Broken Rotor Bar Fault Diagnosis for Induction Motors Using Power Spectral Density and Complex Continuous Wavelet Transform Methods

Abstract

Induction motors are widely used in various industrial sectors, fault diagnosis of induction motors are critical to prevent equipment failure and production downtime. In this paper, a stator current signature analysis method is proposed for squirrel cage induction motors’ broken rotor bar (BRB) fault diagnosis. Two different techniques are implemented: Power Spectral Density (PSD) based stator currents’ amplitude spectrum analysis; and one dimensional Complex Continuous Wavelet Transform (CWT) based stator currents’ time-scale spectrum analysis using Complex Morlet Wavelet (CMW). The performance of the two techniques are compared using experimental stator current data measured in a lab for a 0.25 HP induction motor. The stator current under healthy and faulty states of the motor were measured, the faults include one, two and three BRBs. For 2 and 3 BRB faults, the holes were drilled on the rotor bars 90 degree apart. Two loading conditions of the motor were used during the measurement, 30% and 85%. It is found that the CWT has better performance than the PSD estimates for the BRB fault detection.