Generalized Likelihood Ratio Test Based Approach for Stator-Fault Detection in a PWM Inverter-Fed Induction Motor Drive

This paper proposes a new approach for the detection of stator faults in inverter-fed induction motor under closed-loop control. These faults, generally, start with interturn short circuit and can evolve to phase-to-phase or phase-to-ground faults, leading to stator currents unbalance. In the considered application, the stator windings are supplied by a static converter for the control of the speed and the rotor flux of the motor. The unbalance fault has been artificially created through additional resistance in series with one phase of the stator windings. The proposed diagnosis approach is based on the computation of the symmetrical components of the stator currents. The supply fundamental frequency and the three-phase phasors are estimated based on the maximum likelihood estimator. Then, the generalized likelihood ratio test is applied for unbalance fault detection. Simulation and experimental results on a 1.5-kW induction motor illustrate the effectiveness of the proposed approach, leading to an effective diagnosis procedure for stator faults in inverter-fed induction motor under steady state and closed-loop operation.