Fault Classification of an Induction Motor Using Texture Features of Vibration Signals
This paper proposes an approach for a two-dimensional (2D) representation of vibration signals for the highly reliable fault classification of an induction motor. The resulting 2D data has texture characteristics showing repetitive patterns and this paper extracts these texture patterns by utilizing gray level co-occurrence matrix. Moreover, this paper employs a distance evaluation technique in order to avoid the unnecessary fault signatures in the extracted texture features. To identify multiple faults of an induction motor, this paper finally utilizes support vector regression as a classifier using the extracted fault signatures. Experimental results indicate that the proposed approach achieves high classification accuracy.
KeywordsFault classification Gray level co-occurrence matrix Induction motor Support vector regression Texture patterns
This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MEST) (No. NRF-2012R1A1A2043644), and the Leading Industry Development for Economic Region (LeadER) grant funded the MOTIE (Ministry of Trade, Industry and Energy), Korea in 2013 (No. R0001220).
- 1.Wang H, Chen P (2009) Fault diagnosis method based on kurtosis wave and information divergence for rolling element bearings. WSEAS Trans Syst 8(10):1155–1165Google Scholar
- 5.Kia SH, Henao H, Capolino G–A (2011) Some digital signal processing techniques for induction machines diagnosis. In: IEEE international symposium on diagnostics for electric machines, power electronics and drives (SDEMPED), Bologna, pp 322–329Google Scholar
- 7.Gaeid KS, Pind HW (2011) Wavelet fault diagnosis and tolerant of induction motor: a review. Int J Phys Sci 6(3):358–376Google Scholar