蚁群算法中蚂蚁更新方法之研究  

Study on updating algorithms in ant colony algorithm

作　　者：孟非[1] 李静宜[2] 朱人杰[3] 

机构地区：[1]江苏科技大学经济管理学院,江苏镇江212003 [2]江苏科技大学电子信息学院,江苏镇江212003 [3]山东省邮电规划设计院有限公司,济南250031

出　　处：《计算机工程与应用》2011年第25期54-57,共4页Computer Engineering and Applications

基　　金：江苏省高校自然科学基础研究项目(No.07KJB510032)

摘　　要：蚁群算法是根据蚂蚁的觅食行为而提出的随机优化算法,但其存在早熟收敛和搜索精度低等问题。模拟生物克隆选择中5%的B细胞自然消亡过程,在蚁群算法进化过程中分别基于代间差分、混沌理论、变异原理等方法设计了8种蚂蚁更新算法,按照模拟退火方法进行更新后蚂蚁的选择。通过数值试验得出结论:基于代间差分和混沌变异的蚂蚁更新算法是一种很好的选择,并且当性能较差的20%左右蚂蚁按照这种算法更新时效果较好。这种算法可以有效克服蚁群算法的早熟现象,能够加快收敛速度。Ant Colony Algorithm（ACA） is a stochastic optimization algorithm inspired by the behavior of ant looking for food.However,the standard ACA has some shortcomings,such as premature convergence,searching precision lowness and so forth.Based on the simulation of natural death process of 5% B-cell in biology clone selection,this paper proposes 8 kinds of updating algorithms according to intergeneration differential,theory of chaos,principle of mutation respectively,and selects the updated ants in terms of simulated annealing method.Numerical experiments show that the updating algorithm by using intergeneration differential and chaotic mutation is a good selection.Simultaneously,the updating effect is perfect when the updated ants are about 20%.This algorithm can effectively overcome the premature problem and speed up the convergence.

关 键 词：蚁群算法 克隆选择 混沌 变异 模拟退火 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]