks for storing crude oil and petroleum products, which will be
considered in this research. External floating roof storage tanks and open top tanks are thin-
walled structure and it is subjected to various loads including axial compression, global
bending, external or internal pressure and wind loading. Tank shell is susceptible to out of
roundness or buckling due to these loading conditions as presented in Figure. 1.1.
Consequently, the buckling behavior must be understood to design a stable tank to withstand
the mentioned loading and prevent tank shell buckling. Wind loading is the most common
cause of tank buckling due to its occurrence and consequences in both safety and cost. The

IIARD – International Institute of Academic Research and Development Page 20

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

wind loading will be more critical when the tank is empty or when it is partially filled. Even
in the fixed roof tanks, similar bucking due to wind loads may happen during its construction
phase before erecting the tank roof. This research focuses on the buckling behavior of the
external floating roof storage tank shell under wind loading condition and designs the
required stiffening rings. The simulation will be made in both serious situation, when the
storage tank is empty and when it is partially filled with liquid.

Figure. 1: Shell out-of-roundness caused by wind.

1.2. Problem Statement:

External floating roof storage tank may be subjected to buckling behaviour due to internal or
external loads including wind loading. This buckling behaviour plays a very important aspect
of tank settlement, particularly, when the tank is empty or partially filled of liquid and during
the construction phase. Although many researchers have investigated open top and external
floating roof tanks wind loading and its influence on the tank buckling, more studies need to
be conducted to understand buckling behaviour during wind loading for various tank
geometries at different operating conditions.

1.3. Research Objectives

The objectives of this present research could be summarised in sequence as:
1. Study the external floating roof storage tanks wind loading from the literature.
2. Design stiffening rings (wind girders) to protect the tank shell from buckling due to the
wind loading effects. The calculation design for the wind girder will be performed
using API 650 (2013) standard.
3. The effect of the wind loading in the tank shell particularly, when it is empty will be
simulated using the commercial CFD package finite element analysis (FEA) ANSYS.
This simulation will be made to the tank with and without wind girders at the shell.
4. Finally, the commercial CFD package (FEA) ANSYA simulation prediction results will
be evaluated comparing with API 650 design calculation and with previous works from
the literature.

1.4 Research Approach:

The first step will start with a study of the storage tanks background and buckling behaviour
from the literature in general and concentrating on the wind loading and the design and
calculation of the wind girder against tank wind loading at open top and external floating roof
storage tanks. Upon completion of the literature review of the storage tanks and wind loading

IIARD – International Institute of Academic Research and Development Page 21

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

and its buckling effects on the tank shell, then the research will approach into two ways.
Firstly, the American petroleum Institute API 650 standard for designing a welded steel tank
for oil storage will be used as a guide for designing and calculating the tank wind girders
shape, size and locations in the tank height. This calculation will depend upon the tank
diameter, plate thickness and tank height.
Secondly, an existing tank design data will be modelled using the commercial CFD package
finite element analysis ANSYS to evaluate the package's prediction ability. This simulation
will be done at various tanks filling situation particularly, at tank empty and half filled with
liquid.
Finally, the simulation prediction results from the package will be analysed and evaluated
comparing with this research design results and previous research results from the literature
to evaluate the package prediction capability.

2. Literature review
2.1 Vertical Cylindrical Storage Tanks:
Vertical cylindrical storage tank broken down into several types, comprising the open top
tank, fixed roof tank, external floating roof and internal floating roof tank as shown in Figure
2.1.

Figure. 2.: Storage tank types.

2. 1 Open top tanks:

This type of tank has no roof. They shall not be used for petroleum product, but may be used
for firewater or cooling water tank. The product is open to the atmosphere; therefore, it is an
atmospheric tank as presented in Figure 2.

2.1.2 Fixed roof tanks:

Fixed roof tanks can be divided into two types, cone roof and dome roof. They can be self-
supported or rafter / trusses supported, depending on the size. Figures 3.1 and 3.2 illustrate
each type respectively.

IIARD – International Institute of Academic Research and Development Page 22

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

Figure. 3.1: Open-top tank.

Figure. 3.2.: Fixed roof tank (Cone type).

IIARD – International Institute of Academic Research and Development Page 23

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

Figure. 3.3: Fixed roof tank (Dome type).

2.1.3 Floating roof tanks:

These types of tanks are designed to work at atmospheric pressure. Floating roof tanks is
which the roof lifts directly on top of the liquid product. The floating roof is consisting of a
deck, fittings, and rim seal system. Floating roof decks are erected of welded steel plates and
are of three general types: pan, pontoon, and double deck. The diameter of a floating roof
tank shall at least be equal to its height to enable the use of a normal rolling ladder for access
to the roof. There are two types of floating roof tanks. Internal floating roof is where the roof
floats on the product in a fixed roof tank. See Figure 3.4. As it is a fixed roof tank so it is
supported by the fixed roof.

[Link] External floating roof:

The external floating roof is where the roof floats on the liquid product in an open tank and
the roof is open to atmosphere as can be seen in Figure 3.5. Typical Products stored on an
external floating roof tank are crude oil, gasoline and gasoline components, solvents… etc..

IIARD – International Institute of Academic Research and Development Page 24

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

Figure. 3.4: Internal floating roof tank type.

Figure. 3.5: External floating roof tank type.

2.2. Storage Tanks General Buckling:

Deformation of storage tanks can be occurred due to various load arrangements. Buckling is
one of these loads which cause deformation to storage tanks. Buckling is defined as a rapid
sideways failure of a slim member structural exposed to high compressive stress. When this
high compressive stress in a thin-walled structure such as aboveground storage tanks
increases, buckling may occur and it may lead to structural deformation and shape changes

IIARD – International Institute of Academic Research and Development Page 25

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

from its original geometry as can be presented in Figure 3.6. Buckling can be stimulated by
the structure itself, such as its weight and inside vacuum pressure in the hollow structure or
by external reasons such as earth quack, snow, rain and wind. Due to its occurrence and
severity, wind is considered is one of the most causes of tank buckling.
Buckling in general has been studied by many researchers. An early paper by Gody and
Flores (2002) used the finite element analysis modelling to investigate the buckling of the
empty storage tanks. Four storage tank geometries were simulated to study the radius and
height (R/L) and radius thickness (R/t) of the tank. The results of this study demonstrated that
the thin-walled short tanks with R/L = 3 has a great chance to structure failure. Ayari et al
(2014) conducted a study to a large fixed roof tank with an internal floating roof. The fixed
roof was collapsing and the tank reconstructed, however, the functionality of the internal
floating roof must be examined due to the plastic deformation to the tank shell, which could
prevent the internal floating roof from travelling freely up and down. This study concludes
that the plastic deformation is associated to buckling of the tank shell due to the out-of-plan
settlement (OOP settlement).

2.3. Storage Tanks Wind Loading and Stiffening Ring:

Both open top tank and external floating roof tank are open at the top and closed at the
bottom. The wind load will affect this cylindrical tank stability and its roundness, as can be
seen in Figure 1.1, which will cause cracks to the shell welds and disturbing floating roof
from travelling freely up and down with the liquid level. Due to these effects of the wind
load, a suitable stiffens to these tanks is an essential factor during design and must be
considered. At the beginning of this section, the previous works on the tank buckling
concentrating on the wind loading effects will be presented below, then the stiffening ring
design studies will follow.

Figure. 3.6: Local buckling of a tank.

An initial paper by Ummenhofer and Knoedel (2000) studied the pressure distribution around
five different cylindrical steel structure shells models due to wind loading. The method used
for modelling was the numerical finite element analysis. This simulation study displays that
the maximum axial shell membrane stresses at the bottom flange is prominently affected by
the boundary conditions used during the modelling process. So selecting the appropriate

IIARD – International Institute of Academic Research and Development Page 26

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

modelling boundary conditions is essential variables for designing cylindrical steel structures
such as vertical steel welded storage tanks against axial buckling. Jaca et al (2007) they
investigate the buckling axisymmetric of cylindrical shell of a tank under non-axisymmetric
wind loads for reduced stiffness approach. The finite element analysis code was used and
modified to obtain the mode shape from the eigenvalue analysis. Their result indications that
the critical loads of reduced stiffness approach constitute lower bounds compared with the
both values calculated from nonlinear analysis and the experiment measurement results of the
tank shell. Borgersen and yazdani (2010) conducted a simulation study to a huge steel plate
tank failure during construction. The failure was a localized buckling with a high straight
wind load. The simulation method used was the finite element analysis method. This study
suggests that the tank localized buckling could be prevented by mounting a circumferential
stiffening angel around the top of the tank or connecting a circumferential stiffener flange on
the tank open-end edge.
Jahangiri, et al (2013) examined a storage tank subject to buckling due to wind load. The
analysis method used were finite element analysis method TK-Industrial program written in
ANSYS. The aim of this investigation was to evaluate this prediction package to predict the
buckling of the tank subjected to wind load and to assess the application of a stiffener and its
effects to the wind load. The results of this investigation showed that this package correlates
well and can be used to model any tank types including open-top, fixed roof and floating roof
tanks. Moreover, this investigation suggests that using a suitable stiffener ring will have great
effects to prevent tank-buckling occurrence due to wind load. Zhao and Lin (2014) studied
vertical steel tank subjected to buckling under wind load. They modelled six tanks with
various capacities and dimensions using finite element analysis ABAQUS package. This
research results presents that large storage tanks will be affected by wind loading. The
recommendation of this study suggests that during the high wind, the tank must have
sufficient liquid stored in the tank and for permanent solution; larger tanks must have a wind
girder to increase tank buckling resistance from wind load. Finally, and most essentially
finding that the tank buckling is ruled by windward positive pressure, however the wind
pressure in the other sections has no effect.
Uematsu et al (2015) offered a finite element analysis model with a turbulent boundary layer
for designing crude oil open top storage tanks with different arrangements, for open top tank
arrangements were modelled. This model was utilized to study the static wind loading and the
tank buckling due to the static wind loading. This model shows that the positive wind
pressure in the windward direction has the greatest effect on the buckling behaviour in the
open top tanks. Different techniques can be used, however, a tank stiffens ring (wind girder)
is the common solution which widely used to protect open top and floating roof from wind
load deformation. Mainly, two wind girder is used which are upper and intermediate wind
girders. For calculations of tank stability in strong winds, the velocities given in the local
regulations should be used; if no local regulations exist, local experience should be
considered. Several authors have studied large open top and floating roof storage tanks under
wind loading and designing a wind girder to support open top and floating roof tanks from
wind loading. These studies presented below in chronological order. Burgos et al (2015)
examined a simplified model for fixed roof support structure and open top with wind girder
using finite element analysis. Three loading situations were examined: thermal loading due to
nearby fire, constant external pressure and wind pressure. The results show that the simplified
model of the wind load is less than the real model value for the fixed roof tank. The
prediction correlates well between the real value and the simplified open top tank model with
wind girder. Bu and Qian (2016) used a two dimensional finite element analysis model
ANSYS to design top wind girder for large storage tank. The design includes the wind girder
section modulus magnitude and location. The results of this model compared with the

IIARD – International Institute of Academic Research and Development Page 27

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

calculation presented in the American petroleum institute API 650 for evaluation to the
package prediction. Moreover, this paper examined the tank bottom plate forces by finite
element analysis. The finding of this paper indicates that the tank bottom plate forces for
large storage tank must be considered for any numerical model for designing tank top wind
girder. On the same way, Azzuni and Guzey (2017) used finite element analysis to design top
wind girder for aboveground storage tank according to API 650 obligation. The aim of this
paper is to investigate the tank diameter limits for designing the top wind girder. This paper
results determined that the tank diameter limits which can be used for designing the wind
girder for large aboveground storage tank is 52 m. To conclude up, storage tank buckling is a
very critical factor affecting the storage tank stabilities, particularly for open top and external
floating roof tanks. Moreover, although the fixed roof tank has its own support due to its
fixed roof, it is may also affected by buckling load during its construction phase before
installing the roof. As we can see from all these studies, wind loading upon open and external
floating roof tanks is the major contributor of tank buckling due to its occurrence and
severity. Finally, these tanks must be supported by designing a suitable stiffening ring (wind
girder) to increase tank stabilities and prevent buckling failure.

3. WIND GIRDER DESIGN AND SIMULATION

3.1. Wind Girder Design Overview
Open-top and the external floating roof tank is a vertical cylinder that is open at the top and
closed at the bottom. This tank is constructed with different course plate thicknesses, high
thickness at the bottom and then the plate thickness decreases with the tank height due to the
hydrostatic pressure effect as shown in Figure 4.1. This tank will be affected by various
loadings such as wind loading and forcing it to become out of roundness and shell buckling
deformation, unless the shell alone or other means provide an adequate stiffening.
Theoretically, there are two ways to provide adequate stiffness:
 Increasing the shell thickness to withstand the deformation, or
 An additional stiffening method must be applied to support the tank shell from
deformation as illustrated in Figure 4.2.
On the other hand, increasing tank shell thickness, is not economical. Firstly, it will increase
the project costs due to the increasing of the steel quantities, increasing of transporting and
handling of these thick steel plates, as a result, and increasing welding passes due to the plate
thickness. Secondly, increasing the tank shell plate thickness will increase the weight exposed
to the bottom shells, bottom plates and bottom soil under the tank, which may affect tank
settlements. Finally, during tank construction, erecting thick plates at the top courses will
increase the potential risk exposed. In this research, the floating roof storage tank designed by
Ali and Omar (2017) presented in Table 3.1 will be used to design wind girders according to
API 650 standard calculation and selection. This calculated design result will be compared
with a simulation accomplished by ANSYS v-15.0 software package. Materials selected
according to strength fracture and toughness properties. Tank shell materials selected from
Table 5.2a from API 650.

IIARD – International Institute of Academic Research and Development Page 28

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

Figure. 4.1 Tank shell courses.

Figure. 4.2. Wind girder placements on tank shell.

Table 3.1 Basic storage tank design data

Tank size:
Diameter – 50 m Total height 17 m
Tank capacity: 195,250 USBBL
Shell courses: 7 courses, each Course 2.427 m height
Design liquid level: 15.81 m
Corrosion allowance: 3 mm (According to NOC standard for sour service)
Liquid specific gravity: 1
Maximum operating temperature: 84 C
Minimum operating temperature: 5 C

IIARD – International Institute of Academic Research and Development Page 29

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

Wind velocity: 180 km/h

Maximum rainfall: 80 mm/h
Shell plate thickness and materials
Shell course Thickness Material Tensile stress (N/mm2 ) Yield stress
(mm) (N/mm2 )

1 23 A537 CL1 485 345

2 20 A537 CL1 485 345
3 17 A537 CL1 485 345
4 14 A537 CL1 485 345
5 11 A537 CL1 485 345
6 8 A537 CL1 485 345
7 8 A283 GR C 380 205
Tank bottom:
Center plate thickness: 9 mm
Center plate material: A283 GR C
Annular plate thickness: 9 mm
Annular plate material: A537 CL1
Tank roof:
Roof type: External floating roof
Roof plate thickness: 8 mm
Roof plate material: A283 GR C

3.2 API 650 Design Approach

According to API 650, all open-top tanks must be provided with a stiffening ring (wind
girder). The stiffening rings shall be made on an angle with a minimum nominal plate
thickness of 6 mm and a minimum size of 65 x 65 x 6 mm, Figure 4.3 shows typical
stiffening ring sections for tank shell from the API 650 standard. These stiffening rings shall
be placed outside of the tank shells at or near the top shell course supported by sufficient legs
with appropriate spacing limits. A liquid drain hole with at least about 25 mm diameter shall
be drilled to drain trapped liquid from top of stiffening rings, which could accelerate the
corrosion rate to the wind girder and attached tank shell. The stiffening rings may be used as
a walkway. In this case, a projection width of the stiffening ring from the tank shell shall be
at least 710 mm. The walkway must be provided with a standard railing at the outside end.
Also, if a stairway is opened through the wind girder as shown in Figure 4.4, the distance
outside the opening shall conform to the section model design.
In addition, a top curb angle with dimensions of 75 x 75 x 6 mm must be constructed of shells
more than 5 mm thick, if the stiffening rings are positioned more than 0.6 m below the top of
the tank shell. To design a storage tank wind girders the following steps will be followed:
 Calculate the minimum required section modulus for the top wind girder.
 Determine whether a second intermediate wind girder is required or not.
 If the second intermediate wind girder is required, determine its location.
 Finally, if the second intermediate wind girder is required, calculate its minimum
required section modulus.

IIARD – International Institute of Academic Research and Development Page 30

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

Figure. 4.3. Typical wind girder sections for tank shell.

IIARD – International Institute of Academic Research and Development Page 31

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

Figure. 4.4 Stairways opening through wind girder.

3.2.1 Top wind girder calculations

The top wind girder calculation starts by calculating the required minimum section modulus,
then choosing the required wind girder size to provide the required section modulus. The
required minimum section modulus for the top wind girder can be determined by the
following equation:
2
D2 H 2  V 
Z   Equation (3.1)
17  190 
Where
Z is the required minimum section modulus, in cm3
D is the nominal tank diameter, in m.
H2 is the height of the tank shell, including any freeboard provided above the maximum
filling height in m.
V is the design wind speed, in km/h.
The section modulus (Z) of the wind girder shall include a portion of the tank shell for a
distance of (16t) below and above the shell ring attached.

Z
50 2 17   180  2
 
17  190 
Z = 2,256.25 cm3
From Table 5.20a from API 650 where the seventh course thickness is 8 mm and by choosing
detail-e from the typical stiffening ring sections of tank shells. We can define the length (b)
for the top wind girder, which is 1000 mm. The selected top wind girder type is shown in the
Figure 4.5 below.

IIARD – International Institute of Academic Research and Development Page 32

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

Figure. 4.5 Wind girder sections selected.

3.2.2 Specified top wind girder evaluation

To evaluate the top wind girder acceptance, firstly, it should be divided into four sections
according to its geometry as shown in Figure 4.6. Secondly, the following information for the
selected wind girder is needed to calculate the actual section modulus for the selected wind
girder. Then the calculated shall be made by the equations as shown below. Lastly, the actual
section modulus calculated must be compared with the minimum required section modulus
calculated by API 650 equation to confirm that the selected wind girder geometry and
dimensions are acceptable. The actual section modulus must be greater than the minimum
required section modulus of the selected wind girder details from API 650.
TS : Corroded thickness (nominal thickness minus CA) of the top shell course, cm 0.5
TW: Wind girder thickness, cm 0.6
b: Wind girder width, cm 100
HW: Wind girder height, cm 15
LW: Wind girder lip, cm 6.5

Figure. 4.6 Wind girder geometry.

IIARD – International Institute of Academic Research and Development Page 33

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

a) We calculate the area of each section (A1 , A2 , A3 and A4 ):

A1  32TS  TW TS Equation (3.2)
A1  ( 32  0.8 )  0.6 0.5  13.1cm 2
A2  bTW Equation (3.3)
A2  100  0.6  60 cm 2
A3  H W TW Equation (3.4)
A3  15  0.6  9 cm 2
A4  LW TW Equation (3.5)
A4  6.5  0.6  3.9 cm 2
b) We calculate the distance from the inside tank diameter to the centroid of each
separate wind girder section area (X1 , X2 , X3 and X4 ):
X 1  0.5TS Equation (3.6)
X 1  0.5  0.5  0.25cm
X 2  ( TS  0 . 5 b ) Equation (3.7)
X 2  ( 0.5  ( 0.5  100 ))  50.5cm
X 3  ( TS  b  0.5TW ) Equation (3.8)
X 3  ( 0.5  100  ( 0.5  0.6 ))  100.2cm
X 4  ( TS  b  0.5 LW ) Equation (3.9)
X 4  ( 0.5  100  ( 0.5  6.5 ))  97.25cm
c) We calculate the centroid of the combined area ( X ):
 ( AX )
X  Equation (3.10)
A
4314.36
X   50.17 cm
86
d) We calculate the distance of the centroid of each section area to the centroid of the
combined area (d1 , d2 , d3 and d4 ):
d1  X  X 1 Equation (3.11)
d 1  50.17  0.25  49.92cm
d2  X  X 2 Equation (3.12)
d 2  50.17  50.5  0.33cm
d3  X  X 3 Equation (3.13)
d 3  50.17  100.2  50.03cm
d4  X  X 4 Equation (3.14)
d 4  50.17  97.25  47.08 cm
e) We calculate the moment of inertia of each individual section area about its centroid
(I1 , I2 , I3 and I4 ):
( 32TS  TW )TS
3
I1  Equation (3.15)
12

IIARD – International Institute of Academic Research and Development Page 34

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

(( 32  0.5 )  0.6 )0.5 3

I1   0.17 cm 4
12
TW A 3
I2  Equation (3.16)
12
0.6  100 3
I2   50 ,000 cm 4
12
3
H T
I3  W W Equation (3.17)
12
15  0.6 3
I3   0.27 cm 4
12
3
T L
I4  W W Equation (3.18)
12
0 .6  6 .5 3
I4   13.73cm 4
12
f) We calculate the moment of inertia of the combined area (Io ):
Io   Ad 2   I Equation (3.19)
I o  63 ,823.27  50 ,014.17  113 ,837.44 cm 4
g) We determined the distance from the centroid of the combined area to each edge of
the combined area (C1 and C2 ):
C 1  TS  b  X Equation (3.20)
C1  0.5  100  50.17  50.33cm
C2  X Equation (3.21)
C 2  50.17 cm
h) We determined (C) the maximum distance C1 or C2 :
C  C1  50.33cm
i) Finally, we calculate the section modulus of the combined area (Z):
Io
Z Equation (3.22)
C
Z  113 ,837.44  2 ,261.82cm 4
50.33
The calculation summary results of the wind girder evaluation can be seen in Table 3.2.
Furthermore, since the calculated actual section modulus of the combined wind girder area is
higher than the required area modulus (2,261.82 cm3 ˃ 2,256.25 cm3 ), then the selected wind
girder details are acceptable. The top wind girder shall be located in the top course normally
it is welded below the tank shell top by one meter distance. This wind girder’s location will
increase the stiffness of the thin top shell courses and it will be used as a walkway for the
floating roof tank as well.

IIARD – International Institute of Academic Research and Development Page 35

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

Table 3.2 Top wind girder evaluation calculation summary

Section A X AX D I Ad2
1 13.1 0.25 3.28 49.92 0.17 32,645.28
2 60 50.5 3,030 -0.33 50,000 6.53
3 9 100.2 901.8 -50.03 0.27 22,527.01
4 3.9 97.25 379.28 -47.08 13.73 8,644.45
∑A ∑(AX) ∑I ∑(Ad2 )
Total 86 4,314.36 50,014.17 63,823.27

3.2.3 Intermediate wind girder calculations

When the distance between the top wind girder and the bottom of the tank is too large in a
particular tank data (tank height, tank diameter, and tank shells course thickness), the tank
could not resist wind loading with the top wind girder alone. In this situation, an intermediate
wind girder between the top wind girder and tank bottom must be installed to increase the
tank stiffness and shell deformation resistance to wind loading. This intermediate wind girder
shall be designed according to the following steps as indicated by API 650 standard:
• Determine if an intermediate wind girder is required according to tank data and design
wind speed.
• Calculate the intermediate wind girder position from the top wind girder.
• Calculate the minimum needed section modulus of the intermediate wind girder and
choose a typical shape according to this section modulus.

[Link] Requirements of intermediate wind girders

The aboveground storage tank is a thinned-wall structure with shell plates are thicker at the
bottom and their thickness decrease as the tank going from bottom to top. So, the stiffness of
the upper portion of the tank is weaker than the lower portion. Thus, if the intermediate wind
girder is required it must not install in the middle between the upper wind girder and tank
bottom, but it shall be installed above the middle of the tank and closed to the upper weak
portion to increase tank stability. This tank shell courses variation shall be considered during
the locating of the intermediate wind girder. According to API 650, the actual tank shell
height shall be converted mathematically to transformed tank shell height as illustrated in
Figure 4.7. By this conversion, the transformed tank shell will have the same stiffness
throughout the tank height. So, the intermediate wind girder must be installed in the middle
of the transformed tank shell courses. Then, by installing the wind girder in the same shell
course in the actual tank shell, the wind girder will be located above the mid-height of the
actual tank shell.

IIARD – International Institute of Academic Research and Development Page 36

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

Figure. 4.7 Actual and transformed shell.

The maximum height of the unstiffened tank shell can be calculated by using the following
formula;
3 2
 t   190 
H 1  9.47 t     Equation (3.23)
D  V 
where;
H1 is the vertical distance between the intermediate wind girder and the top wind girder,
in m.
t is the nominal top shell course, in mm
D is the nominal tank diameter, in m.
V is the design wind speed, in km/h.
By substituting the tank date;
3 2
 8   190 
H 1  ( 9.47 )( 8 )    
 50   180 
We obtain;
H1 = 5.4 m
Then, the transformed tank shell course height can also be computed by this formula below;
5
 t uniform 
Wtr  W  
 Equation (3.24)
 t actual 
where;
Wtr is the transformed shell course width, in m.
W is the actual shell course width, in m.
t uniform is the nominal thickness (design thickness minus CA) of top shell course, in cm.

IIARD – International Institute of Academic Research and Development Page 37

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

t actual is the nominal thickness (design thickness minus CA) of the shell course, being
calculated, in cm.

Table 3.3 Transformed shell calculation summary

5
 t uniform 
Course W Wtr  W  

 t actual 
5
 0 .8  0 .3 
1 2.427 Wtr  2.427    0.075
 2 .3  0 .3 
5
 0 .8  0 .3 
2 2.427 Wtr  2.427    0.114
 2  0 .3 
5
 0 .8  0 .3 
3 2.427 Wtr  2.427    0.185
 1 .7  0 .3 
5
 0 .8  0 .3 
4 2.427 Wtr  2.427    0.337
 1 .4  0 .3 
5
 0 .8  0 .3 
5 2.427 Wtr  2.427    0.750
 1 .1  0 .3 
5
 0 .8  0 .3 
6 2.427 Wtr  2.427    2.427
 0 .8  0 .3 
5
 0 .8  0 .3 
7 2.427 Wtr  2.427    2.427
 0 .8  0 .3 
Wtr total   Wtr 5.33 m

From these calculations, we noted that;

Wtr  total ( 5.63m )  H 1 ( 5.4 m )
As the transformed shell height is greater than the vertical distance between the intermediate
wind girder and the top wind girder (H1 ), then an intermediate wind girder is required.
 Wtr total   6.3 
Also,     3.15 m  H 1 ( 5.4 m ) . Moreover, since total transformed shell
 2   2 
width divided by two is equal to or less than the vertical distance between the intermediate
wind girder and the top wind girder (H1 ), then only one intermediate wind girder is required.

[Link] Intermediate wind girder location

The intermediate wind girder must be installed in the midpoint of the transformed shell to
make the shell stiffness similar in both between the top wind girder and the intermediate wind
girder and between the intermediate wind girder and tank bottom. This location in the same
course in the actual shell must be located using relative transformation position. As the
 Wtr total 
   3.15 m  H 1 ( 5.4 m ) Then the intermediate wind girder shall be located at 3 m
 2 

IIARD – International Institute of Academic Research and Development Page 38

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

from the top wind girder, which measures 4 m from the tank top. This made the rest of the
transformed shell height of the intermediate wind girder to the tank bottom equal to 2.3 m
and will be in a suitable stiffness.

[Link] Intermediate wind girder size

The top wind girder calculation starts by calculating the required minimum section modulus,
then choosing the required wind girder size to provide the required section modulus. The
required minimum section modulus for the top wind girder can be determined by the
following equation:
2
D2 H1  V 
Z   Equation (3.25)
17  190 
where,
Z is the required minimum section modulus, in cm3
D is the nominal tank diameter, in m.
H1 is the vertical distance between the intermediate wind girder and the top wind girder,
in m.
V is the design wind speed, in km/h.
By substituting into the equation above, we obtain,

Z
50  3   180 
2 2
 
17  190 
Z = 396.18 cm3
From Table 5.20a in API 650 where the seventh course thickness is 8 mm and by choosing
detail-e from the typical stiffening ring sections of tank shells. We can define the length (b)
for the intermediate wind girder, which is 300 mm. The selected top wind girder type is
shown in the Figure 3.5.

3.3 Computational Model

A 3D floating roof storage tank designed by Ali and Omar (2017) according to API 650
standard presented in Table 3.1 with different shell course thickness will be used in this study
with the authors wind girder design and selection according to API 650 standard. A 3D tank
geometry will be modeled using the commercial software package ANSYS v-15.0. Due to the
tank symmetries and to reduce time and space required from the solver, only half of the tank
will be modeled. Moreover, the model assumed to have the same material properties (A537
CL1) with properties also shown in the Table 3.1.

3.3.1 Finite element structural model

To evaluate the wind girders performance to prevent wind buckling to the storage tank, three
geometries were modeled in this study. Firstly, a tank without wind girders, secondly, a tank
with only top wind girder, and finally a tank with both top and intermediate wind girders
were modeled. A finite element analysis commercial software package ANSYS v-15.0 will
be used to sketch the tank geometries and generate mesh to each model. A three-dimensional
medium size mesh of Tetrahedral CFD cells will be used to model all the geometries
mentioned. The first tank model without wind girders, the second model with only top wind
girders and the third model with both top and intermediate wind girders contains about 9393,
13159 and 15994 elements respectively. The three models geometries with its mesh and wind
girders view are shown in figures from Figure .4.8 to as presented in Figure 4.15. These
three tank models will be subjected to the wind simulated as a velocity pressure with various

IIARD – International Institute of Academic Research and Development Page 39

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

liquid levels to examine the wind girders performance to prevent tank buckling due to wind
loading.

3.3.2. Solution Procedure

Finite element analysis scheme contains three main running stages, preprocessing, solving
and post processing. The first stage is preprocessing where the structure geometry modeled in
the computer design modeler then the required mesh will be generated to divide the domain
into small elements in space and apply the subjected loads according to problem boundary
conditions. The mesh type and size must be selected appropriately to obtain more accurate
results with less computer processer effort. The second stage is the finite element solver
where the package starts solving the problem for each elements individually, then joins the
elements solution together to form the whole solution to the structure modeled. The third and
final stage in finite element analysis is post processer where the solution results from the
finite element solver, after achieving equilibrium condition, can be exported to the post
processer for further interpretation and explanation. In this stage the results can be presented
as a pictures, graphs and plots to study the problem according to the package prediction
capability. Finally, the predicted results shall be compared with numerical calculations or
experimental measurements for verification to the finite element analysis package predicted
outputs.

Figure. 4.8 Schematic view of half of the tank without wind girders.

IIARD – International Institute of Academic Research and Development Page 40

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

Figure. 4.9 Mesh model for half of the tank without wind girders.

Figure. 4.10 Schematic view of half of the tank with top wind girders only.

IIARD – International Institute of Academic Research and Development Page 41

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

Figure. 4.11 Mesh model for half of the tank with top wind girders only.

Figure. 4.12 Schematic view of half of the tank with both top intermediate wind girders.

IIARD – International Institute of Academic Research and Development Page 42

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

Figure. 4.13 Mesh model for half of the tank with both top intermediate wind girders.

Figure. 4.14 Magnified view of the two wind girders geometry combination.

IIARD – International Institute of Academic Research and Development Page 43

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

Figure. 4.15 Magnified view of the mesh model for the two wind girders combination.

3.3.2 Wind simulation

Wind pressure is very important for any structures analysis and must be considered during the
design of these structures, which exposed to wind load. Wind pressure fluctuating randomly
due to the random fluctuating of the wind phenomena with space and time. Therefore, in
practice and for simplicity, wind pressure must be converted to static pressure. API 650
provides the formula to calculate the velocity pressure.

[Link] The velocity pressure

From API 650 standard the velocity pressure can be calculated from the equation below as
following:
P  0.00256 K Z K Zt K d V 2 IG  1.48 kPa Equation (3.26)
where;
Kz is the velocity pressure exposer coefficient = 1.04 for exposure C at height of 12.2 m.
Kzt is 1.0 for all structures except those on isolated hills or escarpments.
Kd is the directionality factor = 0.95 for round tanks.
V is the wind speed at 10 m and = 190 km/h.
I is the important factor = 1.0 for Category-II structures.
G is the gust factor = 0.85 for exposure C.
As the wind pressure is uniform around the tank shells theoretical buckling approach the
shape factor is not required and can be neglected. For open top tanks the inward drag, which
equal to 0.24 kPa, shall be added to the velocity pressure as an internal vacuum. Then, the
total velocity pressure can be calculated as;
P  1.48  0.24
P = 1.72 kPa

[Link] Wind pressure distribution

The middle wind attacking point at the edge of the tank, which shown in Figure 4.16 as point
number (1), is the zero point and the pressure at this point is stagnation pressure Jahangiri, et
al (2013). Figure 4.17 illustrations the storage tank attacking wind direction, showing the zero
point at the attacking edge and angle Ɵ starting from the zero point.

IIARD – International Institute of Academic Research and Development Page 44

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

Figure. 4.16 Air streamlines around the tank.

Figure. 3.17 Sketch of a tank exposed to wind.

The normalized pressure then can be calculated by the equation below;

Actual Pr essure
Normalized Pr essure  Equation (3.27)
Stagnation Pr essure
where;
Actual pressure is the pressure at the point with angle Ɵ.
Stagnation pressure is constant = 1.72 kPa.
Many researches and design codes introduce the distribution pattern of the wind pressure
coefficient around the storage tanks [Cp(Ɵ)] such as Ummenhofer and Knoedel (2000), Sosa
(2005), Jahangiri (2013) and (Rish 1967) as presented in Figure 3.18 by using Fourier series
expression as following;
C P (  )  i0 aiCos( i ) Equation (3.28)
where;
Ɵ is the longitude measured from the windward.

IIARD – International Institute of Academic Research and Development Page 45

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

ai is the Fourier coefficient.

Jahangiri (2013)

Figure. 4.18 External wind pressure coefficients along the circumference of tanks
(Figure from Jahangiri at al 2013)

4.2 API 650 Results and Discussions

The required section modulus for the upper wind girder was calculated and it is value is
2,256.25 cm3 . Then the wind girder detail is selected to be as the geometry shown in Figure
3.5 with length (b) is 1000 mm. Then, the actual section modulus of the combined wind
girder area selected was calculated and found to be 2,261.82 cm3 . So, as the actual section
modulus of the combined wind girder area selected is higher than the required area modulus,
then the selected details of wind girder are acceptable. On the contrary, the transformed shell
height is calculated and it is result indicates that an intermediate wind girder is required due
to it is value 5.63 meter which is greater than the vertical distance between the intermediate
wind girder and the top wind girder which is determined as 5.4 meter. Then it is section
modulus was calculated as 396.18 cm3 and selected to be the same geometry as upper wind
girder with length (b) is 300 mm.

4.3 Simulation Results and Discussions

In this section liquid hydrostatic analysis will be presented first followed by wind pressure
analysis for filled tank and empty tank. The empty tank analysis will consists of tank without
wind girders, with only upper wind girder and with two upper and intermediate wind girders.
In this simulation only half of the tank along the symmetry axis will be modeled to speed up
solution process and reduce analysis result data stored.

4.3.1 Liquid hydrostatic analysis

For hydrostatic analysis we use water as filled liquid with density of 1000 kg/m3 . As can be
seen in Figure 4.1 (a) the hydrostatic pressure increased at tank bottom with maximum value
of 155.01 kPa. To evaluate the package hydrostatic pressure prediction we calculate the
hydrostatic pressure by the formula;
P  gh Equation (4.1)

IIARD – International Institute of Academic Research and Development Page 46

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

where
ρ is the water density, in kg/m3
g is the acceleration due to gravity, in m/s2 .
h is the height of the filled liquid, in m.
By substituting we obtain,
P  1000  9.81  15.81
P = 155.01 kPa
It is evidence that the finite element analysis ANSYS v-15.0 predicts the hydrostatic pressure
at the tank accurately. The equivalent stress results due to hydrostatic pressure increases from
bottom of the tank to the top and presented in Figure 4.1 (b) with minimum value of 17.941
kPa to maximum value of 72.989 MPa at the top of the tank. The increases of the stress with
tank height possibly caused due to decreasing of the tank shell thickness with the tank height.
This equivalent stress caused a total deformation increases with the tank height from zero at
the tank bottom to 6.7 mm at the top of the tank as displayed in Figure 5.1 (c).

a) Hydrostatic pressure distribution b) Equivalent stress

c) Total deformation
Figure 5.1. Hydrostatic pressure results.

4.3.2 Wind pressure analysis for filled tank

To determine the stored liquid influence to support tank against wind load, a filled tank
subjected to wind loading was analyzed. As presented in Figure 5.2 (a) represents the
Equivalent stress through the tank bottom and shells. As can be seen from this figure the

IIARD – International Institute of Academic Research and Development Page 47

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

Equivalent stress almost uniformly distributed and increased along the tank height. Figure 5.2
(b) shows the tank total deformation due to both hydrostatic pressure and wind loading. The
maximum total deformation due to these loads together was 7.1 mm whereas the total
deformation due to hydrostatic was 6.7 mm. So, the wind loading causes a slight deformation
value which is 0.4 mm. As expected it seems that the hydrostatic pressure for the filled tank
increases the tank resistance to the deformation due to wind loading.

a) Equivalent stress b) Total deformation

Figure 5.2 Wind load analysis results for filled tank

4.3.3 Wind pressure analysis for empty tank

As the tank subjected to wind loading buckling could be occurred particularly when it is
empty. This section will focus on the buckling analysis of empty tank when it is subjected to
wind loading. Three tank geometries will be modeled in this section. First geometry is a tank
without wind girders, then the tank with only one wind girder and finally, the tank with two
wind girders.

[Link] Storage tank without wind girders subjected to wind loading

As presented in Figures 5.3 (a) and (b) show the equivalent stress and total deformation for
the haul tank respectively for empty tank without wind girders subjected to wind loading. The
equivalent stress variation along the tank circumference increasing from zero angle of
windward direction up to 90° with maximum value is 2.158 MPa then, declined and return to
nearly zero value at 180°.
This stress variation caused total deformation difference along the tank circumference. This
deformation has higher value of 0.375 mm at windward direction and lower value of 0.083
mm at the other end (at 180°) from the windward direction. The difference between these
values reads 0.292 mm.

IIARD – International Institute of Academic Research and Development Page 48

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

a) Equivalent stress b) Total deformation

Figure. 5.3 Wind load analysis results for empty tank without wind girders

Figure .5.4 presents the distribution pattern for pressure coefficient around the storage tank
circumference from deferent authors. These reading were taken from windward to half of the
tank geometry. Three data were presented in this graph, Jahangiri (2013), Rish (1967) and
ACI-ASCE (1991). The equivalent stress factor along the tank circumference prediction
using the commercial package ANSYS v-15.0 was show in Figure 5.5 after reversing its sign
because the pressure is the intensity of external force whereas stress is the intensity of
internal resisting forces at a point. This graph starts at factor one and decreased sharply up to
one third of the circumference distance to enter the negative section, then increased gradually
until it reaches slightly near zero reading and finally levels up to the end of the graph. The
predicted equivalent stress graph has a fair waveform comparing to the literature data.

Jahangiri (2013)

Figure. 5.4 External wind pressure coefficients along the circumference of tanks
(Figure from Jahangiri at al 2013)

IIARD – International Institute of Academic Research and Development Page 49

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

1.50E+00
Equivalent Stress Coefficient
1.00E+00

5.00E-01
Equivalent Stress Coefficient Factor

0.00E+00
0 10 20 30 40 50 60 70 80

-5.00E-01

-1.00E+00

-1.50E+00

-2.00E+00

-2.50E+00

Tank Circumference (m)

Figure. 5.5 External equivalent stress coefficients along the circumference of the tank

[Link] Storage tank with one wind girder subjected to wind loading
To examine the influence of the tank upper wind girder to tank buckling due to wind loading
pressure, the tank with only one wind girder is simulated and its analysis results can be seen
in Figures 5.6 (a) and (b) respectively. The equivalent stress has a gradual increases to reach a
higher value of 2.307 MPa at near 90° and then decreases to a value closed zero at 180°.this
stress has a corresponding total deformation from 0.360 mm to 0.080 mm at zero and 90°
from the windward direction with a deference of 0.28 mm. Although the amount of
deformation is slightly small value, the upper wind girder increased the stiffness of the tank
and improve the buckling strength of the tank against wind loading.

a) Equivalent stress b) Total deformation

Figure. 5.6 Wind load analysis results for empty tank with one wind girder

[Link] Storage tank with two wind girders subjected to wind loading
The final tank geometry was a tank with two wind girders. The analysis results of the tank
with two wind girders subjected to wind loading is presented in Figures 5.7 (a) and (b)
respectively. In contrary to the previous analysis of the tank with one wind girder, the tank
geometry with two wind girders shows lower equivalent stress and higher deformation at
windward direction comparing with only one wind girder tank. It is possible that adding of

IIARD – International Institute of Academic Research and Development Page 50

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

second wind girder caused more strength to the upper shells which reduces the stress in that
region. This unexpected prediction result may be explained by that the design criteria of the
second wind girder needs to be reviewed.

a) Equivalent stress b) Total deformation

Figure. 5.7 Wind load analysis results for empty tank with two wind girders

4.4 Summary
The equivalent stress variation along the three storage tank geometries, the first without wind
girder, the second with only upper wind girder and the third with two wind girders were
shown in Figures .5.8 (a), (b) and (c) respectively. This data was plotted together for
comparison in Figure 5.9. As can be seen from the graph, all the three geometries starts
together at 0.2 MPa and increasing gradually until the maximum values which are 1.9693
MPa, 1.6987 MPa and 1.7736 MPa approximately at length of 50 m, 45 m and 40 m for the
three geometries respectively. Then all the trends declined with fluctuation until they meet
together again at about 1.15 MPa stress value. The data for the tank without wind girder has a
higher value then the tank with only one wind girder has lower value. However, surprisingly
the tank with two wind girders has higher value than the tank with one wind girder. This
indicates that the upper wind girder improves the tank buckling against wind loading whereas
the second wind girder shows ambiguous prediction. This finding also noted by Zhao and Lin
(2014) which indicates that the design criteria of the second wind girder required more
consideration.

IIARD – International Institute of Academic Research and Development Page 51

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

a) Tank without wind girders

b) Tank with one wind girder c) Tank with two wind girders
Figure. 5.8 Equivalent stress variation along the tank circumference

2.00E+06

Tank without Wind Girders

1.80E+06
Tank with One Wind Girder
1.60E+06
Tank with Two Wind Girders

1.40E+06
Equivalent Stress (Pa)

1.20E+06

1.00E+06

8.00E+05

6.00E+05

4.00E+05

2.00E+05

0.00E+00
0 10 20 30 40 50 60 70 80 90

Tank Circumference (m)

Figure. 5.9 Schematic view of equivalent stress along the tank circumference

IIARD – International Institute of Academic Research and Development Page 52

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

5.1 Conclusion:
In this research the tank buckling resistance against wind loading will be studied. This
complex problem study includes a detailed design and calculations to select tank wind girders
using API 650 standard. Two tank wind girders were selected according to API 650 design
calculations. Then, a wind simulation analysis to the storage tank was conducted using finite
element analysis ANSYS V-15.0 software. Series simulation analysis to the storage tank were
undertaken. The effect of hydrostatic pressure on tank deformation and its effect during wind
loading were studied. The effect of wind girders to resist wind loading when the tank is
empty were then explored. Three tank geometries subjected to wind loading were
investigated, a tank without wind girder, then a tank with upper wind girder and finally a tank
with two wind girders. The main conclusion from this research can be summarized as
following:
1. The stored liquid hydrostatic pressure has a great influence for supporting the storage
tank against wind loading. The finite element analysis ANSYS v-15.0 predicts the
hydrostatic pressure perfectly.
2. The decreases of the upper tank shell courses reduces the tank buckling resistance
against wind loading. So, carful shell course design should be used including
corrosion allowance factor.
3. The storage tank buckling due to wind loading is governed by windward positive
pressure region, while other regions has less effect.
4. The wind girders increase the storage tank buckling resistance against wind loading.
Although, the second wind girder did not show extra improvement to buckling due to
wind loading.

5.2 Recommendations:
It is hoped that the researchers and tank designers will have benefits from this research to
understand the wind loading complex problem, which affection the tank buckling stability.
Moreover, this research may assist for the design of tank wind girders to support the tank
during wind loading. Although a lot of data were available from researchers regarding wind
loading and its effect on buckling stability of storage tanks, there are many areas remained
required further investigation. Some of these areas which required future investigation are
listed below:
 Various tank types (fixed roof and floating roof) and sizes should be examined.
 The effect of material used should be studied in more details.
 Wind pressure simulation using fluid flow analysis should be considered.
 An experimental measurement should be conducted to evaluate the finite element
analysis package prediction capability.

REFERENCES
Ayari, H., Truong, D., Sehn, J., and Truong, K. T., 2014. A Nonlinear Finite Element
Study on
Settlement and Releveling Procedure of a Large Deformed Steel Tank. Journal of
Pressure Vessel Technology, 136(3), 034502.
Azzuni, E., Guzey, S., 2014. Stability of Open Top Cylindrical Steel Storage Tanks: Design
of
Top Wind Girder. Journal of Pressure Vessel Technology, 139(3), 031207.
Borgersen, S. and Yazdani, S., 2010. Finite element analysis of wind induced buckling of
steel
tank. Challenges, Opportunities and Solutions in Structural Engineering and
Construction, CRC Press, Boca Raton FL, pp.157-160.

IIARD – International Institute of Academic Research and Development Page 53

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

Bu, F., and Qian, C., 2016. On the rational design of the top wind girder of large storage
tanks.
Thin-Walled Structures, 99, 91-96.
Burgos, C. A., Batista-Abreu, J. C., Calabró, H. D., Jaca, R. C., and Godoy, L. A., 2015.
Buckling estimates for oil storage tanks: Effect of simplified modelling of the roof and wind
girder. Thin-Walled Structures, 91, 29-37.
Geyer W. B., (2011) Handbook Of Storage Tank Systems-Codes, Regulations, And Designs,
Marcel Dekker Inc., New York, [ISBN 0-8247-8589-4]
Godoy, L. A., and Flores, F. G., 2002. Imperfection sensitivity to elastic buckling of wind
loaded open cylindrical tanks. Structural Engineering and Mechanics, 13(5), 533-542.
Jaca, R. C., Godoy, L. A., Flores, F. G., and Croll, J. G., 2007. A reduced stiffness
approach
for the buckling of open cylindrical tanks under wind loads. Thin-Walled Structures,
45(9), 727-736.
Jahangiri, M., 2013. Computational buckling analysis of wind loaded cylindrical storage
tanks.
Majlesi Journal of Energy Management, 1(4).
Kuan, S. Y., 2009. Design, Construction and Operation of The Floating Roof Tank, Iranian
Journal of Health, Safety & Environment, Vol. 2, No.4, pp.385-402.
Myers P. E., 1997. Aboveground Storage Tanks, McGraw-Hill, [ISBN 0-07-044272-X].
Rish, R. F., 1967. Forces in Cylindrical Shells due to Wind. P. I. Civil Eng-Civ. En., 36, pp
791-803.
Sosa E. M., 2005. Computational Buckling Analysis of Cylindrical Thin-Walled
Aboveground
Tanks. PhD Dissertation, University of Puerto Rico, Mayaguez, Puerto Rico.
Standard API Standard 650, March 2013. Welded Tanks For Oil Storage. Twelfth
.Edition,
American Petroleum Institute.
Uematsu, Y., Yasunaga, J. and Koo, C., 2015. Design wind loads for open-topped storage
tanks
in various arrangements. Journal of Wind Engineering and Industrial Aerodynamics,
138, pp.77-86.
Ummenhofer, T. and Knoedel, P., 2000. Modelling of boundary conditions for cylindrical
steel
structures in natural wind.
Zhao, Y., and Lin, Y., 2014. Buckling of cylindrical open-topped steel tanks under wind
load.
Thin-Walled Structures, 79, 83-94.

IIARD – International Institute of Academic Research and Development Page 54

 International Journal of Engineering and Modern Technology E-ISSN 2504-8848 P-ISSN 2695-2149
Vol 7 No 1 2021 [Link]

APPENDICES
Appendix-A

IIARD – International Institute of Academic Research and Development Page 55