An Adaptive Speed Estimation Method Based on a Reduce-Order Extended Kalman Filter for PMSM Sensorless Drives

A speed sensorless control system for Permanent Magnet Synchronous Machine (PMSM) by using an adaptive parallel reduce order Extended Kalman Filter is proposed in this paper. The noise matrix in the EKF algorithm is one of the important factors that determine its estimation performance, and the optimal solution of the noise matrix obtained by the traditional EKF noise matrix through trial-and-error method is often difficult to cope with the changing working conditions in reality. In addition, EKF requires complex calculations, so it has high requirements for chips, making it difficult to industrialize. Therefore, this paper proposes a novel EKF algorithm, which can not only reduce the computational complexity, but also update the noise matrix online by using the fading factor according to the actual working conditions. The simulation results verify the effectiveness of the algorithm.