A new method for structural modal identification using sparse component analysis

摘要：

　　Operational modal analysis plays an important role in structural health monitoring of large-scale structures.Blind modal identification methods based on the blind source separation(BSS)theory has been applied to output-only modal identification to directly separate the modal responses and mode shapes from vibration signals.Sparse component analysis(SCA)which is one of the most popular methods of BSS has been applied to modal identification successfully because of its capability of handling non-stationary excitation and under-determined problems.In the process of SCA,the clustering number is a very important parameter for modal matrix estimation.However,it is always unknown during the estimation,which brings inconvenience for the application.In this paper,a modified SCA method equipped with a process of estimating the number of clusters is proposed.After transforming the structural responses into time-frequency domain,single-source-points detection process is adopted to estimate the number of clusters and identify the modal matrix.For underdetermined case,modal responses are reconstructed using sparse reconstruction technique.Frequencies and damping ratios are identified from the modal responses subsequently.The proposed method is verified by a real bridge.It is demonstrated that the proposed method is useful for civil structures under non-stationary excitation and with limited sensors.

关键词： Operational modal analysis blind source separation sparse component analysis

作者: Xiao-Jun Yao Chun-Xu Qu Ting-Hua Yi Hong-Nan Li

作者单位: School of Civil Engineering,Dalian University of Technology,Dalian,China School of Civil Engineering,Shenyang Jianzhu University,Shenyang,China

会议类型: 国际会议

会议名称: The 7th World Conference on Structural Control and Monitoring(7WCSCM)(第七届结构控制与监测世界大会)

会议地点: 青岛

会议语种:英文

页码: 248-256

在线出版日期: 2018-07-22（万方平台首次上网日期，不代表论文的发表时间）