摘要
蚁群算法是一种解决组合优化问题的有效算法,已得到日益深入的研究,并逐渐得到应用。蚁群算法的一个不足是,算法参数的设置往往凭借经验,缺乏充足的依据。文章以车辆路径问题(vehicle routing problem,VRP)为例,从一个烟草配送的智能决策系统中抽取一定量的数据,对蚁群算法中各参数与算法收敛性之间的关系进行了大量的仿真实验,通过对实验结果的分析,给出了解决此类问题时的一种优化算法参数的方法。
Ant colony algorithm is an effect way to solve the problem of combination optimization, which has been researched deeply and used increasingly. The deficiency of ant colony algorithm is that the parameters are set by experience without sufficient evidence. Taking vehicle routing problem as an example, a certain amount of data is extracted from an intelligent decision-making system for tobacco distribution, then a lot of simulation experiments for the relation between the ant colony algorithm parameters and the algorithm convergence are done, finally an optimal selection of the algorithm parameters to solve the problems is given by analyzing the results of the experiments.
出处
《计算机时代》
2010年第3期21-23,共3页
Computer Era
关键词
蚁群算法
收敛速度
算法参数
仿真实验
ant colony algorithm
convergence rate
algorithm parameters
simulation experiment
