A novel dynamic load identification approach for multi-source uncertain structures based on the set-theoretical wavelet transform and layered noise reduction. (May 2023)

• Record Type: Journal Article
• Authors: Wang, Lei; Xu, Hanying; Liu, Yaru
• Is Part Of: Structures. Volume 51(2023)
• Page Start: 91
• Page End: 104
• Publication Date: 2023-05
• Subjects: Dynamic load identification -- Wavelet transform -- Uncertainty -- Interval analysis -- Denoising; Structural engineering -- Periodicals; 624.1
• DOI: 10.1016/j.istruc.2023.03.037
• Languages: English
• ISSNs: 2352-0124

Abstract

Abstract: This paper presents a dynamic load identification method based on wavelet decomposition. This method can identify multiple types of dynamic loads. Firstly, the parametric matrix is simplified to a diagonal matrix requiring the number of modal truncations by taking advantage of the orthogonality of the principal mode shapes and the mass, stiffness and damping matrices. By using the excellent characteristics of the wavelet space, such as the integrity of the wavelet space and the multi-resolution analysis, the generalized matrix of each parameter is transformed into the wavelet transform, which can effectively express the response history with the wavelet coefficients, further improve the computational efficiency, reduce the singularity of the inverse matrix, and enhance the anti-noise property of load recognition. In order to get closer to the practical engineering application, the interval analysis method of Taylor series expansion is proposed, and the propagation analysis of uncertain variables is carried out. Finally, the wavelet semi-soft threshold method is used to filter the signal to improve the accuracy of load recognition. The applicability of the proposed method is verified by numerical examples and simulation test cases.