基于差分进化算法的PID参数整定  

Parameter Tuning of PID Controller Based on Differential Evolution

作　　者：姜立强[1] 刘光斌[1] 郭铮[1] 

机构地区：[1]第二炮兵工程学院,陕西西安710025

出　　处：《计算机仿真》2009年第6期204-206,共3页Computer Simulation

摘　　要：研究了PID控制器的参数整定问题。常用的工程整定法没有考虑系统的任何要求,只能提供给系统一个稳定的状态,无法使系统运行最佳。理论设计法只能保证满足系统的某一特性要求。针对这两种方法存在的缺陷,将差分进化算法应用于PID参数整定,提出了基于差分进化算法的PID参数整定方法。新方法根据系统性能指标,设计适应度函数,充分利用差分进化算法优秀的寻优能力,寻找出较优的PID参数。最后,通过实例对三种参数整定方法的控制效果进行了仿真和比较,仿真结果证明了基于差分进化算法的PID参数整定方法明显优于工程整定法和理论设计法。PID parameter tuning is studied in this paper. Engineering tuning method can only provide a stable state for the system and can not make the system run optimally, because it doesn＇t take the system requirements into consideration. Theoretical design method can only meet a certain characteristic requirement. In order to overcome the disadvantages of the two methods, this paper puts Differential Evolution （DE） into use for PID parameter tuning and proposes a new method based on DE. The fitness function of DE is designed according to the system requirements. For DE＇s powerful global search ability, the new method can obtain good PID parameters. At last, simulations are carried out using the three methods and the results show that the new method outperforms both the engineering tuning method and theoretical design method apparently.

关 键 词：差分进化算法 比例积分微分 参数整定 仿真 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]