Research of convergence based on optimal control iterative learning algorithm
                                                                               YANG Liang liang, HU Jian
                           (Faculty of Mechanical and Automatic Control, Zhejiang Sci Tech University, Hangzhou 310018, China)

Abstract: Aiming at the convergence problem of iterative learning control in linear discrete systems, the discrete mathematical model of the linear motor was established, and the iterative learning control algorithm was applied to the motor control system, whose stability and convergence were studied.An iterative learning control algorithm based on optimal control theory was proposed,the stability and convergence conditions of which were analyzed by the convergence condition of iterative learning control.The controller was designed based on a typical two degree of freedom controller structure with feedforward feedback control strategy.At the same time, a weighting matrix coefficient on feedforward control force was introduced to improve the convergence speed of iterative learning algorithm,which was applied to the Matlab simulation platform and the actual electromechanical control system. The results show that the iterative learning control algorithm based on optimal control theory and weighting matrix coefficient has a significant convergence effect and improves the trajectory tracking performance.

Key words: iterative learning control(ILC); optimal control; convergence; weighted matrix coefficient