求解旅行商问题的蚁群优化算法参数设计  

Parameter Design of Ant Colony Optimization for Travelling Salesman Problem

作　　者：汤雅连 杨期江 TANG Yalian;YANG Qijiang(School of Internet Finance and Information Engineering, Guangdong University of Finance, Guangzhou 510520, China;School of Marine Engineering, Guangzhou Maritime University, Guangzhou 510725, China)

机构地区：[1]广东金融学院互联网金融与信息工程学院,广东广州510520 [2]广州航海学院轮机工程学院,广东广州510725

出　　处：《东莞理工学院学报》2020年第3期48-54,共7页Journal of Dongguan University of Technology

基　　金：国家自然科学基金资助项目(51375178);广东省教育厅项目(2017KQNCX145);广东省自然科学基金(2018A030310017);广州市科技计划(201904010133)。

摘　　要：针对蚁群优化算法存在收敛时间较长,容易陷入局部最优的特点,对算法参数进行研究,通过仿真实验来设计适合问题模型的算法参数,然后有针对性地进行改进,通过3-opt策略提高算法的局部搜索能力,并对几个算例进行仿真,结果证明提出的改进策略在求解小规模TSP模型时具有相对优势。In view of the characteristics of ant colony optimization,which has a long convergence time and is easy to fall into local optimization,this paper studies the algorithm parameters,and designs the suitable algorithm parameters for the problem model through a large number of simulation experiments,and then improves them pertinently,improves the local search ability of the algorithm through the 3-opt strategy,several examples are simulated,and the results prove that it has a comparative advantage the improved strategy proposed in solving the small-scale TSP model.

关 键 词：旅行商问题 蚁群优化算法 3-opt 

分 类 号：TP31[自动化与计算机技术—计算机软件与理论]