ISRASE First International Conference on Recent Advances in Science & Engineering -20014 (ISRASE-2014)
Wind and Buckling Analysis of Natural Draught
Cooling towers using ANSYS
Sachin Kulkarni, P.G Student
Department of Civil Engineering (Structural)
BLDEA‟S V.P Dr P. G Halakatti College of Engineering & Technology
Bijapur 586101, Karnataka, India [email protected]
Prof A. V. Kulkarni, Associate Professor & Academic coordinator
Department of Civil Engineering (Structural)
BLDEA‟S V.P Dr P. G Halakatti College of Engineering & Technology
Bijapur 586101, Karnataka, India
[email protected]
ABSTRACT
Reinforced Concrete Cooling towers of hyperbolic shell configuration find wide spread application in utilities engaged in the
production of electric power. Natural Draught hyperbolic cooling towers are characterizing land marks of powe r stations. They
comprise of a thin concrete shell of revolution are common place in civil engineering infrastructure. The wind load is
always the dominant load in the design of the cooling tower due to its large size, complex geometry and thin wall. This paper
deals with the study of wind and buckling analysis of two existing cooling towers of 143.50m and 175.50m high above ground
level. Two existing coolin g towers are chosen from Bellary thermal power station (BTPS) as case study. These cooling towers
have been analyzed for wind loads using ANSYS software by assuming fixity at the shell base. The analysis of two existing
cooling towers has been carried out using 8 noded SHELL 93 element with uniform SHELL thicknesses. The wind loads on
these cooling towers have been calculated in the form of pressure by using design wind pressure coefficient given in IS 115041985 code & design wind pressure at different levels as per IS 875 (Part3)-1987 code. Eigen buckling analysis has been carried out
for two existing cooling towers.
Maximum deflection, Maximum Principal Stress & Strain, Maximum Von Mises stress & strain are obtained. The variation in
max principal stress v/s thickness, maximum deflection v/s thickness is plotted graphically.
Keywords—Cooling tower, SHELL element, Stress, Strain, Mises, wind load
I.
INTRODUCTION
Design and construction of efficient and economic Reinforced Concrete (RC) Hyperbolic cooling towers have driven the
engineers toward the design of tall and thin-shell towers which have considerable high slenderness aspect ratio. Consequently, the
shell of R.C. cooling towers with relative high slenderness aspect ratio is extremely prone to buckling instability due to wind
loading. In the absence of earthquake loading, wind constitutes the main loading for the design of natural draught cooling towers.
To increase the structural stability or buckling safety factor, one economic approach is to design and construct stiffening r ings for
the R.C Hyperbolic cooling towers. The analysis of these towers is an interesting and challenging to any structural engineer in
view of their size and shape.
II.
LITERATURE SURVEY
Research works have been reported in the literature on wind load acting on cooling towers [1 to 5]. G. Murali, C. M Vivek
Vardhan and B. V. Prasanth kumar Reddy [1] Response of cooling towers to wind loads. This paper deals with the study of two
cooling towers of 122m and 200m high above ground level. Meridional forces and bending moments have been calculated. G.
Murali [2] Response of Natural Draught Cooling Towers to Wind loads. This paper deals with the study of five cooling towers of
122m, 177m, and 200m, high above ground level with different throat height to total height ratio‟s, throat diameter to base
diameter ratio‟s and diameter to thickness ratio‟s. The results of the analysis include membrane forces, meridional force and hoop
force and bending moments. Mungam and wittek [3] studied the assessment of the design wind loads acting on RC cooling tower
shell under the turbulent wind. Comparison between several methods was also performed using data measured by wind tunnel on
an isolated RC cooling tower shell. The quasi-static response of an isolated RC cooling tower shell under the turbulent wind was
compared with results obtained with the design wind load by means of GRF, LRC and optimized peal load-distributions methods.
Orlando [4] studied the wind induced interference effect on two adjacent cooling towers through pressure measurements on
cooling tower models in a boundary layer wind tunnel. Further numerical linear analyses were performed to calculate the
structural responses of the isolated and grouped towers. Prashanth N, Sayeed sulaiman [5] This paper deals with study of
2014-15 ICRASE Copy right ©by ISRASE
ISRASE eXplore digital library
ISRASE First International Conference on Recent Advances in Science & Engineering -20014 (ISRASE-2014)
hyperbolic cooling tower of varying dimensions and RCC shell thickness, for the purpose of comparison an existing tower is
considered, for other models of cooling tower the dimensions and thickness of RCC shell is varied with respect to reference
cooling tower.
A Description of the Geometry of the Towers
Bellary thermal power station (BTPS) is a power generating unit near Kudatini village in Bellary district, Karnataka state. Two
existing cooling towers are considered as case study as shown in Fig 1 & 2. BTPS is geographically located at 15º11‟58” N
latitude and 76º43‟23” E longitude.
Details of existing cooling towers
1) The Total height of the tower is 143.50 m. The tower has a base, throat and top radii of 55 m, 30.5 m and 31.85 m respectively,
with the throat located 107.75 m above the base. (Unit No- 2 Cooling tower in BTPS)
2) The Total height of the tower is 175.50 m. The tower has a base, throat and top radii of 61 m, 34.375 m and 41.00m
respectively, with the throat located 131.60 m above the base (Unit No- 3 Cooling tower in BTPS).
The geometry of the Hyperboloid revolution
………….(1)
In which Ro horizontal radius at any vertical coordinate, Y origin of coordinates being defined by the center of the tower throat, ao
radius of the throat, and b is some characteristic dimension of the hyperboloid.
Figure1: Geometry of Existing Cooling Tower (BTPS)
CT 1 (143.50 m)
2014-15 ICRASE Copy right ©by ISRASE
ISRASE eXplore digital library
ISRASE First International Conference on Recent Advances in Science & Engineering -20014 (ISRASE-2014)
Figure 2: Geometry of Existing Cooling Tower (BTPS)
CT 2 (175.50 m)
Table 1: Geometric Details of Cooling Towers
Analysis is carried out for uniform shell thickness from 200mm,
250mm, 300mm, 350mm, 400mm, 450mm, and 500mm.
B Material Properties for Analysis of Cooling Tower (CT)
 Young‟s modulus: 31Gpa.
 Poisson‟s Ratio: 0.15.
 Density of RCC: 25 kN/m3
C Finite Element Modeling (FEA)
Due to the complexity of the material properties, the
boundary conditions and the tower structure, finite element
analysis is adopted. The finite element analysis of the cooling
towers has been carried out using ANSYS V.10. The analysis has
been carried out using 8-node shell element (SHELL 93). In the
present study, only shell portion of the cooling towers has been
modeled and fixity has been assumed at the base.
ANSYS is a commercial FEM package having capabilities
ranging from a simple, linear, static analysis to a complex, non
linear, transient dynamic analysis. It is available in modules; each
module is applicable to specific problem. Typical ANSYS
program includes 3 stages Pre processor, Solution & General Post
processor.
Sl
no
Description
Symbols
Cooling
tower 1
( CT 1)
Cooling
tower 2
( CT 2)
1
Total height
H
143.50m
175.50m
2
Height of
throat
Hthr
107.75m
131.60m
3
Diameter at
top
Dt
63.6m
82.00m
4
Db
110m
122.00m
Dthr
61.0m
68.750m
Hc
9.20m
9.275m
7
Diameter at
bottom
Diameter at
throat level
Column
Height
(Hthr/H) ratio
0.750
0.749
8
(Dthr/D) ratio
0.554
0.563
5
6
D Wind loads
The wind pressures on two existing cooling towers CT 1& CT 2 at a given height [Pz] are computed as per the
stipulations of IS:
875 (part 3)-1987. For computing the design wind pressure at a given height the basic wind speed (Vb) will be taken as Vb=39
m/s at 9.2m height above mean ground level. For computing design wind speed (Vz) at a height z, the risk coefficient k1=1.06
will be considered. For coefficient k2 terrain category 2 as per table 2 IS: 875 (part-3)-1987 will be considered. The wind
direction for design purpose will be the one which would induces worst load condition. Coefficient k3 will be 1 for the
towerunder consideration. The wind pressure at a given height will be computed theoretically in accordance with the IS codal
provision given as under
Pz=0.6 Vz2 N/m2
………………………. (2)
2014-15 ICRASE Copy right ©by ISRASE
ISRASE eXplore digital library
ISRASE First International Conference on Recent Advances in Science & Engineering -20014 (ISRASE-2014)
Where Vz =Vb x k1 x k2 x k3 ………………… (3)
Vb = basic wind speed which is specified for different zones of the country
k1 = probability factor ( risk coefficient) based on the statical concepts which take into account the degree of reliability required
and the time period of wind exposure i.e. the life of the structure.
k2 = the terrain height and structure size parameter that gives the multiplying factor by which the basic wind speed shall be
multiplied to obtain the wind speed at different sizes of buildings and structures.
k3 = the topography factor.
Computation of wind pressure (Pz) along the wind direction by Gust factor method.
E Wind Analysis
Wind analysis is carried out for two existing cooling towers i.e. CT 1 & CT 2. Geometry of the model is created in ANSYS .V.10
by using key points & input material models, shell element & make mesh to model in Pre processor. By assigning the loads &
boundary conditions and input the Pressures alongside to the model and solve the problem in solution & read the results in
General post processor.
Characteristics Models for wind analysis is shown in fig 7 to13.
Fig 3: Key points
Fig 4: Boundary condition
2014-15 ICRASE Copy right ©by ISRASE
ISRASE eXplore digital library
ISRASE First International Conference on Recent Advances in Science & Engineering -20014 (ISRASE-2014)
Fig 5: Element numbers in model
Fig 6: Nodes in model
Fig 7: Wind Pressure applied for Cooling Tower 1
2014-15 ICRASE Copy right ©by ISRASE
ISRASE eXplore digital library
ISRASE First International Conference on Recent Advances in Science & Engineering -20014 (ISRASE-2014)
Fig 8: Deflection for Cooling Tower 1 (200mm thickness)
Fig 9: Maximum Principal Stress for Cooling Tower 1
„Fig 10: Deflection at Top for Cooling Tower 1
2014-15 ICRASE Copy right ©by ISRASE
ISRASE eXplore digital library
ISRASE First International Conference on Recent Advances in Science & Engineering -20014 (ISRASE-2014)
Fig 11: Deflection for Cooling Tower 2 (200mm thickness)
Fig 12: Maximum Principal Stress for Cooling Tower 2
Fig 13: Deflection at Top for Cooling Tower 2
2014-15 ICRASE Copy right ©by ISRASE
ISRASE eXplore digital library
ISRASE First International Conference on Recent Advances in Science & Engineering -20014 (ISRASE-2014)
„
F Tabulation and Results
1) T= 0.09 H/√d = (0.09× 143.50)/√110= 1.23139
2) fo = natural frequency = 1/ T = 0.8120858 for CT 1
3) Damping Value (β) = 0.016 for CT 1
Table 2: Gust factor values for cooling tower 1
Sl
no
Height of tower
(H)
(m)
Gust factor values (G)
1
9.2
1.886
2
29.2
1.883
3
49.2
1.889
4
69.2
1.880
5
89.2
1.842
6
108.47
1.808
7
134.33
1.817
1) T= 0.09 H/√d = (0.09×175.50)/√122.009 = 1.429959
2) fo = natural frequency = 1/ T = 0.699320 for CT 2
As per clause 7, page no-48, IS 875 (part 3)-1987
3) Damping Value (β) = 0.016 for CT 2
As per table 34, page no-52 IS 875 (part 3)-1987
Table 3: Gust factor values for cooling tower 2
2014-15 ICRASE Copy right ©by ISRASE
Sl
no
Height of tower
(H)
(m)
Gust factor values (G)
1
9.2
1.8695
2
29.2
1.8681
3
49.2
1.8587
4
69.2
1.8356
5
89.2
1.7677
6
109.2
1.6897
7
129.2
1.7304
8
149.2
1.7216
9
166.22
1.6884
ISRASE eXplore digital library
ISRASE First International Conference on Recent Advances in Science & Engineering -20014 (ISRASE-2014)
Table 4: Results of wind analysis for cooling tower 1
Sl
no
Shell
Thickness
(mm)
Maximum
Deflection
(mm)
Maximum
Principal
stress (Mpa)
1
200
38.944
1.80
2
250
30.070
1.463
3
300
24.596
1.271
4
350
20.848
1.126
5
400
18.147
1.012
6
450
16.131
0.9195
7
500
14.586
0.8432
Table 5: Results of wind analysis for cooling tower 2
Sl
no
Shell
Thickness
(mm)
Maximum
Deflection
(mm)
Maximum
Principal
stress (Mpa)
1
200
80.996
2.469
2
250
62.965
2.111
3
300
51.284
1.861
4
350
43.188
1.671
5
400
37.304
1.518
6
450
32.875
1.391
7
500
29.452
1.282
Graph 1: Graphical Representation of Stress v/s thickness for CT 1 & CT 2 for wind analysis
2014-15 ICRASE Copy right ©by ISRASE
ISRASE eXplore digital library
ISRASE First International Conference on Recent Advances in Science & Engineering -20014 (ISRASE-2014)
Graph 2: Graphical Representation of Max deflection v/s thickness for CT 1 & CT 2 for wind analysis
G Buckling Analysis
Buckling Analysis is carried out for two existing cooling towers (CT 1 & CT 2) due to its self weight & varying thicknesses.
Eigen buckling analysis is a technique used to determine buckling loads (critical loads at which a structure becomes unstable) and
buckled mode shapes (the characteristic shape associated with a structure‟s buckled response).
1. Modes of Extraction- Block Lanczos method (Default)
2. Number of modes Extracted-50
Characteristics models for buckling analysis is shown in fig 14 to 17
Fig 14: Deflection for CT 1(Buckling mode 1) (200mm thickness)
Fig 15: Maximum Principal Stress (Mode 1) for Cooling Tower 1
2014-15 ICRASE Copy right ©by ISRASE
ISRASE eXplore digital library
ISRASE First International Conference on Recent Advances in Science & Engineering -20014 (ISRASE-2014)
Fig 16: Deflection for Cooling Tower 2 (Mode 1)
Table 6: Results of buckling analysis for CT 1
Shell
Thickness
(mm)
200
250
300
350
400
450
500
Modes
Frequency
(HZ)
1
3
5
10
1
3
5
10
1
3
5
10
1
3
5
10
1
3
5
10
1
3
5
10
1
3
5
10
11.154
11.589
11.645
12.522
15.062
15.177
15.272
17.573
19.013
19.046
19.806
21.879
23.131
23.44
24.961
27.142
27.508
28.32
29.343
33.338
32.172
33.495
33.616
40.068
37.118
37.896
39.279
47.358
2014-15 ICRASE Copy right ©by ISRASE
Max
Deflection
(mm)
1.027
1.036
0.88155
1.06
1.052
1.042
1.012
1.023
1.042
1.033
1.043
1.046
1.035
1.026
1.011
1.037
1.038
1.041
1.021
0.9725
1.031
1.011
1.054
1.005
1.01
1.038
1.053
0.880141
Max
Principal
stress
(Mpa)
0.141107
0.121869
0.152701
0.205809
0.152004
0.176925
0.111382
0.248456
0.179553
0.129468
0.211512
0.099406
0.15139
0.192193
0.227134
0.279461
0.172969
0.212251
0.12819
0.298305
0.192098
0.139156
0.238148
0.35523
0.207458
0.154163
0.263028
0.348003
Shell
Thickness
(mm)
200
250
300
350
400
450
500
Modes
Frequency
(HZ)
1
3
5
10
1
3
5
10
1
3
5
10
1
3
5
10
1
3
5
10
1
3
5
10
1
3
5
10
5.677
6.349
7.144
8.533
7.55
8.597
8.701
11.557
9.645
10.406
11.152
15.128
11.992
12.291
14.033
17.126
14.372
14.603
17.242
20.60
16.654
17.484
20.773
24.619
19.143
20.638
23.365
28.783
Max
Deflection
(mm)
1.006
1.036
1.038
1.126
1.024
1.036
1.036
1.110
1.035
1.036
1.038
1.112
1.058
1.035
1.04
0.98545
1.035
1.069
1.043
1.054
1.035
1.04
1.046
1.054
1.035
1.034
0.98648
1.038
Maximum
Principal
stress
(Mpa)
0.140914
0.170463
0.105226
0.222687
0.140338
0.166921
0.110335
0.252749
0.156524
0.126587
0.191002
0.300151
0.182783
0.143113
0.233165
0.133479
0.159731
0.209092
0.256321
0.351288
0.176373
0.227851
0.288999
0.39471
0.193007
0.249987
0.150298
0.432689
ISRASE eXplore digital library
ISRASE First International Conference on Recent Advances in Science & Engineering -20014 (ISRASE-2014)
Fig 17: Maximum Principal Stress (Mode 1) for Cooling Tower 2
Table 7: Results of buckling analysis for CT 2
Graph 3: Graphical Representation of Stress v/s thickness between CT 1 & CT 2 for buckling (Mode 1)
2014-15 ICRASE Copy right ©by ISRASE
ISRASE eXplore digital library
ISRASE First International Conference on Recent Advances in Science & Engineering -20014 (ISRASE-2014)
Graph 4: Graphical Representation of Stress v/s thickness between CT 1 & CT 2 for buckling (Mode 3)
Graph 5: Graphical Representation of Stress v/s thickness between CT 1 & CT 2 for buckling (Mode 5)
Graph 6: Graphical Representation of Stress v/s thickness between CT 1 & CT 2 for buckling (Mode 10)
2014-15 ICRASE Copy right ©by ISRASE
ISRASE eXplore digital library
ISRASE First International Conference on Recent Advances in Science & Engineering -20014 (ISRASE-2014)
Graph 7: Graphical Representation of Maximum Deflection v/s thickness between CT 1 & CT 2 for buckling (Mode 1)
Graph 8: Graphical Representation of Maximum Deflection v/s thickness between CT 1 & CT 2 for buckling (Mode 3)
Graph 9: Graphical Representation of Maximum Deflection v/s thickness between CT 1 & CT 2 for buckling (Mode 5)
2014-15 ICRASE Copy right ©by ISRASE
ISRASE eXplore digital library
ISRASE First International Conference on Recent Advances in Science & Engineering -20014 (ISRASE-2014)
ΙΙΙ ACKNOWLEDGMENT
The author thank the guide Prof A.V. Kulkarni, Associate professor & Academic coordinator, Department of Civil
Engineering, Principal and Management of BLDEA‟S V.P Dr. P.G. Halakatti College of Engineering & Technology,
Bijapur for the continued support and cooperation in carrying out this research study.
ΙV SUMMARY & CONCLUSIONS OUTCOMES OF
WIND
ANALYSIS
[A] From Graphical Representation of stress v/s thickness and Deflection v/s thickness, it is
evident that
1) As thickness increases, Deflection and Maximum Principal stress decreases for both CT 1 & CT 2. (Refer graph
no-1 & 2)
2) On Comparing two Existing cooling towers CT 1 & CT 2, the degree of distortion increases with height of tower,
hence the deflection is maximum in CT 2. (Refer graph no-2)
3) The Distortion is minimum at bottom part of shell due to fixed base (i.e. fixity), & maximum at top part
of shell. OUTCOMES OF BUCKLING ANALYSIS
1) In Buckling analysis, the buckling of CT 1 is maximum as compared to CT 2, CT 2 shows less buckling due to
its size, symmetric geometry of shell ( for increasing thickness).
2) In Buckling analysis, frequencies for CT 1 & CT 2 for various selected modes increases for increasing
thickness. [B] From Graphical Representation of stress v/s thickness, it is evident that
1) In
Mode
1
a) As thickness increases, Maximum Principal Stress for CT 1 & CT 2 increases, CT 1 shows maximum stress at
500mm compared to CT 2. (Refer graph no-3)
b) For thickness of 200mm, 310mm, 385mm, CT 1 & CT 2 shows similar value of principal
stress. c) For thickness of 350mm, CT 2 shows high value of stress compared to CT 1.
2) In Mode 3
a) As thickness increases, Maximum Principal Stress shows high value for CT 2 as compared to CT 1, and behaves
opposite to
Mode 1. (Refer graph no-4)
b) For thickness of 240mm, 300mm, 400mm, CT 1 & CT 2 shows same value of stress.
3) In Mode 5
a) As thickness increases, Maximum Principal Stress for CT 1 gives high value as compared to CT 2, & behaves
opposite to
Mode 3. (Refer graph no-5)
b) For thickness of 250mm, 350mm, 470mm, CT 1 & CT 2 shows similar value of stress.
4) In Mode 10
2014-15 ICRASE Copy right ©by ISRASE
ISRASE eXplore digital library
ISRASE First International Conference on Recent Advances in Science & Engineering -20014 (ISRASE-2014)
a) As thickness increases, Maximum principal stress shows high value for CT 2 as compared to CT 1 and behaves
opposite to
Mode 5. (Refer graph no-6)
b) CT 1 & CT 2 shows similar value of stress for 250mm, 310mm, and 390mm
thickness. [C] From Graphical Representation of Deflection v/s thickness, it is
evident that
2014-15 ICRASE Copy right ©by ISRASE
ISRASE eXplore digital library
ISRASE First International Conference on Recent Advances in Science & Engineering -20014 (ISRASE-2014)
1) In Mode 1
a) As thickness increases, deflection in CT 1 & CT 2 gradually increases and thereafter reduces. (Refer graph no-7)
b) CT 1 & CT 2 shows similar value of deflection for 310mm, 390mm, and 440mm thickness.
2) In Mode 3
a) For thickness of 400mm, deflection is maximum in CT1 & CT 2. (Refer graph no-8)
b) CT 1 & CT 2 shows similar value of deflection for 200mm, 290mm, and 490mm thickness.
3) In Mode 5
a) For thickness of 200mm, CT 1 shows less deflection as compared to CT 2, & thereafter increasing the
thickness at
500mm both behaves opposite. (Refer graph no-9)
b) CT 1 & CT 2 shows same value of deflection for 270mm, 310mm, and 440mm thickness.
4) In Mode 10
a) As thickness increases, CT 2 shows high deflection as compared to CT 1, & deflection overlap at 350mm, which
shows less deflection for CT 2. (Refer graph no-10)
b)
CT
1
&
CT
2
shows
similar
value
of
deflection
for
320m,
370mm
thickness.
[D] From graphical representation of stress v/s thickness and deflection v/s thickness curves, optimum thickness of
300mm thickness gives same value of stress & deflection for both CT 1 & CT 2 in all selected modes.
ΙV REFERENCES
[1] G. Murali, C. M. Vivek Vardhan and B. V. Prasanth Kumar Reddy “RESPONSE OF COOLING TOWERS TO WIND LOADS”, ARPN
Journal of Engineering and Applied Sciences, VOL. 7, NO. 1, JANUARY 2012 ISSN 1819 -6608.
[2] G Murali “ RESPONSE OF NATURAL DRAUGHT COOLING TOWERS TO WIND LOADS” International Journal of Emerging trends in
Engineering and Development, Issue 2, Vol 4 (May 2012), ISSN 2249-6149.
[3] Mungan and Wittek, 2004, Natural draught cooling towers. Taylor and Francis Group, London, UK.
[4] Orlando M. 2001, Wind-induced interference effects on two adjacent cooling towers, Engineering structures, 23: 979 -992.
[ 5] Prashanth N, Sayeed sulaiman, “To study the effect of seismic loads and wind load on hyperbolic cooling tower of varying dimensions and
RCC shell thickness” : International Journal of Emerging Trends in Engineering and Development Issue 3, Vol.4 (June -July 2013) ISSN 2249-6149.
[6] N Prabhakar (Technical Manager), “Structural aspects of hyperbolic cooling tower”, National seminar on Cooling tower, jan1990, Technical session
IV, paper no 9.
[7] IS: 11504:1985, Criteria for structural design of reinforced concrete natural draught cooling tower, New Delhi, India: Bureau of Indian standards.
[8] IS: 875 (Part3):1987, Code of practice for design loads (other than earthquake loads) for buildings and structures. New Delhi, India: Bureau of
Indian Standards.
2014-15 ICRASE Copy right ©by ISRASE
ISRASE eXplore digital library