Affiliation:
1. Automotive Structure and Energy Storage Engineering Center, School of Mechanical Engineering, Shanghai Dianji University, China
2. Shanghai Aowei Technology Development Co., Ltd, China
Abstract
The control of nonlinear system of permanent magnet synchronous motor (PMSM) can be realized using extended Kalman filter (EKF). However, the traditional EKF algorithm has limited tracking accuracy and is prone to accumulated error. In addition, the existing improved EKF algorithm pursues the adaptability to nonlinear system, which is susceptible to noise interference and overfitting. For this reason, this paper proposes an observation error threshold EKF (OET-EKF) algorithm from the perspective of error reduction. On the basis of the traditional EKF, a threshold is set for the error between the estimated observation and the actual observation of the rotational speed. In addition, the traditional EKF algorithm is corrected by the threshold, which improves the robustness. In this paper, MATLAB/Simulink simulation is used to verify the robustness of the OET-EKF algorithm. In addition, it is found that the anti-interference ability of the system under a certain value of white noise is better than the traditional EKF algorithm.