| 引用本文： |
-
区云鹏,韦兆文,蒋慧超.基于多信息素的蚁群算法[J].广西科学院学报,2008,24(3):240-242. [点击复制]
- OU Yun-peng,WEI Zhao-wen,JIANG Hui-chao.An Ant Colony Algorithm Based on Multi-Pheromones[J].Journal of Guangxi Academy of Sciences,2008,24(3):240-242. [点击复制]
|
|
| 摘要: |
| 针对传统增强型蚁群算法容易出现早熟和停滞现象的缺陷,提出一种多信息素的蚁群算法(MPAS),并以TSPLIB的数据为例对该算法进行实验测试.MPAS算法将信息素分为局部和全局两种不同的信息素,在搜索过程中,对局部和全局信息素采用不同的更新策略和动态的路径选择概率,使得在搜索的中后期能更有效地发现全局最优解.在中大型问题上MPAS算法有着更好的发现最优解的能力. |
| 关键词: 蚁群算法 信息素 旅行商问题 |
| DOI： |
| 投稿时间：2007-09-25 |
| 基金项目:广西自然科学基金项目(桂科自0640026)资助。 |
|
| An Ant Colony Algorithm Based on Multi-Pheromones |
|
OU Yun-peng, WEI Zhao-wen, JIANG Hui-chao
|
| (School of Computer, Electronics and Information, Guangxi University, Nanning, Guangxi, 530004, China) |
| Abstract: |
| This paper improves an ant colony algorithm based on multi-pheromones and solves the problem existed in classical augment ant colony algorithm.The basic idea is to divide the pheromone into local pheromone and global pheromone. Then their pheromones are updated using different strategies during searching optimal path. Many experiments based on the data of TSPLIB show the advantages of this algorithm in sweeping problems. |
| Key words: ant colony algorithm pheromones traveling salesman problem |
