摘要
针对基本磷虾算法收敛效率低下,容易收敛到局部极值的缺点,基于协同进化和量子计算基本理论,提出一种量子行为磷虾算法,称为协同进化量子磷虾算法(cooperative evolution quantum krill herd algorithm,CEQKHA).该算法将磷虾种群划分为主种群和辅种群,各种群能够独立进化,并实现优良个体的交换.利用量子进化行为方式更新磷虾个体位置,引进delta势阱,将粒子势阱中心设置为最优个体位置,获取磷虾进化后的位置,并分别将主种群和辅种群个体采用不同的位置更新方式,提高种群勘探和开采的能力.对所提出的算法进行了收敛性分析,证明了所提出算法的收敛性能.最后利用基准函数进行了仿真对比分析,经仿真验证,所提出的CEQKHA能得到更优解,具备良好的优化性能.
Aiming at the low efficiency and easy trap into local optimum for the basic krill herd algorithm,a new quantum behave krill herd algorithm was proposed.It was named as cooperative evolution quantum krill herd algorithm (CEQKHA)based on cooperative evolution and quantum computation.The population could be divided into two parts,such as main population and sub-population,which could evolve independently and exchange fine individuals.The position of krill herd could be updated by using quantum activity.The best position of krill herd could be set as the center of potential well of delta potential well,and the position of krill herd in the different population could evolve in different way.The convergence of the algorithm was also deduced.Simulation on benchmark functions proved that CEQKHA could get better results and perform better than other algorithms.
作者
刘振
刘文彪
鲁华杰
LIU Zhen;LIU Wenbiao;LU Huajie(College of Coastal Defense Force,Naval Aviation University,Yantai 264001,China)
出处
《郑州大学学报（工学版）》
CAS
北大核心
2018年第6期43-49,共7页
Journal of Zhengzhou University（Engineering Science）
基金
国家自然科学基金资助项目(51605487
61174031)
关键词
磷虾算法
协同进化
量子
多种群
势阱
krill herd algorithm
cooperative evolution
quantum
multi-population
potent.ial well
