峰值法和子空间法在结构动力反分析时的对比  

Comparison between peak-picking method and subspace method in structural dynamic back analysis

作　　者：张舒翔 张其林[1] ZHANG Shuxiang;ZHANG Qilin(School of Civil Engineering,Tongji University,Shanghai 200092,China)

机构地区：[1]同济大学土木工程学院,上海200092

出　　处：《阜阳师范大学学报（自然科学版）》2020年第4期16-20,共5页Journal of Fuyang Normal University:Natural Science

基　　金：国家重点研发计划课题(2018YFD1101004)资助。

摘　　要：结构动力反分析一直是土木工程领域重要内容。尤其在结构建成后,监测系统对结构动力反分析的结果反映了结构的健康状态。本文就动力反分析现有的峰值法与子空间法进行算法实现,同时选取大模型和小模型做传感器监测数据处理,完成了算法的实例化模态识别和结果比较,得出两种算法的实用范围及误差比较。Structural dynamic back analysis is always an important content in the field of civil engineering, especially after the completion of the structure, the results of structural dynamic back analysis by the monitoring system reflect the health of the structure. In this paper, the existing Peak-picking method and Stochastic Subspace Identification method of dynamic back analysis are used to realize the algorithm. At the same time, large model and small model are selected to process the sensor monitoring data. The instantiation mode identification and result comparison of the algorithm are completed, and the practical scope and error comparison of the two algorithms are obtained.

关 键 词：动力分析 动力反分析 算法 模态识别 健康监测 

分 类 号：TU17[建筑科学—建筑理论]