Detection and Differentiation of Winding Inter-Turn Fault and High-Resistance Fault in PM Machines

Rapid and reliable fault detection and differentiation are crucial for enhancing the performance of permanent magnet (PM) machines in high-reliability applications, such as aerospace and rail transportation. Current research often focuses on single fault detection but tends to overlook the fact that different faults may exhibit similar characteristics. To improve the detection reliability of PM machines, this paper proposes a method for detecting and differentiating between inter-turn fault (ITF) and high-resistance fault (HRF) in stator windings, based on the 2nd order positive and negative sequence Park transformation. The method reveals that ITF appears as DC components in both the 2nd order positive and negative sequence planes, while HRF presents DC components only in the 2nd order negative sequence plane. By leveraging the distinct features of voltage residuals in these different planes, the proposed approach effectively detects and differentiates these faults. Compared to existing methods, this approach does not require high-frequency signal injection or additional hardware, and it provides excellent fault differentiation without adversely affecting the machine's operation. Extensive experiments on PM machines validate the effectiveness and reliability of the proposed method.