Moment of Inertia and Load Torque Identification Based on Adaptive Extended Kalman Filter for Interior Permanent Magnet Synchronous Motors

Abstract

The moment of inertia (MI) is an essential parameter in the speed loop controller and is unknown. Manual trial-and-error tuning of the speed loop controller is unattractive due to its haphazard, lengthy, and non-optimal. And the load torque plays a vital role in improving the dynamic performance. To attack these problems, this article proposes an adaptive extended Kalman filter (AEKF) to identify the MI and load torque of the interior permanent magnet synchronous motor (IPMSM). The real-time MI identified by the proposed AEKF method is used as the input of the speed loop self-tuning PI controller to solve the impact of the change of MI on the speed loop PI controller. Additionally, the load torque identified by the proposed AEKF method is used as the torque feed-forward compensation, thereby improving the torque-boosting capability of the system and further improving the dynamic performance of the IPMSM drive system. Compared with the extended Kalman filter (EKF), the AEKF identification method shows better performance, and the effectiveness of the algorithm is validated by simulations and experiments.