资源描述：
《A comparative study of combined feedforward feedback model predictive control for nonlinear systems.pdf》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、AComparativeStudyofCombinedFeedforward/FeedbackModelPredictiveControlforNonlinearSystemsJi-ZhengChu,2Shi-ShangJang1*andYu-NanChen11ChemicalEngineeringDepartmentNationalTsing-HuaUniversity,Hsin-Chu30043Taiwan2DepartmentofAutomation,BeijingUniversityofChemicalTechnology,Beijing.ight
2、erperformancespecificationsfromworldwidecompetitionandeverincreasingconstraintsfromenvironmentalandsafetyModelpredictivecontrol(MPC)providesanaturalTconsiderationsgivethepracticaldrivingforceforthedevelopmentframeworktorealizefeedforwardandfeedbackcontrolofadvancedcontroltechnology
3、(FindeisenandAllgöwer,2001).Largefornonlinearsystemswheretheeffectofdisturbances(DVs)cannotbeseparatedfromthatofmanipulatedamountsofindustrialpracticeandacademicresearchhavemademodelvariables(MVs).Thisstudyexaminestheperformancepredictivecontrol(MPC)thedefactostandardalgorithmfora
4、dvancedofMPCwithmeasuredDVsaspartialinputsofthecontrolintheprocessindustry(Nikolaou,2001).MPCisageneralmodelused,whichistermedascombinedfeedforward/andmathematicallyfeasibleschemetointegrateourknowledgeaboutfeedbackMPC(CMPC)incontrasttoconventionalMPCatarget,processintocontrollerd
5、esignandoperation,whichallowsusingamodelwithoutinputofanymeasuredDV.InthesimulationofapHprocess,wedemonstratetheflexibleandefficientexploitationofourunderstandingofatarget,andclearsuperiorityofCMPCoverMPC.Intheexperi-thusoptimalperformanceofasystemundervariousconstraints.Threementwi
6、thabench-scaleethanolandwaterdistillationkeyfactorsresponsibleforthegreatsuccessofMPCaretheincorpora-column,CMPCandMPCusingartificialneuralnetworktionofaprocessmodel,analgorithmconsideringplantbehaviourover(ANN)modelsareappliedtothedualtemperatureafuturehorizonintime,andexplicittre
7、atmentofconstraints(Qinandcontrolproblem.Externalrecurrentneuralnetworks(ERNs)withandwithoutameasuredDV(feedrateofBadgwell,1999),withthemodelbeingforemostandfundamental.thecolumn)astheirpartialinputarebuiltandemployedForanaccurateandreliablepredictionofamodel,sufficientinforma-inth
8、eexperiment,witharesultthatinclus
