應(yīng)用人工神經(jīng)網(wǎng)絡(luò)的蒸發(fā)器傳熱性能仿真研究.pdf

1、東南大學(xué)碩士學(xué)位論文應(yīng)用人工神經(jīng)網(wǎng)絡(luò)的蒸發(fā)器傳熱性能仿真研究姓名：張鋒申請學(xué)位級別：碩士專業(yè)：制冷與低溫工程指導(dǎo)教師：虞維平2001.3.1AbstractItisdifficulttoemulatetheevaporatoroftheminiair—conditioningdevicepreciselybecauseitscomplicated，time—variedandnonlinearmoreoveritalsoinvolves

2、sometheoreticalquestionsthathavenotbeenexplainedexactlysuchastwo—phasedstream’10utilizetheimitation，itsnecessarytochoosethesimplestmodelsandalgoritlun，andlnakesurethatthemodelscalldepicttherealisticobjectstrictlyatthesam

3、etimeClassicalnumericalemulativemethodologyhasputonmanyattempts，andhasmadequiteafewprogresses，alsoencountermanydifficultiesWeexpecttosetofffromthescientificcrossingtofindanotherwayinthisfieldNeuralnetworktheoryofartifici

4、alintelligencecanrealizemanycomplexfunctionrelationsthroughsomerelationsthataresilnplerSOitisfitfordealingwiththeproblemsthatneedconsideringmanyfactorsandconditions，andareinexactorvagueIhispassage’smainideaistoillll’oduc

5、eneuralnetworktheoryintotheemulationoftheevaporator’SheattransferalldmakesolnefactorsthatareofgreatcomplexandexperientialevenuncertainintiletraditionaltheoreticmodelscomeundertheneuralnetworkmodelsSOastofoundallavailable

6、wayfromqualitativeemulationtoquantitativecalculationTheintroductionofartificialintelligencereplacestheidiomaticmethodthatregulatestheemulativepatternartificiallyanddecreasetheblindnessofartificialregulationWeexpecttoimpr