摘要
提出了一种基于经验模态分解(Empirical Mode Decomposition,EMD)剔除位移响应漂移量的算法。这种算法将漂移成分视为EMD分解的最终残量(Residue),从原时程响应信号中分离并剔除EMD分解的最终残量后,再利用本征模函数(Intrinsic Mode Functions,IMFs)进行信号重构,能够得到接近真实的响应时程信号。该方法引入到实际桥梁结构的地震响应时程分析中,成功地解决了剔除结构地震响应位移漂移的难题。试算和实际计算效果表明此种方法效果良好,因而在结构地震响应分析中具有重要实用价值。
Will come into being ＂displacement drift＂, when the ＂large-mass method＂ is adopted to analyze the structural earthquake responses. A new algorithm to remove this drift components based on empirical mode decomposition （EMD） is put forward, which regards the drift component as the residue of EMD, separates and removes the residue from original signal of time history response data by using EMD, reconstructs the response signal though using intrinsic mode functions （IMF） of EMD. Then the displacement response signal approximated to real one is obtained. By applying the method to analyse earthquake time history responses of real bridge structures, the difficult problem of removing displacement drift from time history data is successfully solved. The results of trial and true calculations are precise, so it can be used for the structural earthquake response analysis.
出处
《振动工程学报》
EI
CSCD
北大核心
2009年第4期438-442,共5页
Journal of Vibration Engineering
基金
陕西省建设厅建设科技专项资助项目(2006-SJK-02)
陕西省自然科学基金资助项目(2005E205)
关键词
桥梁抗震
经验模态分解(EMD)
时程分析
位移漂移
bridge vibration reduction
empirical mode decomposition （EMD）
time history analysis
displacement drift
作者简介
傅光耀（1954-），男，教授，工学博士。电话：（029）85363381；E—mail：fugy@chd．edu．cn
