0% found this document useful (0 votes)
45 views

Module 05 Lecture 32: HEN Optimization: Key Words: HEN, Loop Breaking, PDM, Utility Path

- The document discusses optimization of heat exchanger network (HEN) design beyond maximum energy recovery (MER) to reduce total annual cost. - MER design divides the problem into independent hot and cold regions, but this leads to redundant heat exchangers when combined. Allowing heat transfer through the pinch can reduce units while increasing utilities. - The example HEN has a loop that can be broken to reduce one unit, but this increases utilities. Using the "utility path" between heaters and coolers allows breaking the loop while maintaining minimum temperature differences to avoid large utility increases.

Uploaded by

api-256504985
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
45 views

Module 05 Lecture 32: HEN Optimization: Key Words: HEN, Loop Breaking, PDM, Utility Path

- The document discusses optimization of heat exchanger network (HEN) design beyond maximum energy recovery (MER) to reduce total annual cost. - MER design divides the problem into independent hot and cold regions, but this leads to redundant heat exchangers when combined. Allowing heat transfer through the pinch can reduce units while increasing utilities. - The example HEN has a loop that can be broken to reduce one unit, but this increases utilities. Using the "utility path" between heaters and coolers allows breaking the loop while maintaining minimum temperature differences to avoid large utility increases.

Uploaded by

api-256504985
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 9

HENoptimization Module05 Lecture32

Module05
Lecture32:HENOptimization
Keywords:HEN,Loopbreaking,PDM,utilitypath,

Inlectures29,30and31itwasshownhowtodesignMaximumEnergyRecovery(MER)HENs.
InMERHENscoldandhotutilitiesasexactlysameastargeted.However,MERHENsarenot
thebestHENsasfarastotalannualcost(TAC)isconcerned.Nowthequestioniswhether,the
MERHENsdesignedinLecture2931canbeimprovedfurther?Theanswertothisquestionis
yes.
Using the PDM a HEN structure can be created based on the assumption that no heat
exchanger should have a temperature difference less than T
min
. However, the structure thus
createdcantbetermedoptimumandsomeprocedureforoptimizationcanbeadopted.Letus
onebyoneseethedrawbacksoftheHENcreatedbyPDM.
In an MER design, the pinch divides the problem into two thermodynamically independent
regionsasshowninFig.5.48(a).HENdesignsareperformedindependentlyforthesetwoparts
andoncedonearejoinedtogethertogetacompleteHENfortheproblem.Duetothisprocess
someweaknessescreepintothedesign.
Now,supposethedesignisconvertedintoanonMERdesignbyallowingamountofheatto
pass through pinch division, then the hot and cold utility amount will rise to Q
Hmin
+ and
Q
Cmin
+ respectively as shown in Fig.5.48(b). Now the earlier regions( Hot end & cold end)
defined in MER design is no longer thermodynamically independent. Thus, if the number of
unitstargetisappliedtothewholeproblem,ignoringpinchdivision,thenonewillobservethat:
U
min(
NonMER)U
min
(MER)
The above result is an expected one and can be explained as follows: When one applies unit
targettoaMERdesign,streamsthatcrosspinchdivisionarecountedtwiceandthusinaMER
design the total number of units are always more than a nonMER design. This clearly shows
thatbyallowingheattotransferthroughpinchdivision,thoughtheamountofutilityincreases,
the number of units is decreased and thus offers an opportunity to decrease the fixed cost of
theHEN.Thusthereisatradeoffbetweenenergyrecoveryandnumberofunitsemployed.
TradingoffunitsandEnergy
TheabovefactsareexplainedthroughtheexamplegiveninTable5.10.

HENoptimization Module05 Lecture32

Table5.10Afourstreamproblemtodemonstrateenergyvs.no.ofunitstradeoff(T
min
=10
0
C)
StreamName SupplyTemperature TargetTemperature CP

H1 160 60 2.5
H2 140 30 1.5
C1 20 125 2
C2 80 130 3.5

The hot pinch temperature for the above problem is 900 C whereas, the cold pinch
temperature 80
0
C. The hot utility requirement is 15 kW whereas that of cold utility
requirementis45kW. ThestreamdiagramisshowninFig.5.49.Thehotendaswellcoldend
designoftheHENwascarriedoutfortheproblem.AfterjoiningthehotandcoldendHENsthe
overall MER HEN is shown in Fig.5.50. It can be seen that the overall number of units in this
designis6includingheaterandcoolerswhereasanonMERdesignshouldproduce5(=4+21)
units. This clearly shows that there is a scope of decreasing one unit in a nonMER design
Q
HMIN

Q
CMIN

COLD END
HOTEND
Pinch
U
min
(MER)=U
min
(HotEnd) +U
min
(ColdEnd)
MaximumEnergyRecovery(MER)Design
NonMERDesign (noconceptof pinch division)

Q
CMIN+

Q
HMIN+

U
min
(nonMER)=N1
U
min
(nonMER)U
min
(MER)
(a)
(b)
Fig.5.48Targetingfornumberofunitsin(a)MERaswellasin(b)nonMERdesigns
HENoptimization Module05 Lecture32
however, at the cost of increased hot and cold utility which is to be investigated to strike a
tradeoff.ThereasonforoneextranumberofunitisduetothepresenceofaloopintheMER
design(pleaseseeLecture14).Thusiftheloop(inFig.5.50)isbrokenthenoneunitcanbe
reduced in the design. However, this will convert the MER design to a nonMER design and
someamountofheatwillpassthroughpinchdivision.Nowthequestioniswhatshouldbethe
properwaytodoitsothatwegetthebenefitofreducednumberofunitsintheHENandatthe
sametimetheamountofheatpassedthroughthepinchisalsonotmuch.
Fig.5.51showsanattempttobreaktheloopbyremovingtheheatexchanger(45kW)placedat
onesideoftheloopandaddingthisloadwiththeheatexchanger(75kW)ontheothersideof
theloopmakingitsloadas120kW.Fromcommonsenseitappearsthatthisarrangementwill
work. However, it violates the T
min
criterion and thus the design is not acceptable. Fig.5.52
showsanalternatearrangementwheretheamountofhotaswellascoldutilitiesareincreased
by 45kW( load of the removed heater). Theoretically, this design works and does not violates
theT
min
criterionandthusacceptable.However,theutilityloadsareincreasedsignificantly.
Nowthequestioniscantherebeathirdalternativetothedesignwhichwilldecreasetheutility
loadsbutwillprovideafeasibledesign?Theanswerisyes,itcanbedonebyutilizingtheutility
pathasdiscussedbelow:

Fig.5.49StreamdiagramfortheproblemshowninTable5.10.
HENoptimization Module05 Lecture32

Fig.5.51GriddiagramoftheHEN(MERDesign)
175kW
75kW+45kW
15kW
75kW
45kW
60C
57.5C
T
min
violation
117.5C
Fig.5.50GriddiagramoftheHEN(MERDesign)
U
min
(MER)=6
175kW
75kW
15kW
75kW 45kW
45kW
HENoptimization Module05 Lecture32

42.5
Fig.5.53Identificationofloop(reddottedline)intheMERHENdesign
Loop1
A
B
C
D
Heater
Cooler
90
60
42.5
117.5
80
T
min
=10C
175kW
75kW
15kW
75kW 45kW
45kW
Fig.5.52GriddiagramoftheHEN(NonMERDesign)
175kW
75kW
15kW+45kW
75kW
45kW+45kW
90C
57.5C
95C
HENoptimization Module05 Lecture32

GiventheHENasinFig.5.53itispossibletoidentifyloopsandutilitypaths.IntheFig.5.53a
loopwhichisshownbyreddottedlinesconnectsexchangersBandD.Itispossibletotracea
closepathinthenetworkforaloop.Thoughtheexistenceofaloopintroducessomeelement
offlexibilitybytransferringloadfromoneunittootherintheloop,itaddsonemoreunitper
loop.Forexample,theloadofexchangerBcanbetransferredtoDwhichispresentinthe
sameloop.Further,bymakingtheloadofanexchangerzerotheexchangercanberemoved
fromtheloopandtheloopcanbebroken.ForexampleiftheloadofexchangerDismade
zeroandtocompensatetheloadtheloadofexchangerBisincreasedto120(=75+45)kW,
thentheloop1canbebroken.However,thiswilldisturbtheheatflowinthenetworkand
violationofT
min
(Fig.5.54)willtakeplacewhichhastoberestoredtokeeptheutilitiesnear
targetedvalue.
TherestorationoftheT
min
canbedonethroughautilitypath.Utilitypathsarethosepaths
whichconnectstwodifferentutilities.Thispathcouldbepathjoininghotutility,steamtocold
utilitycoldwaterorapathfromhighpressuresteamtolowpressuresteam.
ForthepresentcaseinFig.5.55autilitypathfromheaterthroughexchangerBtocooler(
purpledottedline)canbetracedtofindaT
min
compensationplanfortheproblem.
Fig.5.54Identificationofloop(reddottedline)
Loop1
A
B
C
D
Heater
Cooler
60
117.5
57.5
T
min
=10C
45kW45kW=0 75kW+45kW
T
min
violation
175kW
75kW
15kW
75kW 45kW
45kW
HENoptimization Module05 Lecture32

IfweaddheatloadofXtotheheater,itloadbecomes15+XkW,thenforenthalpybalance
theloadofexchangerBshouldbedecreasedbyXamountandtheloadoncoolerbe
increasedbyXamount.Letthetemperatureofstream2beT
1
aftertheexchangerB.Now
torestoretheT
min
,T157.5shouldbeequalto10C.
Theaboveispossiblewhenfollowingequationissatisfied.
140(120X)/1.5=T
1
=57.5+10=67.5C
Or30+(45+X)/1.5=T
1
==57.5+10=67.5C
TosatisfytheaboveequationthevalueofXshouldbe11.25kW.
TheabovesolutionisclearlybetterthanthesolutionofferedinFig.5.52

PDM creates network structures based on the assumption that none of the exchanger in the
network should have a temperature difference less than T
min
. Once a HEN structure is
designed based on PDM, it can be subjected to optimization to improve its cost effectiveness.
Theconstraintimposedonitcanberemovedonebyonetogetimprovedsolutionintermsof
fixedcostofHEN,numberofunitsorshellsinHENorfinallyTACofHEN.
175kW
75kW
75kW
90kW
90kW 120kW
175kW 175kW
75kW
15kW
75kW 45kW
45kW
Fig.5.55Identificationofloop(reddottedline)intheMERHENdesign
A
B
C
15+XkW
Cooler T
1

57.5
T
min
=10C
120XkW
45+XkW
HENoptimization Module05 Lecture32
1. Theconstraintthatnoneoftheexchangershouldhavetemperaturedifferencelessthan
T
min
canberelaxed
2. Theexchangerdutiescouldberedistributedorevensettozerotoremoveoneofit.
3. RemovalofextranumberofunitsintheHENbybreakingloops.
4. ProperselectionofutilitypathsintheHENafterloopbreakingtorestoreT
min
.Utility
pathsandloopsprovidedegreeoffreedomforoptimizationandshouldbeutilizedto
improvecosteffectivenessofHEHs.
5. Streamsplittingalsooffersanadditionaldegreeoffreedomintheprocessof
optimizationandshouldbemanipulatedfortheimprovementofHENs.

References
1. AngelMartn*,FidelA.Mato,Hint:Aneducationalsoftwareforheatexchanger
networkdesignwiththepinchmethod,educationforchemicalengineers3(2008)e6
e14
2. Linnhoff,B.andFlower,J.R.,1978,Synthesisofheatexchangernetworks,AIChEJ,24(4):
633.
3. Linnhoff,B.andHindmarsh,E.,1983,Thepinchdesignmethodforheatexchanger
networks,ChemEngSci,38(5):745.
4. Linnhoff,B.,Townsend,D.W.,Boland,D.,Hewitt,G.F.,Thomas,B.E.A.,Guy,A.R.and
Marsland,R.H.,1994,AUserGuideonProcessIntegrationfortheEfficientUseof
Energy.(TheInstitutionofChemicalEngineers,Rugby,Warks,UK).
5. Smith,R.2005,ChemicalProcess:DesignandIntegration(seconded.),(J.Wiley,JWiley.

HENoptimization Module05 Lecture32

You might also like