High-frequency current injection-based online fault diagnosis for power converter in SRM drives

Abstract

Exceptional robust configuration combined with fault tolerance and rare-earth free features make switched reluctance motors (SRMs) a competitive technology for EV/HEV applications, where system reliability is critically important. In this paper, a novel real time fault diagnosis algorithm for power converters in switched reluctance motor drives is proposed. In contrast to the existing diagnostic schemes that use additional sensors, the proposed method extracts the fault signatures from the fundamental current by injecting a sinusoidal high-frequency (HF) current signal into the upper switches of the asymmetric converter. The injected HF current is then filtered and its frequency is calculated for fault diagnosis. In addition, another PWM pulse is injected into the lower switch to locate the faulty switch when an open-circuit fault is detected. Open-circuit and short-circuit faults are analyzed based on the frequency and amplitude of the filtered HF current along with the variation in amplitude of the fundamental current with the occurrence of faults. The effectiveness of the proposed strategy is verified though thorough simulation results based on a three-phase 12/8 SRM drive system.