Online Data-Driven Diagnosis for Common Electrical and Sensor Faults in Dual Three-Phase PMSM Drives
Lei Kong, Yao Mao, Ting Zhang, Xing‐Long Chen, Zheng Wang, Xueqing Wang
Abstract
This article proposes a data-driven online diagnosis method for detecting common electrical faults and current sensor faults (CSFs) in dual three-phase permanent-magnet synchronous motor (DTP-PMSM) drives. The proposed diagnosis method fulfills feature extraction by obtaining the average values of phase currents in the positive and negative half-cycles of fault features to reduce dimensionality. By designing a one-vote veto classification rule, the proposed diagnosis method can accurately distinguish 24 fault modes and effectively prevent misdiagnosis. The classification model consists of only 24 decision functions, making it easy to implement on the mainstream industrial microcontrollers. The experiments are carried out to verify the validity of the proposed method.