Wind Tunnel Study Of Wind Pressure Distribution On Xi an s Tallest Building And The Finite Element Calculation And Analysis

The Eighth Asia-Pacific Conference on Wind Engineering, December 10 14, 2013, Chennai, India Wind Tunnel Study Of Wind Pressure Distribution On Xi an s Tallest Building And The Finite Element Calculation And Analysis Guoqiang Zhang 1, Zhenzhen Dou 2,Jiawu Li 3 1 Student of School of Highway, Chang an University, Xi an710064,china,e-mail:913353723@qq.com 2 Student of School of Highway, Chang an University, Xi an710064,china,e-mail:820067059@qq.com 3 Professor of School of Highway, Chang an University, Xi an710064,china,e-mail:34086100@qq.com ABSTRACT The new Chang an plaza building with a height of 303 m and 64-storeys, located in Xi an of Shaanxi province, will be the tallest building in west of China after the completion. This paper presents selected results from a wind tunnel study of wind pressure distribution on the building. According to the data of the wind tunnel experiments, a detailed study was conducted to investigate the influences of incident wind direction and get the law of wind pressure varying with wind direction angle.this will provide a reference basis for the design of glass curtain wall. The flow-field of the building with the consideration of aerodynamic effects of the wind is numerically simulated by FLUENT, which is an analysis software of computational fluid dynamics. Comparisons between the mean wind pressure of the numerical simulation and the wind tunnel test of the building model are presented in detail. The wind tunnel test along with the finite element analysis, performed to calculate the wind-induced vibration response of the structure, provide the necessary reference basis for the wind resistant design of the building. Keywords: Tall building, Surface wind pressure, Wind tunnel, FLUENT, Finite element analysis Introduction Modern tall buildings are usually constructed with innovative structural systems and high strength materials tending to be more flexible and lightly damped than those in the past. As a consequence, the sensitivity of these tall buildings to dynamic excitations, such as strong wind, has increased.this has resulted in a greater emphasis on understanding the structural behavior of modern tall buildings under strong wind actions. The new Chang an plaza building located in Chang an Area,Xi an, has a height of 303m and is the highest building of Xi an. The main structure of this building has 64 storeys.it is a steel and concrete composite structure with a building plan form very close to a square shape. Therefore, the aspect ratio between the height and transverse width is about 7, which has exceeded the criteria in the current design codes and standards of China. This illustrates that the new Chang an plaza building is a flexible and slender structure. All these facts make a comprehensive study of wind effects on this super tall building of particular importance. Proc. of the 8th Asia-Pacific Conference on Wind Engineering Nagesh R. Iyer, Prem Krishna, S. Selvi Rajan and P. Harikrishna (eds) Copyright c 2013 APCWE-VIII. All rights reserved. Published by Research Publishing, Singapore. ISBN: 978-981-07-8011-1 doi:10.3850/978-981-07-8012-8 253 920

Fig. 1 The new Chang an plaza Wind tunnel testing is an effective method for investigating wind effects on buildings and structures. However, in general, it is difficult to reproduce the exact field conditions such as incident turbulence and terrain characteristics in wind tunnel tests. The results obtained in this study can actually provide some useful results of wind effects on the building. Meanwhile, this also provided an excellent opportunity to compare the real structural performance of the super tall building with wind tunnel test results for the purpose of improving the modeling techniques in wind tunnel tests. This paper presents selected results from the combined wind tunnel, numerical simulation and finite element analysis.the wind tunnel test can generate detailed and additional results that are not available from the field measurements. Therefore, the numerical simulation and the wind tunnel study are complementary so that the understanding of wind effects on the super tall building can be improved. The main objective of this paper is to further study the understanding of wind effects on super tall buildings and the behavior of high-rise structures under wind conditions by means of wind tunnel tests and numerical simulation in order to apply such knowledge to design.the finite element analysis, performed to calculate the wind-induced vibration response of the structure, provides the necessary reference basis for the wind resistant design of the building. WIND TUNNEL EXPERIMENTS The wind tunnel experiment was carried out in the boundary layer wind tunnel at Chang an University in China. The dimensions of the working section in the wind tunnel are 3 m wide 2.5 m high and 15 m long. The typical boundary layer wind flows representing the urban flow configuration,as specified in the China National Load Code as terrain D, were simulated for the model test by means of placing a barrier at the entrance of the wind tunnel, arrayed cubic roughness elements with different sizes on the tunnel floor upstream of the building model. For the isolated building case, no neighboring building around the building was considered in the wind tunnel test. Fig. 2 shows a photo of the models mounted in the wind tunnel representing the existing surrounding conditions. 921

Fig. 2 The models in the wind tunnel test A rigid model with a geometric length scale of 1:280 was made to represent the building. The model was made of polymethyl methacrylate, PVC pipe, balsa and foam. The wind tunnel block ratio is 2.6%<5%, conforming to the experiment requirements. The mean wind speed profiles of the fully developed boundary layer flows were found to follow a power law with exponents of = 0.30 with the gradient heights of 550 m. The measured mean wind speeds and turbulence intensities at various heights over the test section are illustrated in Fig. 3. Fig. 3 Mean wind speed and turbulence intensity profiles The model is equipped with 216 pressure taps of 1 mm diameter on the surface, which is illustrated in Fig. 4. In the wind tunnel test, wind direction was defined from the north along an anti-clockwise direction varied from 0º to 345º with increments of 15º. The wind velocity at the model height is approximately 10 m / s. Data sampling frequency was about 312Hz with data sampling length of 9 000. 922

Fig. 4 Pressure taps of 1mm diameter on the surface Aerodynamic forces on tall buildings are usually described in terms of mean pressure coefficients with incident wind direction as a variable. Mean force coefficients are defined as follows: 2α 0.60 C = ( Z / H ) C = (296.8 / 550) C = 0.691C i P r G i i i Pr Pr Pr i where Z r is the height of measuring point, H G =550m is the the gradient heights, C pr is the i wind pressure coefficient of wind tunnel test, CP is the mean pressure coefficients of measuring point i, =0.30 is the surface roughness coefficient. The actual wind pressure value of the pressure measuring point for R years return period can be calculated by the following formula. p = C w i i R p G, R i where p R is the actual wind pressure value of the pressure measuring point for R years return period, w GR, is the actual wind pressure gradient for R years return period. WIND TUNNEL EXPERIMENTAL RESULTS According to the experimental results of each measuring point pressure obtained in the wind tunnel test, MATLAB is used to get the mean wind pressure coefficient of each measuring point. Using Sufer software to draw windward side of average wind pressure coefficient distribution at 0º wind direction angle, as shown in Fig. 5. 923

North West South East Fig. 5 Mean pressure coefficients distribution of windward side at 0º wind direction angle Origin software are used to draw the mean wind pressure coefficient changes of measuring points varying from the wind direction angle, selecting a103 measuring point at about 2/3 height of the model which shows the typical wind pressure change trend, as shown in Fig. 6.a. Choose point a53 to a133 to draw the mean wind pressure coefficient changes varying from different heights at 0º wind direction angle, as shown in Fig. 6.b. a). point a103 b). point a53 to a133 Fig. 6 Mean pressure coefficient of measuring points On the basis of the time history signals of 24 kinds of wind direction angle working conditions,calculate the maximum and minimum peak wind pressure coefficient of each measuring point. Thus actual building external peak wind pressure distribution is obtained, as shown in Fig. 7. 924

Fig. 7 The maximum wind pressure of 50-year return period NUMERICAL SIMULATION In the first processing module of FLUENT software GAMBIT, the geometric model is established according to the size of the prototype to avoid size effect may be the impact of the results. The aerodynamic interference effect between the actual situation in the building can be ignored because buildings around the high-rise building is less The ratio of computing zone and building size is 10 at 0º wind direction angle, the computing zone is 1500m 1000m 1800m, as shown in Fig. 8. The buildings along the wind is located the first third place of the calculating zone. Grid adopts tetrahedral mesh, mesh used in river basin by line divided step by step to get to the surface to the body, namely: at 0 angle of the wind using the k- turbulence model (standard) to get 1147981 nodes, 1799353 elements. Fig. 8 Computing river basin of 0 wind angle The river inflow place of the computing zone uses FLUENT software import velocity-inlet 925

boundary conditions, where velocity v, k and ε should be defined. Use the typical ground rough categories corresponding to the atmospheric boundary conditions for the coming flow conditions to calculate and simulate the atmospheric boundary layer wind profile exponential distribution. The average wind speed profile V ( z), turbulent kinetic energy k and the turbulent dissipation rateε in the inflow port adopts UDF program profile to interface implementation with FLUENT software. The fully developed flow boundary conditions (outflow) is selected as the export condition. The top and on both sides of the computing river basin uses symmetric boundary conditions, while the building surface and ground adopt no slip wall condition. Considering the wind effects on buildings belong to the steady state, the SIMPLEC algorithm is used with smaller owe relaxation coefficient because of its accelerating the convergence speed. Simulate working condition of 0 wind direction angle and calculate the average wind pressure on the surface of the building. The following are the building surface wind pressure and wind pressure coefficient of CFD calculation chart, as shown in Fig. 9. a. Contours of Pressure Coefficients b. Contours of Total Pressure Fig. 9 Distribution of Pressure Coefficients and Total Pressure at 0 wind angle THE FINITE ELEMENT CALCULATION AND ANALYSIS The dynamic characteristics analysis of the tall building adopts discrete structure finite element method. According to the structure characteristics of the high-rise building and the preliminary design scheme of this research provided by the local design institute, to ensure the quality and stiffness are consistent with the actual structure, finite element analysis model is established by using large-scale finite element software, the basic method is: (1) using three-dimensional space beam element to simulate the columns, beams, beam, oblique beam and concrete beam. (2) using three dimensional elastic shell element to simulate the shear wall. (3) constraint conditions of the model: at the bottom of the shear wall, the lateral column, the composite beam column degrees of freedom of six directions are constrained. The ANSYS model is illustrated in Fig. 10. 926

Windvibrationcoeficie 11 12 22...... 46 80 24 Proc. of the 8th Asia-Pacific Conference on Wind Engineering (APCWE-VIII) Fig.10 ANSYS model The wind pressure coefficient of each measuring point is performed into surface load time history, then load to the network frame roof surface unit. Dynamic time history response of the wind load analysis is conducted to calculate the displacement response of structures generated by mean and fluctuating wind under different wind angles and the wind vibration coefficient of 216 key measuring points generated by displacement response β z.the responses are shown in Fig. 11, here only selecting key measuring point a44 to give an typical example. β z = ( R s + Rd ) / Rs = 1+ Rd / Rs where R s is the structural displacement response generated by average wind, Rd is the displacement response of structure generated by pulsating wind. 2.8 2.6 Wi nd vibration coefficient 2.4 2.2 2.0 1.8 1.6 1.4 1.2 1.0 0 50 100 150 200 250 300 350 Wi nd di r ect i on angl e( o) a44 50-year return period a44 100-year return period Fig.11 The maximum wind vibration coefficient of key measuring point The main conclusions drawn from the finite element analysis results are that for 50-year return period of wind pressure, the most unfavorable wind direction angle of the tall building on the north and south surface are 165º, 255 285º, the most unfavorable wind direction angle of the tall building on the west and east surface are 75º, 105º, 165 195º and that for 100-year return period of wind pressure, the most unfavorable wind direction angle of the tall building on the north and south surface are 165º, 270 300º, the most unfavorable wind direction angle of the tall building on the west and east surface are 90 105º, 195º. Conclusions The objective of this study is to investigate wind effects on the building under strong 927

winds by means of wind tunnel tests, numerical simulation and finite element analysis. In the wind tunnel experiments, wind loads and wind-induced responses of this super tall building, such as force coefficients were presented and discussed in detail. On the other hand, the numerical simulation were compared with the wind tunnel results. Furthermore, the finite element analysis was performed to calculate the wind-induced vibration response of the structure. Some conclusions from the combined wind tunnel, numerical simulation and finite element analysis are summarized as follows. (1) The surrounding tall buildings is less, so it is not easily influenced by other architectural wake flow, but within a certain range of the wind direction angle the environment construction s impact on the mean wind pressure coefficient should not be ignored. It can be seen in the wind tunnel experiment results that wind pressure coefficient is higher in the windward side, which is close to 1.0, while the wind pressure coefficient of the bottom is less affected by other architectural wake flow. The side face and the leeward face show most negative wind pressure coefficient, especially the column part due to the vortex shedding, its value is close to -1.0. (2) Windward pressure on the top of the actual building and the negative pressure of lee side is large. Contraction at the top of the building is considered by numerical simulation, which obviously reduces the windward side and lee side pressure. From the aerodynamic point of view, in order to reduce the bending moment at the bottom of the building, the building with the proper contraction surface to improve the air flow state should be considered in the design. (3) In the midst of the four elevation windward side at the wind direction angle with height H 100 m, the positive wind load is possibly to be W k 1.0kpa / m. While H < 100 m, the positive pressure is rather small. Each building facade especially on both ends may appear small wind load, in many cases the wind load W k 1.0kpa / m, while as for the lee side the wind load W k < 0, but the absolute value is not big. The positive wind load of skirt building is small, but the absolute value of the negative wind load sometimes is quite large, W k 1.0kpa / m. (4) From the wind load of the building, the vertical preiection on the surface of the building can increase surface roughness, playing a role on the flow around characteristics, avoiding very outstanding wind load point. (5) Numerical simulation method can well predict the streamline around complex highrise building and average wind pressure distribution of the surface. Simulation results have a good match with the results of wind tunnel test on the whole. (6) As the wind vibration coefficient increases, the peak value of the fluctuating windinduced vibration response of the structure becomes more significant. When doing wind resistance design, use a simplified wind vibration coefficient i.e. use the wind vibration partition coefficient to design. (7) The wind vibration coefficient and shape coefficient analyzed and calculated by means of wind tunnel test results can directly used for building surface wind pressure calculation. 928

Acknowledgements The work described in this paper was fully supported by a grant from the boundary layer wind tunnel laboratory at Chang an University in China. The financial support is gratefully acknowledged. References Gu M, Zhou Y, Zhang F, Xiang HF. Dynamic responses and equivalent wind loads of the Jin Mao Building in Shanghai. In: Proceedings of the tenth international conference on wind engineering vol. 3. 1999. Q.S. Li, J.Y. Fu,b, Y.Q. Xiao, Z.N. Li, Z.H. Nid, Z.N.Xie, M. Gue,Wind tunnel and full-scale study of wind effects on China s tallest building,engineering Structures 28 (2006). Wind tunnel studies of buildings and structures, ASCE manuals and reports on engineering practice No.67, Task committee on wind tunnel testing of buildings and structures [M]. Aerodynamics committee aerospace division, American society of Civil Engineers, 1999. Li QS, Fang JQ, Jeary AP,Wong CK, Liu DK. Evaluation of wind effects on a super tall building based on fullscale measurements. Earthquake Engineering and Structural Dynamics 2000. AIJ Recommendations for Loads on Buildings. Tokyo (Japan):Architectural Institute of Japan; 1996. Melbourne WH. Turbulence effects on maximum surface pressures a mechanism and possibility of reduction. In: Proc. 5th int. conf. on wind eng. 1979. Li QS, Melbourne WH. Turbulence effects on surface pressures of rectangular cylinders. Wind and Structures 1999. Wind tunnel testing: A general outline. The boundary layer wind tunnel laboratory. Faculty of Engineering Science, The University of Western Ontario, London, Ontario, Canada N6A5B9, May, 1999. Lv Fuyu, Yang Shichao, Numerical Wind-tunnel Simulation of Wind Pressure on High-rise Buildings, Acta Scientiarum Naturalium Universities SUNYATSENI, NOV.2008. Li Zhengliang, Chen Sheng, Test study on the wind pressure distribution of a super high-rise building, Building Structure JAN. 2011. GB50009-2012. Load code for the design of building structures. Beijing:China Architecture & Building Press; 2012 [in Chinese]. 929