改进的遗传粒子群混合优化算法  

Improved hybrid optimization algorithms based on genetic algorithm and particle swarm optimization

作　　者：陈璐璐[1] 邱建林[2] 陈燕云[3] 陆鹏程[1] 秦孟梅 赵伟康[2] CHEN Lu-lu;QIU Jian-lin;CHEN Yan-yun;LU Peng-cheng;QIN Meng-mei;ZHAO Wei-kang(School of Electronic Information, Nantong University, Nantong 226019,China;School of Computer Science and Technology,Nantong University,Nantong 226019,China;Engineering Training Center, Nantong University, Nantong 226019,China)

机构地区：[1]南通大学电子信息学院,江苏南通226019 [2]南通大学计算机科学与技术学院,江苏南通226019 [3]南通大学工程训练中心,江苏南通226019

出　　处：《计算机工程与设计》2017年第2期395-399,共5页Computer Engineering and Design

基　　金：国家自然科学基金项目(NSF61272424);江苏省自然科学基金项目(BK2010277);南通市科技计划基金项目(K2010002;AL2007033)

摘　　要：为解决遗传算法计算时间长和粒子群算法易陷入局部极值的问题,提出一种基于实数编码的改进的遗传算法与粒子群算法混合的优化算法。改进遗传算法中的选择算子,保留适应度值较好的个体,重新组成新的种群,由粒子群算法更新速度和位置,对个体进行进一步的成熟。交叉算子采取精英竞争策略,选取适当个体进行交叉,剩余个体再次通过PSO算法更新速度和位置,将粒子群思想引入变异算子。通过对4个函数的优化,对此算法进行测试,并研究比较其它算法,测试结果表明,该算法在收敛性、运算速度和优化能力方面具有优越性。To solve the problem that the genetic algorithm(GA)?s calculation time is long and that the particle swarm optimiza-tion(PSO)is easy to fall into local optima,an improved hybrid optimization algorithms based on GA and PSO was developed.The selection operator of the GA was improved to retain the better individuals,and speed and position of the individuals were up-dated with PSO for further maturation.Elitist strategy was adopted in crossover,and appropriate individuals were selected to cross,speed and position of the remaining individuals were updated using PSO again.The idea of PSO was introduced to muta-tion operator.Through dealing with the four functions optimization problems,the proposed method was tested and compared with other bibliography,showing its advantages in convergence,computing speed and optimization capabilities.

关 键 词：选择算子 交叉算子 变异算子 遗传算法 粒子群算法 混合算法 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]