摘要： 蚁群算法中参数选择不当会直接影响算法的全局收敛性和求解效率。为此，分析各参数对算法性能的影响，提出确定蚁群算法参数最优组合的两阶段法，在离线调整阶段将蚁群算法基本模型的参数设定问题描述成多因素多水平的均匀设计，然后在在线调整阶段引入混沌扰动以避免搜索过程陷入局部极值。实验结果表明，两阶段法发现最优解的能力明显优于在线调整前的蚁群算法。

关键词: 蚁群算法, 两阶段法, 离线参数调整, 在线参数调整, 混沌理论

Abstract: If the parameters of ant colony algorithm are wrongly chosen, it has great impact on global convergence and efficiency of the system. This paper analyzes the influence of various parameters on algorithm performance, and proposes a two-stage method to determine the optimal combination of parameters in ant colony algorithm that firstly uniform design method is used in the off-line tuning stage to convert the problem of parameter setting into the uniform design of multi-factor and multi-level. The on-line tuning stage chaos perturbation is utilized to avoid the search being trapped in local optimum. Experimental result shows that the proposed two-stage method can achieve good results, being beneficial to the promotion and application of ant colony algorithm.

Key words: ant colony algorithm, two-stage method, off-line parameter tuning, on-line parameter tuning, chaos theory

中图分类号: 

• TP312