Citation

Shen S, Li W. China Saf. Sci. J. 2019; 29(6): 158-164.

Copyright

(Copyright © 2019, China Occupational Safety and Health Association, Publisher Gai Xue bao)

DOI

10.16265/j.cnki.issn1003-3033.2019.06.026

PMID

unavailable

Abstract

In order to improve the accuracy of post-test evaluation of tanks by non-expert inspectors and comprehensively analyze those common types of noise, an acoustic emission signal processing method based on wavelet analysis and pattern recognition was proposed The experiment of Q235B three-point bending sandpaper rub, rubber handle percussion and electric noise was designed, and the experimental data of new acoustic emission waveform feature analysis were collected to obtain different kinds of signal characteristics. Finally, the large sample data was cluster analyzed to distinguish different types of sound sources based on the proposed signal processing method and the new feature quantity. The results show that the proposed method can quantitatively reflect the sudden burst of acoustic emission waveforms and identify the signal type of acoustic emission; it is also found to be able to extract typical features of valid acoustic emission signals and preserve key frequency band wavelet coefficients to reconstruct waveforms, thus making it suitable for acoustic emission testing of tanks. © 2019 China Safety Science Journal

Language: zh