基于rbf神经网络在玻璃窑炉温度控制上研究

《基于rbf神经网络在玻璃窑炉温度控制上研究》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、东北大学硕士学位论文AbstractResearchontheControloftheTemperatureofGlassFurnaceBasedonRBFNeuralNetworkAbstractThisthesistakesglassfumac宅asresearchobject，whichdiscussestheapplicationofintelligentcontroltononlineartime-delaycomplicatedsystemintheindustrialprocess．Thetemperatureofglas

2、sfurnaceisthemainparameteringlassproductionprocess，Maintainingtemperatureonitssetpointisthekeyfactorforassuringhigllqualityofglassproducts．Thethesisfirstlyintroducesthestructure，technologicalprocessandtemperaturecontrolrequirementofglassfurnace，andthenpresentsadynamiccha

3、racteristicanalysisofit．Thecontrolledobjectisaself-balanceprocesswithfeaturesoflargetimelagandinertia；alltheparametersarechangedwiththetemperatureandoperatingmodeoffurnace．Atpresent，themainlymethodusedinthetemperaturecontrolofglassfurnaceisPIDcontr01．AlthoughPIDcontrol伽a

4、chievesatisfiedresultinsmallrangeneartemperaturesetpoint，itcan’tcontroltemperaturetoitssetpointrapidlyandsmoothlywhentheworkconditionchangebyawidemarginduetotheinfluenceoftimelagandparametersvarying．SoPIDcontrolissubjectedtocomparativelybiglimitation．FurtherlyIintroduceS

5、imthpredictionmethod，fuzzycontrol，adaptivecontrolmethod．Fromtheresultsofsimulation,wecanknowthatintelligentmethodscansolvethetime-delaycharacteristicofglassfurnace．ImainlyintroduceRBFneuralnetworkcontrolmethodincludingPIDcontrolmethodbasedonRBFneuralnetwork,model-referen

6、ceadaptivecontrolofPIDbasedonRBFneuralnetwork,andinternalcontrolmethod,Ialsodosomeresearchonthebigtime—delaycharacteristicofglassfurnace．Usingthedatassampledatthespot,includingtheflowofthefuel，theflowoftheair,thetemperatureoftheglassfurnace，IidentifytheglassfurnaceusingR

7、BFneuralnetwork,andcontrolitusingPIDbasedonRBFneuralnetwork，alsocomparethePIDbasedonRBFneuralnetworkandPIDoftradition，concludingthattheformaerhasbettercapability．Theresultofsimulationindicatesthattheself-adaptivePIDcontrolmethodbasedonneuralnetworkwhichthethesisadoptshas

8、bettereffectsandthecontrollerpossessgoodabilityofregulation，self-learningandrobustness．Attheendofthisth