簡(jiǎn)介：分類號(hào)TP391學(xué)號(hào)13030020UDC密級(jí)公開工學(xué)碩士學(xué)位論文偵察型無(wú)人機(jī)監(jiān)督控制中的操作員自主代理偵察型無(wú)人機(jī)監(jiān)督控制中的操作員自主代理技術(shù)研究技術(shù)研究碩士生姓名簡(jiǎn)立軒學(xué)科專業(yè)控制科學(xué)與工程研究方向模式識(shí)別與智能系統(tǒng)指導(dǎo)教師沈林成教授國(guó)防科學(xué)技術(shù)大學(xué)研究生院國(guó)防科學(xué)技術(shù)大學(xué)研究生院二〇一五年十一月二〇一五年十一月偵察型無(wú)人機(jī)監(jiān)督控制中的操作員自主代理技術(shù)研究國(guó)防科學(xué)技術(shù)大學(xué)研究生院

簡(jiǎn)介：中圖分類號(hào)TH113O327論文編號(hào)102870515S153學(xué)科分類號(hào)080203碩士學(xué)位論文基于代理模型的L型雙加筋板結(jié)構(gòu)動(dòng)力學(xué)優(yōu)化設(shè)計(jì)研究生姓名展銘學(xué)科、專業(yè)機(jī)械設(shè)計(jì)及理論研究方向結(jié)構(gòu)動(dòng)力學(xué)指導(dǎo)教師郭勤濤副教授南京航空航天大學(xué)研究生院機(jī)電學(xué)院二О一五年三月NANJINGUNIVERSITYOFAERONAUTICSANDASTRONAUTICSTHEGRADUATESCHOOLCOLLEGEOFMECHANICALANDELECTRICALENGINEERINGSTRUCTURALDYNAMICSOPTIMIZATIONDESIGNBASEDONSURROGATEMODELOFPANELWITHTWOLSTIFFENERSATHESISINMACHINEDESIGNANDTHEORYBYZHANMINGADVISEDBYPROFGUOQINTAOSUBMITTEDINPARTIALFULFILLMENTOFTHEREQUIREMENTSFORTHEDEGREEOFMASTEROFENGINEERINGMARCH,2015

簡(jiǎn)介：南京航空航天大學(xué)碩士學(xué)位論文機(jī)翼氣動(dòng)力雙向代理模型設(shè)計(jì)姓名楊華申請(qǐng)學(xué)位級(jí)別碩士專業(yè)飛行器設(shè)計(jì)指導(dǎo)教師姚衛(wèi)星20070101機(jī)翼氣動(dòng)力雙向代理模型設(shè)計(jì)－II－ABSTRACTSURROGATEMODELISONEOFKEYELEMENTSOFMULTIDISCIPLINARYDESIGNOPTIMIZATIONMDOITISUSEDTOPROVIDEAPPROXIMATEOBJECTIVEVALUEBYREPLACINGTIMECONSUMINGSIMULATIONEVALUATIONSEXPERIMENTSITCANBEEASILYINTEGRATEDINTOAMDOFRAMEWKWHENDEALINGWITHAWINGAERODYNAMICSTRUCTUREOPTIMIZATIONDESIGNSURROGATEMODELOFWINGLIFTDISTRIBUTIONISNEEDEDTHESURROGATEMODELSOFWINGLIFTDISTRIBUTIONPUBLISHEDINTHERECENTWEREREVIEWEDINTHISTHESISANEWAPPROACHTOOBTAIN2DSURROGATEMODELOFWINGLIFTDISTRIBUTIONWASPROPOSEDINWHICHTHERADIALBASISFUNCTIONSHAPEFUNCTIONWEREINDUCEDTHESHAPEFUNCTIONWASUSEDTOTRANSFMEDARBITRARYPLANOFAWINGINTORECTANGLEALSOTHEPROCESSTOGETTHESURROGATEMODELWASINTRODUCEDTHOUGHLYTHEWINGLEFTOFACRUISEMISSILEWASCOMPUTEDSIMULATEDBYTHEPROPOSED2DSURROGATEMODELTHEPREDICTIONRESULTSDEMONSTRATEDTHEFEASIBILITYEFFECTIVENESSOFTHEPROPOSEDMETHODKEYWDSMULTIDISCIPLINARYDESIGNOPTIMIZATION2DSURROGATEMODELRADIALBASISFUNCTIONGEOMETRICALTERNATIONSHAPEFUNCTION

簡(jiǎn)介：分類號(hào)V4211學(xué)號(hào)13013174ＵＤＣ密級(jí)公開工學(xué)碩士學(xué)位論文基于代理模型的飛行器氣動(dòng)外形快速優(yōu)基于代理模型的飛行器氣動(dòng)外形快速優(yōu)化設(shè)計(jì)化設(shè)計(jì)碩士生姓名張龍龍學(xué)科專業(yè)航空宇航科學(xué)與技術(shù)研究方向飛行器氣動(dòng)外形設(shè)計(jì)指導(dǎo)教師楊濤教授國(guó)防科學(xué)技術(shù)大學(xué)研究生院國(guó)防科學(xué)技術(shù)大學(xué)研究生院二〇二〇一五年五年十一月一月RAPIDAERODYNAMICSHAPEOPTIMIZATIONDESIGNOFAIRCRAFTBASEDONSURROGATEMODELCIDATEZHANGLONGLONGADVISYANGTAOATHESISSUBMITTEDINPARTIALFULFILLMENTOFTHEREQUIREMENTSFTHEDEGREEOFMASTEROFENGINEERINGINAEROSPACESCIENCETECHNOLOGYGRADUATESCHOOLOFNATIONALUNIVERSITYOFDEFENSETECHNOLOGYCHANGSHA，HUNAN，PRCHINA（NOVEMBER，2015）

簡(jiǎn)介：分類號(hào)V421學(xué)號(hào)10019053ＵＤＣ密級(jí)博士生姓名公開工學(xué)博士學(xué)位論文基于代理模型基于代理模型和MOEAD的飛行器的飛行器氣動(dòng)氣動(dòng)外形優(yōu)化設(shè)計(jì)研究外形優(yōu)化設(shè)計(jì)研究馬洋學(xué)科專業(yè)航空宇航科學(xué)與技術(shù)研究方向飛行器氣動(dòng)外形優(yōu)化設(shè)計(jì)指導(dǎo)教師楊濤教授國(guó)防科學(xué)技術(shù)大學(xué)研究生院國(guó)防科學(xué)技術(shù)大學(xué)研究生院二〇一二〇一四年十二月RESEARCHONTHEAERODYNAMICSHAPEOPTIMIZATIONOFVEHICLEBASEDONSURROGATEMODELMOEADCIDATEMAYANGSUPERVISPROFYANGTAOADISSERTATIONSUBMITTEDINPARTIALFULFILLMENTOFTHEREQUIREMENTSFTHEDEGREEOFDOCTOFENGINEERINGINAEROSPACESCIENCETECHNOLOGYGRADUATESCHOOLOFNATIONALUNIVERSITYOFDEFENSETECHNOLOGYCHANGSHA，HUNAN，PRCHINADECEMBER，2014

簡(jiǎn)介：南京航空航天大學(xué)碩士學(xué)位論文基于代理模型的加筋板結(jié)構(gòu)布局優(yōu)化設(shè)計(jì)姓名張柱國(guó)申請(qǐng)學(xué)位級(jí)別碩士專業(yè)飛行器設(shè)計(jì)指導(dǎo)教師姚衛(wèi)星20080301基于代理模型的加筋板結(jié)構(gòu)布局優(yōu)化設(shè)計(jì)IIABSTRACTSTIFFENEDPANELSTRUCTUREISWIDELYUSEDINTHEENGINEERINGFIELDSOFAERONAUTICSANDASTRONAUTICSASTHEYCANSAVETHESTRUCTURALMATERIALS,REDUCETHESTRUCTURALWEIGHT,ANDRAISETHESTRUCTURALEFFICIENCYSTIFFENEDPANELSTRUCTUREMAYRESULTINBUCKLINGOREVENCOLLAPSEWHENUNDERTAKINGCOMPRESSIVELOADORSHEARLOADORTHEBOTHBECAUSEOFTHENUMEROUSFAILUREMODESANDOPTIMUMDESIGNVARIABLES,MODELING,ANALYSISANDOPTIMIZATIONARECOMPLICATEDSOTHERESEARCHOFTHEOPTIMUMDESIGNOFSTIFFENEDPANELISVERYIMPORTANTINTHISTHESIS,ANOPTIMUMDESIGNMETHODOFMETALLICSTIFFENEDPANELWASBUILTUPBASEDONSURROGATEMODELITCANBEAPPLIEDTOTHELAYOUTOPTIMIZATIONOFMETALLICSTIFFENEDPANELSTRUCTUREINTHISMETHOD,THEINITIALSAMPLESWEREFIRSTLYCREATEDBYDESIGNOFEXPERIMENTDOE,SECONDLYTHERESPONSESCORRESPONDINGTOTHEINITIALSAMPLES,INCLUDINGTHEWEIGHTANDTHEBUCKLINGFACTOR,WERECOMPUTEDBYFEA,THENTHEINITIALKRIGINGMODELWASDERIVEDTHROUGHTHESAMPLES,LATETHESAMPLESWEREUPDATEDIFTHEKRIGINGMODELWASNOTSATISIFIED,ANDANEWKRIGINGMODELWASREDERIVEDWITHTHEUPDATEDSAMPLESFINALLYTHEOPTIMALSOLUTIONWASOBTAINEDWITHTRADEOFFMETHODANDGENETICALGORITHMSGABASEDONTHEKRIGINGMODELBASEDONTHISMETHOD,THEPOSTBUCKLINGOPTIMIZATIONPROBLEMOFMETALLICSTIFFENEDPANELWASALSOACCOMPLISHEDBYINTRODUCINGPOSTBUCKLINGANALYSISBESIDES,ASTACKINGSEQUENCEOPTIMIZATIONMETHODOFSYMMETRICALANDBALANCEDCOMPOSITELAMINATEWASPROPOSEDINTHISTHESISBASEDONTHEABOVEPRESENTEDOPTIMIZATIONMETHODANDBENDINGLAMINATIONPARAMETERANDTHENATWOLEVELOPTIMIZATIONALGORITHMWASDEVELOPEDALLTHEEXAMPLESDEMONSTRATETHATTHEYCANFINDOUTTHEOPTIMALSOLUTIONANDCANVERIFYTHEFEASIBILITYANDVALIDITYOFTHESEMETHODSPROPOSEDINTHISTHESISKEYWORDSSTIFFENEDPANEL,STRUCTURALOPTIMIZATION,LAYOUTOPTIMIZATION,SURROGATEMODEL,BUCKLING,POSTBUCKLING,STACKINGSEQUENCEOPTIMIZATION

簡(jiǎn)介：碩士學(xué)位論文基于改進(jìn)代理模型航天器結(jié)構(gòu)基于改進(jìn)代理模型航天器結(jié)構(gòu)可靠性可靠性優(yōu)化設(shè)計(jì)優(yōu)化設(shè)計(jì)RELIABILITYBASEDDESIGNOPTIMIZATIONOFSPACECRAFTSTRUCTUREBASEDONIMPROVEDSURROGATEMODEL作者姓名鞏萃穎學(xué)科、專業(yè)工程力學(xué)學(xué)號(hào)21403027指導(dǎo)教師李剛教授完成日期2017年6月大連理工大學(xué)DALIANUNIVERSITYOFTECHNOLOGY大連理工大學(xué)碩士學(xué)位論文I摘要加筋柱殼結(jié)構(gòu)和格柵結(jié)構(gòu)是航空航天領(lǐng)域中常見的結(jié)構(gòu)形式，常常應(yīng)用于大型運(yùn)載火箭的級(jí)間段、整流罩和燃料貯箱等部段。減重任務(wù)對(duì)于飛行器的設(shè)計(jì)尤為重要，另外，飛行器在載荷環(huán)境、計(jì)算模型簡(jiǎn)化、幾何制造公差、材料性能等方面均存在不確定性，基于可靠度的優(yōu)化可以盡可能地考慮這些不確定性，獲得兼?zhèn)浣?jīng)濟(jì)和安全的最優(yōu)方案。本文面向加筋柱殼和格柵等航天器結(jié)構(gòu)，建立了一個(gè)高效穩(wěn)定的基于改進(jìn)精度代理模型的可靠性優(yōu)化框架。它以結(jié)構(gòu)減重為目標(biāo)，考慮結(jié)構(gòu)在幾何制造公差、材料性能和載荷分布方面的隨機(jī)性，主要考慮軸力、彎矩、剪力和側(cè)向外壓工況下結(jié)構(gòu)屈曲失穩(wěn)的失效模式，對(duì)結(jié)構(gòu)的幾何尺寸和筋條數(shù)目進(jìn)行優(yōu)化設(shè)計(jì)。該優(yōu)化框架的主要內(nèi)容是，首先，按照靈敏度分析確定代理模型參變量、選擇代理模型類型、擴(kuò)大訓(xùn)練樣本集的設(shè)計(jì)空間、全局搜索優(yōu)化參數(shù)、放松局部設(shè)計(jì)域的精度要求的步驟建立改進(jìn)代理模型，獲取復(fù)雜載荷信息的統(tǒng)計(jì)規(guī)律以后，將載荷分布當(dāng)量正態(tài)化，然后，內(nèi)層采用自適應(yīng)混沌控制的改進(jìn)均值法獲取可靠度信息，傳遞到外層全局尋優(yōu)的演化算法進(jìn)行可靠度優(yōu)化，得到最優(yōu)方案后，重新在最優(yōu)點(diǎn)附近建立一個(gè)更高精度的局部代理模型，使用MONTECARLO方法對(duì)最優(yōu)方案的失效概率進(jìn)行驗(yàn)證。然后，本文使用該優(yōu)化框架實(shí)現(xiàn)了對(duì)航天飛行器中兩個(gè)典型構(gòu)型金屬正置正交加筋柱殼結(jié)構(gòu)和復(fù)合材料KAGOME格柵柱殼及錐殼結(jié)構(gòu)的可靠性優(yōu)化工作，驗(yàn)證了該優(yōu)化框架解決工程結(jié)構(gòu)可靠性優(yōu)化設(shè)計(jì)問(wèn)題的有效性及高效性，并探索得出通過(guò)改變幾何尺寸提高這兩類結(jié)構(gòu)承載效率和安全可靠性的普適性規(guī)律。另外，本文還提出兩種在可靠性優(yōu)化設(shè)計(jì)中考慮載荷隨機(jī)性的方式一種是按照載荷的真實(shí)分布把載荷允許值作為隨機(jī)變量帶入功能函數(shù)，另外一種是將載荷均值項(xiàng)與標(biāo)準(zhǔn)差項(xiàng)線性相加后作為一個(gè)確定值帶入功能函數(shù)。在本文的優(yōu)化框架下，這兩種方法均能在保證結(jié)構(gòu)安全可靠度要求的同時(shí)，獲得減重效果最好的方案，且效果均優(yōu)于基于15安全系數(shù)法的確定性優(yōu)化效果。關(guān)鍵詞加筋柱殼；KAGOME格柵；可靠性；優(yōu)化設(shè)計(jì)；代理模型

簡(jiǎn)介：中圖分類號(hào)V2113論文編號(hào)102870116S017學(xué)科分類號(hào)080103碩士學(xué)位論文基于KRIGING代理模型的氣動(dòng)外形優(yōu)化設(shè)計(jì)研究生姓名胡遠(yuǎn)根學(xué)科、專業(yè)流體力學(xué)研究方向計(jì)算流體力學(xué)指導(dǎo)教師夏健教授南京航空航天大學(xué)研究生院航空宇航學(xué)院二О一六年三月NANJINGUNIVERSITYOFAERONAUTICSASTRONAUTICSTHEGRADUATESCHOOLCOLLEGEOFAEROSPACEENGINEERINGAERODYNAMICSHAPEOPTIMIZATIONDESIGNBASEDONKRIGINGSURROGATEMODELATHESISINFLUIDMECHANICSBYHUYUANGENADVISEDBYPROFXIAJIANSUBMITTEDINPARTIALFULFILLMENTOFTHEREQUIREMENTSFTHEDEGREEOFMASTEROFENGINEERINGMARCH2016

簡(jiǎn)介：分類號(hào)學(xué)號(hào)M201571880學(xué)校代碼10487密級(jí)碩士學(xué)位論文基于多代理的充電站負(fù)荷定容基于多代理的充電站負(fù)荷定容仿真軟件的設(shè)計(jì)與實(shí)現(xiàn)仿真軟件的設(shè)計(jì)與實(shí)現(xiàn)學(xué)位申請(qǐng)人劉林夕學(xué)科專業(yè)電子與通信工程指導(dǎo)教師涂來(lái)副教授兼職導(dǎo)師徐振華高工答辯日期2017年5月25日

簡(jiǎn)介：點(diǎn)擊查看更多電子產(chǎn)品逆向供應(yīng)鏈代理方的激勵(lì)契約設(shè)計(jì).pdf精彩內(nèi)容。

簡(jiǎn)介：南京航空航天大學(xué)碩士學(xué)位論文多學(xué)科設(shè)計(jì)優(yōu)化代理模型技術(shù)的研究和應(yīng)用姓名穆雪峰申請(qǐng)學(xué)位級(jí)別碩士專業(yè)飛行器設(shè)計(jì)指導(dǎo)教師姚衛(wèi)星20040201墮室墮至墮墨盔蘭堡主堂篁堡苧ABSTRACTSURROGATEMODELIS912IMPORTANTCOMPONENTOFMULTIDISCIPLINEDESIGNOPTIMIZATIONMDOASURROGATEMODELISCREATEDWITHEXPERIMENTDESIGNMETHODSANDAPPROXIMATIONAPPROACHES，ANDITCANBEEASILYINTEGRATEDINTOAMDOFRAMEWORKTHISARTICLEDESCRIBESTHEDEVELOPMENTPROCESSOFSURROGATEMODELSANDINTRODUCESSOMEEXPERIMENTDESIGNMETHODSANDAPPROXIMATIONAPPROACHESTHATCANBEUSEDFORAMDOSURROGATEMODEL，THEYAIEFULLFACTORIALEXPERIMENTDESIGN，ORTHOGONALEXPERIMENTDESIGN，UNIFORMEXPERIMENTDESIGN，CENTRALCOMPOSITEEXPERIMENTDESIGN，ANDPOLYNOMIALRESPONSESURFACEMETHOD，KRIGINGMETHOD，RADIALBASISFUNCTIONMETHODANDARTIFICIALNEURALNETWORKATTHEENDOFTHISARTICLE，ASURROGATEMODELISCREATEDTODESCRIBETHE1IFTDISTRIBUTIONOFAWINGRESULTSHOWSTHATKRIGINGMETHODANDRBFMETHODARETWOUSEFULAPPROXIMATIONAPPROACHES，ESPECIALLYFORHIGHDIMENSIONPROBLEMINADDITION，THISARTICLECREATEDAGENERALIZEDPOLYNOMIALRESPONSESURFACEMODEL，ANDRESULTSHOWSTHATITCANBEALSOUSEDTODESCRIBETHELIFTDISTRIBUTIONKEYWORDSMULTIDISCIPLINEDESIGNOPTIMIZATION，SURROGATEMODEL，DESIGNOFEXPERIMENT，APPROXIMATIONAPPROACHES，POLYNOMIALRESPONSESURFACEMETHOD，KRIGINGMETHOD，RADIALBASISFUNCTIONMETHODN