Copyright 2017, Letton Hall Group. This paper was developed for the UPM Forum, 22 – 23 February 2017, Houston, Texas, U.S.A., and is subject to correction by the author(s). The contents of the paper may not necessarily reflect the views of the UPM Forum sponsors or administrator. Reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Letton Hall Group is prohibited. Non-commercial reproduction or distribution may be permitted, provided conspicuous acknowledgment of the UPM Forum and the author(s) is made. For more information, see www.upmforum.com.

Abstract

Production optimization and reservoir managementrelyinpartonperiodicsurveillanceofproducedoil,gasandwater flowratesof individualgascondensatewells.Flowmeasurementatthewellheadhasproventobechallengingdue toanumberof factors.Conventional (gravity-based)TestSeparators(CTS)andin-lineMulti-PhaseFlowMeters(MPFM)areintrusiveandmaybecostlytomobilize,nottomentiontheassociatedHSErelatedrisks.Forfieldswhereaccess and logistics pose a challenge, the deployment ofeither CTS or MPFM packages may require a significantamount of time. In some cases, the required well testfrequency may not be reached with the availableequipment or due to the increased number of wells inproduction.

This paper describes a convenient and cost-effective

approach to production surveillance of gas condensatewells using the non-intrusive SONAR-based Surveillancesystem. The system integrates the clamp-on sonar flowmeter(SONAR)withaPVTandmultiphaseflowenginetocalculate the properties of the produced fluids, and theindividual phase flow rates. The SONAR is amultiphase-tolerantvolumetricflowdevicethatprovidesthebulkflowrateofthefluidstreamwithintheflowline.ThePVTenginecalculatestheindividualphasepropertiesoftheproducedfluids, at the pressure and temperature conditionsmeasuredwheretheSONARisclamped-on.Themeasuredflowrateistheninterpretedintermsofoil,gasandwaterratesatboththeactualandstandardconditions.

ExamplesoffieldperformanceofSONAR-basedsystem

forthree-phaseproductionsurveillanceofgascondensatewellsarepresented.ThetestresultsshowthattheSONAR-basedsystemrepresentsareliable,safeandcost-effectivesolution for recurring production surveillance wherereservoirconditionsarerelativelystableovertime.

Introduction

Monitoringproducedoil,gas,andwaterfromindividualwells plays an important role in reservoir management.Acquiringtimelyandaccuratewellheadmeasurementscanbechallengingduetoarangeoffactors.Usingtraditionalwelltestseparatorsprovidesonlyperiodicmeasurements

and may result in deferred production due to pressuredrop.

While measuring dry gas flow measurement is well-

servedbyawiderangeofgasflowmeteringtechnologies,accurateandcost-effectivemeasurementofwetgasflowremainsachallengefortheupstreamoilandgasindustry.The paper provides a method to provide minimallyintrusive, clamp-on production surveillance for wet gasflowmeasurement.

Exprodevelopedaclamp-onnon-intrusiveproduction

monitoring system, Total Production Surveillance (TPS),designed to provide rapid and cost-effective wellheadsurveillance forawide rangeofapplications, suchasgasandgascondensatewells.Expro’sclamp-onSONAR-based(Sonar) flow meters, PassiveSONAR™ or ActiveSONAR™,operate effectively in multiphase conditions typical formost oil and gas applications. Therefore, they can beinstalled close to the wellhead, allowing accurate flowmeasurementonanindividualwellbasiswithoutinstallinganinlinedeviceordivertingproduction.

SONAR-basedFlowMeasurementTechnology

History. Sonar flow measurement technology was

introduced to the oil and gas industry in 2000, usingSONAR-basedpassive-listeningtechniquestoprovideflowrate and compositional information for downholeapplications(fiber-opticmeter).

In 2004, a clamp-on version of a strain-based sonar

meter (PassiveSONAR™)was introducedwithpiezo-strainsensors,providingsimilarfunctionalityastheoriginalfiber-optic-strain sonar technology with reduced cost andcomplexity. PassiveSONAR™hasbeenappliedtoawiderangeof flowmeasurementapplications, including singleandmultiphasemixtures,fromgas/liquidmixturestohighsolidscontentslurries.

In2009,ExproMetersintroducedtheActiveSONAR™,

based on the pulsed-array sonar technology, usingexternally generated acoustic pulses to “illuminate” thevortical coherent structures. The pulsed-array sonartechnologyallows themeasurementof lower flowrates

UPM 17050

SONAR-based Wellhead Surveillance for Gas Condensate Fields Gabriel Dragnea, Expro Meters Inc., Siddesh Sridhar, Expro Meters Inc.

Copyright 2017, Letton Hall Group. This paper was developed for the UPM Forum, 22 – 23 February 2017, Houston, Texas, U.S.A., and is subject to correction by the author(s). The contents of the paper may not necessarily reflect the views of the UPM Forum sponsors or administrator. Reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Letton Hall Group is prohibited. Non-commercial reproduction or distribution may be permitted, provided conspicuous acknowledgment of the UPM Forum and the author(s) is made. For more information, see www.upmforum.com.

2

thanstrain-basedsonarflowmeters.TheActiveSONAR™meteriswellsuitedforGasandGasCondensatewellsinheavyschedulepipes.ExproMetersutilizesbothtypesofClamp-on SONAR flow measurement technologies toaddressawiderangeofflowlineconditions.

SONAR-basedFlowMeasurement.Sonar-basedflow

measurementsutilizeanarrayofsensors,alignedaxiallyalong the pipe, to characterize the speed at whichnaturallyoccurring,coherentflowstructuresconvectpastthesensorarrayusingsonarprocessingtechniques.Sinceboth single andmultiphase flows typically exhibit thesecoherent structures, the methodology is suitable for awiderangeofapplications.

Figure1showsthenaturallyoccurring,self-generated,

coherent structures characteristic for turbulent flow ofNewtonianfluids.Theturbulenteddiesaresuperimposedover the time-averagedvelocityprofiles andare carriedalongwiththemeanflow.Theseeddiesremaincoherentfor several pipe diameters and convect at, or near, theaverageflowrateinthepipe.

Figure 1: Sonar-based Flow Meters with Coherent

StructuresSonar flow meters use the convection velocity of

coherentstructures(eddies)todeterminevolumetricflowrate.Thesonar-basedalgorithmsdeterminethespeedofthesestructuresbycharacterizingboththetemporalandspatial frequencycharacteristicsof the flow field. Foraseriesofcoherenteddiesconvectingpastafixedarrayofsensors, the temporal and spatial frequency content ofpressure fluctuations are related through a dispersionrelationship,expressedasfollows:

k

Vconvectw

= (1)

Where, Vconvect is the convection velocity of the

disturbance,ωthetemporalfrequency(rad/sec),kisthe

wave number (defined as k=2π/λ) and λ is the spatialwavelength.

In sonar array processing, the spatial/temporalfrequencycontentofsoundfieldsaredisplayedusing“k-ω" plots in which the power of the sound field isdecomposed into bins corresponding to specific spatialwavenumbersandtemporalfrequencies(Figure2).Onak-ωplot, thepowerassociatedwithcoherentstructuresconvecting with the flow is distributed along “theconvective ridge”. The slope of the convective ridgerepresents the speed of the turbulent eddies, which isthenconvertedtovolumetricflow.

Figure2:k-ωplotwithconvectiveridgeSonarWetGasOver-readingCorrelation.Whenusing

the Sonar forwet gas flowmeasurement, the reportedflow rate is higher than the actual flow rate of the gasphaseofthemixture,similartothedP-basedflowmeters(Venturi,orifice,etc.).ThemagnitudeoftheSonarover-reaadingistypicallylessthanthatofthedP-basedmeters.

The over-reading of a sonarmeter is defined as the

ratiobetweenthereported flowvelocity (Vsonar)andthegassuperficialvelocity(Vsg):

sg

sonar VsonarORV

= (2)

Thefollowingempiricalcorrelationwasdevelopedto

characterize the over reading of pulsed-array sonarmeters for wet gas streams flowing in fully developed,horizontalflowlines:

DK

Dw

Slope– 11.57fps

Copyright 2017, Letton Hall Group. This paper was developed for the UPM Forum, 22 – 23 February 2017, Houston, Texas, U.S.A., and is subject to correction by the author(s). The contents of the paper may not necessarily reflect the views of the UPM Forum sponsors or administrator. Reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Letton Hall Group is prohibited. Non-commercial reproduction or distribution may be permitted, provided conspicuous acknowledgment of the UPM Forum and the author(s) is made. For more information, see www.upmforum.com.

3

2

111 ÷÷

ø

öççè

æ

++÷÷ø

öççè

æ

++= mmsonar Fr

LVFFrLVFOR jb (3)

Where,LVF is the liquid volume fraction (LVF) and Fr

is the gas densimetric Froude number (Fr); β andϕ arecalibrationconstantsand

îíì

³<<-

=5.1 ,1

5.10.5 ,5.0

r

rr

FFF

m (4)

Total Production Surveillance System (TPS). The

system integrates the Sonar flowmeasurementwith anEquation of State (EoS) model for the Pressure,Temperature and Volumetric (PVT) properties of theproducedfluidsingascondensatewellsasshowninFigure3.

Figure 3: Schematic Total Production Surveillance

MethodologyWell bore composition is input by specifying

molecularcompositionofthewellborefluid.TheEoSPVTmodelcalculatesthepropertiesofthewellborefluidatthe location of the sonar meter using the measuredpressure and temperature. The model also calculatesmixture properties such as liquid volume fraction (LVF)andgasFroudenumber(Fr).Thoseparametersareusedinconjunctionwiththeempiricalcorrelationfortheover-reading of the sonar meter to correct the sonar flowvelocityof themixture in termsof actual gas flow rate.Oncethegasflowrateisdeterminedatactualconditions,theoilandwaterratesarethendeterminedfromthePVTmodel. The total mixture is flashed to standardconditions,andgas,oil(condensate)andwaterratesarereportedatstandardconditions.

AschematicofthethetypicalembodimentoftheTPS

systematthewellheadisshowninFigure4:

Figure 4: Schematic Total Production Surveillance

MethodologyTheover-readingcorrectionenablestheActiveSONAR

metertoreportgasratestowithin+/-3%for0<LVF<0.106and0.5<Fr<5.78.Figure5 illustratestheresultsfora4inSonartestedinthewetgasloopatCEESI,CO.

Figure5:Wetgasover-readingcorrection(4inSonar)

TPSSensitivitywithCompositionalParameters.SincedefiningthewellborecompositionisequivalenttospecifyingtheCGR(condensategasratio)andWGR(watergasratio)ofthegascondensatestream,theaccuracyofthecompositionalinformationwouldhaveadirectimpactontheindividualphaseflowratesinferredbytheTPSsystem.

Typically, the sonar-based surveillance relies on the

knowledge of the well bore composition from eitherhistorical well tests or/and fluid sampling. Therefore, abriefsensitivityanalysiswouldbebeneficialtoquantifythemagnitude of flow errors due to errors in compositionalparameters.

MultiphaseFlow&Compositional

Models

P,T

CompostionalInformation

Qoil

Qgas

Qwater

(ClientProvided)

VSONAR

Input

Measurements

Output

TPS1000SystemDiagram

©CopyrightExpro2011ExproMetersWELLFLOWMANAGEMENT

Copyright 2017, Letton Hall Group. This paper was developed for the UPM Forum, 22 – 23 February 2017, Houston, Texas, U.S.A., and is subject to correction by the author(s). The contents of the paper may not necessarily reflect the views of the UPM Forum sponsors or administrator. Reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Letton Hall Group is prohibited. Non-commercial reproduction or distribution may be permitted, provided conspicuous acknowledgment of the UPM Forum and the author(s) is made. For more information, see www.upmforum.com.

4

Forsimplicity,theflowconditionsfromoneofthetrialspresentedhereinwereused(Table1).TherearetwoCGRvalues used for the analysis, 30STB/mmscf and70STB/mmscf, which are varied by ±20% each, whilekeepingthewatercut(WC)constant,at5%.Theneteffectof±20%changeinCGRisthatboththeoilandwaterflowratesshiftby±20%.

If the CGR is held constant (30STB/mmscf and

70STB/mmscf), andWC is varied by ±2% (absolute), thewater flowratechangesby±40%,which is similar to therelativechangeinWC.

Table1:TPSsensitivitywithwellcomposition

Theanalysisshowstheimportanceofhavingaccurate

compositional inputs in The TPS system in order for theinferred phase rates to be accurate. Therefore, thecompositional information shouldbeupdatedevery timethereisasignificantchangeinthewetgasparameters.

Fieldperformance

Inmanyplacesaroundtheworld,the localregulatorybodyrequiresthewellstobetestedatacertainfrequency,buttheexistinginfrastructuremaynotallowtheoperatorsto comply easily with those requirements. In mostinstances, it could lead to a significant increase of theoperatingcosts,especiallyfortheoffshoreandremotelandfieldswherephysicalaccessposeschallenges.

Oftentimes,thefieldpipinglayoutdoesnotincludethe

means(testseparators)totesteachindividualwellandtheproduction allocation becomes a serious issue for bothfinancialandtechnicalreasons.

Production surveillance and monitoring of individual

wellsusingtheclamp-onSONAR-basedTPSsystem,allowsoperators tomeasure the production of eachwellmorefrequentlyandwithoutdeferringproduction.Also,duetothe clamp-on design, the well production is measuredunder normal operating conditions, helping to identifyunderperformingwellsortheneedforworkover.

The test data from two gas condensate field trials ispresentedherein,oneoffshoreandtheotheronland.

OffshoreTrial.Theoffshoretrialwasconductedinagas

condensate field with two satellite platforms and oneproductionplatform.Thesatelliteplatformsareequippedwithproductionseparators,butnotestseparators.

The client was experiencing significant production

losseswhendivertingwellsthroughthetestseparatordueto backpressure effects. Additionally, the client waslooking to obtain production surveillance data at normalwell flowconditions,unaffectedby the limitationsof thetest separators. The client wanted to increase both thefrequency and efficiency of productiontesting/surveillance.

ElevenwellsweretestedonplatformsAandB:A1,A2,

A3,A4,A5,B1,B2,B3,B4,B5andB6,overa7dayperiod.

Figure6:Sonarflowmeter(6in)The flow data recorded by the flow meter was post

processed using Expro Meters' proprietary TPS1000software,alongwithwellheadpressureandtemperature(provided by client), a well stream compositionreconstructedwithCGRandWGR(watergasratio)valuesdeterminedfromtheApril2014welltestreport,toreportgas,oilandwaterratesatstandardconditions.

Well A5was tested before and after a facility layout

modification,aspartofanoptimizationscheme.Reviewofthe data for both tests (Pre- and Post- Modification)indicated that the flow measurement increased by 2.4MMscfdinthepost-testflowdata.

Well CGR Water Cut WGR Qgas Qoil Qw ater Qgas Error Qoil Error Qw ater ErrorSTB/MMscf STB/MMscf MMscfd STB/D STB/D % % %

U 30 5% 1.58 79.00 2370.0 124.824 5% 1.26 78.79 1891.4 99.3 -0.27% -20.19% -20.43%36 5% 1.89 79.17 2850.0 149.6 0.22% 20.25% 19.87%

V 70 5% 3.68 80.11 5607.7 294.856 5% 2.94 79.73 4464.6 234.4 -0.47% -20.38% -20.49%84 5% 4.42 80.50 6762.1 354.2 0.49% 20.59% 20.15%

W 30 5% 1.58 79.00 2370.0 124.830 3% 0.92 79.02 2370.6 72.7 0.03% 0.03% -41.75%30 7% 2.26 78.98 2369.4 178.5 -0.03% -0.03% 43.03%

X 70 5% 3.68 80.11 5607.7 294.870 3% 2.16 80.15 5610.6 173.1 0.05% 0.05% -41.28%70 7% 5.26 80.07 5604.7 421.1 -0.05% -0.05% 42.84%

Vsonar 75 ft/sPressure 1400 psig

Temperature 200 degFPipe Size 6in Sch 160

Sonar Wet Gas Sensitivity

Copyright 2017, Letton Hall Group. This paper was developed for the UPM Forum, 22 – 23 February 2017, Houston, Texas, U.S.A., and is subject to correction by the author(s). The contents of the paper may not necessarily reflect the views of the UPM Forum sponsors or administrator. Reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Letton Hall Group is prohibited. Non-commercial reproduction or distribution may be permitted, provided conspicuous acknowledgment of the UPM Forum and the author(s) is made. For more information, see www.upmforum.com.

5

Figure7:SonarflowmeterresultsPlatformA

Figure8:SonarflowmeterresultsPlatformB.

The total3-phase flowrates for theAandBwells,as

reported by Expro Meters, have been compared to theproductionseparatorA(platformA),B1andB2(platformB)flowrates.ExproreportedgasratesfortheAwellswerewithin 3% of the A platform production separator rates.ThereportedgasratesfortheBwellswerewithin-3%ofthe B1 and B2 platform production separator rates. Thecondensaterateswerewithin4.5%,whilethewaterrateswithin12%forSeparatorB,whichmaybeindicativeofanincorrectWGR.

As a result of the trial, the client decided to employ

Exprotoperformthewellsurveillanceonaquarterlybasisandexpandthescopetotheentirefield.

Land Trial. The land trialwas conducted in a remote

highpressuregascondensatefield;thepressuresrecordeddownstream of the choke were between 1700psig and2150psig. The client was looking to increase well testfrequencyduetothelocalregulationsandpossiblyreducethecost.

Attherequestoftheclient,ExproMetersperformeda

wellhead surveillance trial on twowells: C1 and C2. TheTPS1000 softwarewas then used to provide single pointvolumetricflowrates(oil,gasandwater)onsite,basedonaveragevelocitymesurements.

ForwellC1,theSONARmeterwasinstalledonthiswell

forsixdays. Theflowdatawaspostprocessedusinglinepressure and temperature readings along with a

wellstream composition (CGR - 33.84 STB/MMSCF andWGR-0.94STB/MMSCF)providedby theclient.Thegasand condensate rates were within 4% of the separatorrates,whilethewaterwithin10%.

The SONARmeterwas installed onwell C2 for three

days.Theflowdatawaspostprocessedusinglinepressureand temperature readings along with a wellstreamcomposition (CGR - 47.1 STB/MMSCF and WGR - 8.715STB/MMSCF) provided by the client. The reported flowrates (gas, condensate water) were all within 3% of thereference.

Figure9:Sonarflowmeter(10in)wellC1.

Figure10:SonarflowmeterresultswellsC1andC2.Theclientiscurrentlylookingintowhatrequirements

need to be met in order to qualify the Sonar-basedSurveillance forgascondensatewell testingby the localregulatorybody.

Conclusions

Aclamp-onproductionsurveillancesystemdesignedtomonitorgascondensatewellswaspresented.Thesystememploysaclamp-onsonarflowmeterastheflowmeteringelement,integratedwithanEquationofStatePVTmodel.Intheapproachdescribedherein,producedoilandwater

Copyright 2017, Letton Hall Group. This paper was developed for the UPM Forum, 22 – 23 February 2017, Houston, Texas, U.S.A., and is subject to correction by the author(s). The contents of the paper may not necessarily reflect the views of the UPM Forum sponsors or administrator. Reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Letton Hall Group is prohibited. Non-commercial reproduction or distribution may be permitted, provided conspicuous acknowledgment of the UPM Forum and the author(s) is made. For more information, see www.upmforum.com.

6

ratesareinferredfromthemeasuredgasrateusingauser-definedwellborecomposition.

Whencombinedwithaccuratewellborecomposition

data, the clamp-on production surveillance systemprovides a practical, cost-effective, real-time surveillanceofgascondensatewells.

Two field cases are presented illustrating the

applicationofthissystemtogascondensatewells.Thetwocases utilized historical well bore compositional data toprovide three-phase surveillance on a periodic or surveybasis.TheresultsforbothtrialsprovedthatSonar-basedsurveillance(TPS)isaviablealternativeforgascondensatewellheadsurveillance.

While the system measures variations of produced

liquids due to variations in gas production, it will notmeasure variations in CGR and/or WC. To account forchanges in CGR or/and WC, an updated well-borecompositionmustbeenteredintothesystem.

Traditional well testing approaches, such as CTS and

MPFM may require a significant amount of time to bedeployed in the field. The Sonar clamp-on approachrequires a shorter amount of time (about 90min) forinstallationandcommissioningwhichallowsthepossibilitytoperformmulti-ratetestingofthewellsormultiplewelltesting in one day. Therefore, the sonar clamp-onmethodology offers the opportunity to increase thewelltestfrequencyatafield-widelevelthusallowingabetterfield/productionmanagement.

The Sonar-based Surveillance is not to replace

traditional well testing methodologies, but to augmentthembyofferingaquick,reliableandcosteffectivesolutionfor applications requiring recurring productionsurveillance, especially where reservoir conditions arerelativelystableovertime.

AcknowledgementsTheauthorsgratefullyacknowledgeExproGroupforpermissiontopublishthisworkandacknowledgetheeffortsofourcolleagueswhohavecontributedtothispaper.Nomenclature

P CTS = Conventional Test Separator V MPFM = Multiphase Flow Meter T TPS = Total Production Surveillance V Vconvect = Convection velocity of turbulent eddies T ω = Temporal Frequency MPFM = Multiphase Flow Meter

T TPS = Total Production Surveillance V Vconvect = Convection velocity of turbulent eddies T ω = Temporal Frequency k = Wave Number λ = Wave Length T ORsonar= Sonar Wet Gas Over-reading V Vsonar = Sonar Velocity T Vsg = Superficial Gas Velocity T LVF = Liquid Volume Fraction Fr = Gas Froude Number β = Sonar Wet Gas Correction Constant T φ = Sonar Wet Gas Correction Constant V m = Sonar Wet GasCorrection Coefficient T EoS = Equation of State CGR = Condensate Gas Ratio T WGR = Water Gas Ratio WGR WC = Water Cut

References

[1] Danesh,A.,PVTandPhaseBehaviourofPetroleumReservoir

Fluids,DevelopmentsinPetroleumScience,Vol.47,ElsevierScienceB.V.,ISBN:0-444-82196-1,1998

[2] GyslingD.L.,Loose,D.H.,vanderSpek,Clamp-on,SONAR-

based,VolumetricFlowandGasVolumeFractionMeterIndustrialApplications,presentedatthe13thInternationalFlowMeasurementConference,NEL,EastKilbride,Scotland,UK,June2005

[3] Gysling,D.L.,Loose,D.H.,andVolz,R.,WetGasMetering

UsingaCombinationofDifferentialPressureandSONARFlowMeters,presentedatthe24thInternationalNorthSeaFlowMeasurementWorkshop,NEL,EastKilbride,Scotland,UK,Oct2006

[4] Gysling,D.andMorlino,N.,“Pulsed-Array,Sonar-basedFlow

MeasurementTechnologyforClamp-onWetGasMetering,Presentedatthe2009AmericasFlowMeasurementWorkshop,Houston,TX,USAFebruary,2009.

[5] Kragas,T.K.,etal.,Downhole,FiberOpticMultiphaseFlow

Meter:FieldInstallation,presentedattheSocietyofPetroleumEngineersAnnualConference,SPEPaper#77654,September2002

[6] Kragas,T.K.,etal.,DownholeFiber-OpticMultiphase

Flowmeter:Design,OperatingPrinciple,andTesting,presentedatSocietyofPetroleumEngineersAnnualConference,SPEPaper#77655,September2002

[7] Shoham,O.,MechanisticModelingofGas-liquidTwo-Phase

FlowinPipes,SocietyofPetroleumEngineers,2006,ISBN978-1-55563-107-9

[8] Steven,R.,WetGasFlowMeteringwithGasMeter

Technologies,Nunn,ColoradoEngineeringExperimentStationInc.,9/1/2006,Ciateq2006