Buckling Oil and Gas

Faculty of Science and Technology
MASTERS THESIS
Study program/ Specialization:

Of f shor e Technol ogy/
Subsea Technol ogy

Spring semester, 2013

Open/ Restricted access

Writer:
Obel e I f enna I saac

(Writers signature)
Faculty supervisor:
Dr . Dani el Kar unakar an ( Adj unct Pr of essor )
( Uni ver si t y of St avanger , Subsea 7 Nor way)

External supervisor(s):
Dr . Dashar at ha Achani ( Subsea 7 Nor way)

Title of thesis:

Lat er al Buckl i ng and Axi al Wal ki ng of Sur f ace Lai d Subsea Pi pel i ne

Credits (ECTS):
30
Key wor ds:
I ni t i al i mper f ect i on, Lay r adi us, wal ki ng,
l at er al buckl i ng, Out - of - st r ai ght ness
( OOS) , pi pe- soi l i nt er act i on, end
expansi on, ef f ect i ve axi al f or ce, Feed- i n,
Vi r t ual anchor , snake- l ay, t her mal
gr adi ent , pi pel i ne.

Pages: xviii +115

+attachment/other: 82

Stavanger, J une 15, 2013
Date/year

Master Thesis: Lateral Buckling and Axial Walking of Surface Laid Subsea Pipeline
0bele Ifenna Isaac 0niveisity of Stavangei Page iii
ABSTRACT
Subsea pipelines are increasingly being required to operate at high temperature and pressure
HT/HP. The pipeline installed on the seabed and left exposed have a potential to buckle, walk
and change configuration under high temperature and pressure (HT/HP). This could lead to
failure of the Pipeline if buckling and walking is not properly controlled or mitigated.
The objective of the thesis work is to study and understand the influence of pipeline-soil
interaction on the design of surface laid subsea pipeline susceptible to lateral buckling and
pipeline walking.
The main focus of the thesis work is on the use of snake-lay configuration as a mitigating
measure under controlled buckling design and rock dumping if needed to limit feed-in into
buckle and end expansions; the effect of thermal gradient on axial walking and the use of direct
electric heating (DEH) to reduce rate of walking.
The snake-lay configuration is achieved by installing deliberate horizontal lay imperfection to
trigger a sufficient number of thermal buckles at a pre-determined location along the pipeline.
The desire is to limit pipeline expansion at the connecting ends by using snake-lay design with
intermittent rock dumping.
The work includes performing non-linear finite element analysis (FEA) and modeling the soil-
pipeline interaction of as-laid pipeline using general finite (FE) element software ANSYS. The
results are discussed against the relevant design criteria based on design codes DNV-OS-F101,
DNV-RP-F110 and Subsea 7 Lateral Buckling Analysis Design Guideline.
FE analyses were performed to study the lateral buckling of a 2km VAS model with an initial
out-of-straightness (OOS) under operating temperature and pressure.
Consequently, the effect of thermal gradient of an asymmetric heating process in pipeline
walking phenomenon is investigated. Based on FE analyses findings, the use of direct electric
heating (DEH) system to reduce the rate of axial walking is proposed and explained.

Keywords: Initial imperfection, Lay radius, walking, lateral buckling, Out-of-straightness
(OOS), pipe-soil interaction, end expansion, effective axial force, Feed-in, Virtual
anchor, snake-lay, thermal gradient, pipeline, direct electric heating (DEH).
0bele Ifenna Isaac 0niveisity of Stavangei Page iv

0bele Ifenna Isaac 0niveisity of Stavangei Page v
ACKNOWLEDGEMENT
This thesis was carried out to fulfill the requirement for the award of Master of Science degree in
Offshore Technology at the Department of Mechanical and Structural Engineering and Materials
Science, Faculty of Science and Technology, University of Stavanger, Norway.

The thesis work was carried out in the premises of the world class company, Subsea 7 Norway,
during spring academic year 2013.

My gratitude goes to Subsea 7, Stavanger, Norway for providing me an office space, computer
system, full support and access to ANSYS finite element software and other Software to
facilitate successful execution of this task.

My sincere appreciation goes to my faculty supervisor, Dr. Daniel Karunakaran (Adjunct
Professor), whose guidance, encouragement and wealth of knowledge made every bit of time I
spent on this work extremely meaningful.
I would especially like to thank my day to day external supervisor, Dr. Dasharatha Achani from
Subsea 7, for his guidance, time for sharing knowledge and for his great help and tutorial for
finite element works in ANSYS. I will not forget the time he spent in reviewing my final work.
I would like to appreciate Dr T Sriskandarajah and Pradeep Hegde from Subsea 7 Sutton for their
assistance.
I am thankful to Arild sthus (Central Engineering Manager, the Department of Rigid Pipeline
and Structural Engineering, Subsea 7 Norway) and Tim Van Kempen for their help.
My appreciation also goes to Dr. Qiang Chen for his great tutelage in the areas of buckling
analysis and walking.
I am also grateful to my sponsors (Obi Obele, Geoffrey Nwankwo, Dr. Daniel Obele and
Virginia Madueke) and most importantly my mother, my brothers and sister whom their prayers
and support saw me during my time of studies.

Obele Ifenna Isaac
Stavanger, J une 10, 2013.

0bele Ifenna Isaac 0niveisity of Stavangei Page vi

0bele Ifenna Isaac 0niveisity of Stavangei Page vii
TableofContents
1. Introduction...................................................................................................................1
1.1 General.............................................................................................................................1
1.2 StateoftheArt.................................................................................................................5
1.2.1 PipelineSoilInteraction............................................................................................5
1.2.2 Engineered Buckle initiators and Mitigating Measures............................................7
1.2.3 AxialWalkingControlMeasures.............................................................................12
1.3 ThesisObjectives............................................................................................................15
1.4 OutlineofThesis.............................................................................................................17
2. TheoreticalBackground................................................................................................19
2.1 PipeBehaviour...............................................................................................................19
2.2 Buckling..........................................................................................................................19
2.3 OperatingStresses.........................................................................................................23
2.3.1 HoopStress (oH)...................................................................................................23
2.3.2 LongitudinalStress (oL).........................................................................................24
2.3.3 CombinedStresses..................................................................................................26
3. TheoreticalBackgroundEndExpansions.....................................................................27
3.1 PipelineEndExpansion..................................................................................................27
3.1.1 LongitudinalStrain..................................................................................................28
3.1.2 FrictionalStrainandForce......................................................................................29
3.1.3 EffectiveAxialForce................................................................................................30
3.1.4 EndExpansion.........................................................................................................31
4. TheoreticalBackgroundLateralBuckling....................................................................35
4.1 LateralBucklinganditsMechanism...............................................................................35
4.2 LateralBucklingModes..................................................................................................35
4.3 FeedinZone.................................................................................................................36
4.3.1 FeedinLengthandMaximumAllowableMoment................................................37
4.3.2 HobbsPredictiveAnalyticalMethod.....................................................................39
4.3.3 PipelineInitialImperfection....................................................................................42
4.3.4 VirtualAnchorSpacing............................................................................................43
4.4 SnakeLayControlMechanism.......................................................................................43
4.4.1 LayRadiusImperfection.........................................................................................44
0bele Ifenna Isaac 0niveisity of Stavangei Page viii
4.5 DesignLimitingCriteria..................................................................................................46
5. Pipeline Walking.........................................................................................................47
5.1 PipelineWalkinganditsMechanisms............................................................................47
5.2 SteepThermalGradient.................................................................................................49
5.3 PullingForceatSCR........................................................................................................50
5.4 SteepThermalGradient.................................................................................................50
5.5 InteractionbetweenPipelineWalkingandLateralBuckling.........................................51
6. DesignMethodology....................................................................................................53
6.1 General...........................................................................................................................53
6.1.1 Designobjective......................................................................................................53
6.1.2 DesignAssumptionsandRequirement...................................................................54
6.2 DesignProcessandRoadmapLateralbuckling............................................................55
6.3 DesignRoadmapforPipelineWalkingThermalTransientEffect...............................59
6.4 DesignAnalysis...............................................................................................................60
6.4.1 AnalyticalMethod...................................................................................................60
6.4.2 FiniteElementMethod(ANSYSMechanicalAPDL)LateralBuckling....................63
6.4.3 FiniteElementMethodPipelineWalking.............................................................69
7. DescriptionofCase.......................................................................................................71
7.1 FieldDescription.............................................................................................................71
7.2 DesignParametersLateralBuckling............................................................................72
7.3 DesignParametersWalking........................................................................................75
8. ResultsandDiscussion..................................................................................................77
8.1 ResultsandDiscussion...................................................................................................77
8.1.1 VerificationofPipelineLengthScale......................................................................77
8.1.2 EffectiveAxialForce................................................................................................78
8.1.3 EndExpansions.......................................................................................................79
8.1.4 SusceptibilityofPipelinetoLateralBuckling..........................................................80
8.1.5 RegionsSusceptibletoLateralBuckling.................................................................83
8.2 LateralBucklingBehaviour.............................................................................................84
8.3 SnakeLayControlMeasure............................................................................................86
8.3.1 LayConfigurationfor10kmpipelineR1500m.....................................................86
8.3.2 ParametricStudyofLayConfiguration...................................................................92
8.4 PipelineWalkingResultsandDiscussion.......................................................................96
0bele Ifenna Isaac 0niveisity of Stavangei Page ix
8.4.1 SusceptibilitytoPipelineWalking...........................................................................96
8.4.2 FiniteElementAnalysisAxialWalking..................................................................97
8.4.3 EffectofThermalGradientonPipelineWalking..................................................101
8.4.4 OperationalEffectofWalkingduetoThermalGradient......................................104
8.4.5 ControlofWalkingPhenomenonbyDirectElectricHeating(DEH).....................104
8.4.6 ChallengesfacingtheuseofDEH.........................................................................107
9. Conclusions,RecommendationsandFurtherWork....................................................109
9.1 SummaryandConclusions...........................................................................................109
9.2 Recommendations.......................................................................................................111
9.3 FurtherWork................................................................................................................111
REFERENCES............................................................................................................................113
APPENDIXA:CALCULATIONRESULTS..................................................................................116
APPENDIXB:ANSYSSCRIPT.................................................................................................125

0bele Ifenna Isaac 0niveisity of Stavangei Page xi
List of Figures
Figuie 11: Example of Beep Watei Subsea Fielu Layout (2b1stconsulting, 11 Septembei,
2u12)..............................................................................................................................................1
Figuie 12: 0pheaval anu Lateial Buckling (Floiiano et al., 2u11)...........................................2
Figuie 1S: Example of a local buckleu Pipe (Takahashi et al., 2uu7)......................................2
Figuie 14: Example foi 0valisation Failuie Noue (Kyiiakiues anu Coiona, 2uu7)...............3
Figuie 1S: Example foi Ruptuie Failuie Noue (Ahmeu anu & uaieth, 2u12)........................3
Figuie 16: Illustiation of Pipeline walking (cieep) that coulu leau to excessive enu
movement anu ultimately the failuie of tiein jumpeispool connection (EcoPiasinos,
2u12)..............................................................................................................................................4
Figuie 17: Example of Pipeline Response uuiing Pipesoil Inteiaction (Biuton et al., 2uu7)
.........................................................................................................................................................5
Figuie 18: Illustiation of Pipesoil inteiaction foi Effective axial foice foi a iange of fiiction
in a stiaight Pipe (White anu Biuton, 2uu8)...............................................................................6
Figuie 19: Illustiation of Pipesoil inteiaction foi uisplacement along shoit pipeline with
lateial impeifections (White anu Biuton, 2uu8)........................................................................6
Figuie 11u: Biffeient Regions in a buckle (Kein et al.).............................................................7
Figuie 111: Post Effective foice of a single isolateu buckle (Kein et al.).................................8
Figuie 112: Illustiation of expansion shaiing with multiple buckles (Kein et al.).................8
Figuie 11S: Example of Niuline expansion spool (Kein et al.)...............................................9
Figuie 114: Buckle initiating using veitical sleepeis (Kein et al.).........................................10
Figuie 11S: SB view of Buckle initiation using veitical tiiggei (Kein et al.)........................10
Figuie 116: Buckle initiation using uistiibuteu Buoyancy (Biuton et al., 2uuS).................11
Figuie 117: Snake Lay Configuiation (Kein et al.)..................................................................11
Figuie 118: Schematic view of Suction anchoi, flowline anu Risei System (Subsea7, 2u12)
.......................................................................................................................................................12
Figuie 119: Inline Expansion Spool..........................................................................................13
Figuie 12u: Sketch of the PLET with Sliuing Nechanism (Caineiio anu castelo, 2u1u)......13
Figuie 121: PLET with Sliuing Founuation (Caineiio anu castelo, 2u1u)............................13
Figuie 21: Loau Response in Buckling (Robeit).....................................................................19
Figuie 22: Bifuication Buckling (Ahmeu anu & uaieth, 2u12).............................................19
Figuie 2S: SnapThiough Buckling (Ahmeu anu & uaieth, 2u12).........................................20
Figuie 24: Loau Response in buckling (Robeit)......................................................................20
Figuie 2S: Beam Section unuei Loauing..................................................................................20
Figuie 26: Stiess Inuuceu by Inteinal Piessuie Loauing (Kaiunakaian, 2u12)..................23
Figuie 27: Longituuinal Stiess component (Piof. Shaima)....................................................24
Figuie 28: Piessuie inuuceu by Enu Cap Effect (Kaiunakaian, 2u12)..................................24
Figuie 29: Pipeline expansion uue to Poisson's effect (uuo et al., 2uuS)..............................25
Figuie S1: Example of Expansion Analysis..............................................................................27
0bele Ifenna Isaac 0niveisity of Stavangei Page xii
Figuie S2: Theimal stiain effect...............................................................................................28

Figuie SS: Enu Cap Effect..........................................................................................................28
Figuie S4: Poisson Effect...........................................................................................................29
Figuie SS: Illustiation of Lay Tension inuuceu uuiing Pipe laying (Fyiileiv anu Collbeig,
2uuS)............................................................................................................................................31
Figuie S6: Pipeline Enu Expansion (Cheuk, 2uu7).................................................................33
Figuie 41: Symmetiic anu Asymmetiic buckle moues (Kaye, 1996)....................................36
Figuie 42: Lateial Buckling moues (Kaye, 1996)....................................................................36
Figure 4-3: Feed-in-Zone for Mode-3 Buckling (Ahmed and & Gareth, 2012)...........................37
Figuie 44: Lateial Buckling configuiation (EINSFELB et al., 2uuS)......................................39
Figuie 4S: Foice piofile along pipeline showing viitual anchoi spacing anu ieuuction in
foice at each lateial buckle (Biuton anu Caii, 2u11)...............................................................43
Figuie 46: SnakeLay configuiation (Subsea7, 2u12).............................................................44
Figuie 47: SnakeLay with Lay Rauius Impeifection (Biuton anu Caii, 2u11)....................45
Figuie 48: Ciitical Buckling Foice Against the lateial soil fiiction (Runusag et al., 2uu8)..45
Figuie 49: Schematic view of Pipeline walking uue to sepaiation of viitual Anchoi
(Chauuhuiy, 2u1u)......................................................................................................................47
Figuie 41u: Foice Piofile uuiing the fiist anu seconu heating stage (Caii et al., 2uu6)......48
Figuie 411: Illustiation of Pipeline waking anu its contiibutoiy factois (Biuton anu Caii,
2u11)............................................................................................................................................49
Figuie 412: Foice Sloping Seabeu (Caii et al., 2uu6)...........................................................50
Figuie 41S: Schematic of PipelineRisei tension at the enu connection...............................50
Figuie 414: Theimal Tiansient.................................................................................................51
Figuie 41S: Effective axial foice in a shoit pipeline................................................................51
Figuie 416: Effective Axial foice foi long pipeline..................................................................52
Figuie 61: Besign steps foi Pipeline Lateial buckling contiolleu using Snakelay
configuiation...............................................................................................................................55
Figuie 62: Besign ioaumap foi Pipeline walking unuei theimal tiansient..........................60
Figuie 6S:Pipe288 ueometiy anu the ANSYS mouel..............................................................64
Figuie 64: A Pieloau associateu with pipeline 00S...............................................................65
Figuie 6S: Pipeline showing an Initial 00S of 1m...................................................................65
Figuie 66: Layiauius ueuuction...............................................................................................65
Figuie 67: StiessStiain ielationship in ANSYS mouel...........................................................66
Figuie 68: Bounuaiy conuition showing pipe enus fixeu in all uiiection.............................67
Figuie 69: Seabeu Nouel on zplane........................................................................................67
Figuie 61u: Nouel showing the contact element geneiating fiiction Foice on the Seabeu 68
Figuie 71: A typical Subsea fielu with Pipeline on even seabeu (Subsea7, 2u11)................71
Figuie 72: Tempeiatuie piofile 1ukm Pipeline....................................................................73
Figuie 7S: Pipeline Steel Nateiial BE iating............................................................................74
0bele Ifenna Isaac 0niveisity of Stavangei Page xiii
Figuie 74: Byuiostatic Effect fiom the topsiue connection...................................................75

Figuie 7S: Tempeiatuie Piofile................................................................................................76
Figuie 81: Effective Axial foice of a shoit Pipeline.................................................................78
Figuie 82: Effective Axial Foice of a long pipeline..................................................................79
Figuie 8S: Bobbs Ciitical Buckling Foice................................................................................81
Figuie 84: Region Susceptible to lateial Buckling...................................................................83
Figuie 8S: lateial uisplacement at the buckle site..................................................................84
Figuie 86: Effective axial foice foi uiffeient fiiction factois..................................................85
Figuie 87: PieBuckling anu Post Buckling of a pipeline snake Configuiation....................87
Figuie 88: Effective axial foice uistiibution foi Snake Lay configuiation............................88
Figuie 89: Expansion Bistiibution...........................................................................................90
Figuie 81u: Effect of Snake Lay configuiation.........................................................................91
Figuie 811: Rock Bumping between snake.............................................................................92
Figuie 812: Effective foice uistiibution of snake lay (R2Suum) configuiation with Rock
uumping.......................................................................................................................................93
Figuie 81S: Expansion iesults foi R2Suu................................................................................94
Figuie 814: Theimal tiansient at SS
o
Ckm.............................................................................97
Figuie 81S: Axial uisplacement piioi to full mobilization.....................................................98
Figuie 816: Axial uisplacement at full mobilization...............................................................98
Figuie 817: Axial Bisplacement uuiing the Fiist heating cycle.............................................99
Figuie 818: Theimal giauient showing two uiffeient Scenaiios (SS
o
Ckm anu 2u
o
Ckm)
.....................................................................................................................................................101
Figuie 819: Axial uisplacement foi piioi to full mobilization foi two uiffeient theimal
giauient......................................................................................................................................102
Figuie 82u: Axial uisplacement of the two cases aftei full mobilization.............................102
Figuie 821: Cumulative uisplacement foi SS
o
Ckm anu 2u
o
Ckm......................................103
Figuie 822: Effect of walking uue to theimal giauient in SCR.............................................104
Figuie 82S: Biiect Electiic heating of pipeline ciosssection (Baialu, 2uu8)....................104
Figuie 824: Effect of BEB on theimal tiansient....................................................................105
Figuie 82S: Axial uisplacement foi both BEB anu noimal heating step.............................106
Figuie 826: Axial uisplacement foi heating with BEB anu without BEB...........................106
Figuie 827: Biiect Electiic heating BEB (Baialu, 2uu8)...................................................107

0bele Ifenna Isaac 0niveisity of Stavangei Page xiv

0bele Ifenna Isaac 0niveisity of Stavangei Page xv
List of Tables
Table 41: Constant foi Lateial Buckling Noues (Bobbs, 1984).............................................41
Table 71: Basic Besign paiameteis..........................................................................................72
Table 72: Exteinal coating paiametei......................................................................................72
Table 7S: Fiiction Coefficients..................................................................................................73
Table 74: Applicability of Pipeline............................................................................................74
Table 81: Anchoi Length iesults...............................................................................................77
Table 82: Enu Expansion Long pipeline...................................................................................79
Table 8S: Enu Expansion of shoit pipeline..............................................................................80
Table 84: Constants foi Lateial buckling moue.......................................................................81
Table 8S: Bobbs Ciitical Buckling foice at uiffeient fiiction factois.....................................82
Table 86: Analytical Result Ciitical Buckling Foice..............................................................82
Table 87: Feeuin Results foi R1Suum.....................................................................................87
Table 88: Expansion Summaiy foi the R1Suum configuiation.............................................89
Table 89: Feeuin Results foi lay iauius of 2Suum.................................................................92
Table 81u: Expansion Summaiy foi R2Suu.............................................................................95
Table 811: Analytical Result foi iate of walking at SS
o
Ckm.................................................96
Table 812: Rate of walking foi seveial cycles..........................................................................99
Table 81S: Results fiom FE Analyses anu Analytical Calculation foi Biffeient Axial
Fiictions.....................................................................................................................................100
Table 814: Results foi both cases...........................................................................................106
Table 91: Expansion Result.....................................................................................................109

0bele Ifenna Isaac 0niveisity of Stavangei Page xvi

0bele Ifenna Isaac 0niveisity of Stavangei Page xvii
Nomenclature
Abbreviations

BE Best Estimate

LB Lower Bound

UB Upper Bound
DNV Det Norske Veritas

FEA Finite Element Analysis

OOS Out-of-Straightness

OD Outside Diameter

OS Offshore Standard

RP Recommended Practice

SMYS Specified Minimum Yield Strength

MSL Mean sea level

SNCF Strain Concentration factor

DEH Direct electric heating

Symbols

E Youngs modulus of the steel pipe

H Residual lay tension

I Second moment of area

M
b
Bending moment

N
a
Axial force

P
e
External pressure

P
i
Internal pressure

N
e (x)
Effective axial force
R Ray radius
F
cr
Critical Buckling Force

W
submerged
Weight submerged

WD Water Depth

N
OOS
Force due to out-of-straightness (OOS)

U
eed-|n
Design feed-in length

X65 Steel grade of 65000psi
T
amh
Ambient temperature

N
Hobbs
Force Hobbs

o
H
Hoop stress

o
L
Longitudinal stress

o
LF
Longitudinal stress due to End cap

o
LT
Longitudinal stress due to temperature

s
pu|xxun
Stroin Juc to Poisson ccct
N Iruc woll Forcc

VAS Virtual Anchor Spacing

HP/HT High Pressure / High temperature

PLET Pipeline End Terminal

SCR Steel Catenary Riser

KP Kilometer point on pipeline

ANSYS Analysis System

PLET Pipeline End terminal
APDL ANSYS parametric design language
SMTS Specified Minimum Tensile Strength

PIP Pipe in Pipe
0bele Ifenna Isaac 0niveisity of Stavangei Page xviii
0bele Ifenna Isaac 0niveisity of Stavangei Page 1
1. Introduction
1.1 General
As oil and gas industry moves farther into deep and ultra-deep waters, HP/HT envelope
continually being pushed outward and fields with wellhead pressures and temperatures of order
of 600bars (8700 psi) and 170
o
C are being developed.
Hence, the need for improved technology in handling the delivery of well fluid has become a
challenge to pipeline industries.
Environmental and operational conditions in deep waters make it almost impracticable to operate
pipeline system (as shown in Figuie 11) and hence will require appropriate design guidelines in
regulating the pipeline-soil behavior in order to counter the large uncertainties developed in the
design of such system. These uncertainties from pipe-soil force-displacement response are as a
result of differences in seabed-pipeline temperatures, pressures and higher hydrostatic pressures.
Figure 11: Example of Deep Water Subsea Field Layout (2b1stconsulting, 11 September, 2012)
Pipeline left exposed on seabed under operational conditions have a potential to buckle, walk and
change configuration due to high temperature and pressure (HT/HP) operational conditions. If
the pipeline is restrained, a compressive axial force will be induced in the pipeline. According to
(Palmer and King, 2004), this could lead to buckling of the pipeline if the compressive axial
force induced reaches the critical buckling force. As a result of the induced force, the pipeline
will tend to move upward or sideways to release the excessive axial force induced.
The direction of the movement depends on the pipeline restrictions. As shown in Figuie 12, large
induced axial compressive force for trenched or buried pipeline will therefore lead to upheaval
buckling (Upward) while exposed surface laid-pipeline leads to lateral buckling (sideways). This
will endanger the integrity of the pipeline if not controlled.
The phenomenon above is termed Global buckling. This is not a failure mode in itself, but a
load response and could lead to several failure modes such as local buckling, fracture and fatigue
(DNV-RP-F110, 2007).
Figure 12: Upheaval and Lateral Buckling (Floriano et al., 2011)

Local buckling is normally the governing failure mode resulting from excessive material
utilization (Almeida, 2001). It appears as wrinkling or as a local buckle on the compressive side
of the cross section as shown in Figuie 1S. This failure mode could result in excessive material
Ovalisation and reduced cross-section area which reduces production efficiency or even cause
full production stop in any event of pig getting stuck during pigging/inspection. A locally
buckled pipeline cannot stand an increased bending moment in the pipeline. This also could lead
to pipeline collapse or production lost time.
Figure 13: Example of a local buckled Pipe (Takahashi et al., 2007)

0bele Ifenna Isaac 0niveisity of Stavangei Page S
Fracture is another failure mode; it is the failure on the tensile side of the cross section which is
due to excessive material utilization through cyclic loading of the pipeline system.
Fractured pipeline could cause leakage or full bore rupture leading to reduced production, or
even full production stop (Almeida, 2001).
Figure 14: Example for Ovalisation Failure Mode (Kyriakides and Corona, 2007)
Low cycle fatigue which often occurs for limited load cycles gives strains in the plastic region.
This resulting strain could possibly cause pipe leakage or rupture (see figure 1-5), resulting to
production reduction, or full production stop. Pipeline exposed to seawater and stresses from
buckle could also lead to leakage through hydrogen induced stress cracking.
Figure 15: Example for Rupture Failure Mode (Ahmed and & Gareth, 2012)
There has been buckling cases in several fields in the world for example, as recorded In J anuary
2000, a 17km 16-Inch pipeline in Guanabara Bay, Brazil, suddenly buckled 4m laterally and
ruptured, leading to a damaging release of about 10,000 barrels of oil and a great loss to the
operator.
Field observation showed that as a result of temperature increase, the pipeline displaced laterally,
when failure took place. Operating pressure and temperature of the pipeline were 400bar
(5800psi) and 95C, respectively. The soil beneath the pipeline was very soft clay with about
2kPa undrained shear strength at seabed (Almeida, 2001).
In this thesis concerning the lateral buckling problem stated above, the tendency for pipeline to
buckle will be investigated. A non-linear finite element analysis will be conducted on the area of
the pipeline which is found to be susceptible to lateral buckling.
If buckling cannot be avoided, the most economical mitigating measure (for example, snake-lay
configuration) will be utilized.
Moreover, under repeated start-ups and short-downs and corresponding heating and cooling
during subsea operations, cumulative axial displacement of short pipelines could occur. This un-
wanted mechanism is termed pipeline walking.
This walking mechanism as written by SAFEBUCK J IP (Carr et al., 2006) could lead to
excessive pipeline end movement and ultimately the failure of tie-in jumper/spool connection,
loss of tension in a steel catenary riser (SCR) and increased loading within buckled area.
The driving mechanism of axial walking is the expansion and contraction of the pipeline and also
on the possibility of no movement in case of an anchor point constraint. As a result of pipeline
walking, the expansion at one end of the pipeline would be more than the expected value
calculated during design stage and may cause failure of expansion spool or riser (Almeida,
2001).
The rate of axial walking depends strongly on temperature profiles, the magnitude of axial
resistance, the mobilization distance and the degradation to residual conditions (Carr et al.,
2006).
According to (Almeida, 2001), in year 2000, there were six incident reported in North sea as a result of
excessive expansion of pipeline and at least one loss of containment failure due to pipeline walking.
Figure 16: Illustration of Pipeline walking (creep) that could lead to excessive end movement and
ultimately the failure of tiein jumper/spool connection (EcoPrasinos, 2012)

WALK
0bele Ifenna Isaac 0niveisity of Stavangei Page S
1.2 State of the Art

1.2.1 PipelineSoil Interaction
Pipe soil interaction influences both mobilization load (breakout resistance) and pipeline post-
buckling configuration from the moment installation commences (DNV-RP-F110, 2007) . This
behavior and responses are subject to global buckling in HP/HT conditions. The interaction
between the pipeline touchdown loads, combined with the dynamic of the pipe catenary and the
seabed surface soil defines the initial pipeline embedment (Bruton et al., 2007). During
installation, the remolding of the soil influences the axial resistance affecting the condition at
which the pipeline becomes restrained. In the same way, lateral resistance affects the tightness of
the installation curves that can be achieved during pipe-laying.
Figure 17: Example of Pipeline Response during Pipesoil Interaction (Bruton et al., 2007)
Pipeline-soil interaction is the largest uncertainty in the design of pipelines both due to variation
and uncertainty in characterization (DNV-RP-F110, 2007). The corresponding force-response
models have been developed during phase 1 of the SAFEBUCK J IP. The SAFEBUCK J IP was
initiated to address this challenge and aims to raise confidence in the lateral buckling design
approach and to improve understanding of the related phenomenon of pipeline walking (Bruton
et al., 2007).
Consequently, SAFEBUCK J IP performed analysis about pipeline structural response which
detailed the responses during installation, expansion during first loading, response in a buckle
(first load), response in a buckle (Cyclic behavior - influence of berms) and buckle initiation.
It was shown in their work that pipelines have a total different behavior at lower bound axial
friction and upper bound axial friction. The lower bound friction (for examplep
u
= u.1u ),
means that the pipeline experiences greater end expansion and is susceptible to Pipeline
walking. The Upper bound friction (for example, p
u
= u.S8) means that the pipeline is fully
constrained over some of its length so that the section in contact will not move axially, thus
preventing walking, but the effective axial compressive force will hence increase significantly
making it susceptible to lateral buckling.
In summary, it was deduced that the low axial friction will increase the end expansion and axial
feed-in into a buckle site while high axial friction will tend to reduce end expansion and feed-in
as can be seen from the Figuie 18 below:
Figure 18: Illustration of Pipesoil interaction for Effective axial force for a range of friction in a
straight Pipe (White and Bruton, 2008)
Figure 19: Illustration of Pipesoil interaction for displacement along short pipeline with lateral
imperfections (White and Bruton, 2008)
From Figuie 19, it is seen that the compressive axial force in a pipeline depends on the
temperature condition of the pipeline and the axial friction. The paper (White and Bruton, 2008)
re-stated that if the compressive force is large enough, then the pipeline may be susceptible to
lateral buckling but this will only occur when the compressive force exceeds the critical buckling
force as stated previously above (Palmer, 2004).
Also, as the temperature and pressure fluctuates, it creates a cyclic soil-pipeline interaction
which influences berms formation. Subsequent consolidation increases its strength and will result
in axial feed-in and out of the buckles with each cycle. This is an unwanted scenario in subsea
pipeline operation and should be controlled.
1.2.2 Engineered Buckle initiators and Mitigating Measures

Buckles are deliberately initiated by introducing initiation sites (triggers) along the pipeline route
to ensure that the pipeline laterally buckles in a planned scenario in order to avoid the induced
axial compressive forces concentrating in a particular site.
Buckle is initiated by either one of the following parameters; effective compressive force in the
pipeline, out-of-straightness (OOS) features and lateral breakout resistance. Lateral Buckling
breakout having the highest uncertainty (Bruton et al., 2007).
Potential localization which is related to inhomogeneity in pipeline as-laid configuration, pipe-
soil interaction and temperature leads to longer feed-in length to the largest buckles and increases
the susceptibility to buckling and fracture.
Figure 110: Different Regions in a buckle (Kein et al.)

As the temperature in the pipeline increases the slip length will therefore continue to feed-into
the buckle after the buckle has been developed (Kein et al.). The length of the slip zone depends
on the available frictional resistance to oppose the feed-in. A virtual anchor is developed where
there is sufficient frictional force to constrain the slip completely.
The post effective force will therefore change to take into effect the compressive forces into the
buckle. This scenario for the post effective force for an isolated single buckle is showed in Figuie
111 according to the analysis conducted by J P Kenny group.
According to the paper (Bruton et al., 2007), if lateral buckles are initiated at regular intervals
along the pipelines, the loads are effectively shared between the buckle sites.
Moreover, the shorter the spacing between buckle initiators, the lower is the probability of
buckle forming at t each site as desired. Therefore, selecting appropriate and suitable spacing is
the key but a challenging task in design.
Figure 111: Post Effective force of a single isolated buckle (Kein et al.).
If the temperature is further increased after the post buckling, more pipe length will feed into the
buckle and will increase the moment of the buckle (Kein et al.). This could lead to formation of
more buckles along the pipeline.
If the buckles are spaced such that the distance between successive buckles is less than the total
buckle length (Lo +2Ls) of an isolated buckle, the feed-in is shared between the two buckles as
shown in Figuie 112 (Kein et al.). This is known as expansion sharing (DNV-RP-F110, 2007).
Figure 112: Illustration of expansion sharing with multiple buckles (Kein et al.).
The conventional techniques to avoid buckling have been to restrain the pipeline by trenching,
burying and rock dumping. Alternatively, the thermally induced stress in the pipeline can as well
be relieved with the use of inline expansion spools or mid-line expansion spools (Cheuk, 2007).
In spite of this, these methods are becoming less cost-effective as the operating temperatures and
pressures are being required to increase further as the exploration moves into deeper waters
where trenching and burying are not viable. Hence, the pipeline is left exposed on the seabed and
allowed to buckle laterally.
In accordance with the recommended practice, DNV-RP-F110 (2007) if the response from the
applied loads exceeds the pipe cross-sectional capacity, mitigating measures have to be
introduced.
Apart from the conventional ways of preventing buckling there are number of improved
mitigating measures that have been utilized in the industry during the past years.
The lateral buckling concept has been a design concept that aims to work with the induced
expansion phenomenon rather than working against the induced stresses on the pipeline and
some of the measures that have been used are as follows:
a) Sharing of expansion into adjacent buckles:
This can be achieved by the use of rock dumping at intermittent sections, with the aim to
increase the restraint to axial movement in order to reduce the feed-in into isolated
buckles that may be triggered by imperfection or trawl gear (DNV-RP-F110, 2007).

b) Mid-line Expansion spool:
This utilizes the mid-line spool to absorb the pipe expansion under operational
temperature and pressure. Figuie 11S shows a mid-line expansion spool which was
modeled in U configuration and imposed to thermal expansion at both ends. (Kein et al.).

Figure 113: Example of Midline expansion spool (Kein et al.)
0bele Ifenna Isaac 0niveisity of Stavangei Page 1u
c) Vertical Triggers/Sleepers:
This is a method that utilizes initial vertical imperfection (Out-of-straightness - OOS) to
initiate a lateral buckle. Pipe sleepers pre-laid across the seabed is used to raise the
pipeline off the seabed. This will create a vertical imperfection, OOS, which will initiate
a buckle at this section. Figuie 114 illustrates buckles initiated by trigger. The buckle
crown elevates the pipe above the seabed and causes a reduction in lateral friction
resistance, and hence reduces uncertainties concerning lateral pipe-soil interactions.

Trigger/Sleeper lowers the critical buckling force as a result of reduction in lateral
friction resistance. This allows for higher thermal feed-in into the buckle site, therefore
increasing the buckle spacing and as a result reducing the number of buckle initiator
required (Kein et al.).

Figure 115: 3D view of Buckle initiation using Vertical trigger (Kein et al.)
Figure 114: Buckle initiating using Vertical sleepers (Kein et al.)

d) Buckle Initiation using distributed Buoyancy or additional insulation coating:

Figuie 116 illustrates buckle initiation using distributed buoyancy. The distributed
buoyancy is added to reduce the weight at the intermittent sections. As the critical
buckling force is a function of pipeline weight, the added distributed buoyancy leads to
buckle initiations as the weight reduces.

Figure 116: Buckle initiation using distributed Buoyancy (Bruton et al., 2005)

e) Snake Lay Configuration:
Figuie 117 present typical snake lay configuration. The concept of snake lay is to
deliberately install horizontal lay imperfections to trigger a sufficient number of buckles
at pre-determined locations along the pipeline so that the thermal expansion is distributed
among a number of buckles rather than being concentrated at a few buckle sites (Rundsag
et al., 2008).

Figure 117: Snake Lay Configuration (Kein et al.)

1.2.3 Axial Walking Control Measures

Axial walking cases can as well be controlled by the above mentioned measures where it is
less economical to be carried out. Axial walking will not result in pipeline failure if the pipeline
itself is not susceptible to lateral buckling (Rong et al., 2009). But, due to accumulated
displacement over several cycles, it may lead to failure of tie-in jumpers/spools connected to
pipeline. It could also lead to increased loading in buckle section and cause a potential
localization in the buckled area.
Many mitigating measures have been utilized in the industry to counter the effect of end
expansion resulting from repeated shut-downs and cool down during subsea operations. The
measures could be one of the following:
a) Use of Anchors: Walking is mostly mitigated by attaching the pipeline or its end
structures to anchors. As shown below in Figuie 118, the connection of the suction
anchor to the pipe can be done with a friction clamp. The details of this system can be
found in (Subsea7, 2012) and (Bruton et al., 2010)
Figure 118: Schematic View of Suction anchor, flowline and Riser System (Subsea7, 2012)
b) Increased Jumper/Spool Expansion Capacity: The end expansion buckling capacity
can be increased at additional cost to withstand against the cyclic loading from axial
displacement and hence there is a curtail effect of pipeline walking (Rong et al., 2009).
The end expansion capacity can be achieved by using longer spool/Jumpers. The
Effective axial force during start-up and shut-down will have to be used to design and
accommodate the increased jumper and spool expansion capacity.
0bele Ifenna Isaac 0niveisity of Stavangei Page 1S
c) Increased Axial Friction: As previously mentioned in this report, pipeline interacted

with lower bound friction (for e.g. p
u
= u.1u ) will experience greater end expansion
and will be susceptible to Pipeline walking (Bruton et al., 2005). One of the controlling
measures for such axial walking is to increase the axial friction. The requirement of
increased axial friction at appropriate sections can be assessed by investigating through
FE and pipe-soil interaction analysis. The increased axial friction can be achieved by
using several techniques such as concrete weight coating, trenching and burying and
the use of rock dumps or mattress at the appropriate sections (Rong et al., 2009).
d) Use of Inline Expansion Loop:

Figure 119: Inline Expansion Spool

f) PLET with Sliding Foundation End Structure

Figuie 12u and Figuie 121 show an arrangement of
PLET with sliding foundation. This is a pipeline
walking mitigating measure that utilizes the sliding
foundation end structures (Carneiro and castelo,
2010). This measure is used where the expansion and
contraction are large enough that the expansion
spools cannot accommodate and the use of longer
spools will be expensive to operate. Further, this
measure will take greater expansion forces even deep
water. Depending on the spool size attached to the
PLET, the effect of end expansion will be reduced as
axial friction factor seizes to be a driving force.
In order to accommodate the end expansion and contraction of the pipelines, inline expansion
loops are installed at regular intervals along each pipeline (see
Figuie 119). This has been utilized in many short and long pipelines.
Figure 121: PLET with Sliding Foundation
(Carneiro and castelo, 2010) Figure 120: Sketch of the PLET with Sliding
Mechanism (Carneiro and castelo, 2010)
1.3 Thesis Objectives
The objective of the present thesis work is to study and understand the influence of pipeline-soil
interaction on the design of surface laid subsea pipeline susceptible to lateral buckling and
pipeline walking.
The main focus of the thesis work is on:
The use of snake-lay configuration as a mitigating measure under controlled buckling
design and rock dumping if needed to limit feed-in into buckle and end expansions.

The effect of thermal gradient on axial walking and the use of direct electric heating
(DEH) to reduce rate of walking.
The aim is to limit pipeline end expansion at the connecting ends and feed-in into the buckle
using snake-lay design with intermittent rock dumping.
The acceptability of snakes as engineered buckles will be verified based on lateral buckling
criteria by performing a design check according to DNV RP F110:
Local buckling check (displacement control criteria) which is the main criteria to obtain
the allowable virtual anchor spacing.

The work includes performing analytical investigations of pipeline expansion and lateral
buckling and verifying the results against the predictions from a non-linear finite element
analyses (FEA). The non-linear FE analyses are performed by modeling the soil-pipeline
interaction of as-laid pipeline and using general finite element software ANSYS.
The work further analyzes the lateral buckling of the pipeline with an initial out-of-straightness
(OOS) while applying internal pressure and temperature over a 2km VAS model.
Also, the work include FE based analyses to investigate the effects of thermal gradient in
pipeline walking phenomenon is investigated and assess the use of direct electric heating (DEH)
system to reduce axial walking. This is done by considering two thermal gradients of asymmetric
heating profile having same heating steps.
The effect of axial friction factor in axial walking phenomenon is analyzed using FE software
ANSYS while comparing the results with analytical results obtained using SAFEBUCK
guideline.


1.4 Outline of Thesis
The thesis is organised in 9 chapters based on the problem and solution considering the
objectives presented in the previous page. The details of the chapters are briefly outlined below.
Chapter 2: (Theoretical background Pipe Behaviour) this chapter presents the behaviour of
pipe material under the influence of compressive axial force and the operating stresses that could
affect the pipe to move in different directions.
Chaptere3: (Theoretical background End expansions) this chapter summarizes expansion that
occur at the end connection due the movement of pipe end as a result of induced effective axial
forces. Detailed analyses of the driving factors like temperature and pressure loading, strain and
end-cap effect are discussed.
Chapter 4: (Theoretical background Lateral Buckling) the chapter details some of the
governing theorys behind pipeline lateral buckling behaviour like Hobbs critical force analysis,
effective axial force, anchor force, initiation control mechanism (expansion sharing formula) and
allowable feed-in based on recommended practice or standard.
Chapter 5: (Theoretical background Pipeline walking) presents the theory behind pipeline
axial walking, relevant equations, driving mechanism and previous work done by SAFEBUCK
J IP.
Chapter 6: Design Methodology The chapter describes the relevant steps and procedures
which are to be followed in designing for lateral buckling and pipeline walking on even seabed.
The design requirement, the assumption, the reason for the use of ANSYS software in the work
and the snake Lay configuration principle shall be discussed within this module.
Chapter 7: Description of the study The chapter presents, the thesis problem and the necessary
data. An in-place design of 10km pipeline for lateral buckling and 2km flowline for axial
walking is presented.
Chapter 8: Results and Discussion The chapter presents the results obtained from the
analytical calculations and FE analysis. Furthermore, the results with respect to lateral and
pipeline working mechanism are discussed. Also, the results from snake lay configuration results
are presented and discussed.
Chapter 9: Conclusions and Further work The chapter summarizes the results of the analysis
and presents the conclusions of the thesis and recommendations for any future work.

2. Theoretical Background
2.1 Pipe Behaviour
The present section discusses the theoretical background and basic scientific principle relating to
pipe-soil interaction with respect to end expansion, lateral buckling and axial walking. This will
vary from the driving factors of high temperature and pressures, breakout resistance to thermal
buckling.
The basics of the study were generated by the principle of buckling phenomenon in a simple bar
element. The same principle is applied for a subsea pipeline installed on a seabed.
2.2 Buckling
Buckling occurs physically when a structure becomes unstable under a loading configuration
and mathematically when a bifurcation occurs in the solution to equation of equilibrium
(Ondrej, 2012). Buckling could either be a:
Bifurcation Buckling This is a situation where the elastic stiffness of the structure is
cancelled by the effects of compressive stress within the structure. If the effect of this
causes the structure to suddenly displace a large in a direction normal to the loading
direction, then it is a classical bifurcation buckling. Figuie 21 and Figuie 22 illustrate the
bifurcation buckling and the load response in the bucking

Snap-through Buckling If there is a sudden large movement in the direction of the
loading it is called a snap-through buckling. According to Robert This occurs in
structures experiencing limit point instability, when the load is increased infinitesimally
beyond the critical load, the structure undergoes a large deformation into a different
stable configuration which is not adjacent to the original configuration (Robert).
Figure 22: Bifurcation Buckling
(Ahmed and & Gareth, 2012)
Figure 21: Load Response in Buckling (Robert)
Figure 23: SnapThrough Buckling (Ahmed

and & Gareth, 2012)
Figure 24: Load Response in buckling
(Robert)
For pipeline with small initial imperfection, the buckling is expected to occur as a snap through
buckling jumping from a particular equilibrium of smaller displacement to another equilibrium
position of higher displacement while for those with large initial imperfection, it will undergo a
gradual displacement (Ahmed and & Gareth, 2012). Figuie 2S and Figuie 24 illustrates the snap
through buckling and its load response.
In reality, pipelines lateral imperfection will arise mostly from vessels motion during pipe
laying. Hence, the pipeline will buckle laterally once the effective force reaches the critical
Buckling Force. The term initial imperfection will be discussed later in this report accordingly.
Consider a beam section of length, L and Flexural rigidity, EI and compressive axial force, P.

L

const EI

Buckling is said to occur if the combined bending and compressive stresses reaches the Critical
Buckling Load, P
cr
. By considering the equilibrium of lateral forces and bending moments
acting on the beam section.
The dynamic equation of motion for the beam section exposed to an axial compressive force, P is
given as:

EI
J
4
w
Jx
4
+P
J
2
w
Jx
2
= u . . (1)

From the homogeneous equation above, taking w = e
2x

P P
M M
Figure 25: Beam Section under Loading
Substituting the function w = e

2x
in equation (1), we have

z
4
+
P
EI
z
2
= u . . (2)
Fromequation,2,thesolutionforzbecomesacomplexnumber
z = _
_
P
EI
i
Hence,thegeneralhomogenoussolutionbecomes:
w = Ax|n
_
P
EI
x +Bcos
_
P
EI
x + C + x
Consideringacaseofsimplysupportedbeamand,invokingtheboundaryconditionwehave:
w(x = u ) = u no Jccction ot tbc support
H(x = u) = EI
J
2
w
Jx
2
= u . no bcnJing ot tbc support
w(x = I) = u no Jccction ot tbc support
H(x = I) = EI
J
2
w
Jx
2
= u . no bcnJing ot tbc support
From the above, we can deduce that
B = C = = u
Therefore, we have:
APx|n_
P
LI
I = u
Here A and P cannot be zero, so we conclude that . x|n_
P
LI
I = u
Ibcrcorc,
_
P
EI
I = an
The Critical Buckling that must be exceeded for buckling to occur becomes:
P
cr
=
n
2
n
2
EI
I
2
. (S)
Forn = 1,
P
cr
=
n
2
EI
I
2

mude 1

Forn = 2,
P
cr
=
4n
2
EI
I
2
= 4P
n=1
mude 2

2.3 Operating Stresses

Operating stresses are stresses which result from a combination of internal pressure and thermal
stresses that occur during operation (Guo et al., 2005). During Operations, pressure and
temperature build up and cause the pipeline to expand both radially and longitudinally due to the
differences created between the pipeline and the surroundings. The stresses vary in magnitude
depending on the forces opposing them, for example, the forces from axial soil friction, end
constraints and end cap.
(Karunakaran, 2012) stated that when a pipeline is subjected to internal pressure, three mutually
perpendicular principal stresses will be set up in the cylinder materials (see Figuie 26), namely

a. Circumferential or hoop stress
b. The radial stress
c. Longitudinal stress
Taking into account that the radial stresses which acts normal to the curved plane of the isolated
element are negligibly small as compared to other two stresses especially in the case of relatively
thin-pipe when (t > 2u) .

Figure 26: Stress Induced by Internal Pressure Loading (Karunakaran, 2012)

2.3.1 Hoop Stress (o
H
)
For a thin-wall pipeline(t > 2u), subjected to internal pressure, P, the effect of the radial
force distributed around the circumference will produce a stress called circumferential or Hoop
Stress o
H
(Karunakaran, 2012).
o
H
=
P
2t

Where: o
H
= Eoop strcss
P = Intcrnol Prcssurc
= nct intcrnol iomctcr
t = nominol woll tbickncss
As operations gets to the deeper waters, the mitigating effect of external pressure should be
included as follows:
o
H
=
(P
IocuI
-P
c
)
2t

Where: P
IocuI
= Iocol Jcsign prcsurc
P
c
= Extcrnol Prcssurc
= cxtcrnol Jiomctcr
2.3.2 Longitudinal Stress (o
L
)
Longitudinal stress is referred as the axial stress experienced by the pipe wall. As seen below
from Figuie 27, it comprises the stresses due to the end cap effect, temperature (thermal stress),
bending, axial and Poissons effect (hoop stress).
Figure 27: Longitudinal Stress component (Prof. Sharma)

2.3.2.1 End Cap Effect (o
LF
)
The longitudinal stress due to the end cap effect can be calculated by dividing the total pressure
against the end of the pipe (end cap effect) by the cross-section area of the pipe.
Figure 28: Pressure induced by End Cap Effect (Karunakaran, 2012)

o
LL
=
Forcc
X - scction orco
=
P. (
n
2
4
)
nt
=
P
4t

2.3.2.2 Thermal Stress

In restrained condition, either by soil friction, anchor, buried or trenched, the restriction of the
expansion generated by temperature differences between the installation and operating
temperature induces a compressive stress on the pipeline. As a result of this restriction, the
longitudinal strain is zero and the induced compressive thermal stress generated is given as:
o
L1
= -oEI
For the unrestrained condition, the longitudinal thermal stress is zero but the thermal strain is
given as: o
L1
= oEI
2.3.2.3 Hoop Stress (Poissons effect)
The longitudinal stress induced due to Poissons effect can be calculated according to the
equation below:
o
LH
= :o
H

Where: : = poissons rotio; u.S or stccl pipc
Figuie 29 explains pipeline expansion due to poisons effect. According to (Guo et al., 2005),
When a two-dimensional element is heated but subjected to a restraint in the y-direction, the
strain in the x-direction is increased due to the Poisson ratio v.
Figure 29: Pipeline expansion due to Poisson's effect (Guo et al., 2005)
2.3.2.4 Bending Stress

The Bending stress (o
LB
) associated with the longitudinal stress according to (Yong and Qiang,
2005), can be calculated as follows:
o
LB
=
H
b
Z
s

H
b
= BcnJing momcnt
Z
s
=
I
z
(2)
=
n
64
(
4
-
4
)
(2)
= Scctionol moJulus o rigiJity o tbc pipc
2.3.2.5 Axial Stress (o
LA
)
The Longitudinal stress generated due to the axial stress can be computed as stated in (Yong
andQiang,2005)bytheequation:
o
LA
=
N
u
A
s

Where: N
u
= Axiol orcc
A
s
= Cross - scctionol Arco =
n(
2
-
2
)
4

2.3.3 Combined Stresses
Provided that the sign convention foe compressive and tensile effect are employed properly, the
total longitudinal stress can be determined according to the relation given in (Guo et al., 2005)
as:
o
L
= o
LL
+o
L1
+ o
LH
+ o
LB
+o
LA

Therefore, the combined stress depending on the approved codes and standard for the project
shall meet the requirement below:
o
combncd
= o
H
2
+o
L
2
- o
H
o
L
F
combncd
. SHS
3. Theoretical Background - End Expansions

3.1 Pipeline End Expansion
A Subsea pipelines with free end is considered laid on a flat seabed and free to expand. Under
constant operating temperature and pressure, the pipeline expands and the soil friction forces are
activated which oppose the expansion. The constraint created by the soil friction causes an
effective compressive axial force to be induced in the pipeline which results in axial
displacement (strain) of the pipeline. However, as a result of this restraint, the expansion will
only manifest at the ends. This is referred to as End Expansion.
As shown below in Figuie S1, expansion occurs within a transition region whose length depends
on the limiting axial friction between the soil and pipeline. If the frictional resistance force is
large, the corresponding transition region will be small and if the friction it is less, the transition
length will be larger.
Figure 31: Example of Expansion Analysis

In pipeline design, expansion analysis will help to determine the maximum expansion at the ends
and the associated axial strain loadings that the pipeline can carry without any failure. According
to analyses performed by (Prof. Sharma), expansion analysis will provide information such as:
The maximum axial loading that will buckle the pipeline
The maximum expansion the spools/tie-in Jumpers can accommodate.
As shown in Figuie S1, the degree of this expansion depends on the operational temperature,
pressure, the pipe weight and the restraint on the pipeline. The pipeline will tend to expand until
the anchor point where it is fully restrained by the soil friction. This distance from the pipeline
end to the anchor point is called the ANCHOR LENGTH.
The lower the axial friction (the restraint), the greater the end expansion at the spool or tie-in end
(the greater the anchor length).
3.1.1 Longitudinal Strain
The maximum pipeline end expansion is consequence of the net longitudinal strain and frictional
force between the pipeline and the seabed (Offshorevn, 2010).
The pipeline expansion as stated above will occur at pipeline ends under unrestrained condition
leading to longitudinal strain at the ends. The longitudinal strains are due to Temperature and
pressure effect.
3.1.1.1 Thermal Strain (s
T
)
For an unrestrained pipeline, the temperature change created by the difference between the
operating and installation temperature induces a thermal strain that is linearly proportional to the
change in temperature in axial direction as shown in Figuie S2:

Therefore, the thermal strain induced in the pipeline in an unrestrained condition is given as:
T
= uT = u(T
u
-T
|
)
3.1.1.2 Pressure Strain
The longitudinal strain created due to pressure loadings are as a result of end-cap effect and
Poisons effect.
a. End-Cap Effect:
The pressure differential across the pipe wall induces an axial loading which give rise to
a longitudinal strain in the pipeline as shown in Figuie SS:

The effect can occur at closed end of a pipeline or at a bend. The strain due this can be calculated
as stated below:
e
1
I
o
I
t e
cnd
P
c

Figure 32: Thermal strain effect
Figure 33: End Cap Effect
From the end-cap force which is given as

F
end
= P
- P
c
A
c
=
P
n
1
2
4
-
P
c
n
2
4

s
end
=
F
end
A
s
E

wbcrc tbc cross -scctionol orco, A
s
is ossumcJ to bc, A
s
= n
t
Taking =
, then the end-cap strain becomes:

e
cnd
=
(P
- P
c
)
4tE

b. Poisson Effect:
The internal pressure induces hoop stress and corresponding circumferential strain in the
pipeline. The hoop expansion causes a longitudinal contraction of the pipe, i.e. the pipe expands
in the hoop direction and the Poisson effect results in an axial contraction (opposite to end cap
pressure effect) (Subsea7, 2011).The radial expansion created by this effect will cause a
longitudinal contraction of the pipe as shown in Figuie S4:

The strain due the Poisson effect can be calculated as given by (Yong and Qiang, 2005):
e
posson
= -
:o
H
E

Assuming no effect of external pressure, the total longitudinal strain is given as:
s
|ung|tud|na|
=
T
+e
cnd
+e
posson
= oI +
P
4tE
-
:o
H
E
= oI +
o
H
2
(1 -2:)
E

3.1.2 Frictional Strain and Force
Pipeline resting on seabed experiences a frictional resistance between the soil and the pipeline
outer surface, and the relationship between them is given by the Coulomb relationship. For a
t
P
P
c

Long. Contraction
Hoop Expansion
Figure 34: Poisson Effect
0bele Ifenna Isaac 0niveisity of Stavangei Page Su
pipeline that do not penetrate the seabed, the Coulomb friction relation is applicable and
appropriate (Prof. Sharma).
The frictional strain of a pipeline resting on seabed is given as:
e
]cton
=
p. w
submcgcd
. I
uncho
A
s
E

Where: I
uncho
= oncbor lcngtb
w
submcgcd
= Pipclinc SubmbcrgcJ wcigbt
From the above expression, the frictional force that will be experienced between the soil and
pipeline is given as:
N
r|ct|un
= p. w
submcgcd
. I
uncho
(4)

3.1.3 Effective Axial Force
During pipeline expansion, the combined driving axial force that must be counteracted in order
to avoid end expansion is the effective axial force (Fyrileiv and Collberg, 2005). The effective
axial force increases from pipeline end until it reaches its maximum at the point of full axial
constraint.
According to (Fyrileiv and Collberg, 2005), the effective axial force governs the structural
response of the pipeline by influencing on lateral and upheaval buckling, anchor forces, end
expansion and natural frequencies of free spans.
Considering the external and internal pressures, the effective axial force can be calculated by the
following relation (Fyrileiv and Collberg, 2005):
N
c]] (x)
= N -P
+ P
c
A
c
. (S)
N = Iruc woll Forcc
But, P
- P
c
A
c
= N
cndcup
= EnJ Cop Forcc
Therefore,
N
c]] (x)
= N -N
cndcup

Taking installation barge tension into consideration, the barge tension,F
It
as shown in Figuie SS
below is given as:
F
It
= N + P
c
A
c
= E. (6)
0bele Ifenna Isaac 0niveisity of Stavangei Page S1
Figure 35: Illustration of Lay Tension induced during Pipe laying (Fyrileiv and Collberg, 2005)
During operation, the true axial force if fully constraint goes into compression as a result of the
thermal expansion (-EA
s
oI) and tension due to the hoop stress and Poissons effect
(:o
E
A
s
) (Fyrileiv and Collberg, 2005).
Therefore, from equation (5), the true axial force after installation becomes:
N = E - P
c
A
c
+:o
E
A
s
- A
s
EoI
N = E - P
c
A
c
+ :A
s
P
2t
-A
s
EoI . . (7)
Equating (5) and (7), the effective axial force becomes:
N
c]]
= E - P
+ :A
s
P
2t
- A
s
EoI . . (8)
In general, the effective axial compressive driving force of a pipeline under full axial constraint
is given as:
N
c]]
= A
s
EoI - :A
s
P
2t
+ P
-E. . (9)
Where: P
= Intcrnol prcssurc Jicrcncc rcloti:c to os loiJ, onJ A

s
= n
t

3.1.4 End Expansion
The amount of expansion induced at the ends can be calculated using the strain balanced method.
This is done by integrating the strain between the free ends and the virtual anchor points. The
longitudinal strain is the difference between the applied axial force and the frictional force
induced by soil-pipeline interaction.
H
This combined driving axial force required to fully constrain the pipeline is as a result of the end-
cap effect, Poissons effect, thermal and residual lay tension. For design purposes, according to
(Subsea7, 2011), the residual lay tension, may be assumed to be negligible. Therefore, the
combined driving axial force (Effective Axial Force) that causes end expansion is given as:
N
e (x)
= H- |N
end-cap (x)
+ N
pu|xxun (x)
+ N
therma| (x)
]
This force is also known as the Anchor Force.
Where:
EnJ cop Forcc: F
end-cap (x)
= P. A
|
.(Subsea7, 2011)
P = P
IocuI (x)
-P
cxtcnuI (x)

P
IocuI
= Iocol Jcsign prssurc = P
dcsgn
+p
contcnt
gb
p
contcnt
= Jcnsity o tbc contcnt
b = Icrticol Jistoncc bctwccn Jcsign prcssurc rccrcncc locotion onJ
clc:otion o intcrcst
Forcc Juc to poisson, N
pu|xxun (x)
= -uo
H
A
x
= -P. u. A
x

= -u
P.
. A
x
2t

Forcc Juc to tbcrmol, N
therma| (x)
= EA
s
oI
Neglecting the lay tension the combined driving axial Force that causes end expansion is given
as:
N
c]] (x)
= N
cnd-cup (x)
+ N
posson (x)
+ N
thcmuI (x)

N
c]] (x)
= EA
s
oI +P. A
-:
P.
. A
s
2t

A
s
= n
t
A
=
n
2
4

N
c]] (x)
= EA
s
oI +
P. n
2
4
(1 -2:) . . (1u)
0bele Ifenna Isaac 0niveisity of Stavangei Page SS
Figure 36: Pipeline End Expansion (Cheuk, 2007)

The maximum pipeline end expansion is consequence of the net longitudinal strain and frictional
force between the pipeline and the seabed (Offshorevn, 2010).
The expansion can now be deduced as:
o = _ e
nct
. JI
vAP
0

o = _ (
cndcap
+ e
poison
+e
tbcrmol
-e
riction
). JI
vAP
0

o = _
(N
cndcap
+ N
poson
+ N
thcmuI
- N
]cton
)
EA
s
JI
vAP
0

In general,
o = _
(N
c]] (x)
-N
]cton (x)
)
EA
s
JI
vAP
0

As stated in (Subsea7, 2011), by taking an anchor point at the hot end and the cool end of the
pipe, the expansion at the hot and cold end are given by the following expressions:

o
H01
= _
(N
c]] (x)
-N
]cton (x)
)
EA
s
KP
0
Lu (x1)
. Jx

o
C0L
= _
(N
c]] (x)
- N
]cton (x)
)
EA
s
Lu (x2)
KP
n-1
. Jx

Where:

o
H01
= bot cnJ cxponsion
o
C0L
= CoJl cnJ cxponsion
N
c]] (x)
= Resultant effective axial foice
N
]cton (x)
= Fiiction iestiaint along full beam
Io (x1) = bot cnJ oncbor point
Io (x2) = colJ cnJ oncbor point
KP
0
= Pipclinc stort Kilo Point
KP
n-1
= EnJ o Pipclinc KP
VAP=Virtual anchor point

4. Theoretical Background - Lateral Buckling

4.1 Lateral Buckling and its Mechanism
We have seen that compressive forces are induced in pipelines by the effect of temperature,
pressure and other effects when the pipeline is restrained either by soil friction, rocks or anchors.
Once these compressive forces exceed the critical buckling force of the pipeline, the pipeline will
tend to buckle either upward, laterally or at an angle to release the excessive inbuilt forces.
(Sriskandarajah et al., 1999) defines lateral buckling as the theoretical buckling state that
occurs under axial compressive loadings accompanied by gradual sideways movement as the
pipeline breaks-out.
The driving force for the lateral buckling behavior is due to effect of the effective axial
compressive force generated by thermal expansion, end-cap effect, Poissons effect of hoop
expansion and the residual lay tension. At full axial constrain, the effective axial compressive
force is given as:
N
c]] (x)
= EA
s
oI +P. A
- :
P.
. A
s
2t
- E
Where:
N
c]] (x)
= Eccti:c oxiol comprcssi:c Forcc
E = HoJulus o RigiJity
A
s
= Cross scctionol orco o tbc pipc woll
o = Cocicicnt o lincor cxponsi:ity
I = Icmpcroturc cbongc
P = Prcssurc Jicrcntiol
A
= Arco boscJ on pipc intcrnol Jiomctcr

: = Poissons rotio
= Intcrnol Jiomtcr
E = RcsiJuol loy tcnsion
4.2 Lateral Buckling Modes
Pipeline buckles when the effective force reaches the critical buckling load. According to (Kaye,
1996), these buckles can either be in symmetric or asymmetric mode. Axis of symmetry here
refers to axis drawn through the center of the buckle and normal to the initial centerline of the
pipeline (see Figuie 41)
Figure 41: Symmetric and Asymmetric buckle modes (Kaye, 1996)

The actual buckle modes formed depend on the pipeline horizontal out-of-straightness and the
seabed features.
(Hobbs, 1984) conducted some experimental work for investigating offshore pipeline buckling
behavior by assuming that the pipeline remains elastic with no initial imperfection. It was
concluded that pipeline can buckle into different lateral mode shapes as shown in Figuie 42.
Figure 42: Lateral Buckling modes (Kaye, 1996)

The buckle is considered as sequence of half waves which arises due to inability of the soil
friction to provide a concentrated lateral force required for equilibrium. The amplitude of the half
wave decreases with increasing distance from the center of the buckle.
4.3 Feed-in- Zone
In order to release the excessive compressive force induced in a pipeline, a buckle is formed.
Once the buckle is formed, the compressive forces in the buckle drops and the total length of
pipe within the buckle region will be greater than the initial pipe length over the same section.
This means that there is movement of slip-length from the zone with higher compressive force to
the zone with lower compressive force generated in the buckle section.
As the temperature in the pipeline increases the slip length will therefore continue to feed-into
the buckle after the buckle has been developed (Kein et al.).
The length of the feed-in- zone depends on the available frictional resistance that resists the feed-
in. If a buckle sections loses its ability to carry additional axial load, it is subjected to excessive
lateral deformation and failure of the pipeline (Ahmed and & Gareth, 2012). This can be seen
from the feed-in behavior of the most critical mode shape, which is mode 3 type of buckling
shown in Figure 4-3 (also see Figuie 42).
4.3.1 FeedinLength and Maximum Allowable Moment

Feed-in is the expansion into or the release of stored energy of the pipeline into a buckle formed
as a result of thermal and pressure loading. The amount of pipeline length that is fed-in into the
buckle can be deduced from the relation given below:
I
]ccd-n
= e
c]]cctc uxuI stun
I
BuckIc

The maximum allowable feed-in length shall be established for displacement controlled
condition on the basis of the relation given below
e
mux
e
cu

Where
e
mux
= Hoximum rcportcJ comprcssi:c oxiol stroin in pipclinc
e
cu
= Allowoblc comprcssi:c stroin in pipclinc
Figure 4-3: Feed-in-Zone for Mode-3 Buckling (Ahmed and & Gareth, 2012)
The allowable compressive strain can be deduced from the criterion of displacement controlled
condition (strain based) which is given in (DNV-OS-F101, 2012):
e
cu
=
u.78
y
s
SNCF
_
t
st
0
st
- u.u1] _1 + S.7S
P
mn
-P
c
P
b
] o
h
-1.5
o
gw
eR
c

Where:
e
cu
= tbc ollowoblc comprcssi:c stroin in tbc pipc,
SNCF = Stroin Conccntrotion Foctor,
y
s
= rcsistoncc stroin octor,
P
mn
= minimum intcrnol Jcsign prcssurc,
P
c
= cxtcrnol prcssurc Juc to moximum wotcr JcptE,
P
b
= tbc burst prcssurc or uncorroJcJ pipc scction . scc Equ. S.8 NI F1u1,
o
h
=
R
t0.S
R
m
yiclJ tcnsilc strcngt rotio ,
o
gw
= girtb wclJ octor,
eR
c
= ollowoblc comprcssi:c stroin rcJuction octor,
If the maximum feed-in exceeds the allowable feed-in, local buckling may occur and the pipeline
integrity could be lost. During design for lateral buckling, it is recommended to do a check for
the local buckling and pipe integrity.
Based on (DNV-RP-F110, 2007), local buckling check can be performed against two acceptance
criteria; one is a criterion for load controlled condition (bending moment) and the other is a
criterion for displacement controlled condition (strain based ).
However, Due to the relation between applied bending moment and maximum strain in pipes, a
higher allowable strength for a given target safety level can be achieved by using a strain-based
criterion rather than a bending moment criterion (Sren and Yong, 1999).
The limiting parameter for a pipeline subjected to combined pressure, longitudinal force and
bending moment due to effect from the installation, seabed contours and HPHT operating
conditions is the Bending moment capacity (Sren and Yong, 1999).
The maximum bending moment can as well be used based on the load displacement criteria to
deduce the maximum feed-in into the buckle.
In this thesis work, the strain based criteria will be used in this thesis to establish the maximum
allowable feed-in length.
4.3.2 Hobbs Predictive Analytical Method

In pipeline design for lateral buckling, an analytical method is always used to deduce the
susceptibility of pipeline to lateral buckling. Hobbs method is the analytical method recognized
by DNV standard for this analysis.
The reason for conducting the analytical approach is to:
Determine if a pipeline is susceptible to lateral buckling;
Determine which area of the pipeline is susceptible;
Predict the positioning of the initial out-of-straightness that will be appropriate to trigger a
sufficient thermal buckle along the pipeline.
There exist several analytical methods but Hobbs method is the most widely used
(Sriskandarajah et al., 1999). Hobbs analytical method is based on force equilibrium and
displacement compatibility after a lateral buckle has formed in the assumed straight pipeline (see
Figuie 44).
Figure 44: Lateral Buckling configuration (EINSFELD et al., 2003)

According to (Hobbs, 1984), the governing linear differential equation for the deflected shape of
a pipeline in lateral mode is given as:
y
ii
+
P
EI
y +
p
L
wI
8EI
(4x
2
- I
2
) = u . . (11)
The method deduces the configuration of the buckle with the relation that the increase in arc
length (i.e. reduction in axial force in the buckle) around the buckle must be equal to the axial
feed-in from the slip zone. According to (Hobbs, 1984), using the constants from Table 41, the
reduced effective axial compressive force, P, within the buckle is given as:
P = k
1
EI
I
2

wbcrc: P = Eccti:c comprcssi:c oxiol orcc witbin tbc bucklc
Recall and compare with Equation (4), the Critical Buckling Load derived previously as:
L
pw pw
P
cr
=
n
2
n
2
LI
L
2
. . n = moJc numbcr
The maximum amplitude of the buckle is given as:
y = k
4
p
L
w
EI
I
4

The maximum bending moment H is:
H = k
5
p
L
wI
2

While using the assumptions, (Hobbs, 1984) derived the final relation for the configuration of
the buckle between the effective axial force N
c]]
, at full constraint and the buckle length, L as:
N
c]]
= P + k
3
p
u
wI __1 +k
2
EAp
L
2
wI
5
p
u
(EI)
2
-1.u_ . . (12)
This is the equilibrium relations in terms of buckle length and the fully restrained axial force.
Where:
P = Eccti:c comprcssi:c oxiol orcc witbin tbc bucklc
p
u
= Cocicicnt o oxiol riction
p
L
= Cocicicnt o lotcrol riction
w = SubmcrgcJ wcigbt o tbc pipclinc
I = Bucklc Icngtb corrcsponJing to N
c]]

E = moJulus o closticity
I = ScconJ momcnt o orco
The constants for the lateral buckling modes according to (Hobbs, 1984) can be found in Table
41.
Table 41: Constant for Lateral Buckling Modes (Hobbs, 1984)

Noues K
1
K
2
K
S
K
4
K
S

1 80.76 6.391 X 10
-5
0.5 2.407 X 10
-3
0.06938
2 4a
2
1.743 X 10
-4
1.0 5.532 X 10
-3
0.1088
3 34.06 1.668 X 10
-4
1.294 1.032 X 10
-2
0.1434
4 28.20 2.411 X 10
-4
1.608 1.047 X 10
-2
0.1483
4a
2
4.7050 X 10
-5
4.4495 X 10
-3
0.05066

Finally, the equilibrium solution of Hobbs critical force can be obtained either by an iterative
process or graphical means which will determine the appropriate value for the buckle length for
fully restrained axial forceN
c]]
(Kaye, 1996).
Graphically, the minimum of the obtained curve gives the lowest effective axial forceN
c]]
, at
which a buckle is stable. According to (Kaye, 1996), This is the SAFE force which defines the
maximum theoretical force which can be withstood without risk of buckling.
The above analysis can as well assist in deducing the axial feed-in on either side of the buckle, at
full constraint given by (Kaye, 1996) as:
= _e. Jx
L
=
(N
c]]
- P)
2
2pwEA

Since the axial feed-in from the outside the buckle must be equal to the increase in Length of the
pipe inside the buckle. Based on compatibility of displacement (Kaye, 1996) and from equation
(12), the increase in length inside the buckle 2 is given as:
2= k
2
k
3
2
I
7
_
p
L
w
EI
]
2
- 2k
3
I
(N
c]]
- P)
EA
. . (1S)
The first term, k
2
k
3
2
I
7
[
L
w
LI

2
describes the axial displacement in the pipeline and the second
term describes the tensile elongation due to the reduced compressive force within the buckle
(Kaye, 1996).
Hobbs predictive analysis stated above is based on certain assumptions such as:
Perfectly Straight Pipe (no initial imperfection);
Single Isolated buckle formation (No multiple buckle formation);
Long pipeline having tendency for full axial constraint.
4.3.3 Pipeline Initial Imperfection

Pipelines are initially manufactured or designed to have straight configuration and hence will not
buckle, either laterally or vertically. For a perfectly straight pipeline, its failure mode will be by
material yielding under axial loading (Sriskandarajah et al., 2001). According to newtons first
law of motion, a body remains at rest or in motion with a constant velocity unless acted upon by
an external force. Similarly, in this thesis work, it can be reformulated that A perfectly straight
pipeline will continue in the state of straightness and will not buckle provided there is no initial
imperfection and pipe material yielding. The initial imperfection in pipeline is also referred as
pipeline out-of-straightness (OOS).
The out-of-straightness has a great effect on the critical buckling force at which buckling will
occur. The exact critical buckling force that must be triggered for the onset of buckling depends
on the magnitude of the pipeline initial out-of-straightness (Kaye, 1996).
For small OOS, the effective axial force easily reaches the bifurcation point and the lateral
deflection increases rapidly as shown in Figuie 24: Loau Response in buckling (Robeit) (snap
buckling).
For larger initial imperfection, there is no bifurcation point and the deflection increases
continuously as the axial force increases.
In reality, pipelines resting on the seabed will have a small imperfection and this could be either
lateral or vertical. These imperfections could arise from different operational scenarios:
Installation vessel motion during pipe-lay:
The barge lateral sway motion during pipe-lay and anchor handling/slip can introduce
horizontal out-of-straightness (Lateral initial imperfection).

Seabed Imperfection (Uneven Seabed):
Seabeds are not naturally straight and will affect the outcome of pipeline configuration
during and after installation. This could be as a result of presence of rock, the topography
and soil property.

Third Party Activity:
Interference from fishing activities i.e. pullover of trawl beams and doors or anchor
hooking or dragging can result in horizontal out-of-straightness (Sriskandarajah et al.,
2001).

4.3.4 Virtual Anchor Spacing

This is spacing generated between the virtual anchors of an isolated lateral buckle in a pipeline.
Virtual anchor point (VAP) is the apparent fixity point in a pipeline at which the effective
driving force equals the soil frictional force.
Figure 45: Force profile along pipeline showing Virtual anchor spacing and reduction in force at each
lateral buckle (Bruton and Carr, 2011)
During buckle initiation, lateral buckle are introduced to share the axial loading between the
buckle such that the loadings are distributed along the pipeline instead of concentrating on a
particular buckle site. Between this buckle virtual anchors are formed due to soil friction and the
effective axial force in the pipeline. The virtual anchors are the points where the axial feed-in
occur in each buckle.
The Virtual anchor spacing defines the feed-in length which determines the amount of axial feed-
in into the buckle with respect to the available soil resistance at the feed-in zone.
4.4 Snake-Lay Control Mechanism
Pipelines that are normally laid and exposed on seabed are naturally allowed to move. If the
induced compressive axial forces are higher than the buckling force, it will buckle sideways.
Uncontrolled buckling can lead to loss of pipeline integrity as a result of the limiting state failure
modes.
According to (DNV-RP-F110, 2007), buckling can be controlled by sharing the expansion into
buckles at regular intervals along the pipeline route. The controlled buckles must be located such
that the resulting axial feed-in into each buckle does not exceed the limiting condition for the
displacement controlled criterion.
As discussed previously, lateral buckles can be triggered using buckle initiators at selected
locations. As mentioned before, the buckles can be initiated by several methods such as the use
of mid-line spool, vertical triggers, rock dumping and snake-lay.
Figure 46: SnakeLay configuration (Subsea7, 2012)

This report considers the snake-lay technology in initiating appropriate buckle at the planned
location to control lateral buckling that could result from the axial compressive force induced in
the pipeline. This method utilizes counteracts which are pre-installed on the seabed and the
pipeline is snaked between them.
The main idea of the snake-lay configuration is to introduce deliberate horizontal lay
imperfections to trigger sufficient thermal buckles at a pre-determined location along the pipeline
such that the resulting thermal expansion is distributed among the possible buckles formed rather
than concentrating the axial feed-in at some few buckle sites(Rundsagetal.,2008).
The snake lay pattern is created by the deviation of the lay barge from its nominal route and by
the use of counteracts. The created curvature created makes the crown of snakes to act as a large
curvature expansion spool and the pitch determines the axial feed-in at the crown (Kein et al.).
Snake-lay is one of the less expensive means of controlling lateral buckling by buckle initiation.
It does not have any issue of span introduction that could lead to vortex induced vibration and
dangers of trawl loading like the other methods. The major challenge of this method is the
technique that will be employed to create the buckle spacing, which will ensure that buckles are
initiated at desired locations. This could be a major problem on softer clay due to the high
breakout force that is associated with the soil.
4.4.1 Lay Radius Imperfection
Lay radius is the radius of curvature created on the pipeline as a result of its initial imperfection
or a deliberate radius created to initiate buckling at that location.
The aim of snake-lay method is such that the thermal buckling could occur at pre-determined
location. The critical buckling load of a pipeline section that must be exceeded to form buckle
depends largely on the lateral initial imperfection. In other words, the critical buckling load
depends on the lay radius of the buckle curvature.
Figure 47: SnakeLay with Lay Radius Imperfection (Bruton and Carr, 2011)
According to (Rundsag et al., 2008), large lateral initial imperfection which means a small lay
radius is normally preferred since it allows buckling at lower effective compressive force which
gives a higher probability that intended buckle will occur at the pre-determined location such that
the required distribution is achieved. The lower effective axial force associated with small lay
radius could also yield a gentle buckle than the radius with a higher effective force (Rundsag et
al., 2008).
In order to trigger this buckle at larger initial imperfection (i.e. lower Critical buckling force), the
lay radius should be as small as possible provided the design limit state conditions are not
compromised. The critical buckling force P
c
which determines the onset of buckling is related
to the lateral imperfection (Lay radius, R) and lateral resistance p by the following equation:
P
c
= p
L
w
submcgcd
. R
Figure 48: Critical Buckling Force Against the lateral soil friction (Rundsag et al., 2008)
P
c
P
c
This equation suggests that the pipeline will buckle laterally when the axial induced compressive
stress exceeds the soil resistance force.
Therefore, since the curved section of the snake-lay is an arc, the size of the imperfection of the
curved radius is equal to the inverse of the radius.
Hence, the reliability of buckle formation can be increased by reducing the lay radius or
increasing the length of curvature in each snake.
Therefore, the main thing in the design of snake lay is to establish a minimum Lay radius for an
appropriate breakout force.
4.5 Design Limiting Criteria

Limit state is the condition beyond which a structure is deemed unsafe and no longer fulfills the relevant
design criteria. The limit state that is usually applicable for lateral buckling effect on pipelines
could range from the list below according to (DNV-OS-F101, 2012):
Local Buckling limit state:
The longitudinal compressive strain induced by the effective axial compressive force
along the pipeline is displacement controlled, so the strain based design criteria based on
requirement of (DNV-OS-F101, 2012) has been used for the local buckling failure mode.

Fatigue Limit state:
High temperature and high pressure pipelines are subjected to low frequency-high
amplitude loading mainly from the startup-shutdown cycles and pipeline installation. The
fatigue response of the girth weld is usually very critical in pipeline design with respect
to this high stress-low cycle regime.

Weld fracture:
Fracture analysis could be carried out to examine accurately any undetected crack within
the ranges as directed by (DNV-OS-F101, 2012) and other relevant standards.

Trawl Gear interaction:
The impact of trawl gear on buckled section could also be considered based on the
environment impact and the area of operation and water depth.

5. Theoretical Background - Pipeline Walking

5. Pipeline Walking
5.1 Pipeline Walking and its Mechanisms

The build-up of effective axial forces in pipelines do not only result to lateral buckling but could
also lead to expansion of the pipeline ends when there are not enough constraint to hold back the
axial forces.
As pipeline is heated up during operation, it will tend to expand but the expansion will be
restricted by the soil friction thereby creating axial compressive movement to the ends along the
pipeline. During shutdown (cooling), the pipeline will tend to contract but due to the seabed
friction the pipeline will not be able to return back to its initial position.
Under symmetrical loading condition (i.e. when there is no differential tension between the ends
of a straight pipeline and uniform heating and cooling operation), pipelines will expand and
contract uniformly resulting in movement of the ends with respect to the virtual anchor at the
middle of the line. This will cause the pipeline ends to move in and out while the pipeline
remains in its initial position. This end movement can be managed by the use of jumpers, spool
and sliding mechanism (Chaudhury, 2010).
For asymmetrical loading (i.e. if there is an unbalanced axial force or non-uniform heating and
cooling operation), the expansion and contraction at the two ends will move non-uniformly along
the pipeline resulting in a shift of the virtual anchor points of the two ends.
Figure 49: Schematic View of Pipeline walking due to separation of Virtual Anchor (Chaudhury, 2010)
According to (Chaudhury, 2010), In a heating and cooling cycle operation of a pipeline, if there
exist a shift between the two virtual anchor points, an unbalanced axial force is generated in
between the virtual anchor points resulting in a small amount of movement of the entire pipeline
if the pipeline is fully mobilized (i.e. not enough frictional resistance to reach full constraints
(e.g. short pipeline)). This is known as WALKING.
Heating Mechanism:
The heating process occurs from the hot end (well) towards the cold end (e.g. SCR) in several
transients at constant gradient along the pipeline until a steady state point is attained. During the
first heating and cooling process, compressive axial force builds up from each transient at
constant thermal gradient. As the pipe expands during heating process a virtual anchor will be
formed at the mid-line when the pipeline become fully mobilized and also during cooling another
virtual anchor will be formed at the midline as the pipe contracts. This is shown in Figuie 41u
with the red line for heating and blue line for cooling, and the other colors represents the
transient stages.
Figure 410: Force Profile during the first and second heating stage (Carr et al., 2006)
As the heating continues, the pipeline expands after each transient making the remaining portion
of the pipeline to remain at colder temperature. This scenario causes a non-uniform expansion
of the pipeline creating a virtual anchor from A1 to A7 for each of the transient. Due to the
expansion and movement, subsequent virtual anchors are created from B1 to B7 to maintain an
equilibrium state as shown in Figuie 41u.
Cooling Mechanism:

These cycles of plastic expansion and contraction could range for several start-ups and shut-
downs and will result in the accumulation of axial displacement along the pipeline.
According to (Carr et al., 2006), Pipeline walking is a phenomenon that can occur in short,
high temperature pipeline that does not reach full constraint in the middle, but instead expands
about a virtual anchor point (VAP) located at the middle of the pipeline.
Alternatively, the pipelines unloads (cooldown) at a uniform rate and forms a virtual anchor at
the mid-line. This symmetrical cooling ensures that there is no reversal of the displacement of the
pipeline, hence the shift that occurred during heat-up will not be recovered (Carr et al., 2006). As
this occurs in each cycle, the pipeline walks towards the cold end.
The phenomenon can also occur in long pipeline (pipeline that develops full constraint force)
where lateral buckling has already occurred.
Figure 411: Illustration of Pipeline waking and its contributory factors (Bruton and Carr, 2011)
As shown in Figuie 411, pipeline walking can occur in one of the following conditions:
Short pipeline with steep transient thermal gradient (axial movement towards the cold
end);
Pulling force at the end of a flowline, associated with a SCR (moves towards the pulling
end (SCR));
Seabed slope along the pipeline (downward movement);
Multiphase flow behavior during startup and shut-down operations.
Pipeline walking is not a limit state but under no critical examination could lead to the
following undesired situations:
Overstress of the spools/J umpers due to end expansions;
Loss of tension in the Steel catenary Riser (SCR);
Increased feed-in within lateral buckle.
Initiating factor in pipeline walking is the tendency for short pipeline to become fully mobilized
and not reaching full constrain and this depends on the axial friction force and non-uniformity of
the heating and cooling operation.
5.2 Steep Thermal Gradient
As shown in Figuie 412, the transient thermal gradient down the slope increases the expansion
towards the cold end as the pipeline weight acts in that direction.
The driving force down the slope is a function of the weight and the slope angle, which result to
axial walking down the hill during heating and cooling. This can be seen by the analysis
conducted by (Carr et al., 2006).
0bele Ifenna Isaac 0niveisity of Stavangei Page Su
Figure 412: Force Sloping Seabed (Carr et al., 2006)

5.3 Pulling Force at SCR
For pipelines attached to the steel catenary riser (SCR), the pipeline will experience a pulling
force at the connection point due the induced force in the SCR as shown in Figuie 41S. This
constant tension can build up walking effect during heating and cooling process.
Figure 413: Schematic of PipelineRiser tension at the end connection

The tension applied by the SCR creates asymmetric force profile and hence result in movement
of virtual anchors towards the SCR during heating and towards the pipeline during cooling but
the asymmetric force profile causes walking towards the SCR as stated by (Carr et al., 2006).
5.4 Steep Thermal Gradient
During start-up of an operation, there is always a temperature change as the flow moves from hot
end to the cold end. As this happens, there exists a resultant transient thermal gradient from the
hot to the cold end until the operation reaches a steady state.
The transient thermal gradient builds up expansion after each cycle of heating and cooling
operation. As shown in Figuie 414, due to the heat lost to the environment the transient gradually
reaches a steady state.
Figure 414: Thermal Transient

5.5 Interaction between Pipeline Walking and Lateral Buckling

The lateral buckling of an exposed subsea pipeline could provide avenue for the line to walk
under certain conditions of operational loads and pipe soil interaction. Lateral imperfections,
seabed imperfection and other disturbances (e.g. trawl) will always create such avenues. When a
pipeline becomes fully mobilized, pipeline walking is likely to occur if it is not controlled. The
interaction between lateral buckling and the pipeline walking should be analyzed and it is based
on the type of pipeline described below:
Short Pipeline:

A short pipeline does not reach fully restrained condition but the compressive force may exceed
the lateral buckling limit for certain combination of axial and lateral friction coefficients and this
could result in lateral buckling. The pipeline may also be susceptible to walking under thermal
transients and/or seabed slope. This is shown in Figuie 41S.
Figure 415: Effective axial force in a short pipeline

Long Pipeline

As shown in Figuie 416, a long pipeline will reach a fully restrained condition and the
compressive force is likely to exceed the lateral buckling limit. The lateral buckling tendency
may be mitigated by introducing predefined buckling locations (e.g. laid on sleepers) and hence
the long pipeline is effectively split into several short pipelines with each buckle acting as a
pipeline end.
These short pipelines may be susceptible to walking under thermal transients and/or seabed
slope.
Figure 416: Effective Axial force for long pipeline

6. Design Methodology
6.1 General
The design for lateral buckling and axial walking of surface laid subsea pipeline is carried out
according to Subsea7 design guideline for lateral buckling and axial walking (Subsea7, 2012)
with respect to DNV OS F101, DNV RP F110 and SAFEBUCK guideline.
According to the SAFEBUCK guideline and DNV RP F101, during pipeline design analysis
prior to installation, pipelines shall be checked for its susceptibility to buckling.
If the pipeline is found to buckle under thermal and pressure loading, a control measure of
deliberate lateral buckles will be introduced at predetermined location along the pipeline such
that the total effective compressive causing the buckling will be shared among the buckles along
the pipeline.
6.1.1 Design objective
The thesis objective is to study and understand the influence of pipeline-soil interaction on the
design of surface laid subsea pipeline that is susceptible to lateral buckling and pipeline
walking. The main focus is on:

(DEH) to reduce rate of walking
The aim is to limit pipeline expansion at the connecting ends by using snake-lay design with
intermittent rock dumping.
The acceptability of snakes as engineered buckle will be verified based on lateral buckling
criteria by performing a design check in accordance with DNV RP F110.
This is done using local buckling check (displacement control criteria) which is the main criteria
to obtain the allowable virtual anchor spacing

The work shall include performing analytical investigations and verifying them with a non-
linear finite element analysis (FEA) by modeling the soil-pipeline interaction of as-laid pipeline
using general finite element software ANSYS.
Secondly, the effect of thermal gradient in pipeline walking phenomenon is investigated and the
use of direct electric heating (DEH) system to reduce axial walking shall be discussed based on
FE analysis results.
6.1.2 Design Assumptions and Requirement

Requirement
The critical buckling force shall be taken as equal to the minimum value between the
Hobbs Force and the force associated with pipeline OOS;
N
ctcuI
= min (N
00S
, N
H0BBs
) . (SAFEBUCK-J IP, 2011)
Design feed-in Length (I
]ccd-n
)for each buckle shall be equal to the maximum feed-
in length into the buckle that will not cause pipeline failure under all limiting states;
I
]ccd-n
= e I
buckIc

The limiting states are the allowable axial strain and maximum bending moment in the
pipeline;
The probability of maximum feed-in length exceeding the design feed-in shall be less
than 1u
-4
(DNV-RP-F110, 2007);
Axial friction force during pipeline walking shall be less than the force associated with
the thermal gradient: ( <
0
);
Rock dumping at the hot and cold end shall b used to reduce end expansion while
applying snake-lay configuration.

Design Assumptions
Lateral Buckling
Hobbs analyses for lateral buckling is based on a straight pipe with no initial out-of
straightness (OOS);
The effect of the hydrodynamic forces (drag force, lift force and Inertia) are not
considered;
The temperature profile is assumed to be exponentially distributed along the pipeline;
The capacity of the pipeline shall be calculated in accordance with DNV-RP-F110 which
is expressed as the allowable feed-in length (how much the pipe is allowed to expand into
a buckle).

Pipeline Walking
The heating steps during pipeline walking analyses for FE method and analytical method
utilized linear thermal transient;
Temperature profile with constant gradient is used for the analytical calculation and FE
analyses of pipeline walking;
The pipeline is fully mobilised (short pipeline);
No pressure variation at each heatups and cooldown;
Only the effect of thermal gradient is considered.

6.2 Design Process and Roadmap - Lateral buckling

The design process of lateral buckling with respect to the steps outlined in figure 3.1 involves the
following:

As described in Figuie 61, analytical calculation is done based on Hobbs Lateral buckling
analytical method, to determine the critical buckling force that had to be exceeded before
buckling can occur. The pipeline post buckling behavior under cyclic loadings is investigated
with respect to the susceptibility to walking. This will be done using a Mathcad 15 and FE
analyses.
The design road map is discussed as follows:
No
No
Select Design Parameters
Is pipeline
susceptible
to Buckling?
No Assessment Required
Select Initiation method
[Snake Lay Method]
Modify Snake Lay Radius
Yes
Yes
Are key Limit
states
exceeded?
Design Complete
Review Design
Parameters
Yes
Local buckling check (DNV)
Fatigue Check
Evaluate Pipeline walking
Is pipeline
susceptible
to walking?
Select Anchoring method
to control walking
Yes
No
Figure 61: Design steps for Pipeline Lateral buckling controlled using Snakelay configuration
a. Selection of Pipeline Parameters:

The Pipeline parameters are selected based on recommended project which subsea7 is
undertaking for her client. This will involve the seabed data, pipeline data, friction
coefficients and the temperature profile.

b. Analytical Calculation:
This section details the end expansion calculations, verification of pipeline length scale
for short and long pipeline, Hobbs buckling critical buckling assessment, effective axial
compressive force for full constraint pipeline and fully mobilized pipeline.
The results from the above calculations yields the critical force required to buckle the
pipe, the expansion at the ends which the spool or tie-ins will accommodate and the
possible anchor point for the short pipeline. This is achieved using Mathcad 15
calculation software.

c. Finite Element Analysis:
FE analyses is carried out to validate the results of effective axial compressive force and
end expansion calculated obtained from the analytical method. Further, it is used for the
non-linear analysis of surface laid pipeline that is susceptible to lateral buckling.

If the pipeline is susceptible to buckling, based on displacement controlled condition,
ANSYS mechanical APDL is utilized to obtain the maximum feed-in corresponding to
the applied pressure, thermal and frictional loading along the pipeline.

This is done using a 2km VAS model, initiating or triggering a lateral buckle by applying
a lateral imperfection corresponding to a given lay radius.

Note: A straight pipe will never buckle if there is no imperfection or horizontal/vertical
force applied to the line that will trigger the buckling along the pipeline.

From the ANSYS model, the equivalent axial strain, bending moment, effective axial
force will be extracted. The computation from above extraction gives the maximum feed-
in into the buckle and the corresponding bending moment. This is used to generate the
number of snakes that will be pre-installed to control/share the effective axial
compressive force built up by the thermal and pressure loading during operation.

Expansion sharing is done to control excessive lateral deformation of the pipeline which
could lead to pipeline failure or fracture if not mitigated.

d. Snake-Lay Configuration
Snake lay involves the introduction of horizontal-lateral imperfection into the pipeline in
form of curves having a given radius of curvature at predetermined location.
This is done to deduce the critical buckle spacing required to prevent the maximum strain
and cyclic strain range from exceeding the allowable design strain obtained from the
DNV- OS - F101.

To address the issue of buckling on even seabed, it has been proved a better option is to
adopt a method that allows buckling to happen provided it could be demonstrated that the
resulting high loads and deformations are acceptable.

To ensure that expansions are shared into pre-installed snakes along the pipeline, the
pipeline capacity need to be determined based on DNV lateral buckling check.

Pipe Capacity:
The pipeline capacity which determines the feed-in capacity is determined based on DNV-OS-
F101. In this thesis work, displacement controlled method (strain based criteria) is used to
deduce the capacity.
The allowable compressive strain can be deduced from the displacement controlled condition
criterion (strain based) as given stated in (DNV-OS-F101, 2012):
e
cu
=
u.78
y
s
SNCF
_
t
st
0
st
- u.u1] _1 + S.7S
P
mn
-P
c
P
b
] o
h
-1.5
o
gw
eR
c

Where:
e
cu
= tbc ollowoblc comprcssi:c stroin in tbc pipc
SNCF = Stroin conccntrotion Foctor
y
s
= rcsistoncc stroin octor
P
mn
= minimum intcrnol Jcsign prcssurc
P
c
= cxtcrnol prcssurc Juc to moximum wotcr JcptE
P
b
= tbc burst prcssurc or uncorroJcJ pipc scction . scc Equ. S.8 NI F1u1
o
h
=
R
t0.5
R
m
yiclJ tcnsilc strcngt rotio
o
gw
= girtb wclJ octor
eR
c
= ollowoblc comprcssi:c stroin rcJuction octor

The maximum allowable feed-in length is established for displacement controlled condition on
the basis of the relation below
e
mux
e
cu

Where:
e
mux
= Hoximum rcportcJ comprcssi:c oxiol stroin in pipclinc
e
cu
= Allowoblc comprcssi:c stroin in pipclinc
By comparing the reported strain which shall be determined from the ANSYS model against the
criteria stated above, the allowable feed-in to the buckle is be established.

From the ANSYS model, the maximum feed-in is determined using the allowable compressive
strain obtained previously. This is estimated with the relation below:

I
]ccd-n
= e I
buckIc

Lay Configuration:

The lay configuration is achieved based on the principle of expansion sharing outlined in DNV-
RP-F110 for sharing into adjacent buckle. Following the steps, it is to be ensured that the
allowable feed-in lengths are not exceeded.

According to the guideline, expansion sharing into adjacent buckle can occur if the following
relations are fulfilled:

F
post
LB
+S
LB
F
ctcuI
0B

Where:
F
post
LB
= tbc lowcr bounJ post - buckling orcc or tbc gi:cn snokc conigurotion

S
LB
= cccti:c oxiol orcc builJ -up bctwccn oJ]occnt boscJ on IB octor

F
ctcuI
0B
= uppcr bounJ criticol orcc or tbc gi:cn snokc conigurotion

Note:
To build up effective force between the buckles and limit the pipeline end expansion without
exceeding the allowable feed-in length in the buckle, snake-lay design and intermittent rock
damping is utilized.

e. Lateral Buckling Check

A local buckling is conducted analytically based on recommended displacement
controlled criteria of DNV RP F110 see Appendix A4.

f. Pipeline Walking Check
Analytical check based on SAFEBUCK guideline is to be carried out to check if the
pipeline is susceptible to axial walking. Also, cyclic loading for about 3 cycles is applied
to investigate walking at the crown of the buckle.
6.3 Design Roadmap for Pipeline Walking Thermal Transient Effect

When pipeline is laid on seabed and heated up by hot fluid content passing through it, it expands
axially and circumferentially. Upon cooling, since this expansion is restricted by axial friction
generated by pipe-soil contact, the pipeline will not return to its original position.

With subsequent heating and cooling, an accumulation of axial displacement towards the cold
end is generated. This is known as pipe-walking. The methodology herein is based on the
SAFEBUCK design guideline which Subsea7 was one of the sponsors.

The analysis considers a 2km flowline under thermal asymmetric heating steps with
considerations mainly based on the effect of walking due to thermal loadings.

Analytical Method
The flowline is checked for susceptibility to walking using an analytical method based on
SAFEBUCK design guideline. If walking is identified, the rate of walking for each transient
cycle will be calculated.

The corresponding axial displacement (rate of walking) will be determined and validated with
the results from Finite element analysis.

Finite Element Method
Once the flowline is confirmed to be susceptible to walking, a finite element (FE) analysis is
conducted using ANSYS v13 to analyze the non-linear walking mechanism. The analysis
involves the incorporation of constant thermal gradient into the FE model and deducing the
corresponding rate of walking.
The effect of thermal gradient is investigated using FE for two (2) different thermal gradients
and comparable the result obtained.
Finally, the need to use of DEH to reduce pipeline walking is investigated and explained using
FE result.

The overall road map can be seen from the Figuie 62 below:

6.4 Design Analysis
This section details the design steps taken to achieve the analytical and finite element solution.
This is done in accordance with the design guidelines and relevant procedure specified above.
6.4.1 Analytical Method
The analytical method in this thesis will be based on the several points: End expansions
calculations and Hobbs critical buckling assessment, pipeline susceptibility to walking and
Snake-lay radius determination.
a. End Expansion:
Assuming a straight-fully constrained pipeline, the end expansion due to pressure (end-cap
and Poisson effect) and thermal loadings is calculated. This deduces the maximum expansion
that the effective axial compressive force will yield which the ends must accommodate.

b. Effective Axial Force The expansions at the ends are built-up because of the effective axial
forces generated from the pressure and thermal loadings when fully constrained at the ends.
Thus, the effective axial force along the pipeline is calculated based on the relation below:
N
P0LL
= N
Lu
- N
Lndcup
+N
posson
+ N
1hcmuI

Figure 62: Design roadmap for Pipeline walking under thermal transient
No
No
Select Design Parameters
Is pipeline
susceptible
to walking?
Define rate of walking
(Analytical method
thermal transient)
Yes
Apply direct electric
heating (DEH)
control Measure
Review Design
Parameters
Yes
Are
displacement
Limits
exceeded?
Pipeline walking Design
solution
This force is sometimes called the Anchor Force. The results of the end expansion reveals if the
pipeline is a short or a long pipeline based on the value of the anchor length obtained.
Recall: if the anchor length obtained is greater than the length of the pipeline then the pipeline is
termed a short pipeline i.e. fully mobilized. If the pipeline is a long pipeline, the two virtual
anchor points will be deduced for the hot and the cold end.
The expansion at the ends is calculated using the below equation based on subsea7 design
guideline:
cxponsionE0I =
(N
P0LL
- N
]ctonuI
)
EA
s
KP0
uncho hot cnd

cxponsionC0I =
(N
P0LL
- N
]ctonuI
)
EA
s
uncho coId cnd
KPcnd

Thus, the minimum spool length required can be calculated with the following relations:
I
mnmum
= _
2.2S
o
cxponsionE0I
o
Bcndng

.See Appendix A
Submerged weight calculation The submerged weight is calculated with respect to the
pipeline datas. This is achieved by using the relation:
w
submcgcd
= contcnt wcigbt +stccl wcigbt +cooting wcigbt -buoyoncy
c. Hobbs Critical Buckling Force Assessment:
The Hobbs critical buckling force analysis is done to determine the susceptibility of the pipeline
to lateral buckling as well as determine the Hobbs critical buckling force. The axial forces
described by Hobbs with respect to several modes (modes 1, 2, 3, 4 and infinity) are used to
obtain the critical buckling forces for the corresponding nodes.
The method traces the equilibrium path in terms of buckle length and fully restrained axial force.
The Effective axial force according to Hobbs is given as:
P
0
= P
buckIc
+ K
3
p
uxuI
w
submcgcd
I
buckIc
__
_
1 +K
2
EAp
Lut
2
w
submcgcd
I
buckIc
5
p
uxuI
(EI)
2
_ -1_
.See Appendix A for details.
The corresponding buckle wavelength is also given according to Hobbs as:

I
buckIc
= _
2.7969 1u
5
(EI
s
)
3
(p
Lut
w
submcgcd
)
2
A
s
E
_
The effective axial P
0
is obtained for all the modes and plotted against buckle
wavelength I
buckIc
. From the graph, the Hobbs critical buckling force will be obtained by the
deducing the minimum value of the axial forces of all the modes plotted.
N
H0BBs
= min (P
0
moJc1, P
0
moJc2, P
0
moJc2, P
0
moJcS, P
0
moJc4, P
0
moJcininity)
The Critical buckling force along the pipeline according to SAFEBUCK guideline is computed
with respect to the equation below:
N
ctcuI
= min (N
00S
, N
H0BBs
) . (SAFEBUCK-J IP, 2011)
Where N
00S
= p
mnmuILut
w
submcgcd
R
N
00S
is tbc Axiol Forcc ossociotcJ witb pipclinc initiol Impcrctction
According to (SAFEBUCK-J IP, 2011), Pipelines is susceptible to buckling if the condition below is
exist:
N
mux
N
ctcuI

Ibc moximum orcc in tbc pipclinc: N
mux
= min (N
PuII
, N
]mux
)
]N
]mux
= p
mux.uxuI
w
submcgcd

L
2
=Maximum axial restriction force for a Short pipeline
N
PuII
= Fully constroincJ oxiol Forcc
The analytical method stated above is based on several assumptions (straight pipeline, no initial
imperfection) which did not meet the actual scenario of pipeline-soil interactions on seabed.
Hence, Hobbs method is generally used to determine the susceptibility of pipeline to lateral
buckling and the critical buckling force.
d. Pipeline Walking
Using the procedure outlined in SAFEBUCK guideline, the analytical solution for pipeline rate
of walking at different thermal gradient is obtained. This is done on Mathcad 15 (see Appendix
A).
According to the guideline, pipeline is not susceptible to walking if the axial friction force [f]
exceeds the following value:
>
(FA
xt
uT)
L

If the pipeline is susceptible to walking, the rate of walking per cycle under constant thermal
gradient will be deduced using the relation below:

0
=
I
2
8 EA
s
--- - -- - -- i

0
6

0
=
I
2
16 EA
s
(24
0
-
0
-4) - - -- - --- -i >

0
6

This gives the accumulated axial displacement as a result of the thermal transient based on
constant gradient along the seabed.
6.4.2 Finite Element Method (ANSYS Mechanical APDL) Lateral Buckling
This is a detail description of the non-linear solution method for lateral buckling of subsea
pipeline laid on even seabed with an initial imperfection. The model is done using ANSYS
mechanical APDL.
Reason for Finite Element Method:
Based on the Hobbs analytical assumptions stated previously, Hobbs critical buckling force
analyses did not account for initial imperfection and post-buckling behavior in the pipeline.
Hence, a non-linear finite element method is necessary so as to determine the effect of initial
imperfection on lateral buckling.
The main reason for a non-linear solution is to overcome the shortcomings that were created
while applying the analytical method. The material modeling, the seabed modeling, the non-
linearity of pipe-soil interaction, the initial imperfection, the temperature profile and the pipeline
boundary conditions which could not be implemented in the analytical method.
The finite element (FE) analysis uses iterative newton Raphson solution method by applying the
loading in an increment with respect to time step (transient analysis). This is done to locate an
equilibrium path defined by the load and nodal variable.
Pipeline Model:
A local model approach (VAS model) of 2km pipeline section with an initial imperfection is
modeled with respect to the material properties of the pipeline. The VAS model is used to
determine the effective axial force, the bending moment and the lateral displacement acting on
the buckled section of the pipeline.
The initial out-of-straightness (OOS) is achieved as a function of lay radius for the snake lay
configuration with respect to the available operational corridor.
i. Pipe Element:
The pipeline is modeled with PIPE288 element, which supports nonlinear modeling of Elasticity,
hyper-elasticity, plasticity, creep, and other nonlinear material models. PIPE288 is a linear,
quadratic, or cubic two-node pipe element in 3-D. The element has six degrees of freedom at
each node (the translations in the x, y, and z directions and rotations about the x, y, and z
directions). The element is well-suited for linear, large rotation, and/or large strain nonlinear
applications (see Figuie 6S).

Figure 63:Pipe288 Geometry and the ANSYS model
PIPE288 supports the thin-pipe (KEYOPT (4) =1) option. The thin-pipe option assumes a plain
stress state in the pipe wall and ignores the stress in the wall thickness direction.
To obtain a better lateral displacement distribution and bending moment, the pipe element is
subdivided into a unit length per element. Activating KEYOPT (6) =0 generates end-caps loads
from the internal and external pressure.
Setting KEYOPT (7) =0 generates output section forces/moments, strains/curvatures, internal
and external pressures, effective tension, and maximum hoop. This enables the maximum feed-
in length and bending moment in the buckle section to be deduced.
ii. Initial Horizontal Out-of-Straightness:
In practice, pipelines laid on seabed will always have an imperfection probably induced due to
uneven seabed, barge motion during installation or by fishing gear activity. A perfectly straight
pipeline without any imperfection will not buckle when subjected to pressure and thermal
loading.
To generate an initial imperfection that will trigger lateral buckling, a pre-determined axial force
associated with pipeline out-of-straightness (OOS) is introduced at some section in the
middle-nodes of the pipeline.

, N
00S
Depending on the Lay-radius required and the buckle amplitude, the pre-load is initiated which
will trigger buckling based on the submerged weight and the lateral friction coefficient as shown
in Figuie 64.

Figure 65: Pipeline showing an Initial OOS of 1m

Lay-Radius Deduction: From the model shown in Figuie 6S, an initial pre-load of 7500N per
node along a section of 40 nodes (node 480 node 520) displaced the pipeline 1m laterally.

Middle node
N
00S
Figure 64: A Preload associated with pipeline OOS
0
R
I
w
b
A
R - b
A
R
I
w
Figure 66: Layradius deduction

Applying Pythagoras theorem, we have: R

2
= (R - b
A
)
2
+ I
w
2

The Lay-radius becomes: R =
h
A
2
+L
w
2
2h
A

Where:
b
A
= initiol lotcrol impcrcction
I
w
= bol wo:clcngtb
In the model, b
A
= 1.um
I
w
= Sum.therefore, the Lay-radius becomes R = 12Sum

iii. Material Modeling:
The non-linearity that is associated with lateral buckling due to the thermal expansion and large
bending moment is modeled into the program to care of the plasticity or yielding of the material
that could occur during loading. In ANSYS, this is inputted using TB command based on the
below stress-strain curve as shown in Figuie 67.
Figure 67: StressStrain relationship in ANSYS model
iv. Boundary Condition:

The boundary condition is obtained by fixing the ends of the pipeline in all direction which
forms a basis from our theory that the pipeline is fully restrained by axial fictional resistance.
The submerged weight of the pipeline is introduced along the nodes to act as the frictional force
resisting lateral and axial movement. This is shown in Figuie 68 below:
Figure 68: Boundary condition showing pipe ends fixed in all direction
Seabed Modeling:
The seabed is modeled on the soil surface along the z-axis using element TARGE170.
TARGE170 is used in ANSYS APDL to represent a rigid target surface for associated 3D node-
to-contact element. The element can be set into a 4 node by applying shape as QUAD using a
TSHAPE command. This is shown in Figuie 69 below.

Figure 69: Seabed Model on zplane
Pipe-Soil Interaction (Contact model):
The submerged weight introduced in the pipeline model creates a self-weight on each of the pipe
element. As a result, the need to create a contact between the pipeline and the seabed
(TARGE170) arises. The contact is modeled using CONTA175 in ANSYS. As shown in Figuie
61u, this element supports the orthotropic coulomb friction, a pipeline-seabed model with both
axial and lateral friction coefficient. ESURF command is used to generate a surface contact in y-
direction of the all the contact nodes between all the pipe-nodes and the associated seabed nodes
(TARGE170).
Figure 610: Model showing the contact element generating friction Force on the Seabed
Loading Sequence:
The model is divided into two (2) sections; the lateral buckling section, the cyclic loading and
unloading section.
a. Lateral Buckling Loading Sequence:
The lateral buckling of the 2km VAS model in ANSYS will be operated/simulated exactly the way
installation and operation of pipelines on seabed is done.
1. Model the Pipeline;
2. Lay Pipeline on even seabed;
3. Apply boundary condition at the ends (fixed end nodes);
4. Apply Submerged weight of the entire content;
5. Applying Initial imperfection (OOS) due to the installation motion of the vessel;
6. Apply internal pressure to retain the initial imperfection (OOS) against collapse from external
pressure effect;
7. Apply external pressure p
wutc
g wotcr Jcptb;
8. Remove lateral force that generated initial imperfection;
9. Applying operating pressure and temperature.

b. Cyclic loading and Unloading Sequence:
The lateral buckling of the 2km VAS model in ANSYS is loaded the way installation and
operation of pipelines on seabed is done as shown below:
2. Lay Pipeline on even seabed;
3. Apply boundary condition at the ends (fixed end nodes);

5. Applying Initial imperfection (OOS) due to the installation motion of the vessel;
6. Apply internal pressure to retain the initial imperfection (OOS) against collapse from
external pressure effect;
wutc
g wotcr Jcptb;
8. Remove lateral force that generated initial imperfection;
9. Applying operating pressure and temperature First heat-up;
10. Apply hydrostatic pressure [p
contcnt
g (wotcr Jcptb + 2um)] to replace the
operating pressure. First cooldown;
11. Repeat the process for second and third heat-up and cooldown.

6.4.3 Finite Element Method Pipeline Walking

The finite element analysis for the axial walking of 2km flowline is done with the same pipe-soil
interaction, pipe element and contact element used in the lateral knuckling model. Below are the
process taken to achieve the model and the results.
Reason for Finite Element Method:
The main reason for a finite element analyses in pipeline walking is to overcome some of the
short-comings that was introduced while applying the analytical method. The analytical method
described according to SAFEBUCK guideline were based on certain assumption and certain pipe
material properties that was used to obtain the defining parameter, [.
Realistically, the analytical method is mostly used during design to determine if a non-linear
analysis should be conducted based on the susceptibility and the rate of walking obtained. Hence,
it is not used as a validation method for a finite element analysis due to its sort-comings.
The analytical method gives a guide on axial displacement that could be encountered during the
heating cycles based on certain assumptions stated on the guideline but FEA provide a better
approximation to the walking that is to be experienced during operation.
Design Steps:
A 2km flowline is modeled using Pipe288 in ANSYS without any lateral imperfection;
Seabed Contact and target modeled using Conta175 and Targe170 respectively;
A temperature profile for the flowline is derived using excel spreadsheet corresponding to
different thermal gradient for each heating steps.
T
grad|ent
= T
|n|et
+ q
0
|
|
. . | uar|ex rum nude | tu nude n
A constant thermal gradient is generated until steady state is reached;
A total of 15 asymmetrical heating steps were done for each cycle and a uniform cooling
at the end overall heating steps;
The pressure variation remained at zero along the pipeline and solutions obtained for
each cycle.
Loading Steps without Direct Electric Heating (DEH):
2. Lay pipeline on even seabed with both ends free to move;
wutc
g wotcr Jcptb;
5. Apply the operating pressure along the pipeline;
6. Start the heating steps until steady state is attain;
7. Cooldown to ambient temperature.
Loading Steps with Direct electric heating (DEH):
2. Lay pipeline on even seabed with both ends free to move;
wutc
g wotcr Jcptb;
5. Apply the operating pressure along the pipeline;
6. Apply a temperature of 20
o
C along the pipeline length;
7. Apply the heating steps until steady state is attained;
8. Cooldown to DEH temperature.

7. Description of Case
7.1 Field Description
The work of this thesis is based on a typical subsea field development as shown in Figuie 71. The
scope includes the installation of two 22-inch pipelines of length 10km and 2km. The flowline
are made of X65 grade steel. The Field is a typical subsea field with a water depth of
approximately 800m. A riser is connected from a topside tie-in point of 20m above the mean sea
level (MSL) to a tie-in spool which is connected to the pipeline at the touchdown point. The 2km
flowline is connected between the two tie-in spools before the end structures while the 10km
pipeline is tied to PLET as shown in Figuie 71 below.
Figure 71: A typical Subsea field with Pipeline on even seabed (Subsea7, 2011).
CASE study:
Lateral Buckling
Pipeline end expansion assuming straight pipe on even seabed;
Deduction of the hot and cold end expansion using analytical and finite element model;
Verification of the 10km pipeline susceptibility to lateral buckling due to thermal, pressure
loading during operation;
Control of lateral buckling using snake-lay initiation method during installation;
Sensitivity analysis of the optimum lay radius required during installation of snakes.
Pipeline Walking
Susceptibility of Pipeline to walking (A check);
Deduction of rate of axial displacement in each cycle (rate of walking);
Sensitivity analysis of walking with respect to thermal gradient;
Analysis of the maximum strain that the tie-in spools can accommodate and corresponding
remedies.
2km Flowline
10km Pipeline
7.2 Design Parameters Lateral Buckling

This section describes the design parameters used in the analysis. The parameters include the
data related to pipeline geometry, material, operating and environmental loads, pipe-soil
interaction.
The design parameters used in the analysis are shown in the Table 71 below.
Table 71: Basic Design parameters
Parameter Abbreviation Value
Pipeline Length L
L
10km
Flowline Length L
S
2km
Outside Diameter D
o
22-inches (559mm)
Wall thickness t
st
19.1mm
Corrosion Allowance t
ext
3mm
Pipeline Material - X65
Steel Density p
st
7850kg/m
3

Pipe Submerged weight
w
submcgcd

2.52kN/m
SMYS <20
o
C - 450MPa
SMTS <20
o
C - 535MPa
Minimum Radius of Curvature in Normally
straight Pipe R 1000m
Operating temperature
I
op

95
o
C
Ambient temperature I
umb
5
o
C
Operating Pressure
P
op

15MPa
Water depth WD 800m
The coating parameter below gives the insulation coating and concrete weight coating according to
operational practice in Subsea 7.
Table 72: External coating parameter
Parameter Value
External coating thickness 5mm
External Coating density 910kg/m
3

Concrete coating thickness 55mm
Concrete coating Density 2400Kg/m
3

a. Temperature Profile:
The operating temperature variation along the entire pipeline length was generated using the general
exponential relation between the inlet and the ambient.
I = I
umb
+ (I
nIct
- I
umb
)c
(-L)

The temperature for the 10km pipeline is as shown in Figuie 72 below:
Figure 72: Temperature profile 10km Pipeline

The temperature profile in Figuie 72 was also used for deducing the end
expansion for both short and long pipelines. See Appendix A for details. This profile was used to
estimate pipeline expansion at both cold and hot ends.
b. Seabed Soil Condition and Frictional Data:
The seabed is assumed to be flat with a soft clay type soil. The axial and lateral frictional factors
are represented by their corresponding lower, upper bound and best estimate values of
coefficients as shown in Table 7S below.
Table 73: Friction Coefficients
Direction Lower Bound (LB) Best estimate (BE) Upper Bound (UB)
Axial 0.35 0.45 0.50
Static lateral 0.60 0.70 0.80
Dynamic lateral 0.43 0.85 1.28
Soil Mobilization 2mm 4mm

c. Applicability of Pipeline:
According to SAFEBUCK guidelines, pipeline data should be checked to ensure that pipeline is
considered within the range of validity. This is shown in Table 74:
Table 74: Applicability of Pipeline
Parameter Calculated Value
SAFEBUCK rage of
Validity
D/t 29,26701571 ration between 10 and 50
Design Temperature 95
o
C <180
o
C
Ovality less than 1.5% <1.5%
PIP NA -
FLUID oil HC based product
Installation Plasticity
No significance
plastic
deformation
during installation
No significance plastic
deformation during
installation not considered
Inspection Valid
100% girth weld UT
inspection
Seabed roughness Flat Seabed Relatively Flat seabed only

Hence, the pipeline falls within the range of applicability of SAFEBUCK guideline.
d. Pipe Material Yield Stress:
Based on the stress strain characteristics of X65 material, the de-rating of SMYS yield strength with
temperature is shown in Figuie 7S. Based on the given temperature profile, the de-rated material
properties are used in FE modeling analysis.
Figure 73: Pipeline Steel Material DE rating

e. Hydrostatic pressure:

w = wotcr Jcptb (8uu)
Therefore, the hydrostatic pressure on the pipeline surface is given as:
P
Hdostutc
= p
contcnt
g (w +y)
7.3 Design Parameters Walking
The design parameters for the 2km flowline are the same as the properties used for the lateral
buckling analysis.
The pressure variation for pipeline walking is taken as zero and this helps in deducing the actual
walking due to thermal gradient. The Axial friction coefficient is taken for the all the cases such
that the effect of walking can be deduced with respect to axial frictional forces.
Pipeline walking mechanism depends largely on the steps of heating profile developed during
heat-up cycles. As a result of this, a constant thermal profile is used for the heating steps until the
steady state is reached to ensure the effect was obtained due to thermal gradient alone.
This profile was developed in an excel spreadsheet, by interpolation of the inlet temperature at
each node along the flowline in the FE model.
wbcrc y = Jistoncc bctwccb tbc topsiJc tic -in onJ tbc mcon sco lc:cl (HSI)
y
w
Pipeline on seabed
Figure 74: Hydrostatic Effect from the topside connection
The analytical method used a constant thermal profile as well through out the heating steps.
The temperature profile used in the heating steps was developed in Excel using the relation:
I = I
hot cnd
- q
0
I

i = u, 1. . . lostnoJc
a. Thermal Transient Asymmetric Heating Steps:
The temperature was determined such that the asymmetric heating steps still retains the
remaining section of the pipeline at the ambient temperature. The heating process will continue
with the same profile until steady state is reached. The heating steps will be done in such a
manner that a large movement from a section to another is avoided and hence a continuous
heating interval is obtained.
The linear transient presented below exhibits a constant gradient along the pipeline involving 10
heating steps and the final steady state heating. This is the basis for walking mechanism along
the pipeline and it is visible at the cold end.
The Profiles are as follows:
Figure 75: Temperature Profile
0
10
20
30
40
50
60
70
80
90
100
0 200 400 600 800 1000 1200 1400 1600 1800 2000
T
e
m
p
e
r
a
t
u
r
e
(
C
)
KPdistance(m)
Temperature Profile Thermal transient
Walkingheatingprofile
8. Results and Discussion

8.1 Results and Discussion

This section presents results from the analyses carried out for the design of pipeline lateral
buckling and axial walking. The results are discussed against the design guide lines based on
DNV-FS-101.
The results are presented in relation to the main focus of the thesis work which is:
(DEH) to reduce rate of walking.
8.1.1 Verification of Pipeline Length Scale
Using the pipeline parameters and soil properties, the length scale of 10km pipeline and the 2km
flowline is investigated. The detail of the verification is shown in Appendix A1.
According to the guidelines (Subsea7, 2011), pipeline is regarded as short pipeline if the
calculated anchor length is larger than the length of the pipeline. This also means that the
pipeline is having insufficient friction to attain to full constrained axial force and it is hence said
to be fully mobilized. Otherwise it is regarded as a long pipeline.
Details of end expansion calculation are as shown in Appendix A: The following were obtained
as shown in Table 81.
Table 81: Anchor Length results
Pipeline
Hot end Anchor
length
Cold end Anchor
length
10km Pipeline 3.272km 2.79km
2km Flowline 4.83km
As shown in Table 81, the 10km pipeline is now established as a long pipeline which develops
full constraint axial force while the 2km flowline is a short pipeline which never develops full
constraint axial force.
This is further verified by deducing the corresponding effective axial force and frictional
resistance forces generated by soil-pipe interactions (see Appendix A1).

8.1.2 Effective Axial Force

Applying the temperature profile and the pipeline properties outlined in the previous section, the
effective axial forces for the 10km and 2km pipelines were calculated. The results are shown in
Appendix A1 using.
Short Pipeline:
The effective axial force and the fully constrained axial force (See Appendix A for details)
generated along the 2km pipeline is plotted and shown in Figuie 81. Based on the results for the
anchor length and the effective axial force, it is clear that the 2km pipeline is fully mobilized and
thus it is a short pipeline.
As seen from Figuie 81, the total axial frictional force is not enough to fully constrain the
pipeline. During expansion, the two pipe-ends move in opposite direction to each other and
develop maximum friction at the middle of the pipe where no expansions occur and this point is
termed as virtual anchor point.
Figure 81: Effective Axial force of a short Pipeline

Long Pipeline:
For the 10km pipeline, it is evident from Figuie 82 that the buildup of frictional force exceeds the
axial force required to fully constrain the pipe. As a result, it generates two virtual anchor points
towards the hot and cold ends. This shows that certain section of the pipeline is fully constrained
while the end sections are free to expand.
Virtual Anchor Point
The effective axial force developed during thermal and pressure loading reaches full constraint
as the frictional forces builds-up. This could also yield larger end expansion at both ends if the
lateral and axial frictional forces are decreased. See Appendix A2 for detail calculations.
Figure 82: Effective Axial Force of a long pipeline

8.1.3 End Expansions
The build-up of frictional force induces compressive axial forces while restricting the axial
expansion generated by temperature and pressure loading. This end expansion shows impacts on
the design of tie-in/spool connected to the pipeline. Hence, the results are important during
lateral buckling and axial walking design. The results for end expansion at different coating
thicknesses are shown in Table 82 and Table 8S for long and short pipelines respectively.
Table 82: End Expansion Long pipeline
End Expansion of 10km Pipeline
Analytical FE Model
Concrete thickness Hot-end cold-end Hot-end Cold-end
25mm 2.007m 1.14m 1.96m 0.96m
40mm 1.858m 1.037m 1.864 0.858
55mm 1.725m 0.945m 1.765m 0.816m
Virtual Anchor Point
Note: The results stated above are for end expansions without lateral buckling control measure.
This is also a testament that the 10km pipeline builds up more frictional force and hence has
tendency more to buckle as seen from the results of the two pipelines.
Table 83: End Expansion of short pipeline
End Expansion of 2km Flowline
Analytical FE Model
Concrete
thickness Hot-end cold-end Hot-end Cold-end
25mm 1.021m 0.933m 1.0509m 0.869m
40mm 1.006m 0.918m 1.0547m 0.8862m
55mm 0.99m 0.902m 1.08m 0.909m
Note: See Appendix A1 for the detail calculation steps
Controlled lateral buckling by sharing expansion reduces the end expansion. If the end expansion
is found higher than the spool capacity designed for the pipeline connection, a measure will be
re-established to accommodate the expansions at the ends, either by rock dumping or in-line
spool connection.
8.1.4 Susceptibility of Pipeline to Lateral Buckling
It is a standard requirement to check for susceptibility of pipeline to buckling (See Appendix A2
for detailed calculation steps). Pipelines are said to be susceptible to lateral buckling if the
following inequality holds:
N
muxmum
N
ctcuI

Where:
N
muxmum
= Ibc moximum comprcssi:c cccti:c oxiol orcc,
N
ctcuI
= Ibc criticol Buckling orcc,
N
muxmum
= min (N
]uII constun
, N
]mux
),
N
ctcuI
= min (N
00S
, N
Hobbs
),
N
]uII constun
= Full onstroincJ cccti:c oxiol orcc,
N
]mux
= moximum comprcssi:c cccti:c oxiol orcc in sbort pipclinc,
N
00S
= ciitical buckling foice associateu with Pipeline out of stiaightness (00S),
N
Hobbs
= Eobbs minimum buckling orcc,
N
Hobbs
= min (moJc
1
, moJc
2
, moJc
3
, moJc
4
, moJc
n]nt
),
The Hobbs critical buckling force is calculated based on a spreadsheet developed using Mathcad
15. The critical buckling force for each of the buckling modes is based on the pipeline input and
using the Hobbs constants K1, K2, K3, K4, and K5 as shown in Table 84.
Table 84: Constants for Lateral buckling mode
Modes K
1
K
2
K
3
K
4
K
5

1 80.76 6.391 X 10
-5
0.5 2.407 X 10
-3
0.06938
2 4a
2
1.743 X 10
-4
1.0 5.532 X 10
-3
0.1088
3 34.06 1.668 X 10
-4
1.294 1.032 X 10
-2
0.1434
4 28.20 2.411 X 10
-4
1.608 1.047 X 10
-2
0.1483
4a
2
4.7050 X 10
-5
4.4495 X 10
-3
0.05066

From the spreadsheet, the buckling force within the buckle was plotted against the buckling
length for a certain section as described in Hobbs papers. Below is the result of one of the plots
in Figuie 8S (see Appendix A for details).
Figure 83: Hobbs Critical Buckling Force

Mode 1
Mode 2
Mode 3
Mode 4
0,0
1,0
2,0
3,0
4,0
5,0
6,0
7,0
8,0
0,0 20,0 40,0 60,0 80,0 100,0 120,0 140,0
B
u
c
k
l
i
n
g

F
o
r
c
e

(
M
N
)
Buckling Length (m)
Lateral buckling
Infinite Mode Mode 1 Mode 2 Mode 3 Mode 4
The values for the Hobbs critical buckling force obtained are as shown in Table 8S for different
lateral friction factors:
Table 85: Hobbs Critical Buckling force at different friction factors
Buckling mode Units
Analytical Results
LB = u.6 BE =u.7 0B =u.8
Mode 1 KN 3086.5 3354.0 3603.8
Mode 2 KN 3001.4 3240.2 3482.2
Mode 3 KN 2937.4 3190.0 3425.2
Mode 4 KN 2927.8 3180.7 3417.5
Mude KN 3672.4 3966.6 4240.5
Minimum values (KN) 2927.8 3180.7 3417.5
N
Hobbs
LB
= 2927.8 KN
N
Hobbs
BL
= S18u.7 KN
N
Hobbs
0B
= S417.S KN
Note: Hobbs critical buckling force is based on assumption of a straight pipeline which is not
obtainable in reality due to deformation from lay barge, seabed contours and initial imperfection
(OOS). The results are summarized in Table 86.
As stated previously, the critical buckling force associated with pipeline out of straightness
(OOS) is obtained with the following relation:
N
00S
= p
Lut
w
submcgcd
R
Table 86: Analytical Result Critical Buckling Force
N
OOS

W_xuhmerged
KNm
Radius
R (m)
Analytical Results
LB = u.6 BE =u.7 0B =u.8
N
OOS
3345 1500 3010 3512 4014

N
OOS
3345 2500 4014 4683 5352

Hence,
N
ctcuI
= min (N
00S
, N
Hobbs
)
N
ctcuI
LB
= 2927.8 KN
N
ctcuI
BL
= S18u.7 KN
N
ctcuI
0B
= S417.S KN
Given that the maximum compressive axial force along the pipeline is obtained from calculation
as:
N
Muxmum
= 82S2 KN
The pipeline is therefore, susceptible to BUCKLING because the inequality is true.
N
muxmum
N
ctcuI

8.1.5 Regions Susceptible to Lateral Buckling
Since the 10km pipeline is now known to be susceptible to lateral buckling, the area that is prone
to such lateral movement is to be determined. Critical buckling force is used as a limiting
criterion to define the areas that are susceptible to buckling. These areas are those with effective
axial force greater than the critical buckling force as shown in Figuie 84 below:
Figure 84: Region Susceptible to lateral Buckling

The section that are prone to buckle as seen from Figuie 84 above ranges from KP1 to KP9,
resulting in about 8km pipeline length having the potential to buckle.
9
8
7
6
5
4
3
2
1
0
0 2000 4000 6000 8000 10000
A
x
i
a
l
F
o
r
c
e
(
M
N
)
KP(Km)
RegionSusceptibletoLateralBuckling
EffectiveAxial FullConstrain CriticalForce(LB)
Critical(BE) Criticalforce(UB)
Buckle Prone Region
8.2 Lateral Buckling Behaviour
As observed from the previous analysis, about 8km of the pipeline length is susceptible to lateral
buckling within KP1 and KP9. Naturally, pipeline laid of seabed with an initial imperfection will
buckle on thermal and pressure loadings.
If the induced compressive force is sufficiently high, uncontrolled buckling could occur and the
pipeline integrity could be lost. This effect can lead to the following limit state:
Excessive Plastic deformation of the pipe which could lead to local buckling collapse
Cyclic fatigue failure due to continuous heat-up and cool-down
Figure 85: lateral displacement at the buckle site

As shown in Figuie 8S, the non-linear analyses of the 2km VAS model with a 2m initial
imperfection and 1500m radius generated a lateral displacement of 8.56m under operating
temperature and pressure considering the operation corridor of not more than 20m. This is as
shown in the Figuie 8S with corresponding cyclic loading at shut-down and cooldown steps.
Based on the results shown in Figuie 8S, maximum lateral displacement of 8.56m occurred at a
particular buckle site along the pipeline. This effect will be detrimental to the pipeline and need
to be reduced to an acceptable limit.
4,00
2,00
0,00
2,00
4,00
6,00
8,00
10,00
450,00 460,00 470,00 480,00 490,00 500,00 510,00 520,00 530,00 540,00 550,00
B
u
c
k
l
e
A
m
p
l
i
t
u
d
e
DistanceAlongpipeline(m)
Lateraldisplacementatthebuckle
Heatupatsteadystate Firstcooldown Secondheatup
SecondCooldown Thirdheatup Thirdcooldown
Linear(Secondheatup)
During cyclic loading, it can be seen that the pipe-soil interaction cannot be predicted during
cooldown steps. The cooldown step from operating temperature of 90
o
C to ambient temperature
of 5
o
C resulted in non-linear behavior of the buckle section which resulted in strain build up
within the buckle site.
Since this is a displacement controlled condition, the capacity of pipeline is based on allowable
strain. For the pipeline, the predicted strain from the FE analyses was compared against
allowable strain. It was observed that predicted strain reached maximum allowable strain can be
predicted to occur at lower operating temperature.
The effective axial force at different soil friction is important to extract the result of the post
buckling force from FE analyses. Figuie 86 presents the results for the effective axial force at
different friction factors.
Figure 86: Effective axial force for different friction factors

This is also a conformational statement that the worst case scenario for pipeline to fail occurs at
the lower bound friction. At the lower bound friction, there will be a lower effective axial force
due to less frictional resistance from the soil and hence more length of the pipe is fed-into the
buckle section.
3
2,5
2
1,5
1
0,5
0
0 200 400 600 800 1000 1200 1400 1600 1800 2000
E
f
e
c
t
i
v
e
A
x
i
a
l
f
o
r
c
e
(
M
N
)
KPdistance(m)
Effectiveaxialforceforceatdifferentaxialfrictionfactor
UB BE LB
The above results from FE analyses result are the basis for Snake lay configuration and
expansion sharing into adjacent buckle such that total strain developed by the thermal loading
and pressure loading be shared among the pre-determined buckles.
8.3 Snake Lay Control Measure
There are many ways of reducing or controlling lateral buckling of pipeline on seabed depending
on the criticality of the problem. The conventional way is by rock dumping at some certain pre-
determined areas along the pipeline.
Here in this thesis work, one of the Buckle initiation method (expansions sharing) called Snake-
Lay configuration is applied. The snake Lay configuration is based on the principle of sharing
expansion into adjacent buckles as stated in DNV-RP-F110.
Note: The obtained snake lay configuration in the present work is based on the allowable
capacity of the pipeline. However, the required number of snakes is dependent on allowable end
expansion taken by the spools connected to the pipeline. As the capacity of the connecting Spool
(expansion Limit) to the 10km pipeline is not known, the result of expansion sharing will be
acceptable based on further analysis
The purpose of the snake lay configuration and intermittent rock dumping is to ensure the feed-in
into buckle shall be within allowable capacity of the pipeline and further to ensure that end
expansion limit can be kept within the limit given by the spool capacity.
As stated previously, the expansion sharing between the adjacent buckles shall be established
when the following relation is fulfilled:
F
post
LB
+S
LB
F
ctcuI
0B

This relation is used in developing the spreadsheet for snake lay configuration.
At the hot and cold ends, there is less frictional force available to build up effective forces
between the buckle so there is need to build up frictional forces at those ends. To obtain this,
rock dumping of some height TOP was employed at the hot and cold ends.
8.3.1 Lay Configuration for 10km pipeline R1500m
As shown in Figuie 87, the FE model for snake-lay configuration of radius 1500m was developed
with an initial imperfection of 2m for a 2km VAS model on even seabed. Based on the limiting
condition stated previously from DNV-OS-F101, the post buckling force at the lower bound
friction coefficient were obtained.
By applying several time steps, the FE analyses were run to obtain the post buckling force
corresponding to the maximum allowable strain based on displacement controlled condition
(strain based criteria) as given below:
e
mux
e
cu

Also allowable feed-in was calculated based on displacement controlled condition. Based on the
results from FE analyses, the predicted post buckling forces and the allowable feed-in are
summarized below in Table 87 for several combinations of frictional factors.
Table 87: Feedin Results for R1500m
Axial
Friction
Lateral
Friction
F
cr|t|ca|

Strain Limit
Displacement controlled condition
F
puxt

s
ca
Feed -|n
KN KN m/m m
LB LB -2408.4 -1756 0.004 1.20
BE BE -3010.5 -1864 0.004 1.12
BE UB -3010.5 -1894 0.004 0.9
UB UB -3417.5 -1910 0.004 1.12
Adopting the results from the ANSYS model and using the snake lay excel spreadsheet; the
expansion sharing was generated for each snake based on acceptance criteria stated previously.
PREB0CKLINu
P0STB0CKLINu
Figure 87: PreBuckling and Post Buckling of a pipeline snake Configuration
By applying the R1500m configuration, the upper bound critical buckling force is -3010.5kN and
the lower bound post-buckling force is -1756kN.
To ensure that allowable feed-in of 0.9m at the hot end, rock dumping of 2m rock height TOP
was applied over the section of 100m. Figuie 88 presents the effective axial force distribution
obtained from snake-lay configuration.
Figure 88: Effective axial force distribution for Snake Lay configuration
The result in Figure 88 shows that the 10km pipeline requires 7snakes, rock dumping to limit end
expansions. Table 88 shows the results obtained for the 1500m radius lay configuration.
0
1
2
3
4
5
6
-6
-5
-4
-3
-2
-1
0
0 2 4 6 8 10
E
f
f
e
c
t
i
v
e

A
x
i
a
l

F
o
r
c
e

[
M
N
]
KP [km]
Effective Axial Force
Lower bound Best Estimate
Upper bound Pcrit (Hobbs)
P crit (stab) Fully restrained
Rock Height
R
o
c
k

H
E
i
g
h
t

(
m
)

Table 88: Expansion Summary for the R1500m configuration

Lay radius
Virtual anchor
#1 KP
Virtual anchor
#1 KP
Total feed-
in
Allowable
feed-in
m m m m m

Hot end 1500 Free end 840 0.800 -

Snake 1 1500 840 2064 0.814 0.9

Snake 2 1500 2064 3014 0.778 0.9

Snake 3 1500 3014 4002 0.747 0.9

Snake 4 1500 4002 4870 0.714 0.9

Snake 5 1500 4870 5816 0.683 0.9

Snake 6 1500 5816 6766 0.652 0.9

Snake 7 1500 6766 8236 0.625 0.9

Cold End 1500 8236 Free end 0.60 -
The predicted end expansion and the buckle feed-in length are listed in Table 88 and compared
against allowable values. Note that pipeline end expansions are within the allowable capacity of
the pipe and these shall be compared against the allowable end spool expansions which are not
known.
From Table 88, it is seen that the computed pipeline feed-in lengths satisfy the acceptable limit
which is based on strain based criterion. The effect of rock dumping at the hot end reduces the
number of snakes that are used in the expansion.
The snakes can as well be reduced further but with a higher cost of rock dumping at the buckle
crest. Depending on the project, the client requirement, the capacity of end spools and hub
capacity, the expansion can be re-established if needed.
It is worth to note that the hub capacity where the spool is fitted into must not be compromised
and the expansions that the spool must accommodate are not to be exceeded. Otherwise the
sharing expansion shall be re-considered with respect to the limiting capacities.
In the present case, the end spool capacities are unknown. But according to the results presented
in Table 88, the spool shall accommodate expansion of 0.80m at the hot end and 0.6m at the cold
end. These results for the pipeline end expansion can be used along with operational loads to
design end spools and to assess interface loads at the hub.
This shows that the integrity of our pipeline is recovered and intact according to DNV-RP-F110.
Figuie 89 shows the expansion distribution. At the hot end the 2m rock height TOP was used to
build up the effective axial force in order to reduce end expansion. This helps to reduce the
expansion in the adjacent snake along the pipeline.
From Figuie 89, it can be seen that the expansion towards the cold end is reduced as a result of
the sharing of the feed-in into the snakes. Also due to the use of rock at the end, an expansion of
0.60m was induced to achieve the desired configuration with an acceptable displacement control
condition.
Figure 89: Expansion Distribution

The purpose of snake lay configuration has therefore been achieved which is to trigger sufficient
number of thermal buckles at pre-determined location along the pipeline so that the expansion is
distributed among the buckles rather than being concentrated at a few buckle sites.
This is shown in Figuie 81u where the effective axial for the snakes lay control were compared
with the effective axial force without snake lake measure.
It can also be seen that the planned buckle occurs prior to the unplanned buckle with regard to
the 2km model used in obtaining the post buckling force. This is done to ensure that the required
buckle distribution is achieved.
-1
-0,8
-0,6
-0,4
-0,2
0
0,2
0,4
0,6
0,8
1
0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000
E
x
p
a
n
s
i
o
n

(
m
)
Kp (m)
Expansions
Lower bound Best Estimate Upper bound
Figure 810: Effect of Snake Lay configuration

Figuie 81u, shows the distribution of total feed-in into several buckles. Here, the crown of the
snake behaves like a curvature expansion spool while the feed-in is controlled by the distance
between two successive crowns.
The effective axial force increases as the frictional restraint builds up due to the seabed
interaction. But the introduction of snakes modifies the development of the effective axial force
and causes effective axial force to drop within the buckle site as the pipe feeds-in to the buckle.
As a result, the effective axial force drops to a value much less than the critical buckling force.
This is achieved at the lower bound friction factor which is worst case scenario for the pipe to
fail during an operation.
9
8
7
6
5
4
3
2
1
0
0 2000 4000 6000 8000 10000
F
O
R
C
E
(
M
N
)
KP(Km)
EffectofSnakeLayExpansionSharing
Fullconstrain EffectiveForce SnakeLayControl
8.3.2 Parametric Study of Lay Configuration

The previous section considers snake lay generated by using a lay radius of 1500m and 2m
imperfection. The response of the pipeline shows that a total of 7 snakes are formed while using
rock dumping at both hot and cold-ends. The present section considers a lager lay radius of
2500m having the same initial imperfection of 2m.
Table 89: Feedin Results for lay radius of 2500m
Axial
Friction
Lateral
Friction
Radius F
cr|t|ca|

Strain Limit
Displacement controlled
condition
F
puxt

s
ca
Feed -|n
m KN KN m/m m
LB LB 2500 -2927.8 -2376 0.004 1.52
BE BE 2500 -3180.7 -2501 0.004 1.50
BE UB 2500 -3417.6 -2540 0.004 1.1
UB UB 2500 -3417.6 -2641 0.004 1.48
It is known that Critical Buckling force F
ctcuI
is a function of Lay radius (R), submerged
weight and the friction coefficient as given by:
F
ctcuI
= p
L
w
submcgcd
R
It can be seen from Table 89, that the larger the radius the more the critical buckling force. The
increase in the buckle length from 160m (R1500m) to 200m (R2500m) results in larger radius
and smaller initial imperfection. The results based on this lay radius show less feed-in compared
to the radius of 1500m. In this case rock was applied only at the hot end. The snake lay with lay
radius of 2500m increases the number of snakes to 9 as seen in Figuie 812.
Also, it is noticed that the post buckling force at higher radius of 2500m was higher than the
R1500m due to the unpredicted pipeline embedment and the resultant breakout force.
With the use of more rock dumping at the crown of each snake, we will be able to obtain less
number of snakes as shown as dashed lines in Figure 811. But, this can be achieved at higher
cost of rock dumping which also depend on the operation corridor for the pipeline.

The dashed lines are the new
snakes generated upon using the
rock dump at the crest and along a
buckle region in between two
buckles. There is a tendency of
reaching the critical buckling force
withthissolution.
Figure 811: Rock Dumping between snake
Figure 812: Effective force distribution of snake lay (R2500m) configuration with Rock dumping
For the R2500m configuration shown in Figuie 812, the lowest expansion at the cold and hot
reduced by 0.2m due to the lower feed-in in the buckle site as compared to R1500m. This is
testament that the expansion the spool can accommodate can be re-established to suit the spool
design value by increasing the radius or rock dumping at the opposite end. This is shown at the
expansion result in Figuie 81S below:
0
1
2
3
4
5
6
-6
-5
-4
-3
-2
-1
0
0 2000 4000 6000 8000 10000
E
f
f
e
c
t
i
v
e

A
x
i
a
l

F
o
r
c
e

[
M
N
]
KP [m]
Effective Axial Force R2500
Lower bound Best Estimate
Upper bound Pcrit (Hobbs)
P crit (stab) Fully restrained
Rock Height
R
o
c
k

H
e
i
g
h
t

(
m
)
Figure 813: Expansion results for R2500

The resulting post buckling force within the buckle is the basis in deriving number of snakes on
seabed. The post buckling force is extracted from FE analyses based on predicted strain
corresponding to the allowable design strain from displacement controlled criterion. As
mentioned previously, the expansion sharing between the adjacent buckles shall be established
when the following condition is fulfilled.
F
post
LB
+S
LB
F
ctcuI
0B

Based on this condition, the lower bound the post buckling force is a requirement for a
conservative solution such that the pipeline integrity will be maintained even at lower bound
value of the soil-pipe interaction.
Upon using the lower bound friction factor in the FEA, the worst feed-in length is obtained based
on the stated condition. The result shows that more feed-in is generated with R2500m as
compared with R1500m as shown Figuie 89.
It has also been observed also, that there are uncertainties regarding the post buckling force and
it depends strongly on the soil-pipeline interaction. This is one of the challenges during snake lay
configuration.
-1,2
-1
-0,8
-0,6
-0,4
-0,2
0
0,2
0,4
0,6
0,8
1
0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000
E
x
p
a
n
s
i
o
n

(
m
)
Kp (m)
Expansions
Lower bound Best Estimate Upper bound
Table 810: Expansion Summary for R2500

Lay radius
Virtual anchor
#1 KP
Virtual anchor
#1 KP
Total feed-
in
Allowable
feed-in
m m m m m

Hot end 2500 Free end 1172 1.000 -

Snake 1 2500 1172 2296 0.83 1.1

Snake 2 2500 2296 3060 0.57 1.1

Snake 3 2500 3060 3828 0.55 1.1

Snake 4 2500 3828 4620 0.53 1.1

Snake 5 2500 4620 5398 0.51 1.1

Snake 6 2500 5398 6188 0.48 1.1

Snake 7 2500 6188 6980 0.46 1.1
Snake 8 2500 6980 7712 0.45 1.1
Snake 9 2500 7712 8508 0.43 1.1

Cold End 2500 8508 Free end 0.58 -

The computed expansion and the buckle feed-in lengths are summarized in Table 81u. The results
show that the computed feed-in lengths are within the acceptable feed-in based on the strain
based criterion.
Note: It can be deduced from the FE analysis and the snake lay result that the planned buckles
occurred at the center of the VAS model or prior to the location chosen during the FE modeling.
This is in accordance with the expansion sharing described in DNV-RP-F110.

8.4 Pipeline Walking Results and Discussion

As stated previously, when pipeline is laid on seabed and heated up by hot fluid content passing
through it, it expands axially and circumferentially. Upon cooling, since this expansion is
restricted by axial friction generated by pipe-soil interaction, the pipeline will not return to its
original position. Over number of cycles of subsequent heating and cooling, an accumulation of
axial displacement of pipeline towards the cold end is generated. This is known as pipe-walking.
Note: This thesis work considers only the effect of thermal loading and transient along the 2km
flowline.
8.4.1 Susceptibility to Pipeline Walking
According to SAFEBUCK guideline, pipelines are regarded to not to be susceptible to walking if
the axial friction force exceeds the following value:
>
(FA
xt
uT)
L

The results from the calculation summarized in Appendix A3 for the 2km flowline. From the
results, it can be seen that the 2km flowline is susceptible to walking based on the SAFEBUCK
guidelines for pipeline walking.
Pipeline being susceptible to walking do not necessarily suggest danger to the life of the field
unless otherwise proved that the walking rate will be detrimental to the connecting ends.

Thus, there is need to obtain the extent the pipeline will walk by using an analytical method
based on the relation stated on the SAFEBUCK guideline.

A spreadsheet was developed on Mathcad 15 (see Appendix A4) and the rate of walking per
cycle under constant thermal gradient was deduced using the relation below:
0
=
I
2
8 EA
s
--- - -- - -- i

0
6

0
=
I
2
16 EA
s
(24
0
-
0
-4) - - -- - --- -i >

0
6

The above expression gives the accumulated axial displacement as a result of the thermal
transient based on constant gradient along the seabed and the results are as shown below in Table
811:
Table 811: Analytical Result for rate of walking at 35
o
C/km
Axial Friction Factor
Thermalgradient
[
o
C/km]
Rateofwalking
[mm]
LB 35 51
BE 35 44
UB 35 40
From the above results, it is seen that for lower bound friction, the rate of walking towards the
cold end is higher than that for best estimate and upper bound friction. This response was
expected.
The axial friction coefficient is the dominating factor during pipeline walking and this has to be
taken into consideration in order to minimize the effect of walking.
From the calculation using the analytical method, it is found that at higher axial frictional factor
and lower thermal gradient the SAFEBUCK relations do not give a good approximation based on
the assumption stated on the guideline (see Appendix A4). Hence, it requires performing a finite
element analysis and to obtain a better approximation of the rate walking at different thermal
gradient.
Hence, the analytical method of calculation is not used for final conclusion of results but it is
used as a basis to predict the response of walking along the pipeline.
8.4.2 Finite Element Analysis Axial Walking

Temperature profile presently in Figuie 7S with constant thermal gradient of 35
o
C/km was used
to generate asymmetric heating step along the pipeline. The response from the heating steps is as
shown in Figuie 814 below:
Figure 814: Thermal transient at 35

o
C/km
0
10
20
30
40
50
60
70
80
90
100
0 500 1000 1500 2000
T
e
m
p
e
r
a
t
u
r
e
(
C
)
KPdistance
ThermalTransient 35
o
C/km
heatstep1
heatstep2
heatstep3
heatstep4
heatstep5
heatsteps6
heatstep7
heatstep8
heatstep9
heatstep10
heatstep11
heatstep12
Steadystate
heatstep11
heatstep12
heatstep13
When the 2km flowline is subjected to the above heating steps, it experiences a non-uniform
expansion due to the distribution of the hot fluid from the hot end. The region of the pipe close to
the hot end expands from the center of virtual anchors that are formed along the center pipeline
at each heating step, while the rest of the pipeline remains cold approximately at the ambient
temperature. Due to this reason, non-uniform heating along the pipeline induces walking.
Figuie 81S presents the results for axial displacement prior to full mobilization for heat-up steps.
From the figure, it is seen that, the displacement at the hot region increases from the heat-step 1
to heat-step 7. But the displacement at the cold region remains to be almost zero.
Figure 815: Axial displacement prior to full mobilization

When the pipeline approaches heating-step 8, the entire pipeline is heated up and the cold end
begins to expand with increase in temperature as seen from the results shown in Figuie 816:
Figure 816: Axial displacement at full mobilization

From the results shown Figuie 817, it is further observed that as heating continues the pipeline
experiences expansion at both hot and cold ends and it forms a virtual anchor at the center of the
pipeline.
0,6
0,4
0,2
0
0,2
0 200 400 600 800 1000
A
x
i
a
l
D
i
s
p
l
a
c
e
m
e
n
t
(
m
)
KP(mx2)
AxialdisplacementpriortofullMobization
heatstep1
heatstep2
heatstep3
heatstep4
heatstep5
heatstep6
heatstep7
0,6
0,4
0,2
0
0,2
0 200 400 600 800 1000
A
x
i
a
l
D
i
s
p
l
a
c
e
m
e
n
t
(
m
)
KP(mx2)
AxialDisplacementduringFirstheatingcycle 35
o
C/km
heatstep1
heatstep2
heatstep3
heatstep4
heatstep5
heatstep6
heatstep7
heatstep8
Figure 817: Axial Displacement during the First heating cycle

From the above mentioned analyses, the predicted axial displacements are summarized below in
Table 812 for the case of upper bound (UB) friction coefficient.
Table 812: Rate of walking for several cycles
Axial Friction
Factor
Thermalgradient
[
o
C/km] Cycles
Displacement
[m]
Cumulative
displacement
(mm)
UB 35 1 0.0724 -
UB 35 2 0.1318 59.40
UB 35 3 0.2104 78.50
UB 35 4 0.2686 58.20
UB 35 5 0.3402 71.50
UB 35 6 0.3988 58.50

1,2
1
0,8
0,6
0,4
0,2
0
0,2
0,4
0,6
0,8
1
1,2
0 200 400 600 800 1000
A
x
i
a
l
D
i
s
p
l
a
c
e
m
e
n
t
(
m
)
KP(mx2)
AxialDisplacementduringFirstheatingcycle 35
o
C/km
heatstep1
heatstep2
heatstep3
heatstep4
heatstep5
heatstep6
heatstep7
heatstep8
heatstep9
heatstep10
heatstep11
heatstep12
heatstep13
heatstep14
heatstep15
steadystate
Cooldown
0bele Ifenna Isaac 0niveisity of Stavangei Page 1uu
Operational Implication: Assuming that the pipeline is subjected to a total of 100 full shut-
down and start-up cycles over its lifetime, the pipeline would walk axially a total of 6m towards
the cold end connection. This scenario could result in a serious damage and a threat to the
integrity of the pipeline system which must be mitigated.
Finite element analyses were carried out for different friction coefficients to understand the
sensitivity of method of analyses on predicted results when compared from analytical
calculations. Table 81S compared the results obtained from both analytical calculations and FE
analyses.
Table 813: Results from FE Analyses and Analytical Calculation for Different Axial Frictions.
Methods Axial Friction
Walking displacement
(mm)
Analytical UB 40
FE analysis UB 59

Analytical BE 44
FE analysis BE 60.4

Analytical LB 51
FE analysis LB 106.3

As was concluded by (Carr et al., 2006), the above results confirms that the walking
phenomenon is dependent on the axial frictional force.

0bele Ifenna Isaac 0niveisity of Stavangei Page 1u1
8.4.3 Effect of Thermal Gradient on Pipeline Walking

Several authors have analyzed the pipeline walking mechanism based on its causing effect and
the operational assumptions. One of the factors is the effect of the axial frictional force as was
confirmed in the previous discussion. As explained by (Carr et al., 2006), the causing effect of
pipeline walking in short pipelines could be one of the following:
Tension at the end of the flowline tied to an SCR;
Seabed slope;
Thermal gradient along the pipeline.
In this thesis work, the effect of thermal gradient has been considered by modeling a 2km short
pipeline which is susceptible to walking based on SAFEBUCK guidelines. The study is based on
the two thermal gradients of 35
o
C/km and 20
o
C/km with the same 2km length as shown in Figuie
818:

From the heating profiles presented above in Figuie 818, it is evident that the steeper profile of
35
o
C/km induces more heating steps (8 heating steps) before reaching the end of the pipeline
compared to the five (5) heating steps induced by the 20
o
C/km thermal gradient.
The steepness is a function of temperature change since the steeper the gradient the faster the
temperature is transferred towards the downstream of the pipeline. Irrespective of the
temperature profile, this is a dominating effect as seen in the above plots.
From the result of the finite element analyses for the two thermal gradients, the predicted axial
displacements prior to full mobilization are shown in Figuie 819.

0
100
0 500 1000 1500 2000
T
e
m
p
e
r
a
t
u
r
e
(
C
)
KPdistance
ThermalTransient 20
o
C/km
heatstep1 heatstep2
heatstep3 heatstep4
heatstep5 heatsteps6
heatstep7 heatstep8
heatstep9 heatstep10
Steadystate heatstep11
0
100
0 500 1000 1500 2000
T
e
m
p
e
r
a
t
u
r
e
(
C
)
KPdistance
ThermalTransient 35
o
C/km
heatstep1 heatstep2
heatstep3 heatstep4
heatstep5 heatsteps6
heatstep7 heatstep8
Figure 818: Thermal gradient showing two different Scenarios (35
o
C/km and 20
o
C/km)

The two cases were subjected to the same temperature with increasing 10
o
C in each heating step
down the pipe length. At 70
o
C and heating step 8, the 35
o
C/km case attained full mobilization
while at 50
o
C and heating step 5 the 20
o
C/km attained full mobilization.
As shown in the Figuie 819, the hot end of the 35
o
C/km case expansion of about 0.45m before
reaching full mobilization while the case of 20
o
C/km shows expansion of about 0.23m at full
mobilization. This concludes that the case of 20
o
C/km reaches full mobilization earlier than the
case of 35
o
C/km at the same temperature. Once full mobilization is reached, to attain steady
state, the case of 35
o
C/km requires another 8 heating steps to attain to steady state temperature
while the case of 20
o
C/km requires only 6 heating steps. This is as shown in Figuie 82u.

0,6
0,4
0,2
0
0,2
0 500 1000
A
x
i
a
l
D
i
s
p
l
a
c
e
m
e
n
t
(
m
)
KP(mx2)
AxialdisplacementpriortofullMobization
35
o
C/km
heatstep1 heatstep2 heatstep3
heatstep7
0,3
0,2
0,1
0
0,1
0 500 1000
A
x
i
s
T
i
t
l
e
AxisTitle
Axialdisplacementpriortofullmobilization
20
o
C/km
heatstep2 heatstep3
heatstep4 heatstep5
1
0,5
0
0,5
1
0 500 1000
A
x
i
a
l
d
i
s
p
l
a
c
e
m
e
n
t
(
m
)
KP(mx2)
Axialdisplacementduringthefirstcycle
20
o
C/km
heatstep2
heatstep3
heatstep4
heatstep5
heatstep6
Heatstep7
heatstep8
heatstep9
heatstep10
heatstep11
heatstep12
steadystate
cooldown
1,5
1
0,5
0
0,5
1
1,5
0 500 1000
A
x
i
a
l
D
i
s
p
l
a
c
e
m
e
n
t
(
m
)
KP(mx2)
AxialDisplacementduringFirstheatingcycle
35
o
C/km heatstep1
heatstep2
heatstep3
heatstep4
heatstep5
heatstep6
heatstep7
heatstep8
heatstep9
heatstep10
heatstep11
heatstep12
heatstep13
heatstep14
Figure 819: Axial displacement for prior to full mobilization for two different thermal gradient
Figure 820: Axial displacement of the two cases after full mobilization.
0bele Ifenna Isaac 0niveisity of Stavangei Page 1uS
As a result of the enormous amount of heat induced on the pipeline prior to full mobilization by
the 35
o
C/km gradient, over shutdown/cooling cycles, the pipeline axial walking towards the cold
end is more than the walking from the case of 20
o
C/km as seen in Figuie 821. This can also
restated that the heating that results in walking phenomenon is mostly the heating steps prior to
full mobilization stage.
The case of 20
o
C/km requires 5 heating steps and yields walking of 39.5mm while the case of
35
o
C/km requires 8 heating steps and yields walking of 59mm.
The cumulative displacements obtained from the two cases are shown Figuie 821:

From Figuie 821, the full curvature shows the attainment of full mobilization and also shows that
it took enormous heating for the 35
o
C/km gradient to reach full mobilization. Because of the
asymmetric expansion from the heating steps prior to full mobilization, the 35
o
C/km gradient
induces an axial movement giving rise to 59mm pipe movement towards the cold end while the
20
o
C/km gradient gives 39.5mm pipe movement after the cooling step as shown above Figuie
821. Once full mobilization is attained, the walking stops and the middle section remains fixed
as noticed in Figuie 82u.
250
200
150
100
50
0
50
0 500 1000
C
u
m
u
l
a
t
i
v
e
D
i
s
p
l
a
c
e
m
e
n
t
(
m
m
)
KP(mx2)
CumulativedispalcementduringfirstCycle
35
o
C/km
Heatstep1
Heatstep2
Heatstep3
heatstep4
Heatstep5
Heatstep6
Heatstep7
heatstep8
heatstep9
heatstep10
heatstep11
heatstep11
heatstep12
heatstep13
heatstep14
steadystate
cooldown
250
200
150
100
50
0
50
0 500 1000 1500
A
x
i
a
l
d
i
s
p
l
a
c
e
m
e
n
t
(
m
m
)
KP(mx2)
CumulativedispalcementduringfirstCycle
20
o
C/km
Heatstep2
heatstep3
heatstep4
heatstep5
heatstep6
heatstep7
heatstep8
heatstep9
heatstep10
heatstep11
heatstep12
steadystate
cooldown
Figure 821: Cumulative displacement for 35
o
C/km and 20
o
C/km
8.4.4 Operational Effect of Walking due to Thermal Gradient

As stated previously, assuming that the pipeline is subjected to a total of 100 full shut-down and
start-up cycles over its lifetime, the pipeline for the case 35
o
C/km gradient would walk axially to
about 6m towards the cold end connection, while that for the case of 20
o
C/km gradient would
walk axially to about 4m towards the cold end for the same star-up and shut-down cycles.
Figure 822: Effect of walking due to thermal gradient in SCR

This effect could appear to be little in millimeters but for the lifetime of the field, the
accumulation of this axial displacement will become excessive and disastrous to the end
connections like SCR as illustrated Figuie 822.
Since higher thermal gradient means more walking per cycle, then the thermal gradient should be
reduced in order to reduce the rate of walking.
To reduce thermal gradient such that the rate of walking be reduced, a gradual warming of the
pipeline is required possibly to achieve some extent of uniform heating before reaching full
mobilization.
8.4.5 Control of Walking Phenomenon by Direct Electric Heating (DEH)
Direct electric heating (DEH) is the method direct heating of pipe through the use of electric
cables strapped onto the pipeline. The heat is generated by applied AC current from a power
supply located on a topside platform (see Figuie 82S).

Hotend
Coldend
Figure 823: Direct Electric heating of pipeline
crosssection (Harald, 2008)
DEH system has been used for several flow
assurance problems in controlling hydrate
formation mostly in multiphase flows in pipeline
by maintaining fluid temperature above the
hydrate formation temperature during shutdowns.
This is normally done by increasing the
temperature from ambient to above hydrate
formation temperature.
0bele Ifenna Isaac 0niveisity of Stavangei Page 1uS

Hypothesis:
Gradual warming of cold section of short pipeline reduces the effect of thermal gradient in
pipeline axial walking such that some extent of uniform heating is achieved prior to full
mobilization.
Task:
In order to use this measure as a means of controlling axial accumulation (walking),
investigations with FE analysis is carried out to determine the difference it yields by gradual
warming of the pipeline cold end section prior to start-ups and shutdowns.
Conditions:
Same thermal gradient of 35
o
C/km and the same heating steps are maintained. The cold end is
heated from 5
o
C ambient to a temperature of 20
o
C using direct electric heating (DEH) during the
start-up and shut down by allowing the tail end of heating steps to heat to 20
o
C until full
mobilization.
This is shown in the heating profile in Figuie 824:
Figure 824: Effect of DEH on thermal transient

Process: The pipeline is initially heated to 20
o
C by the DEH and the heating step maintains 20
o
C
at the cold end as shown in Figuie 824.
0
10
20
30
40
50
60
70
80
90
100
0 500 1000 1500 2000
T
e
m
p
e
r
a
t
u
r
e
(
C
)
KPdistance
EffectofDEHonThermalTransient
heatstep2
heatstep3
heatstep4
heatstep5
heatsteps6
heatstep7
heatstep8
heatstep9
heatstep10
heatstep11
heatstep12
Steadystate
heatstep11
heatstep12
heatstep13
Directelectricheatingof pipelinefrom5
o
C(ambient)to20
o
C.

The results from see Figuie 826 shows that prior to full mobilization, the pipeline with direct
electric heating along the pipeline expands at both hot end and cold ends while the other does not
remain cold but moves axially towards the cold end.

It is evident from Figuie 826, that the heating steps with DEH after full mobilization are almost
the same values as compared to the one without DEH. This shows that walking happens at the
heat-up sections and not after full mobilization. Hence, DEH reduces the walking by gradual
warming to uniform temperature of 20
o
C and the results are presented in Table 814:
Table 814: Results for both cases
With DEH Without DEH
Rate of walking (mm) 40 59.2
Thermal gradient (
o
C/km) 35 35
0,2
0
0,2
0 500 1000
A
x
i
a
l
d
i
s
p
l
a
c
e
m
e
n
t
(
m
)
KP(mx2)
WithDEH
heatstep4 heatstep5
0,2
0
0,2
0 500 1000
A
x
i
a
l
D
i
s
p
l
a
c
e
m
e
n
t
(
m
)
KP(mx2)
WithoutDEH
heatstep1 heatstep2
heatstep3 heatstep4
heatstep5
Figure 825: Axial displacement for both DEH and normal heating step
1,5
1
0,5
0
0,5
1
1,5
0 500 1000
A
x
i
a
l
d
i
s
p
l
a
c
e
m
e
n
t
(
m
)
KP(mx2)
WithDEH
heatstep1
heatstep2
heatstep3
heatstep4
heatstep5
heatstep6
heatstep7
heatstep8
heatstep9
heatstep10
heatstep11
heatstep12
heatstep14
Heatstep15
heatstep16 1,5
1
0,5
0
0,5
1
1,5
0 500 1000
A
x
i
a
l
D
i
s
p
l
a
c
e
m
e
n
t
(
m
)
KP(mx2)
WithoutDEH
heatstep1
heatstep2
heatstep3
heatstep4
heatstep5
heatstep6
heatstep7
heatstep8
heatstep9
heatstep10
heatstep11
heatstep12
heatstep13
heatstep14
Figure 826: Axial displacement for heating with DEH and without DEH
Early Expansion of cold end
Early Expansion of Hot end
8.4.6 Challenges facing the use of DEH

There are challenges with welcome development in the use of direct electric heating method as a
means of reducing the effect of walking in short pipelines.
Advantages in the use of DEH:
Direct electric heating system (DEH) has been utilized in subsea operations largely in the control
of flow assurance in multiphase flows in the pipeline. The use of the DEH in short flowline
where pipeline walking are mostly encountered will be of good help in reducing walking and at
the same using the mechanism to reduce hydrate formation within the pipeline.
The heating required from the DEH is only during the start-up and shutdown section and hence it
does not yield much operational cost once installed.
There will be need of installing the conventional heavy anchors if it could be designed to reduce
the walking as low as possible.
Figure 827: Direct Electric heating DEH (Harald, 2008)

Possible Negative impact :
Trawls gear/pulling on the cable. The DEH system can be expensive to install or operate.

9. Conclusions, Recommendations and Further Work

9.1 Summary and Conclusions
Lateral Buckling and Snake Lay Configuration

The design of 10km rigid pipeline for lateral buckling was carried out in accordance with the
design standards, DNV-RP-F110 and SAFEBUCK guidelines. The pipeline was assessed to be
susceptible to lateral buckling based on the calculations from Hobbs Analytical method and out-
of-straightness force. Based on these calculations, the region prone to buckle was identified and
the maximum allowable feed-in was deduced based on allowable design strain from the local
buckling criterion for displacement controlled condition. The results were checked such that the
predicted feed-in lengths are within the maximum allowable limits.
Having obtained the allowable design strain of 0.4% and post buckling force from the FE
analyses for a given lay radius, the number of snakes required for sharing the expansion under
thermal and pressure loading were established based on expansion sharing methodology given in
DNV-RP-F110. The feed-in is distributed among seven snakes such that the predicted feed-in
lengths are not exceeded the allowable values and expansion at the ends are reduced.
Table 91: Expansion Result
End Expansion- 10km Pipeline
Hot-end Cold-end
Before Snake lay 1.96m 0.96m
After snake-lay 0.80 0.6

The combination of snake lay and rock dumping in expansion sharing is a way of building up
effective force such that the critical buckling force will not be exceeded for any pipeline to
laterally buckle under the action thermal and pressure loading.
From the sensitivity analyses, it is deduced that the axial friction, initial imperfection, lay radius
and the buckle wavelength are important factors that influence the amount feed-in into a buckle.
Reducing the initial imperfection is a function of the available corridor of operation for the
project and increases the lay radius required for expansion sharing. The cost of laying several
snakes with longer radius and rock dumping is relatively less compared to laying more snakes
with smaller radius without rock dumping.
This is because lay barge takes longer time for negotiating a smaller bend radius than for a larger
radius.
Increasing the buckle length from 160m (R1500m) to 200m (R2500m) resulted in less initial
imperfection. From the analyses for the selected pipeline, this radius (R2500m) gives less feed-in
compared to the initial radius ((R1500m) but with rock dumping at only the hot-end.
Pipeline Walking
Pipeline walking has been studied in greater depth by SAFEBUCK J IP and a number of
deductions were generated. In the present thesis work, the influence of thermal gradient on
pipeline axial walking has been studied. FE analyses were performed for the selected pipeline
considering the two cases of thermal gradients. The predictions from the two cases are compared
and discussed with respect to the rate of pipeline walking toward cold end.
The calculations using the analytical method based on SAFEBUCK guidelines were performed
and compared against the predictions from FE analyses. Based on the results, it can be confirmed
that analytical relations do not give an accurate estimation of rate of walking, but can be used as
basis for preliminary assessment whether to perform an FE analysis.
Reason for discrepancy between analytical and FE results being that the analytical relations are
based on certain assumptions which cannot be generalized for all pipeline operation cases.
Findings:
The critical buckling force at lower bound friction has been seen as better basis for
pipeline buckling design such that where the pipe-soil interaction are not properly
estimated, the system will still be able to withstand the worst case effect in any failure
mode.

Use of post-lay rock dumping with large lay radius for expansion sharing has been seen
as a better means of lateral buckling control measure. This is especially where the
operational corridor is smaller such that the distance between successive buckles are less
than the total buckle length.

Pre-warming the pipeline system is a greater way of reducing the effect of thermal
gradient on short pipeline before reaching full mobilization.

Use of di-electric heating (DEH) of the short pipeline before operational start could assist
in reducing thermal gradient along the pipeline.

9.2 Recommendations

As a result of the uncertainties surrounding pipe-soil interactions, most long pipelines are
found to be susceptible to lateral buckling. Excessive build-up of effective axial force by the
frictional force from the soils and the extreme high pressure and high temperature (XHPHT)
have been a major driving factor as we drill deep for more oil and gas.
Moreover, the need to maintain the temperature and pressure of the well effluent in order to
avoid the flow assurance issues has made the solution more critical. Hence, pipelines are allowed
to move laterally at will provided that the integrity of the pipe is not jeopardized.
Allowing the pipeline to move laterally without restriction by the use of snake lay configuration
is a great way of sharing expansion such that the total allowable feed-in is distributed among the
snakes generated by the configuration applied.
In order to reduce the cost of lay barge which cost more in negotiating a smaller bend radius,
longer lay radius with post-lay rock dumping can be applied thereby reducing the operational
corridor which could act adversely in case of unforeseen lateral movement of the pipeline.
Axial accumulation at the cold end can be so disastrous if allowed to go on till the life of the
system and it should be mitigated. Pre-warming the pipeline system is a greater way of reducing
the effect of thermal gradient on short pipeline before reaching full mobilization.
The thermal gradient of pipeline installation should always be determined and the effect
calculated based on the axial frictional forces so that the effect can be reduced or the pipeline re-
routed to ensure safety of the connecting ends.
9.3 Further Work

As mentioned above, there are uncertainties hidden on the pipe-soil interaction which was
evident on the results from the post buckling force at different soil frictions. A study should be
carried out to generate a better way of deducing a better approximation of the pipe-soil
movement with respect to temperature profile at extreme well pressures.
The use of di-electric heating (DEH) with pipe-in-pipe system c considered as a way of reducing
the effect of thermal gradient along short pipeline such that the accumulation of the axial
displacement is greatly reduced over the life of the field.


REFERENCES
2B1STCONSULTING. 11 September, 2012. Hess progresses on Australia deep offshore Equus

LNG project [Online]. Available: http://www.2b1stconsulting.com/hessprogresseson
australiadeepoffshoreequuslngproject/[AccessedJanuary282013].
AHMED,M.R.&&GARETH,L.F.2012.InvestigatingintotheDynamicEffectofLateralBuckling
of High Temperature/High Pressure Offshore Pipelines. Proceedings of Acoustic 2012
Fremantle
ALMEIDA,M.S.S.,ETAL2001.SoftSoilEngineering,Swets&Zeitlingler
BRUTON,D.,CARR,M.,CRAWFORD,M.&POIATE,E.2005.TheSafeDesignofHotOnbottom
Pipeline with Lateral Buckling using the Design Guideline developed by the SAFEBUCK
jointIndustryProject.DeepOfshoreTechnologyConference.Vitoria,EsiritoSanto,Brazil.
BRUTON,D.,CARR,M.&WHITE,D.J.2007.TheInfluenceofofPipeSoilInteractiononLateral
Buckling and Walking of Pipelines SAFEBUCK JIP. International Ofshore Site
InvestigationandGeotechnicsConferences.London,UK.
BRUTON, D., SINCLAIR, F. & CARR, M. 2010. Lessons Learned From Observing Walking of
Pipelineswith Lateral Buckles, Including New Driving Mechanism and Updated Models.
OffshoreTechnologyConference.Houston,Texas,USA.
BRUTON, D. A. S. & CARR, M. 2011. Overview of the SAFEBUCK JIP. In: OTC (ed.) Offshore
TechnologyConference.Houston,Texas,USA.
CARNEIRO, D. & CASTELO, A. 2010. THERMOMECHANICAL ANALYSES OF HP/HT PIPELINES
WITH SLIDING FOUNDATION END STRUCTURE. In: VERITAS, B. (ed.) 23o Congresso
NacionaldeTransporteAquaviario,ConstrucaoNavaleOffshore.RiodeJaneiro,Brazil.
CARR,M.,SINCLAIR,F.&BRUTON,D.2006.PipelineWalkingUnderstandingtheFieldLayout
Challenges, and Analytical Solutions Developed for the SAFEBUCK JIP. Offshore
TechnologyConference(OTC)HoustonTexas,USA
CHAUDHURY, G. Managaing Unidiretional Movements (WALK) of HPHT Submarine Flowline
During Startup heating and Shutdown Cooling. International Offshore Pipeline Forum
IOPF,2010Houston,Texas,USA.
CHEUK,J.2007.AdvancedModelsforSoilPipelineInteraction.Subsea&Pipelines.Department
ofBuildingandConstruction,CityUniversityofHongKong.
DNVOSF101,D.N.V..2012.SubmarinePipelineSystemsHvik,Norway:(FM+SGML)
DNVRPF110,D.N.V..2007.GlobalBucklingofSubmarinePipelines.StructuralDesigndueto
HighTemperature/HighPressure.Hvik,Norway:(FM+SGML)
ECOPRASINOS. 2012. Subsea Pipelines [Online]. EcoPrasino. Available:
http://www.ecoprasinos.com/services/subseapipeline[AccessedFebruary13,2013.
EINSFELD, R. A., MURRAY, D. W. & YOOSEFGHODSI, N. 2003. Buckling analysis of high
temperature pressurized pipelines with soilstructure interaction. Journal of the
BrazilianSocietyofMechanicalSciencesandEngineering,vol.25,n.2,pp.164169.
FLORIANO, C., ABED EL, C., STEFANO, G. & ANTONIO, C. 2011. Characterization of Pipe Soil
Interaction and Influence on HPHT Pipeline Design. In: WWW.ISOPE.ORG (ed.)
International Offshore and Polar Engineering Conference. Hawaii, USA: International
SocietyofOffshoreandPolarEngineers(ISOPE).
FYRILEIV, O. & COLLBERG, L. Influence of Pressure in Pipeline Design Effective Axial Force.
International Conference on Offshore Mechanics and Arctic Engineering (OMAE 2005),
2005Halkidiki,Greece.
GUO,B.,SONG,S.,CHACKO,J.&GHALAMBOR,A.2005.OffshorePipelines,Elsevier.
HARALD, K. 2008. Direct Electric HEating (DEH) Basic technology. In: RESEARCH, S. E. (ed.).
SINTEFEnergyResearch.
HOBBS,R.1984.InServiceBucklingOfHeatedPipelines.JournalofTransportationEngineering,
Vol.110,No.2.
KARUNAKARAN, D. 2012. Lecture nots in MOK 160: Pipelines and Risers. University of
Stavanger,Norway.
KAYE,D.1996.LateralBucklingofSubseaPipelines:ComparismbetweenDesignandOperation.
UK:SocietyforUnderwaterTechnology.
KEIN, L. K., MING, S. L. & MASCHNER, E. Design of High Temperature/High Pressure (HT/HP)
PipelineagainstLateralBuckling.
KYRIAKIDES, S. & CORONA, E. 2007. Mechanics of Offshore Pipelines Volume 1: Buckling and
Collapse,Elsevier.
OFFSHOREVN. 2010. End Expansion and Local Buckling [Online]. Available:
http://www.slideshare.net/Offshorevn/endexpansionanalysis [Accessed February, 20
2013].
ONDREJ,K.2012.Buckling[Online].Available:https://wiki.csiberkeley.com/display/kb/Buckling
[AccessedFebruary,192013].
PALMER,A.C.2004.LateralBucklingofAxiallyConstrainedPipeline.JPTForum2.
PALMER, A. C. & KING, R. A. 2004. Subsea Pipeline Engineering, Tulsa, Oklahoma, PennWell
Corporation.
PROF. SHARMA, S. C. Strength of Material [Online]. Available:
http://nptel.iitm.ac.in/courses/Webcoursecontents/IIT
ROORKEE/strength%20of%20materials/lects%20&%20picts/image/lect15/lecture15.ht
m[AccessedFeruary,222013].
ROBERT, M. J. Buckling [Online]. Wikipedia. Available: http://en.wikipedia.org/wiki/Buckling
[AccessedFebruary,192013].
RONG,H.,INGLIS,R.,BELL,G.,HUANG,Z.&CHAN,R.2009.EvaluationandMitigationofAxial
Wlaking with a Focus on Deep Water Flowlines Offshore Technology Conference (OTC).
Houston,Texas,USA:OTC.
RUNDSAG,J.O.,TRNES,K.,CUMMING,G.,A.D.,R.&ROBERTS,C.2008.OptimisedSnakedLay
Geometry.InternationalSocietyofOffshoreandPolarEngineers(ISOPE).vancouver,BC,
Canada.
SAFEBUCKJIP 2011. SAFEBUCK III Safe Design of Pipelines with Lateral buckling Design
Guideline.
SRISKANDARAJAH, T., DONG, S., SRIBALACHANDRAN, S. & WILKINS, R. Effect of Initial
ImperfetcionontheLateralBucklingofSubseaPipelines.In:(ISOPE),I. S.O.O.A.P.E.,
ed. Proceedings of the Ninth (1999) International Society of Offshore and Polar
Engineers(ISOPE),1999Brest,France.
SRISKANDARAJAH,T.,RAGUPARTHY,P.&WILKINS,R.DynamicVersusStaticBucklingofSubsea
Pipelines. In: (ISOPE), T. I. S. O. O. A. P. E., ed. International Offshore and Polar
Conference,2001Stavanger,Norway.
SUBSEA7 2011. Techninal Guideline: Pipeline Expansion In: 7, S. (ed.) Doc. No: GDGLPDCOE
006.2011ed.
SUBSEA7 2012. Lateral Buckling and Pipeline Walking. In: NORWAY, S. (ed.) LONCENGEPLN
0001.
SREN,H.&YONG,B.1999.BendingMomentCapacityofPipes.JournalofOffshoreMechnics
andArcticEngineering.
TAKAHASHI, K., ANDO, K., HISATSUNE, M. & HASEGAWA, K. 2007. Failure behavior of carbon
steel pipe with local wall thinning near orifice. Nuclear Engineering and Design, 237,
335341.
WHITE, D. J. & BRUTON, D. A. 2008. PipeSoil Interaction During Lateral Buckling and Pipeline
walkingSAFEBUCKJIP.In:OTC(ed.)OffshoreTechnnologyConference(OTC).Houston,
Texas,USA.
YONG,B.&QIANG,B.2005.SubseaPipelinesandRisers,SanDiego,USA,ELsevier.

A1|P a g e
APPENDIX A: CALCULATION RESULTS
APPENDIX A
A11|P a g e
APPENDIX A1
PIPELINE END EXPANSION
Pipeline End Expansion Calculations

MSc Subsea Technology, University of Stavanger, Norway
Date : 8 March, 2013
Author : Obele Ifenna
Description :
The Mathcad analysis worksheet presented in this report is used for Pipeline end expansion calculation of
a rigid pipeline under thermal and pressure loading. The purpose is to deduce the maximum expansion that
an end connection can take during maximum loading. The Calculations are based on Subsea 7 Pipeline
expansion guideline: CEO1PD-P-GU-126 and DNV-OS-F101
Limitation:
Pipeline Susceptibility to lateral buckling and walking are not considered at this stage.
Units: MPa 1N mm
2
g 9.81 m s
2
MN 1 10
6
N
Pipeline Data:
D
o
559mm :=
Pipeline Outside Diamter
Wall Thickness
t
wall
19.1mm :=
t
ext.coat
5mm :=
External Coating Thickness
Concrete Coating Thickness
Length of Pipeline
Material Properties:
Pipeline:
Pipe Steel Density
SMYS Steel Pipe
Steel young's Modulus
Steel Pipe Thermal Expansion Coefficient
Steel Poisson Ratio
Insulation or Coating:
Insulation or Coating Density
Concrete Coating Density
Operating Parameters:
Sea Water Density
Max Content Density
Design Pressure
Operating Pressure
Hydrotest Pressure
t
conc
55mm :=
L 2000m :=
st
7850kg m
3
:=
SMYS 450MPa :=
E 2.07 10
5
MPa :=
1.17 10
5
C
1
:=
0.3 :=
inscoat
910kg m
3
:=
conc
2400.kg m
3
:=
water
1027kg m
3
:=
cont
900kg m
3
:=
P
d
15MPa :=
P
op
15MPa :=
P
hyd
0MPa :=
E:\thesis\mathcad\Endexpnasion 1 of 10
T
amb
5 C :=
Sea water Ambient Temperature
T
op
95 C :=
Operating Temperature
External Loads:
M
b
0kN m :=
Bending moment
Axial Force
N
a
0kN :=
Residual Lay Tension
Soil Properties:
Lateral Friction Factor
Safety Factors:
Usage Factor for Hoop stress
Usage Factor for Longitudinal stress
N
lay
0kN :=
axial
0.5 :=
lateral
0.8 :=
h
0.72 :=
L
0.8 :=
Parameter Calculations:
Internal Diameter
Effective Pipe Diameter
Cross-sectional Area of Steel Pipe
Cross-sectional Area of external coating
Cross-sectional Area of Concrete coating
Mass of Steel Pipe
Mass of External coating
Mass of concrete coating
Mass of Content
Mass of water Content
Mass in water (Bouyancy Mass)
Total Mass in air
D
in
D
o
2 t
wall
:=
D
eff
D
o
2 t
ext.coat
t
conc
+
( )
+ :=
A
s

4
D
o
2
D
in
2
|
\
|
.
|
\
|
.
:=
A
ext

4
D
o
2 t
ext.coat
+
( )
2
D
o
2
:=
A
conc

4
D
o
2 t
ext.coat
+ 2 t
conc
+
( )
2
D
o
2 t
ext.coat
+
( )
2
:=
M
st
A
s
st
:=
M
ext.coat
A
ext
inscoat
:=
M
conc
A
conc

conc
:=
M
cont

4
D
in
2

cont
:=
M
water

4
D
in
2

water
:=
M
bouyancy

4
D
eff
2

water
:=
M
air
M
st
M
conc
+ M
ext.coat
+ :=
Submerged Mass
Weight of Dry Pipe
Weight of Content
Weight of Submerged Pipe
Temperature Difference
Moment of Inertia of Steel Pipe
cross section
Sectional Modulus of Steel Pipe
M
submerged
M
air
M
bouyancy
:=
W
dry
M
air
g :=
W
cont
M
cont
g :=
W
submerged
M
cont
g M
ext.coat
g + M
st
g + M
conc
g + M
bouyancy
g :=
W
submerged
3.345
1
m
kN =
T T
op
T
amb
:=
I
s

4
D
o
4
D
in
4
|
\
|
.
:=
Z
s
I
s
D
o
2
:=
EFFECTIVE AXIAL FORCE CALCULATION
559 x 19.1mm PIPE OF 2km Pipe Length 1.
Kp
0
0m :=
n 6 := KP
step
100m := k 0 2000 .. :=
Kp
n 1
2000m :=
T
k
5C ( ) 90C ( ) e
0.0000911607 k
+ :=
x Kp
0
KP
step
, Kp
n 1
.. :=
x
k
k 1 m :=
0 500 1000 1500 2000
70
80
90
100
Temperature profile
Design Temperature Profile
KP distance (m)
T
e
m
p
e
r
a
t
u
r
e

(
C
)
Water depth
WD
800m
800m
|
\
|
|
.
:=
KPw
0km
2km
|
\
|
|
.
:=
WD k ( ) linterp KPw WD , x
k
, ( ) :=
P
o
k ( )
water
g WD k ( ) :=
External Pressure
Internal Pressure P
in
k ( ) P
op

cont
g WD k ( ) + :=
P k ( ) P
in
k ( ) P
o
k ( ) :=
Pressure Difference with KP
Thermal Force N
thermal
k ( ) E A
s
T
k
:=
Poisson Effect Force
N
poisson
k ( ) P k ( ) A
s
D
o
t
wall
( )
2t
wall
(
(
:=
End Cap Force N
Endcap
k ( )

4
P
in
k ( )
( )
D
in
2
P
o
k ( )
( )
D
o
2

:=
N
eff
k ( ) N
lay
N
Endcap
k ( ) N
poisson
k ( ) + N
thermal
k ( ) + :=
Effective Axial Force with KP
F
anchor
D
in
t
wall
E T
op
T
amb
( )
1 2 ( )
P
op
D
in
2
(
(
+ := F
anchor
8.09 MN =
Restraining Force: F
friction

axial
W
submerged
( )
1 m :=
W
submerged
3.345 10
3
1
m
N =
F
friction
1.673 10
3
N =
L
anchor
F
anchor
axial
W
submerged
( )
:= L
anchor
4.837 km =
Since the Restraining Force (Friction Force) cannot attain the Fully constraint Axial Force
(Anchoring Force), then pipeline is termed a Short Pipeline (fully mobilised)
Short Pipeline: Pipeline which will never develop the full constrain force
Effective Axial Force - Fully mobilized Pipeline
Maximum Friction Force at the mid-line:
F
MAXfriction

axial
W
submerged
W
cont
+
( )
L
2
:=
F
MAXfriction
2.613 10
6
N =
F
HOTfriction
k ( )
axial
W
submerged
W
cont
+
( )
x
k
1 ( ) :=
Friction Force with Length at the hot end:
Friction Force with Length at the cold
end:
F
COLDfriction
k ( )
axial
W
submerged
W
cont
+
( )
x
k
L ( ) :=
The Frictional Restrained Force along the Pipeline full
Length:
F
Restfriction
k ( ) if F
HOTfriction
k ( ) F
COLDfriction
k ( ) > F
HOTfriction
k ( ) , F
COLDfriction
k ( ) ,
( )
:=
N
effSHORT
k ( ) if N
eff
k ( ) F
Restfriction
k ( ) < F
Restfriction
k ( ) , N
eff
k ( ) ,
( )
:=
0 0.5 1 1.5 2
10000000
8000000
6000000
4000000
2000000
0
Fully constrain
Effective Force
Effective Axial Force for 2km Flowline
KP distance (km)
F
o
r
c
e

(
N
)
EFFECTIVE FORCE FOR A 559 x 19.1mm Pipe of 10km
Pipelength
n 6 := KP
Lstep
0.001km :=
L
longKpL
n 1
10km :=
i 0 n 1 .. :=
Kp
0
0m = j 0 1 , 10000 .. :=
x
i
Kp
0
KP
Lstep
, L
longKpL
n 1
.. := T
j
5C ( ) 90C ( ) e
0.0000492476 j
+ :=
x
j
j 1 m :=
0.0 2.0 4.0 6.0 8.0 10.0
50.0
60.0
70.0
80.0
90.0
Design Temperature Profile - 10km Pipeline
KP distance (km)
T
e
m
p
e
r
a
t
u
r
e

(
C
)
Water depth
nw 2 :=
iw 0 nw 1 .. :=
WDL
800m
800m
|
\
|
|
.
:=
KPLw
0km
10km
|
\
|
|
.
:=
WDL j ( ) linterp KPLw WDL , x
j
, ( ) :=
N
critical
3.010 MN :=
P
oL
j ( )
water
g WDL j ( ) :=
External Pressure
Internal Pressure P
inL
j ( ) P
op

cont
g WDL j ( ) + :=
PL j ( ) P
inL
j ( ) P
oL
j ( ) :=
Thermal Force N
thermalL
j
E A
s
T
j
:=
N
poissonL
j ( ) PL j ( ) A
s
D
o
t
wall
( )
2t
wall
(
(
:=
N
EndcapL
j ( )

4
P
inL
j ( )
( )
D
in
2
P
oL
j ( )
( )
D
o
2

:=
End Cap Force
Effective Axial Force with KP N
FULL
j ( ) N
lay
N
EndcapL
j ( ) N
poissonL
j ( ) + N
thermalL
j
+ :=
Friction Force with Length at the hot end: F
HOTfrictionL
j ( )
axial
W
submerged
W
cont
+
( )
x
j
1 ( ) :=
Friction Force with Length at the cold end:
F
COLDfrictionL
j ( )
axial
W
submerged
W
cont
+
( )
x
j
x
10000
( ) :=
The Frictional Restrained Force along the Pipeline full Length:
F
RestfrictionL
j ( ) if F
HOTfrictionL
j ( ) F
COLDfrictionL
j ( ) > F
HOTfrictionL
j ( ) , F
COLDfrictionL
j ( ) ,
( )
:=
N
EFF
j ( ) out N
FULL
j ( ) F
RestfrictionL
j ( ) N
FULL
j ( ) < if
F
RestfrictionL
j ( ) otherwise
:=
0 2 4 6 8 10
1 10
7
8 10
6
6 10
6
4 10
6
2 10
6
0
Effective Axial Force
Plot of Effective Axial Force
KP distance (km)
F
o
r
c
e

(
M
N
)
0 2000 4000 6000 8000 10000
10000000
8000000
6000000
4000000
2000000
0
Critical Buckling Force
Effective Axial force
N
critical
N
EFF
j ( )
j
0 2 4 6 8 10
1.5 10
7
1 10
7
5 10
6
0
Fully constrained axial force
Effective Force
Friction Force
Effective Axial Force of a Long Pipeline
KP Distance (km)
A
x
i
a
l

F
o
r
c
e

(
N
)
N
FULL
3028 ( ) 7.274 MN =
F
RestfrictionL
3028 ( ) 7.912 10
6
N =
N
FULL
7469 ( ) 6.079 MN = F
RestfrictionL
7469 ( ) 6.613 10
6
N =
VAP
1
3272m := VAP
2
7210m :=
VIRTUAL ANCHOR POINT
From the Tables and datas above, it can deduced that the virtual Anchor Point for the Long
pipeline where - Fully Constrained Axial Force Equals the Frictional Force are:
VAP
1
= KP 3.272 -
VAP
2
= KP 7.210 -
Therefore, the anchor Length for the long pipeline are now:
L_anchor (1) = The distance betwen KP
0
and VAP
1
L_anchor (2) = The distance betwen KP
n-1
and VAP
2
For the Long Pipeline
L
anchorHotEnd
VAP
1
Kp
0
:= L
anchorColdEnd
L
longKpL
n 1
VAP
2
:=
Anchor Length at the hot end:
Anchor Length at the cold End:
L
anchorHotEnd
3.272 km = L
anchorColdEnd
2.79 km =
For the Short Pipeline
F
anchor
D
in
t
wall
E T
op
T
amb
( )
1 2 ( )
P
op
D
in
2
(
(
+ :=
*
F
anchor
8.09 MN =
F
friction

axial
W
submerged
:=
F
friction
1.673
1
m
kN =
L
anchor
F
anchor
F
friction
:= L
anchor
4.837 km =
Since the anchor length 6.417km is greater than the length of the pipeline (2km), this is a prove that
this is a short Pipeline having insufficient friction to attain the full constrain Axial Force. Hence, it is
Fully mobilised.
F
frictional
F
friction
1 m :=
Total Frictional Force F
frictional
1.673 10
3
MN =
END EXPANSION CALCULATION
Based on the Long Pipeline analysis, the end expansion can be deduce as the following for
each ends:
The Expansion at the Hot end:
expansionHOT
3028
0
j
N
FULL
j ( ) F
RestfrictionL
j ( )
( )
1 m
E A
s
=
:=
expansionHOT
1.714m =
expansionCOLD
10000
7469
j
N
FULL
j ( ) F
RestfrictionL
j ( )
( )
1 m
E A
s
=
:=
The Expansion at the Cold
end:
expansionCOLD
0.936 m =
For Short Pipeline analysis, the end expansion can be deduce as the following for each ends:
expansionHOTshort
1000
0
k
N
eff
k ( ) F
Restfriction
k ( )
( )
1 m
E A
s
=
:=
expansionHOTshort
0.99m =
expansionCOLDshort
2000
1000
k
N
eff
k ( ) F
Restfriction
k ( )
( )
1 m
E A
s
=
:=
expansionCOLDshort
0.902m =
EXPANSION OFFSET DESIGN
The expansion offset is configured to transfer the expansion displacement into a bending stress, below the
code allowable. Typical offset configurations include 'L, 'U" or "Z shaped spools.
For an L-shaped spool the bending is idealised as a pinned / fixed cantilever and the allowable stress is
taken from the combination of hoop, end-cap and longitudinal (incl. temperature stress):
Maximum allowable bending stress:
Bending
F
anchor
A
s
:=
The minimum Spool length for an L shaped spool according Subsea7 guideline is given as:
L
min
2.25 E D
o

expansionHOT
Bending
:=
L
min
42.275m =
Temperature de-rated steel Yield Stress
T
d
0C 1C , 200C .. := SMYS T
d
( )
SMYS T
d
50C s if
SMYS 3
T
d
50C
( )
5
(
(
MPa
C
(
(
50C T
d
< 100C s if
SMYS
T
d
100C
( )
2.5
(
(
MPa
C
30MPa +
(
(
otherwise
:=
0 50 100 150 200
350
400
450
500
De-rated Steel Yield Stress
Temperature (C)
S
M
Y
S

(
M
P
a
)
SMYS T
op
( )
423 MPa =
A21|P a g e
A21|P a g e
APPENDIX A2
HOBBS CRITICAL BUCKLING FORCE
Susceptibility of Pipeline to Lateral Buckling

Hobbs Critical Buckling
Description :
The Mathcad analysis worksheet presented here is used to calculate the Pipeline susceptibility to lateral
buckling due to axial loading from temperature, pressure and frictional resistance.
The Calculations are based on SAFEBUCK DESIGN GUIDELINE which integrated the DNV-RP-F110
using Hobbs Critical Buckling Force.
Assumptions:
The Steel yield stress is only temperature de-rated
Straight Pipeline is considered
Initially perfect Pipe buckles at an indefinte series of half waves.
Lateral Frictional force is fully mobilized
Limitation:
The concrete coating is not applicable in this case.
The Lateral restrain applicable here is only the friction resistance forces
References:
Hobbs, R. E., 'In Service Buckling of heated pipelines', Journal of Transport Engineering, Vol 110, 1.
No. 2, March 1984
Carr, M., Bruton, D., & Baxter, D., 'Safe Design of Pipeline with Lateral Buckling', SAFEBUCK III, 2.
DESIGN GUIDELINE. July, 2011
Units: MPa 1N mm
2
g 9.81 m s
2
MN 1 10
6
N
Pipeline Data:
D
o
559mm :=
Wall Thickness
t
wall
19.1mm :=
t
ext.coat
5mm :=
Length of Pipeline
Pipeline:
Pipe Steel Density
SMYS Steel Pipe
Steel Poisson Ratio
t
conc
55mm :=
L 2000m :=
st
7850kg m
3
:=
SMYS 450MPa :=
E 2.07 10
5
MPa :=
1.17 10
5
C
1
:=
0.3 :=
inscoat
910kg m
3
:=
conc
2400.kg m
3
:=
E:\thesis\mathcad\Hobbs Analysis 1 of 10
Sea Water Density
Max Content Density
Design Pressure
Operating Pressure
Hydrotest Pressure
water
1027kg m
3
:=
cont
900kg m
3
:=
P
d
15MPa :=
P
op
15MPa :=
P
hyd
0MPa :=
T
amb
5 C :=
T
op
95 C :=
External Loads:
M
b
0kN m :=
Bending moment
Axial Force
N
a
0kN :=
Residual Lay Tension
Soil Properties:
Safety Factors:
Usage Factor for Hoop stress
Usage Factor for Longitudinal stress
N
lay
0kN :=
axial
0.5 :=
lateral
0.8 :=
h
0.72 :=
L
0.8 :=
Internal Diameter
Mass of Steel Pipe
Mass of Content
D
in
D
o
2 t
wall
:=
D
eff
D
o
2 t
ext.coat
t
conc
+
( )
+ :=
A
s

4
D
o
2
D
in
2
|
\
|
.
|
\
|
.
:=
A
ext

4
D
o
2 t
ext.coat
+
( )
2
D
o
2
:=
A
conc

4
D
o
2 t
ext.coat
+ 2 t
conc
+
( )
2
D
o
2 t
ext.coat
+
( )
2
:=
M
st
A
s
st
:=
M
ext.coat
A
ext
inscoat
:=
M
conc
A
conc

conc
:=
M
cont

4
D
in
2

cont
:=
M
water

4
D
in
2

water
:=
Total Mass in air
M
bouyancy

4
D
eff
2

water
:=
M
air
M
st
M
conc
+ M
ext.coat
+ :=
Submerged Mass
Weight of Dry Pipe
Weight of Content
Moment of Inertia of Steel Pipe
cross section
Sectional Modulus of Steel Pipe
M
submerged
M
air
M
bouyancy
:=
W
dry
M
air
g :=
W
cont
M
cont
g :=
W
submerged
M
cont
g M
ext.coat
g + M
st
g + M
conc
g + M
bouyancy
g :=
W
submerged
3.345
1
m
kN =
T T
op
T
amb
:=
I
s

D
o
4
D
in
4
|
\
|
.
64
:=
Z
s
I
s
D
o
2
:=
EFFECTIVE AXIAL FORCE CALCULATION
559 x 19.1mm PIPE OF 2km Pipe Length 1.
Kp
0
0m :=
n 6 := KP
step
100m := k 0 2000 .. :=
Kp
n 1
2000m :=
T
k
5C ( ) 90C ( ) e
0.0000911607 k
+ :=
x Kp
0
KP
step
, Kp
n 1
.. :=
x
k
k 1 m :=
Water depth
WD
800m
800m
|
\
|
|
.
:=
KPw
0km
2km
|
\
|
|
.
:=
WD k ( ) linterp KPw WD , x
k
, ( ) :=
P
o
k ( )
water
g WD k ( ) :=
External Pressure
Internal Pressure P
in
k ( ) P
op

cont
g WD k ( ) + :=
P k ( ) P
in
k ( ) P
o
k ( ) :=
Thermal Force N
thermal
k ( ) E A
s
T
k
:=
N
poisson
k ( ) P k ( ) A
s
D
o
t
wall
( )
2t
wall
(
(
:=
End Cap Force N
Endcap
k ( )

4
P
in
k ( )
( )
D
in
2
P
o
k ( )
( )
D
o
2

:=
N
eff
k ( ) N
lay
N
Endcap
k ( ) N
poisson
k ( ) + N
thermal
k ( ) + :=
Effective Axial Force with KP
F
anchor
D
in
t
wall
E T
op
T
amb
( )
1 2 ( )
P
op
D
in
2
(
(
+ := F
anchor
8.09 MN =
Restraining Force: F
friction

axial
W
submerged
( )
1 m :=
W
submerged
3.345 10
3
1
m
N =
F
friction
1.673 10
3
N =
L
anchor
F
anchor
axial
W
submerged
( )
:= L
anchor
4.837 km =
Since the Restraining Force (Friction Force) cannot attain the Fully constraint Axial Force
(Anchoring Force), then pipeline is termed a Short Pipeline (fully mobilised)
Short Pipeline: Pipeline which will never develop the full constrain force
Effective Axial Force - Fully mobilized Pipeline
Maximum Friction Force at the mid-line:
F
MAXfriction

axial
W
submerged
W
cont
+
( )
L
2
:=
F
MAXfriction
2.613 10
6
N =
F
HOTfriction
k ( )
axial
W
submerged
W
cont
+
( )
x
k
1 ( ) :=
Friction Force with Length at the hot end:
Friction Force with Length at the cold
end:
F
COLDfriction
k ( )
axial
W
submerged
W
cont
+
( )
x
k
L ( ) :=
The Frictional Restrained Force along the Pipeline full
Length:
F
Restfriction
k ( ) if F
HOTfriction
k ( ) F
COLDfriction
k ( ) > F
HOTfriction
k ( ) , F
COLDfriction
k ( ) ,
( )
:=
N
effSHORT
k ( ) if N
eff
k ( ) F
Restfriction
k ( ) < F
Restfriction
k ( ) , N
eff
k ( ) ,
( )
:=
EFFECTIVE FORCE FOR A 559 x 19.1mm Pipe of 10km
Pipelength
n 6 := KP
Lstep
0.001km :=
L
longKpL
n 1
10km :=
i 0 n 1 .. :=
Kp
0
0m = j 0 1 , 10000 .. :=
x
i
Kp
0
KP
Lstep
, L
longKpL
n 1
.. := T
j
5C ( ) 90C ( ) e
0.0000492476 j
+ :=
x
j
j 1 m :=
Water depth
WDL
800m
800m
|
\
|
|
.
:=
nw 2 :=
iw 0 nw 1 .. :=
KPLw
0km
10km
|
\
|
|
.
:=
WDL j ( ) linterp KPLw WDL , x
j
, ( ) :=
N
critical
2.017 MN :=
P
oL
j ( )
water
g WDL j ( ) :=
External Pressure
Internal Pressure P
inL
j ( ) P
op

cont
g WDL j ( ) + :=
PL j ( ) P
inL
j ( ) P
oL
j ( ) :=
Thermal Force N
thermalL
j
E A
s
T
j
:=
N
poissonL
j ( ) PL j ( ) A
s
D
o
t
wall
( )
2t
wall
(
(
:=
N
EndcapL
j ( )

4
P
inL
j ( )
( )
D
in
2
P
oL
j ( )
( )
D
o
2

:=
End Cap Force
Effective Axial Force with KP N
FULL
j ( ) N
lay
N
EndcapL
j ( ) N
poissonL
j ( ) + N
thermalL
j
+ :=
Friction Force with Length at the hot end: F
HOTfrictionL
j ( )
axial
W
submerged
W
cont
+
( )
x
j
1 ( ) :=
Friction Force with Length at the cold end:
F
COLDfrictionL
j ( )
axial
W
submerged
W
cont
+
( )
x
j
x
10000
( ) :=
The Frictional Restrained Force along the Pipeline full Length:
F
RestfrictionL
j ( ) if F
HOTfrictionL
j ( ) F
COLDfrictionL
j ( ) > F
HOTfrictionL
j ( ) , F
COLDfrictionL
j ( ) ,
( )
:=
N
EFF
j ( ) out N
FULL
j ( ) F
RestfrictionL
j ( ) N
FULL
j ( ) < if
F
RestfrictionL
j ( ) otherwise
:=
N
FULL
3028 ( ) 7.274 MN =
F
RestfrictionL
3028 ( ) 7.912 10
6
N =
N
FULL
7469 ( ) 6.079 MN = F
RestfrictionL
7469 ( ) 6.613 10
6
N =
VAP
1
3272m := VAP
2
7210m :=
BUCKLING PROGRAMMING
0 1 2 3
0
1
2
3
"Direction" "Minimum" "Median" "Maximum"
"Axial" 0.3 0.45 0.5
"Static Lateral" 0.6 0.7 0.8
ynamic Lateral" 0.43 0.85 1.28
:=
REF
2
HOBBS LATERAL BUCKLING ANALYSIS
Inserting the Constants for lateral Buckling modes
K
0 1 2 3 4 5
0
1
2
3
4
5
"Mode" "K1" "K2" "K3" "K4" "K5"
1 80.76
-5
6.39110 0.5
-3
2.40710
-2
6.93810
2 39.478
-4
1.74310 1
-3
5.53210 0.109
3 34.06
-4
1.66810 1.294
-2
1.03210 0.143
4 28.2
-4
2.14410 1.608
-3
1.04710 0.148
"infinity" 39.478
-5
4.70510
-5
4.70510
-3
4.49510
-2
5.06610
:=
REF
1
Case 1: Infinite mode Lateral buckling with lateral friction of = 0.8
Lateral Friction
Ref 1, Equat. 22
Buckle Wave Length L
buckle
2.7969 10
5
E I
s
( )
3
2 3 ,
W
submerged
( )
2
A
s
E
(
(
(
0.125
:=
L
buckle
55 m =
L
buckle
0.5 28m = L
buckle
1.5 83m =
n
x
0.5 := m
x
2.5 :=
Accordning Hobbs recommendation the Buckle wave length will be manipulated within a range of 20
values between 0.5 of buckle wave length to 1.5 of the wavelength
L
bw
n
x
L
buckle
m
x
L
buckle
n
x
L
buckle

( )
20
(
(
n
x
L
buckle
+
(
(
, m
x
L
buckle
.. :=
Ref 1, Equat. 20
P
buckle
L
bw
( )
K
5 1 ,
E I
s
L
bw
2
:=
Reduced Axial Force within Buckle
Axial force due to Thermal expansion Ref 1, Equat. 21
P
o_infinity
L
bw
( )
P
buckle
L
bw
( )
4.705 10
5
A
s
E

2 3 ,
W
submerged
E I
s
\
|
|
.
2
L
bw
6
+ :=
The Buckle Amplitude
y
max
L
bw
( )
K
5 4 ,
2 3 ,
W
submerged
E I
s
L
bw
4
:=
Case 2: All Buckling modes of mode 1 - 4
Mode 1
The Reduced force within the Buckle in mode 1: P
bucklemode1
L
bw
( )
K
1 1 ,
E I
s
L
bw
2
:=
The Axial Force for mode:
P
omode1
L
bw
( )
P
bucklemode1
L
bw
( )
K
1 3 ,
2 3 ,
W
submerged
L
bw
1 K
1 2 ,
A
s
E
2 3 ,
( ) W
submerged
L
bw
5
E I
s
( )
2
+
(
(
(
0.5
1.0
(
(
(
(
+
... :=
y
maxmode1
L
bw
( )
K
1 4 ,
2 3 ,
W
submerged
E I
s
L
bw
4
:=
Mode 2
The Reduced force within the Buckle in mode 2:
P
bucklemode2
L
bw
( )
K
2 1 ,
E I
s
L
bw
2
:=
The Axial Force for mode:
P
omode2
L
bw
( )
P
bucklemode2
L
bw
( )
K
2 3 ,
2 3 ,
W
submerged
L
bw
1 K
2 2 ,
A
s
E
2 3 ,
( ) W
submerged
L
bw
5
E I
s
( )
2
+
(
(
(
0.5
1.0
(
(
(
(
+
... :=
y
maxmode2
L
bucklewave
( )
K
2 4 ,
2 3 ,
W
submerged
E I
s
L
bucklewave
4
:=
Mode 3
P
bucklemode3
L
bw
( )
K
3 1 ,
E I
s
L
bw
2
:=
P
omode3
L
bw
( )
P
bucklemode3
L
bw
( )
K
3 3 ,
2 3 ,
W
submerged
L
bw
1 K
3 2 ,
A
s
E
2 3 ,
( ) W
submerged
L
bw
5
E I
s
( )
2
+
(
(
(
0.5
1.0
(
(
(
(
+
... :=
y
maxmode3
L
bw
( )
K
3 4 ,
2 3 ,
W
submerged
E I
s
L
bw
4
:=
Mode 4
P
bucklemode4
L
bw
( )
K
4 1 ,
E I
s
L
bw
2
:=
P
omode4
L
bw
( )
P
bucklemode4
L
bw
( )
K
4 3 ,
2 3 ,
W
submerged
L
bw
1 K
4 2 ,
A
s
E
2 3 ,
( ) W
submerged
L
bw
5
E I
s
( )
2
+
(
(
(
0.5
1.0
(
(
(
(
+
... :=
y
maxmode4
L
bw
( )
K
3 4 ,
2 3 ,
W
submerged
E I
s
L
bw
4
:=
50 100 150
2000000
4000000
6000000
mode 1
mode infinity
mode 2
mode 3
mode 4
Hobbs Lateral Buckling for 19.10mm WT Pipe
Buckle Wave Length (m)
A
x
i
a
l

F
o
r
c
e

(
N
)
min_mode_1 3.605MN :=
min_mode_2 3.48MN :=
min_mode_3 3.417MN := min_mode_4 3.4107MN :=
min_mode_infy 4.238MN :=
MinCritivcalBuckleForce min min_mode_1 min_mode_2 , min_mode_3 , min_mode_4 , min_mode_infy , ( ) :=
MinCritivcalBuckleForce 3.411 MN =
Therefore the minimum force for which a buckle can exist in a straight line is given by
the Hobbs Force:
N
Hobbs
MinCritivcalBuckleForce :=
N
Hobbs
3.411 MN =
Based on SAFEBUCK GUIDELINE: A pipeline is is not susceptible to buckling if the inequality
below can be established
N
max
N
critical
s
..........................................................REF 2
max.axial

1 3 ,
:=
N
fmax

max.axial
W
submerged
L
2
:=
N
FULL
F
anchor
:=
For short Pipeline, the maximum for in the system: N
maxshort
min N
FULL
N
fmax
,
( )
:=
For Long Pipeline, the maximum for in the system is the fully constrain force : N
maxlong
N
FULL
:=
Considering only the Long Pipeline of KP 10
N
max
N
maxlong
:= N
max
8.09 MN =
Given the critical buckling force associated with Pipeline out of straightness (OOS) as:
minlateral

2 1 ,
:=
Minimum Radius of Curvature of Nominally Straight Pipe: R 1000m :=
N
OOS

minlateral
W
submerged
R :=
The critical buckling force is defined according to SAFEBUCK GUIDELINE as:
N
critical
min N
OOS
N
Hobbs
,
( )
:= N
critical
2.007 MN =
N
max
8.09 MN =
Since
N
max
N
critical
>
The Long Pipeline of 559 X 19.1mm is Susceptible to Lateral Buckling
For the Short Pipeline of 2km Length:
N
maxshort
1.673 MN = N
critical
2.007 MN =
N
maxshort
N
critical
s
Hence, the short Pipeline is not Susceptible to Lateral Buckling
A31|P a g e

A31|P a g e

APPENDIX A3
PIPELINE WALKING
Pipeline Walking
SUSCEPTIBILITY TO PIPELINE WALKING
References:
Description :
The Mathcad analysis worksheet presented in this report is used for Pipeline end expansion calculation of a rigid
pipeline under thermal and pressure loading. The purpose is to deduce the maximum expansion that an end
connection can take during maximum loading. The Calculations are based on Subsea 7 Pipeline expansion
guideline: CEO1PD-P-GU-126 and DNV-OS-F101
Assumtions:
Linear thermal profile with constant gradient throughout the heat-up
Pressure is assumed to be zero
References:
Hobbs, R. E., 'In Service Buckling of heated pipelines', Journal of Transport Engineering, Vol 110, No. 2, March 1984
Carr, M., Bruton, D., & Baxter, D., 'Safe Design of Pipeline with Lateral Buckling', SAFEBUCK III, DESIGN GUIDELINE.
July, 2011
Units: MPa 1N mm
2
g 9.81m s
2
MN 1 10
6
N
Pipeline Data:
D
o
559mm :=
Wall Thickness
t
wall
19.1mm :=
t
ext.coat
5mm :=
Length of Pipeline
Pipeline:
Pipe Steel Density
SMYS Steel Pipe
Steel Poisson Ratio
Sea Water Density
Max Content Density
Design Pressure
Operating Pressure
Hydrotest Pressure
t
conc
55mm :=
L 2000m :=
st
7850kgm
3
:=
SMYS 450MPa :=
E 2.0710
5
MPa :=
1.1710
5
C
1
:=
0.3 :=
inscoat
910kgm
3
:=
conc
2400.kgm
3
:=
water
1027kgm
3
:=
cont
900kgm
3
:=
P
d
15MPa :=
P
op
15MPa :=
P
hyd
0MPa :=
Pipeline_Walking.xmcdC:\Users\SS7N1342\Desktop\THESIS backup\MATHCAD\ 1 of 5
T
amb
5 C :=
T
op
95C :=
Residual Lay Tension N
lay
0kN :=
Soil Properties:
axial
0.5 :=
lateral
0.8 :=
Internal Diameter
Mass of Steel Pipe
Mass of Content
Total Mass in air
D
in
D
o
2 t
wall
:=
D
eff
D
o
2 t
ext.coat
t
conc
+
( )
+ :=
A
s

4
D
o
2
D
in
2
|
\
|
.
|
\
|
.
:=
A
ext

4
D
o
2 t
ext.coat
+
( )
2
D
o
2
:=
A
conc

4
D
o
2 t
ext.coat
+ 2 t
conc
+
( )
2
D
o
2 t
ext.coat
+
( )
2
:=
M
st
A
s
st
:=
M
ext.coat
A
ext
inscoat
:=
M
conc
A
conc
conc
:=
M
cont

4
D
in
2

cont
:=
M
water

4
D
in
2

water
:=
M
bouyancy

4
D
eff
2

water
:=
M
air
M
st
M
conc
+ M
ext.coat
+ :=
Submerged Mass
Weight of Dry Pipe
Weight of Content
M
submerged
M
air
M
bouyancy
:=
W
dry
M
air
g :=
W
cont
M
cont
g :=
W
submerged
M
cont
g M
ext.coat
g + M
st
g + M
conc
g + M
bouyancy
g :=
W
submerged
3.345
1
m
kN =
T T
op
T
amb
:=
Effective Area
A
eff

4
D
eff
( )
2
:=
SUSCEPTIBILITY TO PIPELINE WALKING
Thermal Transient
Using SAFEBUCK guideline, the pipeline is not susceptible to walking if the axial friction force exceeds the
following value:-
f
E A
eff
T
( )
L
> .........................................................................................................................REF 2
f = Axial Friction force
The constant = Parameter for walking due to thermal transient equation
The parameter can be obtained from the relation below:
2
3
8
2
6 +
q
L
T
+ 0 =
If the system is susceptible, the walking displacement per cycle can be estimated from the relations below:
f L
2
8 E A
s
=
|
\
|
|
|
.
.............................................................................................. whenf
f
6
s
L
2
16E A
s
24f
f f
4 f
( )
=
(
(
(
.............................................................................................. whenf
f
6
>
The axial friction due to the temperature gradient
f
E A
s
q
=
q
constant Heatup gradient

C
km
|
\
|
|
.
=
According to SAFEBUCK guideline, the maximum level of walking occurs when:
f
3
8
f
= .............................................................................................. REF....2
Susceptibility Check
i 1 3 .. :=
q
35
20
|
\
|
|
.
C
km
:=
Given
q
35
C
km
:=
We can obtain the corresponding parameter as follows:
2
3
8
2
6 +
q
1
L
T
+ 0 =
q
0
L
T
0.778 = y
q
0
L
T
:=
k
y
6
8
2
|
\
|
|
|
|
|
.
:= polyroots k ( ) :=
0.112
1.181
2.931
|
\
|
|
|
.
=
E A
eff
T
( )
L
4.433
46.608
115.68
|
\
|
|
|
.
1
m
kN =
f
axial
W
submerged
:=
f 1.673
1
m
kN =
Using SAFEBUCK guideline, the pipeline is not susceptible to walking if the axial friction force exceeds the
following value:-
f
E A
s
T
( )
L
>
Hence, the 2km flowline will be susceptible to walking with an axial friction coefficient of 0.5
We can also see that the length of the pipeline has a strong factor in determining the susceptibility of pipeline.
The axial friction due to the temperature gradient
f
E A
s
q
=
f
E A
s
q := f
2.746
1.569
|
\
|
|
.
1
m
kN =
f
6
0.458
0.262
|
\
|
|
.
1
m
kN =
f 1.673
1
m
kN =
Since
f
f
6
>
The walking displacement per cycle can be estimated from the relations below:
L
2
16E A
s
24f
f f
4 f
( )
:=
A
s
0.032m
2
=
The walking displacement per cycle can for different thermal gradient considered are as
follows:
q
35
20
|
\
|
|
.
C
km
=

39.623
12.032
|
\
|
|
.
mm =
A41|P a g e

A41|P a g e

APPENDIX A4
LOCAL BUCKLING CHECK
DNV-OS-F101

DNV-OS-F101 and DNV-RP-F110 Structural checks of pipeline
1 - Input
Design moment
M
sd
0.15kN m :=
Design effective axial force
S
sd
15 kN :=
Internal pressure
p
ip
345bar :=
External pressure
p
ep
0bar :=
Minimum internal pressure
p
min
0bar :=
Yield strength
R
t05
450MPa :=
Tensile strength
R
m
535MPa :=
Strain at yield strength point

rt05
0.005 :=
Strain at tensile strength limit

rm
0.180 :=
Outer diameter of pipe
D 559mm :=
Wall thickness of pipe
t
w
19.1mm :=
Corrosion allowance
t
corr
5mm :=
Specified minimum yield strength
SMYS 450.0MPa :=
at 100degC derarting
Specified minimum tensile strength
SMTS 700.0MPa :=
at 100degC derarting
Young's modulus
E 207000MPa :=
Functional load factor

f
1.1 :=
Safety class resistance factors

sc
1.14 :=
Seabed condition factor

c
0.9 :=
Pressure load factor (OS-F101 - 2000)

pr
1.05 :=
Material resistance factor

m
1.15 :=
Material reduction factor

u
0.96 :=
Resistance strain factor
e
2.5 :=
Axial strain resistance factor
ax
3.5 :=
Concrete strain intention factor

cc
1.25 :=
2 - Load controlled combined buckling check in accordance with
DNV-OS-F101 - 2007
Design wall thickness
t t
w
t
corr
14.1mm = :=
Design internal pressure
p
i
p
ip
345 bar = :=
p
e
p
ep
0 bar = :=
cloadcheck "The combined loading buckling criterion is applicable"
D
t
45 s p
ip
p
ep
> . if
"The combined loading buckling criterion is not applicable" otherwise
:=
cloadcheck "The combined loading buckling criterion is applicable" =
C:\Users\SS7N1342\Desktop\THESIS 1 of 4
Design yield stress:
f
y
SMYS
u
432 MPa = :=
Design tensile stress:
f
u
SMTS
u
672 MPa = :=
The pressure containment resistance
f
cb
min f
y
f
u
1.15
,
|
\
|
|
.
432 MPa = :=
p
b
2 t
D t
f
cb
2
3
25.8 MPa = :=
Plastic capacities for a pipe
S
p
f
y
D t ( ) t 10427.2 kN = :=
M
p
f
y
D t ( )
2
t 1808.6kN m = :=
Normalised moment
M
sdn
M
sd
f

c
M
p
0.0001 = :=
Normalised effective force
S
dn
S
sd
f

c
S
p
0.0014 = :=
Normalised pressure
q
h
p
i
p
b
2
3
1.157 = :=
0.5
D
t
15 < if
60
D
t
90
15
D
t
s 60 s if
0
D
t
60 > if
0.23 = :=
p
1
p
i
p
e
p
b
2
3
< if
1 3 1
p
i
p
e
p
b
\
|
|
.

p
i
p
e
p
b
2
3
> if
1.228 = :=
c
1 ( )
f
u
f
y
+ 1.126 = :=
Utilisation in accordance with DNV-OS-F101 - 2007
UF1
m
sc
M
sd
f

c
c
M
p
sc

f

c
S
sd
c
S
p
\
|
|
.
2
+
(
(
(
p
p
i
p
e
c
p
b
\
|
|
.
2
+ 2.126 = :=
UF2
m
sc
M
sdn
sc
S
dn
c
|
\
|
|
.
2
+
(
(
(
p
p
i
p
e
c
p
b
\
|
|
.
2
+ 2.126 = :=
Maximum allowable moment:
M
bsmax

c
sc
1
p
p
i
p
e
c
p
b
\
|
|
.
2

m
sc
S
dn
2
(
(
(
M
p
f

c
0 1664.7i + ( ) kN m = :=
2 - Displacement controlled combined buckling check in accordance
with DNV-OS-F101
cloaddischeck "The displ. contr. buckling criterion is applicable"
D
t
45 s p
ip
p
ep
> . if
"The displ. contr. buckling criterion is not applicable" otherwise
:=
cloaddischeck "The displ. contr. buckling criterion is applicable" =
Yield strength / tensile strength ratio:
h
R
t05
R
m
0.841 = :=
Girth weld factor:
gw
1
D
t
20 s if
1
D
t
20
|
\
|
|
.
0.01 20
D
t
< 60 < if
0.6 otherwise
0.804 = :=
Design compressive strain - pi >pe:
c
0.78
t
D
0.01
|
\
|
|
.
1 5.75
p
min
p
e
p
b
+
|
\
|
|
.

h
1.5

gw
0.0124 = :=
sd

c
cc
0.004 = :=
3 - Axial strain limit - DNV-RP-F110
Ramberg-Osgood hardening parameter:
n
ro
log

rm
R
m
E
rt05
R
t05
E
\
|
|
|
|
|
.
log
R
m
R
t05
|
\
|
|
.
23.926 = :=
Compressive axial strain limit:
cr
4
3
1
n
ro
t
D
0.007 = :=
d

cr
ax
0.002 = :=
4 - Ratcheting criterion in accordance with Klever et. al.
Hoop stress (Barlow's formula):
h
p
i
p
e
( )
D
2 t
683.9 MPa = :=
Ratcheting design factor:
f
3
0.7 :=
Limit strain wrt. ratcheting:
m
ra

h
SMYS
1.52 = :=
rl
SMYS 1 0.75m
ra
2
f
3
2
0.75 m
ra
2
+
|
\
|
.
E
0.00428i = :=
A41|P a g e

B1|P a g e

APPENDIX B: ANSYS SCRIPT
B21|P a g e

APPENDIX B1
LATERAL BUCKLING ANSYS SCRIPT

Lateral_buckling_model.txt
!##########################################################################################################
!# #
!# Title : Lateral Buckling and Axial Walking of Surface Laid Subsea Pipeline #
!# #
!# Company : Subsea 7 Norway #
!# Date : March 2013 #
!# Originator: Ifenna Obele #
!# Master Student in Offshore Tech. University of Stavanger, Norway #
!##########################################################################################################
! #
!Filename: Lateral_Buckling_model #
!Description: The response analysis of lateral buckling on HT/HP subsea pipeline #
!triggered by horizontal out-of-straightness on a flat seabed. #
!##########################################################################################################
*SET,model_id,'Lateral_Buckling_model'
/TITLE,%model_id%
/FILNAM,%model_id%
/ESHAPE,1 !Display elements as solids
/TRIAD,rbot !Display XYZ triad in right bottom corner
/PSYMB,NDIR,1 !Display nodal coord. system if other than global
/UNITS,MKS !MKS system (m, kg, s, deg C).
!###########################################################################################################
!Defining parameters
!Units are [m] [N] [KG] [S] [deg]
!###########################################################################################################
pi=4*ATAN(1.0) !Pi
g=9.81 !Gravitational Acceleration (ms^-2)
WD=200 !Water Depth (m)
RADc=100 !RAdius of Curvature in a normally straight pipe
igap=0 !Initial gap between pipeline and seabed
bgap=0 !Gap between pipe to the peakseabed profile
/PREP7 !Enter model creation preprocessor
!ANTYPE,0,NEW !0=STATIC
ACEL,,g !Define gravity
ET,1,PIPE288 !Pipe elements
SECTYPE,1,PIPE !Define pipe Section type
SECDATA,559E-3,19.1E-3 !Define Pipe Section:Outer Dia. and Wall Thickness [M]
ET,2,TARGE170 !Seabed element
ET,3,CONTA175 !Contact elements
!###########################################################################################################
!Defining PIPELINE DATA
!###########################################################################################################
Page 1
!#PHYSICAL DATA
OD=559E-3 !Pipe Outer Diameter (m)
twall=19.1E-3 !Pipe Wall Thickness (m)
Din=OD-2*twall !Pipe Internal Diameter (m)
L=2000 !Pipe Model Length (m)
!
t_ext=5E-3 !External Coating Thickness (m)
t_conc=55E-3 !Concrete Coating Thickness (m)
!#OPERATIONAL DATA
D_w=1027 !WaterDensity (kgm^-3)
D_cont=900 !Content Density (kgm^-3)
D_st=7850 !Pipe steel Density (kgm^-3)
P_des=15E6 !Design Pressure (Nm^-2)
P_op=15E6 !Operational Pressure (Nm^-2)
P_hyd=0E6 !Hydrotest Pressure
T_amb=5 !Ambient Temperature
T_op=95 !Operating Temperature
!N_Ray=0 !Residual Lay Tension
!#MATERIAL PROPERTIES
MPTEMP,1,0,95 ! Define temperatures for Young's modulus
MP,EX,1,207E9 !Young's Modulus (Nm^-2)
MP,ALPX,1,1.17E-5 !Thermal expansion Coefficient (1/deg)
MP,PRXY,1,0.3 !Poisson Ratio
!MP,DENS,1,D_st
D_ext=910 !Insulation or Coating Density (kgm^-3)
D_conc=2400 !Concrete coating density (kgm^-3)
!------------------------------------------------------------------------------------------------------------
! SEABED DATA
!------------------------------------------------------------------------------------------------------------
!DEFINE SEABED SOIL FRICTION
FRICLAX=0.5 ! Soil friction coefficient in axial direction
FRICLLAT=0.8 ! Soil friction coefficient in lateral direction
TB,FRIC,2,,,ORTHO ! Define orthotropic soil friction
TBDATA,1,FRICLAX,FRICLLAT

!############################################################################################################
!**RELEVANT CONNECTING EQUATION
!############################################################################################################
D_eff=OD+2*(t_ext+t_conc) ! Effective Pipe Diameter (m)
Ast=pi*(OD**2-Din**2)/4 ! Cross-sectional Area of Pipe Steel (m^2)
Page 2
Ast_ext=pi*((OD+2*t_ext)**2-OD**2)/4 ! Cross-sectional Area of External Coating (m^2)
Ast_conc=pi*((OD+2*t_ext+2*t_conc)**2-(OD+2*t_ext)**2)/4! Cross-sectional Area of Concrete Coating (m^2)
M_st=Ast*D_st ! Pipe Steel Mass (Kg/m)
M_ext=Ast_ext*D_ext ! External Coating Mass (Kg/m)
M_conc=Ast_conc*D_conc ! Concrete Coating Mass (Kg/m)
M_cont=pi*(Din**2)*D_cont/4 ! Content Mass (Kg/m)
M_water=pi*(Din**2)*D_w/4 ! Water Mass (Kg/m)
M_bouy=pi*(D_eff**2)*D_w/4 ! Buoyancy Mass (Kg/m)
M_air=M_st+M_ext+M_conc ! Pipeline Total Mass (Kg/m) (weight on air)
M_sub=M_air-M_bouy ! Submerged Mass (Kg/m) (weight in water)
W_cont=M_cont*g ! Content Weight (N/m)
W_water=M_water*g ! Flooded Weight (N/m)
W_sub=M_sub*g ! Empty Pipe Submerged Weight (N/m)
DEN_equiv=M_sub/Ast ! Submerged pipe Equivalent Density (kg/m^3)
D_insul=((t_ext*D_ext)+(t_conc*D_conc))/(t_ext+t_conc) ! Insulation Eqv. Density (Corr. & Concr. Coat.)
(N/m)
t_insul=t_ext+t_conc ! Insulation thickness (Corr. & Concr. Coat.) (m)
A_insul=Ast_ext+Ast_conc ! Insulation Area (Corrosion coat.& Concrete
Coat.)(m^2)
!############################################################################################################
!**UPDATE WEIGHT ON PIPELINE !EQUIVALENT DENSITY APPLIED TO SUBMERGED WEIGHT
!############################################################################################################
MP,DENS,1,DEN_equiv ! Pipe Material density (Kg/m^3)
SECCONTROLS,M_cont ! overrides default section properties.added mass: Content(kg/m)
!############################################################################################################
!**STRESS-STRAIN CURVE** STRAIN HARDENING
!############################################################################################################
TB,KINH,1,2,4 !Activates a data table for nonlinear material properties
!KINH Multilinear kinematic hardening using von Mises or Hill plasticity.
TBTEMP,20.0 !Temperature = 20 deg
TBPT,,0.0,0.0 !Strain=0.00,Stress=0.00
TBPT,,0.002174,450E6 !Elastic: Strain = 0.0217%, Stress = 450E6 (Nm^-2)
TBPT,,0.020,450E6 !Yield Strain: strain = 2.0%, stress = 450E6 (Nm^-2)
TBPT,,0.060,535E6 !Plastic strain: strain = 6.0%, stress = 535E6 (Nm^-2)
TBTEMP,100.0 !Temperature = 100 deg for material data
TBPT,,0.0,0.0 !Strain=0.00,Stress=0.00
TBPT,,0.00203,420E6 !Elastic: Strain = 0.0203%, Stress = 420E6 (Nm^-2)
TBPT,,0.020,420E6 !Yield Strain: strain = 2.0%, stress = 420E6 (Nm^-2)
TBPT,,0.060,505E6 !Plastic strain: strain = 6.0%, stress = 535E6 (Nm^-2)
TBLIST,KINH,1 !Lists the material data tables.
/XRANGE,0,0.01 !Specifies a linear abscissa (X) scale range of TBPLOT
Page 3
TBPLOT,KINH,1 !Displays the Data table
!###########################################################################################################
!**ELEMENT REAL CONSTANT
!###########################################################################################################
!# FOR PIPELINE !
!###############!
KEYOPT,1,1,0 ! Temperature Through wall gradient
!KEYOPT,1,3,0 ! linear shape functions
KEYOPT,1,4,1 ! Thin Pipe Theory
KEYOPT,1,6,0 ! Internal and External pressure cause loads on end caps
KEYOPT,1,7,0 ! Output control for section forces/moments and strains/curvatures

KEYOPT,1,8,0 ! Output control at integration points (1=Maximum and minimum
stresses/strains)
KEYOPT,1,9,2 ! Maximum and minimum stresses/strains + plus stresses and strains
at each section node
KEYOPT,1,15,0 ! One result for each section integration point

!################
!# SEABED !
!################
!R,22,,,1,0.2 ! Define Normal Contact Stiffness Factor and Penetration Tolerance
Factor
! (use ANSYS default)
KEYOPT,3,10,2 ! Set option 10 (Contact Stiffnes Update) for element type 3 to 2
(Each substep based on mean
! stress of underlying elements from the previous substep (pair
based))
! Update stiffness automaticly based on maximum penetration
KEYOPT,3,2,1 ! Penalty method, static stiffness of seabed

!KEYOPT,3,3,1 ! Contact MOdel: (0)Contact Force Based (1)Contact traction based
KEYOPT,3,4,2 ! Normal from contact nodes
!KEYOPT,3,5,3 ! Either Close the gap or reduces initial penetration
!KEYOPT,3,9,4 ! Include offset only (exclude initial geometrical penetration or
gap), but with ramped effects
KEYOPT,3,10,2 ! Applying the normal contact stiffness by a factor of 0.2 for each
bisection
KEYOPT,3,12,0 ! Behaviour of Contact Surface (0=standard)
!#############################################################################################################
Page 4
!Generate nodes and pipe element:
!############################################################################################################
nod1= 1 !first node number
nodn= 999 !last nodenumber
nelem=nodn-1 !number of elements in pipe
midnode=(nodn+1)/2 !midnode
elength=L/nelem !length of an element
n,nod1,0,0,0 !position of first pipenode
n,nodn,L,0,0 !position of last pipenode
fill,nod1,nodn !fill a row of nodes between nod1 and nodn
numstr,elem,1 !element numbering from 1
e,1,2 !create pipeelement nod1 and nod2
*repeat,nelem,1,1 !create the all the pipeelement
nsel,all !select all nodes
nsel,s,node,,1,nodn !select the pipenodes
cm,pipenodes,node !make it a single component
nsel,all
esel,s,type,,1 !select element by type
cm,pipeelem,elem !make it a pipeeleme
esel, all
!############################################################################################################
!**MESHING SEABED ELEMENT
!############################################################################################################
! Define nodes for seabed area
N, 3001 , -30.0 , -igap , 30
N, 3002 , 2030 , -igap , 30
N, 3003 , 2030 , -igap , -30
N, 3004 , -30.0 , -igap , -30
!#DEFINE TARGET ELEMENT##
!########################
numstr,elem,2990
TYPE,2 ! Select material and properties for seabed
MAT,2
REAL,22
TSHAPE,QUAD ! SET TARGET SHAPE
E,3001,3002,3003,3004 ! Define Element
numstr,elem,3001
type,3
real,22
Page 5
mat,2
NSEL,R,LOC,Y,0 ! Reselect nodes (DOF) in Y-direction
ESURF ! Generate contact elements overlaid on the free faces of
existing selected elements
ALLSEL ! - Seabed done
!#############################################################################################################
! DISPLAY MODEL
!#############################################################################################################
/ESHAPE,1 ! Display elements as solids
/TRIAD,rbot ! Display XYZ triad in right bottom corner
/PSYMB,NDIR,1 ! Display nodal coord. system if other than globa
WAVES ! Initiates reordering for the solution phase
WSORT ! Sorts elements based on geometric sort
!WMID,YES
SAVE ! Save all current database information
PARSAV,ALL,Latbuck,txt ! Save parameters to latbuck.txt
FINISH ! Exit the preprocessor
!/EOF
!#############################################################################################################
! SOLUTION
!#############################################################################################################
/CONFIG,NRES,30000
/solu !Enter solution processor
ANTYPE,TRANS !NEW STATIC SOLUTION
solcontrol,on !solution control on activates optimized defaults
!for a set of commands applicable to nonlinear solutions
nlgeom,on !Includes large-deflection effects in a static or full transient
analysis.
autots,on !automatic timestepping on
NROPT,UNSYM ! Specifies the Newton-Raphson options in a static or full
transient analysis
! (FULL or UNSYM= the stiffness matrix is updated at every
equilibrium iteration)
!NSUBST,10,20,10 ! Specifies the number of substeps to be taken every
loadstep (nbr this step, maximum number of
! substeps to be taken (i.e. min. time step), minimum number of
step (i.e. max time step)
Page 6
neqit,1000 !Specifies the maximum number of equilibrium iterations for
nonlinear analyses.
pstres,on !Calculate (or include) prestress effects
lnsrch,on !Activates a line search to be used with Newton-Raphson.
parres,,Latbuck,txt !Reads parameters from a latbuck.txt file.
tref,T_amb !Defines the reference temperature for the thermal strain
calculations.
!Thermal strains are given by *(T-TREF)
sfcum,pres,repl !cummulative surface load on
bfcum,temp,repl
cncheck,auto !Automatically sets certain real constants and key options to
recommended values
nsel,all
nsel,u,node,,465,535
D,all,UZ,0
allsel,all
!#############################################################################################################
! LOAD STEPS AND BOUNDARY CONDITION
!#############################################################################################################
!The word loads as used in ANSYS documentation includes boundary conditions
!(constraints, supports, or boundary field specifications) as well as other
!externally and internally applied loads
!-------------------------------------------------------------------------------------------------------------
TIME,1
/stitle,1,Lay pipeline on seabed and Apply boundary condition,set imperfection on pipeline and apply
internal and external presure
d,nod1,all !Initial state: Fix end 1
d,nodn,all !Fix end 2
!f,480,fz,(100000)/(520-480),,520 ! apply a load between node 1490 and 1505
f,480,fz,7000,,520
sfe,pipeelem,1,pres,,P_op !Apply internal Pressure
sfe,pipeelem,2,pres,,D_w*g*(WD) !The hydrostatic pressure @ 800m WD (N/m)
NSUBST,10,20,10
solve
!save
!fini
!/EOF
Page 7
!------------------------------------------------------------------------------------------------------------
TIME,2
/stitle,1,Remove lateral force and apply internal pressure
f,480,fz,0,,520 ! apply a load between node 1490 and 1505
NSUBST,10,20,10
allsel
solve
!save
!fini
!/EOF
!-------------------------------------------------------------------------------------------------------------
TIME,3
/stitle,1,Appy operating pressure and temperature
sfcum,pres,repl
bfcum,temp,repl
sfe,pipeelem,1,pres,,P_op
bf,pipenodes,temp,90
NSUBST,50,100,50
allsel
solve
save
!/EOF
!END OF LATERAL BUCKLING MODEL
!--------------------------------------------------------------------------------------------------------------
TIME,4
/stitle,1,FIRST COOLDOWN
sfcum,pres,repl
bfcum,temp,repl
sfe,pipeelem,1,pres,,D_cont*g*(WD+20)
bf,pipenodes,temp,T_amb
NSUBST,20
allsel
solve
!save
!/EOF
!------------------------------------------------------------------------------------------------------------
TIME,5
/stitle,1,SECOND HEAT-UP
sfcum,pres,repl
bfcum,temp,repl
Page 8
bf,pipenodes,temp,(T_op-T_amb)
NSUBST,20
allsel
solve
!save
!/EOF
!------------------------------------------------------------------------------------------------------------
TIME,6
/stitle,1,SECOND COOLDOWN
NSUBST,20,100,20
allsel
solve
!save
!/EOF
!------------------------------------------------------------------------------------------------------------
TIME,7
/stitle,1,THIRD HEAT-UP
bf,pipenodes,temp,(T_op-T_amb)
NSUBST,25,100,25
allsel
solve
!save
!/EOF
!-------------------------------------------------------------------------------------------------------------
TIME,8
/stitle,1,THIRD COOLDOWN
NSUBST,25,100,25
allsel
solve
!save
!/EOF
!------------------------------------------------------------------------------------------------------------
Page 9
B21|P a g e

B21|P a g e

APPENDIX B2
PIPELINE WALKING ANSYS SCRIPT
Pipeline_walking_model.txt
!#################################################################################################################################
#####
!#
#
!# Title : Lateral Buckling and Axial Walking of Surface Laid Subsea Pipeline
#
!# Company : Subsea 7 Norway
#
!# Date : March 2013
#
!# Originator: Ifenna Obele
#
!# Master Student in Offshore Tech. University of Stavanger, Norway
#
!#################################################################################################################################
#####
!
#
!Filename:Pipeline_Walking
#
!Description: Axial response due to Pipeline walking
#
!#################################################################################################################################
#####
*SET,model_id,'Pipeline_Walking'
/TITLE,%model_id%
/FILNAM,%model_id%
/ESHAPE,1 !Display elements as solids
/TRIAD,rbot !Display XYZ triad in right bottom corner
/PSYMB,NDIR,1 !Display nodal coord. system if other than global
/UNITS,MKS !MKS system (m, kg, s, deg C).
!#################################################################################################################################
####
!Defining parameters
!Units are [m] [N] [KG] [S] [deg]
!#################################################################################################################################
####
pi=4*ATAN(1.0) !Pi
g=9.81 !Gravitational Acceleration (ms^-2)
WD=800 !Water Depth (m)
RADc=100 !RAdius of Curvature in a normally straight pipe
igap=0 !Initial gap between pipeline and seabed
bgap=0 !Gap between pipe to the peakseabed profile
/PREP7 !Enter model creation preprocessor
!ANTYPE,0,NEW !0=STATIC
ACEL,,g !Define gravity
ET,1,PIPE288 !Pipe elements
SECTYPE,1,PIPE !Define pipe Section type
SECDATA,559E-3,19.1E-3 !Define Pipe Section:Outer Dia. and Wall Thickness [M]
ET,2,TARGE170 !Seabed element
ET,3,CONTA175 !Contact elements
!#################################################################################################################################
Page 1
###
!Defining PIPELINE DATA
!#################################################################################################################################
###
!#PHYSICAL DATA
OD=559E-3 !Pipe Outer Diameter (m)
twall=19.1E-3 !Pipe Wall Thickness (m)
Din=OD-2*twall !Pipe Internal Diameter (m)
L=2000 !Pipe Model Length (m)
!
t_ext=5E-3 !External Coating Thickness (m)
t_conc=55E-3 !Concrete Coating Thickness (m)
!#OPERATIONAL DATA
D_w=1027 !WaterDensity (kgm^-3)
D_cont=900 !Content Density (kgm^-3)
D_st=7850 !Pipe steel Density (kgm^-3)
P_des=15E6 !Design Pressure (Nm^-2)
P_op=15E6 !Operational Pressure (Nm^-2)
P_hyd=0E6 !Hydrotest Pressure
T_amb=5 !Ambient Temperature
T_op=95 !Operating Temperature
!N_Ray=0 !Residual Lay Tension
!#MATERIAL PROPERTIES
MPTEMP,1,0,95 ! Define temperatures for Young's modulus
MP,EX,1,207E9 !Young's Modulus (Nm^-2)
MP,ALPX,1,1.17E-5 !Thermal expansion Coefficient (1/deg)
MP,PRXY,1,0.3 !Poisson Ratio
MP,DENS,1,D_st
D_ext=910 !Insulation or Coating Density (kgm^-3)
D_conc=2400 !Concrete coating density (kgm^-3)
!---------------------------------------------------------------------------------------------------------------------------------
--
! SEABED DATA
!---------------------------------------------------------------------------------------------------------------------------------
-
!DEFINE SEABED SOIL FRICTION
FRICLAX=0.5 ! Soil friction coefficient in axial direction
FRICLLAT=0.8 ! Soil friction coefficient in lateral direction
TB,FRIC,2,,,ORTHO ! Define orthotropic soil friction
TBDATA,1,FRICLAX,FRICLLAT

!#################################################################################################################################
##
!**RELEVANT CONNECTING EQUATION
!#################################################################################################################################
##
D_eff=OD+2*(t_ext+t_conc) ! Effective Pipe Diameter (m)
Ast=pi*(OD**2-Din**2)/4 ! Cross-sectional Area of Pipe Steel (m^2)
Ast_ext=pi*((OD+2*t_ext)**2-OD**2)/4 ! Cross-sectional Area of External Coating (m^2)
Ast_conc=pi*((OD+2*t_ext+2*t_conc)**2-(OD+2*t_ext)**2)/4! Cross-sectional Area of Concrete Coating (m^2)
M_st=Ast*D_st ! Pipe Steel Mass (Kg/m)
M_ext=Ast_ext*D_ext ! External Coating Mass (Kg/m)
Page 2
M_conc=Ast_conc*D_conc ! Concrete Coating Mass (Kg/m)
M_cont=pi*(Din**2)*D_cont/4 ! Content Mass (Kg/m)
M_water=pi*(Din**2)*D_w/4 ! Water Mass (Kg/m)
M_bouy=pi*(D_eff**2)*D_w/4 ! Buoyancy Mass (Kg/m)
M_air=M_st+M_ext+M_conc ! Pipeline Total Mass (Kg/m) (weight on air)
M_sub=M_air-M_bouy ! Submerged Mass (Kg/m) (weight in water)
W_cont=M_cont*g ! Content Weight (N/m)
W_water=M_water*g ! Flooded Weight (N/m)
W_sub=((M_sub*g) + (M_cont*g)) ! Empty Pipe Submerged Weight (N/m)
DEN_equiv=M_sub/Ast ! Submerged pipe Equivalent Density (kg/m^3)
D_insul=((t_ext*D_ext)+(t_conc*D_conc))/(t_ext+t_conc) ! Insulation Eqv. Density (Corr. & Concr. Coat.) (N/m)
t_insul=t_ext+t_conc ! Insulation thickness (Corr. & Concr. Coat.) (m)
A_insul=Ast_ext+Ast_conc ! Insulation Area (Corrosion coat.& Concrete Coat.) (m^2)
!#################################################################################################################################
##
!**UPDATE WEIGHT ON PIPELINE !EQUIVALENT DENSITY APPLIED TO SUBMERGED WEIGHT
!#################################################################################################################################
##
!MP,DENS,1,DEN_equiv ! Pipe Material density (Kg/m^3)
!SECCONTROLS,M_cont ! overrides default section properties.added mass: Content(kg/m)
!#################################################################################################################################
###
!**ELEMENT REAL CONSTANT
!#################################################################################################################################
###
!# FOR PIPELINE !
!###############!
KEYOPT,1,1,0 ! Temperature Through wall gradient
!KEYOPT,1,3,0 ! linear shape functions
KEYOPT,1,4,1 ! Thin Pipe Theory
KEYOPT,1,6,0 ! Internal and External pressure cause loads on end caps
KEYOPT,1,7,0 ! Output control for section forces/moments and strains/curvatures
KEYOPT,1,8,0 ! Output control at integration points (1=Maximum and minimum stresses/strains)
KEYOPT,1,9,2 ! Maximum and minimum stresses/strains + plus stresses and strains at each section node
KEYOPT,1,15,0 ! One result for each section integration point

!################
!# SEABED !
!################
!R,22,,,1,0.2 ! Define Normal Contact Stiffness Factor and Penetration Tolerance Factor
! (use ANSYS default)
KEYOPT,3,10,2 ! Set option 10 (Contact Stiffnes Update) for element type 3 to 2 (Each substep based on
mean
! stress of underlying elements from the previous substep (pair based))
! Update stiffness automaticly based on maximum penetration
KEYOPT,3,2,1 ! Penalty method, static stiffness of seabed

!KEYOPT,3,3,1 ! Contact MOdel: (0)Contact Force Based (1)Contact traction based
KEYOPT,3,4,2 ! Normal from contact nodes
!KEYOPT,3,5,3 ! Either Close the gap or reduces initial penetration
!KEYOPT,3,9,4 ! Include offset only (exclude initial geometrical penetration or gap), but with ramped
Page 3
effects
KEYOPT,3,10,2 ! Applying the normal contact stiffness by a factor of 0.2 for each bisection
KEYOPT,3,12,0 ! Behaviour of Contact Surface (0=standard)
!#################################################################################################################################
####
!Generate nodes and pipe element:
!#################################################################################################################################
####
nod1= 1 !first node number
nodn= 999 !last nodenumber
nelem=nodn-1 !number of elements in pipe
midnode=(nodn+1)/2 !midnode
elength=L/nelem !length of an element
n,nod1,0,0,0 !position of first pipenode
n,nodn,L,0,0 !position of last pipenode
fill,nod1,nodn !fill a row of nodes between nod1 and nodn
numstr,elem,1 !element numbering from 1
e,1,2 !create pipeelement nod1 and nod2
*repeat,nelem,1,1 !create the all the pipeelement
cm,pipenodes,node !make it a single component
cm,coldnodes,node !make it a single component
nsel,all
esel,s,type,,1 !select element by type
cm,pipeelem,elem !make it a pipeeleme
esel, all
!#################################################################################################################################
#
!**MESHING SEABED ELEMENT
!#################################################################################################################################
#
! Define nodes for seabed area
N, 3001 , -30.0 , -igap , 30
N, 3002 , 2030 , -igap , 30
N, 3003 , 2030 , -igap , -30
N, 3004 , -30.0 , -igap , -30
!#DEFINE TARGET ELEMENT##
!########################
numstr,elem,2990
TYPE,2 ! Select material and properties for seabed
MAT,2
REAL,22
TSHAPE,QUAD ! SET TARGET SHAPE
E,3001,3002,3003,3004 ! Define Element
numstr,elem,3001
type,3
Page 4
real,22
mat,2
NSEL,R,LOC,Y,0 ! Reselect nodes (DOF) in Y-direction
ESURF ! Generate contact elements overlaid on the free faces of existing selected
elements
ALLSEL ! - Seabed done
!#################################################################################################################################
#####
! DISPLAY MODEL
!#################################################################################################################################
#####
/ESHAPE,1 ! Display elements as solids
/TRIAD,rbot ! Display XYZ triad in right bottom corner
/PSYMB,NDIR,1 ! Display nodal coord. system if other than globa
WAVES ! Initiates reordering for the solution phase
WSORT ! Sorts elements based on geometric sort
!WMID,YES
SAVE ! Save all current database information
PARSAV,ALL,Latbuck,txt ! Save parameters to latbuck.txt
FINISH ! Exit the preprocessor
!/EOF
!#################################################################################################################################
######
! SOLUTION
!#################################################################################################################################
######
/CONFIG,NRES,30000
/solu !Enter solution processor
ANTYPE,TRANS !NEW STATIC SOLUTION
solcontrol,on !solution control on activates optimized defaults
!for a set of commands applicable to nonlinear solutions
nlgeom,on !Includes large-deflection effects in a static or full transient analysis.
autots,on !automatic timestepping on
NROPT,UNSYM ! Specifies the Newton-Raphson options in a static or full transient analysis
! (FULL or UNSYM= the stiffness matrix is updated at every equilibrium iteration)
!NSUBST,10,20,10 ! Specifies the number of substeps to be taken every load step (nbr this step, maximum
number of
! substeps to be taken (i.e. min. time step), minimum number of step (i.e. max time step)
neqit,1000 !Specifies the maximum number of equilibrium iterations for nonlinear analyses.
pstres,on !Calculate (or include) prestress effects
lnsrch,on !Activates a line search to be used with Newton-Raphson.
parres,,Latbuck,txt !Reads parameters from a latbuck.txt file.
tref,T_amb !Defines the reference temperature for the thermal strain calculations.
!Thermal strains are given by a *(T-TREF)
sfcum,pres,repl !cummulative surface load on
bfcum,temp,repl
cncheck,auto !Automatically sets certain real constants and key options to recommended values
nsel,all
nsel,u,node,,1,999
D,all,UZ,0
Page 5
allsel,all
!#################################################################################################################################
######
! LOAD STEPS AND BOUNDARY CONDITION
!#################################################################################################################################
#####
!The word loads as used in ANSYS documentation includes boundary conditions
!(constraints, supports, or boundary field specifications) as well as other
!externally and internally applied loads
TIME,1
/stitle,1,Lay pipeline on seabed and Apply boundary condition,set imperfection on pipeline and apply internal and external presure

f,pipenodes,fy,-(W_sub*elength)
sfe,pipeelem,2,pres,,D_w*g*(WD) !The hydrostatic pressure @ 800m WD (N/m)
NSUBST,15,20,10
solve
!save
!fini
!/EOF
!---------------------------------------------------------------------------------------------------------------------------------
-
TIME,2
/stitle,1,Heatstep 2
heatstep2n.mac
NSUBST,15
allsel
solve
save
!/EOF
!---------------------------------------------------------------------------------------------------------------------------------
----
TIME,3
heatstep3n.mac
NSUBST,15
allsel
solve
save
!/EOF
!---------------------------------------------------------------------------------------------------------------------------------
-------
TIME,4
heatstep4n.mac
Page 6
NSUBST,15
allsel
solve
save
!/EOF
!---------------------------------------------------------------------------------------------------------------------------------
------
TIME,5
heatstep5n.mac
NSUBST,15
allsel
solve
save
!/EOF
!---------------------------------------------------------------------------------------------------------------------------------
--------
TIME,6
heatstep6n.mac
NSUBST,15
allsel
solve
save
!/EOF
!---------------------------------------------------------------------------------------------------------------------------------
--------
TIME,7
heatstep7n.mac
NSUBST,15
allsel
solve
save
!/EOF
!---------------------------------------------------------------------------------------------------------------------------------
---------
TIME,8
heatstep8n.mac
NSUBST,15
allsel
solve
Page 7
save
!/EOF
!---------------------------------------------------------------------------------------------------------------------------------
--------
TIME,9
heatstep9n.mac
NSUBST,15
allsel
solve
save
!/EOF
!---------------------------------------------------------------------------------------------------------------------------------
------
TIME,10
heatstep10n.mac
NSUBST,15
allsel
solve
save
!/EOF
!---------------------------------------------------------------------------------------------------------------------------------
------
TIME,11
heatstep11n.mac
NSUBST,15
allsel
solve
save
!/EOF
!---------------------------------------------------------------------------------------------------------------------------------
-----
TIME,12
heatstep12n.mac
NSUBST,15
allsel
solve
save
!/EOF
!---------------------------------------------------------------------------------------------------------------------------------
----
Page 8
TIME,13
heatstep13n.mac
NSUBST,15
allsel
solve
save
!/EOF
!---------------------------------------------------------------------------------------------------------------------------------
----
TIME,14
heatstep14n.mac
NSUBST,15
allsel
solve
save
!/EOF
!---------------------------------------------------------------------------------------------------------------------------------
--
TIME,15
heatstep15n.mac
NSUBST,15
allsel
solve
save
!/EOF
!---------------------------------------------------------------------------------------------------------------------------------
-
TIME,16
heatstep16n.mac
NSUBST,15
allsel
solve
save
!/EOF
!---------------------------------------------------------------------------------------------------------------------------------
TIME,17
/stitle,1,Finalheatstep
finalheatstepn.mac
NSUBST,15
allsel
Page 9
solve
save
!/EOF
!---------------------------------------------------------------------------------------------------------------------------------
TIME,18
/stitle,1,First cycle -cooldown
NSUBST,15
allsel
solve
save
!/EOF
! !Second Cycle
!#################################################################################################################################
TIME,19
heatstep2n.mac
NSUBST,15
allsel
solve
save
!/EOF
!--------------------------------------------------------------------------------------------------------------------------------
TIME,20
heatstep3n.mac
NSUBST,15
allsel
solve
save
!/EOF
!---------------------------------------------------------------------------------------------------------------------------------
TIME,21
heatstep4n.mac
NSUBST,15
allsel
solve
save
!/EOF
!--------------------------------------------------------------------------------------------------------------------------------
Page 10
TIME,22
heatstep5n.mac
NSUBST,15
allsel
solve
save
!/EOF
!---------------------------------------------------------------------------------------------------------------------------------
TIME,23
heatstep6n.mac
NSUBST,15
allsel
solve
save
!/EOF
!------------------------------------------------------------------------------------------------------------------------------
TIME,24
heatstep7n.mac
NSUBST,15
allsel
solve
save
!/EOF
!---------------------------------------------------------------------------------------------------------------------------
TIME,25
heatstep8n.mac
NSUBST,15
allsel
solve
save
!/EOF
!-----------------------------------------------------------------------------------------------------------------------------
TIME,26
heatstep9n.mac
NSUBST,15
allsel
solve
save
Page 11
!/EOF
!---------------------------------------------------------------------------------------------------------------------------
TIME,27
heatstep10n.mac
NSUBST,15
allsel
solve
save
!/EOF
!------------------------------------------------------------------------------------------------------------------------------
TIME,28
heatstep11n.mac
NSUBST,15
allsel
solve
save
!/EOF
!-----------------------------------------------------------------------------------------------------------------------------
TIME,29
heatstep12n.mac
NSUBST,15
allsel
solve
save
!/EOF
!-------------------------------------------------------------------------------------------------------------------------------
TIME,30
heatstep13n.mac
NSUBST,15
allsel
solve
save
!/EOF
!------------------------------------------------------------------------------------------------------------------------------
TIME,31
heatstep14n.mac
NSUBST,15
Page 12
allsel
solve
save
!/EOF
!-------------------------------------------------------------------------------------------------------------------------------
TIME,32
heatstep15n.mac
NSUBST,15
allsel
solve
save
!/EOF
!------------------------------------------------------------------------------------------------------------------------------
TIME,33
heatstep16n.mac
NSUBST,15
allsel
solve
save
!/EOF
!-----------------------------------------------------------------------------------------------------------------------------
TIME,34
/stitle,1,Finalheatstep
finalheatstepn.mac
NSUBST,15
allsel
solve
save
!/EOF
!-----------------------------------------------------------------------------------------------------------------------------
TIME,35
/stitle,1,First cycle -cooldown
NSUBST,15
allsel
solve
save
!/EOF
!------------------------------------------------------------------------------------------------------------------------------
Page 13
B31|P a g e

Buckling Oil and Gas

Uploaded by

Copyright:

Available Formats

Buckling Oil and Gas

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Buckling Oil and Gas

Uploaded by

Copyright:

Available Formats

What is the objective of the thesis work?

What is the objective of the thesis work?

What is the main focus of the thesis work?

What is the main focus of the thesis work?

Faculty of Science and Technology

Figuie S2: Theimal stiain effect...............................................................................................28

Figuie 74: Byuiostatic Effect fiom the topsiue connection...................................................75

APDL ANSYS parametric design language

SMTS Specified Minimum Tensile Strength

Figure 12: Upheaval and Lateral Buckling (Floriano et al., 2011)

Figure 13: Example of a local buckled Pipe (Takahashi et al., 2007)

1.2 State of the Art

1.2.2 Engineered Buckle initiators and Mitigating Measures

Figure 110: Different Regions in a buckle (Kein et al.)

Figure 114: Buckle initiating using Vertical sleepers (Kein et al.)

d) Buckle Initiation using distributed Buoyancy or additional insulation coating:

Figure 117: Snake Lay Configuration (Kein et al.)

1.2.3 Axial Walking Control Measures

c) Increased Axial Friction: As previously mentioned in this report, pipeline interacted

Figure 119: Inline Expansion Spool

1.3 Thesis Objectives

1.4 Outline of Thesis

Figure 23: SnapThrough Buckling (Ahmed

Substituting the function w = e

2.3 Operating Stresses

Figure 26: Stress Induced by Internal Pressure Loading (Karunakaran, 2012)

Figure 27: Longitudinal Stress component (Prof. Sharma)

Figure 28: Pressure induced by End Cap Effect (Karunakaran, 2012)

2.3.2.2 Thermal Stress

2.3.2.4 Bending Stress

3. Theoretical Background - End Expansions

Figure 31: Example of Expansion Analysis

From the end-cap force which is given as

, then the end-cap strain becomes:

= Intcrnol prcssurc Jicrcncc rcloti:c to os loiJ, onJ A

Figure 36: Pipeline End Expansion (Cheuk, 2007)

4. Theoretical Background - Lateral Buckling

= Arco boscJ on pipc intcrnol Jiomctcr

Figure 41: Symmetric and Asymmetric buckle modes (Kaye, 1996)

Figure 42: Lateral Buckling modes (Kaye, 1996)

4.3.1 FeedinLength and Maximum Allowable Moment

4.3.2 Hobbs Predictive Analytical Method

Figure 44: Lateral Buckling configuration (EINSFELD et al., 2003)

Table 41: Constant for Lateral Buckling Modes (Hobbs, 1984)

4.3.3 Pipeline Initial Imperfection

4.3.4 Virtual Anchor Spacing

Figure 46: SnakeLay configuration (Subsea7, 2012)

5. Theoretical Background - Pipeline Walking

Figure 412: Force Sloping Seabed (Carr et al., 2006)

Figure 413: Schematic of PipelineRiser tension at the end connection

Figure 414: Thermal Transient

Figure 415: Effective axial force in a short pipeline

Figure 416: Effective Axial force for long pipeline

6.1.2 Design Assumptions and Requirement

6.2 Design Process and Roadmap - Lateral buckling

a. Selection of Pipeline Parameters:

e. Lateral Buckling Check

The corresponding buckle wavelength is also given according to Hobbs as:

Figure 65: Pipeline showing an Initial OOS of 1m