Multivariable wavelet finite element-based vibration model for quantitative crack identification by using particle swarm optimization. (4th August 2016)

• Record Type: Journal Article
• Authors: Zhang, Xingwu; Gao, Robert X.; Yan, Ruqiang; Chen, Xuefeng; Sun, Chuang; Yang, Zhibo
• Is Part Of: Journal of sound and vibration. Volume 375(2016)
• Page Start: 200
• Page End: 216
• Publication Date: 2016-08-04
• Subjects: Crack -- Quantitative identification -- Multivariable wavelet finite element method -- Particle swarm optimization -- Experiment; Sound -- Periodicals; Vibration -- Periodicals; Son -- Périodiques; Vibration -- Périodiques; Sound; Vibration; Periodicals; Electronic journals; 620.205
• DOI: 10.1016/j.jsv.2016.04.018
• Languages: English
• ISSNs: 0022-460X

Abstract

Abstract: Crack is one of the crucial causes of structural failure. A methodology for quantitative crack identification is proposed in this paper based on multivariable wavelet finite element method and particle swarm optimization. First, the structure with crack is modeled by multivariable wavelet finite element method (MWFEM) so that the vibration parameters of the first three natural frequencies in arbitrary crack conditions can be obtained, which is named as the forward problem. Second, the structure with crack is tested to obtain the vibration parameters of first three natural frequencies by modal testing and advanced vibration signal processing method. Then, the analyzed and measured first three natural frequencies are combined together to obtain the location and size of the crack by using particle swarm optimization. Compared with traditional wavelet finite element method, MWFEM method can achieve more accurate vibration analysis results because it interpolates all the solving variables at one time, which makes the MWFEM-based method to improve the accuracy in quantitative crack identification. In the end, the validity and superiority of the proposed method are verified by experiments of both cantilever beam and simply supported beam. Highlights: MWFEM-based method is proposed for quantitative crack identification. The intelligent PSO is utilized in optimization of the crack parameters. Compared with traditional methods, MWFEM can achieve more accurate results. Experiment results indicate that the relative errors are less than 14%.