UKF -based PMSM Magnetic Chain Tracking and Monitoring Study

A permanent magnet chain tracking method is proposed for the demagnetization fault of permanent magnet synchronous motors to improve the reliability of motor operation and prevent further deterioration of the fault. The magnetic chain parameter is one of the important indexes to measure its operating performance. Accurate tracking of the permanent magnet magnetic chain enables precise control and fault monitoring of the motor. In order to achieve a better tracking effect under different operating conditions, this paper proposes the magnetic chain tracking of permanent magnet synchronous motors based on Unscented Kalman filtering, which solves the fluctuation error generated during the tracking process of the magnetic chain of the PMSM by other previous methods. By varying the load and causing partial demagnetization of the PMSM by treating torque as an external perturbation, the error fluctuations in the tracking of the magnetic chain are kept to a minimum by using the Unscented Kalman Filter (UKF) algorithm in compared to the Extended Kalman Filter (EKF) algorithm. The simulation is verified in the Matlab/Simulink environment and the results confirm that the Unscented Kalman filter is more accurate in the online monitoring of permanent magnet chains.