Adaptive Fourier ILC for Mover Position Estimation Error Suppression for Sensorless PMLSM Drives

Permanent magnet linear synchronous motor (PMLSM) suffers from serious thrust ripples which severely affect the estimation accuracy and present challenges for sensorless control. This article proposes an adaptive Fourier iterative learning control (AFILC) to suppress the mover position estimation error of high-frequency signal injection (HFSI)-based sensorless PMLSM drives. First, the influence of the end effect on sensorless PMLSM drives is analyzed by establishing a dynamic inductance matrix. According to the characteristic of mover position estimation error, a proportional differential (PD)-type iterative learning control (ILC) is constructed to suppress the errors. On this basis, ILC combined with Fourier expansion extraction is designed to weaken the accumulation of aperiodic errors. The ILC control gain self-tuning method is investigated to improve the error suppression ability in different cases. The proposed method has a stronger suppression effect on estimation errors and can improve the operation performance of sensorless PMLSM drives. Finally, the feasibility and effectiveness of the proposed method are verified on a 750-W PMLSM experimental platform.