资源描述:
《Genetic_Algorithm_overview》由会员上传分享,免费在线阅读,更多相关内容在学术论文-天天文库。
1、EleventhInternationalIBPSAConferenceGlasgow,ScotlandJuly27-30,2009AGENETICALGORITHMFOROPTIMIZATIONOFBUILDINGENVELOPEANDHVACSYSTEMPARAMETERSMattiPalonen,AlaHasan,KaiSirenHVACTechnology,HelsinkiUniversityofTechnology,P.OBox4400,FIN-02015HUT,FinlandThesuccessoftheoptimizat
2、ionisstronglyaffectedABSTRACTbythepropertiesoftheproblem,formulationoftheobjectivefunctionandbytheselectionofanTheaimofthispaperistodescribethefeaturesofappropriateoptimizationalgorithm.aGeneticAlgorithm(GA)developedtosolvesimulation-basedoptimizationproblemsfortheAgene
3、ticalgorithm(GA)isasearchtechniqueusedoptimaldesignofbuildingparameters.ThisGAincomputingtofindsolutionstooptimizationhasbeendevelopedusingguidelinesfromtopproblems.Geneticalgorithmscanbecategorizedasresearchsinthefieldofevolutionarycomputation.Itmetaheuristicswithgloba
4、lperspective.GeneticismostlybasedonNSGA-IIandOmni-optimizer.Italgorithmsareaparticularclassofevolutionarycanbeusedforsingleandmulti-objectivealgorithmsthatusetechniquesinspiredbyoptimizationproblemswithandwithoutconstrains.evolutionarybiologysuchasinheritance,mutation,B
5、othdiscreteandcontinuousvariablescanbeselection,andcrossover.Geneticalgorithmsarehandled.implementedasacomputersimulationinwhichapopulationofabstractrepresentationsofcandidateReal-worldoptimizationproblemsinthefieldofsolutionstoanoptimizationproblemevolvestowardbuilding
6、performancesimulationarecarriedouttobettersolutions.Traditionally,solutionsareverifytheperformanceofthedevelopedGA.Resultsrepresentedinbinaryasstringsof0sand1s,butforasingle-objectiveoptimizationproblemareotherencodingsarealsopossible.Theevolutionpresented,wheretheaimis
7、minimizationoflifecycleusuallystartsfromapopulationofrandomlycostofadetachedhouse.Besidesdiversesetsofgeneratedindividualsandhappensingenerations.Innon-dominatedsolutionsresultsforamulti-objectiveeachgeneration,thefitnessofeveryindividualinthebuildingdesignproblemareals
8、opresented.populationisevaluated,multipleindividualsarestochasticallyselectedfromthecurrentpopulation(basedont