Diagnosis and Identification of Common Electrical Faults in PM Machine Drives

With the widespread integration of machine drives in aerospace and electric vehicles, high reliability has become a paramount concern. Fault diagnosis and tolerant control play crucial roles in enhancing the dependability of machine drives. However, prevailing research on fault detection typically concentrates on a single fault. In fact, various faults might share similar fault characteristics. This limitation could result in misdiagnosis and incorrect fault-tolerant of machine drives. To address this issue, this article delves into the analysis of several common electrical faults and proposes a detection method based on high frequency (HF) and low frequency (LF) signals for identifying stator winding inter-turn fault, stator winding high resistance fault, current sensor bias fault, current sensor gain fault which may exhibit similar fault characteristics. By scrutinizing the HF and LF voltage residuals, four fault indicators at different frequencies were designed, enabling reliable fault diagnosis and differentiation. The effectiveness of the proposed method is validated through extensive experiments on a permanent magnet machine.