Applications of unscented Kalman filter on position sensorless control of permanent magnet linear syncheonous motor

Abstract

For a variety of nonlinear estimation problems, the unscented Kalman filter (UKF) is a superior alternative to the extended Kalman filter (EKF). The effectiveness of UKF for position sensorless control of permanent magnet linear syn- chronous motor (PMLSM) is investigated. At first, the PMLSM model used in the estimation algorithm is established, then, the position sensorless PMLSM drive system based on UKF is presented. Via simulations and experiments, the effectiveness of UKF for position sensorless control of PMLSM is verified. And, a comparison of estimation performance of UKF and EKF for estimating position and speed is carried out. The simulation and experiment results indicate that UKF performs equivalently with EKF, but the computational time of UKF is more than that of EKF. For estimating position and speed of PMLSM, the EKF is a better choice than UKF.