高炉炉顶压力智能解耦控制方法及应用

39220104InformationandControlVol.39,No.2April,20101002-0411(2010)-02-0180-07410083blastfurnacetopgasrecoveryturbineunitTRTPIDTRT2200m3§2kPaTRTTP273TF325AApplicationofIntelligentDecouplingControlMethodforTopPressureofBlastFurnaceANJianqiWUMinXIONGYonghuaWANGChangjun(InstituteofInformationScienceandEngineering,CentralSouthUniversity,Changsha410083,China)Abstract:Adynamicmathematicalmodeloftoppressureandthepressurebeforeblastfurnace(BF)topgasrecoverytur-bineunit(TRT)isfirstlydesignedinthispaperbasedontheprincipleofblastfurnacetoppressuresystem.Thenconsideringthefeaturesofstrongcoupling,disturbanceandnonlinearityoftheblastfurnacetopsystem,adecouplingmethodcombiningPIDcontrol,fuzzycontrolandexpertcontrolisproposed,whichcanremovethecouplingoftoppressureandthepressurebeforeTRT,andcanalsosolvenonlinearproblemsaboutopeningandflowrateofannulargapandstaticblade.Whenthisdecouplingcontrolmethodisappliedtoa2200m3blastfurnaceinasteelcorporation,itisprovedeffectivetocontrolthefluctuationofthetoppressureofblastfurnacewithin§2kPa.Keywords:toppressureofblastfurnace;pressurebeforeTRT;decoupling;fuzzycontrol;expertcontrol1Introduction[1][2][3]PID[4-6]2TRTTRTTRT[7-10][e-mail]2009-06-12/2009-09-01/2010-01-131802181TRT2Descriptionofsystemprocess2200m3TRT11Fig.1PathoftopgasundernormalconditionofBFTRTp0TRTp1TRTp0+5kPa¡7kPa3ModelingandanalysisoftoppressureofBF————TRT3.1PV=nRT(1)PVnRTDP=RTVDn(2)p0TRTp1dp0dt=K1(Qd¡Q1)(3)dp1dt=K2(Q1¡Q2)(4)K1K2QdQ1Q2Qd[11]Q1=p2AS1px1rpp0¡p1(5)Q2=p2AS2px2rpp1¡p2(6)S1S2x1x2Arp2TRTp2CC=p2ASprx(7)ClClC(l)Cll1l2(5)(6)(7)(3)(4)dp0dt=K1¡Qd¡C1(l1)pp0¡p1¢(8)dp1dt=K2¡C1(l1)pp0¡p1¡C2(l2)pp1¡p2¢(9)(8)(9)dp0dt=¡K1C1(l1)p0¡p12pp00¡p10+K1Qd(10)18239dp1dt=K2C1(l1)p0¡p12pp00¡p10¡K2C2(l2)p1¡p22pp10¡p2(11)p00p10TRTRR1=2pp00¡p10(12)R2=2pp10¡p2(13)dp0dt=¡K1C1(l1)p0¡p1R1+K1Qd(14)dp1dt=K2C1(l1)p0¡p1R1¡K2C2(l2)p1¡p2R2(15)24˙p0˙p135=264¡K1R1(p0¡p1)0¡K2R1(p0¡p1)¡K2R2(p1¡p2)375¢2664C1(l1)C2(l2)3775+2664K1Qd03775(16)U01=C(l1)(17)U02=C(l2)(18)U01U02U01U02l1l23.2(16)TRTTRT(1)p0TRTp1p0¡p1p1¡p2PID(2)p0TRTp1p0¡p1p1¡p2PID(3)p0¡p1p0TRTp1(4)KRQd[12-13][14-15]PIDPIDp0PIDp0TRTp14Intelligentde-couplingcontroloftoppressure2PIDTRT2Fig.2Overallcontrolstructureofthesystem21834.1PIDp0¡p1=a(19)p1¡p2=b(20)ab2124˙p0˙p135=264¡K1R1a0¡K2R1a¡K2R2b37524u1u235+24K1Qd035(21)PIDPIDPIDPIDPID2eecPIDu0A1A1=8:0jej6ea1jeje(22)a1e2571[¡8;+6][¡0:6;+0:6]6[¡1;+1]a1=0:710%200ms1Tab.1RuletableoftoppressurefuzzycontrolHHHHHHEC1E1NBNMZOPMPBNBPBPBPSZOZONSPBPMZOZONSZOPMPSZONSNMPSPSZOZONMNBPBZOZONSNBNB(23)u01=nåi=1m(ui)¢uinåi=1m(ui)(23)uiii=1;2¢¢¢nnm(ui)uiu014.2(16)p0TRTp1TRT[16]222PIDU1U22U1U2DU1DU2U1U25[¡10;10]DU1DU27[¡1;1]0.1200ms(16)U2U1p0p1U1U2p1p02(23)DU1DU2(24)(25)U1U2U01U02184392Tab.2Ruletableofdecouplingfuzzycontrol@@@U2U1NBNMZOPMPBNBDU1=ZODU1=ZODU1=PSDU1=PMDU1=PBDU2=NBDU2=NMDU2=NSDU2=ZODU2=PSNSDU1=ZODU1=ZODU1=PSDU1=PSDU1=PMDU2=NMDU2=NSDU2=ZODU2=PSDU2=PSZODU1=NSDU1=NSDU1=ZODU1=PSDU1=PSDU2=NSDU2=ZODU2=ZODU2=PSDU2=PMPSDU1=NMDU1=NSDU1=NSDU1=ZODU1=PSDU2=ZODU2=ZODU2=PSDU2=PMDU2=PMPBDU1=NBDU1=NMDU1=NSDU1=ZODU1=ZODU2=ZODU2=PSDU2=PSDU2=PMDU2=PBU01=U1+DU1(24)U02=U2+DU2(25)4.33[17]3Fig.3Relationshipofflowrate,pressuredifferenceandopeningofannulargap4Fig.4Expertcalibrationcurveofannulargap34U01l1200ms5Operationresultsandanalysis2200m3PAULWURTH2S7-4002185PLCCPUPCS7215kPa113.5±2.5tPIDPCS7CFCCTRLPIDGAIN(kp)=¡550TN(ki)=25TV(kd)=75PIDGAIN(kp)=¡4TN(ki)=50TV(kd)=1.5200ms57PIDPIDPID8103TRTkPas5PIDPID+5¡7kPa6PID§4kPa7§2kPa810TRT3112332u1U1U01(1)PIDPIDPID(2)PID5PIDFig.5ToppressuredeviationcurveunderPIDcontrol6PIDFig.6Toppressuredeviationcurveunderfuzzy-PIDcontrol7Fig.7Toppressuredeviationcurveunderintelligentdecouplingcontrol8PIDTRTFig.8PressurebeforeTRTdeviationcurveunderPIDcontrol9PIDTRTFig.9PressurebeforeTRTdeviationcurveunderfuzzyPIDcontrol10TRTFig.10PressurebeforeTRTdeviationcurveunderintelligentdecouplingcontrol1863911TRTFig.11ControlcurveofthethreealgorithmstoppressureandpressurebeforeTRTdeviationcurve3Tab.3Toppressurecontrolresults(kPa)PIDPID6.38972.21841.4036jej2[0;1]25.70%35.24%49.95%jej2(1;2]30.84%41.64%45.98%jej2(2;¥)43.46%23.12%4.07%CO242.41%42.65%42.98%TRT33PID§2kPaCO26ConclusionTRTPID2200m3References[1]PLC[J]200724(3):56-58[2][J]2006,8(3):38-40[3]TsumuraK.Modelingforcontrolofblastfurnace[C]//IEEEInternationalConferenceonControlApplications.Piscataway,NJ,USA:IEEE,1999:1-6.[4][J]2000(6):15-16[5]TRT[J]2005(5):88-90[6]TRT[J]2007(130):28-30[7]TRT[J]200640(9):1536-1539[8]TRT(STPC)[J]200717(10):50-53[9]TRT[J]200827(4):15-19[10][D]2003[11][M]1993:70-75[12]ClarkeW.Applicationofgeneralizedpredictivecontroltoin-dustrialprocesses[J].IEEEControlSystemsMagazine,1988,8(2):49-55.[13]ARX[J]200837(2):219-223[14][J]2008,37(2):141-145[15]VadimU.Variablestructuresystemswithslidingmodes[J].IEEETransactionsonAutomaticControl,1978,22(2):212-222.[16]KuoBC,GolnaraghiF.Automaticcontrolsystems[M].NewYork,NY,USA:JohnWiley&Sons,2005.[17]PaulWurthCo.D021BFTopPressureAdjustmentCurve(deltap/deltah)ofTopGasCleaningSystem[R].2003.1981–1963–1979–