Interactive genetic algorithms with individual’s fuzzy fitness

《Interactive genetic algorithms with individual’s fuzzy fitness》由会员上传分享，免费在线阅读，更多相关内容在行业资料-天天文库。

1、ComputersinHumanBehavior27(2011)1482–1492ContentslistsavailableatScienceDirectComputersinHumanBehaviorjournalhomepage:www.elsevier.com/locate/comphumbehInteractivegeneticalgorithmswithindividual’sfuzzyfitnessDun-weiGong⇑,JieYuan,Xiao-yanSunSchoolofInformationandElectricalEngineering,ChinaUniv

2、ersityofMiningandTechnology,Xuzhou,ChinaarticleinfoabstractArticlehistory:InteractivegeneticalgorithmsareeffectivemethodstosolveanoptimizationproblemwithimplicitorAvailableonline30October2010fuzzyindices,andhavebeensuccessfullyappliedtomanyreal-worldoptimizationproblemsinrecentyears.Intradit

3、ionalinteractivegeneticalgorithms,manyresearchersadoptanaccuratenumbertoKeywords:expressanindividual’sfitnessassignedbyauser.ButitisdifficultforthisexpressiontoreasonablyreflectOptimizationauser’sfuzzyandgradualcognitivetoanindividual.WepresentaninteractivegeneticalgorithmwithanGeneticalgorithm

4、sindividual’sfuzzyfitnessinthispaper.Firstly,weadoptafuzzynumberdescribedwithaGaussianmem-Individual’sfitnessbershipfunctiontoexpressanindividual’sfitness.Then,inordertocomparedifferentindividuals,weFuzzynumbergenerateafitnessintervalbasedona-cutset,andobtaintheprobabilityofindividualdominanceby

5、Fashiondesignuseoftheprobabilityofintervaldominance.Finally,wedeterminethesuperiorindividualintournamentselectionwithsizetwobasedontheprobabilityofindividualdominance,andperformthesubsequentevolutions.Weapplytheproposedalgorithmtoafashionevolutionarydesignsystem,atypicaloptimi-zationproblemw

6、ithanimplicitindex,andcompareitwithtwointeractivegeneticalgorithms,i.e.,aninteractivegeneticalgorithmwithanindividual’saccuratefitnessandaninteractivegeneticalgorithmwithanindividual’sintervalfitness.Theexperimentalresultsshowthattheproposedalgorithmisadvan-tageousinalleviatinguserfatigueandlo

7、okingforuser’ssatisfactoryindividuals.Ó2010ElsevierLtd.Allrightsreserved.1.Introductionfitnessaremorelikelytobeselectedtogenerateindividualsinthenextgeneration.AnewgenerationofindividualsisgeneratedOptimizationproblemsareverycommoninreal-worldapplic