歩行者高さにおける低頻度で発生する高風速のワイブル分布に基づく統計的予測

Abstract

The low-occurrence strong wind speed (LOSWS) at the pedestrian level is important regarding the human and structure safety. Although several statistical methods were developed in previous studies, their accuracy and applicability for the cases with various building layouts remain unclear. This study analyzed the performances of two statistical methods (3W and 2W methods) for estimating the LOSWS at the pedestrian level of an isolated building case (Case-IB) and an actual urban case (Case-TPU). The 3W and 2W methods are based on the three-parameter and two-parameter Weibull distributions, respectively. The LES result of Case-IB was from a validated LES database. The LES of Case-TPU was validated with a wind tunnel experiment, and was found to have acceptable accuracy. It was found that the peak factors with exceedance probabilities 𝑞 = 10%, 1% and 0.1% of the 3W and 2W methods are close to those of the LES results. The relative errors of the LOSWSs with 𝑞 = 10%, 1% and 0.1% predicted by both methods are within 10% at most probe points for both cases. The high accuracy and robustness of the 3W and 2W methods were confirmed. The findings in this study are supportive to the further applications of these statistical methods in urban area cases.