永磁直线同步电动机的自适应学习控制

由于没有传动机构,使永磁直线交流同步电机（PMLSM）控制器设计较为复杂.PMLSM对模型不确定性和外扰更加敏感;推力波动等非线性因素对运动精度影响很大.针对上述问题,用自适应学习方法改善PMLSM的轨迹跟踪性能,并对迭代模式和单次运行模式下算法的收敛性进行了证明,通过实验进行了算法验证.该控制方法基于迭代学习,控制器分为两个部分,通过执行重复任务自适应学习项补偿系统的非线性;另一项用于增强系统的鲁棒性,保证系统在单次运动模式下稳定.实验结果表明,这种控制方法可以有效提高PMLSM轨迹跟踪精度.

ADAPTIVE-LEARNING CONTROL FOR PERMANENT-MAGNET LINEAR SYNCHRONOUS MOTORS

There is more complexity in controller design for Permanent-magnet linear synchronous motors (PMLSM) due to the elimination of mechanical transmission mechanisms.Firstly,PMLSM are more sensitive to disturbances and parameters uncertainties.Furthermore,significant nonlinear effects (e.g.thrust ripple) worsen its motion precision.In order to resolve above-mentioned issues,an adaptive-learning control scheme was applied to improve trajectory tracking performance of PMLSM.Convergence analysis of the algorithm was given in both repetitive operational mode and single operational mode.The control scheme was verified in experiments.The scheme is based on iterative learning control.The controller consists of two terms,the adaptive-learning term compensates nonlinear effects of the system in the repetitive tasks,and another term enhances robustness to guarantee stability in the single operational mode.The experimental results show that tracking precision of PMLSM can be effectively improved by the scheme.