混合遗传算法思想的改进鸡群算法  

Improved Chicken Colony Algorithm Based on Hybrid Genetic Algorithm

作　　者：崔心惠 张祝威 李文萱[1] CUI Xinhui;ZHANG Zhuwei;LI Wenxuan(College of Electrical Engineering,Chuzhou Polytechnic,Chuzhou,Anhui 239000,China;Nanjing Dianyan Electric Power Automation Co.,Ltd,Nanjing,Jiangsu 210000,China)

机构地区：[1]滁州职业技术学院电气工程学院,安徽滁州239000 [2]南京电研电力自动化股份有限公司,江苏南京210000

出　　处：《宜宾学院学报》2021年第12期25-30,共6页Journal of Yibin University

基　　金：安徽省科技攻关重大项目(1301041023);2020年度安徽省高校优秀青年人才支持计划重点项目(gxyqZD2020068)。

摘　　要：为了解决标准的鸡群算法在寻优中存在早熟收敛和较容易陷入局部极值问题,提出一种混合遗传算法思想的改进鸡群算法(GA-CSO).算法在母鸡位置中引入惯性权值因子,并将小鸡位置中加入公鸡对其影响的学习因子,最后利用遗传思想选取适应度值优良的个体以一定概率进行交叉和高斯变异操作.利用4个典型的测试函数分别进行仿真并和其他算法对比,仿真结果表明,改进的GA-CSO算法可避免局部最优,且加强了全局的极值搜索能力,较大地改善算法收敛速度和精度范围.In order to solve the problem of premature convergence and being easy to fall into local extremum in the standard chicken swarm algorithm, an improved chicken swarm algorithm based on hybrid genetic algorithm(GA-CSO) was proposed. In this algorithm, the inertia weight factor was introduced into the position of hen, and the learning factor of rooster’s influence was added into the position of chicken. Finally, the individuals with good fitness value were selected by genetic thought, and the crossover and Gaussian mutation operations were carried out with a certain probability. The simulation results show that the improved GA-CSO algorithm can avoid local optima, enhance the global extremum search ability, and greatly improve the convergence speed and accuracy range of the algorithm.

关 键 词：改进鸡群算法 惯性权重 遗传算法 交叉变异 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]