Index PDF

OVERVIEWOFSOLUTIONSTOPREVENTLIQUIDLOADINGPROBLEMSINGASWELLS
ATHESISSUBMITTEDTO
THEGRADUATESCHOOLOFNATURALANDAPPLIEDSCIENCES
OF
MIDDLEEASTTECHNICALUNIVERSITY

BY

ZMENBNL

INPARTIALFULFILLMENTOFTHEREQUIREMENTS
FOR
THEDEGREEOFMASTEROFSCIENCE
IN
PETROLEUMANDNATURALGASENGINEERING

DECEMBER2009

ApprovaloftheThesis:


submitted by ZMEN BNL in partial fulfillment of the requirements for the degree of
Master of Science in Petroleum and Natural Gas Engineering Department, Middle East
TechnicalUniversityby,
Prof.Dr.Cananzgen ______________________
Dean,GraduateSchoolofNaturalandAppliedSciences
Prof.Dr.MahmutParlaktuna ______________________
HeadofDepartment,PetroleumandNaturalGasEngineering
Assoc.Prof.Dr.Evrenzbayolu ______________________
Supervisor,PetroleumandNaturalGasEngineeringDept.,METU
ExaminingCommitteeMembers:
Prof.Dr.MahmutParlaktuna ______________________
PetroleumandNaturalGasEngineeringDept.,METU
Assoc.Prof.Dr.Evrenzbayolu ______________________
Prof.Dr.MustafaVeranKk ______________________
Prof.Dr.SerhatAkn ______________________
EmreDorukTanyel,M.Sc. ______________________
DrillingEngineer,TrkiyePetrolleriA.O.
Date: ______________________
ii
Iherebydeclarethatallinformationinthisdocumenthasbeenobtainedandpresentedin
accordance with academic rules and ethical conduct. I also declare that, as required by
theserulesofconduct,Ihavefullycitedandreferencedallmaterialandresultsthatarenot
originaltothiswork.
Name,LastName:
Signature:
iii
ABSTRACT
Binli,zmen
M.Sc.,DepartmentofPetroleumandNaturalGasEngineering
Supervisor:Assoc.Prof.Dr.Evrenzbayolu
December2009,84pages
Everygaswellceasesproducingasreservoirpressuredepletes.Theusualliquidpresencein
the reservoir can cause further problems by accumulating in the wellbore and reducing
productionevenmore.Thereareanumberofoptionsinwellcompletiontopreventliquid
loading even before it becomes a problem. Tubing size and perforation interval
optimization are the two most common methods. Although completion optimization will
preventliquidaccumulationinthewellboreforacertaintime,eventuallyasthereservoir
pressuredecreasesmore,thewellwillstartloading.Asliquidloadingoccursitiscrucialto
recognizetheproblematearlystagesandselectasuitablepreventionmethod.Thereare
various methods to prevent liquid loading such as; gas lift, plunger lift, pumping and
velocity string installation. This study set out to construct a decision tree for a possible
expertsystemusedtodeterminethebestresultforaparticulargaswell.Thefindingsare
testedtoconfirmbyfieldapplicationsasattemptsoftheexpertsystem.
Keywords:Gasproduction,liquidloading,artificiallift,decisiontree,expertsystem

iv
GAZKUYULARINDASIVIDOLUMUNUNENGELLENMESN
ZMYNTEMLERNEGENELBAKI

Binli,zmen
YksekLisans,PetrolveDoalGazMhendisliiBlm
TezYneticisi:Do.Dr.Evrenzbayolu
Aralk2009,84sayfa
Rezervuar basncnn azalmasyla zamanla gaz kuyularnn retimleri azalr. Formasyon

svlarzamaniindekuyudibindebiriktiindentr,hidrostatikbasnyaratarakretimin
beklenenden erken dmesine yol aar. Bu ve benzeri problemler ciddi retim sorunlar
oluturmadan nce doru yntemler kullanlarak engellenmelidir. Kuyuda doru retim
dizisininkullanmbukonudakibalcanemtar.Kullanlaneitliyntemlerkuyudibinde
svbirikmesinibelirlibirsreengelleyecek,ancakrezervuarbasncazalmayadevamettike
kuyu yeniden sv ile dolmaya balayacaktr. Sv dolumunun erken safhalarda tehis
edilmesivekuyuzelliklerineuygunbirmcadeleyntemiseilmesininnemibyktr.Bu
sorun ile mcadele iin kuyuya gaz enjekte etme, svy bir serbest bir piston yardmyla
kaldrma,pompalama,retimdizisininkltlmesigibibirdiziyntemkullanlmaktadr.Bu
aratrmada teorik verilerle elde edilenler sorunun sistematik zm iin bir karar
mekanizmas oluturulmas zerinde kullanlmtr. Saha verileri ile dorulanmaya allan
bukararmekanizmasnnbellidurumlariineniyizmnbulunmasnaallmtr.
Anahtarkelimeler:Gazretimi,svdolumu,yapaykaldrma,kararaac,uzmansistem
v
toMe,
vi
ACKNOWLEDGMENTS
TheauthorwouldliketothankAssoc.Prof.Dr.Evrenzbayoluforhissupervisionthrough
theresearch.
The author would also like to thank Prof. Dr. Mahmut Parlaktuna for his guidance and
support.
TheauthorwouldalsoliketothankMr.MuratFatihTuanforhistechnicalassistance.
Lastly, the author would like to express his deepest gratitude to Ms. Helin Kara for her
encouragement,patienceandgreatlyappreciatedassistance.
vii
TABLEOFCONTENTS
ABSTRACT iv
Z v
ACKNOWLEDGEMENTS vii
TABLEOFCONTENTS viii
LISTOFTABLES x
LISTOFFIGURES xi
LISTOFSYMBOLS xiv
CHAPTER
1 INTRODUCTION 1
1.1 MultiphaseFlow 1
1.2 LiquidLoadingConcept 3
1.3 SourceofLiquids 4
2 LITERATUREREVIEW 7
3 STATEMENTOFPROBLEM&SCOPE 11
4 THEORY 12
4.1 PredictingLiquidLoading 12
4.2 CriticalRateTheory 14
4.3 TubingSizeOptimization 17
4.4 WellCompletion&ProductionRateOptimization 23
4.5 OverviewofSolutionstoPreventLiquidLoading 27
5 DECISIONMAKING:EXPERTSYSTEM 44
5.1 GeneralInformation 44
5.2 SummaryofTheory:DesignProcess 44
6 APPLICATION:CASESTUDY 48
viii
6.1 BackgroundInformation 48
6.2 Well#10 49
6.3 Well#28 54
7 DISCUSSIONOFRESULTS 63
8 CONCLUSIONS 70
REFERENCES 71
APPENDICES
A DEVELOPMENTOFCRITICALVELOCITYEQUATIONS 74
B PLUNGERLIFTEQUATIONSANDFEASIBILITYCHARTS 77
B.1 MinimumCasingPressure 77
B.2 MaximumCasingPressure 78
B.3 PlungerFeasibilityCharts 78
C INFLOWPERFORMANCECALCULATIONSOFWELLS#10&#28 81
C.1 Well#10 81
C.2 Well#28 83
ix
LISTOFTABLES
Table4.1CompressionHorsepowerandFuelGas 30
Table6.1GeneralWellInformationof#10and#28 48
Table6.2Well#10PressureSurvey#1SampleData 50
Table6.3ProductionDataofWell#28 58
Table6.4DailyProductionRatesof#28PreandPostLogicControl 61
TableC.1CalculatedInflowPerformanceDataforWell#10 82
TableC.2CalculatedInflowPerformanceDataforWell#28 84
LISTOFFIGURES
Figure1.1BasicProfileofMultiphaseFlowintheWell 2
Figure2.1ShapeofEntrainedDropMovementinHighVelocityGas 8
Figure4.1ATypicalGasWellDeclineCurvealongwithIndicationofLoading 13
Figure4.2ATypicalPressureSurveyGraph 14
Figure4.3LiquidDropletMovement 16
Figure4.4CriticalFlowrateforDifferentTubingSizes 18
Figure4.5CriticalGasFlowratesforTurner&Coleman 19
Figure4.6TypicalNodalAnalysisCurves 20
Figure4.7TubingPerformanceCurve 21
Figure4.8NodalAnalysisGraphforDifferentTubingSizes 22
Figure4.9TheEffectofSettingtheEndofTubing 24
Figure 4.10 Impact of Length of Perforation Interval on Cumulative Water and Gas
ProductioninYears 26
Figure4.11ImpactofPerforatedIntervalandPermeabilityonUltimateGasandWater
Recovery 26
Figure4.12ImpactofGasProductionRateandPermeabilityonWellLifeandUltimate
GasRecovery 27
Figure4.13VelocityStringApplicationSchematic 29
Figure4.14EnergyRequiredforDifferentTubingSizestoStayaboveCriticalRate 31
Figure4.15ATypicalPlungerLiftInstallation 33
xi
Figure4.16ASimpleIllustrationofPlungerLiftCycles 34
Figure4.17GasRequirementforPlungerLiftwithorwithoutPacker 36
Figure4.18ContinuousGasLiftSchematic 38
Figure4.19IntermittentFlowGasLiftSchematic 39
Figure4.20FlowingPressureGradientofWaterandFoam 41
Figure4.21ASimpleBeamPumpingSystem 42
Figure5.1DecisionTreefortheExpertSystem 47
Figure6.1TubingFlowingPressureShowingErraticFlowin#10 49
Figure6.2PressureSurveyfor#10ShowingLiquidLevel 50
Figure6.3Turneretal.sCriticalFlowrateforDifferentTubingSizes 52
Figure6.4Colemanetal.sCriticalFlowrateforDifferentTubingSizes 52
Figure6.5NodalAnalysisfor#10withDifferentTubingSizes 54
Figure6.6ErraticFlowBehaviorshowninPressurevs.TimeGraphof#28 55
Figure6.7PressureSurveyfor#28ShowingLiquidLevel 56
Figure6.8NodalAnalysisfor#28withDifferentTubingSizes 56
Figure6.9ActualRatesvs.CriticalRatesfor#28 57
Figure6.10GasandLiquidProductionChartof#28 58
Figure6.11DailyGasProductionBeforeandAfterInstallation 61
Figure7.1ComparisonofCriticalRateEquations 64
Figure7.2ActualRatesvs.CriticalRatesfor#10 65
Figure7.3Powervs.RecoveryforTightReservoir 67
Figure7.4IntermittentFlowCyclesfor#28 68
FigureB.1LiquidProductionChartforPlungerLift 78
xii
FigureB.22PlungerFeasibilityChartfor23/8Tubing 79
FigureB.32PlungerFeasibilityChartfor27/8Tubing 80
FigureC.1IPRCurveofWell#10 83
FigureC.2IPRCurveofWell#28 84
xiii
LISTOFSYMBOLS
Symbols Definition
2
A Tubingarea(ft )
t
2
A Dropletarea(ft )
d
C Dragcoefficient
d
d Dropletdiameter(ft)
F Downwardgravityforce(lbf)
G
F Upwarddragforce(lbf)
D
2
g Accelerationduetogravity(=32.17ft/s )
h Thicknessofliquiddroplet(in)
ID Tubinginternaldiameter(ft)
GLR Gasliquidratio(scf/bbl)
OD Tubingouterdiameter(ft)
P Wellheadpressure(psia)
P Reservoirpressure(psia)
R
P Wellflowingpressure(psia)
wf
q Gasflowrate(Mscf/d)
3 o
R Gasconstant(=10.73psiaft /lbmol R)
xiv
T Wellheadtemperature(F,R)
TR TurnerRatio
V Velocity(ft/s)
3
V Volume(ft )
V Velocityofgas(ft/s)
g
V Velocityofdroplet(ft/s)
d
V Terminalvelocity(ft/s)
t
V Criticalvelocity(ft/s)
c
z GasCompressibilityfactor
Specificgravity
Interfacialtension(dyne/cm,lbf/ft)
3
Density(lbf/ft )
2
Viscosity(lbfsec/ft )
Subscripts:
a Air
c Critical
g,G Gas
l,L Liquid
t Tubing
w Water

xv
CHAPTER1
INTRODUCTION
Liquid loading, by definition, is the inability of a gas well to remove liquids that are
producedwiththegasfromthewellbore.Theproducedliquidwillaccumulateinthewell,
therefore creating a hydrostatic pressure in the well against formation pressure and
reducing production until the well ceases production. In order to reduce these effects of
liquidloadingongasproduction,loadingproblemsshouldbediagnosedintimeanddealt
properlyandefficiently.
One fact about liquid loading is that it can present itself as a problem for high rate/high
pressure wells as well as low rate/low pressure wells. The differences depend on tubing
string size, surface pressure, amount and density of liquids produced along with gas.
Thereforeitisimportanttorecognizeliquidloadingsymptomsatearlystages,anddesign
propersolutionforthegaswellsinordertominimizethenegativeeffectsofliquidsfilling
upthewellbore.
1.1MultiphaseFlow
In order to understand the liquid loading phenomena properly and dealing with it
effectively, it must be understood how liquid and gas behave when flowing together
upwards in the production string of the well. This concept is called multiphase flow.
Multiphaseflowis,basically,aflowphenomenonthatdenotesthereismorethanonefluid
phase flowingthrougha media; inthis case themedia beingtheproduction stringof the
gas well. Multiphase flow is usually represented by four main flow regimes which are
bubbleflow,slugflow,transitionflow,annularmistflow.Theseflowregimesoccurwhen
certain flow velocityof liquid and gasphasesand the amount ofthesephasesrelativeto
eachotherinthemedia,againinthiscasethegaswellproducing.
Figure1.1BasicProfileofMultiphaseFlowintheWell
AnnularMist Flow: The gas phase is the dominant phase in the well and the continuous
one.Liquidispresentamongthegasasamist.Insideofthetubulariscoveredwithathin
layerofliquidtravellingupthepipe.Inthisflow,thepressuregradientisdeterminedfrom
gas.
Transition Flow: Although the flow starts to change from mist to slug therefore the
continuous phase changesfromgastoliquidorviceversa.Liquidparticles may still be in
gasasmistformbutthepresenceofliquiddeterminespressuregradient.
SlugFlow:Thegasisfoundaslargeslugsinliquidbutthedominantandcontinuousphase
is liquid. Gas slugs may cause drops in pressure gradient therefore liquid and gas both
determinepressuregradient.
BubbleFlow:Thetubularinthewellisalmostcompletelyfilledwithliquid.Gasispresent
assmallbubblesintheliquidthereforeitcancausepressuredropsintheliquid,decreasing
pressure gradient along the well. However, the liquid is the continuous phase along the
tubular and completely determines pressure gradient, although presence of gas bubbles
maycausedropsinpressure.
Consideringtheseflowregimes,onemustrememberthatduringitslifetime;itisrarelythe
caseforonlyoneflowregimeispresentinagaswell.Usually,agaswellmaygothrough
almostalloftheseflowregimesduringitsproductivelife.Alsomorethanoneflowregime
may be present at the same time in the well, since gas bubbles will be expanding when
travellingupalongtheproductionstring.Alsoitshouldnotbeforgottenthatflowvelocity
isdirectlyrelatedtocrosssectionalarea,soflowregimesmaydifferaboveandbelowthe
productionpacker,ifthereisone.Anotherpointtoconsideristheflowregimeseenatthe
surface may not be the flow regime near the perforations, considering bottomhole
conditionswouldbedifferentdownhole.
Asstatedabove,asgasvelocitydecreasestheflowregimesgoesfrommisttobubble.Since
the liquid presence is much more in bubble flow, the amount of produced liquid will
increaseastheflowregimeschanges.Thismeans,ofcourse,asthegasratedeclineswith
decreasing reservoir pressure, the amount of liquid produced along with gas will
dramaticallyincrease,increasingthecostofwellalso.Atsomepoint,theincreasingamount
ofliquidwillstarttoaccumulateinthewellastheflowregimedownholeshiftstobubble
flowandincreasingthebottomholepressureinthewell.Thewellwilleventuallybeunable
toovercomethatpressureandstopproducingaltogether.
1.2LiquidLoadingConcept
Asmentionedbefore,thegas,whichisthedominantphase initiallyinthewell,willcarry
theproducedliquidpresentinthereservoirtothesurfaceaslongasthegasvelocityishigh
enoughtoletitdoso.Ahighgasvelocitywillcausemistflowinthewellinwhichliquidis
dispersedinthegas.Thisalsomeanstheliquidinthewellwillbelowrelativetothegasand
will be carried out without accumulating downhole. This will result in a low pressure
gradientinthewellsincethereismoregasthanliquid.Atthispoint,itshouldbenotedthat
whenawellisflowingatahighgasrate,andthereforevelocity,thefrictionalpressureloss
willbehighalso.Thispressurelosswillnotbeabigproblemsincethecomponentissmall
duetolowpercentageofliquidcomparedtogas.Asthegasvelocitydropswithtime,the
liquidcarriedoutalongwithgaswillstarttodropandaccumulateinthewell,causingthe
pressure gradient component to increase. Since high pressure gradient means a high
hydrostatic pressure in the well, the reservoir pressure will encounter a much larger
pressure against itself downhole. Obviously, this will cause a decline in the gas rate and
cripple gas production. Lower the gas rate falls, more liquid will be accumulated and this
holdupwillbecomeacycle,causingthewellceaseproducingeventually.
1.3SourceofLiquids
Onlyasmallnumberofgaswellsproducecompletelydrygas.Thismeansthatalmostevery
gas well produces liquids along with gas even if the produced amount of liquids is very
small.Theseliquidsmaybefreewater,watercondensateand/orhydrocarboncondensate.
Condensatemaybeproducedasliquid,orvapordependingonthereservoirandwellbore
pressure. Produced liquids along with gas may have several sources depending on the
conditionsandtypeofthereservoirfromwhichgasisproduced:
Theremaybeanaquiferbelowthegaszonewhichmayeitherleadtowaterconing
orwaterencroachment.
The source of liquids may be another zone or zones, especially if the completion
typeofthewellisopenhole.
Thewaterproducedalongwithgasmaybefreewaterpresentintheformation.
Dependingonthereservoir,bottomhole andtubinghead pressures water and/or

hydrocarbon vapor may enter the well and condense while travelling up the
productiontubing,comingoutasliquid.
1.3.1WaterConing
If the production rate of any vertical or deviated gas well is high enough to result in a
drawdownpressure highenoughtopull thecontact waterin thereservoirbelow thegas
even if the perforations do not extend to the underlying zone. Horizontal wells generally
reducewaterconingeffectsbutitcanstilloccuranditiscommonlycalledaswatercresting
insteadofwaterconing.
1.3.2AquiferWater
If the reservoir has a waterdrive mechanism, the aquifer giving pressure support to
produced gas will eventually reach the perforations and into the wellbore. This
phenomenon is also called water encroachment. After water reaches wellbore, liquid
loadingproblemswillrise,reservoirpressurewillstarttodropsharperthanbeforeasthe
drivemechanismisdepletingwithproducedgas.
1.3.3CondensedWater
Since nearly every reservoir contains free formation water, natural gas present in the
reservoirmaybesaturatediftheconditionsaresuitableforwatertodissolveinnaturalgas.
Inthiscase,waterwillenterthewellasvapordissolvedinnaturalgasandtherewillbeno
or very little water in liquid phase at the bottom, near the perforations. As the solution
flowsthroughtheproductionstringthewaterwillstartcondensingifthetemperatureand
pressureconditionsinthewelldropbelowdewpoint.Iftheamountofcondensedwateris
high in the well, it will create a high hydrostatic pressure in the string, increasing the
pressure,thereforecausingwatersolubilityingastodecreaseevenmoreandcausingmore
watertocondense.Eventually,condensedwaterwillaccumulateatthebottomofthewell.
1.3.4CondensedHydrocarbons
Just like water, hydrocarbons that are in liquid phase at atmospheric conditions can also
enter the well in vapor phase. As the gas solution flows to the surface, vapor state
5
hydrocarbons may start condensing when or if conditions drop below dew point. At this
time, the condensed hydrocarbons are shortly called condensate. Condensate, although
less than water, has a much higher pressure gradient than gas, so it will create a higher
hydrostaticpressureandeventuallystartloadingupthewelljustlikewater.
1.3.5WaterProductionfromAnotherZone
Especiallyinopenholecompletions andsomecaseswellswithmultipleperforations,it is
possibletoproduceliquidsfromanotherzoneunintentionally.
1.3.6FreeFormationWater
Differentthantheconditionstatedabove,watercanalsobeproducedalongwithgasfrom
thesameperforations,ifthereisfreewaterinthereservoir.
As mentioned, there are different sources for liquid loading, and there exist various
solutionmethods forremovingliquids oreliminatingliquid loadingproblems ingaswells.
However, there is an uncertainty in which methodology will give the best result for a
particulargaswell.Thisstudyaimstoaddressthisissue.
CHAPTER2
LITERATUREREVIEW
Aproducinggaswellceasingproductionprematurelybecauseofliquidloadingwouldmean
afinanciallossandtheinefficientuseofresources.Inordertoovercomethisissue,first,it
mustbe identifiedproperly.Itisknown thatasreservoirpressuredeclines itis easier for
the well to be killed by loaded liquids since the velocity of the gas passing through the
productionstringwilldecrease.In1969,Turneretal.1analyzedwhetherthegasflowrate
wouldbesufficienttoremovetheliquidscontinuouslyfromgaswells.Twophysicalmodels
are proposed for the analysis of the removal of liquids; liquid film along the walls of the
pipeandsphericalliquiddropletsentrainedintheflowinggascore.Acomparisonofthese
two models with the field test data concluded that liquid droplet theory yielded a better
modelforpredictingtheonsetofliquidloading.Itisalsoconcludedthatthereexistsagas
velocitysufficienttoremovethedropletscontinuouslytoavoidloadup,buta20%increase
should be added to insure removal of all drops. Coleman et al.2 proposed a new look at
predictingloadupin1991,whichisbasicallyTurneretal.smodelwithoutthe20%increase
intheminimumgasflowrate,knownascriticalrate.Itisalsostatedthatliquid/gasratios
below22.5bbl/MMscfhavenoinfluenceindeterminingtheonsetofloadup,meaningthe
gasflowrateisthedominantfactor.
In2000,Nosseiretal.3suggestedanewapproachforaccuratepredictionofloadingingas
wells under different flowing conditions. Turner et al.s basic concepts are adopted but
differentflowconditionsareconsideredresultingindifferentflowregimes.Widevariation
offlowconditionsingaswellswouldmakeitdifficulttoassumeaconstantflowregimefor
allwells and conditions,thereforetheirnewapproachmostlyconsistedofacasebycase
basis.Uponcalculatingthecriticalflowrate,itisstatedtheappropriateequationshouldbe
appliedforeachcase.Inwellswiththepossibilityofhavingmorethanoneflowregime,itis
recommended that the calculations are carried out at the wellhead pressure since gas
slippagewillbeatmaximumnearthesurface,andalsowatershouldbeconsideredasthe
loadingphasetoguaranteeremovingallthedropletsoflighterphasesalso.
Anewviewoncontinuousremovalofliquidsisproposedin2001,whenLietal.4adopted
theliquiddropletsentrainedingascoretheorybutpredictedtheliquiddropletstendtobe
flat instead of spherical (shown in Figure 2.1) and deduced new simple formulas for the
continuous removal of these droplets accordingly, for field application. Models and
approachesbyTurneretal.andColemanetal.didnottakethedeformationofafreefalling
droplet into consideration. The results calculated from these formulas were smaller than
findingsofTurneretal.However,theystatedthatpredictedresultswereinaccordwiththe
practicalproductionperformanceofChinasgaswellsdealingwithliquidloadingproblems.
Figure2.1Shapeofentraineddropmovementinhighvelocitygas4
In2003,Veekenetal.5acceptedTurnersmethod,butdevisedaratiotermcalledTurner
Ratio(TR)whichistheratioofactualflowrateandminimumflowratepredictedbyTurner
etal.forcontinuousremovalofliquids.Veekenetal.scorrelationdataincludeddeviated
wells, also and the predicted the critical rate for deviated wells is about the same for
vertical wells. An inflow performance parameter is also added to their Turner Ratio
equation which allows evaluating critical flow rate at bottomhole conditions. Veeken et
al.s model showed a much higher flow rate is needed than Turners (and therefore
Colemans)modelpredictedtoremoveliquidsproperlyandcontinuouslyatlowpressures.
Belfroidetal.6(2008)statedthatwhenmakingpredictionsoncriticalflowrates,inclination
angle, flow regime transitions, tubing outflow and reservoir inflow relations should be
taken into account. Also, they argued that the influence of dynamic disturbances on the
stability is not taken into account by the classical prediction models. Belfroid et al.
8
concludedthattheonsetofliquidloadingisdeterminedbythetransportoftheliquidfilm.
Theystatedforlargerinclinationstheeffectofgravityisreducedandthereforecriticalgas
rate will be lower; however, at large inclinations, the liquid film starts to thicken at the
bottomofthetubecomparedtotop,whichincreasesthecriticalgasflowrate.Thisresults
in erroneous flow rate calculations in classical models. Also it is concluded that high
permeabilityreservoirswillshowliquidloadingbehaviormuchfasterthanlowpermeability
reservoirs.Theirresultsregardingcriticalflowrateweremuchhigherthanclassicalmodels
especiallyinhighpermeabilitylowpressurereservoirs.
In 2009; Sutton et al.7 proposed a guideline for the proper application of critical velocity
calculations.Theystatedthatalthoughfieldpersonnelgenerallyusesconditionsatthetop
of the well as an evaluation point for calculating critical flow rate for a well, a change in
geometry downhole or other conditions may lead to erroneous conclusions. Using
conditions at the bottom with fundamental equations requires accurate correlations for
PVT properties such as surface tension and density for gas and liquid phases. They
concluded that for almost every case, the critical velocity can be calculated using water
propertiessincewaterhasahigherdensitythanliquidhydrocarbons;gaswillbeabletolift
hydrocarbonsifitisabletoliftwater.Theevaluationpointfordeterminingcriticalvelocity
canbeeitherthewellheadorbottom.Theystatedwellheadconditionsshouldbeusedin
highpressurewells(Pwhfgreaterthan1000psia)andbottomconditionsshouldbeusedin
low pressure wells (Pwhf less than 100 psia) when calculating critical velocity. For wells
producingfreewater,usingbottomconditionswouldbemoreaccurate.Alsoaccordingto
thestudy,theoriginalsafetyfactorTurneretal.providedisneededtoensurethewellis
unloadedalongtheentireflowpath.
Thegeneralaimofalltheseresearchistodeterminetheconditionsforremovingliquidsin
gas wells continuously. However, as liquid loading problems in a well progress after a
certainpointitmaybeimpossibletokeepthewellflowingonitsown.Whenthathappens,
thereareavarietyofsolutionsthatcanbeusedinordertosolveliquidloadingproblemsof
thewell.Lea&Nickens8(2004)compiledsomeofthesesolutionsinastudytodescribeand
discusstheproblemsofliquidaccumulationinagaswell.Someofthesemethodsinclude
sizingproductionstringstochangetheflowpatternand increasegasvelocity,installinga
compressor,plungerliftmechanism,andfoaming.Theyproposednodalanalysisasaliquid
loading prediction method and stated at initial stages surfactants can be tried as a cost
effective solutions after evaluating economics. Smaller diameter tubing may be used to
increasetheflowingvelocity;however,eventuallyhastobedownsizedevenmore.Plunger
lift may be preferred over tubing sizing since it can be used in already installed larger
tubing.Itisconcludedalthoughthereareseveralmethods;noneofthemistheultimateor
only solution since solving liquid loading problems is more of a case by case project
involvingdifferentreservoirparameters,wellheadconditionsandliquidproperties.
As Belfroid et al.6 have stated in their study, even though virtually all of the worlds gas
wells are either at risk of or suffering from liquid loading, the modeling of liquid loading
behaviorisstillquiteimmatureandthepredictionoftheminimumstablegasratenotvery
reliable. Therefore, predicting onset of liquid loading and solving loadup problems are
critical and only credible when approached on a case by case basis and constructed a
methodologyaccordingly.
10
CHAPTER3
STATEMENTOFPROBLEM&SCOPE
Thisstudyisundertakentoinvestigatetheeffectsofliquidloadingondepletinggaswells,
determinecertainmethodstominimizetheseeffectsinordertoproposeadecisiontreeas
the algorithm of apossible expert systemtochoose aproper solution foreach individual
gas well. The critical rate theory for unloading liquids from wells is discussed due to
discrepanciesindifferentmodels.Theobjectivesofthestudyare;
Determinethebestmethodstopredictliquidloadingandrecognizethesymptoms
earlyontoavoidproductionlosses.
Evaluating completion methods and production practices to find an optimum
designwhendealingwithgaswellswithliquidloading.
Comparesolutionsonliquidloadingsolutionsandartificialliftmethodstoseethe
advantagesanddisadvantagesonparticularcases.
Designadecisiontreeforthealgorithmofapossibleexpertsystemforsystematic
selectionofproperliquidloadingsolutionsundervariousconditions
11
CHAPTER4
THEORY
4.1PredictingLiquidLoading
Over the life of a typical gas well, gas flow rate will eventually decrease while liquids
produced along with gas will increase. At some point, this situation would cause
accumulation of liquids at the bottom of the well since the producing gas rate would be
insufficienttoliftalloftheliquid,whichwillleadtoerraticflowbehaviorandinevitablyloss
ofproduction.Ifthesymptomsofliquidloadingarerecognizedatearlystages,lossesingas
productionthatmayeventuallycostthelifeofthewellmaybeavoided.Aproperanalysis
ofthedeclinecurveofagaswellcanbeinformativeaboutdownholeflowproblemsofthe
well.
Thechangesinthegeneralshapeofthedeclinecurve(asseeninFigure4.1)ofthewellcan
be an important indication of loading, if properly analyzed. Characteristically, a typical
decline curve of a dry gas production well should be a smooth exponential curve as
reservoir depletes over time. During decline of the curve sharp changes and fluctuations
indicate possible liquid loading downhole due to erratic flow behavior caused by liquid
slugs.Figure4.1showstheexpecteddeclinecurveandpossiblefluctuationsduetoliquid
loading.Eventuallythesesharpdeclineswillcausethewelltodepleteearlierthanreservoir
estimationsandpossiblydieprematurely.Installingmethodsremedialforliquidloadingcan
restorethedeclinecurveofthewelltoitsoriginalshape.
12
Figure4.1ATypicalGasWellDeclineCurvealongwithIndicationofLoading
If liquids begin accumulating in the bottomhole, the increased pressure caused by

hydrostaticheadpressureoftheliquidontheformationwillcauseadropinsurfacetubing
pressure.Inwellswithpackerlesscompletion,theincreasedpressureinthetubingwould
causegasbubblestostartaccumulatinginthetubingcasingannulus,causinganincreasein
the casing surface pressure, contrary to tubing pressure. Therefore, in packerless
completionsanincreaseincasingsurfacepressureandacorrespondingdecreaseintubing
surface pressure could indicate possible liquid loading. Although this method is a good
indicatorwhenthepressuresareobservedclosely,apressuresurveyshouldgivedefinitive
dataonthemattertoseeifthewellisreallybeganloading.
Aflowingorstaticwellpressuresurveydonewithelectronicdownholegaugesispossibly
themostaccuratemethodtodeterminewhetherthewellisloadingwithliquids.Pressure
surveys,byusingdownholegauges,measurethepressurewiththecorrespondingdepthof
thewellwhilethewellisflowingorshutin.Thedatacanbeusedtoconstructapressure
gradientgraph,whichisafunctionofthedensityofthefluidinthewellatthatparticular
depth.Theconstructedpressuregradientcurvewillexhibitasharpchangewhenthefluid
in the well turns to liquid from gas since the density of liquids are much higher than the
13
densityofgasesoccupyingthewell.Thepressurevs.depthgraph(Figure4.2)willalsogive
theliquidlevel,sincethepointwherethesharpchangeoccursisbasicallythepointwhere
theliquidisloadedinthewell.
Figure4.2ATypicalPressureSurveyGraph
Insummary,agaswellsufferingfromliquidloadingproblemsgivesmanyindicatorswhich
provideearlywarning.Theproducinggaswellsshouldbemonitoredregularlyinorderto
catchtheseindicatorsatearlystagestopreventliquidloadingproblemsfromdamagingthe
reservoirpermanentlyandcausingprematuredeclinesinproduction.
4.2CriticalRateTheory
As stated earlier, when producing gas phase hydrocarbons from a reservoir, some liquid
phase hydrocarbons which we called condensate and also liquid phase water may be
produced along. Presence of liquids in the well will put a pressure against the reservoir
pressure and if the well is unable to unload the liquid, it will die unless some certain
measuresaretaken.Also,eveniftheflowingpressureofthewellishighenoughtounload
theliquids,therestillmaybesluggingofdiscontinuityintheflowduetotheflowregime.At
14
thispoint,thefirstthingtoconsidershouldbedeterminingifthewellwillbeabletounload
this liquid on its own. The answer to that question lies within the critical velocity theory.
Many authors have suggested several methods to determine if the flow rate of a well is
sufficient to remove the liquid phase materials produce on a continual basis. In 1969,
Turner et al.1 proposed two physical models for removal of liquids; (1) liquid forming a
continuousfilminsidethewalloftheproductionstringmovedupwardbyinterfacialstress
and (2) liquiddropletspresent inthe stringasfree falling particles moving up becauseof
thehighvelocityofthegas.Afterdevelopingthesetwomodels,Turneretal.comparedthe
actual field data with the models independently to see which one is a closer match and
whichisthecontrollingmechanismfortheremovaloftheliquids.
4.2.1TheContinuousFilmModel
Liquidphaseaccumulationonthewallsof stringduringatwophaseflow isofinterestin

theanalysisofliquidremovalaccordingtotheworkofTurneretal.Theannularliquidfilm
mustbemovingupwardalongthewallofthestringinordertokeepthewellfromloading.
The minimum gas rate necessary to accomplish this is by calculating the velocity of the
liquid,thevelocityofthegasandtheshearstressinbetween.Turnerconcludedthatthe
predictionsofthefilmmodeldonotprovideacleardefinitionbetweentheadequateand
inadequaterates.
4.2.2LiquidDropletModel
ThestudiesofTurneretal.statethattheexistenceofliquiddropsinthegasstreampresent
adifferentproblem,whichisbasicallydeterminingtheminimumgasflowratethatwilllift
thedropsoutofthewelltothesurface(Figure4.3).Accordingtothestudy,afreefalling
particle reaches a terminal velocity which is the maximum velocity it can attain against
gravity.Therefore,thatterminalvelocity,orinothertermsthecriticalgasvelocitywhichis
determinedbytheflowconditionsnecessarytoremovetheliquidsonacontinualbasis,is
basedondrag&gravitationalforcesonthedroplet.
15
Figure4.3LiquidDropletMovement1
ThestepbystepderivationofTurneretal.sliquiddropletmodelcanbefoundinAppendix
A.Theirexpressionsontheliquiddropletmodelcanbesummarizedasfollows:

1.593 (1)
AccordingtoTurneretal.1,analysisofdatarevealedthatthesefactorsrequiredanupward
adjustmentof20%tofitthefielddata.Thentheequationbecomes:

1.912 (2)
However,in1991,Colemanetal.2suggestedthat20%(18.92%tobeexact)safetyfactoris
unnecessaryandstatedthattheinitialequation(1)fitsbettertothefielddata.Colemanet
al.alsoconcludedwellheadconditionscanbeusedforpredictingloadupconditions,unless
tubing/packer has significant distance from the completion interval, flowing conditions of
the largest segment should be used to predict the wellbore critical rate. For field
applications, Turner et al. consolidated some of the fourth root variables into constants.
Usingconstantsforthefourthrootofsurfacetensionanddensityforbothcondensateand
water; two simple separate equations for condensate and water can be used. Combining
thesefindingsandfollowingtheworkofbothTurneretal.andColemanetal.thefollowing
criticalrateequationisbeingusedtodetermineloadup:
16
.
(3)
Where:

. .
(4)
.
. .
(5)
.
Turneretal.finallystatedthattheliquid/gas(L/G)ratiohasnoinfluenceindeterminingthe
liquidloaduponsetinthewellaslongastheL/Gratioisbelow130bbl/MMscf.Colemanet
al.statedthattheliquid/gasratiohasnoeffectifitisbelow22.5bbl/MMscf.
17
4.3TubingSizeOptimization
Proper tubing size selection is crucial to effectively produce gas from the reservoir and
maximizerecovery.Tubingsizeselectionmaybesomewhatsimplerinverticalwellswitha
single pay zone and single fluid flowing through the wellbore. However, in wells with
multiple reservoirs and liquid loading problems, tubing size selection can become quite
complex.Atfirstglanceatthecriticalrateandterminalvelocityequations,smallertubing
sizes can be economically favorable with time where liquid loading will be more
problematic since reservoir pressure will deplete eventually, causing the well to loadup
withliquidsproducedfromthereservoir.Theaimistodetermineasimple,fieldapplicable
modeltoproperlyselectoptimumtubingsizeifpossible.
UsingcriticalrateequationofTurneretal.andapplyingthefielddata,terminalvelocityand
thencriticalrateofthewellcanbecalculatedfordifferentsizesoftubingstringandcanbe
plotted(Figure4.4).Thecriticalratethancanbecomparedwiththeactualproducingrates
of the well and determined if the gas can lift the liquid from the wellbore with smaller
tubing inside. Gunawan et al.9 showed that field data validates Turner method when
predictingcriticalgasflowrateswithdifferenttubingsizes.ItisseeninFigure4.5thatthe
datasetfitsbetterwiththe20%adjustmentColemanetal.deemunnecessary.
Figure4.4CriticalFlowratesforDifferentTubingSizes9
18
Figure4.5CriticalGasFlowratesforTurner&Coleman9
4.3.1NodalAnalysis
Liquid loadup can also be determined by nodal analysis. Since critical gas rate equations
only give a simple idea fortheminimumrates, nodal analysis will bemoredetailed since
normallyinawell,gasmayhavetoflowagainstmanyrestrictionsotherthanliquiditself,
suchasdifferenttubingsizes,subsurfacesafetyvalves,rockmatrixofreservoiretc.Each
componentthatgasandliquidsflowthroughwillhavepressurelossdependingonflowrate.
In order to determine overall well performance, all of these components must be
consideredasasystem.Nodalanalysisdividesthissystemintotwosubsystemsatacertain
locationcallednodalpointorsimplynode.Oneofthesesubsystemsconsidersinflowfrom
reservoirtothenodalpointselectedwhiletheothersubsystemconsidersoutflowfromthe
nodal point to the surface. Each subsystem gives a different curve plotted on the same
pressurerategraph(Figure4.6).Thesecurvesarecalledtheinflowcurveandtheoutflow
curve, respectively. The point where these two curves intersect denotes the optimum
operatingpointwherepressureandflowratevaluesareequalforbothofthecurves.
Itispossiblethenodalpointcanbelocatedanywhereinthesystem.However,practically,
locating nodal point at the bottom hole (at the midperforation depth) is very common
19
since that way the inflow curve represents the flow from reservoir into the hole and the
outflowcurverepresentstheflowfromthebottomholetothesurface.
Figure4.6TypicalNodalAnalysisCurves
Thenodalanalysiscanbeusedwithbothsingleandmultiphaseflowequations;moreover,
correlations of different components such as well completion and skin effects and also
effects of surface components can be implemented into nodal analysis. The information
gatheredcanbeusedtodetermineandevaluateoverallwellperformanceforavarietyof
different conditions that eventually will lead to optimum completion and production
practices.Itisanimportantpracticenotonlyforanalyzingtheeffectsofliquidloadingbut
alsoforfindingpossiblesolutionstotheproblem.Asmentioned,nodalanalysiscanbeused
toanalyzetheeffectsofdifferenttubingsizesanddifferentflowconditions.Moreover,itis
useful for determining the effects of surface pressure on the system, since excessive
surfacepressurecancauseabackpressureonthereservoir.
Asmentionedabove,thenodalpoint,thepointthatdividesthenodalanalysissysteminto
tosubsystems,iscommonlyplacedatthebottomhole.Inthatcase,theoutflowcurvethat
canalsobecalledthetubingperformancecurve(TPC)showstherelationshipbetweenthe
pressure drop in the tubing string and surface pressure value. The pressure drop in the
20
tubingstringbasicallyconsistsofthesurfacepressurevalue,thehydrostaticpressureofthe
loadedliquidinthestringandthefrictionalpressurelossduetoflow(Figure4.7).
Figure4.7TubingPerformanceCurve10
TheTPCpassesthroughaminimumatthemiddleofthecurve.Thetotaltubingpressure
loss increases due to increased friction losses at the higher flowrates to the right of that
minimumpoint.Theflowtotherightoftheminimumisgenerallyinthemistflowregime
thateffectivelytransportssmalldropletsofliquidstothesurfacebecauseofhigherrates.
AtthefarleftoftheTPCtheflowrateislowandthetotalpressurelossisdominatedbythe
hydrostatic pressure of the liquid column loaded in the well. The flow regime at the left
mostsectionofthecurveistypicallybubbleflow,whichallowsliquidstoaccumulateinthe
wellbore.SlightlytotheleftoftheminimumintheTPC,theflowisoftenintheslugflow
regime.Inthisregimeliquidistransportedtothesurfaceperiodicallyintheformoflarge
slugs.Fluidtransportremainsinefficientinthisunstableregimeasportionsoftheslugsfall
backtothewellboreasthepressuredropsandmustbeliftedagain.Thisfallbackresultsin
a higher producing bottomhole pressure. It is common practice to use the tubing
performancecurvealone,intheabsenceofuptodateandaccuratereservoirperformance
data, to predict gas well liquid loading problems. The general idea when interpreting the
curveisthatflowratestotheleftoftheminimumareunstableandpronetoliquidloading
problems. Flow rates to the right of the minimum of the tubing performance curve are
21
considered to be stable and significantly high enough to effectively transport produced

liquidstothesurface.10Thismethodis considerablyinexactbutintheabsentofaccurate
reservoir data it can be useful for predicting and determining liquid loading problems.
Whenreservoirperformancedataispresent,the intersectionpointofthetubingoutflow
curveandthereservoirinflowperformancecurveallowsanaccuratedeterminationofthe
point the well is flowing and what would be the optimum pressure and rate values.
Calculatingbottomholeflowingpressurefordifferenttubingsizesfordifferentproduction
ratesandplottingthesevaluesonthesamegraphwiththereservoirinflow(IPR)curveto
determine the optimum tubing size for the well to produce gas and remove the liquid
effectivelyasshowninFigure4.8.
Figure4.8NodalAnalysisGraphforDifferentTubingSizes10
After constructing the nodal analysis plot for all the different tubing sizes, the optimum
flowing tubing pressure and desired production rate could be selected accordingly. Also,
analyzing the plot would give information about the expected flow conditions for certain
pressureandproductionratevalues.Calculatingthecriticalratevaluesforeachtubingsize
would letus predict if theselected tubingwouldbeabletolift the liquidthatenters the
wellcontinuouslyorletthewellloadupanddie.Onesignificantadvantageofusingnodal
analysis determining proper tubing size is that it would allow to see the lifespan of the
22
tubingselected,thereforeitcanbepredictediftheselectedtubingsizewillbeadequateto
lifttheliquidandproduceeffectivelynotonlytoday,butforanacceptableperiodoftime.
Animportantthingtoconsideriscalculatingthecriticalratebothatsurfaceanddownhole
conditionsforaselectedtubingsizetomakesurethegasflowsabovethecriticalvelocity
frombottomtotopandnootherrestrictionsarecontributingtoloadup.Alsoselectinga
smallertubingdiameter may notcausea sudden increase intheproductionrate,butthe
new decline curve of the well will give an idea when conclusions are made on the new
stringinstalled.
4.4WellCompletion&ProductionRateOptimization
Properoptimizationofwellsforbothoilandgasproductionisahighlycomplexissuethat
involves countless parameters for different cases. Also, it is important to remember that
there is no absolute solution when dealing with oil or gas wells since processing and
interpretationofwellboredataisnotanexactmatter.Consideringcircumstancesitisnot
possible to propose an optimized completion design that could fit every case. Obviously,
thebestwaytoaddressthisissuewouldbedealingwithasinglewellbasedonproblems
encounteredby offering certainremedial measuresand solutions.Whendealing with gas
wells having liquid loading problems, there are a few key factors that need attention
regardingcompletionoptimization.
4.4.1TheSettingDepthoftheEndofCompletion
Itisgenerallyrecommendedthattheendofthetubingstringinthewellshouldbesetright
atthetopoftheperforations.Manystudies,likeChristiansenetal.s11(2005)suggestthat
the liquid transport is severely constrained in the casingtubing junction because the gas
velocityismuchlowerthanthecriticalvelocityneededduetolargecrosssectionalareaof
thecasing.Also,iftheperforationsarefloodedwithwatercontinuously,thepermeability
of the formation will severely decrease due to fluid invasion (Figure 4.9 a). There are
possible measures to be taken to overcome this issue to a certain level but setting the
tubingjustatthetopoftheperforations(orpayzoneinopenholecompletions)wouldbe
thebettersolution(Figure4.9b).Ontheotherhand,settingthetubingtoodeepcould
23
causeproblems.AccordingtoLeaetal.10(2008),itisnotrecommendedfortheendofthe
tubingtopassthetoponethirdofperforations.Ifthetubingissettoodeep(Figure4.9c),
liquid would collect overthe perforations during shutin. When the well brought back on
production, the large volume of collected liquid must enter the tubing string where the
crosssectionalareaofthetubingstringisrelativelysmall.Thereforethehighlevelofthe
collectedliquidwillcausegreaterhydrostaticheadpressuremakingthewellverydifficult
to flow. If the end of the string is set below the entire perforation interval, the pressure
during shutin cannot push the liquids below to enter the tubing string and it would be
impossibleforthewelltoflowandunloadthecollectedliquid.
Figure4.9TheEffectofSettingtheEndofTubing10
4.4.2TheEffectsofPerforationInterval
Determiningtheimpactofperforationintervalonliquidloadingingaswellsisparticularly
crucial in water drive gas reservoirs. There are basically two possibilities which are either
perforating long intervals and producing at high rates in order to minimize abandonment
pressureorperforatinglimitedintervalsatthetopandproducingatlowratestoprevent
waterconingwhichwouldresultearlyabandonmentforthegaswell.Animportantissueto
considerwhenchoosingeitherapproachshouldbedeterminingtheaquiferencroachment
rate if it is a water drive gas reservoir. It is known that lowering reservoir pressure by
producinggasathigherratesthanaquiferencroachmentcouldincreaseultimaterecovery
24
significantly.Thatmeansthepossibilityofsurpassingtheaquiferencroachmentratebygas
productionrateisanimportantconsiderationonultimategasrecovery.
Thestudy byMcMullan &Bassiouni12(2000) showed thatone ofthemain characteristics

that affects recovery when trying to optimize the perforation interval is reservoir
permeability(Figure4.10).However,verticaltohorizontalpermeabilityratio,fluiddensity
contrast,relativepermeabilityandformationdipdidnotaltertheirconclusionsaccording
to the sensitivity analysis. In their investigations, basic singlephase fluid flow equations
providedinsightontherelativityofgasandwaterflowintothewell.UsingworkofCraft&
Hawkins13itcanbesummarizedasDarcyslawonsteadystateradialflowforbothwater
andgas:
.
(6)
.
(7)
These two Darcy equations describe steadystate conditions and one dimensional radial
flow of water and gas; therefore they are not adequate to describe the complex three
dimensional, unsteadystate multiphase flow, which is exactly the case in a typical gas
reservoir.However,usingtheseequationsprovideinsightonthenatureofwaterandgas
flowrelativetoeachother.
The analysis of the study showed that unless the perforation interval extends to a water
zone,asmallwatergasratiocanbeexpecteduntilnearlytheentireperforationintervalis
flooded with water12. As gas has an extremely lower viscosity relative to liquids, in high
permeabilityreservoirs,veryhighgasratescanbeachievedevenfromaverythinlayerof
flowingintervalthatproducesgas(Figure4.10).Also,itisstatedthatincreasedperforation
interval does not have a crucial impact on ultimate gas recovery however; in high
permeability systems longer perforation intervals may cause increased ultimate water
production(Figure4.11).Theirfindingsfavorthewelltobecompletedwithsufficientlylong
25
perforationintervals,notextendingtowaterzone/zones,inordertoobtainmaximumgas
flowrateandinsuremaximumgasrecovery.
Figure4.10ImpactofLengthofPerforationIntervalonCumulativeWaterandGas
ProductioninYears12
Figure4.11ImpactofPerforatedIntervalandPermeabilityonUltimateGas(left)and
Water(right)Recovery12
4.4.3TheEffectsofProductionRate
Although gas production rate is dependent considerably on the length of perforation

intervalduetodeliverability,itisanothersubjectofinterestsinceultimategasrecoveryis
directlyrelatedwithgasproductionrate.McMullan&Bassiounishowedintheirmodelthat
generally higher production rates do not impair ultimate gas recovery, therefore in low
water disposal situations gas rates should be maximized to insure maximum ultimate gas
26
recovery.Inlowtomoderatepermeabilityreservoirs,itisshownthatwhengasproduction
rate increases, water production in the well is decreased because gas flows much easier
through perforations due to significantly lower viscosity. However, this case does not
necessarilyappliestoreservoirsthathavepermeabilityvaluesof1000millidarciesormore,
sincepermeabilitythathighcausesasignificantincreaseinwaterproductionratealso.Itis
obvious that constraining gas production rates will increase the life of a well, but the
economicconsequencesofdelayedrecoveryarealsosignificant.Restrainedgasproduction
rateswillcausethewelltohaveanincreasedlifebutcouldalsomeandecreasedultimate
gasrecoverysinceitcanhavedetrimentaleffectsonthereservoir(Figure4.12).
Figure4.12ImpactofGasProductionRateandPermeabilityonWellLife(left)and
UltimateGasRecovery(right)12
These findings are in consistency with the effects of length of perforation interval on
ultimate gas recovery. However, in high water disposal cost situations, it should be
recognized that long perforation intervals and elevated gas production rates in high
permeabilityreservoirscouldleadtoelevatedwaterproductionandthereforemuchhigher
wellcosts.
4.5OverviewofSolutionstopreventLiquidLoading
Liquid loading in a gas producing well is a progressing problem as reservoir pressure

depletes continually with produced gas and eventually the well will inevitably need an
artificial lift method to lift the loaded liquid from the well to resume gas production.
Although a properly designed tubing string can increase gas velocity to exceed critical
27
velocity and lift the produced liquids, this may not be a long term solution since, as
mentioned, the reservoir pressure will keep decreasing to a point where it would be
impossible forthe kinetic energyof thegas alone will not besufficient enoughtolift the
produced liquids completely. Different solutions should be evaluated and compared in
order to find the best course of action when dealing with wells that have liquid loading
problemstoachievethehighestultimategasrecoverypossibleforthewell.Inthissection,
several wellknown solutions or remedial measures to prevent liquid loading in gas wells
will be discussed and evaluated in order to find which particular solution is the best
solutiontowhichparticularcase.
4.5.1VelocityStringApplication
Avelocitystringisbasicallyatubingstringwithasmallerdiameterruninsidetheoriginal
largediameterproductionstring.Itisusedasaremedialmeasure,sincereducingtheflow
area of gas will cause the velocity to increase and exceed the critical velocity which is
neededforcontinuousremovalofproducedliquidsinthewellbore.Applicationmaydiffer
as velocity string installation can be up to the surface of just up to a certain point in the
currentproductionstring,asseeninFigure4.13.
The study of Arachman et al.14 (2004) showed that especially for bigbore completions
velocity string installation can be very beneficial. It is generally less expensive than other
solutions and treatment methods for liquid loading, since it could be done in a livewell
withcoiledtubing.However,velocitystringapplicationsarecriticalbecauseasdiameterof
the tubing decreases, the pressure loss value due to friction will increase which would
cause high pressure drop and limited gas production rates. The solution to this problem
would be installing the velocity string from the perforation interval up to a certain point
insteadofinstallingitallthewaytothesurface.Ifthevelocitystringistooshort,however;
itcouldbeinsufficienttolifttheproducedliquidseffectivelyandwouldneedreplacingwith
another longer or smaller string. Also, it may not be a permanent solution as reservoir
pressure continues to deplete an even smaller diameter tubing string would be needed.
These criteria make the velocity string an inexpensive solution with critical design
considerations.
28
Figure4.13VelocityStringApplicationSchematic14
4.5.2Compression
Compression is a vital application in all gas well production practices as lowering the
surface pressure will also lower bottom hole flowing pressure causing an increase in gas
productionrate.Compressingthewellcansubstantiallyincreasetheultimategasrecovery.
However,thisliftingmethodrequiresaninitialinvestmentforthecompressorandalsohas
relatively expensive operating costs for maintenance and power needed for keeping the
compressor running. Using this lifting method as a liquid loading solution can provide
beneficial.
Asmentioned,compressorsincreasegasproductionratebyloweringthesurfacepressure
and bottomhole flowing pressure. This means an increase in gas velocity and therefore
betterremovingofliquidscollectedatthebottomofthewellbore.Theremovalofliquids
andthedecreaseinthebottomholepressureexposesmoreofthegasinthereservoirto
productionwhichwasinitiallyunavailable.
29
Inordertolowersurfacepressurewiththehelpofcompression,energyisrequiredinterms
of horsepower.Energyneeded in terms ofhorsepowerisdirectlyrelatedtotherationof
suctionanddischargepressurealsoknownasthecompressionratio.Assuctionpressureof
the compressor decreases or the discharge pressure increases, the amount of energy
needed increases dramatically. Lea10 constructed the table for the energy required to
compressgasatdifferent surfacepressuretoapipelinethathas 1000psigaswell asthe
percentageofcompressedgasrequiredtopowerthecompressor:
Table4.1CompressionHorsepowerandFuelGas10
Combining the amount of horsepower required at a given pressure (Table 4.1) with the
critical rate equations of Turner et al., it is possible to estimate the minimum amount of
energyrequiredtokeepthewelldrybyremovingproducedliquids.Figure4.14,sensitivity
analysis for different tubing sizes shows the compressor energy requirement to keep the
gasvelocityabovecritical.
30
Figure4.14EnergyRequiredforDifferentTubingSizestoStayaboveCriticalRate10
Theeffectofpermeabilitycanbevariableincompressionapplications.Inhighpermeability
reservoirstheaimofinstallingacompressorcanbeacceleratingtheproductionrateaswell
askeepingthewelldry.Wellswithhighproductivityvaluescanmaintainproductionrates
above critical until the very end of their life. These factors are important regarding the
optimizationofcompressionapplications.Also,anythingthatcausesasignificantpressure
dropduetofrictioninthewayof suctionfromthesurfacetothebottomholewillimpair
the efficiency of the compressor. Restriction in the surface and in the well would cause
increasedenergyrequirementsandreducedpowerinliftingthereforewouldcausequicker
loadingofthewell.
Compression and reducing surface pressure is generally one of the first solutions used
during the production life of a gas well regardless if it has liquid loading problems. The
importance of compression applications is that it could be used not only for keeping the
well free of liquids but also increasing production rate that has decreased with depleting
gasreservoir.Moreover,compressionapplicationscouldbeusedwithothermethodsand
remedialtreatmentssuchasfoamingagents,gaslift,plungerlift,beampumping,electrical
submersible pumps and velocity strings. However, different wells will give different
responsestoanyparticularapplicationanditiscrucialthatthecompressortype,sizeand
properties are selected properly and optimized for maximum efficiency. System nodal
31
analysis can be a useful tool determining the best course of action when choosing
compressors.
4.5.3PlungerLift
Plungerliftisanintermittentartificialliftmethodandaliquidloadingsolutionthatusesthe
energyofthegasreservoirtoproducetheliquidscollectedatthebottomhole.Aplungeris
apistontypetoolthattravelsfreelyinthetubingstringandfitstheinsidediameterofthe
pipe.Ittravelsupwhenthewellpressureissufficientenoughtoliftandtravelsbackdown
duetogravitationalforce.Theplungerliftinstallationoperatesasacyclicprocesswhenthe
wellpressureisbuiltupduringshutinandisflowingwhenthepressureissufficienttolift
theplungerandtheliquidcolumncollectedabovetheplunger.Duringshutinperiod,the
plungerisatthebottomonaspringassembly,thegaspressureaccumulatesintheannulus
and liquids accumulate at the bottom of the tubing. The pressure accumulated in the
annulus depends on different parameter such as shutin period, reservoir pressure and
reservoirrockpermeability.Afteracertainperiodoftime,whenthepressureisincreased
sufficiently, the motorized surface valve (motor valve) is opened to allow flow of gas
throughthetubingliftingtheplungertothesurface,unloadingtheliquidsaccumulatedin
thetubingstringandproducingthegasaccumulatedintheannulus.Allthiscyclicprocess
requires an installation of surface equipment that consists of valves and downhole
equipmentthatconsistsofaplungerandaspringmechanism.
Atypicalconventionalplungerinstallation(Figure4.15)includescomponentswhichare:
A spring assembly called the bumper spring that can be installed via wireline to
catchthefallingplungerandhelpitlandatthebottomwithoutdamagingitself
Asurfacecatcher/lubricatorsystemdesignedtocatchtheplungerwhenitrisesup
tosurfaceandallowflowtocontinuebyholdingitaslongasthewellisflowing.
Amotorizedvalveatthesurfacethatiscontrolledelectronicallytoopenandclose
thewellwhenneeded.
Anelectronicsensoratthesurfacetomonitorplungerarrival.
An electronic controller with logic that will set cycles that consists of production
and shutin periods for best operation by opening and closing the motor valve. It
32
willalsorecordthedatafromthesensortohelpdeterminingtheconditionofthe
plunger.
Figure4.15ATypicalPlungerLiftInstallation10
As mentioned, a typical conventional plunger lift application consists of cycles with

productionandshutinperiodswhichareneededforbuildinggaspressureinthecasingand
lifting the liquids accumulated in the tubing efficiently. Although these periods may have
smalldifferencesduetodifferentpropertiesofdifferentgaswells,thesteps(alsoshownin
Figure4.16)aregenerallyasfollows:
33
Figure4.16ASimpleIllustrationofPlungerLiftCycles10
(1) The well is shutin and the pressure inside the casing is building. The motorized
valveatthesurfacewillopenwhenthepressureinsidethecasingissufficienttolift
theplungerandtheaccumulatedliquidcolumnatavelocityexceedingcritical.
(2) The valve is open and the plunger begins to rise with the liquid column. The gas
builtintheannulusexpandsintothetubingstringprovidingtherequiredenergyfor
lifting.
(3) Alloftheliquidcollectedabovetheplungerreachessurfaceandflowthroughthe
surface line. The plunger is held at the surface due to pressure and flow rate
underneath.Thewellcontinuesproducinggasduringthisperioduntilthepressure
decreasesandthevalveisclosed.
(4) Theflowvelocitybeginstodecreaseasliquidsenterthewellfromperforationsand
startaccumulationatthebottom.Atthisdeclineperiod,alargeamountofliquid
willbeaccumulatedatthebottomoftheholeandinthetubingstringifthewellis
opentoflowtoolongandwillrequirealargerbuilduppressure.
(5) Themotorvalveatthesurfaceclosesandthewellisshut.Theplungerfallstothe
bottom of the well onto the spring. The pressure starts increasing in the annulus
onceagainuntilthenextcycle.
34
Usingplungerliftasanartificialliftmethodtoovercomeliquid loading solutionsrequires

initialcapitalcostswhicharerelativelyinexpensive.However,operatingcostswilladdupto
theinitialcapitalcostsandfieldtestingthismethodtoseeifitissuitableforthewellwould
becostly.Inordertodeterminethefeasibilityofplungerliftinstallation,therearecertain
methods that proven useful. Lea et al.8, 10 developed a rule of thumb regarding the
gas/liquidratio(GLR)ofthewelltodetermine ifthecollectedenergyofthegaspressure
would be sufficient to lift the accumulated liquid effectively. This simple GLR rule states
thatthewellmusthaveagas/liquidratioof400scfperbblforeach1000ftofdepththat
liquids have to be lifted. As an example, a 7000 ft deep gas well would require a GLR of
2800scf/bblfortheplungerliftinstallationbefeasible.Thissimpleruleofthumbmaybe
useful,however;itcangivefalseindicationswhenconditionsareclosetopredictedvalues.
ToovercometheshortcomingsoftheGLRruleofthumb,chartsoffeasibilityofplungerlift
(Figure4.17)canbeusedthataredevelopedbyBeesonetal.15(1957)fordifferenttubing
sizes.Anotherchartthatcanbeusedtodeterminethemaximumpossibleproductionrate
thatcanbeachievedwithaparticulartubingsizeatacertaindepthisthedepthvs.fluid
production charts16 that can be found in product manuals. These charts can be found in
AppendixB.
Thereisarathernewtypeofplungerconsistsoftwopiecesthatisdesignedtofalltothe
bottomofthewellwhilethewellisstillproducinggas.Inconventionalplungerapplications
assoonastheplungerbeginsfallingbackdownitconstrictsflowofgas.Inthistwopiece
plunger that consists of a ball at the bottom and a piston at the top the flow continues
aroundtheballandthroughthepistonasthesetwopiecesfalldownfreefromeachother.
When travelling up, however; the ball is pushed upwards not allowing the liquid to go
throughthepistonandthepiecesactasoneunit.Theplungerpiecescanfalldownatthe
bottom at a velocity of 1000 ft/min or more, while the conventional type plungers are
advisednottoexceed750ft/mintoavoiddamagestothestringandtheequipmentitself.
This will allow the plunger system to move faster and make more trips to the surface
allowing it to lift more liquid from the bottom of the well than conventional plunger
applications.
35
Figure4.17GasRequirementforPlungerLiftwithorwithoutPacker16
In plunger lift applications there is also one more issue to consider, which is the use of
plungerliftincompletionswithpackerinstalled.Whenproposingplungerliftinstallationto
aparticularwell,packerlesscompletionsarehighlypreferredovercompletionwithpacker
becausethelargevolumeoftubing/casingannulusallowsmuchmoregastobestoredin
thewell.However,perforatingthetubingstringabovethepackeranddrainingtheannulus
fluidmayimprovetheefficiencysystembyallowingannulustobeusedforstorage.Ifthe
reservoirenergyissufficienttoproduceenoughgastolifttheaccumulatedliquidswiththe
help of a plunger, plunger lift installation can still be used with completion with packer
installed.However,thegasrequirement(gastoliquidratio)ofthewellwillbesignificantly
higherthanpackerlesscompletionsandthisisanimportantfactortoconsider.
4.5.4GasLift
Gasliftisanotherartificialmethodusedtotreatwellshavingliquidloadingproblems.Gas
fromanothersourceisinjectedtothewellatsomedepthandadditionalgasincreasesgas
productionrateofthewellallowingthewelltoremoveliquidsmoreeasily.Theincreased
36
gasvelocitywill be above critical velocitytherefore theliquidswillnot accumulateat the

bottomofthewellbore.Animportantissueaboutthisapplicationisthatgasliftwilllower
the density of the fluids above the injection point. Therefore, the end of injection string
shouldbedeterminedcarefully.
Duetothenatureofthismethod,gasliftmaybeunabletoreducethebottomholeflowing
pressure to lower values than most pumps do. However, there are certain elements that
makegasliftafavoredartificialmethod.Firstofall,amongallartificialliftmethods,gaslift
istheclosestonetothenaturalflowofthewellsincethewellcouldkeepflowingbyitself
withalittleboostfromanoutergassource.Moreover,incaseswheretheremaininggasin
thereservoirisstillhighenoughwithrespecttoliquidvolume,theGLRratioofthewellwill
betoohighforconventionalpumpsystemstoworkeffectivelybecauseofgasinterference
problems,commonlyknownasgaslock.Forgasliftinstallation,ontheotherhand,ahigh
GLR reduces the volume of gas injection needed to lift the accumulated liquids. In
horizontal or deviated wells where pumps cannot work efficiently due to increased
frictional pressure, and wells with solids such as sand where pumps will be clogged and
damaged are conditions which will be more suitable for a gas lift installation for lifting
liquids.
Fundamentally,forliftingaccumulatedliquidsfromgaswells,therearetwotypesofgaslift
techniquesusedexcessivelyinthe industrywhicharecontinuousgas liftandintermittent
gaslift.Incontinuousgaslift,theflowfromsurfacetothepointwheregasisinjectedinto
thewellbore iscontinuousandthehigherpressuregasfromtheoutersourcemixeswith
the gas inside the well, making it easier to lift the liquid column in the production string
(Figure 4.18). This application can be utilized with either a conventional smaller tubing
stringandasimplevalvemechanismoracoiledtubingapplication.
37
Figure4.18ContinuousGasLiftSchematic14
In intermittent gas lift, on the other hand, is used with an automated logic system and
multiplecheckvalves(Figure4.19).Thesysteminjectgasfromanothersourceintothewell
fromacertaindepthuntilthepressureatthebottomissufficienttolifttheaccumulated
liquidcolumntothesurface,andthenflowsthewell,producingtheinjectedgas andthe
gasfromthereservoir,liftingtheliquidstothesurfacealongtheway.Whenthepressure
drops to a certain value, the system closes the well once again to pressurize it with the
outer gas source and this cycle repeats. Generally, continuous gas lift applications are
convertedtointermittentgasliftsometimealongthelifeofawellwhenthebottomhole
pressureofthewelldeclinestoapointwhereitcannolongerlifttheliquidscontinuously
evenwiththeaidofanoutergassourceandthepressureneedstobebuiltupbeforethe
wellcanbeflown.Thepointwhenitistimetoconvertcontinuousflowgasliftsystemto
intermittent flow differs as it is a decision based on the remaining reservoir energy, GLR,
production flow rate and the production string of the well. Using nodal analysis, which
mentioned before, can be beneficial for determining the optimum point to make the
conversion,andalsotheoptimumtubingsizetobeused.
38
Figure4.19IntermittentFlowGasLiftSchematic14
Atypicalgasliftinstallationrequirescertaincomponents12:
Anoutergassourcewithhigherpressure.
Asurfaceinjectionsystemwithappropriatevalvesandtubular.
Asurfaceproductionsystem.
Agasproductionwellwithaninnerstringandgasliftcomponents.
Despiteitscertainadvantages,gasliftinstallationmaynotbeapplicableorfeasibleinmany
liquidproducinggaswellsduetoitsbasicrequirement:anouterhighpressureflowinggas
source.Unlessthewellisclosetoanothergaswellproducingdrygasathighrates,orthe
well itself has another higherpressure gas pay zone; only option left for gas lift to be
utilizediswithinstallationofcompressorswhichhasratherhighcapitalcostsandhaveto
be monitored regularly. If the conditions comply, gas lift is a very useful method for gas
wells with high GLR and production rates just under the critical values. Otherwise, field
applicationofgasliftwouldunlikelyproveuseful.
39
4.5.5Foaming
Foams are used in a wide variety of useful applications in oil business. They are used in
drilling and well completion operations as circulation fluids, fracturing fluids and more.
Foams are also used in producing gas wells as a medium for removing liquids. The main
differenceofusingfoamasaliquidloadingtreatmentfromotherapplicationsistheneed
to generate the foam the bottom of the hole by injecting surfactants and mixing with
liquidsdownhole.
Foamisbasicallyanemulsionofliquidandgas.Surfaceactiveagents,commonlyknownas
surfactantsareusedinwatertoenablemoregastobedispersed.Theexcessamountofgas
dispersed in liquid results in a drastic decrease in the density of the liquid, making the
reservoirpressuretobeabletoliftthefoam allthewaytothesurface.Campbelletal.17
(2004)describestheeffectoffoamingusingthecriticalvelocityequationofTurneretal.1
(Equation 1). According to Campbell, the surface tension is reduced, reducing critical
velocityrequiredtoremoveliquidsaccumulatedatthebottom.
Wells having loading problems with water reacts better to foaming than hydrocarbons
since water foams better and more easily than liquid hydrocarbons. That is due to the
polarity of the water molecules and the attraction in between. Also, according to Lea10,
wellswithGLRbetween1000and8000scf/bblarebettercandidatesforfoaming;though
thereisnoupperlimit,inhighergas/liquidratioswellsmaygivebetterperformancewith
other methods as bottomhole flowing pressure would be lower. Although generally a
simple and inexpensive method, conditions such as increased complex chemical costs for
foamingofliquidhydrocarbons,possibilityofemulsionatthebottomhole,andthepossible
needforaninjectionsystemtoincreaseefficiencymaketheoptimizationoffoamingagents
a challenge. Solesa and Sevic18 showed that for proper optimization of foaming agents in
fieldapplicationsofgaswellswithliquidloadingproblemsextensivelaboratorytests,field
trialsandnodalsystemanalysismayberequired(Figure4.20).
40
Figure4.20FlowingPressureGradientofWaterandFoam18
4.5.6BeamPumping
Beampumpingismaybethemostcommonmethodusedtoliftoilfromwellsworldwide,
and the conventional surface equipment of beam pumping is possibly the best known
imageforoilfieldoperations.Beampumpingisalsoausefulmethodcommonlyusedingas
wells having liquid loading, especially for the cases where the well is loading with liquid
hydrocarbonswhichareasvaluableastheproducedgas.Inwaterloadinggaswellsinareas
wherewaterdisposalcostsarehigh,however;beampumpingmaynotbebeneficial.
Themainprincipleofusingbeampumpingasaliquidloadingsolutionisinstallingthebeam
pumpbelowtheproductionzone,makingitpossibletoproduceliquidsfromtubingstring
and gas from casing (Figure 4.21). Since the gas in the well flows up the casing to the
surface, the well cannot have a packer that would seal the casing/tubing annulus. A gas
anchormaybeusedbelowthebeampumptohelpseparatingthegasfromtheliquidand
makingitdifficulttoenterthetubingstring.Thiswouldpreventpossiblegaslockproblems
with the beam pump. A gas anchor is a simple tool with perforations that is used to
separategasandmakeiteasiertodraintheliquid.
41
Figure4.21ASimpleBeamPumpingSystem12
Thebeampumpingunitisdesignedtochangerotarymotionintoreciprocatingmotionto
give the sucker rods their movement up and down the hole. Beam pumping units are
generallyenergizedwithmoversusingelectricalenergy.Electricityispreferredduetothe
abilityofbeampumptoputelectricalenergytogoodusebyhighefficiency.However,in
certain remote areas where electricity can only be provided with the presence of a
generator, a gas driven engine that uses a portion of natural gas produced from the well
canbeutilizedtopowerthebeampumpingunit.
Animportantconsiderationaboutbeampumpingis,asmentioned,keepingtheproduced
gasfromenteringthetubingstringwhereliquidsareproduced.Entranceofexcessamounts
ofgas intothetubing stringmayresultingaslocking,reducedefficiency andproduction.
42
Thesimplestyetmaybethebestsolutionwouldbesettingtheendoftubingandtherefore
thepumpbelowtheperforations.Duetonaturalmovement,gaswillquicklystarttravelling
up the casing while liquids migrate slowly down the hole. However, there may be cases
wheresettingthepumpbelowtheperforationsisnotpossible.Inthesecases,usingagas
anchororanothertypeofgasseparatorbelowthepumpmaykeeptheproducedgasfrom
enteringintothetubingstring.
Beam pumps are used worldwide in different artificial lift applications, and using beam
pumps can be a useful method for treating gas wells with liquid loading problems. If gas
separationissuesaresolvedproperlywiththeuseofdownholegasseparatorequipment,
beam pumping may lift the accumulated liquids from the bottom of the well efficiently.
However,iftheliftedliquidsarewateroranotherliquidthatcannotbereusedandneedto
be disposed; beam pumping may be impractical and expensive especially in high water
disposalcostsituations.
There are other methods or treatment techniques that are used in gas well to solve
problems due to liquid loading. These methods include blowing the well down, which is
flowingthewelltoatmosphere,usingelectricalsubmersiblepumpsforartificialliftingand
shuttingthewellforusingthebuiltupgaspressuretoliftliquidasslugs.However,these
methods are temporarymeasuresratherthan solutionstoliquid loadingproblems.As an
example;blowingthewelldownonaregularbasisisinefficientduetotheneedofconstant
monitoring and personnel. Also, blowing the well down requires the well to be flown to
atmosphere which would result in gas and liquid to pollute the environment. Electrical
submersible pumps are large applications with high costs and need low gas liquid ratios
that the income from the produced gas cannot cover the operation costs, let alone the
initialcapitalcost.Therefore,thesemethodsarenotmentionedindetail.
43
CHAPTER5
DECISIONMAKING:EXPERTSYSTEM
5.1GeneralInformation
In the light of the theoretical information and studies conducted by several authors, the
aimofthisstudyistohighlightthenecessityofthedecisionmakingprocessandproposea
systematicapproachforselectingpropersolutionsforproblemscommonlyrelatedtoliquid
loadingconcept.
Inordertodesignaproperexpertsystem,adecisiontreeshouldbeconstructedasthe
algorithmoftheexpertsystem.Theaimofthisprocessistodevelopthedecisiontreethat
showsthebasisoftheexpertsystem.
5.2SummaryofTheory:DesignProcess
Overtheyears,severalauthorshaveobservedthebehaviorofgaswellsshowingsymptoms
ofliquidloading.Inchapter4,itisexplainedthatafterseeingtheinitialindicationsofliquid
accumulation in a gas well initial response should be carrying out a downhole pressure
survey without any delay. After confirming the gas well at hand is having problems
associated with liquid loading, the logical next step would be trying to find the optimum
solutiontotheproblem.However,inordertodoassuch;itshouldbedefinedwhetherthe
remainingreservoirenergywouldbesufficientforthewelltoproducegasonitsowneven
with the presence of liquid production on the side. This should be defined using inflow
performance and tubing performance curves with logistically available tubing sizes. If the
remaining energy allows the well to flow on its own with smaller tubing size, this option
shouldbeconsideredpriortoanyartificialliftmethodsinceitwouldprobablybethelow
costoption.Changingtubingsizewithworkoverorvelocitystringisaquestionoflogistics
andeconomicsdependingontheavailabilityofcoiledtubingapplication.
44
Unlesstheremainingreservoirenergymakesitunfeasibleforthegaswelltoproducewith
itsownpower;thenextstepwouldbetodeterminetheartificialliftmethodtobeusedto
getridoftheaccumulatedliquidsonthebottomofthegaswellcausingtheproblem.The
possiblepresenceofanouterhighpressuregassourcenearthetargetgaswellmakesgas
liftapplicationsthefavoredsolutionsincedependingonthedistanceofthegassourceit
would be an inexpensive solution especially in cases with two different gas producing
intervalsinthesamegaswell.Intheabsenceofanouterhighpressuregassourcehowever,
one of the remaining options should be chosen: plunger lift, compression, foaming or
pumping.
The selection criterion for the next step mainly depends on gasliquid ratio of the
production well. As it is stated in Chapter 4, for gas wells with high gasliquid ratio, the
favorable solution would be either plunger lift or compression. That is mainly due to the
impracticalityofothersolutionsinhighgasenvironments.Lea10statesthatwellswithGLR
between1000and8000scf/bblarebettercandidatesforfoamingandaddsinhigherGLR
situations other methods such as plunger lift may give a better performance. The other
question for gas wells with high GLR ratio is determining the better solution between
plunger lift and compression. Although it is stated that in high permeability systems the
effects of compression would be limited; the question at hand is more of an economics
issue rather than technicality. Plunger lift is a low capital cost solution that has to be
monitoredconstantlyforpossiblestuckproblems,whereastheinitialinvestmentcostsfor
compression could be relatively high. When selecting a proper method, the intent of the
clientwouldmakethedifference.
As stated above, the moderate gasliquid ratio solution is generally foaming. Although
there are possible options regarding the use of foaming in gas wells as a liquid loading
solutionmethod,thepracticediffersmainlybecauseofthetypeoftheaccumulatedliquid
in the wellbore. For water loaded gas wells, a solution as simple as dropping soap sticks
from the surface would prove to be useful; if the loaded liquid consists of hydrocarbons,
theremaybeaneedtoinjectcomplexchemicalstothebottomtoreducethedensityofthe
liquid,thusreducingthepressuregradientoftheliquid.Thatmeans,forfoaming,themost
importantissueaftergasliquidratioistheliquidtype.
45
Forgasproductionwellsattheendoftheirproductivelives,gasliquidratiowouldbealot
lower.Thatmakespumpingtheliquidsoutofthegaswellpossibleandalsofeasiblesince
there would benomoregaslocking problemsthat are related to highgascontentin the
productionstring.Obviously,itshouldbedecidedwhetherthecosttoremovetheliquids
from the well by pumping is economically feasible considering the decreased amount of
producedgasfromthewell.However,itiswidelyknownthatespeciallybeampumpingisa
relativelylowcostsolution;thereforeitcouldkeepthewellfreeofliquidsforalongtime
withouttheneedofahighcapitalinvestment.
All the information gathered from theoretical findings and field trials related to solving
production problems associated with liquid loading can be summarized to systematically
approachagaswelltofindthepropersolution.Thisattemptleadstothedevelopmentof
thefollowingdecisiontree,inFigure5.1,foraprospectiveexpertsystem.
46
Figure5.1DecisionTreefortheExpertSystem
47
CHAPTER6
CASESTUDY
6.1BackgroundInformation
Wells#10and#28aretworelativelydeepverticalgaswellswithmoderatetoseriousliquid
loading problems in the same block, producing from the same reservoir. The wells have
depthsof10500and10300ft,respectivelyandbothwellshavetubingstringswithanouter
diameter of 2 7/8 inches as production string in 6 5/8 inch casing. The casing tubing
annulusissealedwithapackerattheendofcompletion.Thegeneralinformationonthe
wells#10&#28areasfollows:
Table6.1GeneralWellInformationof#10and#28
#10 #28
WellDepth ,ft 10500 10300
TubingStringDepth ,ft 9550 9650
PerforationInterval ,ft 9720 9760 980010010
CasingSize ,in 65/8 65/8
TubingSize ,in 27/8 27/8
Packer/Completion PermanentProductionP. PermanentProductionP.
Thereservoirfromwhere bothwells are producing gas along with waterandcondensate

hasanaveragepermeabilityof1millidarcy.Bothwellsareconnectedtoasurfacepipeline
system with a line pressure of 300 psia. Gas samples collected from the wells show a
specificgravityof0.65forwell#10and0.66forwell#28.Also,thewatersamplesthatare
gathered from the wells have densities of 8.52 ppg, 8.48 ppg for well #10 and 8.54 ppg,
48
8.51 ppg for well #28; so an average density value of 8.5 ppg for water is used in
calculationstosimplifythemodel.
6.2Well#10
Afterproducinggasforquiteatime,dailygasproductionrateof#10hasstartedtodecline
as reservoir pressure depletes. The well has begun showing erratic flow behavior and
eventually,duetothehighpressureofthesurfacelinesthat#10hasbeenflowingin;the
well ceased production. The erratic behavior the well has been showing is, as mentioned
earlier,asymptomofpossibleaccumulationofliquidsatthebottomofthewell.Theever
decliningtubingheadpressurewasanothersignpromotingliquidaccumulation.Sincethe
well was completed with a packer installed at the end of the tubing string, the casing
pressurecouldgivenoindication. InFigure 6.1,tubingheadpressure againsttimeclearly
showstheerraticflowbehaviorofthewell.
Figure6.1TubingFlowingPressureShowingErraticFlowin#10
Inthelightoftheseclues,ourdecisiontreesuggeststakingapressuresurveyinorderto
confirm the presence of liquids in the well. Several pressure surveys with both static and
49
f
flowingwell conditionshaavebeenperrformed inorder to deterrmine ifthe well is actually
s
suffering from
m liquid load
ding. In presssure survey #1
# (Figure 6.2
2), performed
d in static weell
c
conditions att a depth off 9500 feet, it is clear th
hat there is liquid accum
mulation in th
he
p tring.Accordiingtothesurrvey,thestaticliquidlevel inthewellisat4150feet.
productionst
6.2Well#10PressureSu
Table6 urvey#1SampleData
Dep
pth,ft TTemperature,F Pressu
ure,psia
0 72 1067
1640 82 1109
3280 118 1149
4920 161 1499
6560 199 2200
9190 257 3348
9500 266 3491
A
Aftergatheringdataofth
hepressuresu
urvey,thepressurevs.deepthchartcan
nbeplotted to
s
seeclearlythepressuregrradientoftheewellboreandtheliquidleevel.
Figure6.2PressureSurveyfor#1
10ShowingLiiquidLevel
50
Afterdiscoveringthattheproblemsassociatedwiththewellisduetoliquidsaccumulated
atthebottomholeandtheinabilityofthewelltoliftthoseaccumulatedliquidstosurface;
Turners1criticalrateequation(1)hasbeenapplied:

1.593

Consideringthewellproducesanaverageof550bblliquidina30dayperiodand96%of
thisproducedliquidiswater;andtakingintoaccountasmentionedearlierwhenwateris
present at the wellbore as accumulated liquid, even if there are also condensed
hydrocarbons present, critical velocity should be calculated according to water since the
waterdensityishigherthancondensate,thusmakingitsurethattheresultoftheequation
will be adequate for condensate, also. Using bottomhole conditions as proposed in the
theory,gascompressibilityziscalculatedas0.74.Surfacetensionofwaterunderbottom
holeconditionsistakenas60dynes/cm.Thewelldatagatheredisappliedtotheequation
withoutthe20%adjustmentwhichwouldactuallybeColemans2equation:

1.593 60 63.5 0.00297 /

0.00297 /

4.43 63.5 0.0039

0.0039
Addingthe20%adjustmentasTurnerstated,theequationbecomes:

5.32 63.5 0.0039

0.0039
Thewellstartssluggingandloadingupwhenflowingtothesurfacelinewithapressureof
300psiaandalthoughthewelliskeptunloadedwithacombinationmethodofblowingthe
welldownandflowingthewellintermittentlyintothesurfacepipeline,itstartsloadingup
afterashortwhile.Theproductiondatagatheredasthewellisblowndowntoatmosphere
showsthatthewellhasapotentialdailyproductionrateof0.6MMscf/dwhenflowingon
24/64 choke with a pressure of 210 psia and 1.05 MMscf/d when the well is flowing full
openwithaflowingpressureof80psiaatthewellhead.
51
U
Using the critical rate eq
quations of both
b Turner et al. and C
Coleman et al.,
a critical rate
r
requiredford
differenttubingsizesarep
plottedinFigure6.3andFFigure6.4.
Fiigure6.3Tu
urneretal.sC
CriticalFlowrateforDifferrentTubingSiizes
Figgure6.4Colemanetal.ssCriticalFlow
wrateforDiffeerentTubingSSizes
52
Colemanetal.suggestedintheirstudiesthatforhighflowingwellheadpressuresTurneret
al.s 20% upward adjustment may be required but for wells with low flowing wellhead
pressure the adjustment is unnecessary. After examining the critical rate plots and
comparingwithactualwelldata,itisseenthatalthoughtheactualgasflowratevaluesare
closetoColemanetal.scriticalrateestimations,thewellisunabletolifttheliquidsinthe
flow path entirely. The more recent study of Sutton et al. shows that the 20% upward
adjustment should be used to ensure the entire flow path (production string) is free of
liquids.
Theflowdatagatheredduringablowdownoperationshowedthatthewellisflowingwith
a rate of 1.05 MMscf/d and a flowing tubing head pressure (FTHP) of 80 psia. In these
conditions, the well is able to lift the liquids accumulated at the bottom, but when it is
flowingintothesurfacepipelinewhichhasapressureof300psia,thewellstartsloading
due to backpressure caused by the high pressure of the surface line. The well may need
certain artificial lift methods in the future, but it is clear from the data and the plots
generatedusingcriticalrateequationsthattheflowrateofthewellisattheborderlineand
with a relatively inexpensive tubing sizing operation or velocity string installation it can
match the required critical rate. Inflow performance curve is calculated from single point
testsandcanbefoundindetailinAppendixC.Asfortubingperformancecurves,a3rdparty
programPipephaseisusedtoconstructthecurvestopredicttherequiredtubingsize.
Assuming the pressure of the surface lines will remain unchanged at 300 psia; the nodal
analysiswithinflowperformanceandoutflowtubingcurvesareconstructed.IPRcurvesare
plottedwithandwithoutturbulenceeffects,andTPRcurvesareplottedforvarioustubing
sizes ranging from 1.66 to 3 including the current tubing size. The generated Nodal
AnalysiscurvesareshowninFigure6.3.
53
Figure6.5NodalAnalysisfor#10withDifferentTubingSizes
As can be seen in the nodal analysis and the critical flow chart shown in Figure 6.3; the
cross points of IPR & TPR curves are under critical conditions according to Turner et al.s
model.Theplotshowsthatfortubingstringwithdiameters largerthan 2 3/8inches,the
flowing bottomhole pressure values give indications of loading. Examining the data
gatheredfromnodalanalysis,selectingasmallersizetubingstringwillhelptheunloading
process.
6.3Well#28
Similartowell#10,well#28isagasproductionwellthatstartedhavingproblemsdueto
liquidaccumulation.Thewellhasbeenshowingerraticflowbehavioralso(Figure6.6),and
asreservoirpressurecontinuestodecline,#28isstrugglingwithliquidloadingproblemsas
the liquids accumulated at the bottom cannot be lifted properly. The case in well #10
showed that even with proper tubing sizing and reducing the wellhead flowing pressure,
the solution would be relatively temporary depending on the remaining energy of the
formation.Thatwouldleadtotheconclusionthattheremainingenergyoftheformationis
onthevergeofinabilitytolifttheliquidaccumulatedatthebottomofthewellbore.Unable
54
toproduceeffectivelyonitsownduetosevereliquidloadingproblems,thenextstepinthe
decisiontreeisfollowedtocomparevariousmethodsforlookingforapermanentsolution
onliquidloading.
Figure6.6ErraticFlowBehaviorshowninPressurevs.TimeGraphof#28
Liketheformerwell,apressuresurveyinstaticconditionshasbeencarriedouttoconfirm
theindicationsofliquidloadinginthewell,andalsodeterminethepressureatthebottom,
as shown in Figure 6.7. The pressure survey done in #28 showed that the liquid level is
deeperthanthatof#10.Theliquidlevelisat7200ftwhichindicatedthereservoirpressure
islower,pointingouttheneedtofindadifferentsolutiontothematter.
55
Figure6.7PressureSurveyfor#2
28ShowingLiiquidLevel
U
Using the data at hand, inflow and ou
utflow curvess that belongg to #28 have
e been plotteed
u
usingthesam
meprocedureeasinwell#10.InFigure 6.8,nodalan
nalysiswithd
differenttubin
ng
s
sizesisshown
ntoobservettheflowcond
ditionswithvvarioustubinggdiameters.
Figure6.8NodalAnalyysisfor#28w
withDifferentTubingSizes
56
A
Asitcanbec
clearlyseenfromnodalan
nalysis,thein
nflowperform
manceofthe well#28islo
ow
d
duetolower reservoirpre utflowcurvesthatbelong to
essure.Theanalysisalsoshowsthatou
d
different tubing sizes are close to eacch other with
h the exceptio
on of 1.66 outerdiamet
o er
t
tubing. As mentioned,
m s
since he reservoir is
the resservoir presssure is quite low and th
d
depletingand
dkeepingin mindthatthewellisflow
wingwithapressureof20
025psiawheen
o
open mosphere; a lift method should be selected
to atm s insteead of installing a smaller
d
diametertub
bingstring,sin
nceasmallerrdiametertubingstringw
willbecomein
nsufficientaftter
a short time.. Figure 6.9 shows
a s the co
omparison off actual flow rates with Turners
T criticcal
r
rates at the junction poin PR curves for various tubiing sizes. Thee graph simp
nts of IPRTP ply
s
showsthatth
heidealcon
nditionswilln
notmeetcriticalconditionsduetolowdeliverabilityy:
Figurre6.9ActuaalRatesvs.CrriticalRatesfor#28
A
Accordingto Leaetal.,asmentionedeearlier,selectiingthepropeerartificialliftttosolveliqu
uid
l
loading prob
blems is mosstly a matter of the gass/liquid ratio
o of the well. In order to
d heGLRofthe well;gasand
determineth dliquidprodu
uctionratehaasbeenrecorrded.Table6
6.3
s
showsaport ionofthepro
oductiondataawhichgivesdifferentGLR
Rvaluesatdifferenttimess:
57
Table6.3ProductionDataofWell#28
Days GasRate,MMscf/d LiquidProduction,bbl GLR,scf/bbl

1 0.571 11 51915
2 0.592 12 49296
3 0.585 10 58477
4 0.544 9 60459
5 0.544 10 54441
10 0.570 9 63315
15 0.589 4 147269
20 0.386 0
The erratic flow behavior can be clearly seen from the production data also. It is seen in
Figure6.10thatthewellisliftingsomeoftheliquidsinitiallysincethegasrateisrelatively
high and as ratedecline the liquid lifted to surface is decreasing andrate starts declining
even more shortly after that. Also, from Figure 6.10 it can be seen that daily condensate
productionisverylowcomparedtowater,soitsfairtotakealltheliquidaccumulationas
watersince;asstatedbefore,duetohigherdensityofwater.Plottinggasproductionrate
along with liquid production would help to see the effects of loading on the rate more
clearly:
Figure6.10GasandLiquidProductionChartof#28
58
It is clear in the plot that the liquid is accumulated in the production string as the well
continuestoflowandafterawhilethebackpressurecausedbythehydrostaticpressureof
theliquidcolumncausesbubbleflowinthewellsothatnoliquidsareliftedasthegasrate
declines.Aroundthe33rdday,thewellisblowndowntoflareasaremedialmeasuretoget
rid of the liquids accumulated in the production string; and as a result gas rate increases
sharply and liquids are lifted once again. It can be said that blowing the well down is
successfulasthegasrateisincreasedbutsinceblowdownisnotalongtermsolutionbut
onlyaremedialtreatmentthatrequiresconstantmonitoringandmanpower.Also,blowing
downthewellwouldbetemporarysinceasblowdown,liquidscontinuetoenterthewell
assomeofthemareliftedtosurface,themoststillaccumulatedownatthebottomofthe
well as the production data and the continued erratic flow behavior is evidence to that
condition.
Following the decision tree in order to find a more permanent solution to the issue at
hand,thenextstepistodeterminepossibilityofagasliftapplication.Knowingthatgaslift
requiresanouterhighpressuregassourcetoinjecttothewellhavingliquidproblems;the
lack of an outer high pressure source near #28 makes this installation costly and
inapplicable. Since gas lift application is not the logical choice, gas/liquid ratio of the gas
well should be determined as the decision tree suggests. According to gas and liquid
productiondata,thewellhasanaverageGLRvalueof50000scf/bbl.ThehighGLRmakesit
nearlyimpossibletousepumpsbecauseofthehighprobabilityofgaslocking.TheGLRis
toohigh,meaningthereistoomuchgasinthewellasopposedtoliquidsthatbeampumps,
ESPs, and such will have severe gas interference problems like gas locking. Another
possibility is to inject surfactants down the string which is also known as foaming.
According to Lea et al., the wells with GLR between 1000 and 8000 scf/bbl are better
candidatesforfoaming,andalthoughthereisnoupperlimit;thewellswithhigherGLRwill
work best with other methods such as plunger lift or compression. Deciding between
plunger lift and compression is more of a question of capital and operational costs and
logisticavailabilitythantechnicality.Plungerliftapplication,thelowcapitalcostsolution,is
seentobethebestoptiontofollowaccordingtothewishesoftheclient.
59
As,thehighGLRofthewell#28makesitabettercandidateforplungerlift,asmentioned
above,theonlyproblemisthatforplungerlift,packerlesscompletionsarefavoreddueto
increasedcapacityofwellborestorage.Theentirecasingvolumecouldbeusedtostoregas
duringpressurebuildupperiodandduringflowgasratewouldbehigher.Thewell#28has
a permanent production packer installed at the end of tubing. Although this restricts the
amountofgasandrequiresahigherGLRthanpackerlesscompletionsdo;theplungerlift
feasibilitycharts(inAppendixB)showthatthehighGLRofthewellisabenefitandplunger
lift would be feasible even with a packer installed. With a 10000 ft production string,
accordingthefeasibilitycharts,itrequiresaminimumgasliquidratioof4000scf/bblfora
packerless completion and 6000 scf/bbl with packer installed. The well is clearly a good
candidateforplungerlift,althoughitiscompletedwithapacker.However,becauseofthe
commonproblemsassociatedwithplungers,theplungerlifthasnotbeeninstalleddueto
thepossibilityofastuckplunger.Theriskofastuckplungerinthetubingstringwouldlead
tohighcostsofpossibleworkoveroperationsinwell#28andthereforeit isthoughtthat
the riskofa stuckplunger that wouldneed fishing operations willendanger thewell and
mayevencausethelossofthewell.Inthelightofthesepossibilities,anewapproachhas
beenproposedwhichisinstallingtheautomatedlogiccontrolledwellheadtothewell#28
without the plunger and its components, creating an automated intermittent flow in the
well.
Theideaisbasedupontheworkingmechanismsofplungerliftandgasliftinstallations.The
need of a motorized valve and the buildup periods that allow the well to flow at higher
ratesarethekeyelementsofthissocallednewmethod.Theideaissimpleyeteffective,
the well is shutin until the pressure builds up and the well is flown to lift the liquids
accumulatedatthebottomtothesurface.
The production data after the installation of the intermittent flow logic control shows
notableincreaseinthedailygasproduction.Inaselected30dayperiod,thecumulativegas
productionaftertheinstallationismeasuredas30.2MMscf/daywherebeforeinstallation
it was measure as 15.4 MMscf/day. It is noticed that liquid production is also increased.
Table6.4isadaybydaycomparisonoftheproductiondataof#28.
60
Table6.4DailyProductionRatesof#28PreandPostLogicControl
BeforeInstallation AfterInstallation
Liquid
Days GasRate,MMscf/d GasRate,MMscf/d LiquidProduction,bbl
Production,bbl
1 0.571 11.0 0.946 13.8
2 0.592 12.0 1.069 19.2
3 0.585 10.0 0.723 15.6
4 0.544 9.0 0.709 16.4
5 0.544 10.0 0.809 17.2
6 0.523 10.0 1.055 14.8
7 0.591 10.0 1.227 14.8
8 0.483 12.0 1.254 15.2
9 0.529 10.0 1.068 11.6
10 0.570 9.0 1.056 10.8
15 0.589 4.0 0.887 4.8
20 0.386 0.0 0.944 7.2
25 0.340 0.0 1.186 12.8
30 0.339 0.0 1.154 14.0

Figure6.11DailyGasProductionBeforeandAfterInstallation
61
Itisseenthatbothgasandliquidproductionisincreasedandthisisduetotheabilityofgas
liftingtheaccumulatedliquidmoreeasily.TheaverageGLRofthewellisalsoincreasedto
70000 scf/bbl and this increase shows that as liquids are lifted from the well and the
backpressure exerted by the hydrostatic column, the gas production rate is increasing.
Below;15daysofproductionisshowninthegraphinFigure6.11toshowtheimprovement
inproductionbycomparingproductionratesbeforeandaftertheinstallation.
62
CHAPTER7
DISCUSSIONOFRESULTS
Indevelopingcriticalratesregardingwell#10,equationsderivedbyTurneretal.areused;
adetaileddevelopmentoftheseequationscanbefoundinAppendixA.Intheequations,
data gathered from well tests and pressure surveys are used. Actual production data
containingdailygasproductionrates,liquidproductionandflowingwellheadpressuresare
analyzed and compared with previous correlations of Turner et al. and Coleman et al..
Althoughitisadvisedthatforlowerflowingwellheadpressures(FTHP<500psig)equations
proposedbyColemanetal.areused;productiondatashowedthatTurneretal.sequation
fitsbetter when determining liquid loadingduring production,using averagevalues of 63
lb/ft3forwaterdensity,0.9forzfactorand0.65forgasspecificgravity.
TheselectionofTurneretal.scriticalrateequations(with20%adjustment)isduetothe
flowingconditionsofthewells.Inordertodeterminewhichofthecriticalrateequations
fitsthefielddata,naturalflowconditionsofwell#10isappliedtocriticalrateequationsof
Turner,ColemanandLi.Thenaturalflowconditionsofwell#10accordingtonodalanalysis
curvesforatubingstringwith27/8outerdiameter.Noticethattheactualgasflowrate
of the well with 2 7/8 tubing obtained from nodal analysis is higher than the current
flowrate of the well. This is due to the accumulation of liquids in the wellbore, and the
nodalanalysisshowsthecasewithnoliquidspresent,whichmaybecalledasthepotential
flowratethewellcanhavewithoutliquidaccumulationatthebottomofthewell.Figure7.1
shows that flowrate and the critical flowrates of different models. In these conditions,
assumingthewellisfreeofliquidsandstartsflowing;thewellshouldnotloadupandbe
freeofliquidaccumulationaccordingtothemodelsofColemanetal.,NosseirandLietal.
Turneretal.smodelistheclosesttotheactualdata,andverifiesourinitialconceptionof
thesituation.
63
Figurre7.1Comp
parisonofCritticalRateEqu
uations
C
Changing cro by installing a different tubing stringg with smaller
osssectional flow area b
d
diametertha ntheexistinggstringwould ndercertaincconditionsthat
dprovetobeebeneficialun
s
should be an
nalyzed usingg IPRTPR cu
urves for diffferent tubingg sizes. Comp
paring IPRTP
PR
a
analysis with t keep the well free frrom liquid accumulation is
h critical ratees required to
n
necessary to observe if proposed
p bing string would be succcessful or not. In this casse,
tub
c
changingthe herequiredgas
current27//8tubingstrringto23/8issufficientttoobtainth
v
velocity; the effects will most
m likely be temporary. Keeping in mind that deepletion of th
he
r
reservoiristh quidloading; alongtermssolutioncanb
hemainreasonbehindliq beachievedb
by
i
installing a tu
ubing string with a diameeter of 2 1/1
16 or less. TTo have a bettter visual, th
he
f
flowing cond
ditions for diffferent tubin
ng sizes which are the co
ontact points of inflow an
nd
o esarecompaaredagainstccriticalrates calculatedbyyTurneretal.scriticalrate
outflowcurve
e
equationinF
Figure7.2.
64
1.4
1.2
GasRate,MMscf/d
1
0.8
0.6
0.4
0.2
0
1.66" 21/16" 23/8"" 27/8
8" 31//2"
Q
Qcritical Q
Qactual

Figurre7.2ActuaalRatesvs.CrriticalRatesfor#10
ous tubing sizes shows th

T data gatthered from outflow curvves for vario
The hat decreasin
ng
t
tubing size has little effecct in decreassing the flowrate and doees not necesssarily constraain
p
productionra
ateduetolim
miteddeliveraabilityofthe lowpermeab
bilityreservoir.Therefore,,a
t
tubing string with smaller diameter, 2
2 1/16 or leess, would bee the solution to the liqu
uid
l
loading prob
blems encoun
ntered in weell #10 due to
t low veloccity of gas. Along
A the waay
i
installing a compressor to
o use compreession in com
mbination with smaller diiameter tubin
ng
s
stringwould benecessaryyinorderto decreaseflow
wingtubingw
wellheadpressuremoreffor
t ultimate gas recoverry to be high
the her. All analyysis on inflow
w deliverability and tubin
ng
p
performance suggests that, as the decision tree follows, thee remaining energy
e on th
he
f
formation is sufficient to
o continue production if certain cond
ditions are met
m with smaall
c uchassizingttubingstringforanincreaseingasvelo
correctionssu ocity.
T problem
The ulation in weell #28 howeever, are more
ms encountered due to liquid accumu
s
severe.Thep mholepressureisaround 1500psia,an
pressuresurvveyshowsthaatthebottom nd
t
that low presssure hinderss the possibility of a tubing sizing solu
ution for thee well #28. Th
he
n
nodal analyssis shows that inflow peerformance is low due to
t reservoir depletion an
nd
c
comparing gaas flowrates for various tubing sizes with critical flowrates caalculated usin
ng
T
Turneret al.sequation showsthat on
nly installing atubingstringwitha diaametersmaller
t
than 2 (as an example 1.66) will eensure the continuous reemoval of liq
quids from th
he
65
wellbore. Installing a tubing string with diameter that small would probably constrict gas
flowrateduetosmallcrosssectionalarea,butmaintainingasteadydeclinecurveismore
important than increasing daily gas production for a short period of time. Although the
inflowperformanceanalysisshowsthattheremainingenergyontheformationisenough
tocontinuetoproduceonitsown,sincetubingsizingisjustoneofmanymethodsanalyzed
in this study, another approach is adopted for #28, which is basically proposing to find a
proper artificial lift method instead of enhancing natural flow for removal of liquids. This
means followingthe stepsof thedecisiontree as if theremainingreservoirenergy isnot
sufficient fornatural flowbecause aproperartificial lift methodforliquidremovalwould
helpachievingahighergasflowrateasopposedto1.66outerdiametertubingstring.Also,
thisleadstoasmallcorrectiononthedecisiontree;theneedtocheckifthenaturalflow
bytubingsizingoptionwouldleadtoadecreaseingasratecomparedtopossibleartificial
liftmethodsthatcanbeused.
The selection criterion of a proper artificial lift method is a matter of effectiveness and
power consumption as opposed to recovery. As stated by Lea et al. in his studies and
mentionedearlierinthisstudy;selectingthemostsuitablemethodismostlyreliesongas
liquidratio,sinceitbasicallyshowstheamountofliquidthatshouldbeliftedwithacertain
amountofgas.Thehighgasliquidratioofwell#28,whichmeanslowamountofliquidsin
respect to gas production, makes certain methods favorable and the others unfavorable.
Methodslikefoamingandbeampumpingareadvisedforwellswithlowergas/liquidratios,
sinceitispossibletoreachlowerflowingbottomholepressureswithmethodslikegaslift
andplungerlift.Inordertodeterminebetweenthesetwomethods,however;oneshould
look into the power perspective. According to a study by Dotson and Nuez22 (2007),
selectionofartificialliftmethodsfromthepowerperspectiveinvolveseitherthebetteruse
ofremainingreservoirenergyorapplyinganexternalenergytothewell(Figure7.3).They
concludedthathighestultimaterecoveryisachievedbypumpingsincethelowestbottom
hole pressure is reached. However, it is also feasible to harvest a portion of reservoir
energy to lift liquids from wellbore, which can be approximated by plunger lift. Their
research showed in tight reservoirs gas lift would require too much power to increase
ultimaterecovery.
66
Figure7.3Powervs.RecoveryforTightReservoir22
Consideringpowerrequirementsandwithouttheexistenceofanexternalhighgassource,
plungerlifthasmoreadvantagesovergaslift.However,asmentionedearlier,becauseof
thecommonproblemsassociatedwithplungers,theplungerlifthasnotbeeninstalledto
avoid stuck problems that may occur in the well. Instead another approach has been
adoptedasacombinationofautomatedcontrolmechanismofplungerliftapplicationswith
intermittentflow.Themotorizedvalvesandcyclesthebuildupandflowperiodsallowthe
welltoflowathigherrates.Althoughtheideaissimple;theresultsshowthatitiseffective;
the well is shutin until the pressure builds up and then it is flown to lift the liquids
accumulatedatthebottomtothesurface.Figure7.4showsthecyclesconsistingofseveral
shutinandflowstagesina24hourperiod.
67
Figure7.4IntermittentFlowCyclesfor#28
Theresultissomewhatsuccessful,duetoincreaseddailygasproductionratesinthewell,
as well as increased liquid production, meaning the ability of the well to lift liquids more
effectively.However,althoughthereisasignificantincreaseintheproductionrate,thewell
still shows erratic flow behavior and signs of liquid accumulation even if it is minimal or
lower than the former case. The effective use of a plunger lift application would lift the
accumulatedliquidbettersincetheflowbehaviorismostlikelycausedbyfallback,whichis
the liquids falling back to the bottom in flowing period of the intermittent flow cycle as
flowingpressurestartstodecline.Thepresenceofaplungerbelowtheliquidcolumnwill
keep the liquid from falling back to the bottom. However, the completion type and
presence of a production packer downhole is a known disadvantage for plunger lift
installation. Charts for Gas liquid ratio and plunger lift feasibility shows that current
potentialandGLRofthewellissufficientforplungerliftinstallation.Still,thetubingstring
couldbeperforatedandthecompletionfluidinthecasingtubingannuluscouldbedrained
forincreasingthegasstoragevolumeformoreeffectiveplungerliftapplication.
In brief, well #28 shows significant improvement after automated intermittent flow
applicationisinstalledatthewellbore.Thewellstillshowssignsofliquidaccumulation,and
as the reservoir pressure depletes, the well will probably need the plunger lift to be
68
installed as the decision tree suggests in the first place. However, until the well shows
severeproductionproblemsandceasestoproducesteadily,intermittentflowwillbeused
asaliquidloadingsolution.
69
CHAPTER8
CONCLUSIONS
Thepurposeofthisstudywastodeterminethemethodsforpredictingtheonsetofliquid
loading in gas wells, evaluating completion types for optimization and comparing various
methodsasapossiblesolutionforloading.Thefollowingconclusionsaredrawnfromthe
studybasedupontheanalysisofactualfielddataofgasproductionwellsandcomparison
ofvariousstudiesoncriticalvelocitytheorytodeterminecriticalratesofgaswellshaving
productionproblemsduetoliquidloading:
The first thing to do after observing initial signs of liquid loading is a downhole
pressure survey to confirm liquid accumulation at the bottom of the well. This is
alsoessentialtoseeifaccumulatedliquidsfloodedtheentireperforationinterval.
Analyzingdifferentcriticalratetheoriesisimportanttoseewhichmodelfitswhich
case. In this study, Turner et al.s droplet model for determining critical rate fits
flow behavior of two wells better, however for every individual case all models
shouldbecompared.
Eventhoughselectedartificialliftmethodisunabletoliftallaccumulatedliquids
fromthebottom;relievingthebackpressurecausedbytheliftedportionmaystill
provetobeusefulifitprovidesasteadyproductionincrease.
Thisstudywasunabletoanalyzetheeffectsofaccumulatedliquidsinthereservoirdueto
insufficient reservoir data. It is known that accumulation of liquids in the reservoir
decreasestheeffectivepermeabilityofgasduetoincreasedskinfactor.Furtherresearch
onthisfieldregardingtheroleofskinfactoronliquidloadingwouldofhelpindeveloping
criticalrateequationswiththeeffectsofskinfactor.
70
REFERENCES
1. Turner,R.G.,Hubbard,M.G.andDukler,A.E.,AnalysisandPredictionofMinimum
FlowRateforContinuousRemovalofLiquidsfromGasWells,paperSPE2198,JPT
(November1969)
2. Coleman, S.B., Clay, H.B., McCurdy, D.G. and Norris III, H.L., A New Look at
PredictingGasWellLoadUp,paperSPE20280,JPT(March1991)
3. Nosseir,M.A.,Darwich,T.A.,Sayyouh,M.H.andElSallaly,M.,ANewApproachfor
Accurate Prediction of Loading in Gas Wells Under Different Flowing Conditions,
paperSPE66540,SPEProduction&FacilitiesVol.15Number4(November2000)
4. Li, M., Li, S.L. and Sun, L.T., New View on ContinuousRemoval Liquids From Gas
Wells, paper SPE 75455, presented at the 2001 Permian Basin Oil and Gas
RecoveryConference,Midland,Texas,May1516
5. Veeken, K., Bakker,E., andVerbeek,P.,Evaluating Liquid LoadingField Data and

Remedial Measures, paper presented at the 2003 Gas Well DeWatering Forum,
Denver,CO,March34
6. Belfroid, S.P.C., Schiferli, W., Alberts, G.J.N., Veeken, C.A.M. and Biezen, E.,
Prediction Onset and Dynamic Behavior of Liquid Loading Gas Wells, paper SPE
115567presentedatthe2008SPEAnnualTechnicalConference,September2124
7. Sutton,RobertP.,Cox,StuartA.,Lea,JamesF.andRowlanO.Lynn,Guidelinesfor
the Proper Application of Critical Velocity Calculations, paper SPE 120625,
presentedatthe2009SPEProductions&OperationsSymposium,USA,April48
8. Lea, James F. and Nickens, Henry V., Solving GasWell LiquidLoading Problems,
paperSPE72092,DistinguishedAuthorSeries,JPT(April2004)
71
9. Gunawan, R., and Dyer, G.R., Tubing Size Optimization in Gas Depletion Drive
Reservoirs, paper SPE 37001, presented at the 1996 Asia Pacific Oil & Gas
ConferenceinAustralia,October2831
10. Lea, James F., Nickens, Henry V. and Wells, Mike R., Gas Well Deliquification,
SecondEdition,ElsevierPress,Cambridge,2008
11. Christiansen,R.L.,Girija,E.,RigginsL.andElsener,G.,LiquidLifting fromNatural

Gas Wells: TubingCasing Junction, paper SPE 96938, presented at the 2005 SPE
AnnualTechnicalConferenceinDallas,Texas,USA,October912
12. McMullan, J.H. and Bassiouni, Z., Optimization of GasWell Completion and
Production Practices, paper SPE 58983, presented at the 2000 SPE International
PetroleumConferenceandExhibitoninMexico,February13
13. Craft,B.C.andHawkins,M.F.,AppliedPetroleumReservoirEngineering,Prentice
HallPress,EnglewoodCliffs,NJ,1959
14. Arachman, F., Singh, K., Forrest, J.K. and Purba, M.O., Liquid Unloading in a Big
Bore Completion: A Comparison Among Gas Lift, Intermittent Production, and
InstallationofVelocityString,paperSPE88523,presentedatthe2004AsiaPacific
OilandGasConferenceandExhibitioninAustralia,October1820
15. Beeson,C.M.,Knox,D.G,andStoddard,J.H.Part1:TheplungerLiftMethodofOil
Production,PetroleumEngineer,1957
16. OtisPlungerLiftTechnicalManual,1991
17. Campbell, S., Ramachandran, S., and Bartrip, K., Corrosion Inhibition/Foamer
CombinationTreatmenttoEnhanceGasProduction,paperSPE67325,presented
attheSPEProductionandOperationsSymposium,2427March2004
18. Solesa, M. and Sevic, S., Production Optimization Challenges of Gas Wells with
Liquid Loading Problem Using Foaming Agents, paper SPE 101276, presented at
theAsiaPacificOil&GasConferenceandExhibiton,1820October2006
19. McCain,W.,PropertiesofPetroleumFluids,PennWell,Tulsa,OK,US,1990
20. Hinze, J.O., Fundamentals of the Hydrodynamic Mechanism of Splitting in

DispersionProcesses,AICheJour.Sept.1955,No.3,289
72
21. Hinze, J.O., Critical Speeds and Sizes of Liquid Globules, Applied Scientific
Research,1949,No.4,273
22. Dotson, B. and NunezPaclibon, E., Gas Well Liquid Loading From the Power
Perspective,SPE110357,2007
23. Mishra, S. and Caudle, B. H., A Simplified Procedure for Gas Deliverability
Calculations Using Dimensionless IPR Curves, paper SPE 13231, presented at the
59thAnnualTechnicalConferenceandExhibitionofSPE,Houston,Sept1619,1984
73
APPENDIXA
DEVELOPMENTOFCRITICALVELOCITYEQUATIONS
ThisappendixsummarizesthedevelopmentofcriticalvelocityequationsofTurneretal.1to
calculate the minimum gas velocity to remove liquid droplets from a vertical wellbore.
Turneretal. analyzedtwophysicalmodelstodeterminetheminimumvelocitywhichare
liquidfilmmovementalongthewallsoftheproductionstringandliquiddropletsentrained
inflowinggascore.Itisstatedbeforethatliquiddropletmodelisadoptedsinceitisknown
tofitfielddatabest.Whendevelopingthemodel, Turneretal.usedfluidflowequations
developed by Hinze20, 21 whom stated that liquid drops moving relative to a gas are
subjectedtoforcesthattrytoshatterthedrop,whilethesurfacetensionoftheliquidacts
to hold the drop together. Therefore, a droplet is subjected to two forces which are
gravitationalforce(FG)anddragforce(FD).
(A1)
(A2)
Where;
gravitationalforce
dragforce
dragcoefficient
gravitationalconstant=32.17lbmft/lbfs2
Accordingtothetheory,thedropletisentrainedinthegascore,meaningthesetwoforces
areequaltoeachother.Theequationthenbecomes:
(A3)
74
Thisequationgivesthecriticalvelocityequationas:
(A4)
Hinze20showedthatthedropletdiameterisdependentupongasvelocityandisexpressed
intermsofthedimensionlessWebernumber:
(A5)
Hinze showed that the droplet shatters when NWE is greater than 30; thus solving the
equationforavalueof30forthelargestdropletdiametergives:
30 (A6)
Turner assumed the drag coefficient CD as 0.44 valid for all turbulent conditions.
SubstitutingdropletdiameterfoundformWebernumber,dragcoefficientandconverting
surfacetensionfrom1lbf/ftto0.00006852dyne/cmgivestheequation:

1.593 (A7)
In order to simplify the equation for field application, Turner has taken typical values for
temperature,gasgravityandzfactorthusconsolidatingtheterms.Takingtypicalvaluesof:
Gasgravity=0.6
Temperature=120F
Zfactor=0.9
Waterdensity=67lb/ft3
Watersurfacetension=60dyne/cm
Condensatedensity=45lb/ft3
Condensatesurfacetension=20dyne/cm
75
Introducingthesevaluestotheequation,itbecomes:

. .
(A8)
.

. .
(A9)
.
Turnerfoundthata20%adjustmentshouldbemadeintheseequationsforthefielddata
tobematched.Withthe20%adjustmenttheequationsbecome:

. .
(A10)
.

. .
(A11)
.
ThestudyofColemanetal.claimedthatforwellswithlowflowingsurfacepressuresthis
20% adjustment is not needed. For both equations, with or without the adjustment; the
criticalrateequationcanbewrittenas:
.
(A12)
76
APPENDIXB
PLUNGERLIFTEQUATIONSANDFEASIBILITYCHARTS
Thisappendixgivesasummaryonplungerliftequationsandpresentsplungerliftfeasibility
charts.
B.1MinimumCasingPressure
The moment the plunger and the liquid column above the plunger reaches surface,
requiredminimumcasingpressureatthesurfaceis:
, 14.7 1 (B1)
Where;
PP=pressurerequiredtolifttheplunger,psia
PC=pressurerequiredtolift1bblofliquidovercomingfriction,psia
SV=liquidvolumeaboveplunger,bbl
K=factorofgasfrictionbelowtheplunger
D=plungerdepth,ft
KandPciscalculatedfrom:
1.030 10 (B2)
0.433 3.594 10 (B3)
77
B.2MaximumCasingPressure
Themaximumcasingpressureisthencalculatedbytheequation:
, , (B4)
Theseequationsassumethatallthepotentialenergyofgasisconvertedintokineticenergy
when lifting the plunger to surface. The losses due to efficiency and possible gas leaks
aroundplungerofothercomponentsareomitted,thoughtheycanbetakenintoaccount
withcorrections.
B.3PlungerFeasibilityCharts
FigureB.1LiquidProductionChartforPlungerLift16
78
FigureB.22PlungerFeasibilityChartfor23/8Tubing15
79
FigureB.32PlungerFeasibilityChartfor27/8Tubing15

80
APPENDIXC
INFLOWPERFORMANCECALCULATIONSOFWELLS#10&#28
This Appendix summarizes the calculation of inflow performance curves for wells used in
the case study section. Mishra and Caudle23 proposed a simplified procedure on gas
deliverability calculations using single point tests rather than isochronal tests which
consume considerable amount of time especially on low permeability systems. In their
work,theyproposedanequationfordeterminingthegasdeliverabilityinunfracturedgas
wells:

1 5 (C.1)
Where,m(P)istherealgaspseudopressureevaluatedatpressure,P,(psia2/cp).
C.1Well#10
Using the equation proposed in the study of Mishra and Caudle23 (1984), the inflow
performancecurveofwell#10iscalculated.Forthesecalculations,thereservoirpressureis
takenfromthepressuresurveycarriedoutinitiallytoproveliquidaccumulation.Also,the
datagatheredduringblowdownoperationsareusedtofindabsoluteopenflowpotential,
sincethegasflowrateismeasuredas1.05MMscf/dwhenthewellisflowntoatmosphere.
Firstrealgaspseudopressurearecalculated,TableC.1showspseudopressureandgasflow
ratevaluesforvariousflowingbottomholepressures:
PR=3491psia
PR2=1.22E+07psia2
m(PR)=6.09E+08psia2/cp
81
TableC.1CalculatedInflowPerformanceDataforWell#10
Pwf m(Pwf) m(Pwf)/m(PR) q/qmax q(Mscf/d) q(MMscf/d)
3491 6.09E+08 1.00E+00 0.00E+00 0.000 0.000
3400 5.78E+08 9.49E01 9.93E02 104.810 0.105
3200 5.12E+08 8.40E01 2.83E01 299.007 0.299
3000 4.50E+08 7.38E01 4.29E01 453.040 0.453
2800 3.92E+08 6.43E01 5.46E01 576.000 0.576
2600 3.38E+08 5.55E01 6.40E01 674.727 0.675
2400 2.88E+08 4.73E01 7.15E01 754.401 0.754
2200 2.42E+08 3.97E01 7.76E01 818.966 0.819
2000 2.00E+08 3.28E01 8.26E01 871.443 0.871
1800 1.62E+08 2.66E01 8.67E01 914.158 0.914
1600 1.28E+08 2.10E01 8.99E01 948.907 0.949
1400 9.80E+07 1.61E01 9.26E01 977.081 0.977
1200 7.20E+07 1.18E01 9.48E01 999.757 1.000
1000 5.00E+07 8.21E02 9.65E01 1017.764 1.018
800 3.20E+07 5.25E02 9.78E01 1031.739 1.032
600 1.80E+07 2.95E02 9.88E01 1042.158 1.042
400 8.00E+06 1.31E02 9.95E01 1049.368 1.049
200 2.00E+06 3.28E03 9.99E01 1053.603 1.054
100 5.00E+05 8.21E04 1.00E+00 1054.651 1.055
Usingcalculatedinflowperformancedataforvariousflowingbottomholepressurevalues,
theinflowperformancecurvecanbeplottedtofindgasdeliverability.FigureC.1showsthe
plottedIPRcurveofwell#10.
82
FigureC.1
1IPRcurveofWell#10
C
C.2Well#28
T same pro
The ocedure is fo well #28 for inflow performance calcullations, as weell
ollowed for w
#
#10.Therese
ervoirpressureistakenfrromthedow
wnholepressu
uresurveyan
ndthegasflo
ow
r
rateistaken fromthedattagatheredfrromblowing downtheweelltoatmosphere.TableC
C.2
s
showsthecalculateddataforinflowpeerformance.
PR=1425psiaa
06psia2
PR2=2.03E+0
m R)=1.02EE+08psia2/cp
m(P
83
TableC.2
CalculatedIInflowPerformanceDatafforWell#28
Pwf m(Pwf) m(Pwf)/m(PR) q//qmax q(Mscf/d) q(MMscf/d))
1400 9.80E+0
07 9.6
65E01 6
6.80E02 48.994 0.049
9
1300 8.45E+0
07 8.3
32E01 2
2.96E01 212.940 0.213
3
1200 7.20E+0
07 7.0
09E01 4
4.67E01 336.439 0.336
6
1100 6.05E+0
07 5.9
96E01 5
5.98E01 430.353 0.430
0
1000 5.00E+0
07 4.9
92E01 6
6.98E01 502.363 0.502
2
900 4.05E+0
07 3.9
99E01 7
7.75E01 557.953 0.558
8
800 3.20E+0
07 3.1
15E01 8
8.35E01 601.070 0.601
1
700 2.45E+0
07 2.4
41E01 8
8.81E01 634.578 0.635
5
600 1.80E+0
07 1.7
77E01 9
9.17E01 660.564 0.661
1
500 1.25E+0
07 1.2
23E01 9
9.45E01 680.555 0.681
1
400 8.00E+0
06 7.8
88E02 9
9.66E01 695.663 0.696
6
300 4.50E+0
06 4.4
43E02 9
9.82E01 706.691 0.707
7
200 2.00E+0
06 1.9
97E02 9
9.92E01 714.202 0.714
4
100 5.00E+0
05 4.9
92E03 9
9.98E01 718.568 0.719
9
U
Using calculaated inflow performancee data, IPR curve
c for weell #28 can be plotted to
d
determinega
asdeliverability.FigureC.2 plottedIPRcurveforwell##28.
2showsthep
FigureC.2
2IPRcurveofWell#28
84

Index PDF

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Index PDF

Hochgeladen von

Copyright:

Verfügbare Formate

OVERVIEWOFSOLUTIONSTOPREVENTLIQUIDLOADINGPROBLEMSINGASWELLS

Rezervuar basncnn azalmasyla zamanla gaz kuyularnn retimleri azalr. Formasyon

Dependingonthereservoir,bottomhole andtubinghead pressures water and/or

If liquids begin accumulating in the bottomhole, the increased pressure caused by

Liquidphaseaccumulationonthewallsof stringduringatwophaseflow isofinterestin

considered to be stable and significantly high enough to effectively transport produced

Thestudy byMcMullan &Bassiouni12(2000) showed thatone ofthemain characteristics

Although gas production rate is dependent considerably on the length of perforation

Liquid loading in a gas producing well is a progressing problem as reservoir pressure

As mentioned, a typical conventional plunger lift application consists of cycles with

Usingplungerliftasanartificialliftmethodtoovercomeliquid loading solutionsrequires

gasvelocitywill be above critical velocitytherefore theliquidswillnot accumulateat the

Thereservoirfromwhere bothwells are producing gas along with waterandcondensate

Days GasRate,MMscf/d LiquidProduction,bbl GLR,scf/bbl

ous tubing sizes shows th

5. Veeken, K., Bakker,E., andVerbeek,P.,Evaluating Liquid LoadingField Data and

11. Christiansen,R.L.,Girija,E.,RigginsL.andElsener,G.,LiquidLifting fromNatural

20. Hinze, J.O., Fundamentals of the Hydrodynamic Mechanism of Splitting in

0.433 3.594 10 (B3)

Pwf m(Pwf) m(Pwf)/m(PR) q/qmax q(Mscf/d) q(MMscf/d)

3491 6.09E+08 1.00E+00 0.00E+00 0.000 0.000

3400 5.78E+08 9.49E01 9.93E02 104.810 0.105

3200 5.12E+08 8.40E01 2.83E01 299.007 0.299

3000 4.50E+08 7.38E01 4.29E01 453.040 0.453

2800 3.92E+08 6.43E01 5.46E01 576.000 0.576

2600 3.38E+08 5.55E01 6.40E01 674.727 0.675

2400 2.88E+08 4.73E01 7.15E01 754.401 0.754

2200 2.42E+08 3.97E01 7.76E01 818.966 0.819

2000 2.00E+08 3.28E01 8.26E01 871.443 0.871

1800 1.62E+08 2.66E01 8.67E01 914.158 0.914

1600 1.28E+08 2.10E01 8.99E01 948.907 0.949

1400 9.80E+07 1.61E01 9.26E01 977.081 0.977

1200 7.20E+07 1.18E01 9.48E01 999.757 1.000

1000 5.00E+07 8.21E02 9.65E01 1017.764 1.018

800 3.20E+07 5.25E02 9.78E01 1031.739 1.032

600 1.80E+07 2.95E02 9.88E01 1042.158 1.042

400 8.00E+06 1.31E02 9.95E01 1049.368 1.049

200 2.00E+06 3.28E03 9.99E01 1053.603 1.054

100 5.00E+05 8.21E04 1.00E+00 1054.651 1.055

Pwf m(Pwf) m(Pwf)/m(PR) q//qmax q(Mscf/d) q(MMscf/d))

Das könnte Ihnen auch gefallen