DOC

Simulation Study on Heat Value Control System of Natural Gas Used for Color TV Tubes Production

By Lauren Hicks,2014-08-05 07:55
11 views 0
Simulation Study on Heat Value Control System of Natural Gas Used for Color TV Tubes Productionon,of,Study,Heat,Value,Gas,Used,for,Color,gas

    Simulation Study on Heat Value Control System of Natural Gas Used for Color TV

    Tubes Production

    JournalofBeljingInstituteofTechnology,2006,Vo1.15,No.3

    SimulationStudyonHeatValueControlSystemof

    NaturalGasUsedforColorTVTubesProduction

    ZHENGBin(郑斌)

    (PowerFactoryofHenanAncaiGroup,Anyang455000,China)

    Abstract:Inordertoknowthecharacteroftheheatvaluecontrolsystem,determinetheinfluenceofnatural

    gasqualityandflowO1"1theheatvalue,andlearnhowtoadjustthepar~rtetersofcontrolsystem,themodelof

    thewholesystemisestablished,andsimulationofthesystemisadoptedinMaflab/Simulink.Thesimulation

    resultshowsthatthefeedbacksystemwithfeed-forwardblockcontrolstheheatvalueverywell,andthesimu.

    1ationresultcandfectivdyguidetheengineeringdesignoftheheatvaluecontrolsystem,andtheefficiencyof

    engineeringisimproved.

    Keywords:naturalgas;heatvalue;controlsystem;simulation

    CLCnumber:TK229.8Documentcode:AArticleID:1004.O579(2006)03.0283.05 1WorkingPrincipleoftheHeatVal-

    neControlSystem

    Theadjustmentoftheheatvalueisrealizedby

    mixingpureairwithnaturalgas.Thefinalcontrol

    parameteristheopeningoftheairadjustingvalve.

    Theadiustingprocessisshownasfollows[.Firstly, theflowandheatvalueoftheinletnaturalgasismea- suredbytransformersindividually,andthepre- adjustingairamountisdeterminedbythetwosig- rials.ThecontrollerPID2controlstheairamountto thevicinityofthedemandedvaluebyadjustingthe openingoftheairvalve.Thenthepre-determinedair amountismixedwiththeinletnaturalgasinthemix- er,andtheheatvalueofthemixedgasisnearthede- mandedvalue.Theheatmeasurementdevicemea. sure8theactualheatvalueofthemixedgas,andthen feedsbackthemeasuredvaluetothecontrollerPID1, whichaddsanadjustmentvaluetothepre-determined airamountaccordingtotheerrorbetweenthemea- suredandthedemandedheatvalues,causesthe changeofairadjustingvalveopeningandmixingair amount.Andfinallytheheatvalueofthemixedairis adjustedtothedemandedvaluebytheaboveprocess. Theschematicofthesimulationsystemisillustrated asFig.1.

    heat-valuemeasurement

    Fig.1Schematicofthesimulationsystem

    Received2004.06.21

    mcgra~ayZHENGBin(1965),mi0rengineer,zbpowerr~rip.sina.oDm ?-——

    283?-——

    JournatofBeijingInstituteofTechnology,2006,Vo1.15,No.3 2ModelingoftheControlSystem

    2.1MathematicalModelsoftheHeatVallieAd-

justingSystem

    AccordingtoFig.1,thesimulationmodelsin

    cludethefeedforwardcontrolmodel.thetransfer modelofnaturalgasfrommeasurementpointtothe mixe~thecontrollermodel,theairadjustingvalve modekthemixermodel,andtheairflowandheat valuetsetmodels.Theindividualmodels isintroducedasfollows.

    2.2FeedForwardControlModel

    Thecombustionofnaturalgaswillproduceheat. andairisincombustible,sotheheatvaluewillde

    creasebymixingairwithnaturalgas.Supposethe heatvalueandtheflowoftheinletnaturalgasare QIlgandFIlg,andtheheatvalueofthemixinggas (objectheatvalue)isQso,theairfl0wisFair,then accordingtotheruleofenergyconservationwehave QngFIlg=Qso(FIlg+Fair).(1)

    Iftheflowandheatvalueareknown.theair flowfortheobjectheatvalueQsoofthemixtureis Fair0=(2)

    Thisisthefeed-forwardcontrolmode1.

    Accordingtotheabovediscussion,thefactors thatinfluencetheheatvalueofthemixtureare:?

    thevariationoftheheatvalueoftheincominggas,?

    thevariationoftheflowofthegas,?thedisturbance

    ofthedilutionair(pressure,temperature).S0,the controlcharactersundersuchconditionsaremainly analyzedhere.

    Eq.(2)justexplainsthecalculationmethodfor thefeed-forwardairflowundersteadyconditions.

Themainfactorsthatinfluencethetransientpedor

    manceofthefeedforwardblockistheheatvalueand flowofthegas,theresponsetimeofthemeasure

    mentset,andthepuredelaycausedbythe1ongsam

    plepipe,thedelayoftheelectricsignalfromfeed

    forwardblocktotheairvalvecontrollerPID2canbe overlooked.Sothetransientcharacterofthefeed- forwardblockcanbemodeledasapuredelaytimeLff andafirst-orderdelaywithatimeconstantTlf.Llf representsthepuredelaycausedbythelongsample pipe.Asthelengthofthepipeis20m,Lff=6s;Tlf isthebiggeroneofthedelaytimesoftheflowand heatvaluemeasurementset,Tff=0.2s.

    2?3TransferModeloftheBurningGas

    Thedistancebetweenthesamplingpointand mixingdevicemainlydeterminesthetransfermodelof theburninggas.Duringthisperiodthecomponentsis unchanged,andthetransferspeedisconstant,sothis procedurecanbeexpressedbyadelaytime,L.

    LIlgdisinfluencedbytheflowofthegas,thetime willbeshortenedwithbiggerflow.ThenLiscal

    culatedby

    L=Lgd0,(3)

    'Ilg

    whereFandLaretheflowanddelaytimere

    spectivelyunderthereferencecondition.IfF=

    3000m3/h,L.=0.3s,Eq.(3)canbeconverted

    to

    LIlgd=900/FIlg.(4)

    2.4AirControlValveModd

    TheaircontrolvalveisCV3000serieswhichis pneumatic-actuated.Therelationshipbetweenthe openingandthevolumeflowisgivenas

    =1+0.681g(Cr/C),(5)

    whereisthevalveopening(0100%);cisthe

    flowco-efficiency;Cvisthenominalflowco-efficien

    cy,C=200.

    TherelationshipbetweencandQisexpressed as

    Q=287tc/G(273+t)/[(P1+P2)(P1P2)],

    (6)

    whereQistheflowunderstandardcondition;Gis thegravityoftheair;tisthetemperatureoftheheat flow;Pl,P2arethepressuresupanddowntheair controlvalve.

    AccordingtoEqs.(5)(6),therelationshipbe

    tweentheopeningandtheflowthroughthevalveis Q=287tc10'/.?.)/

    ??——

    284??——

    ?G(273+t))/[(P1+P2)(p1p2)].(7)

    Thisisthesteadystatetransferfunctionofthe

    ZHENGBin()/SimulationStudyonHintValueControlSystemofNaturalGasUsedforColorTVTubesP

    roduction

    aircontrolvalve.

    Adelaytimeemergesundertransientconditions, accordingtomaterials,thetimeisabout0.2s. 2.5MixerModel

    Themixerisapipe,inwhichthereisaturbine, theairandthegasaremixedinthemixer,andsothe heatvalueofthegasischanged.Accordingtotheen? ergyconservation,theheatvalueofthemixtureis Q.=FnaQr~

    ,(8)

    whereFIlgandQIlgaretheflowandheatvalueofthe burninggasrespectively;Fairistheairflowusedto mixtheairwiththeburninggas.

    Becausethereisaturbineinthemixer,themix. ingisveryquick,andtheheatvaluechangesvery quickly,thedelaytimecausedbythemixerissmal1. Accordingtoreferences,thistimeisabout0.2s. 2.6AirMeasurementEquipmentModel

    Byestablishingenergyconservationandflow continuityequationtotheventuri,theflowthrough theventuricanbeexpressedas

    Q=0.0125acmD,/-~Plr,(9)

    whereeisthecalibrationco-efficiencydealtwiththe expantionoftheflow.forthecompressibleflow,e< 1;istheflowconstant,whichismeasuredthrough experiments;mistheratioofventuriareaF0topipe areaF;Distheinnerdiameterofthepipe;Apis pressuredifferenceupanddowntheventuri;7is thegravityofthemeasuredflowunderworking condition.

    Theairmeasurementequipmentisinthefield. andtheelectricsignalisfedbacktocontrollerPIDl2. thedelaytimeisshort,itismainlydeterminedbythe timeconstantofthedifferentialpressuregaugeandis

about0.1s.

    2.7HeatValueMeasurementEquipmentModel Inordertomeasurethegasheatvalueaccurately andquickly,theequipmentusescombustionanddilu? tionmethodstomeasurethegasheatvalueindirectly. Theheatvaluearecalculatedby[3,4

    Hg=?T(10)

    whereATismeasuredbythethermocouplesupand downtheburner;Sistheproductofgravitygauge readingandairgravity;Apaismeasuredbyairdiffer? entialpressureventuri;Apgismeasuredbygasdif

    ferentialpressureventuri.

    Asthegassamplingpipeis20m1orig,thedelay timeoftheheatvaluemeasurenuntisLq,Lq=6s. Becauseofthegasburningandtheheatcapacity0f theair,thedelayoftheheatvaluemeasurementsetis expressedbyatimeconstantTq.Accordingtomate

    rialsTqisabout0.5s.

    2.8ControllerModel

    ThecontrollerusesthePLC,thecontrolscheme usesthePIDcontrolarithmatic,sothemathematical modelofthecontrolleris

    Am=Ke+K.ed+Ka,)

    whereAmistheoutputofthecontroller;eistheer. rorofthecontroller,theerrorsofcontrollerPID1and PID2aretheheatvalueerrorandtheairflowerror respectively.Kp,Ki,Kdaretheproportional,the integralandthederivativeco-efficiencyrespectively. 3StudyoftheControlSystemby

    Simulation

    Accordingtotheabovemathematicalmodels, thefollowingsimulationmodelisestablishedinthe Matlab/Simulinkenvironments[.Thetargetvalue Qoftheheatvalueis34.57MJ/m3,theparameters 0ftheburninggasis:FIlg=3000/h,QIlg=

    35.95MJ/m3.Inordertostudytheinfluencesofthe gaspropertiesonthecontrolsystem,twosignalgen

    eratorsareusedtosimulatethedisturbancesofthe flowandheatvalueoftheinletgasandthetwoscopes areusedtoobservethechangesoftheheatvalueof themixtureandtheflowoftheairrespectively. ?First,accordingtotheactualsystem,thede

    laytimeofthesamplingpipeTissetas6s,thegas

    transferdelaytimeLissetas0.3s.Applyadis

    turbancetothegas,whoseamplitudeandperiodare 300and100srespectively.Theroleofthefeed- forwardblockcanbeobservedbycomparingthefluc

    tuationsoftheheatvaluebeforeandafterthedistur. banceareperformed.Thesimulationresultsare ?--——

    285--——

JournalofBeijingInstituteofTechnology,2006,Vo1.15,No.3

    illustratedinFig.2.Itcanbeseenthattheamplitude ofthefluctuationisreducedgreatlybythefeed..for.. wardblock.

    F.2Roleofthefeed-forwardblocktolow

    frequency(O.01Hz)disturbence

    ?Inordertostudythehighfrequencydistur. banceonthesystem,theperiodofthedisturbance

    isreducedtothesameorderofthetransferdelayof thesystem(about710s),theperiodofthedistur.

    banceis10s.Thesimulationresultsareillustratedin Fig.3.

    .

    3Roleofthefeed-forwardblocktohigh

    frequency(O.1Hz)distlIrbance

    Itcanbeseenthatthefeed.forwardblockcannot suppressthedisturbancewiththeperiodapproximate. 1yequaltothesystemdelay;eventhefluctuationof theheatvalueisenlarged.Thereasonisthatthereis adelaytime(68s)inthefeed.forwardblock,so

    thefeed-forwardblockcannotresponsethechangesof theheatvalue.Sothesystemisimprovedbyincreas- ingthedistancefromthegassamplingspottothe mixer,orbyincreasingtheflowspeedinthesam.- piingpipebypump,thenthetimethegasreachthe mixerisequaltothesumofthegassamplingdelay andtheaircontrolvalvedelay.Themethodusedhere isthefirstone,i.e.increasingthedistancefromthe gassamplingspottothemixer,andLn=7s.

    Thentheinfluencesofthelowfrequency(0.01Hz) andhighfrequency(0.1I-I~)disturbancesonthesys- temarestudiedagain.Theresultsareillustratedin Fig.4andFig.5.

    Itcanbeseenfromthefiguresthattheimproved systemcansuppressbothlowandhighfrequencydis. turbancestotheadjustedheatvalue.Sothegaspa. rametershouldbemeasuredandtransferredtothe feed-forwardblockintime,thustheroleofthefeed.

forwardcanbebringintoplayfully.

    Fig.4Roleofthefeedforwardblocktolowfrequency

    (O.01Hz)disturbance(improvedsystem)

    Fig.5Roleofthefeed-forwardblocktohighfrequency (O.01Hz)disturbance(improvedsystem)

    ?Onthebasisoftheimprovedsystem,theef. feetoftheserialcontrolmethodisstudiedbysimula. tion.Adisturbancesign~withfrequencyof0.1Hz andamplitudeof300isappliedtotheheatvalueof theburninggas.

    First,inthesituationoffeed.forwardinfunc. tioning,acomparisonismadewithinthecontrolled heat-valuesbetweentheresultsobtainedwithand withoutPID2,theresultsareshowninFig.6;andin thesituationoffeed.forwardoutoffunction,asimilar comparisonisshowninFig.7.

    Fig.6RoleofthePID2whenthefeedforward

    blockisinfunctioning

    FromFig.6andFig.7wecanseethatwhilethe feed-forwardblockisineffect,PID2suppressesthe effectofthedisturbanceontheheat-valuegreatly; ?-——

    286?-——

    (c-e

    ,_日譬一pa宣;_8

    【;_e

    ,I_aq墨一;_8

    争星

    ,鳘售至一;瞢8

Report this document

For any questions or suggestions please email
cust-service@docsford.com