经验模分解及小波分析在结构损伤识别中的应用:试验研究  

Structural damage detection using empirical mode decomposition and wavelet analysis: experimental investigation

作　　者：陈隽[1] 徐幼麟[2] 李杰[1] 

机构地区：[1]同济大学建筑工程系,上海200092 [2]香港理工大学土木与结构工程系

出　　处：《地震工程与工程振动》2007年第1期110-116,共7页Earthquake Engineering and Engineering Dynamics

基　　金：国家自然科学基金项目(50308020;50321803;50420120133)

摘　　要：本文利用一个三层剪切型结构模型在各种激励下的振动台试验,研究经验模分解(empiricalmode decomposition,EMD)和小波分析(wavelet analysis,WA)在结构损伤识别中的应用。研究针对结构刚度突然损失的损伤类型,并在试验中通过连接在模型两侧的弹簧的断裂来模拟。利用EMD和WA分析试验记录到的结构加速度信号来识别结构损伤发生的时刻和位置。试验结果表明,EMD和WA方法均可利用分解信号中的尖峰准确识别结构损伤发生的时刻,并利用信号尖峰在结构上的空间分布来确定损伤位置。研究表明,EMD和WA都是进行结构在线检测的良好方法。This paper presents an experimental investigation on the applicability of empirical mode decomposition （EMD） and wavelet analysis （WA） for identifying the structural damage caused by a sudden change of structural stiffness. A series of free vibration, random vibration and earthquake simulation tests were performed in a three story shear building installed on a shaking table. Structural damage was simulated by suddenly releasing two pre-tensioned springs those were horizontally connected to the first floor of the building. EMD and WA were then applied to the measured acceleration histories to identify the damage instant and the damage location. The results show that the damage instant may be accurately detected in terms of the data by EMD and WA, whilst the damage location may be damage spikes extracted directly from the measurement determined by the spatial distribution of the spikes along the building.

关 键 词：经验模分解 小波分解 损伤识别 

分 类 号：TU311.3[建筑科学—结构工程]