stationary wavelet transform

 

 

OU Fenlan1, GUO Jinquan1, ZHONG Jianfeng1, ZHONG Shuncong1，2

 

(1. School of Mechanical Engineering and Automation, Fuzhou University, Fuzhou 350108, China;

 

2. Fujian Key Laboratory of Medical Instrument and Pharmaceutical Technology, Fuzhou 350000, China)

 

 

Abstract: Aiming at the difficulty in effective determinations of the existence of the cracks /defects in cantilever beams with small cracks due to weak crack effects, a methodology was proposed for small crack identification in cantilever beams by using stationary wavelet transform (SWT). Modal analysis of cantilever beams were modelled and analyzed by using finite element (FE) method. The obtained modal shapes of cantilever beams were orthogonally decomposed by using SWT into approximation coefficients (which include the lowfrequency components of modal shapes) and detail coefficients (which include the highfrequency components of modal shapes). Finally, the SWT detail coefficients were used to extract small defect features, therefore, small crack localization and identification in cantilever beams were realized. The FE and SWT results indicate that the detail coefficients for the first four modal shapes of damaged cantilever beams could be effectively used for the localization of small cracks. The SWTbased method has excellent detection precision and robustness. Small cracks in engineering structures could be identified avoiding the catastrophic failures, especially when the crack is initiated. Therefore, it could be recommended in real engineering applications.

 

Key words: vibration modal shape; stationary wavelet transform; finite element analysis; crack detection