搜索
2010届信息与计算科学专业毕业设计I毕业论文题目粒子群算法及其参数设置专业信息与计算科学班级计算061学号3060811007学生xx指导教师徐小平2010年任侃：粒子群优化算法及其参数设置II粒子群优化算法及其参数设置专业：信息与计算科学学生：xx指导教师：徐小平摘要粒子群优化是一种新兴的基于群体智能的启发式全局搜索算法，粒子群优化算法通过粒子间的竞争和协作以实现在复杂搜索空间中寻找全局最优点。它具有易理解、易实现、全局搜索能力强等特点，倍受科学与工程领域的广泛关注，已经成为发展最快的智能优化算法之一。论文介绍了粒子群优化算法的基本原理，分析了其特点。论文中围绕粒子群优化算法的原理、特点、参数设置与应用等方面进行全面综述，重点利用单因子方差分析方法，分析了粒群优化算法中的惯性权值，加速因子的设置对算法基本性能的影响，给出算法中的经验参数设置。最后对其未来的研究提出了一些建议及研究方向的展望。关键词：粒子群优化算法；参数；方差分析；最优解2010届信息与计算科学专业毕业设计IIIParticleswarmoptimizationalgorithmanditsparametersetSpeciality:InformationandComputingScienceStudent:RenKanAdvisor:XuXiaopingAbstractParticleswarmoptimizationisanemergingglobalbasedonswarmintelligenceheuristicsearchalgorithm,particleswarmoptimizationalgorithmcompetitionandcollaborationbetweenparticlestoachieveincomplexsearchspacetofindtheglobaloptimum.Ithaseasytounderstand,easytoachieve,thecharacteristicsofstrongglobalsearchability,andhasneverwidefieldofscienceandengineeringconcern,hasbecomethefastestgrowingoneoftheintelligentoptimizationalgorithms.Thispaperintroducestheparticleswarmoptimizationbasicprinciples,andanalyzesitsfeatures.Paperaroundtheparticleswarmoptimizationprinciples,characteristics,parameterssettingsandapplicationstoconductathoroughreview,focusingonasinglefactoranalysisofvariance,analysisoftheparticleswarmoptimizationalgorithmintheinertiaweight,accelerationfactorsettingthebasicpropertiesofthealgorithmtheimpactoftheexperienceofthealgorithmgivenparametersetting.Finally,itsfutureresearchedandprospectsareproposed.Keyword：Particleswarmoptimization;Parameter;Varianceanalysis;Optimalsolution任侃：粒子群优化算法及其参数设置IV目录摘要..................................................................................................................................IIAbstract.............................................................................................................................III1.引言..................................................................................................................................11.1研究背景和课题意义..........................................................................................11.2参数的影响..........................................................................................................11.3应用领域..............................................................................................................21.4电子资源..............................................................................................................21.5主要工作..............................................................................................................22.基本粒子群算法..............................................................................................................32.1粒子群算法思想的起源......................................................................................32.2算法原理..............................................................................................................42.3基本粒子群算法流程..........................................................................................52.4特点......................................................................................................................62.5带惯性权重的粒子群算法..................................................................................72.7粒子群算法的研究现状......................................................................................83.粒子群优化算法的改进策略..........................................................................................93.1粒子群初始化......................................................................................................93.2邻域拓扑..............................................................................................................93.3混合策略............................................................................................................124.参数设置.......................................................................................................................144.1对参数的仿真研究............................................................................................144.2测试仿真函数....................................................................................................154.3应用单因子方差分析参数对结果影响............................................................334.4对参数的理论分析............................................................................................345结论与展望....................................................................................................................39致谢...................................................................................................................................43附录...................................................................................................................................442010届信息与计算科学专业毕业设计11.引言1.1研究背景和课题意义“人工生命”是来研究具有某些生命基本特征的人工系统。人工生命包括两方面的内容：1、研究如何利用计算技术研究生物现象。2、研究如何利用生物技术研究计算问题。现在已经有很多源于生物现象的计算技巧。例如，人工神经网络是简化的大脑模型。遗传算法是模拟基因进化过程的。现在我们讨论另一种生物系统-社会系统。也可称做“群智能”(swarmintelligence)。这些模拟系统利用局部信息从而可能产生不可预测的群体行为。粒子群优化算法(PSO)也是起源对简单社会系统的模拟。最初设想是模拟鸟群觅食的过程。但后来发现PSO是一种很好的优化工具。优