中文摘要：

针对永磁直线同步电机易受端部效应、负载扰动和参数不确定等因素影响的特点,提出一种使用学习滤波器的简约学习前馈补偿控制策略.为消除端部效应的影响,设计多个低维B样条网络进行学习前馈补偿,从而达到良好的补偿效果;为保证系统的稳定性,在学习前馈控制中增加学习滤波器.仿真结果表明,该策略对非周期性输入信号具有良好的跟踪特性,同时有效解决了学习前馈补偿控制中的＂维数灾难＂.

英文摘要：

For the characteristics of permanent-magnet linear synchronous motor（PMLSM） suffering from the effects of end effect,load disturbances and parameter variations and so on,the parsimonious learning feed-forward control（LFFC） with learning filter was designed.The learning feed-forward compensation based on several lower-dimensional B-spline networks was adopted to overcome the influence of the end-effect,and a learning filter was added to the learning feed-forward controller to ensure stability of the system.Simulation results show that the proposed strategy enables servo system to have accurate tracking performance for non-periodic input signal,while solves effectively the ＂curse of dimensionality＂ in learning feed-forward control.