基于迭代学习控制的PID控制器设计

Design of PID controller based on iterative learning control

针对传统的经验PID整定方法,提出了一种新的PID参数整定算法。该算法首先利用PD型迭代学习控制来进行期望轨迹的跟踪控制,然后根据迭代学习控制的输入输出数据序列,通过强跟踪滤波器来进行参数辨识,可获得对应于期望轨迹的优化的PID控制参数。给出了迭代学习控制的收敛条件,以及如何利用强跟踪滤波器来进行参数辨识。仿真和实验结果表明,采用该算法设计PID控制器,被控系统可以获得较佳的动态性能和较强的鲁棒性。

Contrary to the traditional PID tuning methods by experiences, a novel algorithm to tune PID parameters is proposed. By PD-type iterative learning control （ILC）, tracking the desired trajectory is achieved and the input-output data series of ILC are simultaneously acquired so that optimized PID parameters corresponding to the desired trajectory are identified by the proposed strong tracking filter （STF）. The condition for convergence of ILC and how to identify the parameters with STF are also presented. The simulation and experimental results both show that the controlled system with the proposed PID controller can achieve perfect dynamic performance and strong robustness.