基于信息素强度的改进蚁群算法  

An Improved Ant Colony Algorithm Based on Pheromone Intensity

作　　者：郑卫国[1] 田其冲[1] 张磊[1] 

机构地区：[1]中国矿业大学计算机科学与技术学院,江苏徐州221116

出　　处：《计算机仿真》2010年第7期191-193,229,共4页Computer Simulation

摘　　要：现有的基本蚁群算法和MMAS算法都存在收敛速度慢、易陷于局部最优解等缺点,为了提高算法搜索效率,提出了一种求解旅行商问题的改进蚁群算法。在基本蚁群算法和MMAS算法的基础上,通过对蚂蚁进行区分,直接控制信息素的浓度,并进行有选择的更新,有效地抑制了算法收敛过程中的停滞和早熟现象,提高了全局搜索能力和解的质量。最后通过经典的CTSP31实例验证了该改进算法的有效性,仿真实验结果表明,它在最优解、平均解和最优迭代次数等性能上比经典蚁群算法都有较大的改善。The basic ant colony algorithm and the MMAS algorithm have some disadvantages,such as slow convergence rate and easy to fall into a local optimal solution.In order to improve the performance of ant colony,this paper proposed a new ant colony algorithm which solves the traveling salesman problem better.Based on the basic ant colony algorithm and MMAS algorithm,the ants were differentiated in the new algorithm.The pheromone intensity was controlled directly and updated selectively.The new algorithm effectively inhibits the stagnation and premature phenomena in the process of convergence,and improves the global search capability and the quality of solution.Finally, the example CTSP31 proved the validity of the improved algorithm.The experimental results show that the improved algorithm is better than the classical algorithms on the performance of the optimal solution,the average solution and the optimal iterations.

关 键 词：蚁群算法 旅行商问题 信息素 改进算法 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]