（北京建筑大学土木与交通工程学院， 北京 100044）

［摘要］为了研究网架结构杆单元的损伤检测方法，对网架结构在瞬时激励下的损伤检测进行了数值模拟分析和试验分析。为了判断网架结构的损伤杆件位置,选取合适的激励方式，设置了两种不同的工况进行分析，分别提取了结构在损伤前后的时域信号，运用频响函数和主成分分析进行数据处理，选取频响函数向量夹角作为结构损伤指标。数值模拟分析时，提取了结构全部节点的时域信号用来判断损伤杆单元的位置；试验分析时，在基于节点应变能系数的传感器优化布置方法的基础上，提取少量节点的时域信号，分两步逐渐缩小所要检测的损伤范围，最终确定损伤杆单元的位置。数值模拟分析和试验结果显示基于频响函数与主成分分析的损伤检测方法可以准确判断损伤的杆单元的位置。

［关键词］损伤检测； 网架结构； 频响函数； 主成分分析

中图分类号：TU356文献标识码：A文章编号：1002-848X(2017)04-0096-06

Research on damage detection method for grid structures based on frequency response function and principal component analysis

Wang Menghong， Song Chunyue， Ji Chenmeng

（School of Civil and Transportation Engineering, Beijing University of Civil Engineering and Architecture, Beijing 100044, China）

Abstract:In order to study damage detection method for bar element of grid structures, numerical simulation analysis and experimental analysis were carried out for damage detection on grid structure under instantaneous excitation. In order to judge damaged bar member location of the grid structure, the proper excitation method was selected  and two working conditions were set to analyze. The time domain signals of the structure before and after damage were extracted respectively. The frequency response function and principal component analysis were used for data processing, and the angle between the frequency response function vectors was selected as a damage index. In the numerical simulation, the time domain signals of all nodes of the structure were extracted to determine  damaged bar member  location. In the experimental analysis, the time domain signals of a portion of the nodes were extracted on the basis of an optimal sensor placement method of the node strain energy coefficient. The range of damage detection was narrowed gradually in two steps and finally damage bar element location was determined. The results of the numerical simulation and experimental analysis show that the damage detection method based on the frequency response function and principal component analysis can judge damaged bar element location accurately.

Keywords:damage detection; grid structure; frequency response function; principal component analysis

*国家自然科学基金项目(51578038)。

作者简介：王孟鸿，博士后，教授，硕士生导师，Email: wangmh@bucea.edu.cn。