基于小波降噪和主成分分析的结构损伤识别  

Structure damage detection based on wavelet noise reduction and principal component analysis

作　　者：赵怀山 郭伟超[1] 高新勤[1] 杨振朝[1] 李言[1] ZHAO Huaishan;GUO Weichao;GAO Xinqin;YANG Zhenchao;LI Yah(School of Mechanical and Precision Instrument Engineering, Xi~an University of Technology, Xi~an 710048, China)

机构地区：[1]西安理工大学机械与精密仪器工程学院,陕西西安710048

出　　处：《西安理工大学学报》2017年第4期437-442,共6页Journal of Xi'an University of Technology

基　　金：国家自然科学基金资助项目(51505377);陕西省自然科学基础研究计划资助项目(2015JQ5182);陕西留学人员科技活动择优资助项目(203-253081605);中国博士后科学基金会资助项目(2016M592821)

摘　　要：对服役工程结构的状态进行长期监测的过程中,所测信号的数量往往非常巨大,而且信号包含有各种频率成分的环境噪声干扰,严重阻碍了准确识别工程结构状态的效率和准确率。针对这个问题,本文提出了基于小波降噪和主成分分析的结构损伤识别方法。首先采用小波降噪对测试信号进行处理并转换成顺序统计量,然后运用主成分分析对顺序统计量进行降维,提取有用特征矢量,最后利用统计方法构造损伤指标和控制线,通过控制线识别结构的状态变化。同时,论文使用某在役钢架桥的数值模拟及真实实验测量数据对该识别方法进行了验证,结果表明,该损伤识别方法能够有效提取振动信号的特征信息,能准确识别结构的状态变化。In the long-term monitoring of the structural, the recorded signals usually contain thousands of data and various environment noise. This leads to the fact that it is impossible to ef- ficiently and exactly identify the change happened in structure. To solve this problem, this paper proposed a structural damage identification method based on the wavelet de-noising technique with the signal converted into an order statistic. After that, the principal component analysis is adapted to extract the feature vectors of the order statistics containing the changes of structure and to reduce the data dimension. ~＇,,la~ feature vectors using statistic knowledge. The change of structure could be observed by the dam- age index and the damages level of structure could be evaluated by the control line. The proposed method is verified by using the data obtained from a numerical simulation and the measurement {or a real bridge. The result shows that the proposed damage identification method can efficiently extract the characteristic information of the vibration signal and that it can accurately the state changes of structure.

关 键 词：小波降噪 主成分分析 损伤识别 顺序统计量 

分 类 号：TU392.4[建筑科学—结构工程]