Criterion Function for Broken-Bar Fault Diagnosis in Induction Motor under Load Variation Using Wavelet Transform

In this article, a novel criterion function is introduced to diagnose the breakage in rotor bars of induction motors. This criterion function facilitates the precise diagnosis of the fault in induction motors under load variation. It uses wavelet transform to process the stator current signal in the faulty induction motors to extract the wavelet coefficients in a specific frequency band. Furthermore, spectrum analysis of the stator current around the fundamental frequency current is used to diagnose the fault non-invasively. It is shown that the amplitudes of the frequency harmonics components (1 ± 2ks)f s are increased due to the load increases. A time-stepping finite element method is used for modeling the faults in induction motors. In this modeling, the effects of the spatial distribution of the stator winding, nonuniform air-gap permeance, geometrical and physical characteristics of different parts of the motor, and, finally, nonlinearity of the core materials are taken into account. The proposed algorithm is applied to the stator current of a healthy and a faulty induction motor. The simulation results are obtained, and their accuracy is verified by the experimental results.