Rotor speed signature analysis-based inter-turn short circuit fault detection for permanent magnet synchronous machines

Abstract

A rotor speed signature analysis- (RSSA-)based inter-turn short circuit (ITSC) fault diagnostic method is proposed, which is robust with regard to speed- and current-loop bandwidths of permanent magnet synchronous machine (PMSM) drive systems. Conventional ITSC fault detection solutions that rely on current signals or extra devices. Thus, an initial step of investigation on the validity of RSSA is taken for residual insulation capacity monitoring of the ITSC fault at the incipient stage. The Vold–Kalman filtering order tracking method is employed for the real-time extraction of fault features. Besides, the impact of speed- and current-loop controller bandwidths on ITSC fault diagnosis is also analysed and an Adaline estimator is designed to decouple their impacts. Finally, the effectiveness of the proposed method is verified on a faulty PMSM, which exhibits satisfying results in tracking the degradation of residual insulation, even if the phase currents are significantly distorted due to low switching frequency of the inverter.