Fault diagnosis of three-phase PWM inverters using wavelet and SVM

In this paper, a fault diagnosis scheme of three-phase PWM inverters is proposed, which is based on a discrete wavelet transform (DWT) and a pattern recognition algorithm called support vector machine (SVM). First, a discontinuity of current signal, which is usually caused by switch open-circuit faults, is detected by the DWT, and then fault detection and isolation are processed by the SVM. The proposed method is verified by simulation results.