Two-stage damage localization method by integrating stochastic resonance theory and RobustICA algorithm

Qing-hua SHEN,Wei-hong CHEN,Shao-fei JIANG,Sheng-lan MA
DOI: https://doi.org/10.7631/issn.1000-2243.2014.06.0916
2014-01-01
Abstract:In order to realize the structural damage identification problem in the case of strong noise and low signal- to- noise ratio( SNR) environment,a two- stage damage location approach based on independent component analysis and nonlinear denoising derived from SR theory is proposed in this paper. In the first stage,structural dynamic responses and background noise are processed through the nonlinear SR system for enhancing the response signal. In the second stage,the feature components involving damage information are extracted via the robust independent component analysis( Robust ICA) algorithm and then are used to detect the structural response anomalies. Afterwards,the maximum of normalized source distribution vector( NSDV) is computed. On the basis of determined the damage index,the maximum of NSDV is employed to locate the structural damage. Numerical results of a frame show that the proposed algorithm can be successfully implemented the instant and location of damage in the case of low SNR with 5d B.
What problem does this paper attempt to address?