蚁群系统自适应策略的改进与分析

Improvement and Analysis of Self-adaptive Strategy in Ant Colony System

蚁群系统能够通过自适应调整不断优化算法的性能。为寻求算法自适应过程的内部规律,结合旅行商问题,采用参数控制、设置信息素范围的方法进行探讨。通过调控信息素的变化,以及对信息素最值、分布状态的统计分析,揭示算法优化过程的内部状态。实验表明,改进后的算法更稳定,问题解的搜索能力更强。

Ant Colony System（ACS） can develop excellent performance via self-adaptive behavior. In order to find the internal rules of self- adaptive behavior, this paper introduces parameter-control and sets pheromone＇s range, which are applied to the Traveling Salesman Problem（TSP）. The ACS internal state is revealed via pheromone＇s micro-control and statistical analysis of pheromone＇s most values and distribution. Experimental results prove that the improved ACS does well in stability and searching solution.