資源描述：
《不確定性優(yōu)化方法研究及其在車身結構耐撞性優(yōu)化中應用》由會員上傳分享，免費在線閱讀，更多相關內(nèi)容在學術論文-天天文庫。
1、AbstractTheuncertainty．basedoptimizationmethodiscrucialinpromotingthequalityoftheproducts．Asplentyofdeepresearchesontheuncertainty-basedoptimizationmethodrecently，thesurrogate—baseduncertaintyoptimizationmethodisgraduallVbecomlngpopularandisfaVoredbymorean
2、dmoredesignersduetoitssimplityandhighefficiency．NeVertheless，relativeresearchesoninvestigatingthismethodarequiteinsumcient；therefore，toconductsuchresearchesismeaningfula11dvaluable．ThecrashworthinessoptimizationdesignofCafbodyisoneofthekeyderectionsofVehlc
3、lecrashlngsatetyresearchareas，whichisacombinationoftheoDtimizationmethodandthecrashworthinessdesignengineeringproblem．However，theeverpublishedresearchesmainlyfocusonimprovingthecrashworthinessbehavioursthroughadeterminedoptimizationprocess；thus，theintroduc
4、tionofuncertaintvoptlmizationmethodinthisareaisofcrucialmeaningf．u1．Basedontheabovementlonedstatements，themainresearchis“stedasthefollowthreesections．Sec．1．Basedonthe109icanalysis，thedefectsoftheconventionalsurrogatebaseduncertaIntyoptlmlzationmethodarefig
5、uredoutandtheextremalmappingmethod1sorl91natedtoso】Vethjsproblem．Onbasisofthis，theDOEreconstructionbasedsurrogatereconstructionmethodisoriginated，whichisappliedtoconstructa11theneededf-unctionsintheneⅥ，lyorigjnateduncertaintyoptimizationprocess．F。urthermor
6、e，thestatisticaltheoryisemployedheretoinvestigatetheprecisionteatureofthereconstructedDOE．Additionally，thebenchmarksareemployedtoValidatetheresonablenessofthenewmethodandtostudyandcomparetheprecisioncharacteristicsof4kindsorrogatesintheuncertaintyoptimizat
7、ionprocess，whichisalmedtoproVidedesinerswitharef．erenceinselectionsurrogatetVpe．Sec．2．ToValidatethepracticalabiltyofthenewlyoriginatedmethod，afbam-filledtaperedthinwalledstructuresisemployedherein，andboththedetermisticoptlmlzatIonmethodanduncertaintyoptimi
8、zationmethodisappliedtomaximizeitscrashworthlness1ndicators．TheresultindicatesthattheuncertaintVbasedoptlmlzatlonhasworsecrashworthinessbehavioursbutmorereliableconstraintsandmorerobustobjectiVe．Besides，thene
