Title
Detection of Localized Bearing Faults in Induction Machines by Spectral Kurtosis and Envelope Analysis of Stator Current
Abstract
Early detection of faults in electrical machines, particularly in induction motors, has become necessary and critical in reducing costs by avoiding unexpected and unnecessary maintenance and outages in industrial applications. Additionally, most of these faults are due to problems in bearings. Thus, in this paper, experimental bearing fault detection of a three-phase induction motor is performed by analyzing the squared envelope spectrum of the stator current. Spectral kurtosis-based algorithms, namely, the fast kurtogram and the wavelet kurtogram, are also applied to improve the envelope analysis. Experimental tests are performed, considering outer bearing faults at different stages, and the results are promising.
Year
DOI
Venue
2015
10.1109/TIE.2014.2345330
IEEE Transactions on Industrial Electronics
Keywords
Field
DocType
induction motors,three phase induction motor,algorithm design and analysis,signal processing,ball bearings,kurtosis,fault detection,vibrations
Induction motor,Algorithm design,Fault detection and isolation,Control theory,Control engineering,Electronic engineering,Bearing (mechanical),Engineering,Vibration,Stator,Kurtosis,Wavelet
Journal
Volume
Issue
ISSN
62
3
0278-0046
Citations 
PageRank 
References 
22
1.10
22
Authors
7