一种基于2-opt算法的混合型蚁群算法  

A Hybrid Ant Colony Algorithm Based on 2-opt Algorithm

作　　者：秦东各 王长坤[1] 

机构地区：[1]南昌航空大学信息工程学院,江西南昌330063

出　　处：《工业控制计算机》2018年第1期98-100,共3页Industrial Control Computer

摘　　要：针对基本蚁群算法计算时间较长、容易陷入局部最优和收敛速度慢的缺点,提出一种混合型蚁群算法。该算法将改进的2-opt算法和改进蚁群算法进行结合,改进方法如下:对蚂蚁迭代的路径进行升序排列,只对其中一半的较短路径采用2-opt算法进行局部搜索,并对基本蚁群算法提出一种按优先级动态调整信息素挥发因子的改进策略。将该算法在TSP实例上进行模拟实验,MATLAB仿真结果表明该算法增加了全局搜索能力,提高了搜索效率,取得了较好的实验结果。A hybrid ant colony algorithm is proposed to solve the shortcoming of the basic ant colony algorithm with long time,easy to get into local optimal and slow convergence speed in this paper.This algorithm combines improved 2-opt algorithm and improved ant colony algorithm.The improvement method is as follows:the path of the ant iteration is arranged in ascending order,only half of the shorter paths are used for local search using 2-opt algorithm and an improved strategy for the basic ant colony algorithm based on the priority dynamic adjustment of pheromones is presented.The algorithm is simulated on the TSP instance.

关 键 词：蚁群算法 混合型蚁群算法 局部搜索 优先级 动态调整 信息素挥发因子 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]