Abstract

Design an optimized iterative learning control for linear and nonlinear dynamical systems is a challenging task. Norm-optimal iterative learning control (NOILC) is a valuable criterion for these dynamical systems. An iterative learning control algorithm based on optimal control theory is proposed, and the stability and convergence conditions of the proposed control algorithm are analyzed by using the convergence conditions of iterative learning control, and the control design is carried out based on feedforward and feedback control structure. At the same time, by introducing a weighted matrix coefficient to the feedforward control action, the convergence speed of iterative learning control algorithm based on optimal control theory is improved, and it is applied to the Matlab simulation control system. The results show that the convergence effect of the basic optimal control theory and the iterative learning control algorithm based on the weighted matrix coefficient is significant and the performance of the trajectory tracking is improved. The numerical example simulated on MATLAB@2019 and mollified results confirm the validation of the designed algorithm.

摘要

为线性和非线性动力系统设计优化的迭代学习控制是一项艰巨的任务。规范最优迭代学习控制（NOILC）是这些动力学系统的重要标准。提出了一种基于最优控制理论的迭代学习控制算法，并利用迭代学习控制的收敛条件分析了该控制算法的稳定性和收敛条件，并基于前馈和反馈控制结构进行了控制设计。 。同时，通过在前馈控制动作中引入加权矩阵系数，提高了基于最优控制理论的迭代学习控制算法的收敛速度，并将其应用于Matlab仿真控制系统。结果表明，基本最优控制理论和基于加权矩阵系数的迭代学习控制算法的收敛效果显着，改善了轨迹跟踪的性能。在MATLAB @ 2019上模拟的数值示例和改进的结果证实了所设计算法的有效性。