ISSN:
1662-7482
Source:
Scientific.Net: Materials Science & Technology / Trans Tech Publications Archiv 1984-2008
Topics:
Mechanical Engineering, Materials Science, Production Engineering, Mining and Metallurgy, Traffic Engineering, Precision Mechanics
Notes:
In this paper, a cluster-based feature extraction from the coefficients of discrete wavelettransform is proposed for machine fault diagnosis. The proposed approach first divides the matrixof wavelet coefficients into clusters that are centered around the discriminative coefficient positionsidentified by an unsupervised procedure based on the entropy value of coefficients from a set ofrepresentative signals. The features that contain the informative attributes of the signals arecomputed from the energy content of so obtained clusters. Then machine faults are diagnosed basedon these feature vectors using a neural network. The experimental results from the application onbearing fault diagnosis have shown that the proposed approach is able to effectively extractimportant intrinsic information content of the test signals, and increase the overall fault diagnosticaccuracy as compared to conventional methods
Type of Medium:
Electronic Resource
URL:
http://www.tib-hannover.de/fulltexts/2011/0528/01/38/transtech_doi~10.4028%252Fwww.scientific.net%252FAMM.10-12.548.pdf
Permalink