低频线振动台系统的鲁棒自适应重复学习控制

Robust adaptive repetitive learning control for low-frequency linear vibration table system

由于没有传动机构,永磁直线同步电机(PMLSM)作为低频线振动台的驱动部件对扰动和参数不确定性很敏感,摩擦力及纹波推力扰动等非线性因素严重影响了PMLSM的运动精确度。针对上述问题,提出一种鲁棒自适应重复学习控制方法,用于提高低频线振动台系统的精度。所设计的控制律由参数自适应控制、积分滑模控制、重复学习控制组成。参数自适应控制用来估计未知的模型参数并予以补偿;积分滑模控制用来镇定低频线振动台系统,抑制非周期扰动;重复学习控制用来抑制周期性扰动,提高对周期性位置信号的跟踪性能。采用Lyapunov理论设计的鲁棒自适应重复学习控制律能够保证闭环系统的渐近稳定性和位置跟踪性能。仿真结果表明,鲁棒自适应重复学习控制方法明显提高了系统的跟踪性能,改善了加速度失真度。

Due to the elimination of mechanical transmission mechanisms, permanent-magnet linear synchronous motors （PMLSM） as the driving part for low-frequency linear vibration table were more sensitive to disturbances and parameters uncertainties. Furthermore, significant nonlinear effects （e. g. friction and force ripple） worsen their motion precision. Aiming at these problems, a robust adaptive repetitive learning control scheme was presented to improve the precision of low-frequency linear vibration table. The control algorithm consisted of an adaptive control, an integral sliding mode control and a repetitive learning control. The adaptive control was used to estimate and compensate unknown model parameters ; The integral sliding mode control could stabilize the low-frequency linear vibration table system and suppress aperiodic disturbance; The repetitive learning control was used to restrain the periodic disturbance and improve the tracking performance of periodic position signals. The robust adaptive repetitive learning control law designed by using Lyapunov theory guaranteed the system stability and the position tracking pefformance. The simulation results demonstrate the robust adaptive repetitive learning control scheme can provide the better tracking performance and improve the accelaration distortion for low-frequency linear vibration table.