Turn-to-turn fault diagnosis of an Induction motor by the analysis of Transient and Steady state Stator current

Abstract

This work proposes an online diagnosis of turn-to-turn stator winding fault of an induction motor through the combine use of Wavelet Transform (WT) and Fast Fourier Transform (FFT). Both steady state and non-stationary transient part of motor stator currents are assessed for detection of this inter turn short circuit fault. First non-stationary part is assessed by formation of contour of the coefficients of Continuous Wavelet Transform (CWT) of stator current – result shows significant change of amplitude at certain frequencies. Secondly, the supply frequency is filtered off from the steady part of motor current. Then on this filtered signal (i) Fast Fourier Transform (FFT) is performed where from spectrums are observed at different percentage of inter turn short condition and (ii) performing Discrete Wavelet Transform (DWT) detailed wave energy has been calculated using Parseval’s theorem. Work has been performed both under load and no-load condition of the motor. The proposed method has been validated in a laboratory prototype. Results indicate that the proposed technique is suitable for real-time application

Keywords: CWT, DWT, FFT, fault diagnosis, induction motor, inter-turn short, wave energy