Diagnosis of Short Circuit Faults in Stator Winding of Motor based on Hidden Markov Model

Abstract

This paper proposes a new diagnosis method for short circuit faults in stator winding of motor based on Hidden Markov Model. Short circuit fault of a motor is one of the most probable faults in motor drive systems. When the fault occurs, the current waveform running in the motor is no longer sinusoidal which is observed in the healthy motor. The variation of the waveform in the faulty case depends on the location and degree of short circuit fault in the winding. In this paper, a Hidden Markov Model (HMM), which is widely used in the field of speech recognition, is exploited to capture and recognize the variation in the faulty current waveform. Thanks to the similarity between the speech signal and the current waveform, the HMM is highly expected to work as a robust fault diagnoser. Finally, the usefulness of the proposed diagnosis method is verified through some experiments using real faulty current waveforms.