31 December 2010 Identification defect character of MMM signals based on wavelet singular entropy and RBFNN
Author Affiliations +
Proceedings Volume 7544, Sixth International Symposium on Precision Engineering Measurements and Instrumentation; 75443G (2010) https://doi.org/10.1117/12.886053
Event: Sixth International Symposium on Precision Engineering Measurements and Instrumentation, 2010, Hangzhou, China
Abstract
Metal magnetic memory is a novel NDT method that can be used to detect residual stress distribution of ferromagnetic components.Wavelet decomposition and entropy theory are used and wavelet singular entropy is introduced to extract characteristic from abnormal signals of defect. Furthermore, RBF neural network is utilized to identify defect character. Experimental results showed that, compared to the traditional gradient value, the proposed new method can be used to effectively reflect defect character and it is immune to the effect of noises.
© (2010) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Lan Zhang, Lan Zhang, Yongrui Zhao, Yongrui Zhao, Chong Tian, Chong Tian, } "Identification defect character of MMM signals based on wavelet singular entropy and RBFNN", Proc. SPIE 7544, Sixth International Symposium on Precision Engineering Measurements and Instrumentation, 75443G (31 December 2010); doi: 10.1117/12.886053; https://doi.org/10.1117/12.886053
PROCEEDINGS
6 PAGES


SHARE
Back to Top