Motor current signatures and their envelopes as tools for fault diagnosis

Abstract

AbstractInduction motor inter-turn short circuit stator fault is identified using motor current signals with the help of Discrete Wavelet Transform along with Power Spectral Density. This methodology is adopted for motor current signals and their envelopes in order to identify a better tool for fault diagnosis. First, the analysis is performed using the original stator current signal, which has the fundamental frequency components along with the slip frequency and other harmonic/fault components. The fundamental component in the stator current can be eliminated by using the envelope of the signal. The transient in the wavelet decomposition can be suppressed by modifying the current envelope. This is done by pre-multiplying the original envelope with the Tukey window. The Power Spectral Density is calculated for both the signatures. Comparison of these two methods show that the proposed analysis based on modified envelope using Discrete Wavelet Transform along with Power Spectral Density offers better resu...