当前位置: 首页> 国外交通期刊数据库 >详情
原文传递 Data-Driven Approach for Generating Tricomponent Nonstationary Non-Gaussian Thunderstorm Wind Records Using Continuous Wavelet Transforms and S-Transform
题名: Data-Driven Approach for Generating Tricomponent Nonstationary Non-Gaussian Thunderstorm Wind Records Using Continuous Wavelet Transforms and S-Transform
正文语种: eng
作者: Y. X. Liu;H. P. Hong
作者单位: Univ. of Western Ontario;Univ. of Western Ontario
关键词: Record-based simulation;Tricomponent thunderstorm winds;Nonstationary non-Gaussian process;Synthetic records
摘要: Abstract Strong thunderstorm winds cause damage to structures. However, the available number of the tricomponent thunderstorm wind record with a subsecond sampling time interval is limited. In the present study, a record-based procedure for generating tricomponent nonstationary non-Gaussian thunderstorm wind records was proposed. The procedure was based on the iterative power and amplitude correction algorithm framework but with modifications. The modifications were aimed at increasing the variability of the sampled record components by randomizing the power spectral density functions of processes through a digital filter in the frequency domain and improving the convergence by using a relaxation factor for the synchronized phase shift. The formulation and algorithm for the proposed procedure were given by considering the continuous wavelet transform with the harmonic wavelet and generalized Morse wavelet, and the generalized S-transform, which can provide good time localized resolution at high frequencies (low scales) and good resolution at low frequencies (high scales) simultaneously. The proposed procedure, unlike some of the algorithms available in the literature, matches the marginal mixture cumulative distributions of the seed record components and does not require the separation of low- and high-frequency wind components. The use of the proposed procedure to sample tricomponent thunderstorm wind records was shown.
出版年: 2023
期刊名称: Journal of structural engineering
卷: 149
期: 12
页码: 1.1-1.15
检索历史
应用推荐