| description abstract | A short-rise-time gust is a typical nonsynoptic wind with abrupt changes in mean speed and turbulent intensity, which differs from normal winds at plain sites. Field measurements were conducted in a high-altitude deep-cut canyon, in which short-rise-time gusts frequently occur. Specifically, 49 gust segments were recorded and classified into three types based on the entire longitudinal wind-speed time history: rapid rise, slow fall (RRSF); slow change (SC); and slow rise, rapid fall (SRRF). The empirical mode decomposition (EMD) method was used to extract the time-varying mean wind speed in the longitudinal direction, and a rational function is suggested to fit the nonstationary mean wind speed. The evolutionary power spectral density (EPSD) of wind fluctuation indicates that the wind-speed history is a nonstationary process. A modulation function is adopted to convert the fluctuating wind speed into a stationary process. The fluctuating wind components are fitted using a normalized power spectral density (PSD) function. Finally, a simplified nonstationary modeling method for rapid reconstruction of short-rise-time gusts is presented, including time-varying mean wind speed in the longitudinal direction, modulation function, and normalized PSD functions in different spatial directions. A case study was conducted to demonstrate that the proposed method can reproduce nonstationary characteristics well. This study proposes a simplified model for generating nonstationary wind-speed time histories from the viewpoint of application feasibility, which is of significance to further study wind load characteristics and structural wind effects under nonsynoptic winds with typical nonstationary characteristics, particularly short-rise-time gusts in mountainous areas. | |
