• / 6
  • 下载费用:5 下载币  

1-s2.0-S1876380412600858-main

关 键 词:
s2 S1876380412600858 main
资源描述:
RESEARCHPAPERPETROLEUMEXPLORATIONANDDEVELOPMENTVolume39,Issue5,October2012OnlineEnglisheditionoftheChineselanguagejournalCitethisarticleas:PETROL.EXPLOR.DEVELOP.,2012,39(5):624–629.Receiveddate:07July2011;Reviseddate:30Mar.2012.*Correspondingauthor.E-mail:wangjing8510@163.comFoundationitem:SupportedbytheNationalKeyBasicResearchandDevelopmentPlan(973)Project(2011CB201006)andtheNationalMajorS&TProjectofChina(2011ZX05014-003-008HZ).Copyright©2012,ResearchInstituteofPetroleumExplorationandDevelopment,PetroChina.PublishedbyElsevierBV.Allrightsreserved.Formationmechanismanddistributionlawofremainingoilinfracture-cavityreservoirWANGJing1,*,LIUHuiqing1,XUJie1,ZHANGHongfang21.MOEKeyLaboratoryofPetroleumEngineeringinChinaUniversityofPetroleum,Beijing102249,China;2.PetroleumExplorationandProductionInstitute,Sinopec,Beijing100083,ChinaAbstract:Drillinganddewaxingmethodsareusedtocreatethequantitativemodelandstochasticmodelforfracture-cavityreservoirsrespectively,whichareusedtoinvestigateoildisplacementinfulldiametercoresandtostudytheformationmechanismanddistributionlawofremainingoilinfracture-cavityreservoirs.Thedensitydifferencebetweendisplacingphaseanddisplacedphaseandtheconnec-tionpositionbetweenfracturesandcavitiesdeterminethefluiddistributioninthecave.Afterthefirstwaterflooding,theremainingoiliscomposedofthefollowingparts:the“atticoil”causedbythedensitydifferencebetweenoilandwater,the“occlusionoil”formedbe-causetheoil-to-waterreplacementisnotinstantaneous,the“corneroil”causedbytheirregularityofthecave,the“blindcavityoil”formedbecauseofthecomplexconnectionorlowconnectivityofthefracture-cavitysystem,andthe“oilfilm”causedbythewettabilityofrocks.Whentheconnectionishigher,theamountofthe“atticoil”isless.Whentheinjectingvelocityislowerandtheviscosityofcrudeoilislower,theamountof“occlusionoil”isless.Thecontentof“corneroil”willbecomelesswhenthecavefigureismoreregular,andtheamountofcornerisless.Theamountof“blindcavityoil”islesswhentheconnectionissimpleortheconnectivityishigh.Inad-dition,onthewater-wetrocksurface,thecontentof“oilfilm”isless.Keywords:fracture-cavityreservoir;fulldiametercore;remainingoil;formationmechanism;distributionlawIntroductionNumerousfracturesandcavesaredevelopedintheOrdovi-ciancarbonatereservoirintheTaheOilfield,andfluidsarestoredinbothfacturesandcaves.Matrixhasnocapabilityofstorageandpermeability[1g163].Carbonatefracture-cavityreser-voirsarecharacterizedbystrongheterogeneity,complicatedspacestructure,randomdistributionoffracturesandcaves,differentsizesofcavesandlargedipangleoffractures.Theexplorationanddevelopmentofthecarbonatefracture-cavityreservoirsisstillinitsinfancyallovertheworld[4g166].Inrecentyears,manyexpertshavedonealotofinvestigationsonitbyscalemodeling,butduetothelimitationofcoring,thespacestructureofthephysicalmodelsadoptedbythemareallregu-laranddeterminate,orapproximatedassand-packmodels[7g1610].Thesesimplifiedmodelscannotshowthespacedistributionofcavescorrectly,makingitdifficulttodesignandadjustdevelopmentscheme.Inthispaper,thequantitativemodelandstochasticmodeloffracture-cavityreservoirsarebuiltbythedrillinganddewaxingmethodrespectively,whichareusedtostudytheoildisplacementoffulldiametercore.Inthemean-time,theformationmechanismanddistributionrulesofdif-ferenttypesofremainingoilinthereservoiratdifferentdis-placementperiodsareanalyzed,whichcontributetothepro-ductionoffracture-cavityreservoirs.1FFulldiametercoremodelforfracture-cavityreservoir1.1PreparationofthequantitativemodelAsthedisplacingprocessbythefulldiametercoremodelisinvisible,thedistributionofoilandwaterinfracturesandcavescannotbeobserveduntilthecoreistakenoutattheendofexperiment,someman-madeerrorsmightexistinthisprocess.Soaquantitativemodelinwhichthedistributionandstructuredimensionsofthefracturesandcavesareknownisfirstlydesigned.Throughtheexperimentresultsofwaterfloodingandgasinjection,theoilandwaterdistributioncha-racteristicsindifferentdisplacingperiodsareanalyzed,whichcanprovideevidenceforinvestigatingoilandwaterdistribu-tionregularityinthestochasticmodel.Themodelismadethroughthefollowingsteps:first,cutthecoreat45g113and60g113anglewiththecrosssectionofthecorerespectively(length300mm,diameter100mm),asshowninFig.1;sec-WANGJingetal./PetroleumExplorationandDevelopment,2012,39(5):624–629g16625g16Fig.1Quantitativemodelofafracture-cavityreservoirond,drillingholes(about17mmindiameter)oneachsectionwithg30116bit;third,bondingeachpartofthecoretogetherwithcoppersheet(thethicknessis0.2mm)oneachbondingarea;andatlast,bundlingthecorewithfilterscreen(613g541minporediameter).Themodelfeaturessimpledistributionoffracturesandcaves,simpleconnectiontype,andregularcavestructure.1.2PParametersofthequantitativemodelThesizeofeachcaveisshowninTable1.Bothcave3(3')andcave8(8')areconnectedtoobliquesectionaswellaslongitudinalsection,andcaves11,12,13areallperpendiculartothelongitudinalsectionandthetotaldepthsofcave11,cave12andcave13are4.0cm,4.5cm,3.5cm,respectively.Accordingtotheabovedata,thevolumeofthecavesineachpartofthecorecanbecalculated.Thetotalvolumesofthecavesbelowandabovethefractureare23.61cm3and26.11cm3respectively.Thevolumeofthecavescrossedbyfracturesis72.16cm3,andthetotalvolumeoffracturesisTable1SizeofeachcaveCavenumberbeneaththefractureAC(cm)CavenumberabovethefractureAD(cm)13.11'2.522.72'2.532.53'0.542.94'2.452.75'2.263.16'0.573.07'0.683.08'0.293.19'0.6103.410'0.211.42cm3.Sothetotalvolumeofcavesandfracturesis133.30cm3,andtheporosityofthecoreis5.14%.1.3PreparationofthestochasticmodelThepreparationofthestochasticmodelinvolves:first,mixsolidparaffinofdifferentsizesandshapeswithpowder(mix-tureofcarbonatepowderandcement,whilethecementactsascementation)andshapethemintocylinder,andwaitinguntilitconcretes.Second,cutthecylinderaccordingtothestandardofhighanglefracturebycorecutter.Third,putallpartsofthecylinderintotheincubatorof85°Cfor12h,melt-ingthesolidparaffinandmakingitflowout.Atlast,bondallpartsofthecylindertogether.Thenastochasticmodeloffulldiametercoreiscompletedwithalengthof330mmandadiameterof100mm.AsshowninFig.2.Fig.2Stochasticmodelofafracture-cavityreservoirWANGJingetal./PetroleumExplorationandDevelopment,2012,39(5):624–629g16626g161.4FeaturesofthestochasticmodelThestochasticmodelmadebydewaxinghasthefollowingfeatures:(1)Cavesandfracturesaredistributedrandomlyinthemodel.(2)Cavesarediverseinsizeandshape,includingcubesof1cm3and8cm3,g3011cmspheres,g3011.5cmcylinders,andirregularpolyhedronof1cm,2cmand3cmlong.Lotsofthecaveshavecomplicatedcorners.(3)Fracturesandcavesconnectwitheachotherincomplexways,includingfrac-ture-cave,cave-cave,cave-fracture-cave,fracture-cave-frac-ture.Thecoordinationnumberofthecavesrangesfrom1to4,andthetypesofcontactbetweenfractureandcaveunderthesamecoordinationnumberincludepointcontact,linecontactandsurfacecontact,etc.(4)Cavesandfracturesareembeddedinsidethecore,fluidsarenotabletoconvergeorflowinthematrix,andthecorehasastrongheterogeneity.Duetotheabovefourfeatures,thestochasticmodelismoreclosetorealreservoirsthanthequantitativemodel.2OOildisplacementexperimentsoffracture-cavityreservoirs2.1ExperimentalconditionsandproceduresTheexperimentalschemeisdesignedaccordingtothephysicalpropertiesofthefluidandoperatingparametersofBlock4inTaheOilfield.Theexperimentconditionsareasfollows:temperature25°C,confiningpressure1.2MPa,backpressure0.5MPa,oilviscosity23.8mPa·s(oilviscosityinthefieldis2–30mPa·s),oildensity0.815g/cm3,injectionwaterviscosity1.2mPa·s,salinity500mg/L.Theexperimentunitmainlyincludesinjectionsystem,dis-placementsystemandmeteringsystem,asshowninFig.3.2.2Experimentalprocedures(1)Inject2–3PVintothecoreandcalculatetheporousvolumeofthemodelonthebasisofvolumetricbalance.Fig.3Theschematicdiagramofoildisplacementexperiments(2)Afteragingthecorefor15h,injectwaterfromthebot-tomforthefirsttime(primarywaterflooding)todisplaceoiluntilthewaterratiointheoutletrisesupto100%,thencalcu-latethetotalliquidproductionandoilproduction.(3)Stopdisplacingandholdupfor15h,thenrepeatstep2.(4)Stopwaterflooding,andinjectgasfromthetopofthecore,untilthemassofthefluidsintheconicalflaskdoesn’tincreaseanymore.Thenmeasurethetotalliquidproductionandoilproduction.2.3RResultsandanalyses2.3.1OildisplacementexperimentsonthequantitativemodelWaterisinjectedattherateof1mL/min,andthewaterlineadvancingspeedisabout2.5cm/min,thatis,36.0m/d,whichiswithinthelimitsofthenormaladvancingspeed(30–150m/d)inthefield[11].TheexperimentresultsareshowninFig.4.FromFig.4wecanseeatthebeginningisthewaterfreeproductionperiod,whichislongandhasahighoilproducingrate.Afterwaterbreakthrough,thewatercutrisesfastandtheoilproducingratedecreasesrapidly,showingsharpfluctuation.Thesephe-nomenaaresimilartothetypicalwater/oildisplacementcha-racteristicsinthefield[12].Afterprimarywaterflooding,theoilproductionis96.8mL,withanoilrecoveryof72.6%,andawaterfreerecoveryof31.8%.Insecondarywaterflooding,theoilproductionis5.5mL,withanoilrecoveryof4.1%.Thetotaloilproductionbywaterfloodingis102.3mL,whichiscloseto107.19cm3,i.e.thetotalvolumeofthecavesbeneaththefracture,thecavescrossedbythefracturesandthefractures.Inthegasinjectionperiod,thetotalliquidproduc-tionis105.8mL,whichiscloseto109.69cm3,i.e.thetotalvolumeofthecavesabovethefracture,thecavescrossedbyfracturesandthefractures.Andinthisperiodtheoilproduc-tionis23.0mL,whichiscloseto26.11cm3,thevolumeofthecavesabovethefracture.Attheendofthegasinjection,thetotaloilproductionis125.3mL,andtheultimaterecovery94%.Observationofthecoreupontaking-outshowsthatthere-mainingliquidismainlywaterandmostofwhichisintheFig.4ResultsofprimarywaterfloodingonthequantitativemodelWANGJingetal./PetroleumExplorationandDevelopment,2012,39(5):624–629g16627g16cavesbeneaththefracture(Fig.5a,5b),andalltheotherspacehasnoliquid(Fig.5c,5d).Thismaybeexplainedbythedif-ferenceinoilandwaterdensity.Waterinjectedfromthebot-tomofthecoreflowsintothebottomofthecavesfrompointB,anddisplacestheoilout.Whentheoil/waterlevelrisestopointA,theinjectedwaterflowsoutfrompointA.Therefore,inthewaterfloodingperiod(primarywaterfloodingandsec-ondarywaterflooding),theoilrecoveryismainlyfromthecavesunderpointAandfromthefractures,andtheoilabovepointAcannotberecovered,whichisthesocalled“atticoil”.Butastheoilviscosityismuchlargerthanwaterviscosity,theinjectedwatercannoteasilydisplacetheoilinthecavein-stantly,soafterholdingupforaperiodoftime,thereisstillacertainamountofoilrecoveredintheprocessofsecondarywaterflooding.Sincethegasdensityismuchlowerthanliq-uiddensity,thegasinjectedfromthetopofcoreflowsinpointA,displacingoutthe“atticoil”generatedintheperiodofwaterflooding.Whentheoil/waterleveldropsdowntopointB,thegasflowsoutofthecore,soliquidunderpointBcannotberecovered.Asaresult,inthegasinjectionperiod,liquidsrecoveredarefromcavesabovepointBandfractures,liquidsunderpointB(almostwater)cannotbedrivenout,whichiscommonlycalled“occlusionwater”.Sothedensitydifferencebetweendisplacingphaseanddisplacedphaseandthepositionofthejunctionsoffracturesandcavescanshedlightonthemagnitudeofresidualoilsaturationandconnatewatersaturationafterwaterfloodingorgasinjection.Fig.5alsoshowsthatthereisstillminoroilabsorbingonthesurfaceoffracturesandcavesintheformof"oilfilm"aftergasinjec-tion.2.3.2OildisplacementexperimentsonthestochasticmodelWaterisinjectedattherateof1mL/min,andthewaterlineadvancingspeedisabout2.5cm/min,thatis,36.0m/d.Andthevolumeofporecalculatedbyvolumetricbalanceis310cm3,andporosityis11.96%.TheexperimentresultsareshowninFig.6.Itisfoundthattheoilproducingrate,watercutandrecoveryofthismodelareallsimilartoquantitativemodel,onlywithweakerfluc-tuation.Itisbecausethattherearemorecavesinthestochas-ticmodel,thusmakingitmoresimilartocontinuousmedium.Afterprimarywaterflooding,theoilproductionis206.2mL,andtheoilrecoveryis66.5%,whilethewaterfreerecoveryis26.3%.Intheperiodofsecondarywaterflooding,theoilpro-ductionis11.5mLandtheoilrecoveryis3.7%.Thetotaloilproductionis217.7mLafterthewaterfloodingperiod.Intheprocessofgasinjection,theoilproductionis48.1mL,andthetotaloilproductionaftergasinjectionis265.8mL,theulti-materecoveryis85.7%.Fig.5FluidsdistributioninthecoreaftergasinjectionWANGJingetal./PetroleumExplorationandDevelopment,2012,39(5):624–629g16628g16Fig.6ResultsofprimarywaterfloodinginthestochasticmodelFig.7RecoverycomparisonbetweentwomodelsComparedtheresultsofthetwomodels,asthequantitativemodelhassimpleconnectionandregularcavestructure,thewaterfreerecovery,itswaterfreerecovery,water-drivere-coveryandtheultimaterecoveryareallhigherthanthoseofthestochasticmodel(Fig.7).Todiscusstheinfluenceofoilphysicalpropertiesandin-jectionrateonwaterfreerecovery,recoveryofprimaryandsecondarywater-flooding,oildisplacementexperimentsareperformedatdifferentinjectionratesanddifferentoilviscos-ity(Fig.7).Ataninjectionrateof1mL/minandoilviscosityof23.8mPa·s,waterfreerecoveryis26.3%,primarywa-ter-floodingrecovery66.5%andsecondarywater-floodingrecovery3.7%.Ataninjectionrateof2mL/minandoilvis-cosityof23.8mPa·s,theabovethreekindsofoilrecoveryare21.2%,62.5%and5.2%,respectively.Atthesameinjectionrateof2mL/minbutmuchhigheroilviscosityof39.5mPa·s,theresultsare17.9%,51.5%and6.2%.Soiftheoilviscosityisconstant,thelowertheinjectionrateis,thehigherthewaterfreerecoveryandtheprimarywater-driverecovery,andthelowerthesecondarywater-driverecovery.Andiftheinjectionrateisconstant,thehighertheoilviscosity,thelowerthewa-terfreerecoveryandtheprimarywater-driverecovery,andthegreaterthesecondarywater-driverecovery.Reasonsareasfollows:Wheninjectionrateislow,waterfiguringisslowandoil/watercontactadvancessteadily,alsotheinjectedwaterspendsmoretimeflowinginonecavethusmoreefficientlydisplacestheoilinthebottomofit.Athighoilviscosity,greaterviscousforceandseriousmicroscopicfingeringleadtoquickwaterchanneling.Inthemeantime,sincehigherre-sistanceexistsintheoildisplacementbasedongravity,thedisplacingefficiencyislowandacertainamountofoilre-mainsinthebottomofthecaves,whichismostlyreplacedbywaterinjectedfromaboveintheprocessofsettingandrecov-eredintheperiodofsecondarywaterfloodingperiod.Accordingtotheaboveanalysis,attheendofprimarywa-terflooding,thereisstilllargeamountofoilremaininginthecore,including(1)“Occlusionoil”,retainsbecausetheoil-to-waterreplacementisnotinstantaneous.Theamountofthiskindofoildependsoninjectionrateandoil-waterviscos-itydiscrepancy.Thelowertheinjectionrateandoilviscosityare,thelessthe“occlusionoil”thereis.Andthe“occlusionoil”canberecoveredbysecondarywaterflooding.(2)“Atticoil”storedonthetopofcaves,whichiscausedbythedensitydifferencebetweenoilandwater.Theamountofthiskindofoilisdependentonthepositionofthehighestjunctionbe-tweencavesandfractures.Thehigherthepositionis,theless“atticoil”thereis.And“atticoil”canberecoveredbygasinjection.Afterthegasinjection,theremainingoilincludes(1)“Oilfilm”absorbedonthesurfaceofcavesandfractures(Fig.8a).Theamountof“oilfilm”dependsonwettabilityofrocks.Thereisless“oilfilm”onthesurfaceofwaterwetrocks.(2)“Corneroil”leftinthecornersofirregularcaves(Fig.8b),itsquantitydependsontheshapeofcaves.Thereisless“corneroil”inrelativelyregularcaves.(3)“Blindcavityoil”formedbecauseofthecomplexconnectionorlowconnectivityofthefracture-cavity(Fig.8c,8d).Ther
展开阅读全文
  石油文库所有资源均是用户自行上传分享,仅供网友学习交流,未经上传用户书面授权,请勿作他用。
0条评论

还可以输入200字符

暂无评论,赶快抢占沙发吧。

关于本文
本文标题:1-s2.0-S1876380412600858-main
链接地址:http://www.oilwenku.com/p-70525.html

当前资源信息

吾王的呆毛

编号: 20180607204106142863

类型: 共享资源

格式: PDF

大小: 843.68KB

上传时间: 2018-06-08

广告招租-6
关于我们 - 网站声明 - 网站地图 - 资源地图 - 友情链接 - 网站客服客服 - 联系我们
copyright@ 2016-2020 石油文库网站版权所有
经营许可证编号:川B2-20120048,ICP备案号:蜀ICP备11026253号-10号
收起
展开