摘要
为了求解满足一定时间限制的最大概率路径问题,在建立该问题数学模型的基础上,提出了一种改进蚁群算法。首先根据随机网络的定义建立了随机网络最大概率路径问题的数学模型,然后结合随机网络最大概率路径问题的特点,设计了一种新的启发式信息和信息素更新规则的改进蚁群算法,最后选择了4组数据,将改进蚁群算法与一种混合遗传算法进行对比试验,分别求取对应的全局最大概率路径和反映算法总体性能的多项数据。实验表明,改进蚁群算法的收敛速度和总体性能均优于混合遗传算法,为求解随机网络最大概率路径问题提供了一种快速、可行的方法。
To solve the problem of the maximum probability of path satisfying a certain time limit, an improved ant colony algorithm was proposed based on establishing the mathematical model of the problem. Firstly, on the basis of the definition of stochastic network, the model of the maximum probability path problem in the stochastic network was established. Then,according to the characteristics of the maximum probability path problem of stochastic networks, a new ant colony algorithm based on the new heuristic information and pheromone updating rule was designed. The improved ant colony algorithm and a hybrid genetic algorithm were compared on 4 sets of data, and then the corresponding global maximum probability path and a number of data to reflect the overall performance of the algorithm were obtained respectively. According to the experimental result, the convergence speed and the global performance of the improved ant colony algorithm are better than that of the hybrid genetic algorithm. It provides a feasible and fast method for solving the maximum probability path problem of stochastic networks.
出处
《解放军理工大学学报（自然科学版）》
EI
北大核心
2016年第4期391-395,共5页
Journal of PLA University of Science and Technology(Natural Science Edition)
基金
江苏省高校重点学科专项经费资助项目(苏教研[2011]14号)
关键词
最大概率路径问题
随机网络
蚁群算法
遗传算法
maximum probability path problem
stochastic network
ant colony algorithm
geneticalgorithm
