Bearing fault detection in a 3 phase induction motor using stator current frequency spectral subtraction with various wavelet decomposition techniques

Abstract

Induction motors consumes 90% of total power consumed by industries due to large scale utilisation. Even though these motors are rugged in structure, they often face unexpected failure due to long usage without maintenance. Bearing failure is a major problem among various faults, which cause catastrophic damage to machine when unnoticed at incipient stage. So the bearing faults in induction machines should be continuously monitored. Motor current signature analysis (MCSA) has become popular for detection and localisation of these faults and has attracted concentration of many researchers. In this paper stator current is monitored by means of frequency spectral subtraction using various wavelet transforms to suppress dominant components. The spectral subtraction using discrete wavelet transform (DWT), stationary wavelet transform (SWT) and wavelet packet decomposition (WPD) is performed and a comparative analysis is carried out by means of different fault indexing parameters. The proposed topology is examined using 2.2 kW induction machine test bed. Keywords: Condition monitoring, Motor current signature analysis, Bearing faults, Discrete wavelet transform, Stationary wavelet transform, Wavelet packet decomposition