Stator Phase Current STFT analysis for the PMSM Stator Winding Fault Diagnosis

Current trends in industry indicate that the condition monitoring and fault diagnosis of electric drive systems is becoming increasingly important. Nowadays, highly efficient permanent magnet synchronous motors are among the most often used in these systems. Therefore, the detection of their faults is also becoming a very crucial issue. The most common fault of this type of motor is the stator winding fault. This paper presents the effectiveness of the STFT analysis application of stator phase current signal for the extraction of symptoms from stator inter-turn short-circuits of permanent magnet synchronous motors. The time-based features obtained from the classical FFT analysis are not enough as they do not provide sufficiently unique information of the time of failure and do not allow for the further investigation of its source. The superiority of the proposed detection algorithm over the classical method based on the motor current analysis using FFT is described. The proposed solution is experimentally verified under various motor operating conditions. Additionally, a concept of a diagnostic system for on-line stator winding condition monitoring is proposed.