智能优化方法的改进王颖.docVIP

河南财经政法大学成功学院本科生毕业论文(设计) 智能优化方法的改进 院系名称 信息工程系 姓　　名 王颖 学　　号 2010341108 专　　业 计算机科学与技术 指导教师 李喜艳 讲师 2014年4月20摘　要 遗传算法GAs(Genetic Algorithms)是一种模拟自然界生物进化的智能优化算法,产生于一些生物学家用计算机模拟生物进化过程的仿真实验。1960年，美国的J．H．Holland把它创造性地应用于人工系统，并成功地利用它解决了一些实际问题，例如“旅行商问题，加上该算法简单易行、鲁棒性强，特别是他不需要专门的领域知识而仅用适应度函数作评价来指导搜索过程。从此，它的实用价值逐渐被人们所认识。特别是近十年，由于计算机性能的提高，以及并行分布式计算的推广，GAs由于自身独特的优势而越来越受 到人们的重视[1,2]。遗传算法目前已经在优化控制、机器学习、模式识别、神经网络和并行处理等领域得到了越来越广泛的应用[3-6]。 遗传算法是一种新兴的技术,正处于发展期。虽然在应用领域获得了丰收，但其理论基础还比较薄弱，有许多地方需要研究和发展充实。例如简单遗传算法SGAs(Simple Genetic Algorithms)由于自身固有的缺陷，通常优化过程的收敛速度较慢，而且算法稳定性较差。 论文的主要工作是遗传算法的基本概念、构成要素、基本流程、收敛性、特点并举例验证了两种简单遗传算法。并针对遗传算法的一些问题做了一些简单的验证，通过对两种简单遗传代码的分析，举例说明，实践结果验证等，更加深入地研究和讨论遗传算法。并对这两种结果进行比较，得出两种算法的优劣，比以后的研究做准备。 关键词：遗传算法；收敛性[7]；简单遗传算法； Abstract Genetic algorithm (ga) GAs a kind of intelligent optimization algorithm of mimic natural biological evolution, produced in some biological home computer analog simulation experiment of biological evolution.In 1960, the United States J.H.Holland make it creatively applied to the artificial systems, and use it successfully solves some actual problems, such as the traveling salesman problem, and this algorithm is simple, strong robustness, especially he doesnt need special domain knowledge and in evaluation of fitness function to guide the search process. Since then, its practical value is gradually known by people. Especially the past 10 years, due to the improvement of the performance of the computer, and the expansion of parallel and distributed computation, GAs due to its unique advantages and more and more importance by people. Genetic algorithm (ga) is now in the optimization control, machine learning, pattern recognition, neural networks and parallel processing, and other fields has been more and more widely used . Genetic algorithm (ga) is an emerging technology and in development.Although had a good harvest in the field of application, but its theoretical basis is relatively weak, there are so many places need to