Key Words:: Module 04: Targeting Lecture 14: No. of Units Target

Download as pdf or txt
Download as pdf or txt
You are on page 1of 8

No.

ofunitstarget

Module:04

Lecture14

Module04:Targeting Lecture14:No.ofunitstarget KeyWords:HEN,Subset,Loop,Separatecomponent,MER design


Thefixedcostofaheatexchangernetwork(HEN)dependsuponthenumberofheatexchanger itemploys.Thus,thereexistsapossibilitythataHENwithminimumno.ofheatexchangerwill cost less. Thus there exists a strong incentive to reduce the number of heat exchangers ( matchesbetweenhotandcoldstreams)inaHEN.Thefirststeprequiredforthisprocessforits initiationistoidentifythenumberofheatexchangersaHENwillrequirefromthenumberof Hot,ColdandUtilitystreamsithandles. Letusexplaintheproblemwithanexample.Fig.4.23showstheflowsheetofapalmoilrefinery [1,2].

Fig.4.23Flowsheetofapalmoilrefinery

No.ofunitstarget

Module:04

Lecture14

Theflowsheetusesthreeheatexchangers,threecoolersandfourheatersmaking10unitsin all.Nowthequestioniswhether10unitsaretheminimumnumberofunits,oradesignercan reduceitwithouthamperingthefunctionalityoftheprocess? ThestreamtablefortheaboveproblemisgiveninTable4.14: Table4.14Streamdataforpalmoilrefinery(Fig.4.23)forTmin=10C Stream StreamType CP ActualTemperatures Enthalpy,H SerialNo. (KW/K) kW Ts(0C) Tt(0C) 1 Hot1 10.99 120 86 373.66 2 Hot2 6.04 260 160 604 3 Hot3 13.13 230 70 2100.8 4 Hot4 6.56 160 50 721.6 5 Cold1 11.83 50 97 556.01 6 Cold2 14.89 104 124 297.8 7 Cold3 5.69 86 230 819.36 PTAanalysisoftheproblemshowsthatitisathresholdproblemandneedsonlycoolingandno. heating. The minimum cooling load required for the above system computed using PTA is 2126.89. The heat loads of different streams along with cold utility load is shown within the circles representing the streams in Fig.4.24. The predicted cold utility load is also shown similarly. Hot1 Hot2 Hot3 Hot4 373.66 604kW 2100.8 721.6 kW kW kW HX7 HX6 HX5 HX4 HX3 HX2 HX1 2003.04 123.85 249.81 306.2 297.8 97.76 721.6 Cold2 Cold3 Cold1 Coldutility 297.8 819.36 556.01 2126.89kW kW kW kW Fig.4.24Schematicmatchingofheatloadsforstreamtabledata,Table4.14

No.ofunitstarget

Module:04

Lecture14

Notethatthecompletesystemisinenthalpybalance(i.e.thetotalloadofcoldstreamsplus cold utility is equal to the total load of hot streams). If we presume that temperature constraintswillallowanymatchtobemadebetweenhotstreamsandcoldstreamsincluding cold utility, then we can match the whole of cold streams 3 (total 819.36 units) with Hot streams3&4,leavingaresidualheatloadof2003.04unitsonHot3.MatchingHot3&Hot4 with Cold3 and maximizing the load on this match so that Cold3 & Hot4 is ticked off the 2003.4 residual heat available with Hot3 is sent to cold utility which requires 2126.89 units. Cold2istickedoffbymatching with Hot2leaving306.2 unitsofheat init.The remaining heatinHot2alongwith249.81unitsofheatfromHot1ispassedtoCold1toTickitoff.This leaves 123.85 units of heat with Hot1 which is passed to cold utility. This heat along with 2003.04unitsofheatfromHot3ticksoffcoldutility.Sofollowingtheprincipleofmaximizing loads,thatistickingoffstreamorutilityloadsorresiduals,leadstoadesignwithatotalof seven matches (connections between streams and utilities show matches are denoted by HX withanumber).Thisisinfactistheminimumforthisproblem.Noticethatitisonelessthan thetotalnumberofstreamsplusutilitiesintheproblem. Thusitcanbeshownthat: umin=N1 .(4.8) whereumin=minimumnumberofunits(includingheatersandcoolers)and N=totalnumberofstreams(includingutilities)

No.ofunitstarget

Module:04

Lecture14

Infact,itisusuallypossibleinheatexchangernetworkdesigntofindauminsolution.However, certain refinements to Eq.4.8 are required as discussed below to broaden its applicability. In Fig.4.25 (a), problem having two hot streams ( H1 & H2), two cold streams(C1 & C2), Hot utility(HU)andColdutility(CU)isshown.Inthiscase, H1 H2 HU H1 H2 HU 40 80 160 40 80 160 40 50 30 90 70 40 80 90 70 C1 CU CU C2 C1 C2 70 70 90 120 90 120 Matches=05 (a) Matches=04 (b) HU H1 H2 40 80 160 40Y Y 50+Y 30Y 90 70 Loop CU C1 C2 70 90 120 Matches=06 (c) Fig.4.25Subsetandloopsduringmatching putting matches as before by ticking off loads or residuals leads to a design with N 1 units which satisfy Eq.4.8 However, in Fig4.25(b) a design is revealed having one unit less. The justification for the fact that the number is less than minimum is not hard to find. Even as overalltheproblemisinenthalpybalance,thesubsetcontainingstreamsH2,C1andCUare inenthalpybalance.SimilarlyHU,H1andC2areinenthalpybalance(thisisaknownfactas the total problem is in load balance). This means that for the given stream data set we can design two completely separate networks, employing the Eq.4.8 to each subset individually. Thetotalnumberofunitsfortheoverallsystemistherefore(31)+(31)=4units,whichisone lessthanfoundinFig.4.25(a).Thisconditioniscalledsubsetequality,thisappearswhenfora given stream data set it is possible to identify two subsets which are separately in enthalpy balanceandthuscanformseparatenetworks.Sincetheflowsheetdesigner,cancontrolofthe quantityoftheheatloadsinhisplanttosomeextent,itispossibletochangetheheatloadsso as to create subset equality and thus create an opportunity to save a unit. Finally, in Fig.4.25(c)amatchingschemeisshownwhichrequiresoneunitmorethantheschemeshown

No.ofunitstarget

Module:04

Lecture14

inFig.4.25(a),thenewextraunithasbeenintroducedasthematchbetweenHUandC2which introduces a loop in the system. It is so because one can trace a closed path through the networkstartingfrom,sayHU,theloopcanbetracedtoC1,fromC1toH1,fromH1toC2, andfromC2backtoHU.Though,thepresenceoftheloopintroducesanelementofflexibility into the design it increases number of units in the system. Suppose the new extra match, betweenHUandC2,isassignedaloadofYunits,thenthroughenthalpybalance,theloadon thematchbetweenHUandC1hastobe40Y,betweenC1andH1,50+Y,andbetweenH1 andC2,30Y.FromFig(c)itcanbeinstitutivelysaidthatYcanvaryfrom0toavalueof30. WhenY=0thematchbetweenHUandC2vanishesandwhenitis30thematchbetweenH1 and C2 disappears. The flexibility introduced by loops is many times useful, particularly in revampstudiesandcleaningoperations. ThefeaturesdiscussedaboveandshowninFig.4.25(a),(b)&(c)canbeconvertedintoa formulatocomputenumberofheatexchangeunits,usingtheEulersGeneralNetwork theoremappliedtoheatexchangernetwork: umin=N+LS (4.9) where; u=numberofheatexchangeunits(includingheatersandcoolers), N=numberofstreams(includingutilities), L=numberofindependentloops,and S=numberofseparatecomponents

NormallyadesignerwanttoavoidextraunitsbyreducingLtozero.Unlessoneislucky,there willbenosubsetequalityinthestreamdatasetandthusthevalueofswillbe1.Thisleadsto thenumberofunitstargetingequation: umin=N1 (4.8) Inthedesignofheatexchangernetworkstechniquesdiscussedabovewillbeusedtoreducethe number of heat exchangers by allowing small energy penalty at various sections of the networkfortradingoffenergyagainstcapitalcost. Examples:Afivestreamproblemistakenupcomputeno.ofunitstarget. Table4.15Afivestreamproblemforno.ofunitstargetforTmin=10C Stream Stream Heat Capacity Source Target H,kW Number Type FlowRate Temperature Temperature (kW/0C) (0C) (0C) 1 HOT1 147.74 70 10 8864.34 2 HOT2 165.85 60 33 4478 3 COLD1 50 57 60 150 4 COLD2 215 41 60 4085 5 COLD3 194.74 10 30 4479 UsingPTAtheminimumhotandcoldutilitiesarecomputedasgivenbelow:

No.ofunitstarget

Module:04

Lecture14

Hotutility(HU) =822.61kW Coldutility(CU) =5450.95kW HotPinchTemp. =60C Coldpinchtemp. =50C FromthetableitappearsthatiftheheatloadofCold3canbebroughtdownto4478thereisa chanceforsubsetequalityresultingS=2andtherebydecreaseofno.ofunitsbyone.Forthis case:N=7(includingHUandCUstreams);L=0andS=2 umin=N+LS=7+02=5 IfsubsetequalityisnotcreatedthenN=7,L=0andS=1.Forthiscase No.ofunitsare: umin=7+01=6

Targeting for the minimum number of units for a MER design


HoweveriftheaboveequationEq.4.8isappliedtoamaximumenergyrecovery(MER)design the results will be somewhat different. For this purpose the problem of Table 4.15 is considered. In a MER design the pinch divides the problem into two heat balanced regions. Since these balanced regions are independent, numbers of units targeting should be applied separatelytoeachregionasshownbelow:

No.ofunitstarget

Module:04

Lecture14

Fig.4.26showsthestreamlayoutaboveandbelowthepinch.

HU 1 70C

HotPinch=60C

ColdPinch=50C 10C

AbovePinch 2

BelowPinch 60C

33C

60C

57C

60C 41C 4 33C 10C 5 CU Umin=41=3 Umin=51=4 Fig.4.26Processhotandcoldstreamsandutilitystreaminaboveandbelowpinch Thustotaluminforthenetwork=3+4=7 IfPinchdivisionisnotconsideredthenno.ofstreamsincludinghotandcoldutilitiesis=7, S=1andL=0. Thus,thetotalno.ofunitsfornonMERdesign=umin=7+01=6 ThusitcanbeprovedthatuminuminMER ThenumberofunitsobtainedintargetingfortheMERdesignismorethatuminduetothefact thatstreamsthatcrossthepincharecountedtwiceinMERdesign.Theconclusionisthatthere isatradeoffbetweenenergyrecoveryandnumberofunitsemployedinaMERdesign.Howto reduceno.ofunitsinaMERdesignwillbeexplainedwhenMERdesignwillbediscussed.

No.ofunitstarget

Module:04

Lecture14

References 1. K.K.Trivedi,E.Fouche,K.E.Parmenter,ProcessEnergyEfficiency:PinchTechnologyin HandbookofEnergyEfficiencyandRenewableEnergy,CRCPress,BocaRaton,2007,pp. 1511530. 2. SharifahR.WanAlwi,ZainuddinA.Manan,STEPAnewgraphicaltoolfor simultaneoustargetinganddesignofaheatexchangernetwork,ChemicalEngineering Journal162(2010)106121

You might also like