精英学习人工蜂群算法的PID控制器参数优化  

An Elite Learning Artificial bee Colony Algorithm for Parameter Optimization of PID Controller

作　　者：王贞 李旭飞 WANG Zhen;LI Xu-fei(Institute of Mathematics and Information Science,North Minzu University,Yinchuan 750021,China)

机构地区：[1]北方民族大学数学与信息科学学院,宁夏银川7500211

出　　处：《数学的实践与认识》2020年第16期177-186,共10页Mathematics in Practice and Theory

基　　金：宁夏高等学校科学研究项目(NGY2017168)。

摘　　要：为有效求解PID控制器参数优化问题,本文提出一种精英学习人工蜂群算法.针对人工蜂群算法开发能力较弱的问题,设计一种精英学习策略,通过环形拓扑结构和遗传算法的交叉、变异、选择操作产生精英个体.进而,利用精英个体和全局最优个体构造具有精英引导和全局引导的搜索方程,平衡算法的探索能力和开发能力.通过对22个标准测试问题和PID控制器参数优化问题实验可知,精英学习人工蜂群算法具有较好的寻有能力.For solving parameter optimization of PID controller effectively,an elite learning artificial bee colony algorithm is proposed.Since artificial bee colony algorithm does poorly in exploitation,a elite learning strategy is designed,which generating the elite individuals by ring topology and crossover,mutation,selection in GA.Furthermore,the exemplars and global best individual are added into search equation to balance the the exploitation ability and exploration ability.By experimenting on 22 benchmark test problems and parameter optimization of PID controller,the elite learning artificial bee colony algorithm proposed in this paper shows the better performance.

关 键 词：人工蜂群算法 精英学习策略 PID控制器 环形拓扑 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TP273[自动化与计算机技术—控制科学与工程]