基于多头绒泡菌网络模型的蚁群算法优化  

An Optimized Ant Colony Optimization Algorithm Based on the Physarum Network Model

作　　者：刘玉欣[1] 张自力[1] 高超[1] 钱涛[1] 吴雨横 

机构地区：[1]西南大学智能软件与软件工程重点实验室,重庆400715

出　　处：《西南大学学报（自然科学版）》2014年第9期182-187,共6页Journal of Southwest University(Natural Science Edition)

基　　金：国家科技支撑计划资助项目(2012BAD35B08);重庆市自然科学基金资助项目(cstc2012jjA40013;cstc2013jcyjA40022);高等学校博士学科点专项科研基金新教师基金资助(20120182120016);中央高校基本科研业务费专项资金基金资助(XD-JK2012B016;XDJK2012C018;XDJK2013D017)

摘　　要：基于多头绒泡菌网络模型在求解迷宫问题时所展现的"重点管道重点培养"特性,设计一种优化的蚁群算法.该优化算法在更新信息素矩阵时考虑蚂蚁释放的信息素和多头绒泡菌网络中流通的信息素.通过对重点管道信息素浓度的加强,提高路径寻优过程中重点管道的被选概率,从而提高蚁群算法对最优解的开发力度.针对旅行商问题的对比实验验证了该优化算法可提高传统蚁群算法的寻优能力,并具有更高的鲁棒性.The Physarum network （PN） model exhibits a unique feature that the critical pipelines are reserved with the evolution of network during the process of solving the maze problem .In the present study , drawing on this feature ,an optimized ant colony optimization （ACO） algorithm denoted as PNACO algorithm is proposed based on the PN model .When the pheromone matrix is updated ,the PNACO algorithm updates both the pheromone released by ants and the flowing pheromone in the Physarum network .By adding extra pheromones in the Physarum network ,the critical pipelines are the shortest routes that have a higher opportunity to be selected w hen ants travel cities. Hence, the exploitation of the optimal solution will be promoted .Experimental results show that the solutions of PNACO algorithm are better than those of ACO algorithm for solving the traveling salesman problem（TSP）. In addition, the PNACO algorithm is more robust than the ACO algorithm .

关 键 词：多头绒泡菌网络模型 蚁群算法 旅行商问题 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]