多目标细菌觅食优化算法  

Multi-objective bacteria foraging optimization algorithm

作　　者：李珺[1] 党建武[1] 王垚 包敏 Li Jun;Dang Jianwu;Wang Yao;Bao Min(School of Electronic ＆ Information Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China)

机构地区：[1]兰州交通大学电子与信息工程学院,兰州730070

出　　处：《计算机应用研究》2018年第7期1996-2000,共5页Application Research of Computers

基　　金：甘肃省科技计划资助项目(1506RJZA084);甘肃省教育厅科研项目(1204-13);甘肃省教育科学"十二五"规划课题(GS[2015]GHB0907);兰州市科技计划资助项目(2015-2-74)

摘　　要：传统的细菌觅食优化算法仅针对单目标优化问题寻优,为进一步发掘细菌群体智能在多目标优化问题中的寻优优势,提出了改进的多目标细菌觅食优化算法。在个体间互不支配时给出归一化的择优策略;引入差分思想完成复制操作,提高种群的多样性;采用栅格划分法进行迁徙操作,提高解集的分散性;同时使用外部集存放当前找到的非支配解,并不断对外部集进行优化。通过对多个标准函数进行测试并与其他几种算法的对比结果表明,所提出的多目标细菌觅食优化算法在解的收敛性和分散性指标上都有一定提升,能够有效解决多目标优化问题。Conventional bacterial foraging optimization algorithm simply optimizes the single target optimization problems. In order to exploit the further strengths of bacterial colony in multiple target optimization,this paper proposed the improved multiple target bacterial foraging algorithm. It put forward the optimization strategy via normalization method when individuals had no inter-dominance. It increased population diversity at maximum with the introduction of difference in the completion of replication. It enhanced the solution set dispersiblity with the assistance of grid portioning method in the targeted migration operation. Simultaneously,it put the found non-dominant solution at present in the external data set and continuously optimized the non-dominant solution set in the external data set applying the given update strategy. The outcome of comparison between several other algorithm and the test of numerous standard function manifest that,the proposed multiple target bacterial foraging algorithm raises both astringency and dispersiblity of solution,which can address multiple target optimization problem.

关 键 词：多目标优化问题 细菌觅食优化算法 归一化 差分进化 外部集 栅格 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]