基于改进遗传算法的PID参数优化研究  

Research of PID Parameter Optimization Based on Improved Genetic Algorithm

作　　者：王非[1] 冯冬青[1] 马雁[1] 

机构地区：[1]郑州大学电气工程学院,450001

出　　处：《微计算机信息》2008年第22期36-38,共3页Control & Automation

基　　金：河南省自然科学基金资助项目(0511010800)

摘　　要：针对基本遗传算法收敛速度慢和寻优能力不足的问题,提出了一种基于实数编码的改进遗传算法。新算法中,初始种群由空间距离控制使其能够均匀分布于解空间;交叉操作采用等分组的方法,对每组内每两个个体均进行交叉,并择优选择,以扩大搜索空间;变异步长随进化代数自适应调整。将改进后的遗传算法运用于PID控制器参数优化中,通过仿真实验表明,新算法整定效果明显优于基本遗传算法,不仅解决了基本遗传算法存在的缺陷,而且提高了收敛速度与寻优精度。An improved genetic algorithm based on real coding is proposed to overcome slow convergent speed, poor seeking optimiza- tion capabilities of the standard genetic algorithm. In the new algorithm, initial population controlled by space distance can be distributed in the solution space uniformly. The crossover operation use the method of subgroups, which cross each two individuals in every group, and then choice the best individuals. The mutation step can adjust adaptively with the evolution. The improved genetic algorithm is applied to PID controller parameter optimization. The simulation indicates that the new algorithm not only resolve the deficiency of the standard genetic algorithm, but also improve convergence performance and optimization precision.

关 键 词：改进遗传算法 等分组 变异步长 PID控制器 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]