文档介绍：摘要遗传算法是一种新兴的搜索寻优技术,它模拟达尔文的进化论,根据“优胜劣汰”的原则,借助选择、交叉、变异等操作逐步逼近最优解。具有隐并行机制和自适应性,因此他非常适合于多维,非线性和具有多峰值的问题。,并在八十年代得以完善,发展成为标准式的遗传算法,从九十年代中期得到广泛研究与应用。遗传算法具有全局优化性和易操作性。最初应用于非数值计算方面,直到进几年才转向全局优化问题,并取得了显著的成果,吸引了越来越多的研究者,逐渐成为人工智能领域的一个研究热点。本文对遗传算法的发展现状进行分析。在对基本遗传算法的优缺点进行分析后,主要针对它的局部搜索能力差,全局搜索速度较慢和早熟现象提出改进。为了克服这些缺点,我提出了改进的基因库的概念同时对原来的适应度函数进行了简化。主要是针对遗传算法进行改进:适应度函数和遗传算子(选择算子、交叉算子和变异算子)的改进。在此基础上,本文提出了一些新的方法。同时也提出了一种新的算子一加速算子。这些都是基于空间搜索能力展开的。它提高计算效率和精度,力求思路简捷,逻辑清晰,并通过算例加以验证关键词:遗传算法交叉变异选择距离基因库 Abstract ic Algotithm(GA')is one ofnew technologies isthe imitation of Darwin’S optimal sulution through reproduction, crossover and mutation impticit parallel mechanism and itisfitforthose kinds ofmultidimensional nonlinear problems with multiple peak which was presented by Holland in1960s ard generalized to Standard GA in1980s ing apopular domain increasingly in themiddle of 1990s。Then,many researchers focused on theimprovement and1Feapplication ofGA. GA has theability ofglobal search and iseasy ,.it isused for puting and then inopeimization domain recently,.C A ing a hot area inartificialintelligence and interested in ty mol eand more researchers. The paper introduces development analyzing tieadvantages and thedisadvantages of hasthreekind ofweakness which are lowability oflocal searching,Icw speed of global searching and premature order to e these redefine ic stock wNch has been 1he same time simplmed the fitness function ofindividual other than have construct newoperator--accelerale operator isintroduced it’S space paper introduces some improved methods published last on thebasis ofother people’S study, proposes some ECW methods which can improve putational efficiency andprecisk new methods aim ateasyundersantding and lucidlogic,and they are alltestedby