Thrust Ripple Reduction of Permanent Magnet Linear Motor for Servo System Based on Sliding Mode Iterative Learning Control With Feedforward Compensation Strategy

Thrust ripples caused by the end-effect and cogging-effect have a serious impact on the control of Permanent Magnet Linear Motor (PMLM). In this paper, thrust ripple of PMLM is analyzed and reduced for servo system. First, thrust ripple is analyzed by 2-D finite element method and then is identified as the superposition of multiple sinusoidal functions. Second, for a linear servo system driven by PMLM, a feedforward compensation strategy based on sliding mode iterative learning control algorithm is applied to reduce tracking error. Disturbance observer and sinusoidal parameter direct feedforward are compared as two different feedforward methods. A speed loop control method based on sliding mode controller is used to further reduce the influence of thrust ripple. Finally, the simulation results confirm that the control strategy can effectively reduce the thrust ripple in a linear servo system.