Pipl Stress Analysis
Pipl Stress Analysis
Pipl Stress Analysis
By
9127
(Mechanical Engineering)
JANUARY 2012
31750 Tronoh,
By
Approved by,
____________________
(Dr. Mokhtar bin Awang)
i
CERTIFICATION OF ORIGINALITY
This is to certify that I am responsible for the work submitted in this project, that the
original work is my own except as specified in the references and acknowledgements,
and that the original work contained herein have not been undertaken or done by
unspecified sources or persons.
_______________________________
NURUL ALIA BINTI MOHD ANUAR
ii
ABSTRACT
Offshore oil and gas pipelines are being subjected to deeper water depths, more extreme
environmental conditions, and harsher operating requirements than ever before. Given
these conditions, free spanning pipelines are becoming more common and are often
unavoidable during pipeline installation. Free spans occur as a result of irregular seafloor
topography at installation or during pipeline operation as a result of vibration and scour
[1].
A linear-elastic finite element model is applied to the solution of stress analysis problems
involving submarine pipelines freely resting upon irregular seabed profiles. This report
describes a finite element (FE) modelling procedure and parametric study leading to the
investigation of stress distribution and deformation subjected on pipeline. The objective
of this project is to model underwater pipeline using pipe stress analysis software,
CAESAR II. The pipeline will be examined on various conditions according to the
geometry of the seabed. The input or load cases of the pipeline system are ocean current
and wave. The FE analyses are carried out for both the fully fixed and simply supported
pipes, which form the two extreme conditions of pipelines under service conditions.
Expected result is that the stress of the pipelines should not exceed the maximum
allowable stress set by the regulations.
iii
TABLE OF CONTENTS
CHAPTER 1: INTRODUCTION.................................................................................................1
1.1.Project Background ............................................................................................1
1.2.Problem Statement .............................................................................................3
1.3.Objectives and Scopes of study .........................................................................4
iv
CHAPTER 4: RESULT AND DISCUSSION ...........................................................................29
4.1. Stress Analysis Result .....................................................................................29
4.2. Maximum mStresses ......................................................................................29
4.3. Maximum Displacement ................................................................................31
4.4. Seabed Topography Cases .............................................................................32
4.5. Stress Summary Result ..................................................................................36
4.6. Support Location ............................................................................................37
4.7. Distance between Nodes ................................................................................38
REFERENCES .............................................................................................................................40
APPENDICES .............................................................................................................................43
Appendix A: ....................................................................................................... .A1
Appendix B: ........................................................................................................ B1
Appendix C: ........................................................................................................ C1
Appendix D: ........................................................................................................ D1
v
LIST OF FIGURES
vi
LIST OF TABLE
vii
CHAPTER 1: INTRODUCTION
Marine pipelines for the transportation of oil and gas have become a safe and
reliable part of the expanding infrastructure put in place for the development of the
valuable resources below the world’s seas and oceans [2]. Route selection for pipeline is
a crucial activity. A poorly chosen route can be much more expensive than a well chosen
route. Understanding of the seabed geotechnics and the oceanographic conditions:
knowledge of the locations of geotechnically uniform and smooth seabed, free of
obstructions or existing pipelines and not in conflict with other fields, existing or planned
subsea installations [3].
Stress is classified into three major categories namely primary stress, secondary
stress and tertiary stress. Primary stress is developed by imposed loading and necessary to
satisfy the equilibrium between external and internal forces and moments of the pipeline.
Secondary stress is a self-limiting stress which is developed by constraint of the
displacement of a structure. The displacement is caused by thermal expansion or by
outwardly imposed restraint. Tertiary stress is a peak stress which causes no significant
distortion. It is the highest stress under consideration and responsible for causing fatigue
failure [4].
A pipeline rests on or in the seabed. Based on research typically for Malay basin,
most of the underwater pipelines are not supported [5]. Depending on the seabed
topography, sometimes rocks are dumped surrounding the pipeline as a means of support.
The pipelines are also being anchored on the seabed as a means of fixed support or rather
being laid by the concrete mattress. Thus, this particular study is generally focus on the
stress analysis accounted for pipeline that is laid on different type of seabed topography
such as inclined slope, uneven seabed, etc.
1
Sea current and pressure difference around a pipeline will create hydrodynamic
forces. The stress on the pipeline is determined by the relative magnitude of the agitating
hydrodynamic force and the resulting force due to the submerged of pipeline. The
pipeline will be displaced when the resultant of drag and lift forces exceed the resisting
force due to the submerged weight of pipeline [6].
CAESAR II is used rather that ANSYS are for various reasons. CAESAR II user
creates a model of the piping system using simple beam elements and defines the loading
conditions imposed on the system. With this input, CAESAR II produces results in the
form of displacements, loads and stresses throughout the system. Additionally, CAESAR
II compares these results to limits specified by recognized codes and standards [7].
2
1.2 Problem Statement
From the pipeline point of view, the ideal seabed is level and smooth so that no
spans are formed and is composed of stable medium clay. The pipe settles into the clay
and gains enhanced lateral stability [2]. However, the seabed has many types of geometry
and not consistently that the pipeline encounters flat and even ocean floor. Some seabeds
are highly mobile and include sandwaves (which may be 15 m high and 100 m long) and
smaller ripple features (which range in size on many scales from millimeters to meters
high) [6].
All of the offshore activities are mainly concerns on the safety measure. Thorough
inspections are done to ensure that all the facilities and equipment used offshore are safe
and reliable. Pipelines in service are subjected to wave and current loadings. Thus, an
analysis is required for detailed examination of external hydrodynamics loading on the
pipeline.
The input graphics model of CAESAR II facilitates intuitive pipe stress analysis
modeling. CAESAR II stress analysis shows piping system flexibility, plus any areas of
concern. Pipe stress analysis results, in the form of displacements, loads and stresses, are
compared with international piping standards and piping codes [8].
3
1.3 Objective and Scope of Study
Objectives:
1. To model pipeline and perform the stress analysis using pipe stress analysis
interface, CAESAR II.
2. To study the stress distribution of underwater pipeline laying on irregular seabed
geometry under ocean current loads and also verifies the design code compliance.
Scope of study:
The scope of this project encompasses all the necessary activities to understand,
assess and analyze subsea pipeline stress distribution. The beginning phase of the project
includes an extensive research effort. This research begins with the knowledge and
experience of engineers from oil and gas industry and also study through recent journals.
The focal point of this project is the simulation of pipe model using CAESAR II
that focuses on the pipe stress analysis subjected to the pipeline. For economic reasons,
the material that will be used for the fabrication of pipelines (for production and
transmission of oil and gas) is carbon steels (API 5L X65). The area of seabed
investigated is Malay Basin which is located on the north-west of peninsula Malaysia.
Water depth is approximately 100-300 m. The piping code used for the pipeline analysis
is ASME B31.3. The subsea pipeline coverage is from the riser that is attached to the
processing unit (CPP) to the wellhead platform (WHP). The length for the pipe spool is
20 feet.
4
CHAPTER 2: LITERATURE REVIEW
Prior to engineering works, the seabed must be thoroughly surveyed along the
entire pipeline route to map seabed topography and identify potential obstacles. This is to
avoid free spans and seabed peaks and troughs. Uneven topographic conditions mean that
the rigid pipelines cannot always be in direct contact with the seabed. Based on some
research, the author hasn’t found any study on stress analysis of pipeline particularly for
irregular seabed topography. Most previous investigations have only concerned either on
the seabed topography alone or the pipeline stress analysis alone. Recent study by F.P.
Gao, D.S. Jeng and H. Sekiguchi focus on the wave-seabed pipeline interaction problem.
In this study, a proposed finite element model is adopted to investigate the interaction
between nonlinear ocean waves, a buried pipelines and a porous seabed. The numerical
results indicate the importance to the effect of pipeline on the seabed response [9].
5
Pipeline routing is a major factor that can directly influence cost and feasibility of
a pipeline project. For example, this may impact technical considerations such as
excessive water depth or the presence of geohazards, or geopolitical reasons such as
national boundaries. Furthermore, these factors generally become more pronounced when
pipeline routes traverse continental slopes to the abyssal or deep ocean depths [12].
Submarine pipeline are usually just laid above or under the seabed. There are
basically no pipe supports used for underwater pipelines. Some only used anchor as fixed
support and others used rock dumping to ensure the pipelines are in-place. To cater for
thermal expansion and hydrodynamics forces (process related), bends are sometimes
intentionally introduced.
6
Pipelines resting on the seabed are subjected to fluid loading from both waves and
steady currents. For regions of the seabed where damage may result from vertical or
lateral movement of the pipeline it is a design requirement that the pipe weight is
sufficient to ensure stability under the worst possible environmental conditions. In some
circumstances, the pipeline may be allowed to move laterally provided stress (or strain)
limits are not exceeded. [10]
Figure 3: Free body diagram of pipeline for on-bottom stability analysis [10]
7
The loads acting on the pipeline due to wave and current action are; the
fluctuating drag, lift and inertia forces. In a design situation a factor of safety is required
by most pipeline codes, the components of hydrodynamics forces are shown below:
8
2.3 Pipeline Stress Analysis
Hoop stress
Longitudinal stress
Span Analysis
Stability analysis
Expansion and buckling analysis
Pipelines do not always rest continuously in contact with the seabed. There may
be spans where pipeline bridges across low points in profile. Spans can give rise to
various structural problems and may need to be corrected [14].
9
According to C.Kalliontzia, E. Andrianis, K. Spyropoulos and S. Doikas [17], the
mathematical treatment of pressurized submarine pipelines, which are freely laid on sea
floors, poses a considerable problem since the contact points are not known a priori. The
geometrical irregularities of the assumed frictionless seabed profile, which may either be
rigid or deformable, influence to a large extent the bending stress distribution along the
pipeline.
Research has been carried out in the past, aimed mainly at providing solutions
regarding the accurate prediction of pipeline configurations resting freely on seabeds. The
use of a reliable FE model for design predictions could allow the engineer to study
material and structural behaviors, especially in the remote regions of the structure where
physical observation or measurement is not possible [18].
Preliminary tests have been carried out by Oliver [18] on simply supported (SS)
and rigidly clamped pipes under quasi-static and impact loading conditions using rigid
patch and wedge indentors.
10
Figure 7: Result of free span analysis [19]
11
2.4 Seabed Topography Analysis
Figure 8: SIMLA with SimVis: Planning of pipe routes, trenching and rock dumping [20]
Alam M.R. and Mei C.C. [21] estimate the impact of long-period internal waves
on gas pipelines. They study on the evolution of internal solitary waves and the effect of
harmonic-generation in time-periodic waves travelling over random topography.
12
The irregular seabed profile is seen on the continental slope; a steep slope where
the mild slope continental shelf reaches ultra deep waters as seen in figure 10. Figure 11
shows visualizations of a rough seabed topography and subsea pipeline of the Ormen
Lange field (Norway) passing a rough seabed [14].
Figure 11: Subsea pipelines on rough seabed, Ormen Lange field, Norway [14]
13
2.5 Pressure Design of Pipeline [22]
2.5.1 Thin Wall Approximation
Consider a straight section of pipe filled with a pressurized liquid or gas. The
internal pressure generates three principal stresses in the pipe wall: as illustrated in Figure
14: a hoop stress σr . When the ratio of the pipe diameter to its wall thickness D/t is
greater than 20 the pipe may be considered to thin wall. In this case, the hoop stress is
nearly constant through the wall thickness and equal to
σh = PD
2t
The longitudinal stress is also constant through the wall and equal to half the hoop
stress
σ1 = PD
4t
The radial stress varies through the wall, from P at the inner surface of the pipe to zero
on the outer surface.
Figure 12: Hoop (h), Longitudinal (l) and Radial (r) Stress Directions
14
2.5.2 Pipeline design equation
For oil and gas pipelines, the thickness of the pipe wall is obtained by writing that
the hoop stress, which is the largest stress in the pipe, must be limited to a certain
allowable stress S. Using the thin wall approximation, this condition corresponds to
PD < S
2t
For hazardous liquid pipelines (hydrocarbon, carbon dioxide, etc.) the allowable stress
is set at [ASME B31.4]:
S = 0.72 SYE
S = SY F E T
15
E = Weld joint factor, Table 1
T = Temperature derating factor, Table 4
16
Table 3: Location Design Factor F [ASME B31.8]
Location F
Class 1 Div.1: Deserts, farm land, sparsely populated, etc 0.8
Class 1 Div.2: Class 1, with line tested to 110% design 0.72
Class 2: Industrial areas, town fringes, ranch, etc. 0.6
Class 3: Suburban housing, shipping centers, etc. 0.5
Class 4: Multistory buildings, heavy traffic, etc. 0.4
Note : Lowe location design factors apply at crossing, compressor station, etc. The
pipeline designer must refer to codes and regulations for the applicable location design
factor.
Temperature (ºF) T
250 or less 1.0
300 0.967
350 0.933
400 0.9
450 0.867
17
2.5.3 Lame’s formula
Without the thin wall approximation, the more general form of the three principal stresses
in a closed cylinder subject to internal pressure P is given by Lame’s formula.
σt = P
σr = P
σl = P
The allowable stress for pipelines is 72% Sy and does not depend on the material’s
ultimate strength. The allowable stress for power and process plant piping systems is
18
Sy (T ) = Minimum specified yield stress at design temperature T,psi
Su(T ) = Minimum specified ultimate strength at design temperature T, psi
Where the values of yield stress Sy or ultimate strength Su at design temperature are
larger than at room temperature, the room temperature values are used. Some values of
allowable stress are listed in Table 5.
19
CHAPTER 3: METHODOLOGY
The author had read some journals and articles to enhance the understanding on
irregular seabed topography and also stress analysis imposed on pipelines. In order to
complete the project according to the given time frame, the author had planned on the
project flow process as follows:
Project Identification
Literature Review
Define parameter and
data collection
Study the design and
parameter of pipeline
Calculation and
problem definition
Simulation and stress
analysis using CAESAR II
Satisfactory simulation
result
Conclusion and
recommendation
20
3.2 Material Selection
Generally, carbon steels are used for subsea pipelines. API-5L “Specification for
Line Pipe” (2000) is used for standard specifications. API-5L covers Grade B to Grade
X80 steels with Outside Diameter (OD) ranging from 4.5 to 80 inch. Table 1 shows
tensile strength properties according to API-5L. Generally the most common steel grade
used for deepwater subsea pipelines is X65, regarding its cost-effectiveness and adequate
welding technology [14]. Thus, for this project, pipe material used is API 5L X65.
Grade Yield Strength Yield Strength Ultimate Tensile Ultimate Tensile Elongation in
Min. (Psi) Max. (Psi) Strength Min. (Psi) Strength Max. (Psi) 2 in. min. (%)
B 35,000 65,000 60,000 110,000 a
X42 42,000 72,000 60,000 110,000 a
X46 46,000 76,000 63,000 110,000 a
X52 52,000 77,000 66,000 110,000 a
X56 56,000 79,000 71,000 110,000 a
X60 60,000 82,000 75,000 110,000 a
X65 65,000 87,000 77,000 110,000 a
X70 70,000 90,000 82,000 110,000 a
X80 80,000 100,000 90,000 120,000 a
The minimum elongation in 2 in. (50.8 mm) shall be that determined by the following
equation:
.
625,000 .
21
SI Unit Equation
.
1,944 .
where;
a. For both sizes of round bar specimens, 0.20 in.2 (130 mm2);
b. For full section specimens, the smaller of (i) 0.75 in.2 (485 mm2) and
(ii) the cross-sectional area of the test specimen, calculated using
specified outside diameter of the pipe and the specified wall thickness
of the pipe, rounded to the nearest 0.01 in.2 (10 mm2); and
c. For strip specimens, the smaller of (i) 0.75 in.2 (485 mm2) and (ii) the
cross-sectional area of the test specimen, calculated using the specified
width of the test specimen and the specified wall thickness of the pipe,
rounded to the nearest 0.01 in.2 (10 mm2).
By using higher grade steels, the required wall thickness is reduced. Therefore,
the cost of pipeline per meter is slightly reduced. Higher grade steels result in a lighter
pipeline, thus the tension is lower. This factor is very important in deep waters, where
required tension can be a limiting factor.
22
3.3 Line Pipe Specification
Table 2 below shows the physical information regarding the pipe geometry, steel
material strength and all other information required to define the necessary input for
stress analysis.
23
3.5 Load Cases
The pipeline system is analyzed for various load cases listed below, in accordance
with the Pipeline Design Code DEP 31.40.10.19/ISO 14692/ASME B 31.3
Legend:
24
3.6 Applicable Codes, Standards and Specifications
Following codes and standards, with the requirements in these design criteria,
shall form the basis for stress analysis. The International System of units (SI) shall be
used for all unit measurement.
Figure below shows the step by step procedures to complete the analysis from
starting point until the end where all the results are generated.
Compare obtained
result with
START Run simulation
maximum allowable
stress
Data collection
•Piping code/size Report based
Locate support
•Allowable stress output
•Load cases
Input to CAESAR II
•Wave load
Model pipelines END
•Pipe dimensions
25
Figure 14: CAESAR II workflow
26
3.8 Project Gantt Chart
No Detail/Week 1 2 3 4 5 6 7 8 9 10 11 12 13 14
1 Selection of Project Topic
2 Project Identification and
Planning
3 Preliminary Research Work
4 Submission of Preliminary
Report
5 Project Work:
Further research and
Mid-semester break
study
Literature review
6 Seminar (compulsory)
7 Project work continues:
Defining project
constraints and
criteria to be
evaluated
Developing the
analysis technique
9 Submission of Draft Report
10 Submission of Final Report
Milestone
Process
27
Table 10: FYP 2 Gantt Chart
No Detail/Week 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
11 Project Work Continues
Data gathering for
analysis
Start pipeline
modelling
Verifying results
12 Submission of Progress
Mid-semester break
Report
13 Project Work Continues:
Results of analysis
14 Pre-EDX
15 Submission of Draft Report
16 Submission of Dissertation
(Soft Bound)
17 Submission of Technical
Paper
18 Oral Presentation
19 Submission of Project
Dissertation (Hard Bound)
Milestone
Process
28
CHAPTER 4: RESULT AND DISCUSSION
The stresses, displacements, forces and moments for the system analyzed are
found to be within code allowable limits. The maximum stresses (refer Appendix A) and
maximum displacement from the analysis results are tabulated below.
During hydro test, the stress of the pipeline is subjected up to 12959.4 Psi at node
120. Hydrostatic testing is used to determine and verify pipeline integrity. Generally,
pipelines are hydrotested by filling the test section of pipe with water and pumping the
pressure up to a value that is higher than maximum allowable operating pressure
(MAOP).
Compared to all the applied load cases on the underwater pipeline, operating load
case 6 gives the maximum value of stress which 16701.6 Psi with stress ratio of 60.73
percent. Pipelines are loaded by operating conditions; basically, internal pressure and
temperature. The stress distribution is at maximum during operation of the pipeline due to
29
the increasing temperature. These lead to the expansion of the pipeline which later creates
stress upon certain areas such as the joints.
Sustained loads consist of internal pressure and dead-weight. Dead weight is from
the weight of pipes, fittings and components. Internal design or operating pressure causes
uniform circumferential stresses in the pipe wall, based on which a pipe thickness is
determined. Additionally, internal pressure gives rise to axial stresses in the pipe wall. A
pipe’s dead-weight causes it to bend between supports and nozzles, producing axial
stresses in the pipe wall. In the stress analysis, node 120 gives the highest value of stress
for all sustained load cases. Case 4 is the highest amongst other that reads stress of
12356.7 Psi with ratio nearly 50 %.
Expansion loads refer to the cyclic thermal expansion and contraction of pipe.
When the pipeline is restrained in the directions it thermally deforms, such constraint on
free thermal deformation generates cyclic thermal stress range, the system is susceptible
30
to failure by fatigue. To avoid fatigue failure, pipeline system should be made flexible.
Table shows the least pipe stress generated at support node 9050.
31
Figure 16: Front view of the pipeline routing
32
4.4 Seabed Topography Cases
There are three (3) cases of the seabed geometry and topography. Each of these
cases concentrates on a specific loading scenario and has different type of free span.
4.4.1 Case 1:
Offshore oil and gas pipelines are being subjected to deeper water depths, more
extreme environmental conditions and harsher operating requirements than ever before.
Thus, free spanning pipelines are becoming more common and are often unavoidable
during pipeline installation. For case 1, free spans are induced by elevated obstructions.
Loads are more focused at node 80 (refer figure 19), hence the stress distribution is
higher at that point of the pipeline. Fixed support which is an anchor and guide support is
located along node 40 to 90 to sustain the pipe from buckling or fatigue failure.
Figure 19: Slide-down shaped topography applied on the pipeline routing (Brown line
indicates the seabed topography)
33
4.4.2 Case 2
Case 2 is basically the same conditions with case 1, only the stress distribution is
higher at 90 degree shaped of seabed. This is due to the hanging pipeline in between the
corner of the 90 degree seabed (refer figure 21). Without suitable type of support, the
pipeline at the point may buckle and later will cause fatigue failure. Guide support is
located in the middle between node 140 and node 150 to ensure the pipeline can
withstand load subjected to it. Higher stress results were obtained at node 150 which
means critical point of the pipeline.
Figure 21: Stair-case shaped topography applied on the pipeline routing (Brown line
indicates the seabed topography)
34
4.4.3 Case 3
The gap between the pipeline and seafloor will affect the free stream velocity of
the current passing around the free spanning pipe. This gap can also limit the amount of
deflection that may occur due to static and dynamic loading. In general, as pipe tension
increases, the maximum allowable span length increases. The stresses on the free span
due to static loading are not affected significantly by the increase in pipe tension. For
case 3, no support is located along node 150 to 160. The humps of the seabed in a way
creates a +Y rest support to the pipeline. The stress distribution is stable along the
pipeline due to uniform load subjected.
Figure 23: Wavy shaped topography applied on the pipeline routing (Brown line indicates
the seabed topography
35
4.5 Stress Summary Result
PASSED
The pipeline met all the criteria and passed all the analysis. With suitable type of
support attached to the pipeline, it can withstand the maximum possible load subjected.
The material used for the pipeline has the best strength to make the lifetime of the
pipeline last longer. Appendix A shows the maximum calculated stress for all load cases
and appendix B shows the displacement of each node in the pipeline.
36
4.6 Support Location
The pipeline is laid on the seabed and fully constrained. There are one anchor
block at the connecting point or tie in point between the pipeline and the riser. Only three
types of supports that is used in this pipeline which are resting (+Y), guide with 3mm gap
and limit stop. Support type and support location is listed in the table below and support
shall be designed considering the loading.
37
22 180 +Y, Lim Rest support and limit stop
23 9050 +Y, Guide Rest support, guide with 3mm gap
38
CHAPTER 5: CONCLUSION
This report has covered the pipeline modeling and performed pipe stress analysis.
Analytical solutions were developed using pipe stress analysis software, CAESAR II
version 5.10 to generate stress distributions along the underwater pipeline of 250mm
diameter. The design code compliance was verified for the subsea pipeline laying on
irregular seabed geometry under ocean current loads.
It is concluded from the stress analysis result that the system is within design
envelope and stress are acceptable under operating/design conditions. With the
recommended support system, stresses are kept low within code allowable limits. The
pipe stress analysis of underwater pipeline with DN250 has successfully been analyzed
using CAESAR II. The X65 Carbon Steel pipeline has maximum strength that is able to
withstand the highest value of current load.
The maximum allowable stress along the pipeline was calculated to be 27500 Psi.
The maximum stress on the pipeline were up to 16701.6 Psi located at node 130 with
60.73 % stress ratio. While, the maximum displacement can be seen at node 9030 (refer
appendix B of page B11) of 3 inches in +Z direction, 0.0129 inches in +Y direction and
0.2797 inches and +X direction. Node 9030 is a means of guide support. It is not laid on
the seabed due to the gap of free spanning (refer case 2). All of the nodes have higher
displacements during load case 5 and 6 which is the operating conditions compared to
other load cases.
The stress analysis result was compared using manual template calculation to
validate the value. This result should only be used as a guide in determining the most
practical and reasonable maximum allowable stress for a given case.
39
REFERENCES
[1] Project Consulting Services, Inc., December 1997, Analysis and Assessment of
Unsupported Subsea Pipeline Spans, United States Department of The Interior Minerals
Management Service.
[3] Andrew C. Palmer and Roger A. King, 2004, “Subsea Pipeline Engineering”,
PennWell Corporation, Oklahoma.
[4] JP Kenny, 1993, Structural Analysis of Pipeline Span, JP Kenny and Partners Ltd.
[5] Michele G. Bishop, 2002, Petroleum System of the Malay Basin Province, Malaysia,
Central Region Energy Resources Team, U.S. Department of U.S. Geological Survey
[6] Andrew C. Palmer, Rofer A. King, 2008, “Subsea Pipeline Engineering”, 2nd Edition,
Pennwell Corporation
[9] F.P. Gao, D.S. Jeng and H. Sekiguchi, 2003, “Numerical study on the interaction
between non-linear wave, buried pipeline and non-homogenous porous seabed”,
Computers and Geotechnics, Volume 30, 535-547
[10] Yong Bai, 2001, Pipelines And Risers, ELSEVIER Ocean Engineering Book Series,
Volume 3, USA.
40
[11] DNV-RP-F109, 2007, Rules of Submarine Pipeline, On-Bottom Stability Design of
Submarine Pipelines.
[12] Intecsea, Offshore Pipelines Capability and Experience, Worley Parsons Group
[13] Lisa King, Dr Jeremy Leggoe, W/Prof Liang Cheng, “Hydrodynamic Forces on
Subsea Pipes due to Orbital Wave Effects”, The University of Western Australia,
Woodside Energy Ltd
[15] G. Maier and F. Andreuzzi, 1978, Elastic and Elasto-plastic Analysis of Submarine
Pipelines as Unilateral Contact Problems, Computers & Structures Vol. 8, 421-431
[16] P.H. Chuang and D.L. Smith, 1992, Elastic Analysis of Submarine Pipelines, J.
Structural Engineering ASCE 119(1), 90-107
[18] O.O.R Famiyesin, K.D. Oliver, A.A. Rodger, 4 May 2002, Semi-empirical
Equations for Pipeline Design by Finite Element Method, Computers & Structures,
Volume 80, 1369-1382
41
[21] Alam, M.-R. and Mei, C.C., 2008, "Ships advancing near the critical speed in a
shallow channel with a randomly uneven bed", J. Fluid Mechanics, Volume 616, 397-417
<http://www.sintef.no/home/MARINTEK/MARINTEK-Research-
Programmes/DEEPLINE/> , date retrieved: 4 August 2011
[22] Preliminary of Piping and Pipeline Engineering, Kikuchi Industry (Thailand) Co.,
Ltd.
[23] JP Kenny, 1993, “Structural Analysis of Pipeline Span”, JP Kenny and Partners Ltd.
[24] Iwan R., Lambrakos K.F., Billy L., 1999, “Prediction of Hydrodynamic Forces on
Submarine Pipelines Using an Improved Wake II Model”, Ocean Engineering Journal,
Volume 26, 431-462
[25] Ian A.R., John B.H., 1998, “Wave and wave-current loading on a bottom-mounted
circular cylinder”, International Journal of Offshore and Polar Engineering, Volume
8(2), 122
42
CAESAR II Ver.5.10.00, (Build 070917) Date: DEC 29, 2011 Time: 10:53
Job: D:\FYP2CAESAR\PIPELINE1
Licensed To: DEALR/EVAL COPY -- ID #4369
1 (SUS) WNC
2 (SUS) W
3 (SUS) WW
4 (SUS) W+P1
5 (OPE) W+T1+P1
6 (OPE) W+T2+P1
7 (HYD) WW+HP
8 (EXP) L8=L5-L4
9 (EXP) L9=L6-L4
A1
CAESAR II Ver.5.10.00, (Build 070917) Date: DEC 29, 2011 Time: 10:53
Job: D:\FYP2CAESAR\PIPELINE1
Licensed To: DEALR/EVAL COPY -- ID #4369
A3
CAESAR II Ver.5.10.00, (Build 070917) Date: DEC 29, 2011 Time: 10:53
Job: D:\FYP2CAESAR\PIPELINE1
Licensed To: DEALR/EVAL COPY -- ID #4369
A4
1
1 (SUS) WNC
2 (SUS) W
3 (SUS) WW
4 (SUS) W+P1
5 (OPE) W+T1+P1
6 (OPE) W+T2+P1
7 (HYD) WW+HP
8 (EXP) L8=L5-L4
9 (EXP) L9=L6-L4
2
CAESAR II Ver.5.10.00, (Build 070917) Date: DEC 28, 2011 Time: 10:53
Job: D:\FYP2CAESAR\PIPELINE1
Licensed To: DEALR/EVAL COPY -- ID #4369
Temperature (oC)
STD 40 9.27 8385 8385 7923 7131 7048 5450 4527 3652
250 80 15.09 13863 13863 13098 11781 11644 9012 7483 6028
100 18.26 16922 16922 15992 14386 14214 10996 9136 7359
120 21.44 20036 20036 18934 17032 16825 13022 10817 8716
XXS 140 25.4 23998 23998 16474 20394 20153 15599 12960 10438
160 28.58 27229 27229 25734 23143 22875 17700 14703 11844
C1
APPENDIX D: The density of some common liquids can be found in the table below:
Temperature Density
Liquid -t- -ρ-
(oC) (kg/m3)
D1
Temperature Density
Liquid -t- -ρ-
(oC) (kg/m3)
1 kg/m3 = 0.001 g/cm3 = 0.0005780 oz/in3 = 0.16036 oz/gal (Imperial) = 0.1335 oz/gal (U.S.) =
0.0624 lb/ft3 = 0.000036127 lb/in3 = 1.6856 lb/yd3 = 0.010022 lb/gal (Imperial) = 0.008345
lb/gal (U.S) = 0.0007525 ton/yd3
D2