基于随机动力响应时程统计指标的结构损伤初步识别  

TIME-BASED STOCHASTIC DYNAMIC RESPONSE STATISTICAL CHARACTERISTICS FOR PRELIMINARY IDENTIFICATION OF STRUCTURAL DAMAGE

作　　者：肖青松 雷家艳[1] 施伟 王子豪 Xiao Qingsong;Lei Jiayan;Shi Wei;Wang Zihao(School of Architecture&Civil Engineering,Xiamen University,Fujian 361005,China)

机构地区：[1]厦门大学建筑与土木工程学院

出　　处：《动力学与控制学报》2019年第6期567-574,共8页Journal of Dynamics and Control

基　　金：国家自然科学基金青年科学基金项目(51508486);福建省自然科学基金青年创新基金项目(2016J05136);厦门市交通基础设施智能管养工程技术研究中心开放基金项目(TCIMI201806)~~

摘　　要：本文以八层剪切型钢框架模型在白噪声激励作用下的随机振动实验为基础,分别从结构加速度响应的均值、方差、偏度、峰度、概率密度函数和累积分布函数等多个统计特征,对结构无损及9种模拟损伤工况进行识别,结果表明:结构随机振动加速度响应时程的偏度和峰度对结构损伤的表征效果不明显,而概率密度函数能初步判别损伤,方差指标在进行损伤判别的基础上,具有良好的损伤定位能力.Based on the white noise random vibration experiment of an 8-story steel shear building model,the structural undamaged and 8 damaged cases were analyzed from the structural acceleration responses′mean,skewness,kurtosis,probability density function(PDF)and cumulative distribution function(CDF)respectively.The results demonstrated that the difference of skewness and kurtosis is not obvious to structural damage,but the acceleration probability density function could be used for preliminary damage identification to determine whether the damage has happened or not,and the statistical characteristics of variance could locate the damages.

关 键 词：损伤识别 统计特征 随机振动 加速度 方差 

分 类 号：G63[文化科学—教育学]