• / 8
  • 下载费用:5 下载币  

1-s2.0-S1876380414600756-main

关 键 词:
s2 S1876380414600756 main
资源描述:
RESEARCHPAPERPETROLEUMEXPLORATIONANDDEVELOPMENTVolume41,Issue5,October2014OnlineEnglisheditionoftheChineselanguagejournalCitethisarticleas:PETROL.EXPLOR.DEVELOP.,2014,41(5):634–641.Receiveddate:18Feb.2014;Reviseddate:08Aug.2014.*Correspondingauthor.E-mail:fzf@petrochina.com.cnFoundationitem:SupportedbythePetrochinaScienceandTechnologyMajorProject(2011E-2504).Copyright©2014,ResearchInstituteofPetroleumExplorationandDevelopment,PetroChina.PublishedbyElsevierBV.Allrightsreserved.AstudyonremainingoildistributioninacarbonateoilreservoirbasedonreservoirflowunitsFANZifei1,*,LIKongchou1,LIJianxin1,SONGHeng1,HELing1,2,WUXuelin11.PetroChinaResearchInstituteofPetroleumExploration2.YangtzeUniversity,Hubei434023,ChinaAbstract:Reservoir=north,whichbelongstoZhanzhol,Kazakhstan,afracture-poretypedcarbonateoilfield,hasavarietyofreservoirtypes.Accordingtothecombinationpatternofdifferentvoidsandtherelationshipbetweenporosityandpermeability,carbonatereser-voirsareclassifiedintofourtypeswhicharefracture-cavity-poretyped(referredtoascompositetypedhereafter),fracture-poretyped,poretypedandfracturetyped,andtheidentificationofwhichbywellloggingdataisrealized.Thereisalittledifferenceintheproducingdegreewithinthesamereservoirtypeforcompositetypedandfracturetypedreservoirs.However,therearelargedifferencesinthepro-ducingdegreeforfracture-poretypedandporetypedreservoirs.Inordertoappraisetheproducingdegreeofreservoirsmoreexactly,keyparametersaffectingtheproducingdegreeoffracture-poretypedandporetypedreservoirsareselectedseparately,which,forthefrac-ture-poretypedarereservoirqualityindex(RQI),porethroatradiusatmercurysaturationof50%(R50),totalpermeability(Kf+Km),andtheratiooffracturepermeabilitytomatrixpermeability(Kf/Km),andfortheporetypedareRQI,R50,andmatrixpermeability(Km).Withtheintegratedconsiderationofmatrixandfracture,aflowunitclassificationmethodfordualmediumreservoirisestablished,andtakingtheseparametersasclusteringvariants,fourtypesofreservoirsarefurtherdividedintosixkindsofreservoirflowunitsbythemeansofneuralnetworkclusteringtechnology.Basedonreservoirflowunitmodeling,anumericalsimulationmodelisbuiltwitheachtypeofres-ervoirflowunitsadoptingitscorrespondingrelativepermeabilitycurve,basedonwhich,thedistributionregularityofremainingoilofreservoir=northisstudiedanddemonstrated.Eightnewwellsaredesignedaccordingtotheresearchresult,andputintoproductionin2013withtheinitialdailyoilproductionratebeing2.3timesthatofnearbyoldwells.Keywords:carbonateoilreservoir;reservoirtype;reservoirflowunit;neuralnetworkclusteringtechnology;reservoirflowunitmodel;remainingoilIntroductionCarbonatereservoirs,withastrongheterogeneity,haveavarietyofstoragespacesandtheircombinationtypes[1].Inordertodescribeoilreservoirsmoreaccurately,theconceptofreservoirflowunitisintroduced,which,aftercontinuousim-provementanddevelopment,forthemomentgenerallyreferstothebasicreservoirunitwithconsistentgeological,petrological,andhydrodynamicfeatureswithinagivenoilandgasreservoirsthatisdifferentfromotherrocks[2−3].Inthispaper,basedonthedivisionofreservoirtypes,takingmatrixandfractureintoconsideration,thedivisionmethodofreservoirflowunitisstudied;basedonwhich,fracture-poretypedoilreservoir=northofcarbonateoilfieldZhanzhol,Kazakhstan,isdividedintosixtypesofflowunitsandits3Dgeologicalmodelandnumericalsimulationmodelarebuiltinordertocharacterizethedistributionofremainingoil.1OverviewZhanzholOilfield,locatedintheZharamysskayaUpliftzoneintheeasternmarginofthePre-CaspianBasin,isalargecarbonateoilfieldwithcondensategascapandedge-bottomwater.Itwasputintodevelopmentin1983.Themajoroilreservoir,=northcanbedividedintogroups=Upperand=Lowervertically,andthenfurtherdividedintosixandfoursubzones,respectively.Thereservoir,composedofgrainlimestoneofshoalmicrofaciesandtidalchannelmicrofacies,belongstoopenplatformfaciesdeposition,withstoragespacesofintergranularpore,visceralforamen,intragranularpore,micro-fractureandminordissolvedcavity.Reservoirporosityis6.0%−20.73%,withanaverageof9.70%;perme-abilityis(0.05−518.31)×10−3m2,withanaverageof28.45×10−3m2.FANZifeietal./PetroleumExplorationandDevelopment,2014,41(5):634–641−635−2Reservoirtypedivisionandidentification2.1ReservoirtypedivisionAffectedbycementation,fillingandcompaction,theprimi-tiveporesofoilreservoir=northalmostdisappeared,there-fore,itsporesaremainlydissolvedpores,dissolvedcavitiesanddiageneticfracturesformedbydissolutionatburialstageandstructuralfracturesarepoorlydeveloped.Thedevelop-mentlevelofdissolvedpore,cavityanddiageneticfractureiscloselyrelatedtodepositionalmicrofacies[4].Differenttypesofreservoirshowdifferentpermeationcharacteristicsduetothevarianceofvoidcombinationpatternandreservoirstruc-turalfeatures.Accordingtothecombinationpatternsofdif-ferenttypesofpores,carbonatereservoirscanbedividedintoseventypesbymeansoftriangulardiagramtaxonomy[5](Fig.1).Accordingtoporeshapes,thecombinationpatternofdif-ferentpores,aswellastherelationshipbetweenporosityandpermeability,reservoirsof=northarereducedtofourtypeswhicharecomposite,fracture-pore,poreandfracturetypedinordertoidentifyreservoirtypesbyconventionalwellloggingdata,whichlieintheircharacteristiczonesonporos-ity-permeabilitycrossplot(Fig.2).2.2WellloggingidentificationofdifferentreservoirtypesDifferenttypesofcarbonatereservoirshavedifferentporespacestrcutresandwellloggingresponsefeatures[6].Basedoncoreanalysisdataandreservoirtypedivisionresults,combinedwithimaginglogginginterpretationresults,thewellloggingresponsefeaturesofdifferenttypesofreservoirsin=northoilreservoiraresummerizedinTable1.Table1andFig.3showthatcompositetypedreservoirsarecharacterizedbylowGRvalue,highPEvalue,obviousam-plitudedifferenceinthreeresistivitycurves(LLD,LLSandmicrosphericalresistivity),highvalueofthreeporositycurves(bulkdensity,compensatedneutronandacoustic)(Fig.3a).Fracture-poretypedreservoirsarecharacterizedbylowGRvalue,highPEvalue,medium-lowLLDvalue,smalleram-plitudedifferenceinthreeresistivitycurvesthanthatofcom-positetypedreservoirs,medium-highvalueofthreeporositycurvesandmedium-lowporosityvaluebeingsmallerthanthatofcompositetypedreservoirs(Fig.3b).PoretypedreservoirsarecharacterizedbylowGRvalue,highPEvalue,smallam-plitudedifferenceorcoincidenceofthreeresistivitycurves,medium-lowLLDvalue,medium-lowvalueandcoincidenceofneutronanddensitycurves,medium-lowACvalueandporosityvalueisclosetothatoffracture-poretypedreservoirs(Fig.3c).FracturetypedreservoirsarecharacterizedbylowGRvalue,medium-highPEvalue,lowLLDvalue,sawtooth-shapedmicrosphericalresistivitycurvewithahighervaluethanthatofLLDsometimes,mostlycoincidenceofthreeporositycurvesandlowporosityvalue(Fig.3d).Fig.4showstherelationshipbetweenporosityandperme-abilityofdifferenttypesofreservoirsidentifiedbywelllog-gingdata,whichisconsistentwithwhatisdrawnfromcoredata(Fig.2),indicatingthatthereservoirtypesidentifiedbywellloggingarereasonable.3Flowunitdivisionandcharacterizationbasedonreservoirtype3.1Reservoirflowunitdivisionbasis3.1.1ReservoirporestructureTable2isthecapillarypressurecurveparameterstatisticsofdifferenttypesofreservoirobtainedbymercuryinjectiondata.compositetypedreservoirshavethemostfavorableporeFig.1PoretypeclassificationtrianglediagramofcarbonatereservoirsFig.2Cross-plotofcoreporosity-permeabilityTable1StatisticsofwellloggingresponseofdifferenttypesofreservoirsReservoirtypeGR/APILaterologdeep/(·m)Laterologshallow/(·m)Compensatedneutronlog/%Denstiy/(g·cm−3)Deltatime/(s·m−1)Porosity/%PE/(b·e−1)Composite<3050–100010–100•10”2.54•19912–305.0±Frac.-pore<3010–50010–5006–152.61–2.46182–2126–155.0±Pore<2030–50030–5006–132.35–2.65182–2086–155.08Fracture10–705–2005–201<62.65–2.70165–175<63.14–5.08FANZifeietal./PetroleumExplorationandDevelopment,2014,41(5):634–641−636−Fig.3ConventionalloggingandimagingloggingresponsecharacteristicsofdifferenttypesofreservoirsFig.4Cross-plotofporosity-permeabilityofvarioustypesofreservoirsidentifiedbywellloggingstructureforoilandgasflow,followedbyfracture-poretypedandporetyped,fracturetypedistheworst.compositetypedreservoirshavethelargestporosity,permeability,maximumconnectedporethroatradiusandporethroatradiusatmercurysaturationof50%andthesmallestdisplacementpressure,capillarypressureatmercurysaturationof50%inthreetypesofreservoirs.Forthesamereservoirtypes,poretypedandfracture-poretypedreservoirshavesignificantlywiderangesforeachparameter.3.1.2ReservoirproducingdegreeDifferenttypesofreservoirshavedifferentporestructures,leadingtosignificantlydifferentproducingdegreeduringtheTable2StatisticsofcapillarypressurecurveparameterofdifferenttypesofreservoirrocksamplesPorosity/%Permeability/10−3m2Displacementpressure/MPaMaximumconnectedporethroatradius/mCapillarypressureatmercurysaturationof50%/MPaPorethroatatmer-curysaturationof50%/mReservoirtypeMin.Max.MeanMin.Max.MeanMin.Max.MeanMin.Max.MeanMin.Max.MeanMin.Max.MeanComp.11.2017.5013.2125.90436.0089.010.020.080.039.2337.1231.280.080.610.321.209.563.53Frac.-pore6.0014.609.520.0362.5017.980.020.640.151.1536.9311.730.3115.903.080.052.340.69Pore6.0014.309.220.0229.601.430.081.280.300.5718.423.840.5033.765.460.021.460.34Frac.2.805.904.280.058.111.460.322.560.710.282.311.544.0538.8714.430.010.230.09Total2.8017.5010.280.02436.0024.800.022.560.210.2837.1213.030.0838.873.950.019.561.17FANZifeietal./PetroleumExplorationandDevelopment,2014,41(5):634–641−637−processofdevelopment.Reservoir’sfluidproductionintensityiscalculatedusingthefluidproductionprofiledataof122oilwellsdrilledin=north,accordingtoproducingdegree,thereservoirscanbedividedintofourtypes:reservoirswithgoodproducingdegree,whosefluidproductionisgreaterthanorequalto2m3/(d⋅m);reservoirswithfairproducingdegree,whosefluidproductionis0.5−2.0m3/(d⋅m);reservoirswithpoorproducingdegree,whosefluidproductionislessthan0.5m3/(d⋅m)andreservoirswithzeroproducingdegree.Evaluationresultsofreservoirproducingdegreeshow(Fig.5)thatcompositereservoirshavethelargestfluidproductionintensity,highestandconcentratedproducingdegree,andreservoirswithgoodandfairproducingdegreeaccountfor94.1%oftheirthickness;fracture-poretypedreservoirshavehigherproducingdegree,andreservoirswithgoodandfairproducingdegreeaccountfor83.3%oftheirthickness,butreservoirswithpoorandzeroproducingdegreealsoaccountfor16.7%;poretypedreservoirshavesimilarproportionofreservoirswithdifferentproducingdegrees,andreservoirswithgood,fair,poorandzeroproducingdegreeaccountfor25.7%,32.0%,12.9%and29.4%respectively;fracturetypedreservoirsareminor,andshowsnoproducingdegreeonliquidproductionprofile.Poretypedandfracture-poretypedreser-voirshavelargetotalthicknessandsignificantlywiderangeofproducingdegree.Forthepurposeofevaluatingtheproducingdegreeofreservoirsmoreaccurately,itisnecessarytosubdi-videreservoirflowunitbasedonreservoirtypes[7−8].3.2FlowunitdivisionandcharacterizationofdualmediareservoirsOnthebasisofcoreanalysisdata,thegeologicalparame-Fig.5Producingsituationofvarioustypesofreservoirsinthenorth=oilreservoirterscharacterizingreservoirflowfeaturesarecalculatedap-plyingwelllogginginterpretationparameters,then,takingproducingdegreeasdiscriminatingparameters,thegeologicalparametersaffectingreservoirproducingdegreeareselectedasclusteringvariables,finally,bytheapplicationofneuralnetworkclusteringtechnology,thereservoirflowunitofindi-vidualwellisdivided[9]anda3Dgeologicalmodelofcarbon-atereservoirflowunitsisbuiltinordertocharacterizethedistributionpatternofreservoirflowunits.3.2.1SelectionofclusteringvariablesWiththeaidofliquidproductionprofile,theproductionin-tensityofreservoirsiscalculatedinordertocharacterizetheproducingdegreeofreservoirs,then,variousgeologicalpa-rameterscalculatedfromwellloggingparametersareusedtodrawcrossplotinordertoanalyzetheirrelativitytoproduc-ingdegreeofreservoirs.Finally,fourgeologicalparametersforfracture-poretypedreservoirsareselectedasclusteringvariables:reservoirqualityindex(RQI),porethroatradiusatmercurysaturationof50%(R50),theratiooffractureperme-abilitytomatrixpermeability(Kf/Km)andtotalpermeability(Kf+Km).Forporetypedreservoirs,threegeologicalparame-tersareselected:RQI,R50andKm.Reservoirqualityindex(RQI)iscalculatedasfollow[1]:e=0.0314/RQIKφ(1)where,RQI—reservoirqualityindex,m;K—absoluteper-meability,mD;φe—effectiveporosity,f.TakingWinlandformula[2]asreference,therelationshipbetweenthroatradiusatmercurysaturationof50%andabso-lutepermeabilityandporosityisestablishedbasedonregres-sion,whosecorrelationcoefficientis0.951:lgR50=É0.0481+0.0143e+0.3951lgK(2)Forfracture-poretypedreservoirs,thecrossplot(Fig.6)betweenRQI,R50,Kf/Km,Kf+KmandproducingdegreeshowsthattheproducingdegreepositivelycorrelateswithRQI,R50andKf+KmandwithKf/Kmtosomeextent.Forporetypedreservoirs,thecrossplot(Fig.7)betweenRQI,R50,KmandproducingdegreeshowsthattheproducingdegreepositivelycorrelateswithRQI,R50andKm.Whetherforfracture-poredtypeorporetypedreservoirs,thedatapointsofdifferentpro-ducingdegreesarepartiallyoverlapping,asaresult,thepro-ducingdegreeofreservoirscan’tbeaccuratelycharacterizedbyonegeologicalparameterindependently,therefore,clusterFig.6Crossplotofdifferentgeologicalparametersandreservoirproducingdegreesoffracture-poretypedreservoirsFANZifeietal./PetroleumExplorationandDevelopment,2014,41(5):634–641−638−Fig.7Crossplotofdifferentgeologicalparametersandreservoirproducingdegreesofporetypedreservoirsanalysistechnologyisadoptedtodividereservoirflowunits,whichwillconsidermultiplefactors.3.2.2ClassificationandcharacterizationofflowunitsTheflowunitdivisionofoilreservoir=northfollowssuchprinciples:thecompositetypedreservoirswithandconcen-tratedproducingdegreeandthefracturetypedreservoirswithasmallamountaretakenasonetypeofflowuniteach,repre-sentedbylettersofAandD,respectively;forthefrac-ture-poretypedandporetypedreservoirswithsignificantlydifferentproducingdegreesandlargeproportionintotalres-ervoirthickness,withtheselectedgeologicalparametersasclusteringvariables,neuralnetworktechnologyareappliedtoperformclusteringanalysis,bywhichtheyaresubdividedintotwotypesofflowunitseach,representedbylettersofB1,B2andC1,C2,respectively.Basedontheequivalentpermeabilitygeologicalmodelingoffracture-poretypedcarbonatereservoirs,accordingtothereservoirflowunitdivisionresultsofindividualwells,the3Dgeologicalmodeloftheflowunitsofoilreservoir=northisbuiltadoptingsequentialindicatorsimulationmethodinordertocharacterizethedistributionofflowunits.Differentflowunitshavedifferentdistributionscopeandlocationlaterally(Fig.8)aswellasdifferentdistributionlocationandprobabil-ityvertically(Fig.9).4RemainingoildistributionbasedonflowunitsAfterreasonableupscalingofthegeologicalmodelof=north,differenttypesofreservoirflowunitsadoptdifferentrelativepermeabilitycurvestoconductmodelinitializationandreservesmatchinginordertoobtaintheoilsaturationfiledunderoriginalreservoircondition.Thenhistorymatch-ingisconductedinordertogetremainingoilsaturationfiledundercurrentreservoircondition.Finally,theremainingpro-duciblevolumecanbecalculatedbysubtractingresidualoilfromremaininggeologicalvolume,basedonwhich,varioustypesofmapscanbedrawninordertodescribetheremainingoildistributionregularityofeachsubzoneandea
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-S1876380414600756-main
链接地址:http://www.oilwenku.com/p-70403.html

当前资源信息

吾王的呆毛

编号: 20180607204119611430

类型: 共享资源

格式: PDF

大小: 4.11MB

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