• / 10
  • 下载费用:5 下载币  

1-s2.0-S1876380416300799-main

关 键 词:
s2 S1876380416300799 main
资源描述:
RESEARCHPAPERPETROLEUMEXPLORATIONANDDEVELOPMENTVolume43,Issue4,August2016OnlineEnglisheditionoftheChineselanguagejournalCitethisarticleas:PETROL.EXPLOR.DEVELOP.,2016,43(4):679–688.Receiveddate:19Oct.2015;Reviseddate:30May2016.*Correspondingauthor.E-mail:jich.0704@163.comFoundationitem:SupportedbyChinaNationalScienceandTechnologyMajorProject(2009ZX05009,2011ZX05001);theNationalNaturalScienceFoundationofChina(51104173,51274229).Copyright©2016,ResearchInstituteofPetroleumExplorationandDevelopment,PetroChina.PublishedbyElsevierBV.Allrightsreserved.Multistageinterwellchemicaltracingforstep-by-stepprofilecontrolofwaterchannelingandfloodingoffracturedultra-lowpermeabilityreservoirsPUChunsheng1,JINGCheng1,*,HEYanlong1,GUXiaoyu1,ZHANGZhiying2,WEIJikai31.SchoolofPetroleumEngineering,ChinaUniversityofPetroleum,Qingdao266580,China;2.NortheastBureauofSinopec,Changchun130000,China;3.TuhaOilfieldEngineeringTechnologyResearchInstitute,Hami838202,ChinaAbstract:Tomonitorthedynamicvariationsofsweepareaandformationparametersduringtheprocessofmultipleslugsstep-by-stepprofilecontrol,themultistageinterwellchemicaltracingtechniquewasproposedandtestedinfield,inlinewiththefeaturesofthestep-by-stepprofilecontrolinfracturedultra-lowpermeabilityreservoirsandthebasicprincipleofchemicaltracingtest.Accordingtothedesignschemeofstep-by-stepprofilecontrolandthecharacteristicsofwaterchannelingandfloodinginfracturedultra-lowpermeabilityreservoirs,thisstudyworkedouttimes-designmethodforinterwelltracing,optimizedtheselectionprincipleofchemicaltracerandcalculationformulaoftracerdosage,andsetuptheparameteroptimizationforecastingmethodofstep-by-stepprofilecontrolbasedonmultistageinterwelltrac-ing.Theapplicationresultsofthemethodshowmultistageinterwellchemicaltracingcanreflectdynamicvariationoffracturedparametersef-fectively,andthemonitoringresultsmatchwiththedynamicproductiontestingresults,demonstratinggoodadaptabilityofthemethod.Keywords:fracturedreservoir;ultra-lowpermeabilityreservoir;step-by-stepprofilecontrol;multistageinterwelltracing;dynamicmonitoringIntroductionInfracturedultra-lowpermeabilityreservoirsfeaturingtightmatrix,complexandvariablefracturesystemsandstrongheterogeneity,waterchannelingpathwaysarelikelytooccurduringwaterflooding,causingseverewaterchannelingandfloodingofthereservoir[14].Atpresent,multipleslugstep-by-stepprofilecontrolisamajortreatmenttotacklese-riouswaterchannelingandfloodinginthiskindofreser-voir[56],andtherealtimedynamicmonitoringistakentoensuretheeffectivenessofit.Asoneimportantmeansformonitoringthedynamicsbetweendifferentwells,acompletetheorysystemonchemicalinter-welltracermonitoringhasbeenestablished,andgraduallyacceptedinfieldpractices[717].Previously,thereservoirwastreatedasaseriesofnon-com-municatinglayersindifferentkindsofinterpretationmethods(analyticalmethod,numericalmethodandsemi-analyticalmethod)[1117],butthebasichypothesisoftheabovemethodsisnotsuitablefortheinterpretationoffracturedultra-lowpermeabilityreservoir[23].Meanwhile,thesweepingrangeandparametersoftheformationduringtheprocessofstep-by-stepprofilecontroldynamicallychangewiththetime,thetraditionalinter-welltracermonitoring,lackingcontinuityanddynamicity,can'teffectivelymonitorthedynamicvaria-tionduringthestep-by-stepprofilecontrol[710].Multistageinter-welltracermonitoringforthestep-by-stepprofilecontrolofwaterchannelingandfloodingofultra-lowpermeabilityreservoirhasbeenproposedinthisstudy,whichallowsreal-timemonitoringofpluggingsituation,fractureparameters,anddynamicsweepingofinjectionwaterofvari-ouslevelsofchannelingpathways,optimizationandpredic-tionofcontrolparameters,andimprovementofeffectivenessofstep-by-stepprofilecontrolatlast.Themultistagein-ter-welltracermonitoringforthestep-by-stepprofilecontrolofwaterchannelingandfloodingofultra-lowpermeabilityreservoircanprovideanimportantbasisforparameteropti-mizationandeffectevaluationandpredictionofthestep-by-stepprofilecontrol.1.DesignoftracertesttimesMultistageinter-wellchemicaltracertestforstep-by-stepPUChunshengetal./PetroleumExplorationandDevelopment,2016,43(4):679–688680profilecontrolisthepreconditiontoensuresmoothrunningoftracermonitoringandstepbystepprofilecontrol[18].Thede-signoftracermonitoringtimesshouldnotonlyconsiderthecycleoftracermonitoringandstepbystepprofilecontrol,butalsoavoidtheinfluenceofinjectedgelslugsontracermoni-toringresults.Basedontheinjectionschemeofstep-by-stepprofilecon-trol,theinjectionsequenceofgelslugsandtracerslugsandtracermonitoringtimesmustbeworkedout.Whatthetracermonitoredisthedistributionofinjectedfluidundercertainpressure,buttheinjectionofgelslugwouldcausepressurefluctuationsinevitably.Consideringthecrosslinkingtimeofgelandretardanceofpressurepropagation,theinjectionofgelslugwillhavealittleinfluenceontracermonitoringresults,ifagelslugisinjectedwhenthetraceroflasttimereachedthepeakconcentration.AssumingthenumberofinjectedgelslugisEandthetimesofinjectedtracerisF,inordertoensureallworkscanbecompletedwithinthetimeT,thesumofinjec-tiontimeofallslug,gellingtimeandtracermonitoringtimeshouldbeequalorlessthantimeT,then:p11EFfeefefVVatTQQ(1)whereppmaxfjitt(2)Assumingtheproductionwelljhasifracturebeltsandeachbelthasmultiplefractures.ThefracturebeltiisequivalenttoaflowingtubeclusterconsistingofnjitubesinDjidiameter,then:442111p4221π4π4jjjNNMjijijijijiijijijijiNjijijijijeijinDnDssstnDQDDfVs(3)where1=Mjjjjfmm(4)2.TracerselectionprincipleandcalculationoftracerdosageSelectingdifferenttracersfordifferenttimesandcalculat-ingtracerdosageateverystagearethemostimportantpartofschemedesignofthemultistageinter-wellchemicaltracermonitoringforstep-by-stepprofilecontrolofwaterchannel-ingandfloodingoffracturedultra-lowpermeabilityreser-voir[1819].2.1.EvaluationandscreeningprincipleofmultistageinterwellchemicaltracerInthescreeningoftraditionaltracersusedinone-timemonitoring,onlythephysicochemicalpropertiesofthetracer,reservoirrockproperties,thepropertiesofformationwaterandinjectedwater,economy,securityandenvironmentalpro-tectionetcneedtobeconsidered,whileinpickingtracersformulti-stageinterwellmonitoring,besidesallthefactorsmen-tionedabove,theinteractionbetweendifferenttimesoftracersandphysicochemicalpropertiesofthetracerandinjectedgelalsoneedtobeconsidered.Toeffectivelyavoidtheinterfer-enceofdifferenttimesoftracers,atleasttwotypesoftracersshouldbechosen,andthemethodofmultistagealternativeinjectionshouldbeadopted.Inadditiontotheevaluationofheatresistance,compatibilitybetweentracerandinjectedwa-ter,compatibilitybetweentracerandformationwater,rockadsorptionoftracerandreservoirdamage,multistagetracerscreeningmustevaluatethecompatibilitybetweenthetracersandgeladsorptionofthetracers[18].Theselectedmultistagechemicaltracersshouldhavethebiggestmaximumabsorptionwavelengthdifferenceandlowergeladsorption.2.2.CalculationofmultistageinterwelltracerdosageBasedonthecharacteristicsofwaterfloodinginfracturedultra-lowpermeabilityreservoirs,theinjectedwatermainlyflowalongthefracturesystemswiththelowestresistance,andtracerininjectionwatercanmonitorthefracturesystems.Therefore,undertheconditionofBrigham-Smithwaterflooding[11,19],theformulaoffive-pointpatternchemicaltracerdosageisadaptedtothecalculationformulaofinterwellchemicaltracerdosagethatcanbeusedinfracturedultra-lowpermeabilityreservoirs:1.735owsTfsmaxows6wows=4.44102+1fdhLffmCBa(5)Theformulaaboveisapplicabletocalculatetracerdosageoffive-pointwellpattern.Forothertypesofwellpattern,onlytheconversioncoefficientofmaximumrecoveryconcentra-tionbetweengroundandreservoirneedstobechanged[18].Watercutcanbecalculatedfromdynamicproductiondataofeachproductionwellbyformula(5).Duetothecomplexityoffractureporositycalculation,thefractureporosityofthefirststagetracermonitoringbeforestep-by-stepprofilecon-trolisusuallycalculatedmainlyaccordingtothefractureden-sityandfractureparametersintheresearcharea,thuscalcula-tionformulaofthefirststagetracerdosagecanbeexpressedasfollows:ows1ows6ffwows=4.442+110lwhHfhVffmB1.735smaxdLCa(6)Fracturesofvariouslevelsarepluggedstepbystep,frac-tureporosityandwatercutdecreasegraduallywiththeongo-ingofstep-by-stepprofilecontrol.Basedonthelasttracerinterpretationresults,thefracturevolumecanbeworkedout,andthenthecorrespondingfractureporositycanbecalculated.Therefore,tracerdosagecalculationformulasofdifferentinterwellchemicaltracerinjectiontimescanbededuced:ows1owwwsf6os10=4.442+1kefeVVVBfhffmPUChunshengetal./PetroleumExplorationandDevelopment,2016,43(4):679–6886811.735smaxdLCa(f=2,3,…,F)(7)3.Multistagetracerclassificationinterpretationmodelforstep-by-stepprofilecontrolComprehensiveinterpretationoftracercurvesisthecoreofmultistageinterwellchemicaltracermonitoringforstep-by-stepprofilecontrolofwaterchannelingandfloodingoffrac-turedultra-lowpermeabilityreservoir.Theanalysisresultsofmanytracerproductioncurvesofwaterchannelingandfloodingwellsinfracturedultra-lowpermeabilityreservoirsshowthatthewaterchannelingfracturesystemsmainlyin-cludechannelingtype,differential-interactivetypeandrela-tiveuniformpropulsiontype,whichcorrespondtotracercurvesofsinglepeak,multi-peakandwideplatformshapes,respectively(Table1).Consideringthecharacteristicsofdif-ferentwaterchannelingfracturesystems,fracturebeltdistri-butionphysicalmodelsof1/4injection-productionpattern(Table1)havebeenestablishedbasedonthefollowingas-sumptions:(1)injectedwaterisincompressiblefluidflowingcontinuously;(2)thetracerissimilartowaterintheprocessofinjection,therefore,theanalysisofthetracerflowisequivalenttothatofwaterflow;(3)thecomplexwaterchan-nelingfracturesystemofthefracturedultra-lowpermeabilityreservoirisequivalenttoaseriesofmutuallydisjointfracturebeltsbetweeninjectionwellsandproductionwells,eachfrac-turebeltisequivalenttoaflowingtubeclusterwithacertainnumberoftubes,sotheflowinfracturesisequaltotheflowintubes;(4)ignoringfluidgravityandcapillaryforce,theinjectionwaterandtracerinthetubedriveinpistonmatterwithamobilityratioof1;(5)thefluidflowinthetubeac-cordswithHagen-Poiseuilleequation;(6)theinfluenceoftracerlatitudinaldispersionandmoleculardiffusionarenegli-gible,andthetracerneveradsorbsontherocksurface;(7)theimbibitionbetweenmatrixandfractureisignored,andthefluidonlyflowsinthefracture.Byusingequivalenttubemethod,fracturebeltsareequiva-lenttotubesconformingtoHagen-Poiseuilleequation[1114],bysuperposingtracerconcentrationofdifferentfracturebeltsinproductionwells,themacroscopicproductioninformation(fj,QandVd)andmicroscopicfractureinformation(Nj,nji,Djiandsji,etc)arecombinedtogetthetracerclassificationinter-pretationmodelforstep-by-stepprofilecontrolofwaterchannelingandfloodingoffracturedultra-lowpermeabilityreservoir(Table1).Thefittingprocessoftracerproductioncurvecanbecon-sideredasnonlinearoptimization:*21min()rqqqCC(8)Atpresent,optimizationalgorithmscommonlyusedincludegeneticalgorithm,differentialevolution,antcolonyalgorithm,simulatedannealingmethod.Inthisstudy,geneticalgorithmwasadoptedtoconductautomaticfittingoftheinterwelltrac-erproductioncurves.Basedondynamicproductiondataofwellgroupsandtrac-ertestdata,thetracerclassifiedinterpretationmodelsforstep-by-stepprofilecontrolofwaterchannelingandfloodingoffracturedultra-lowpermeabilityreservoirhavebeenusedtomakethetracerinterpretationsoftwareforstep-by-stepprofilecontrolofwaterchannelingandfloodingoffracturedultra-lowpermeabilityreservoirbasedonthegeneticalgo-rithm,thespecificprocessoftracerinterpretationisasfollows:(1)Accordingtotheinputdynamicproductiondataofeachproducingwellandtracermonitoringdata,distributioncoeffi-cientfjofinjectedwaterofdifferentproductionwellsarecalculated;(2)thepredictionoperatorofthetracercurveshapeisrun,inwhichtheinterpretationmoduleischosen(singlepeakmodule1,multi-peakmodule2andwideTable1.Classifiedtracerinterpretationmodelsforstep-by-stepprofilecontrolofwaterchannelingandfloodinginfracturedultra-lowpermeabilityreservoirsModeltypeModelequivalentSchematicdiagramofthephysicalmodelandtheshapeoftracerproductioncurvesMathematicalmodelChannelingtype(typeI)Equivalenttoonefracturebelt22666d2220π1041010exp16ππjjjjnDsfQtfVCCnDnDfQtDifferential-interactivetype(typeII)EquivalenttoNfracturebeltswithlargeseepagedifferences24212651d44412210111π10410exp1610ππNiiijiNijiiiNNNiiiiiiiiiiijiijiinDsfQDtfVsnDCnDCsstnDnDfQsDtfQsssRelativeuniformpro-pulsiontype(typeIII)EquivalenttoNfracturebeltswithsmallseep-agediffer-ences266661d120111π1041010exp16ππNiijNijiNNNiiiiijiijiiiiDRnfQtfVnCCDtDfQntnfQnPUChunshengetal./PetroleumExplorationandDevelopment,2016,43(4):679–688682platformmodule3)accordingtotheshapeoftracercurvepeak;(3)theinitialvalueofthefittingparametersarerandomgenerated,thentheinitialvalueisfittedintocorrespondingtracerinterpretationmodeltogetthetracerconcentrationcurve,thefittingerrorbetweentheobtainedcurveandmeas-uredcurveisworkedout,ifthefittingerrormeetsthere-quirement,thentheinformationofthefractureparameterisoutput,otherwise,anewvalueisgeneratedbyusinggeneticalgorithm,andthefittingerrorisanalyzedagainuntilthefit-tingerrormeetstheerrorrequirement.4.Parameteroptimizationandforecastmethodofstep-by-stepprofilecontrolAccordingtothetestresultsoftracermonitoringforstep-by-stepprofilecontrol,andthedynamicandstaticpro-ductiondata,thevariationofpressureaftersluginjectionispredicted,andtheinjectionparametersareoptimizedandadjustedintimetoensuretheeffectofthestep-by-stepprofilecontrol[20].4.1.Optimizationoftheinjectionpressureofdisplacementagentforstep-by-stepprofilecontrolDuringtheprocessofstep-by-stepprofilecontrol,inordertomakethefrontofthedisplacementagentpluggetintothedeepformation,ononehand,thedelaycrosslinkingtimeshouldbecontrolled,andontheotherhand,theinjectionpressureshouldbehighenough.Forfracturedultra-lowper-meabilityreservoirs,iftheinjectionpressureistoohigh,morenaturalfracturesmaybeopened,formingnewwaterchannels.Therefore,accordingtotheexperimentdata,thecontrollabledelaycrosslinkingtimerange[68],suitableforstep-by-stepprofilecontrolsystemoftheobjectivereservoir,canbeob-tained.Therealizableupperandlowerlimitofthecontrolla-bledelay-crosslinkingtimecanbechosen,andthepressurerangeoftheprofilecontroldisplacementagentisderivedasformula(9):zJ1=1(=)jNjijijikiMjnAvptV(9)where2π4=jijiDA2zzw)()2=(3jijjijiDppvpszwdwz=+jjppghghMeanwhile,consideringproductionwellsaregenerallyputintoproductionafterfracturing,andlargescaleartificialfrac-turesexistaroundtheproductionwells,inordertoavoidtoomuchprofilecontroldisplacementagentbeingrecovered,thefrontofprofilecontroldisplacementagentslugshouldbe2/3ofinjector-producerspaceawayfromtheinjectionwellborehole,thatisitshallsatisfy:zJ23jijisvpt(10)Intheend,theoptimalinjectionpressureofprofilecontroldisplacementagentindifferentstagesduringstep-by-stepprofilecontrolcanbeobtainedaccordingtoformulas(9)and(10).4.2.Injectionpressurepredictionaftercross-linkingAftercross-linkingoftheprofilecontroldisplacementagentinthefracturebelts,theinjectionpressureofnormal
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-S1876380416300799-main
链接地址:http://www.oilwenku.com/p-70327.html

当前资源信息

吾王的呆毛

编号: 20180607204859360115

类型: 共享资源

格式: PDF

大小: 1.10MB

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