A Model-Free Adaptive Iterative Learning Control for PMLSM

In the application of model-based iterative learning control to a repetitive permanent magnet linear synchronous motor, it faces the challenges of unmodeled dynamics and oscillation. To address the two challenges, this paper proposes a feedforward-feedback based model-free adaptive iterative learning control approach that only requires the data of current and motor speed. In order to utilize the repetitive characteristics of the motor, the feedforward part is conducted with a model-free adaptive iterative learning controller, learning the tracking error in the iteration domain to improve the tracking accuracy at each time instant. The feedback part is considered with a PD controller for improving the oscillation phenomenon in tracking the motor speed. Simulation results show that the proposed approach reduces the tracking error by 90% at the 200th iteration compared to the first iteration.