Adaptive state compensation using parameterized iterative learning control for periodic velocity ripple of permanent magnet linear motor

This paper focuses on adaptive velocity compensation along state axis aiming at dent and friction force ripples using parameterized iterative learning control (ILC) for permanent magnet linear synchronous motor (PMLSM) executing same control instruction repeatedly. Velocity ripples are primarily attributed to periodic disturbances such as dent force and the gliding friction in PMLSM. The dent force consists of end effect and cogging effect forces. The key idea of this disturbance compensation method is to use past information for one trajectory period along state axis to update current adaptation law. The new method consists of three different steps: in first repetitive trajectory, an adaptive compensator is designed to guarantee the L₂-stability of overall system; from second repetitive trajectory onwards, a trajectory-periodic adaptive ILC compensator stabilizes the system; and to make use of the stored past state-periodic ripple force information, a search process is utilized for adapting current controller coefficient. As a consequence, the velocity pulsation is reduced significantly. The proposed approach is evaluated by experiment, which validates its effectiveness.