指向性
强地震动
峰值地面加速度
反应谱
地震动
格林函数
地震学
相关函数(量子场论)
地质学
随机模拟
加速度
功能(生物学)
声学
光谱密度
物理
计算物理学
数学
工程类
电信
统计
经典力学
生物
进化生物学
天线(收音机)
作者
Longfei Ji,Xu Xie,Xiaoyu Pan
出处
期刊:Seismological Research Letters
[Seismological Society]
日期:2022-09-07
卷期号:94 (1): 331-349
被引量:2
摘要
Abstract Inputting reasonable ground motion is very significant in the seismic design of engineering structures under near-fault earthquakes. At present, the stochastic Green’s function method has been successfully applied to the simulation of moderate to high-frequency ground motions, but its accuracy is poor for low-frequency ground-motion simulation. In this study, an improved stochastic Green’s function method that is used to simulate broadband ground motion is established by considering the variation of the correlation of phase spectra among small earthquakes in different subfaults with the frequency and distance as well as the variation of the radiation pattern with the frequency and distance. Taking the 1994 Northridge earthquake in America and the 2013 Lushan earthquake in China as examples, the simulation results by the improved stochastic Green’s function method are compared with observed ground-motion records. The results show that considering the influence of near-field and intermediate-field terms has a little effect on the accuracy of ground-motion simulation. The directivity effect of near-fault ground motion can be reflected to a certain extent by considering the variation of the correlation of phase spectra among small earthquakes in different subfaults with the frequency and distance. Considering both the variation of the correlation of phase spectra among small earthquakes in different subfaults with the frequency and distance and the variation of the radiation pattern with the frequency and distance, the simulated acceleration response spectra generally show good agreement with the observed records. Therefore, the improved stochastic Green’s function method proposed in this study can simulate the broadband ground motion effectively.
科研通智能强力驱动
Strongly Powered by AbleSci AI