Modeling and fault diagnosis of multi-phase winding inter-turn short circuit for five-phase PMSM based on improved trust region

Abstract

Multi-phase Permanent Magnet Synchronous Motors (PMSMs) are widely used in fault-tolerant control applications. As an electrical fault with a high probability, inter-turn short circuit (ITSC) fault seriously affects the reliability of five-phase PMSM. However, due to high coupling of fault parameters, most of the fault models and diagnosis methods in fault-tolerant control focus on one-phase winding ITSC fault of three-phase PMSM. In this paper, a novel model and a fault diagnosis method are proposed for multi-phase winding ITSC (MP-ITSC) of five-phase PMSM. First, by analysing the related physical parameters of the motor, a mathematic model of MP-ITSC fault is established. Then, a parameter estimation method based on improved trust region (ITR), which accelerates the convergence speed of trust region algorithm and reformulates the fault diagnosis as the extreme seeking for fault parameters is proposed to detect the ITSC fault level. Experimental results provide verification of the proposed model and fault diagnosis method.