Evaluation of Aerodynamic Force Coefficients on Tall Buildings Based on Probabilistic Method and Its Application to Wind-Resistant Design

Abstract

The aerodynamic force coefficients prescribed in most codes and standards are an ensemble average of 10 or 20 10-min samples, which means that a fundamental uncertainty of 10-min samples cannot be properly considered. Thus, it also is very difficult to properly evaluate the uncertainty of wind load. In the present study, 10-min samples of fluctuating pressures on tall buildings with three side ratios were measured using wind tunnel tests. The probability density distributions of the aerodynamic force coefficients were evaluated first, and then new design parameters are proposed to properly evaluate the uncertainly of wind load. Calculation results indicated that the probability density distributions of local and generalized force coefficients follow a normal distribution, and the probability density distributions of peak pressure coefficients follow extreme value distributions. The positive peak pressure coefficients mainly follow a reverse Weibull distribution, and the negative peak pressure coefficients follow all the extreme value distributions depending on pressure tap positions and wind directions. The coefficient of variation of wind speed was found to have a larger influence on the adjusting coefficient than that of the non-exceedance probability of wind load. The coefficients of variation of the mean and fluctuating force coefficients were less than 0.1, and those of the peak pressure coefficients ranged from about 0.1 to 0.3, slightly larger than those of the mean and fluctuating force coefficients. Using the present results, aerodynamic force coefficients corresponding to the specific non-exceedance probability of wind load can be evaluated from the ensemble averaged values for square and rectangular tall buildings about 220 m high at full scale under a wind environment corresponding to an urban area.