一种模拟羊群行为的新型群集智能算法  

A New Swarm Intelligence Algorithm by Simulating Sheep Behaviors

作　　者：曲大鹏[1] 许伦祥 鲁垚光 原晓坤 黄敏[3] 王兴伟[4] QU Da-peng;XU Lun-xiang;LU Yao-guang;YUAN Xiao-kun;HUANG Min;WANG Xing-wei(College of Information,Liaoning University,Shenyang Liaoning110036,China;School of Computer Science and Engineering,Northeastern University,Shenyang Liaoning 110169,China;College of Information Science and Engineering,Northeastern University,Shenyang Liaoning110169,China;Cbllege of Software,Northeastern University,Shenyang Liaoning110169,China)

机构地区：[1]辽宁大学信息学院,辽宁沈阳110036 [2]东北大学计算机科学与工程学院,辽宁沈阳110169 [3]东北大学信息科学与工程学院,辽宁沈阳110169 [4]东北大学软件学院,辽宁沈阳110169

出　　处：《电子学报》2018年第6期1300-1305,共6页Acta Electronica Sinica

基　　金：国家自然科学基金(No.61572123);国家自然科学基金重点国际合作研究项目(No.71620107003);国家杰出青年科学基金(No.71325002);教育部-中国移动科研基金(No.MCM20160201);辽宁省博士科研启动基金(No.20170520323);辽宁大学博士科研启动项目

摘　　要：针对当前多种群集智能算法往往包含过多策略而掩盖算法核心机理的现状,设计了一种模拟羊群行为的新型群集智能算法.该算法从群集智能算法的核心出发,通过分别模拟羊群的三类行为:头羊引领、羊群互动和牧羊犬监督,设计群集智能算法中相对应的三种策略:全局探索、局部开发和跳出局部优化.基准测试函数的实验结果表明,较之粒子群算法,该算法能获得更高质量的解,同时具有更快的收敛速度和更好的稳定性.The status of current various swarm intelligence algorithms contains many strategies which cover their core issues,therefore,a new swarm intelligence algorithm by simulating sheep behaviors optimization（ SO） is devised. Based on the core of swarm intelligence algorithms,SO designs three kinds of strategies,namely global exploration,local exploitation and escaping from local optimization,by simulating three different kinds of corresponding sheep behaviors,namely bellwether guidance,moving to each other,and shepherd dog supervision respectively. The experimental results of benchmark test functions demonstrate that,compared with particle swarm optimization,SO gets better solution,faster convergence speed and better stability.

关 键 词：群集智能 测试函数 全局探索 局部开发 

分 类 号：O224[理学—运筹学与控制论]