资源描述：
《sensor based on neural networks in rare-earth countercurrent extraction process》由会员上传分享，免费在线阅读，更多相关内容在教育资源-天天文库。
1、sensorBasedonNeuralNetworksinRare-earthCountercurrentExtractionProcessVol_32.No.4ACTAAUT0MATICASINICAJuly,2006ComponentContentSoft-sensorBasedonNeuralNetworksinRare.earthCountercurrentExtractionProcess1)YANGHuiICHAITian—YlOu(SchoolofElectronicsandElectric
2、alEngineering,EastChinaJiaotongUniversity,Nanchang330013)(ResearchCenterofAutomation,NortheasternUniversity,Shenyang110004)(E-mail:yhshuo@263.net)AbstractThroughtfusionofthemechanismmodelingandtheneuralnetworksmodeling.acompo-nentcontentsoft—sensor.whichi
3、scomposedoftheequilibriumcalculationmodelformulti-componentrareearthextractionandtheerrorcompensationmodeloffuzzysystem,isproposedtosolvetheprob-lemthatthecomponentcontentincountercurrentrare-earthextractionprocessishardlymeasuredon-line.Anindustryexperim
4、entintheextractionyprocessbyHABusingthishybridsoft-sensorprovesitseffectiveness.KeywordsRare-earth,countercurrentextraction,soft—sensor,equilibriumcalculationmodel,neuralnetworks1IntroductionChinahasthemostabundantrare—earthresourceintheworldl.Buttheextra
5、ctionprocessautomationisstillinthestagethatcomponentcontentismeasuredoff-line,andtheprocessiscontrolledbyexperienceandparametersareregulatedbyhands.Thissituationleadstolowefficientproductionrate,highresourceconsumptionandunstableproductionquality,andlimit
6、stherare—earthindustrydevelopmentt1.T0implementautomationintherare—earthextractionprocess.theon—linecomponentcontentmeasuringmustbeachievedatfirst.MaincomponentcontentmeasuringmethodsincludetheX—rayabsorbspectrometer,X—rayenergyspectrometer,Iso—topicX-flu
7、orescenceenergyspectrometer,on-linespectrophotometers.etc[,.Theseinstrumentshavenotbeenwidelyusedinindustrybecauseofhighcost,lowreliabilityincontinuousoperation,complexmaintenance,delayinmeasurementandlowaccuracy.Thesoftsensormethod,ontheotherhand,provide
8、sanewwaytoon-linemeasurecomponentcontentintherare-earthcountercurrentextractionprocess.Wlewillfurtherourresearchin[4]byfusionofthemechanismmodelingandtheintelligentmodelingandproposeahybridsoft—sensoroftherareearthc
