基于多目标粒子群算法的PID控制器设计

Design of PID Controller Based on Multi-Objective Particle Swarm Optimization Algorithm

随着对控制系统的要求越来越高,进行比例积分微分(Proportion Integration Differentiation,PID)控制器的设计的时候应该同时考虑到系统时域指标和频域指标,常规的PID整定方法往往很难实现。为解决上述问题,采用多目标粒子群算法进行PID控制器参数的设计,算法将系统的超调量、上升时间和稳定时间作为目标函数,频域指标作为约束条件。算法的运算结果为一组Pareto最优解,运行者可以根据当前对系统的要求从中选取合适的解。通过与常规PID整定方法和采用单目标粒子群算法的方法进行比较,证明了改进方法的有效性。

With the increasing high demands of control systems,both time-domain indices and frequency-domain indices should be taken into account when tuning the parameters of PID controller.This is hard to achieve by using conventional PID tuning methods.In this paper,a multi - objective particle swarm optimization algorithm was employed to optimize the parameters of PID controller.The overshoot,rise time and settling time of the system were taken as objective functions.Meanwhile frequency-domain indices were used as constraints.A group of Pareto optimal solutions can be obtained and the operator can choose the most satisfactory solution according to the requirements of the system.Compared with conventional PID tuning methods and method using single target particle swarm optimization algorithm,the experiment results verify the feasibility of the method proposed in this paper.