计算机工程 ›› 2008, Vol. 34 ›› Issue (14): 206-207.doi: 10.3969/j.issn.1000-3428.2008.14.073

• 人工智能及识别技术 • 上一篇    下一篇

基于蚁群与鱼群的混合优化算法

修春波1,张雨虹2   

  1. (1. 天津工业大学计算机技术与自动化学院,天津 300160;2. 唐山学院信息工程系,唐山 063000)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2008-07-20 发布日期:2008-07-20

Hybrid Optimization Algorithm Based on Ant Colony and Fish School

XIU Chun-bo1, ZHANG Yu-hong2   

  1. (1. School of Computer Technology and Automation, Tianjin Polytechnic University, Tianjin 300160; 2. Department of Information Engineering, Tangshan College, Tangshan 063000)
  • Received:1900-01-01 Revised:1900-01-01 Online:2008-07-20 Published:2008-07-20

摘要: 基于鱼群算法和蚁群算法提出一种混合优化算法用于求解组合优化问题。将鱼群算法中拥挤度的概念引入到蚁群算法中,在优化过程的初期,设置较强的拥挤度限制,保证大部分蚂蚁不受信息素浓度的影响而进行随机寻优。随着寻优迭代次数的增加,拥挤度的限制逐渐减弱,最后蚁群完全由信息素和启发信息来指导寻优。在寻优初期该算法具有较强的遍历寻优能力,能够较快发现全局最优解的存在,而寻优后期,算法利用信息素正反馈的作用保持了较快的收敛速度。仿真结果验证了该方法的有效性。

关键词: 人工鱼群算法, 蚁群算法, 组合优化

Abstract: This paper proposes a hybrid optimization algorithm to resolve combinatorial optimization problem. Aswarm degree in the artificial fish school algorithm is used in ant colony algorithm. During the initial process of the optimization, the aswarm degree plays the main role to guide the ants to search the new path randomly, which makes the algorithm have the stronger ergodicity searching ability. The role of the aswarm degree gradually decreases to zero, the algorithm becomes the conventional ant colony and completes the optimal process by the principle of pheromone positive feedback, which insures the algorithm to have a quick convergence rate. Simulation results prove the validity of the algorithm.

Key words: artificial fish school algorithm, ant colony, combinatorial optimization

中图分类号: