Estimation Of Wind Load Effects And Wind Resistant Design On Buildings Considering Directionality

In the Load code for the design of building structures (GB50009-2012), the wind load on the roof was determined by the most unfavorable pressure coefficient and maximum wind velocity in each direction, namely, assuming the maximum design wind speeds was exactly from the direction of largest structural response. It was proved that this simplified method would result in a conservative prediction result of wind load. In order to make structural design projects more reasonable and economical, and provide a reference for future structural design and optimization, the reduction effect of wind direction on wind load of the roof was analyzed in this study.The main contents of this thesis are shown as follows:(1). Different homogeneity tests methods including the t test, standard normal homogeneity test and penalized maximal F test are presented. After comparing the advantages and disadvantages of each method, the t test method was adopted to test the homogeneity of the original records of annual wind speed data, then the date which does not meet the homogeneity requirements was corrected. Through the homogeneity correction, the distribution of the maximum wind speed changed, the tail of extreme wind speed distribution changed significantly as well.(2). A precise probability model of extreme wind speeds was proposed based on the classical extreme-value theory, according to this model, the extreme wind speeds whose return period was 50 years were obtained. Moreover, an alternative model of extreme wind speeds was established based on the sampled independent peaks, the extreme wind speeds whose return period was 50 years could be obtained as well by using this model. By comparing the results of both methods, it was shown that the extreme-value theory that based on small sample size would underestimate the design wind speed in every direction. Plus, selecting a appropriate threshold was the critical problem for the analysis based on small sample size.(3). Four theoretical methods which taking the reduction effect of wind direction into account were analyzed, including Simiu’s approach, Sector by sector method, Up-crossing method and Storm passage method. The strengths and weaknesses of each method were compared, the conclusion could be served as the basic theory for improvement of wind resistance design.(4). In order to analyze the reduction effect that geometrical character and building orientation put on wind directionality factor in AIJ RLB 2004, the definition of this factor in AIJ RLB 2004 was selected as a reference, and then the reasonability of the factor was verified. By combining the results of measured wind speeds and wind tunnel tests, it was shown that the wind directionality factor was not sensitive to geometrical character and building orientation. Therefore, the reduction factor of wind direction could be proposed in every city regardless of the roof type. Furthermore, with the pressure coefficients gained wind tunnel tests, for all roof types, the direction which had representative wind pressure coefficients were summarized. On this basis, supplementation and optimization of AIJ RLB 2004 were proposed.(5). In order to analyze the reduction effect that geometrical character and building orientation put on wind directionality factor in ASCE 7, the definition of this factor in ASCE 7 is selected as a reference, and then the reasonability of the factor was verified. It was shown that the geometrical character and building orientation have great effect on wind directionality factor in ASCE 7. The effect that changing length-width ratio, slope angle and rise span ratio put on wind directionality factor was summarized, and consequently a qualitative judgment for improvement design of structures was gained. In addition, according to the variation trends of wind directionality factors in every building orientation, the concept of optimal building orientation was put forward in this study. By using the judgment standard of optimal building orientations which was put forward in this study, for any city whose wind speed was known, the building orientations that had great wind reduction effects could be selected quickly.(6). The wind directionality factors in AIJ RLB 2004 and ASCE 7 were calculated by using the extreme value analysis based on small samples, eg. Peaks-Over-Threshold Approach. It was shown that the results gained from both codes were similar to the results obtained from classical extreme-value theory Gumbel method.