摘要
蚁群算法是一种新型的模拟进化算法,该算法采用分布式并行计算和正反馈机制,具有较强的鲁棒性,易于与其他方法结合,目前在很多优化领域中得到了广泛应用,但是进化速度慢,易陷入局部最优是其最主要的缺点。本文在基于网格划分策略的蚁群算法的基础上,结合混沌理论,提出了混沌蚁群算法。在算法初始化和信息素更新方面提出了改进,采用了MAX-MINAntSystem的思想对路径上可能的残留信息素进行了限制,通过实例验证,证明了该算法是有效性。
Ant colony algorithm is a novel stochastic optimization algorithm using artificial ants releasing pheromone on the path, characterized with a positive feedback, distributed computation and parallel algorithm. It has strong robustness and easy to combine with other methods in optimization. The slow step of convergence and easy to be trapped in local optimum is the most shortcoming , although it is widely applied to optimization problems. Based on ant colony algorithm using gridding method and combined with chaos theory, Chaos-Ant Colony Algorithm is put forward. In the Chaos-Ant Colony Algorithm, some MAX-MIN Ant System idea is used to limit the pheromone remained in the path. Improvements are made in initialization and update of pheromone. The feasibility of the proposed Chaos-Ant Colony Algorithm is validated by experiment.
出处
《机电产品开发与创新》
2010年第4期29-31,共3页
Development & Innovation of Machinery & Electrical Products
关键词
蚁群算法
混沌
信息素
混沌蚁群算法
colony algorithm
chaos
pheromone
chaos-ant
colony algorithm
