基于模态分析理论和神经网络的斜拉桥拉索损伤识别研究  

Research on Damage Identification in Stay Cables of Cable-Stayed Bridge Based on Modal Analysis Theory and Neural Networks

作　　者：张治国[1] 张谢东[2] 

机构地区：[1]同济大学,上海200092 [2]武汉理工大学,湖北武汉430063

出　　处：《中南公路工程》2007年第2期67-72,共6页Central South Highway Engineering

基　　金：国家自然科学基金项目(10372075)

摘　　要：将振动模态分析和神经网络技术结合起来,以振动模态构造的损伤标识量作为神经网络识别输入的特征参数,进行结构健康监测。根据云阳长江公路大桥设计资料,考虑桥梁拉索结构的单构件损伤、2个构件损伤、3个构件损伤3类损伤工况,分别采用了模态频率、位移振型模态、曲率模态3种指标作为神经网络的输入参数,共建立9个BP神经网络模型进行了桥梁损伤识别的研究。研究结果表明基于振动模态分析理论和BP神经网络的桥梁损伤识别方法可用于识别斜拉桥拉索结构的损伤位置和损伤程度。Vibration modal analysis is integrated with neural networks in the thesis. Damage signatures for damage identification formed by vibration modal parameters are inputted to neural networks as parameters for structural health monitoring. According to the design data of highway cable-stayed bridge on Yunyang Changjiang river, the focus of the research is placed on three instances, including： one component of the bridge is damaged; two components and three components are damaged. Modal frequency, displacement modal shape and curvature modal shape are used as BP neural networks import vector respectively; sample data of each damaged state are collected; nine BP neural networks models are established for researching bridge damage detection, The result indicates that the method based on vibration modal analysis theory and BP neural networks can detect not only the damage position but the damage degree.

关 键 词：损伤识别 模态分析 神经网络 斜拉桥 

分 类 号：U448.27[建筑科学—桥梁与隧道工程]