论文部分内容阅读
在直驱XY平台系统的交叉耦合迭代学习控制(crosscoupled iterative learning control,CCILC)过程中,由于每次运行时轮廓误差的累积,会导致系统出现收敛速度降低甚至发散的现象。针对这一问题,提出一种与经验模态分解(empirical mode decomposition,EMD)算法相结合的CCILC控制方法。首先设计CCILC控制器,直接降低轮廓误差。然后,利用EMD算法分解CCILC过程中的轮廓误差,筛选并剔除其中的发散分量,提高收敛速度和轮廓精度。仿真和实验结果表明,与传统CCILC相比,所提出的控制方法能够使直驱XY平台系统的轮廓跟踪效果更好,并且使输出轨迹在较少的迭代次数下快速且精确地收敛到期望轨迹。
In the cross-coupled iterative learning control (CCILC) process of the direct-drive XY platform system, the accumulation of contour errors in each run will result in the convergence or divergence of the system. In response to this problem, a CCILC control method combined with empirical mode decomposition (EMD) algorithm is proposed. First design CCILC controller, directly reduce the contour error. Then, the EMD algorithm is used to decompose the contour errors in the CCILC process to filter out and remove the divergence components therein to improve the convergence speed and the contour accuracy. Simulation and experimental results show that compared with the traditional CCILC, the proposed control method can make the profile tracking of the direct-drive XY platform system better and make the output trajectory converge to the desired trajectory quickly and accurately with fewer iterations .