预览加载中，请您耐心等待几秒...

交叉反向学习和同粒社会学习的粒子群优化算法.docx

预览
1/2
2/2

在线预览结束，喜欢就下载吧，查找使用更方便

下载提示 文本预览

如果您无法下载资料，请参考说明：

1、部分资料下载需要金币，请确保您的账户上有足够的金币

2、已购买过的文档，再次下载不重复扣费

3、资料包下载后请先用软件解压，在使用对应软件打开

交叉反向学习和同粒社会学习的粒子群优化算法 ParticleSwarmOptimization(PSO)isameta-heuristicalgorithminspiredbythesocialbehaviorofbirdflocksandfishschools.PSOhasbeenwidelyusedinoptimizationproblemsduetoitssimplicityandeffectivenessinfindingoptimalsolutions.However,likeothermeta-heuristicalgorithms,PSOhaslimitationsinsolvingcomplexoptimizationproblems. Toovercometheselimitations,severalmodificationshavebeenproposed,suchasCross-OverandOpposition-BasedLearning(COOBL)andSelf-OrganizingSocialPSO(SOS-PSO).COOBLintroducescrossoveroperatorstothePSOtoexchangeinformationbetweenparticles,whileSOS-PSOattemptstomimicthesocialbehaviorofparticlesinaswarmtoimproveconvergencespeedandaccuracy.Inthispaper,weproposeahybridalgorithmthatcombinesCOOBLandSOS-PSOtosolveoptimizationproblems. Thehybridalgorithm,calledCross-OrganizingSocialPSO(COS-PSO),adoptstwodifferentlearningmechanisms:cross-overandsame-kindsociallearning.InCOOBL,crossoveroperatorsareintroducedtogeneratenewparticlesbyexchanginginformationbetweenparticles.ThecrossoveroperatorsusedinCOS-PSOaretwo-pointcrossoveranduniformcrossover.Intwo-pointcrossover,twoparentsareselectedrandomly,andarandomsubsetoftheirpositionsisswapped.Inuniformcrossover,eachpositionintheoffspringisrandomlyselectedfromthecorrespondingpositionofeitherparent. Ontheotherhand,same-kindsociallearningisasophisticatedsociallearningmechanisminspiredbythebehaviorofparticlesinaswarm,inwhicheachparticlelearnsfrombothitslocalbestpositionandtheglobalbestposition.InSOS-PSO,eachparticlehasamemoryofitspreviouspositionsandvelocitiesandthepositionsandvelocitiesofitsneighbors.InCOS-PSO,same-kindsociallearningisintroducedtoenableparticlestolearnfromtheirpeersintheswarm. ToevaluatetheeffectivenessofCOS-PSO,wecompareitwithPSO,COOBL,andSOS-PSOonasetofbenchmarkfunctions.TheexperimentalresultsshowthatCOS-PSOoutperformsPSO,COOBL,andSOS-PSOintermsofconvergencespeedandaccuracy.Thecomparisonmetricsincludemeanabsoluteerror(MAE),rootmeansquareerror(RMSE),andnumberoffunctionevaluations(FEs)requiredtoreachaspeci