Application of Finite Element Magnetic Flux Leakage Simulation in Fault Detection of Brushless DC Motor Inter-Turn Short-Circuit

Brushless DC motor (BLDCM) has the advantages of small size, high efficiency and good dynamic characteristics compared with other kinds of motors. There will be a series of faults when motor works for a long time. Inter-turn short-circuit is a common motor fault type. The online fault detection of motor is always a challenging task. In this paper, a new method based on external magnetic flux leakage and machine learning is proposed to diagnose the inter-turn short-circuit fault of BLDCM. The ANSOFT@Maxwell-2D is used to model the BLDCM. According to the motor health state and the fault state of short circuit between phases A, B and C of the motor, the finite element method is used to analyze the magnetic flux leakage of the motor, and the magnetic flux leakage signals of various states are obtained. The neural network model is used to process the magnetic flux leakage signals and determine the fault phase in order to achieve the purpose of detection and diagnosis. Compared with traditional MCSA, MFL detection method has the advantages of robustness and non-invasion.