DNV-RP-F105-2006 FREE SPANNING PIPELINES

RECOMMENDEDPRACTICEDNV-RP-F105FREESPANNINGPIPELINESFEBRUARY2006DETNORSKEVERITASFOREWORDDETNORSKEVERITAS(DNV)isanautonomousandindependentfoundationwiththeobjectivesofsafeguardinglife,prop-ertyandtheenvironment,atseaandonshore.DNVundertakesclassification,certification,andotherverificationandconsultancyservicesrelatingtoqualityofships,offshoreunitsandinstallations,andonshoreindustriesworldwide,andcarriesoutresearchinrelationtothesefunctions.DNVOffshoreCodesconsistofathreelevelhierarchyofdocuments:—OffshoreServiceSpecifications.ProvideprinciplesandproceduresofDNVclassification,certification,verificationandcon-sultancyservices.—OffshoreStandards.ProvidetechnicalprovisionsandacceptancecriteriaforgeneralusebytheoffshoreindustryaswellasthetechnicalbasisforDNVoffshoreservices.—RecommendedPractices.ProvideproventechnologyandsoundengineeringpracticeaswellasguidanceforthehigherlevelOffshoreServiceSpecificationsandOffshoreStandards.DNVOffshoreCodesareofferedwithinthefollowingareas:A)Qualification,QualityandSafetyMethodologyB)MaterialsTechnologyC)StructuresD)SystemsE)SpecialFacilitiesF)PipelinesandRisersG)AssetOperationH)MarineOperationsJ)WindTurbinesAmendmentsandCorrectionsThisdocumentisvaliduntilsupersededbyanewrevision.Minoramendmentsandcorrectionswillbepublishedinaseparatedocumentnormallyupdatedtwiceperyear(AprilandOctober).Foracompletelistingofthechanges,seethe“AmendmentsandCorrections”documentlocatedat:http://www.dnv.com/technologyservices/,“OffshoreRules&Standards”,“ViewingArea”.Theelectronicweb-versionsoftheDNVOffshoreCodeswillberegularlyupdatedtoincludetheseamendmentsandcorrections.Commentsmaybesentbye-mailtorules@dnv.comForsubscriptionordersorinformationaboutsubscriptionterms,pleaseusedistribution@dnv.comComprehensiveinformationaboutDNVservices,researchandpublicationscanbefoundathttp://www.dnv.com,orcanbeobtainedfromDNV,Veritas-veien1,NO-1322Høvik,Norway;Tel+4767579900,Fax+4767579911.©DetNorskeVeritas.Allrightsreserved.Nopartofthispublicationmaybereproducedortransmittedinanyformorbyanymeans,includingpho-tocopyingandrecording,withoutthepriorwrittenconsentofDetNorskeVeritas.ComputerTypesetting(FM+SGML)byDetNorskeVeritas.PrintedinNorwayIfanypersonsufferslossordamagewhichisprovedtohavebeencausedbyanynegligentactoromissionofDetNorskeVeritas,thenDetNorskeVeritasshallpaycompensationtosuchpersonforhisproveddirectlossordamage.However,thecompensationshallnotexceedanamountequaltotentimesthefeechargedfortheserviceinquestion,providedthatthemaximumcompen-sationshallneverexceedUSD2million.Inthisprovision"DetNorskeVeritas"shallmeantheFoundationDetNorskeVeritasaswellasallitssubsidiaries,directors,officers,employees,agentsandanyotheractingonbehalfofDetNorskeVeritas.RecommendedPracticeDNV-RP-F105,February2006Introduction–Page3RelationshiptoRulesunderpreparationPartsofthisRecommendedPracticeshouldbereadinconjunc-tionwiththefollowingtwodocuments:—DNV-RP-F110"GlobalBucklingofPipelines",and—DNV-RP-C205-"EnvironmentalConditionsandEnvironmentalLoads".However,thesetwodocumentsarecurrentlyunderpreparationatDNV.Readersarerequestedtocontact:RP-F110:LeifCollberg,TNCNO714,DetNorskeVeritas,Veritasveien1,N-1322Høvik,mailto:Leif.Collberg@dnv.comRP-C205:ArneNestegaard,TNCNO785,DetNorskeVeritas,Veritasveien1,N-1322Høvik,mailto:Arne.Nestegard@dnv.comformoreinformation,ifrequired.MotivesTheexistingdesigncode,DNV-RP-F105for"FreeSpanningPipelines"fromMarch2002,canbeeffectivelyappliedtodealwithbothsingleandmultiplespansvibratingpredominantlyinasinglemode.Inthecasewherethereisacombinationoflongspansandhighcurrents,notonlydoesthefundamentaleigenmodesbecomeactivated,butalsothehighermodes.Suchamultimoderesponseisnotprohibitedbythedesigncode.However,nodetailedguidanceisprovidedinthepresentRec-ommendedPractice(RP)aboutthefatiguedamagefrommul-timodevibrations.DuringtheOrmenLangeprojectastrongfocuswasputonde-signproceduresforlongfreespansinordertomakethisprojectfeasibleandinordertosaveseabedinterventioncosts.ArelativelylargeR&Dprojectonfreespanvortex-inducedvi-brations(VIV)wasperformed.TheresultsofthisprojectwassynthesisedintoaprojectspecificguidelineissuedbyDNV,concerningthecalculationproceduresanddesignacceptancecriteriaforlongfreespanwithmulti-moderesponse.Themainobjectiveofthisproposalistoincludetheexperienc-esgainedfromtheOrmenLangeProjectregardingmulti-modeandmulti-spanresponseandincorporatethelatestR&DworkrelatingtoFreeSpanningPipelines.ThisproposalhasbeenworkedoutthroughaJIPwithHydroandStatoil.Inadditiontothemulti-moderesponseaspects,ageneralupdateandrevi-sionoftheDNV-RP-F105hasbeenperformed,basedonfeed-backandexperiencefromseveralprojectsandusers.MainchangesThemostimportantchangesinthisupdateare:—Computationalprocedureformulti-modeanalysis,selec-tionofvibrationmodesandcombinationofstresses.—Guidanceonmitigationmeasuresforvortex-inducedvi-brations.—Ultimatelimitstatecriteria(over-stress)updated.—Recalibratedsafetyfactors.—Detailedcalculationprocedureforpipe-soilinteraction.—Practicalguidanceonvariousaspectsbasedonprojectex-periencefromseveralsources.—Updateofresponsemodelsforvortex-inducedvibrationsbasedonavailabletestresults.—Clarifiedlimitationsandrangeofapplication.—Addedmassmodificationduringcross-flowvibrations.ThisupdatestrengthensthepositionofthisRPasthestate-of-the-artdocumentforfreespanningpipelines.Itallowslongerspanstobeacceptedasdetailedcalculationproceduresformultiplemoderesponseisgiven.Acknowledgement:ThefollowingcompaniesaregratefullyacknowledgedfortheircontributionstothisRecommendedPractice:DHIHydroReinertsenStatoilSnamprogettiDNVisgratefulforthevaluableco-operationsanddiscussionswiththeindividualpersonnelofthesecompanies.Themultispanandmulti-modestudieswerefundedbytheOr-menLangelicensegroup.HydroanditslicensepartnersBP,Exxon,Petoro,ShellandStatoilareacknowledgedforthepermissiontousethesere-sults.DETNORSKEVERITASRecommendedPracticeDNV-RP-F105,February2006Page4–IntroductionDETNORSKEVERITAS,RecommendedPracticeDNV-RP-F105,February2006Page5CONTENTS1.GENERAL..............................................................71.1Introduction.............................................................71.2Objective...................................................................71.3Scopeandapplication.............................................71.4ExtendingtheapplicationscopeofthisRP..........81.5Safetyphilosophy.....................................................81.6Freespanmorphologicalclassification..................81.7Freespanresponseclassification...........................91.8Freespanresponsebehaviour................................91.9Flowregimes..........................................................101.10VIVassessmentmethodologies.............................101.11RelationshiptootherRules..................................101.12Definitions..............................................................111.13Abbreviations.........................................................111.14Symbols...................................................................112.DESIGNCRITERIA............................................132.1General...................................................................132.2Non-stationarityofspans......................................142.3Screeningfatiguecriteria......................................152.4Fatiguecriterion....................................................152.5ULScriterion.........................................................162.6Safetyfactors..........................................................183.ENVIRONMENTALCONDITIONS.................193.1General...................................................................193.2Currentconditions................................................193.3Short-termwaveconditions..................................203.4Reductionfunctions...............................................213.5Long-termenvironmentalmodelling...................223.6Returnperiodvalues.............................................224.RESPONSEMODELS.........................................234.1General...................................................................234.2Marginalfatiguelifecapacity...............................234.3In-lineresponsemodel.........................................244.4Cross-flowresponsemodel...................................254.5Addedmasscoefficientmodel..............................275.FORCEMODEL..................................................275.1General...................................................................275.2FDsolutionforin-linedirection..........................275.3Simplifiedfatigueassessment...............................295.4Forcecoefficients...................................................296.STRUCTURALANALYSIS...............................306.1General...................................................................306.2Structuralmodelling.............................................306.3Functionalloads....................................................316.4Staticanalysis........................................................316.5Eigenvalueanalyses.............................................326.6FEMbasedresponsequantities...........................326.7Approximateresponsequantities........................326.8Approximateresponsequantitiesforhigherordermodesofisolatedsinglespans...................346.9Addedmass............................................................347.PIPE-SOILINTERACTION..............................347.1General...................................................................347.2Modellingofpipe-soilinteraction.......................357.3Soildamping..........................................................357.4Penetrationandsoilstiffness................................357.5Artificialsupports.................................................388.REFERENCES.....................................................38APP.AMULTI-MODERESPONSE.............................39APP.BVIVMITIGATION............................................42APP.CVIVINOTHEROFFSHOREAPPLICATIONS...............................................................43APP.DDETAILEDASSESSMENTOFPIPE-SOILINTERACTION.................................................................44DETNORSKEVERITASRecommendedPracticeDNV-RP-F105,February2006,Page6DETNORSKEVERITASRecommendedPracticeDNV-RP-F105,February2006Page71.General1.1Introduction1.1.1Thepresentdocumentconsidersfreespanningpipelinessubjectedtocombinedwaveandcurrentloading.ThepremisesforthedocumentarebasedontechnicaldevelopmentwithinpipelinefreespantechnologyinrecentR&Dprojects,aswellasdesignexperiencefromrecentandongoingprojects,i.e.:—DNVGuideline14,seeMørk&Fyrileiv(1998).—ThesectionsregardingGeotechnicalConditionsandpartofthehydrodynamicmodelarebasedontheresearchper-formedintheGUDESPproject,seeTuraetal.(1994).—ThesectionsregardingFreeSpanAnalysisandin-lineVortexInducedVibrations(VIV)fatigueanalysesarebasedonthepublishedresultsfromtheMULTISPANproject,seeMørketal.(1997).—NumericalstudybasedonCFDsimulationsforvibrationsofapipelineinthevicinityofatrench,performedbySta-toil,DHI&DNV,seeHansenetal.(2001).—Further,recentR&Danddesignexperiencee.g.fromÅs-gardTransport,ZEEPIPE,TOGIandTROLLOILpipe-lineprojectsareimplemented,seeFyrileivetal.(2005).—OrmenLangetestsaimedatmoderateandverylongspanswithmultimodalbehaviour,seeFyrileivetal.(2004),Chezhianetal.(2003)andMørketal.(2003).Thebasicprinciplesappliedinthisdocumentareinagreementwithmostrecognisedrulesandreflectstate-of-the-artindustrypracticeandlatestresearch.Thisdocumentincludesabriefintroductiontothebasichydro-dynamicphenomena,principlesandparameters.Forathor-oughintroductionseee.g.Sumer&Fredsøe(1997)andBlevins(1994).1.2Objective1.2.1Theobjectiveofthisdocumentistoproviderationalde-signcriteriaandguidanceforassessmentofpipelinefreespanssubjectedtocombinedwaveandcurrentloading.1.3Scopeandapplication1.3.1DetaileddesigncriteriaarespecifiedforUltimateLimitState(ULS)andFatigueLimitState(FLS)duetoin-lineandcross-flowVortexInducedVibrations(VIV)anddirectwaveloading.Thefollowingtopicsareconsidered:—methodologiesforfreespananalysis—requirementsforstructuralmodelling—geotechnicalconditions—environmentalconditionsandloads—requirementsforfatigueanalysis—VIVresponseanddirectwaveforceanalysismodels—acceptancecriteria.1.3.2Freespanscanbecausedby:—seabedunevenness—changeofseabedtopology(e.g.scouring,sandwaves)—artificialsupports/rockbeamsetc.—strudelscours.1.3.3Thefollowingenvironmentalflowconditionsarede-scribedinthisdocument:—steadyflowduetocurrent—oscillatoryflowduetowaves—combinedflowduetocurrentandwaves.1.3.4ThisRecommendedPracticeisstrictlyspeakingonlyapplicableforcircularpipecross-sectionofsteelpipelines.However,itcanbeappliedwithcaretonon-circularcross-sec-tionssuchaspiggy-backsolutionsaslongasotherhydrody-namicloadingphenomena,e.g.galloping,areproperlytakenintoaccount.Basicprinciplesmayalsobeappliedtomorecomplexcrosssectionssuchaspipe-in-pipe,bundles,flexiblepipesandum-bilicals.1.3.5TherearenolimitationstospanlengthandspangapwithrespecttoapplicationofthisRecommendedPractice.Bothsinglespansandmultiplespansscenarios,eitherinsinglemodeormultiplemodevibration,canbeassessedusingthisRP.Unlessotherwisedocumented,thedamagecontributionfordifferentmodesshouldrelatetothesamecritical(weld)loca-tionalongthespan.1.3.6Thefreespananalysismaybebasedonapproximatere-sponseexpressionsorarefinedFEapproachdependingonthefreespanclassificationandresponsetype,seeSec.6.Thefollowingcasesareconsidered:—singlespans—spansinteractingwithadjacent/sidespans.ThestressrangesandnaturalfrequenciesshouldnormallybeobtainedfromanFE-approach.RequirementstothestructuralmodellingandfreespananalysesaregiveninSec.6.1.3.7Thefollowingmodelsareconsidered:—responsemodels(RM)—forcemodels(FM).Anamplituderesponsemodelisapplicablewhenthevibrationofthefreespanisdominatedbyvortexinducedresonancephe-nomena.Aforcemodel(Morison’sequation)maybeusedwhenthefreespandominatedbyhydrodynamicloadssuchasdirectwaveloads.Theselectionofanappropriatemodelmaybebasedontheprevailingflowregimes,see1.9.1.3.8Thefatiguecriterionislimitedtostresscycleswithintheelasticrange.Lowcyclefatigueduetoelasto-plasticbehaviourisconsideredoutsidethescopeofthisdocument.1.3.9Fatigueloadsduetotrawlinteraction,cyclicloadsdur-inginstallationorpressurevariationsarenotconsideredhereinbutmustbeconsideredasapartoftheintegratedfatiguedam-ageassessment.1.3.10Proceduresandcriteriaforstructuraldesignorassess-mentoffreespanningHT/HPpipelineshavenotbeenincludedwithinthescopeofthisRecommendedPractice.Freespansduetoupliftarehoweverwithinthescopeofthisdocument.Note:Forinformationconcerningareferencedocument(currentlyinpreparation),regardingproceduresandcriteriaforstructuraldesignorassessmentofHP/HTpipelines,contactTNCNO714,DNV,Høvik,Norway.1.3.11Themainaspectsofafreespanassessmenttogetherwithkeyparametersandmainresultsareillustratedinthefig-DETNORSKEVERITASTheflowregimesarediscussedin1.9.urebelow.RecommendedPracticeDNV-RP-F105,February2006Page8Figure1-1Overviewofmaincomponentsinafreespanassessment1.4ExtendingtheapplicationscopeofthisRP1.4.1TheprimaryfocusofthisRPisonfreespanningsubseapipelines.1.4.2ThefundamentalprinciplesgiveninthisRPmayalsobeappliedandextendedtootheroffshoreelementssuchascylin-dricalstructuralelementsofthejackets,risersfromfixedplat-formsetc.,atthedesigner’sdiscretion.HoweversomelimitationsapplyandthesearediscussedinAppendixC,C.2.1.4.3ForamoredetailedaccountofriserVIV,referenceismadetotheDNV-RP-F204“RiserFatigue”.1.5Safetyphilosophy1.5.1ThesafetyphilosophyadoptedhereincomplieswithSec.2inDNV-OS-F101.1.5.2ThereliabilityofthepipelineagainstfailureisensuredbyuseofaLoadandResistanceFactorsDesignFormat(LR-FD).—Forthein-lineandcross-flowVIVacceptancecriterion,thesetofsafetyfactorsiscalibratedtoacceptabletargetreliabilitylevelsusingreliability-basedmethods.—Forallotheracceptancecriteria,therecommendedsafetyfactorsarebasedonengineeringjudgementinordertoob-—Useofcasespecificsafetyfactorsbasedonquantificationofuncertaintyinfatiguedamage,canalsobeconsidered.1.6Freespanmorphologicalclassification1.6.1Theobjectiveofthemorphologicalclassificationistodefinewhetherthefreespanisisolatedorinteracting.Themorphologicalclassificationdeterminesthedegreeofcom-plexityrequiredofthefreespananalysis:—Twoormoreconsecutivefreespansareconsideredtobeisolated(i.e.singlespan)ifthestaticanddynamicbehav-iourisunaffectedbyneighbouringspans.—Asequenceoffreespansisinteracting(i.e.multi-span-ning)ifthestaticanddynamicbehaviourisaffectedbythepresenceofneighbouringspans.Ifthefreespanisinteract-ing,morethanonespanmustbeincludedinthepipe/sea-bedmodel.1.6.2Theclassificationmaybeusefulinthefollowingcases:—forevaluatingmulti-moderesponseofsingleandmulti-spans—forevaluationofscourinducedfreespansandinapplica-tionofapproximateresponsequantitiesdescribedinSec.6.7.1.6.3ThemorphologicalclassificationshouldingeneralbeEnvironmentaldescriptionch.3ProjectdataCurrentstatisticsCurrentprofileWavestatisticsWavespectrumDirectionalityTurbulenceWave&currentReturnperiodvaluesWave&currentlong-termdescriptionScreeningFatigueULSWave&currentReturnperiodvaluesStructuralresponsech.6&7PipedataSeabed&pipeprofileSoildataLaytensionOperationalconditionsNaturalFrequencyNaturalFrequencyStressamplitudeNaturalFrequencyStressrangesStaticBendingResponsemodelForcemodelch.4&5VRKCαDampingFreespanparametersStressrangesNoofcyclesExtremestressesAcceptancecriteriach.2SafetyclassSafetyfactorsSNcurveOK/notOK(spanlength)OK/notOK(fatiguelife)OK/notOK(localbuckling)MainComponentsKeyParametersDesignCriteria&MainresultsDETNORSKEVERITAStainasafetylevelequivalenttomodernindustrypractice.determinedbasedondetailedstaticanddynamicanalyses.RecommendedPracticeDNV-RP-F105,February2006Page9Guidancenote:WhenlongsegmentsofthepipelineareanalysedinautomatedFEanalysistools,certainlimitationscanariseforidentifyingin-teractingspans.Thefollowingexampleillustratesthesame:Twofreespanswhichareseparatedbyadistanceof1000mandeachwithaspanlengthofabout50m.TheFEanalysisestimatesthatsomeoftheresponsefrequenciesforthesetwospansareidentical.Further,duetonumericalapproximationsorduetoroundofferrors,theresultsmaybepresentedassinglemodere-sponseatthesetwospans,i.e.itisasingleinteractingmodeatthesetwospans.However,inrealitythesetwospansarephysicallyseparatedbyaconsiderabledistance,andnotinteracting.Ifisolatedspansareincorrectlymodelledasinteractingmulti-spans,itmayalsoleadtosignificanterrorsinestimatingthefa-tiguedamage.Thefatiguedamageisdependentontheunitstressamplitudes,asdiscussedinSec.4.Theunitstressamplitudesaredependentonthenormalisedmodeshape,whichisrelatedtothespanlengthoverwhichthenormalisationisconsidered.Whenlongmultispansareanalysed,thenormalisationwillnotbesameasforanisolatedsinglespan,withinthemulti-spanningsystem.Thiswillinturnleadtoerrors.Experiencehasshownthatincaseofclosefrequenciesforspans,theFEanalysismaypredictinteractioneventhoughthephysicaldistancebetweenthespansisquitelong.Incaseofmodeshapeswithdeflectioninspansthatseemtobephysicallywellseparat-ed,useofappropriateaxialpipe-soilstiffnessand/orlocalre-straintsinbetweenthespansshouldbeconsideredtoseparateindividualmodes.Hence,cautionshouldbeexercisedwhenusingautomatedFEtoolsforidentifyinginteractingspans.---e-n-d---of---G-u-i-d-a-n-c-e---n-o-t-e---1.6.4Inlieuofdetaileddata,Figure1-2maybeusedtoclas-sifyspansintoisolatedorinteractingdependingonsoiltypesandspan/spansupportlengths.Figure1-2isprovidedonlyforindicativepurposesandisapplicableonlyforhorizontalsup-ports.Thecurvesare,instrictterms,onlyvalidforthevertical(cross-flow)dynamicresponsebutmayalsobeusedforas-sessmentofthehorizontal(in-line)response.NotethatF