Fault diagnosis of a permanent magnet synchronous generator connected to a uncontrolled rectifier bridge

Abstract

This paper presents a simple and reliable method for detecting the faults occurring in a permanent magnet synchronous generator (PMSG) connected with a three-phase uncontrolled rectifier bridge (TPURB), which is based on monitoring the output DC current. A PMSG connected with a TPURB at constant speed is discussed under two types of faults, namely phase winding faults and diode faults. An adaptive neuro-fuzzy inference system (ANFIS) is utilized to make decisions about the fault type. In order to demonstrate the effectiveness of the proposed method, finite element (FE) method is applied to generate virtual data applied to train the ANFIS. The simulation results show that the proposed method can effectively diagnose these faults.