Short-Time Adaline Based Fault Feature Extraction for Inter-Turn Short Circuit Diagnosis of PMSM via Residual Insulation Monitoring

Abstract

Inter-turn short circuit (ITSC) is a common fault in electrical machines, which may result in further devastation to the whole winding and magnets. In this article, a method for real-time residual insulation capacity monitoring of ITSC fault in permanent magnet synchronous machines (PMSMs) is proposed, which is robust to speed/torque variation and independent of machine parameters and extra sensors. It is based on the theory that the ITSC fault initiates as an insulation deterioration among adjacent turns and the fault severity can be described by the residual insulation capacity being relevant to the 2nd harmonic component in the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">dq</i> -axis currents. Thus, a short-time Adaline-based harmonic extraction method is proposed for the separation of the needed 2nd harmonic component from time-varying current signals, being used as the fault indicator afterward. Finally, the performance of the proposed method is experimentally evaluated on a PMSM and shows quite good performance in fast-tracking and detection of faults in various stages.