山东理工大学信息与计算机科学,山东理工大学信息与计算科学专业毕业论文.docx...

SHANDONGＵＮＩＶＥＲＳＩＴＹ　ＯＦ　ＴＥＣＨＮＯＬＯＧＹ

毕业论文

群智能算法用于优化计算问题研究

学 院： 理学院

专 业： 信息与计算科学

学生姓名：

学 号：

指导教师：

2012 年 6 月

PAGE I

摘 要

优化算法是一门以数学为基础、用来求解各种工程问题最优解的计算机应用技术。随着技术进步，组合优化问题变得越来越复杂，传统的计算方法在求解这类问题时面临着诸多问题，如计算复杂度高、计算时间长等，特别是对于一些NP难和NP完全问题。为了在求解时间和计算精度上取得平衡，研究学者提出了各种启发式特征的计算方法。常见的启发式特征算法有禁忌搜索、模拟退火算法、进化算法和群智能等。

本课题主要研究了群智能算法在组合优化计算问题方面的应用。群智能算法是一种基于群体搜索的随机优化算法，主要有蚁群算法和粒子群优化算法两种。文章中详细地介绍了蚁群算法和粒子群优化算法的原理、发展及基本流程，我的工作主要是利用群智能算法在VC++平台下实现其在连续空间及离散领域的常见应用，如连续多峰函数极值求解、0-1背包和典型的旅行商问题(TSP)。

关键词：群智能，蚁群算法，粒子群优化算法，0-1背包，TSP

PAGE II

Abstract

The optimization algorithm is a computer application technology which based on the mathematics and used to solve various engineering problems for optimal solution. With the progress of technology, combinatorial optimization problems become more and more complex. The traditional methods in solving such problems are faced of much trouble, such as high computational complexity and long computing time and so on, especially for some NP-hard and NP-complete problems. The researchers introduce several of heuristic algorithms in order to strike a balance in the computing time and calculation accuracy. The ordinary heuristic algorithms are tabu search, simulated annealing, evolutionary computation and swarm intelligence and so on.

In this paper, I mainly research the application of swarm intelligence algorithm in combinatorial optimization problem. Swarm intelligence algorithm which mainly includes ant colony optimization and particle swarm optimization is a stochastic optimization algorithm based on colony search. The article describes the theory, development and the basic process of ant colony optimization and particle swarm optimization in detail. My main work is achieving common applications of continuous space and discrete area in the platform of VC++ by use of swarm intelligence algorithm. Common applications are continuous peak funct