摘要
基本蚁群算法容易陷于局部最优解是其较为突出的缺点。针对这一问题,文章提出使用双种群蚁群同时进行搜索。在迭代过程中,若判断出算法陷入可能局部最优时,则交换不同种群对应路径上的信息素,并且同时双向动态自适应调整信息素挥发系数的改进策略。通过信息素的震荡变化和挥发系数的自适应调整,扩大搜索空间,提高算法搜索的全局性。通过实验仿真,证明了此算法改进是可行和有效的。
The prominent shortcoming of the basic ant colony algorithm is easily trapped into local optimal solution.Dual population searching at the same time strategy is proposed in this paper in order to solve this problem.In the iterative process, when the algorithm is trapped into a local optimum, then change the pheromone of the corresponding path of different populations.At the same time, bi-directional dynamic adjust adaptively the volatile coefficient of the pheromone.The concussion change of the pheromone and the adaptive adjustments of the volatile coefficient can expand the search space and improve the overall searching performance .It is proved that the algorithm is feasible and effective in the emulation experiments.
作者
饶跃东
RAO Yue-dong（Wuhan University of Technology, Wuhan 430070, China）
出处
《电脑知识与技术》
2010年第1期181-183,共3页
Computer Knowledge and Technology
关键词
蚁群算法
自适应
双种群
局部最优解
挥发系数
ant colony algorithm
adaptive
dual population
local optimal solution
volatile coefficient
