带有禁忌规则的改进蚂蚁算法  

An Improved ant Colony System with Tabu Rule

作　　者：范彬毅[1] 姜同强[1] 

机构地区：[1]北京工商大学计算机与信息工程学院,北京100048

出　　处：《计算机仿真》2011年第1期32-34,共3页Computer Simulation

摘　　要：提出了一种带有禁忌规则的改进蚂蚁算法,改进的算法在每次迭代后,通过对信息素值的判断,来禁止信息素浓度过高的路径被"人工蚂蚁"过多访问,以此加大蚂蚁搜索领域的能力从而减少算法过早收敛于非最优解的现象,在此基础上,算法结合了当前被证实为最有效解决TSP问题的蚂蚁系统和最大最小蚂蚁算法的部分规则,对算法做出进一步改进。改进的算法通过数学推导证明该算法值收敛成立,并利用C#编程实验,结果表明,算法具有较好的求解性能。To increase the capacity of ants search, an improved ant colony optimization algorithm with the tabu rule is presented by judging the pheromone value in each iteration. The combination of ant system and the maximum and minimum rules for ant algorithm can further improve the algorithm. By mathematical derivation, the value of the convergence of improved algorithm can be proved and the C # programming experiment results show that the algorithm has good solving performance.

关 键 词：蚂蚁算法 最大最小蚂蚁算法 信息素 禁忌规则 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]