Structural Damage Identification Based on Extended Kalman Filter and Response Reconstruction

Aiming at the problems of low signal precision and uncertain sparsity in traditional structural damage identification methods, a structural damage identification method based on an extended Kalman filter and response reconstruction technology is proposed. Firstly, the extended Kalman filter is used for signal filtering analysis, and the standard deviation of the filtered signal is calculated to preliminarily analyze the damage location of the structure. Secondly, the Kalman filter is used for data fusion of the filtered signal, and the orthogonal matching pursuit algorithm of compressed sensing reconstruction is introduced to set the signal sparsity range and reconstruct the response. When the reconstructed signal meets the accuracy requirements of the Euclidean norm relative error, the standard deviations of the reconstructed signal are used to identify the damage location, and the damage degree of the structure is determined by the relationship between the signal standard deviation before and after the damage. The identification effect is verified by the damage location and degree of ASCE benchmark structure story and truss structure rod elements. The results show that the proposed method has certain reliability.