IRJET-Critical Analysis of Wind on Vertical Tall Structures

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 05 Issue: 06 | June -2018 www.irjet.net p-ISSN: 2395-0072
© 2018, IRJET | Impact Factor value: 6.171 | ISO 9001:2008 Certified Journal | Page 42
Critical Analysis of Wind on vertical tall structures
Ashish Ghonmode1, Sanjay Bhadke2
1M.tech Scholar 2ndyear Tulsiramji Gaikwad-Patil college of Engineering and Technology, Nagpur,
Maharashtra
2Professor,Dept of Civil Engineering, Tulsiramji Gaikwad-Patil college and Technology, Nagpur, Maharashtra,
India
---------------------------------------------------------------------***--------------------------------------------------------------------
Abstract: Wind induced structural responses, including
pressure, are directional dependent. First win speed will not
be uniform in all directions. Second the shape and structural
properties of the structure will not be axi -symmetric.
Consideration of the directionality effect will help to achieve
an economical and safe design of structure The wind
pressure acting on individual units of a structure can be
determined using the pressure coefficient which depends on
the overall dimensions of the structure as well as the
openings present in the walls of the structure. Thenumerical
example considered in this chapter illustrates the
determination of static wind loads by both force coefficient
and pressure coefficient methods. Dynamic along-wind
analysis procedures using Random Vibration Analysis and
codal provisions explained in this paper. For the purpose of
along-wind analysis of the structures by the analytical
procedure based on random vibration analysis, a FORTAN
program was developed. For this purpose, t hree structures
have been considered, out of which, two are buildings and
one is a chimney. The output of the program is the response
of the structure in terms of mean response, peak factor,
standard deviation of fluctuating response along the height
of the structure.
1. Introduction
Characteristics of Wind and Wind Velocity
Wind is the motion of air relative to the earth’s surface
caused by rotation of the earth and differential solar
heating. The shear action of surface roughness retards the
wind velocity to nearly zero at the earth’s surface. The
wind velocity gradually increases with height andattains a
nearly constant value at a height known as the Gradient
Height (Nigam and Narayanan 1994). Layer. The variation
of wind velocity within the earth’s boundary layer with
respect to both height and the approach terrain .
1.2 Wind Response of Structures
Wind analysis methods are well established and available
in textbooks and in various international codes and
standards on wind load. IS 875 (Part 3) – 1987 is the
present Indian Standard available for wind load analysis
for buildings and structures. Draft code for wind load [IS
875 (Part 3) – Draft (2015)] was been published in 2015
with some modifications in the wind load assessment
procedures. Both the codes have provisions forStaticwind
load assessment on moderate height structures using a
simplified approach and Dynamic load calculation for tall
flexible structures using Gust Factor Method (GFM).
2. Literature Review
Solari (1983)has given direct formulae for calculation of
the along wind response in terms of gust factor,
displacement and acceleration for point-like structures,
vertical structures and horizontal structures. The ease of
use of these formulae as opposed to the procedures based
on use of graphs has been mentioned.
Badruddin et al. (1984) have performed experimental
and analytical study on along-wind and across-wind
response of three structures located in different terrains.
This includes a reinforced concrete tower (330m high), a
reinforced concrete chimney (180m high) and a latticed
steel tower (45.72m high). The results obtained
experimentally and analytically have been compared and
presented graphically.
Holmes (1987) proposed the need for a correction factor
to accommodate non-linearity in the modeshape of tall
buildings which are generally considered to have a linear
modeshape in the fundamental vibration mode. An
expression for modeshapecorrectionfactorhasbeengiven
by the author which is a function of the power law
exponent. The results have been compared to wind tunnel
measurement data and they werefoundtobeinagreement
with each other.
Gaikwad (2013) has described the procedure of random
vibration analysis and the Indian codal provision (IS 4998
(Part 1)) for determination of along-wind response of tall
RC chimneys. He has discussed aboutthevariousspectra of
longitudinal velocity fluctuations proposed in past
literatures and has presented the results of variation in
response of chimneys depending on the PSD being used in
calculation of wind response.
Kwok et al. (1988) carried out an experimental wind
tunnel study on a benchmark building (CAARCbuilding)to
study the effect of edge configuration of buildings on the
response of such tall buildings due to action of wind.
Different plan configurations such as plain rectangular
plan, rectangular plan with slotted corners and chamfered
corners were subjected to wind tunnel testing. The
response of the model in along-wind and across-wind

directions was studied and it was observed that these
modifications in the cross section of the building caused a
reduction in the wind-induced responseinbothdirections.
The effect of angle of incidence of wind on the model was
also studied and it was seen that the response is maximum
only when the wind direction is exactly normal to the face
of the structure.
3 .Methods of Along-Wind Analysis
Wind analysis methods are well established and are
available in textbooks and codes of practices. In the
following sections, the rigorous method of Random
Vibration Analysis (RVA),whichistheanalytical procedure
for determination of wind response of a structure
subjected to action of wind load has been discussed in
detail in the following section. Also, the procedure
described in the Indian Standard for wind response
estimation is studied.
3.1 Random Vibration Analysis
Wind load acting at any point on a structure can be
considered as a sum of a mean component and a
fluctuating component. The mean component is time-
independent but varies along the height of the structure,
whereas the fluctuating component varies with both time
and height. Thus wind velocity and wind load can both be
referred in mathematical terms as Random Processes,
whose instantaneous valuecannot bepredictedaccurately.
Thus, quantification of the randomly varying wind load is
done statistically in terms of Average (mean), Variance,
Standard Deviation, Peak value, Power Spectral Density,
etc. The response of a structure acted upon by wind load is
also obtained in terms of statistical quantities.
Then the drag force acting on the structure due to wind
velocity is given by,
3.1
Instantaneous wind velocity is given as the sum of mean
and fluctuating component as,
3.2
where, = Mean Component of wind velocity
(independent of time)
U’ = Fluctuating Component of wind velocity
(varies with time and height)
On substituting Eq (3.2) in Eq (3.1), and neglecting higher
order terms we get,
Where, Bz is the width of the structure at height z, is
the mean wind load and is the fluctuating wind
load component acting per unit height of the structure.
The response of a structure subjected to an along wind
load described by Eq (3.3) is determined by modeling the
structure as a multi degree freedom system acted upon by
a mean wind load component and a fluctuating wind load
component. Thus, the mean deflection of the structure in
the along wind direction due to the mean wind load
component is given by Eq (3.5)
(3.5)
where , and are the mode shape, natural frequency
and generalized mass of the structure in mode and N is the
total number of modes considered. If m(z) is the mass per
unit height of the structure, then the generalized mass of
the structure injthmode is given by,
(3.6)
The fluctuating components ofwindvelocityand windload
are stochastic quantities whose value is randomly varying
with respect to time. Thus their values cannotbepredicted
accurately. Hence they are defined in terms of Power
Spectral Density (PSD). The PSD of the fluctuating along
wind deflection [Sx(z,n)] is obtained in terms of PSD of the
fluctuating component of the wind velocity [Su (z)]as
shown below.
where njis the damping ratio in jthmode, Coh (y1, y2, z1, z2,
n)is the across-wind crosscorrelation coefficient and CDFis
the reduced drag coefficient. PSD of wind velocity hasbeen
discussed in detail in section 3.3.2.
If M1 and M2 are two points on a structure, having
coordinates (y1,z1) and (y2,z2)respectively, then the PSD of
longitudinal velocity fluctuations at those two points
arecorrelatedinthealong-windandacross-winddirections
by the along-wind and acrosswind
cross correlation coefficients which are denoted by N(n)
and Coh (y1, y2, z1, z2, n)respectively.
Along wind correlation is accounted by considering the
reduced drag coefficient (CDF)by the following relation.
(3.8)
If M1 and M2 are on the same face of the structure, then
N(n) = 1 as there is noseparation between the points in the

along wind direction. In other cases, N(n) iscalculated
using the following equations.
(3.10)
3.2 IS : 875 (Part 3) – Draft 2015
Dynamic wind load calculation by the draft code is also
done using the Gust Factormethod but with slight
modifications. The present wind code uses graphs for
calculationof various terms used in Gust factor estimation.
This leads to a high degree of manualerror. But the draft
code suggests direct and simple formulae for the same
which allowsease calculation using computer programs
and reduces possibility of errors.
The present code suggests a single value of gust factor for
the entire height range of thestructure which is not the
actual case. This problem has been sorted out in the draft
code,in which Gust factor increases with the height of the
structure.
The procedure for wind load calculation by IS:875(Part3)
– Draft 2015 has beenprovided in Table 3.1 along with the
procedure prescribed in the present code.
3.3.1.1 Power Law
If Uris the wind velocity at a reference height zr then the
mean wind velocity acting atany height z is given by the
following empirical expression.
(3.25)
where α is the power law coefficient, whose value vary
depending on the terraincategorybeingconsidered.Values
of α suggested by Davenport (1965) and ASCE 7-10 has
beenprovided in Table 3.2.
Shear velocity of a terrain which is useful in
determining the spectra of windvelocity and variance of
the spectral curve is given by equation (3.26) which
depends onthe terrain roughness factor (k). Davenport
(1965) has suggested values of k as 0.005,0.015 and 0.05
for open, towns and large cities respectively.
(3.26)
Where is the wind velocity at 10m height.
Table 3.1 : Values of α suggested in Codes and
Literatures
Terrain
Category
Power Law Exponent (α)
Davenp
ort
(1965)
Nigam
and
Narayana
n (1994)
ASCE 7-10
Coastal
Exposure
- 0.12 0.1
Open
exposure
0.16 0.16 (1/7)=0.143
Sub-
urban
terrain 0.28 0.28 (1/4.5)=0.222
Centers
of towns
Centers
of large
cities
0.4 0.4 (1/3)=0.333
Table 3.2 : Shear Velocity correction
Factors for various terrains
Terrain Category
Coastal Terrain 0.85
Open Terrain 1.00
Center of Towns 1.33
Center of large cities 1.45
3.3.1.2 Comparison of Wind Profiles
Mean Wind Velocity Profiles were calculated as per
Logarithmic law and Power law fordifferentterrainsbased
on the following data.
Terrain
Logarithmic law
Power
law
z0(m) α
Coastal
Terrain
0.002 2.26 0.12
Open
Terrain
0.07 2.66 0.16
Center of
Towns
0.8 3.54 0.28
Center of
large cities
2.00 3.86 0.4

Based on the data given above, wind velocities have been
calculated and it is observedthat both the laws give
comparable results. The results have been tabulated in
Table 3.5and graphical result basedonLogarithmic lawhas
been presented in Fig. 3.1.
Figure 3.1 : Mean wind velocity profiles for various
terrains using Logarithmic Law
4. Numerical Results Of Static WindLoadAnalysis
4.1 Introduction
The difference between pressure coefficient method and
force coefficient method is explained in section 4.2 by
considering a building as an example. The various types of
static wind loads acting on different sections of the
building have been calculated using this example.
4.2 Static Wind load estimation for a building
The Wind code IS 875 (Part 3) gives provision for
calculation of wind load on a building for:
a) The building as a whole,
b) Individual structural elements such as roofs and walls,
and
c) Individual cladding units including glazing and their
fixings.
The above wind loads have been calculated for an example
building given the Explanatory Handbook on IS 875 (Part
3):1987 by IIT Roorkee.
Consider an RCC building locatedinDelhi,withdimensions
10m x 50m x 18m. The building has 40 openings 1.5m x
1.5m. The structure consists of RC column-beam frame at
5mc/c horizontally and 3mc/c vertically, supporting the
walls.
Figure 4.1: 3-Dimensional view of the building
Table 4.2:Calculation of Net Pressure Coefficients on
the walls
External Pressure Coefficients (Cpe)
Wall Cpefor Wind
angle (θ) = 0°
Cpefor Wind
angle (θ) =
90°
A 0.7 -0.5
B - 0.48 (= - 0.4
- 0.08)
0.5
C -0.7 0.8
D 0.7 -0.1
Net Pressure Coefficients (Cpnet= Cpe-Cpi)
Wall Net Pressure Coefficients
(Cpnet)
A, B + 0.7 – (-0.2) = +0.9 (Pressure)
- 0.5 – (+ 0.2) = - 0.7 (Suction)
C, D + 0.8 – (-0.2) = + 1.0 (Pressure)
- 0.7 – (+ 0.2) = - 0.9 (Suction)
Table 4.3: Design wind pressure acting on individual
structural units
The individual structural elements like walls, and
individual cladding units like glazing and their fixings are
designed for the load obtained by multiplying the design
wind pressure from Table 4.3 at the required height above
the ground surface with their respective surface areas.
Random Vibration Analysis Results
The details of the structures provided in section 5.3 have
been given as input to the FORTRAN program. The output
of the program is the along-wind response of the
structures in terms of mean response, peak factor,
standard deviation of fluctuating response, gust factor,
peak response of the structure,Bendingmomentand Shear
force along the height of the structure. These results have
been provided in the following sub-sections for the
structures described in the previous section.

4.3 Structure 1: 200m Building
The PSD that was actually used Simiu and Scanlan (1986)
was the PSD proposed by Simiu (Eq (3.32)). The output of
the FORTRAN programusing PSDssuggestedbyDavenport
(1961), Harris (1968), Kaimal (1972) and Simiu (1974)
have been presented in Table 5.1 along with the numerical
results given in the literature from
Table 4.3: RVA results of 200m building
Note: Xmean is the mean response of the building in m,
Xpeakis the maximum response (mean +
fluctuating) of the building in m,
Kxis the peak factor,
is the standard deviation of response in m, and
Structure 3: 400m Chimney
The 400m chimney was modeled using SAP2000, FEM
model of the two-noded single-element beam was
developed by discretizing it into 36 elements, as
considered in the literature [Menon and Rao (1996)].
The natural frequency details given in the literature has
been tabulated in Table 5.3 with SAP2000 and the values
seem to be very much comparable.
For both the buildings first vibrational mode with linear
mode shape and constant lumped masses along the height
was considered. In case of chimney, SAP2000 was used
model it as a vertical cantilever beam and modal analysis
was performed on the FEM model. The requireddata ofthe
structures were given as input to the FORTRAN program
and the along wind response of the structures were
obtained. Responses from various PSDs discussed in
Chapter 3 were obtained and the results have been
compared with each other and also with results presented
in the literature from which they have been taken. Also the
results codal analyses have been presented.
6.1 Conclusion
In the present work, methods of along wind analysis of tall
and slender structures have been discussed in detail. This
includes the rigorous method of Random Vibration
Analysis (RVA) and methods available in Indian Standard
for wind load calculation [IS : 875 (Par1987 and IS : 875
(Part 3) – Draft 2015]. The RVA procedure considers the
modal properties and geometric properties of the
structure, and the wind characteristics in the terrain in
which the structure is located in order to give theresponse
of the structure in terms of mean and fluctuating
displacement, Gust factor, Shear force and Bending
Moment. Only wind load, Shear force andBending moment
results can be determined using codal procedures.
Two important wind velocity profiles and various Power
Spectral Density functions proposed in past literatures
useful
6.2 Scope of Future work
 Random vibration analysis for across wind response
analysis of structures can be done.
 Random vibration analysis for torsional response of
structures can be studied.
 Combined response (along-wind, across-wind and
torsional responses) of slender structures due to action of
wind can be studied.
 Wind response analysis ofotherstructuressuchasbridges,
cables, transmission towers, etc. can be performed.
 Ways to include higher modes of vibration for buildings
can be worked out.
Xmean
(m)
Kx
(m)
G
Xpeak
(m)
Simiu and
Scanlan
(1986)
0.184 3.63 0.074 2.46 0.452
Simiu's PSD 0.186 3.64 0.083 2.62 0.488
Kaimal's
PSD
0.186 3.66 0.077 2.51 0.467
Harris'sPSD 0.186 3.67 0.092 2.82 0.524
Davenport's
PSD
0.186 3.70 0.094 2.88 0.535
Mode
Natural Frequency
(Hz)
Damping
Ratio (%)
Menon and
Rao (1996)
Present
study
Mode
1
0.148 0.163 1.5
Mode
2
0.654 0.715 2.1
Mode
3
1.649 1.713 4.3

REFERENCES
1. Ahmad, M. B., Pande, P. K., & Krishna, P. (1984). “Self-
Supporting Towers Under Wind Loads.” Journal of
Structural Engineering, 110(2), 370-384.
2. American Society of Civil Engineers (2010). Minimum
Design Loads for Buildings and Other Structures (ASCE/SEI
7-10
3. Bhandari, N. M., Krishna, P., Kumar, K., & Gupta, A. “The
spectrum of horizontal gustiness near the ground in high
winds.”.
4. Davenport, A. G. (1995). “The response of slender
structures to wind.” In Wind ClimateinCities (pp.209-239).
Springer Netherlands.
5. Draft IS 875. (Part-3) (2015). Code of practice for design
loads (Other than earthquake) for buildings and structures,
Part 3 Wind loads (Third Revision), BIS, New Delhi, India.
6. Gaikwad, P. J. (2013). Analysis of Dynamic wind loading on
RC chimneys. M.Tech Dissertation, Dept. of Applied
Mechanics, Visvesvaraya National Institute of Technology,
Nagpur, Maharashtra, India.

IRJET-Critical Analysis of Wind on Vertical Tall Structures

More Related Content

What's hot (18)

Similar to IRJET-Critical Analysis of Wind on Vertical Tall Structures (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET-Critical Analysis of Wind on Vertical Tall Structures