Induction machine stator winding failure detection using motor current signature analysis

Abstract

In addition to the domestic applications, developments in power converter topologies and governments focus on developing renewable energy sources, Induction machines use increases gradually. Earlier wind farms used Squirrel cage induction generators and now the doubly fed Induction generators replaced the SQIMs because of the flexibilities and controllable opportunities they have. But due to rapid fluctuations in speed and power flows stator windings prone to high electrical stresses and leads to outages. In this article wavelet based current signature analysis method adopted to investigate the induction machines fault frequencies. Discrete wavelet transform, stationary wavelet transforms and wavelet packet decompositions are used for the said purpose. The wavelet decomposition is used to pull out the information on a signal over a ample range of frequencies. Analysis is carried out in both time and s domains. The stator current analysis is done using the daubechies wavelets. Stator windings inter turn faults considered and simulated in simulink.