資源描述：
《基于微粒群算法的多目標(biāo)優(yōu)化問(wèn)題研究》由會(huì)員上傳分享，免費(fèi)在線閱讀，更多相關(guān)內(nèi)容在學(xué)術(shù)論文-天天文庫(kù)。
1、中南民族大學(xué)碩士學(xué)位論文基于微粒群算法的多目標(biāo)優(yōu)化問(wèn)題研究姓名：楊勛申請(qǐng)學(xué)位級(jí)別：碩士專業(yè)：計(jì)算機(jī)應(yīng)用技術(shù)指導(dǎo)教師：王江晴20080528基于微粒群算法的多目標(biāo)優(yōu)化問(wèn)題研究策方案；在實(shí)驗(yàn)和理論分析的基礎(chǔ)上我們給出了Pareto-ε動(dòng)態(tài)策略的評(píng)價(jià)。4、基于PSO算法和Pareto-ε優(yōu)勝關(guān)系提出了一種新的PεPSO算法框架；基于面向?qū)ο罄碚撎岢隽艘环N相應(yīng)的數(shù)據(jù)結(jié)構(gòu)，提高了算法實(shí)現(xiàn)的通用性、復(fù)用性以及兼容性；選用了經(jīng)典測(cè)試函數(shù)集中的部分ZDT系列函數(shù)進(jìn)行了測(cè)試，實(shí)驗(yàn)顯示PεPSO算法是有效的。5、在采用動(dòng)
2、態(tài)調(diào)整ε的策略后，通過(guò)動(dòng)態(tài)調(diào)整ε的值，使算法開始時(shí)快速向Pareto真實(shí)前沿逼近，最終讓?duì)旁谒惴ㄟ\(yùn)行過(guò)程中逐步回歸為0，從而更好的逼近真實(shí)Pareto前沿，不受ε的影響。既可以提高算法的搜索和收斂速度，又可以消除ε值對(duì)最終解的質(zhì)量的影響。關(guān)鍵詞：計(jì)算智能；進(jìn)化算法；微粒群算法；多目標(biāo)優(yōu)化；Pareto-εII中南民族大學(xué)碩士學(xué)位論文AbstractItisanindispensablecapabilityforthemoderndecisionandassistancesystemtoprovided
3、ecision-makerswithscientific,properandtimelydicisionschemes.Sincemostoftheactualdecision-targetsaremulti-objective,theresearchonMulti-objectiveOptimizationProblem(MOP)hasgainedmoreandmoreattention.Forthefactthatthesub-objectivesinMOParecontradictorytoea
4、chotherandtheyhavenounifiedmeasurestandards,ithasbecomeaprincipalconcerntodefinitetheoptimalsolutionofMOPwhensolutingMOP.Thedefinitionofnon-dominancedsolution,basedontheideaofchoosingPareto,isbecomingincreasinglypeople’sgeneralconsensus.TheoptimalPareto
5、solutionofMOP,however,isnotanexclusivesolution.Sometimesthereareevennumerousones.Inthiscase,theobtainedsolutionsdonotfacilitatebuttroublethedicision-makers.Furthermore,ittakesalongtimetoobtaintheoptimalsolutions.Soitisratherimportanttoprovideproperandfe
6、asiblesolutionschemesrapidlyforthedecisionmakers.ParticleSwarmOptimization(PSO)isakindofswarmintelligentalgorithmwhichhasbeendevelopedinrecentyears.Itisanewintelligentsearchalgorithm.Thealgorithmutilizestheeffectiveinformation,whichissharedbyeverypartic
7、leinthepopulationfromitspastexperiencesandotherparticles’experiences,tosearchtheoptimalsolutionssynergically.Atpresent,thealgorithmresearchbasedonPSOhasbeenattachedmoreandmoreimportanceinthefieldofmulti-objectiveoptimizationanditisevenahottopicintherese
8、arch.TherelevantPareto-εconceptsareproposedinthepapergroundedonthedefinitionofParetooptimalsolutions.Bymeansofanalysisandexperiments,itcanbeprovedthattheprocessofoptimizationinsolutiingMOPisimprovedfortheusageofPareto-εconcepts,t
