智能计算课程论文

《智能计算课程论文》由会员上传分享，免费在线阅读，更多相关内容在学术论文-天天文库。

1、武汉理工大学智能计算课程论文智能计算课程论文基于C++编程语言用蚁群算法解决TSP问题学院：机电工程学院学号：0121504970524姓名：文科1武汉理工大学智能计算课程论文指导老师：黄樟灿2018年5月8日1武汉理工大学智能计算课程论文摘要在人工智能迅猛发展的今天，智能优化计算中的各类算法在其中起到了至关重要的作用。本文以著名的TSP问题为研究对象，通过蚁群算法将之高效优化。在VS2017的编译环境下，用C++编程语言对此算法进行执行，并输出调试结果，通过程序源代码和各项输出数据表明蚁群算法在解决TSP问题上的高效性和优化性。关键字：蚁群算法、TSP、C++武汉

2、理工大学智能计算课程论文目录第一章绪论·················································································11.1何为TSP问题·······································································11.2TSP问题概述········································································11.3TSP问题举例··········

3、······························································1第二章蚁群算法···············································································22.1群体智能算法概述································································22.2蚁群算法的起源··················································

4、·················22.3从自然现象到算法原理··························································22.4蚁群算法的应用领域·····························································42.4蚁群算法的研究背景·····························································4第三章利用蚁群算法进行C++编程································

5、······················43.1从真实蚁群到人工蚁群··························································43.2程序设计思想·········································································53.3程序调试结果及数据分析·······················································7附录程序源代码·······························

6、·············································10参考文献························································································18武汉理工大学智能计算课程论文第一章绪论1.1何为TSP问题TSP问题（全称：TravellingSalesmanProblem），又称为旅行商问题、货郎担问题、tsp问题，是一个多局部最优的最优化问题：有n个城市，一个推销员要从其中某一个城市出发，唯一走遍所有的城市，再回到他出发的城市，求

7、最短的路线。也即求一个最短的哈密顿回路。1.2TSP问题概述TSP问题是一个组合优化问题。该问题可以被证明具有NP计算复杂性。因此，任何能使该问题的求解得以简化的方法，都将受到高度的评价和关注。旅行推销员问题是数图论中最著名的问题之一，即“已给一个n个点的完全图，每条边都有一个长度，求总长度最短的经过每个顶点正好一次的封闭回路”。Edmonds，Cook和Karp等人发现，这批难题有一个值得注意的性质，对其中一个问题存在有效算法时，每个问题都会有有效算法。迄今为止，这类问题中没有一个找到有效算法。倾向于接受NP完全问题（NP-Complet或NPC）和NP难题。