The Seventh International Colloquium on Bluff Body Aerodynamics and Applications (BBAA7)

Shanghai, China; September 2-6, 2012

Along-wind aerodynamic damping of high-rise buildings

Cao Huilan a, Quan Yong a, Gu Ming a, Wu Di a
a
State Key Laboratory of Disaster Reduction in Civil Engineering, Bridge Building, Tongji
University, No. 1239 Siping Road, Shanghai, China

ABSTRACT: Along-wind aerodynamic damping ratios are identified from wind-induced re-
sponses of 46 aeroelastic models in simulated turbulent flow using the random decrement tech-
nique (RDT). Their validity is examined through comparison with previous research achieve-
ments and the results evaluated by quasi-steady theory. Based on the identified results,
characteristics of along-wind aerodynamic damping of isolated high-rise buildings are studied.
The effects of density ratio, generalized stiffness, structural damping ratio, aspect ratio, side ra-
tio, roughness exposure, aerodynamically modified cross-sections and tapering on aerodynamic
damping ratio of high-rise buildings are investigated. Results indicate that: aerodynamic damp-
ing ratio increases monotonically with reduced wind velocity; structural damping ratio, density
ratio, side ratio, reduced wind velocity and aspect ratio are very important parameters for along-
wind aerodynamic damping ratio, while no clear effect of generalized stiffness on aerodynamic
damping ratio is observed. Aerodynamic damping ratio increases as corner-cut ratio or taper ra-
tio increases; low corner-cut ratios significantly decrease aerodynamic damping; however, modi-
fications of building corners and tapering are not always effective at reducing the aerodynamic
damping of tall buildings. According to the database, an empirical aerodynamic damping func-
tion for high-rise buildings is proposed.

KEYWORDS: high-rise building; wind-induced vibration; aerodynamic damping; aeroelastic

model; wind tunnel test; along-wind.

1 INTRODUCTION
Since structural damping ratio of high-rise building is always small and predominant frequency
of aerodynamic force under strong wind is close to the natural frequency of structure, wind load
would be one of the control loads in design of high-rise building for significant wind-induced re-
sponse. In addition, coupling effect between incident turbulent flow and structure may generate
negative aerodynamic damping, which would decrease total damping, and furthermore cause
larger response. Generally, along-wind, across-wind and torsional structure motion all could in-
duce aerodynamic damping. And aerodynamic damping in along-wind attracts researchers’ at-
tention most early.
Aerodynamic damping in along-wind can generally be estimated with satisfactory accuracy
by adopting the quasi-steady theory. Based on it, Davenport (1979) [1] found a=0.0175 with
V/(f0B)=10, M/aB2=182 and CD~2. Holmes (1996, 2001) [2,3] estimated the aerodynamic damp-
ing in along-wind based on the assumption that the aerodynamic drag force is proportional to the
square of the difference between the mean oncoming wind speed and the speed of the moving
structure. Gabbai and Simiu (2010) [4] presented a methodology for estimating aerodynamic
damping in which the relative wind speed with respect to the moving structure is modeled, more
accurately, as the difference between the total oncoming wind speed and the speed of the moving
structure. In addition, the structural motion is assumed to be proportional to the relative oncom-
ing wind speed raised to powers larger than two. In theory, this methodology is more accurate
than quasi-steady method and the method proposed by Holmes, but more complex.

813
 There are also many researchers study along-wind aerodynamic damping of high-rise build-
ings through wind tunnel test. Marukawa (1996) [5] evaluated the aerodynamic damping of rec-
tangular tall buildings from wind tunnel tests of aeroelastic model. Random decrement technique
is used to analyze effects of side ratio, aspect ratio and structural damping ratio on along-wind
aerodynamic damping. Cooper (1997) [6] studied the impact of motion amplitude on along-wind
aerodynamic damping by forced vibration test. Besides, the aerodynamic damping obtained from
aeroelastic response measurements was compared with those predicted by quasi-steady theory.
The results indicated that aerodynamic damping ratios in this test are independent of motion am-
plitude. Quan (2002, 2004, 2005) [7,8,9] and Gu (2004) [10] proposed a formula for aerodynam-
ic damping as a function of reduced velocity, roughness exposure and structural damping using
RDT method. Zou (2003) [11] investigated the aerodynamic damping of an isolated rectangular
high-rise building with side ratio of 2:1 at three reduced wind velocities. But up to now, existing
researches only focus on one or a few influence factors and are far from systemization. Moreo-
ver, the research achievements are lack of comparison with each other and the theoretical results.
Generally, the use of aerodynamically modified cross-sections, such as slotted corners, cham-
fered corners and tapering, is an effective means to reduce the wind-induced response. However,
modifications of building corners might not be entirely effective, as adverse effects may also oc-
cur (Kareem et al., 1999; Kim and Kawai, 1999) [12,13]. Kim and You (2002) [14] investigated
the taper-ratio effect for reducing wind-induced excitations in along- and across-wind directions
based on the force-balance test using rigid models with taper ratios of 5%, 10%, and 15%, and
one basic model of a square cross-section without taper. They discovered that the tapering is
more effective in reducing wind-induced excitations in the suburban flow environment than in
the urban flow environment. They also showed that tapering reduces the across-wind responses
more than the along-wind responses. And, compared with a basic model without a taper, the ta-
pering is not always effective at reducing the wind-induced excitations of tall buildings. Howev-
er, the literatures mentioned above could not systematically indicate the variation of wind-
induced response of high buildings and aerodynamic damping. Huang (2003) [15] studied effects
of modified cross-sections (corner-cut ratio of 10%) on along- and across-wind aerodynamic
damping. Results showed that aerodynamic damping of high buildings with slotted corner and
chamfered corner are much less than that of square high buildings, and in the view of engineer-
ing application, the aerodynamic damping in along- and across-wind all could be neglected for
buildings with corner-cut ratio of 10%. However, up to the present, relevant researches all focus
on rectangular building (Davenport, 1979; Marukawa, 1996; Quan, 2002, 2004, 2005; Gu, 2004;
Zou, 2003 etc.). And whether theoretically or experimentally, quite limit studies are about the in-
fluence of corner-cut and tapering on aerodynamic damping. Therefore, further researches are
needed to consider about these factors.
Along-wind aerodynamic damping ratios are identified from wind-induced response of 46
aeroelastic models in simulated turbulent flow. The effects of density ratio, generalized stiffness,
structural damping ratio, aspect ratio, side ratio, roughness exposure and aerodynamically modi-
fied cross-sections and tapering on aerodynamic damping ratio of high-rise buildings are investi-
gated. According to the database, an empirical aerodynamic damping function for high-rise
buildings is proposed.

2 OUTLINE OF THE WIND TUNNEL TESTS

2.1 Simulation of wind characteristics
The test is carried out in TJ-1 Boundary Layer Wind Tunnel in Tongji University, whose work-
ing section is 1.8m in width, 1.8m in height and 18m in length, and the wind speed ranges from 3

814
The Seventh International Colloquium on Bluff Body Aerodynamics and Applications (BBAA7)
Shanghai, China; September 2-6, 2012

to 32 m/s. Four kinds of wind conditions, corresponding to exposure category A, B, C and D in

the Chinese code [16] are simulated in the wind tunnel at a length scale of 1/800. The exponents
of the mean wind profiles for the exposure category A, B, C and D are 0.12, 0.16, 0.22 and 0.30,
and the corresponding gradient heights are 300, 350, 400, 450m, respectively. The wind charac-
teristics are achieved by a combination of turbulence generating spires, a barrier at the entrance
of the wind tunnel, and roughness elements along the wind tunnel floor upstream of the model.
The longitudinal turbulence intensities at the height of building (480m) are separately 9.23%、
9.36%、9.52% and 9.78%, while the values at the height of roughness length are 18.52%、
22.11%、27.07% and 27.50%. As profile of turbulence integral scale is difficult to be accurately
simulated, the values at the 2/3 height of building for the exposure category A, B, C and D are
respectively 229m，231m，285m，274m. Figure 1 shows the simulated mean wind speed pro-
files and longitudinal component profiles of turbulence intensity, as wind speed profiles provid-
ed by Chinese code and turbulence intensity profile given by AIJ (2004) [17].
Tubulence Intensity Iu Tubulence Intensity Iu
0.0 0.1 0.2 0.3 0.4 0.0 0.1 0.2 0.3 0.4
1.2 Exposure A 1.2 Exposure B
1.0 9.23% 1.0 9.36%
0.8 0.8
Power Law Fit Power Law Fit
0.6 0.6
Z/ZH

Z/ZH

0.12 0.16
UZ=(Z/ZH) UZ=(Z/ZH)
0.4 0.4
0.2 0.2 22.11%
18.52%
0.0 0.0
0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00
Normalized mean speed Uz/UH Normalized Mean Speed Uz/UH
Turbulence Intensity Iu Turbulence Intensity Iu
0.0 0.1 0.2 0.3 0.4 0.0 0.1 0.2 0.3 0.4

1.2 1.2 Exposure D

Exposure C
1.0 1.0 9.78%
9.52%
0.8 0.8 Power Law Fit
Power Law Fit 0.30
0.22 0.6 UZ=(Z/ZH)
Z/ZH

0.6 UZ=(Z/ZH)
Z/ZH

0.4 0.4

0.2 0.2
27.07% 0.0 27.50%
0.0
0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00
Normalized Mean Speed Uz/UH Normalized Mean Speed Uz/UH

Figure 1 Profiles of wind velocity and longitudinal turbulence intensity

1 38 39 40 41 42 43 44 45 46

Figure 2 Model cross-sections (mm)

815
 2.2 Building models
Length and velocity scale are separately 1/800 and 1/8. Dimensions of standard model are
0.075m×0.075m ×0.6m, with frequency of 13Hz, density of 213Kg/m3 and generalized mass of
0.24Kg. Corresponding to real structure, these parameters represent building with size of
60m×60m ×480m, natural frequency of 13 Hz, and density of 213Kg/m3. Generalized stiffness is
selected to be 1600 Kg/s2 according to natural frequency. In addition, structural damping ratio of
standard model is taken as 1%, while density of air is 1.227Kg/m3.
Wind-induced responses of 46 high-rise buildings are measured from aeroelastic model wind
tunnel test. The structure characteristic parameters are shown as Table 1. As to the other cases,
the heights of models keep the same, and cross sections are changed according to aspect ratios
and side ratios. The maximum block ratio of model to test section of wind tunnel is 2.22％. Case
38 to 46, cross sections are changed in accordance with corner-cut and taper ratios, considering
one type of aerodynamically modification each case, as shown in Figure 2. The base is used to
model the elastic parameters of buildings, as structural damping ratios are simulated by width of
damping plates and their depth dipped into oil, while springs are used to achieve stiffness. In or-
der to avoid energy transmission in the two orthogonal directions, one degree of freedom in hori-
zontal direction is fixed. All of the models are built with base plates, hollow aluminum alloy as
the cores, foamed plastics, light wood plates of 1mm thickness as their “clothes” and balancing
weight. Two piezoelectric accelerometers with sampling frequency of 1000Hz are placed at the
two ends inside windward surface, as sampling time set as 7 minutes.
Table 1 Wind tunnel test cases for high-rise buildings
Model Rough- Side Aspect Structural Structure Generalized Corner-cut
Case Height ness Ex- Ratio Ratio Damping Density Stiffness or Taper
H(mm) posure B/D H/(BD)0.5 s (%) s (Kg/m3) K (Kg/s2) Ratio
1-4 600 A,B,C,D 1 8 1 213 1600 0
1/3,3,
5-12 600 B,D 8 1 213 1600 0
1/2,2
13-18 600 B,D 1 5,10,12 1 213 1600 0
0.5,0.9,1.1,1.7
19-25 600 B,D 1 8 213 1600 0
5,2.3,2.3,2.8
26-31 600 B,D 1 8 1 360,255,160 1600 0
32-37 600 B,D 1 8 1 213 947,2130,3027 0
38-40 600 C 1 8 1 213 1592,1568,1472 5%-20%
41-43 600 C 1 8 1 213 1584,1536,1344 5%-20%
44-46 600 C 1 8 1 213 1600 1%-5%

3 TEST RESULTS

The generalized formula with four variables proposed by Tamura et al (1996) [18] is used to es-
timate damping from random acceleration responses.
a ( )  Ae   cos 1   2 0  B sin 1   2 0 
  (1)
Firstly, amplitude dependent structural damping ratio is derived from free acceleration response.
Then, RMS of wind-induced acceleration response after band-pass filtering is taken as the initial
amplitude of RDT to calculate total damping ratio. Afterwards, structural damping ratio corre-
sponding to this amplitude is derived through interpolation. The aerodynamic damping ratio a is

816
The Seventh International Colloquium on Bluff Body Aerodynamics and Applications (BBAA7)
Shanghai, China; September 2-6, 2012

calculated by subtracting the value of structural damping ratio s from the value of total damping
ratio, i.e. a =s. Following research mainly focuses on the study of aerodynamic damping
variation with density ratio, generalized stiffness, structural damping, aspect ratio, side ratio,
roughness exposure, aerodynamically modified cross-sections and tapering.
3.1 Verification of result
Marukawa (1996) studied the effects of side ratio in the range of 0.33-3, aspect ratio from 4 to 6,
and structural damping ratio of 0.5% to 2% on along-wind aerodynamic damping, and compared
the results with that evaluated by quasi-steady theory. Here, the drag coefficient was derived
from the mean displacement assuming that the wind force is distributed in proportion to the ve-
locity pressure. Quan (2002, 2004) discussed impacts of exposure category (A、B、C、D) and
structural damping on along-wind aerodynamic damping, and made comparison with Marukawa
(1996). The validity of research results in this paper is examined through comparison with previ-
ous research achievements and the results evaluated by quasi-steady theory as shown in Figure 3,
corresponding structure characters shown in Table 2.
As shown in Table 2 and Figure 3, there are certain differences between research results
with different roughness exposures and model parameters. But in general, the results show a
good agreement with previous research achievements and that derived by quasi-steady theory,
especially with the result evaluated by quasi-steady theory.
Table 2 Structure parameters for tests
Aspect Roughness Structural Structure
Side Ratio Model Height Length
Ratio Exposure Damping Density
B/D H (mm) Scale
H/(BD)0.5 (α、Ih) s s (Kg/m3)
Standard Model 8 1 A、B、C、D 600 1/800 0.88% 213
Marukawa α=0.167,
6 1 480 1/500 1% 200
(1996) Ih=10.7%
Quan (2002) 6 1 A 600 1/500 0.6% 180
1  a B 2 VH 0.5
a  C D ，as reduced velocity defined as VH/f0(BD) ，derives
4 M f 0 B
Quasi-steady
1  a B VH
Theory  a  C D (According to Marukawa (1996), for B / D =1, CD=1.05; as to the
4  s D f 0 BD
Chinese code, for B / D =1, CD=1.3)

0.030
Quan(2002)
0.025 Marukawa et al (1996)
quasi-steady theory (Cd=1.05)
0.020
quasi-steady theory (Cd=1.3)
0.015 Exposure A
Exposure B


0.010 Exposure C
0.005 Exposure D

0.000
0 5 10 15
0.5
V/f0(BD)

Figure 3 Comparison with research achievements and theoretical result

817
3.2 Characteristics of aerodynamic damping

3.2.1 Effect of roughness exposure

Figure 3 indicates the variation of aerodynamic damping ratio with reduced velocity VH/f0(BD)0.5
in simulated turbulence wind environments A、B、C、D. The aerodynamic damping ratio in
the along-wind in Figure 3 represents positive damping at reduced velocities of 4 or more, and it
increases monotonically with reduced wind velocity. At low reduced wind velocities, aerody-
namic damping ratios in exposure D are bigger than that in exposure category A. Furthermore, at
reduced wind velocities of more than 10, aerodynamic damping ratios are apparently bigger than
other cases. That is, aerodynamic damping ratio gradually increases slower with reduced velocity
for exposure category A、B、C、D. But overall, wind environment has no clear effect on
along-wind aerodynamic damping, as discrepancies are within 0.003. It may be caused by the
similar profiles of wind velocity (Chinese code) and turbulence intensity (AIJ2004) for this
height of building. So, the influence of roughness exposure still needs further research.

3.2.2 Effect of side ratio B/D

Figure 4 shows the variation of aerodynamic damping ratio with reduced velocity, as aspect ratio
H/(BD)0.5=8, structural damping ratio s =1%, side ratio B/D ranges from 0.33 to 3. The corre-
sponding values obtained by quasi-steady theory are also provided in this Figure. Here, the drag
coefficients adopt the values given in Marukawa (1996). Figure 4 indicates that side ratio is an
important parameter for aerodynamic damping, as aerodynamic damping increases with side ra-
tio, exclude the case with B/D=0.33. In most cases, aerodynamic damping increases monoton-
ically with reduced wind velocity, except that it starts to decrease at reduced velocity of 8-10 for
the case with B/D=0.33. For side ratios B/D≥1, aerodynamic damping increases with side ratio;
for side ratios B/D≤1, there is no clear variation in aerodynamic damping ratio with side ratio. As
to side ratios B/D>1, aerodynamic damping ratios derived from experiment are bigger than that
evaluated by quasi-steady theory in exposure category B; and opposite phenomena can be ob-
served in exposure category D. As to side ratios B/D=1, 0.5, aerodynamic damping ratios derived
from experiment are a little bigger than that evaluated by quasi-steady theory, as discrepancies
are within 0.002 in exposure category B and much less in exposure category D. Generally,
there’s no big difference between each other, so along-wind aerodynamic damping can be well
predicted by quasi-steady theory for these cases. But as to the case of B/D=0.33, experimental
values are much bigger than theoretical values. All in all, there is a much better agreement be-
tween aerodynamic damping evaluated from wind tunnel test and the theoretical value than that
obtained by Marukawa (1996). This is because, reduced velocity adopted by Marukawa (1996) is
VH/f0(BD)0.5, while the one used for quasi-steady theory is VH/f0B.

0.030 quasi-steady theory(From the 0.030

quasi-steady theory (From the
0.025 bottom up Cd=0.70,0.75,1.05,1.02,0.97) 0.025 bottom up Cd=0.70,0.75,1.05,1.02,0.97)
B/D=0.33 B/D=0.33
0.020 Exposure B 0.020 Exposure D
B/D=0.5 B/D=0.5
0.015 B/D=1 0.015 B/D=1
B/D=2 B/D=2




0.010 B/D=3 0.010

B/D=3
0.005 0.005
0.000 0.000
0 5 10 15 0 5 10 15
0.5 0.5
VH/f(BD) VH/f(BD)

Figure 4 Effect of side ratio B/D (H/(BD)0.5=8, s =1% )

818
The Seventh International Colloquium on Bluff Body Aerodynamics and Applications (BBAA7)
Shanghai, China; September 2-6, 2012

0.030 0.030
quasi-steady theory (Cd=1.05) quasi-steady theory (Cd=1.05)
0.025 quasi-steady theory (Cd=1.3) 0.025 quasi-steady theory (Cd=1.3)
0.5 0.5
0.020 H/(BD) =5 0.020 H/(BD) =5
Exposure B Exposure D
0.5 0.5
0.015 H/(BD) =8 0.015 H/(BD) =8
0.5 0.5
H/(BD) =10 H/(BD) =10



0.010 0.5 0.010 0.5
H/(BD) =12 H/(BD) =12
0.005 0.005
0.000 0.000
0 5 10 15 20 0 5 10 15 20 25
0.5 0.5
VH/f(BD) VH/f(BD)

Figure 5 Effect of aspect ratio H/(BD)0.5 (B/D=1, s=1%)

0.030 0.030
quasi-steady theory (Cd=1.05) quasi-steady theory (Cd=1.05)
0.025 quasi-steady theory (Cd=1.3) 0.025 quasi-steady theory (Cd=1.3)
0.020 s=0.0050 0.020 s=0.0091
Exposure B Exposure D
s=0.0088 s=0.0121
0.015 0.015
s=0.0112 s=0.0227




0.010 s=0.0175 0.010 s=0.0279

0.005 s=0.0229
0.005
0.000 0.000
0 5 10 15 0 5 10 15
0.5 0.5
VH/f(BD) VH/f(BD)

Figure 6 Effect of structural damping s (B/D=1, H/(BD)0.5=8)

3.2.3 Effect of aspect ratio H/(BD)0.5

Figure 5 shows the variation of aerodynamic damping ratio with reduced velocity, as side ratio
B/D=1, structural damping ratio s =1%, aspect ratio H/(BD)0.5 ranges from 5 to 12. Figure 5 in-
dicates that aerodynamic damping ratio gradually increases slower with wind velocity as aspect
ratio increases. At reduced wind velocitiy lower than 10, aerodynamic damping ratio increases
as aspect ratio increases; while opposite trend is observed at reduced wind velocity higher than
10. Above all, the effect of aspect ratio is not so obvious, with difference not more than 0.003,
and they can be relatively accurate evaluated by quasi-steady theory.

3.2.4 Effect of structural damping s

Figure 6 shows the variation of aerodynamic damping ratio with reduced velocity, as side ratio
B/D=1, aspect ratio H/(BD)0.5=8, structural damping ratios s ranges from 0.005 to 0.0279. At
the same reduced velocity, difference of aerodynamic damping is about 0.005, which states une-
lectable effect of structural damping on aerodynamic damping. At low reduced wind velocity,
aerodynamic damping generally decreases with structural damping; while the smaller the struc-
tural damping, the bigger the aerodynamic damping at high reduced wind velocity. The smallest
aerodynamic damping ratio is observed when the structural damping ratio is 0.023. In addition,
variations of aerodynamic damping for s=0.0091，0.0121，0.0227 in exposure category D are
similar with cases of s=0.0088，0.0112，0.0229 in exposure category B.

819
3.2.5 Effect of density ratio s/a
Figure 7 shows the variation of aerodynamic damping ratio with reduced velocity when density
ratio varies from 294 to 131. Density ratio has a clear effect on aerodynamic damping, with dif-
ference between 0.003 and 0.005 at the same reduced wind velocity. Aerodynamic damping ra-
tio increases monotonically with density ratio. Moreover, the value derived in exposure category
D is a little smaller than that in exposure category B.

3.2.6 Effect of generalized stiffness K

Figure 8 shows the variation of aerodynamic damping ratio with reduced velocity when general-
ized stiffness varies from 947 to 3027. As shown in Figure 8, no clear effect of generalized stiff-
ness on aerodynamic damping is observed in the along-wind direction, with difference not more
than 0.002. Likewise, aerodynamic damping ratio in exposure category D gradually increases
slower with reduced velocity than in exposure category B. The values estimated in exposure cat-
egory D are much closer to the values derived by quasi-steady theory, while much more discrep-
ancies are observed in exposure category B.

3.2.7 Effect of slot rate

Figure 9 shows the variation of aerodynamic damping ratio with reduced velocity, as aspect ra-
tios H/(BD)0.5=8, structural damping ratio s=1%, side ratio B/D=1, slot rate ranges from 0% to
20%. The corresponding values obtained by quasi-steady theory are also provided in this Figure.
Here, the drag coefficients adopt the value given in Marukawa (1996). Figure 9 indicates 5%
and 10% slot rates significantly decrease aerodynamic damping, and this trend is more obvious
at
0.030 0.030
quasi-steady theory (Cd=1.05) quasi-steady theory (Cd=1.05)
0.025 0.025
s/a=294 s/a=294
0.020 s/a=208 Exposure B 0.020 s/a=208 Exposure D
0.015 s/a=174 0.015 s/a=174




0.010 s/a=131 0.010 s/a=131

0.005 0.005
0.000 0.000
0 5 10 15 20 0 5 10 15 20
0.5 0.5
VH/f(BD) VH/f(BD)

Figure 7 Effect of density ratio s/a

0.030 0.030
quasi-steady theory (Cd=1.05) quasi-steady theory (Cd=1.05)
0.025 0.025
quasi-steady theory (Cd=1.3) quasi-steady theory (Cd=1.3)
0.020 K=947 0.020 K=947
Exposure B Exposure D
0.015 K=1600 0.015 K=1600
K=2130 K=2130




0.010 K=3027 0.010 K=3027

0.005 0.005
0.000 0.000
0 5 10 15 20 0 5 10 15 20
0.5 0.5
VH/f(BD) VH/f(BD)

Figure 8 Effect of generalized stiffness K

820
The Seventh International Colloquium on Bluff Body Aerodynamics and Applications (BBAA7)
Shanghai, China; September 2-6, 2012

0.030 0.030
quasi-steady theory︵ Cd=1.05︶ quasi-steady theory (Cd=1.05)
0.025 0.025 quasi-steady theory (Cd=1.3)
quasi-steady theory︵ Cd=1.3
︶

0.020 standdard model 0.020 standard model

chamfer ratio 5% Exposure C
slot ratio 5% 0.015
0.015 Exposure C chamfer ratio 10%
slot ratio 10%


a

0.010 slot ratio 20% 0.010 chamfer ratio 20%

0.005 0.005

0.000 0.000
0 5 10 15 0 5 10 15
0.5 0.5
U/f0(BD) U/f0(BD)

Figure 9 Effect of slot rate Figure 10 Effect of chamfer rate

higher reduced wind velocity. For slot rate of 20%, although it decreases in certain extend, but
difference is not significant, i.e. larger slot does not lead to more decrease. Also in this case, de-
crease effect of aerodynamic damping strengthens as reduced wind velocity increases. In gen-
eral, slot rates in certain range significantly decrease aerodynamic damping, especially for the
case with slot rate of 5%.

3.2.8 Effect of chamfer rate

Figure 10 shows the variation of aerodynamic damping ratio with reduced velocity, as aspect ra-
tios H/(BD)0.5=8, structural damping ratio s=1%, side ratio B/D=1, chamfer rate ranges from
0% to 20%. The corresponding values obtained by quasi-steady theory are also provided in this
Figure. As shown in Figure 10, 5% and 10% chamfer rates significantly decrease aerodynamic
damping, and this trend is more obvious at higher reduced wind velocity. For chamfer rate of
20%, although it decreases in certain extend, but difference is not significant, which means larg-
er chamfer is not always effective at reducing the aerodynamic damping of tall buildings. When
chamfer rate varies from 5% to 20%, aerodynamic damping ratio increases as chamfer rate in-
creases. As compared with cases with slotted corner, aerodynamic damping ratios of building
with 5% chamfer rate are smaller than that with 5% slot rate, the opposite phenomena can be ob-
served for chamfer rate 10% and 20%. Overall, the effects of slotted corner and chamfered cor-
ner on aerodynamic damping ratio are similar; for low corner-cut ratios, aerodynamic damping
ratios are bigger with slotted corner, while for high corner-cut ratios, they are bigger with cham-
fered corner.

3.2.9 Effect of tapering

Figure 11 shows the variation of aerodynamic damping ratio with reduced velocity, as aspect ra-
tios H/(BD)0.5=8, structural damping ratio s=1%, side ratio B/D=1, taper rate ranges from 0% to
20%. The corresponding values obtained by quasi-steady theory are also provided in this Figure.
As shown in Figure 11, taper rate of 1% reduces aerodynamic damping, but taper rate of 3% and
5% increase aerodynamic damping. When taper rate varies from 1% to 5%, aerodynamic damp-
ing ratio increases as taper rate increases. Therefore, tapering is not always effective at reducing
the aerodynamic damping of tall buildings, as lower rates of tapering reduce aerodynamic damp-
ing and it will increase when taper rate exceeds 3%.

821
 0.030
quasi-steady theory (Cd=1.05)
0.025 quasi-steady theory (Cd=1.3)
0.020 standard model
taper rate 1% Exposure C
0.015 taper rate 3%
a

0.010 taper rate 5%

0.005
0.000
0 5 10 15
0.5
U/f0(BD)

Figure 11 Effect of tapering

0.025 0.025
fitted values fitted values
0.020 cases with square section 0.020 standard model
bevel rate 5%, 10%, 20%
0.015 chamfer rate 5%, 10%, 20%
0.015
Fitted value

taper rate 1%, 3%, 5%

Fitted value

0.010
0.010
0.005
0.005
0.000

-0.005 0.000
0.00 0.01 0.02 0.000 0.005 0.010
Test value Test value

Figure 12 Comparison between fitted values and tested values for aerodynamic damping

3.3 Formula fitting of aerodynamic damping ratio

As described above, the aerodynamic damping ratios of high-rise building increase monotonical-
ly with reduced wind velocity. In most cases, they can be relatively accurate evaluated by quasi-
steady theory. Research shows that structural damping s, density ratio s/a, side ratio B/D, re-
duced velocity VH/f0(BD)0.5, aspect ratio H/(BD)0.5, aerodynamically modified cross-sections and
tapering are important effects for aerodynamic damping ratio in along-wind, while the influence
of generalized stiffness and roughness exposure are not so significant. As quasi-steady theory
only takes density ratio, side ratio, reduced velocity and drag coefficient into consideration, an
empirical aerodynamic damping function for high-rise buildings, written as formula (2), is pro-
posed according to the database after many times comparison. The conformity between aerody-
namic damping ratios derived from wind tunnel test and by fitting is shown in Figure 12. Stand-
ard error for this for this empirical formula is
N
1
 a 
N
 
i 1
a  calc (i )   a test (i )   0.0018
2

Where, N is the number of case (415 for here); a-calc and a-test are separately the fitted values
and tested values for aerodynamic damping.

822
The Seventh International Colloquium on Bluff Body Aerodynamics and Applications (BBAA7)
Shanghai, China; September 2-6, 2012

0.5
1 a B U H
 a  (  s0.15   C D  0.002)
4  s D f 0 BD B (2)
Here, drag coefficient CD adopts 1.05; λ is modification coefficient of cross section, as λ=1.0 for
square building, with λ for other cases and corresponding standard error shown in Table 3.
Table 3 Fitting parameter of aerodynamic damping for corner-cut and tapering high-rise buildings
Slot rate Slot rate Slot rate Chamfer rate Chamfer rate Chamfer rate Taper rate Taper rate Taper rate
5% 10% 20% 5% 10% 20% 1% 3% 5%
λ 0.38 0.44 0.75 0.30 0.53 1.10 0.71 1.17 1.26
ai 0.8×10-3 1.2×10-3 1.1×10-3 1.2×10-3 1.2×10-3 1.7×10-3 0.9×10-3 1.5×10-3 1.0×10-3

4 CONCLUSION

46 aeroelastic model tests are conducted to study the effects of important factors on along-wind
aerodynamic damping ratios using RDT method with four parameters. The influences of density
ratio, generalized stiffness, structural damping ratio, aspect ratio, side ratio, roughness exposure,
aerodynamically modified cross-sections and tapering on aerodynamic damping are investigat-
ed. All parameters of high-rise building, which could affect aerodynamic damping, are systemat-
ically taken into consideration. The following conclusion can be drawn:
1 Aerodynamic damping ratio gradually increases slower with reduced velocity in exposure
category A、B、C、D. But overall, wind environment has no clear effect on along-wind
aerodynamic damping, which may be caused by the similar profiles of wind velocity (Chinese
code) and turbulence intensity (AIJ2004) for this height of building. So, the influence of
roughness exposure still needs further research.
2 For side ratios B/D≥1, aerodynamic damping ratio increases with side ratio; for side ratios
B/D≤1, there is no clear variation in aerodynamic damping ratio with side ratio. In addition,
along-wind aerodynamic damping can be well predicted by quasi-steady theory.
3 Structural damping ratio and aspect ratio has great influence on aerodynamic damping. Aero-
dynamic damping ratio gradually increases slower with wind velocity as aspect ratio increas-
es. At reduced wind velocity lower than 10, aerodynamic damping ratio increases as aspect
ratio increases; while at reduced wind velocity higher than 10, opposite trend is observed.
4 Although density ratio and generalized stiffness all reflect in change of structural damping,
the impact of density ratio is relatively significant while no clear effect of generalized stiff-
ness can be observed.
5 The effects of slotted corner and chamfered corner on aerodynamic damping ratio are similar;
aerodynamic damping ratio increases with corner-cut ratio, as corner-cut ratio of 5% and 10%
significantly decrease aerodynamic damping, while there’s little difference between square
building and cases with corner-cut ratio of 20%.
6 Aerodynamic damping ratio increases as taper rate increases; taper rate of 1% reduces aero-
dynamic damping, but taper rate of 3% and 5% increase aerodynamic damping.
7 Modifications of building corners and tapering are not always effective at reducing the aero-
dynamic damping of tall buildings.
Structural damping s, density ratio s/a, side ratio B/D, reduced velocity VH/f0(BD)0.5, as-
pect ratio H/(BD)0.5, aerodynamically modified cross-sections and tapering are important effects
for aerodynamic damping ratio in along-wind, while the influence of generalized stiffness is not

823
so significant. Through fitting of estimated aerodynamic damping ratios, an empirical aerody-
namic damping function for high-rise buildings is proposed with taking effects of structural
damping, aspect ratio and modification of cross section into quasi-steady theory.

5 ACKNOWLEDGEMENTS

The author would like to gratefully acknowledge the support of the National Natural Science
Foundation of China (50878159, 90715040) and Special Research Fund for Doctoral Program of
Senior School by the Minstry of Education of China (200802471005).

6 REFERENCES

[1] A.G. Davenport, The influence of Turbulence on the Aeroelastic Responses of Tall Structures to Wind, in IAHR-
IUTAM Symp. Pract. Exp. with Flow Ind. Vib, Germany, Karlsruhe, University of Karlsruhe, 1979, pp.681-695.
[2] J.D. Holmes, Along-wind response of lattice towers—II: Aerodynamic damping and deflections, Eng. Struct., 18
(1996): 483-488.
[3] J.D. Holmes, Wind loading of structures, E & FN Spon, London, 2001.
[4] R.D. Gabbai and E. Simiu, Aerodynamic damping in the along-wind response of tall buildings, J. Struct. Eng.,
136 (2010): 117-119.
[5] H. Marukawa, N. Kato, K. Fujii and Y. Tamura, Experimental evaluation of aerodynamic damping of tall build-
ings, J. Wind Eng. and Ind. Aerodyn., 59 (1996): 177-190.
[6] K.R. Cooper, M. Nakayama, Y. Sasaki, A.A. Fediw, S. Resende-Ide and S.J. Zan, Unsteady aerodynamic force
measurements on a super-tall building with a tapered cross section, J. Wind Eng. and Ind. Aerodyn., 72 (1997):
199-212.
[7] 全涌, 超高层建筑横风向风荷载及响应研究, 上海, 同济大学, 2002.
Y. Quan. Across-wind loads and responses on super high-rise buildings, Shanghai, Tonji University, 2002.
[8] 全涌, 顾明, 方形断面高层建筑的气动阻尼研究, 工程力学, 21 (2004): 26-30+47.
Y. Quan and M. Gu, Wind tunnel test study of aerodynamic damping of super highrise buildings, Eng. Mech.,
21(2004): 26-47.
[9] Y. Quan and M. Gu, Experimental evaluation of aerodynamic damping of square super high-rise buildings, Wind
and Struct., 8(2005): 309-324.
[10] M. Gu and Y. Quan, Across-wind loads of typical tall buildings, J. Wind Eng. and Ind. Aerodyn., 92(2004):
1147-1165.
[11] 邹浩良, 梁枢果, 顾明, 高层建筑气动阻尼评估的随机减量技术, 华中科技大学学报 (城市科学版), 20
(2003): 30-33.
L.H. Zou, S.G. Liang and M. Gu, Evaluation of aerodynamic damping in wind-induced vibration of tall buildings
by random decrement technology, J. Hu St. (Urban Science Edition), 20(2003): 30-33.
[12] A. Kareem, T. Kijewski and Y. Tamura, Mitigation of motions of tall buildings with specific examples of recent
applications, Wind Struct., 3(1999): 201-251.
[13] Y.M. Kim and H. Kawai, Aerodynamic methods for reducing bending and torsional vibrations of tall building,
Proc. 10th Int. Con. on Wind Eng., Denmark, Copenhagen, 1999, pp. 673– 667.
[14] Y.M. Kim and K.-P. You, Dynamic responses of a tapered tall building to wind loads, J. Wind Eng. and Ind.
Aerodyn., 90(2002): 1771-1782.
[15] 黄鹏, 顾明, 全涌, 高层建筑气动阻尼的实验研究, 第十二届全国结构风工程学术会议, 2003, pp. 359-
364.
P. Huang, M. Gu and Y. Quan, Wind tunnel study of aerodynamic damping of super high-rise buildings, Proc.
12th National Conf. on Struct. Wind Eng., 2003, pp. 359-364.
[16] GB50009-2001, 建筑结构荷载规范, 中国, 2001.
GB50009-2001, Architectural structure load standards load code, China, 2001.
[17] AIJ 2004 Recommendations for Loads on Building, Architectural Institute of Japan, 2004.
[18] Y. Tamura and S.Y. Suganuma, Evaluation of amplitude-dependent damping and natural frequency of buildings
during strong winds, J. Wind Eng. and Ind. Aerodyn., 59 (1996): 115-130.

824