Permanent Magnet Synchronous Motor Parameter Identification Based on Improved Adaptive Extended Kalman Filter

Abstract

The electromagnetic parameters of the permanent magnet synchronous motor (PMSM) will change influenced by certain factors like temperature and magnetic circuit saturation, which will reduce the motor control system performance, and even cause permanent damage to the motor. Therefore, accurate acquisition of motor parameters remarkably affects the high-performance operation of the motor. The paper adopts an improved adaptive extended Kalman filter (IAEKF) algorithm for identifying the parameters of the nonlinear permanent magnet synchronous motor (PMSM) system. This method performs Taylor expansion on the high-order nonlinear function of the system model, thereby converting the nonlinear problem into a linear problem for solving. Moreover, adaptive technology is adopted to the method. The process noise covariance is estimated in real time through an improved noise statistic estimator (NSE). The improved NSE is composed of a biased and an unbiased estimator, which can improve the accuracy of the noise parameter estimation, while ensuring the positive semi-definiteness of the process noise variance matrix to ensure the robustness of the algorithm. Finally, simulation analysis helps to verify the whether the algorithm is feasibly.