蚁群优化算法的收敛性分析  

Analysis of Convergence of Ant Colony Optimization Algorithms

作　　者：朱庆保[1] 

机构地区：[1]南京师范大学计算机科学系,南京210097

出　　处：《控制与决策》2006年第7期763-766,770,共5页Control and Decision

基　　金：江苏省自然科学基金项目(01KJB520007)

摘　　要：有关蚁群优化算法收敛性分析的研究还很少,不利于进一步改进其算法.为此,较详细地分析了用蚁群优化算法求解TSP问题的收敛性,证明了当0<q0<1时,算法能够收敛到最优解.分析了封闭路径性质、启发函数、信息素和q0对收敛性的影响,据此给出了提高算法收敛速度的几点结论.The convergence problems of ant colony optimization algorithm is studied. To improve the algorithm, the convergence of this algorithm applied to TSP is analyzed in detail. The algorithm will be certain to converge to the optimal solution under the condition 0〈q0〈1. In addition, the influence on its convergence caused by the properties of the closed path, heuristic functions, the pheromone and q0 is analyzed. Based on it, some conclusions about the improvement of the speed of convergence are obtained.

关 键 词：蚁群优化算法 收敛性分析 启发函数 TSP问题 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]