一种基于搜索多样性的改进蚁群算法研究  

An Improved Ant Colony Algorithm Based on the Search for Diversity

作　　者：王涌鑫[1] 王希武[1] 钟启艮[1] 

机构地区：[1]军械工程学院,石家庄050003

出　　处：《计算机与数字工程》2013年第6期896-898,共3页Computer & Digital Engineering

摘　　要：传统蚁群算法是以正反馈策略进行指导,实质上这种指导有利于算法的收敛却不利于搜索的多样性。为了缩短寻优路径,论文通过提高搜索多样性提出了一种改进的蚁群算法。该算法,正反馈、逆反馈同时作用使得整个蚁群的搜索空间和搜索结果的多样性大大增加,蚁群搜索的时间大大减少。通过对TSP问题的仿真结果表明,新算法对解决复杂组合优化问题具有显著效果。The traditional ant colony algorithm is based on the positive feedback mechanism, in essence, this guidance is conducive to the convergence of the algorithm is not conducive to the diversity of the search. In order to shorten the length of the path o{ the optimization, this paper proposes an improved ant colony algorithm to improve search diversity. The algorithm, the positive feedback, the inverse feedback simultaneously makes the entire ant colony search time is greatly reduced, will greatly increase the diversity of the ant colony search space and search results. Simulation results of the TSP problem show that, the new algorithm for solving complex combinatorial optimization prob- lem has a significant effect.

关 键 词：蚁群算法 多样性 正反馈 逆反馈 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]