International Journal of Current Engineering and Technology

ISSN 2277 - 4106

© 2013 INPRESSCO. All Rights Reserved.
Available at http://inpressco.com/category/ijcet

Research Article
Effect of Temperature on Wind Forces on Tall Building using Numerical
Approach - Computational Fluid Dynamics (CFD)
Jigar K. Sevalia*a, Atul K. Desaia and S. A. Vasanwalaa
a
Applied Mechanics Department, Sardar Vallabhbhai National Institute of Technology, Surat, Gujarat, India.

Accepted 05 July 2013, Available online 01 August 2013, Vol.3, No.3 (August 2013)

Abstract

In this paper an attempt has been made for comprehensive numerical study of wind effects on tall building considering
temperature effects. For that a tall structure having different geometric plan configuration of same plan area have been
considered. All the tall buildings with different plan configuration have been modeled in CFD Code namely Fluent /
Gambit and then comparative study has been executed. Distribution of the fluctuating surface pressure and the wind
forces acting on bluff-shaped bodies are of great practical interest in the field of structural design in wind engineering
because much civil and industrial structure can be assimilating to this shape. Computational wind engineering as a new
branch of computational fluid dynamics (CFD) has been developed recently to evaluate the interaction between wind and
buildings numerically. The techniques of Computational Fluid Dynamics (CFD), such as Standard k-ε Simulation, were
adopted in this study to predict wind loads on and wind flow around the building. For the study, commercial
Computational Fluid Dynamics (CFD) code Fluent has been used to generate computational domain around building
then effect of temperature on static pressure, drag force and wind force coefficient on tall building has been examined.

Keywords: Tall Building; Computational Fluid Dynamics (CFD); Numerical Simulation; Temperature; Wind; Wind
Force; Wake Region

1. Introduction been proved quite useful for development, evaluation,

validation and general performance assessment of CFD
1
Modeling the wind atmosphere around buildings is of methods.
great importance for the Wind Engineering, Civil Distribution of the fluctuating surface pressure and the
Construction sectors as well as Structural Engineering wind forces acting on bluff shaped bodies are of great
Sectors. The potential market for wind engineering studies practical interest in the field of structural design in wind
around buildings is very large. engineering (Swaddiwudhipong S et al,2002) especially
Computational Wind Engineering (CWE) as a branch when building is located in different atmospheric
of Computational Fluid Dynamics (CFD) has been condition i.e. under different temperature conditions. Any
developed rapidly over the last three decades to evaluate increase in building height increases the effect of wind
the interaction between wind and structures numerically, loading. Wind loads on tall structures cause concerns
offering an alternative technique for practical applications about the integrity of the structure envelope and safety of
(Shenghong Huang et al,2007). the whole structural system. Under the influence of the
CFD simulations can provide information on all flow dynamic wind loads, typical high-rise buildings vibrate in
parameters in the entire computational domain. Moreover, the along wind, across wind, and torsional directions.
a reliable numerical evaluation of the interaction between Modern high-rise buildings designed to satisfy static
fluids namely winds and buildings can be achieved with lateral drift requirements still might oscillate excessively
CFD modeling in a time- saving as well as economic during windstorms. The level of these oscillations may be
manner. Thus, CFD can offer more flexibility when significant enough to cause discomfort to the occupants.
exploring a variety of building designs and modifications An assessment of building motion is an essential
and their impact on the flow around them. CFD could also prerequisite for serviceability.
potentially supersede traditional wind tunnel studies as a
more cost-effective and powerful design tool for wind 2. Methodology
engineering studies. However, wind tunnel studies have
In order to study the effect of Wind on Tall Building with
respect to Wind Force Coefficient (Cf), Drag Force, Lift
*Corresponding author Jigar K. Sevalia is a Research Scholar, Dr. Atul Force, Across wind force etc., a tall building with different
K. Desai and Dr. S. A. Vasanwala are Professors.

819
Jigar K. Sevalia et al International Journal of Current Engineering and Technology, Vol.3, No.3 (August 2013)

geometric plan shapes having dimensions as shown in fig. The gradient height ZG was assumed to be 900m and the
1 has been considered. The plan area of all geometric plan mean wind velocity UG at the gradient height was 57.91
shapes of tall building is same. The height of the tall m/s.
building considered is 300 m. To execute study, Since the k-ε model was used, the values of k and ε
Computational Fluid Dynamics Code namely Fluent and were required to account for the turbulence in the
Gambit have been used. The different geometric plan approaching wind. The turbulence intensity was assumed
shapes of tall building considered are as follows, to be 12% at 20 m above ground. The root-mean-square
1. Circular Plan Shape of Building (Model – 1) value of the fluctuating wind at the longitudinal direction
2. Square Plan Shape of Building (Model – 2) (Cheng-Hu Hu et al,2005) was obtained by
3. Swastik Plan Shape of Building (Model – 3)
4. Hexagonal Plan Shape of Building with Sharp √̅̅̅̅̅
(u ) * (2)
Windward Edge (Model – 4)
5. Hexagonal Plan Shape of Building with Blunt Where I = 12 %
Windward Edge (Model – 5)
6. Octagonal Plan Shape of Building with Sharp The other components u2’ and u3’ were assumed as
Windward Edge (Model – 6) following as no relevant experimental data is available.
7. Octagonal Plan Shape of Building with Blunt
Windward Edge (Model – 7) ̅̅̅̅̅̅
( ) ̅̅̅̅̅̅
( ) ̅̅̅̅̅̅
( )

Therefore, turbulence kinetic energy (k) was calculated

from

(3)
This estimation can be crude though, it is one of the
methods often suggested when the k–e model is used.
Another suggestion to determine k is

k = (U X I )2 (4)

Which is based on the assumption that

̅̅̅̅̅̅
( ) ̅̅̅̅̅̅
( ) ̅̅̅̅̅̅
( )

The two Eqs. (03) and (04) are often used to calculate k if
lacking the experimental data. Eq. (03) is a more
conservative estimate, as it implies a higher level of
Fig. 1 Tall Building Geometry with Different Plan Shapes turbulence in the approaching flow. n this study, the k
with Same Plan Area was calculated from Eq. (03), since the conservative
estimation was preferred for the external flow simulations
2.1Governing Equations (Cheng-Hu Hu et al,2005).
The other important value required is the dissipation
The wind in the atmospheric boundary layer is turbulent rate ε; which can be obtained from the assumption that the
and the wind velocity profile and its scale of turbulence wind is neutrally stratified and homogeneous in the surface
are determined by the surface roughness of the ground layer, where the rate of energy production is
(Cheng-Hu Hu et al,2005). A common practice of approximately equal to its dissipation rate (Cheng-Hu Hu
studying the turbulent wind is to decompose it into a et al,2005), therefore
quantity of mean wind velocity and its fluctuating
components. Statistically, an instantaneous wind velocity
at one direction can be considered as consisting of a mean
(5)
value (U) with a fluctuating component (u’) in which its
mean value (u’) is zero. With this manipulation, the time-
Where, k = Von Ka´ma´n constant (=0.41)
averaged mean flow equations can be derived from the
u* = Friction Velocity
time-dependent Navier–Stokes equations.
The approaching wind was created from a power-law
The friction velocity can be calculated from,
model to approximate the mean velocity profile (Cheng-
Hu Hu et al,2005):

(6)
(1) Where, Cµ = Constant = 0.09

820
Jigar K. Sevalia et al International Journal of Current Engineering and Technology, Vol.3, No.3 (August 2013)

2.2 Boundary Conditions The computational grid is a key element in CFD since it
determines the level of resolution of a flow field. -D
The boundary conditions for the computational domain is Structured grids are created in the testing domains and 3-D
considered as follows, unstructured meshes are arranged in the vicinity of
 The ground at the bottom of the computational Building Model (Shenghong Huang et al,2007). The grids
domain was simulated with a smooth wall using log in vertical plane are closely spaced near ground and
law wall function. coarser mesh is modeled away from ground. The
 The free slip boundary conditions are applied to top Computational Grid Patterns for the Building Unit situated
and side surfaces of computing domain. The flux in computational domain is shown in fig. 4 and fig. 5,
normal to the boundary is considered zero.
 The no slip boundary conditions are applied to the
surfaces of Building Model.

2.3 Domain Size

There are no explicit rules dictating the size of a

computing domain. Many researchers determine their
domain size by a trial-and-error approach because the
domain size does influence the computed results (Cheng-
Hu Hu et al,2005). For this study, size of the
computational domain considered is 1875 m X 1275 m X
900 m in the longitudinal (X), lateral (Z), and vertical (Y)
directions, respectively. Fig. 4 Computational Grids in X-Z Plane (Horizontal
Plane)

Fig. 2 View of Computational Domain along with

Square Plan Shaped Tall Building in X-Z Plane Fig. 5 Computational Grids in X-Y Plane (Vertical Plane)
(Horizontal Plane)
2.5 Properties of Air

Different properties of air considered is as shown in table

1,

Table 1 Air Properties

Temperature
Properties
˚ C (273 k) 7˚ C ( k)

Density (ρ) - kg/m3 1.293 1.1777

Fig. 3 View of Computational Domain along with Specific Heat Capacity (Cp) –
Square Plan Shaped Tall Building in X-Y Plane (Vertical 1005 1005
J/kg.K
Plane)
Thermal Conductivity (W/m-K) 0.0243 0.0262
2.4 Computational Grid
Kinematic Viscosity (kg/m.s) 1.33E-05 1.58E-05

821
Jigar K. Sevalia et al International Journal of Current Engineering and Technology, Vol.3, No.3 (August 2013)

Table 2 Wind Force Coefficient for Different Temperature Condition

Force Coefficient %
Building Geometry
˚ C ( 7 k) 7˚ C ( k) Difference

Circular Plan Shape – Model 1 0.57667 0.57706 0.067

Square Plan Shape – Model 2 1.05021 1.05043 0.021

Swastik Plan Shape – Model 3 0.92334 0.92304 -0.032

Hexagonal Plan Shape with Sharp Windward Edge – Model 4 0.88962 0.89019 0.064

Hexagonal Plan Shape with Blunt Windward Edge – Model 5 0.71583 0.71596 0.018

Octagonal Plan Shape with Sharp Windward Edge – Model 6 0.70287 0.70317 0.043

Octagonal Plan Shape with Blunt Windward Edge – Model 7 0.73933 0.73964 0.042

Wind Force Coefficient (Cf) Vs Atmoshperic Temperature (˚C)

Model - 1
Wind Force Coefficient

Model - 2

Model - 3

Model - 4

Model - 5

Model - 6

Model - 7

Atmoshperic Temperature (˚C)

Graph 1 Wind Force Coefficient (Cf) Vs Temperature (˚C)

Table 3 Total Drag Force on Tall Building Geometry for Different Temperature Condition

Total Drag Force %

Building Geometry ˚ C (273 k)
7˚ C ( k) Difference
Circular Plan Shape – Model 1 20323.34 18524.17 8.85
Square Plan Shape – Model 2 32800.59 29882.87 8.90
Swastik Plan Shape – Model 3 38689.25 35228.83 8.94
Hexagonal Plan Shape with Sharp Windward Edge – Model 4 29855.79 27211.82 8.86
Hexagonal Plan Shape with Blunt Windward Edge – Model 5 27740.48 25272.43 8.90
Octagonal Plan Shape with Sharp Windward Edge – Model 6 26105.77 23788.76 8.88
Octagonal Plan Shape with Blunt Windward Edge – Model 7 25369.28 23117.60 8.88

Results Shape Geometry. The Plan Area of each Tall Building

Geometry is Constant i.e. 5625 square meter and Height of
The result consists of Wind Force Coefficient (Cf) and each Tall Building Unit is 300 meter.
Drag Force on Tall Building due to interaction between The variation of Wind Force Coefficient for each Tall
wind and building under different temperature conditions. Building Geometry for different temperature condition as
The two different temperature conditions have been obtained by using CFD Code Fluent is tabulated in table 2
considered i.e. ˚C and 7˚C for different Building Plan and also shown graphically in Graph 1.

822
Jigar K. Sevalia et al International Journal of Current Engineering and Technology, Vol.3, No.3 (August 2013)

Total Drag Force (N) Vs Building Plan Configuration 0 Degree Celcius Temperature

27 Degree Celcius Temperature

Total Drag Force (N)

Building Plan Configuration

Graph 2 Total Drag Force (N) Vs Building Plan Configuration

Table 4 Suggested Design Wind Pressure Equation for Different Atmospheric Temperature Condition

Suggested Equation for finding

Atmospheric Temperature (°C)
Design Wind Pressure on Structures (Pz)
-30 0.719 * Vz2
-20 0.695 * Vz2
-10 0.671 * Vz2
0 0.646 * Vz2
10 0.624 * Vz2
20 0.602 * Vz2
30 0.583 * Vz2
40 0.563 * Vz2
50 0.548 * Vz2
60 0.533 * Vz2

From table 2 and table 3, it can be seen that increase in ilding and properties of air are changing simultaneously
temperature from ° C to 7° C doesn’t make any more and according to definition of wind force coefficient, the
difference in wind pressure coefficient value. But it affects change in the value of wind force coefficient is very
the total drag force acting on the building. Hence, it is negligible.
necessary to develop an equation to calculate design wind tatic Pressure
pressure on surfaces of tall building considering properties Wind Force Coefficient
⁄ * ρ* v
of air in different atmospheric temperature conditions.
From the study, it can also be concluded that even though
Accordingly the suggested equation for design wind
the change in temperature is not affecting the value of
pressure (N/m2) for the given design wind velocity (m/sec)
wind force coefficient (Cf), the total drag force acting on
will be as follows in table 4,
the tall building is changing along with temperature as
shown in table 3. At low temperature, the total drag force
Conclusion
on tall building is high due to the reason of high density of
air at low temperature. At the same time, the total drag
The study of effect of temperature on Wind Force
force on tall building is less due the low density of air at
Coefficient has been executed. The two different
high temperature. Hence, the structural engineer has to
atmospheric temperature conditions have been considered
consider the change in properties of air with respect to
to study the effect of temperature.
temperature to calculate the wind forces on the surfaces of
It has been found that the change in the value of wind
building.
force coefficient (Cf) is very negligible due to change in
The equation for finding wind pressure on surfaces of
temperature. The reason is that due to change in
structure based on research work shall be as mentioned in
temperature both static pressures on the surfaces of tall bu-

823
Jigar K. Sevalia et al International Journal of Current Engineering and Technology, Vol.3, No.3 (August 2013)

table 4 which take care of properties of air at different Hoxeya R. P., Robertsona A. P., Basarab B. and Younisb B. A.
atmospheric temperature conditions. (1993), Geometric parameters that affect wind loads on low-
rise buildings: full-scale and CFD experiments, Journal of
Wind Engineering and Industrial Aerodynamics, Vol. 50,
References
pg.243-252.
Lien F.S., Yee E. and Cheng Y. (2004), Simulation of mean flow
Aishe Zhang, Cuilan Gao and Ling Zhang (2005), Numerical
and turbulence over a 2D building array using high-resolution
simulation of the wind field around different building
CFD and a distributed drag force approach, Journal of Wind
arrangements, Journal of Wind Engineering and Industrial
Engineering and Industrial Aerodynamics, Vol. 92, pg. 117-
Aerodynamics, Vol. 93, pg. 891–904.
158.
Alexandre Luis Braun and Armando Miguel Awruch (2009),
Swaddiwudhipong S. and Khan M.S.(2002), Dynamic response
Aerodynamic and aeroelastic analyses on the CAARC
of wind-excited building using CFD, Journal of Sound and
standard tall building model using numerical simulation,
Vibration, pg. 735-754.
Computers and Structures, Vol. 87, pg. 564–581.
Shenghong Huang, Q.S. Li and Shengli Xu(2007), Numerical
Appupillai Baskaran and Ahmed Kashef , Investigation of air
evaluation of wind effects on a tall steel building by CFD,
flow around buildings using computational fluid dynamics
Journal of Constructional Steel Research, Vol. 63, pg. 612–
techniques, Engineering Structures, Vol. 18, No. 11, pg. 861-
627.
875.
Tore Wiik and Ernst W. M. Hansen(1997), The assessment of
Baskaran T. and Stathopoulos T.(1989), Computational
wind loads on roof overhang of low-rise buildings, Journal
Evaluation of Wind Effects on Buildings, Building and
of Wind Engineering and Industrial Aerodynamics, Vol. 67
Environment, Vol. 24 , No. 4, pg. 325-333.
& 68 , pg. 687-696.
Cheng-Hu Hu and Fan Wang (2005), Using a CFD approach for
Yoshie R., Mochida A., Tominaga Y., Kataoka H., Harimoto K.,
the study of street-level winds in a built-up area, Building and
Nozu T. and Shirasawa T.(2007), Cooperative project for
Environment 40, pg. 617-631.
CFD prediction of pedestrian wind environment in the
Gloria Gomes M., Moret Rodrigues A. and Pedro Mendes
Architectural Institute of Japan, Journal of Wind Engineering
(2005), Experimental and numerical study of wind pressures
and Industrial Aerodynamics, Vol. 95, pg. 1551–1578.
on irregular-plan shapes, Journal of Wind Engineering and
Industrial Aerodynamics, Vol. 93, pg. 741–756.

824