Journal of Pipeline Science and Engineering 4 (2024) 100196

Contents lists available at ScienceDirect

Journal of Pipeline Science and Engineering

journal homepage: www.keaipublishing.com/en/journals/journal-of-pipeline-
science-and-engineering/

Strain concentration factor at field joints for offshore concrete coated

pipelines – Literature review
Mohamed Elgazzar a , Ahmed Reda b, c, * , Ibrahim Sultan a , Truong Phung a
a
Institute of Innovation, Sustainability and Science, Federation University, VIC, Australia
b
School of Civil and Mechanical Engineering, Curtin University, WA, Perth, Australia
c
Cladtek International PTY LTD, Perth, Australia

A R T I C L E I N F O A B S T R A C T

Keywords: This paper introduces a literature review of the accomplished research to calculate the strain concentration factor
Strain concentration factor which shall be considered during the concrete coated subsea pipelines design. The strain concentration results
Stress concentration factor from the discontinuity of the concrete coating which is applied to pipe joints to enhance the pipeline stability on
Subsea pipelines
the seabed. The paper introduces a definition of the strain concentration factor and a conclusive explanation of
On-bottom stability
Concrete weight coating
the work performed to calculate the strain concentration factor using the following three methods: full-scale
Pipeline installation tests, analytical models, and numerical models. The paper also introduces other works addressing the contri-
Pure bending tests bution of concrete coating to pipe stiffness. Literature indicates that while full-scale tests produce the most ac-
Push-off tests curate values for the strain concentration factor, due to cost implications, the method is limited to a small range
Corrosion coating of pipe sizes and coating properties. Although the analytical method produces acceptable results, prediction of
Analytical modelling strain concentration factor due to concrete coating sliding and beyond steel yield stress and concrete crushing
Semi-analytical modelling limit is unpredictable. The numerical method using finite element (FE) analysis indicates acceptable strain
Numerical modelling
concentration factor values, however, careful consideration shall be taken due to possible modelling inaccura-
3D modelling
cies. The review shows that the strain concentration factor is mainly influenced by the thickness and strength of
Linear & non-linear behaviour
Finite element analysis
the concrete coating as well as the shear capacity of the external corrosion coating, whilst due to its low impact,
the effect of the concrete coating reinforcement is not considered for most of the analysis due to low significance.

1. Introduction that results from the effect of surface waves and bottom currents
(DNV-RP-F109, 2021). The forces investigated during the pipeline
Subsea pipelines as a vital asset in the oil and gas industry are on-bottom stability analysis are as follows:
employed for the transmission of hydrocarbon products. Flowlines are
used to transport unprocessed fluid from the wells to a processing or 1) Acting forces: The applied hydrodynamic forces from waves and
storage facility, which can be either onshore (refinery, process plant, bottom steady current (lift, drag, and inertia) that are acting to
and tanks) or offshore (platforms). The gathering pipelines transport the displace the pipeline.
fluid from the different storage or processing facilities to a common 2) Resisting forces: The pipeline submerged weight and soil friction
trunk pipeline, finally, trunklines transmit fluids from the gathering which are acting against the hydrodynamic forces and are trying to
pipeline to shore facilities. A pipeline laid on the seabed will be sub- maintain the pipeline’s position.
jected to horizontal and vertical loading due to wave and current ac-
tions. The pipelines should be designed to have sufficient specific gravity According to Reda et al. (2020, 2021a), two different methods of
to remain stable under the worst combination of environmental loadings pipeline on-bottom stability are utilised, and the choice between the

* Corresponding author.
E-mail address: [email protected] (A. Reda).

https://doi.org/10.1016/j.jpse.2024.100196
Received 16 January 2024; Received in revised form 15 April 2024; Accepted 16 April 2024
Available online 20 April 2024
2667-1433/© 2024 The Authors. Publishing Services by Elsevier B.V. on behalf of KeAi Communications Co. Ltd. This is an open access article under the CC
BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
M. Elgazzar et al. Journal of Pipeline Science and Engineering 4 (2024) 100196

analysis methods is dependent on the degree of detail required in the subject to the following limit states as per DNV-ST-F101 (2021):
results:
1) Ultimate limit state (ULS)
2) Serviceability limit state (SLS)
1.1. Absolute stability
3) Accidental limit state (ALS)
4) Fatigue limit state (FLS)
In this method, forces are acting in equilibrium on the pipeline
resulting in zero pipeline displacement. Static analysis is used to
determine the pipeline stability requirements assuming that absolute
1.3. Concrete weight coating
stability is required at certain locations along the pipeline route.
The application of concrete weight coating to the pipelines has been
1.2. Dynamic stability adopted in the offshore pipeline industry for decades since the concrete
coating can achieve the purpose of pipe stability on the seabed, and it is
Dynamic stability can be employed to reduce the thickness of the more cost-effective when it is compared to other secondary stabilisation
concrete weight coating in locations where absolute stability is not methods such as rock dumping or trenching (Fig. 1). The submerged
necessary. In this method, the pipeline is allowed to displace on the seabed weight of the pipeline can be increased by either increasing the thickness
provided that adequate integrity to its structural components and the of the concrete coating, or by using a higher concrete density. Practical
adjacent facilities is maintained. Therefore, this method provides a po- limits of concrete application would limit the thickness to about 125 mm
tential for significant reductions in the concrete weight coating thickness (Reda et al., 2021a, 2021b). Concrete coating density may be increased
and the overall stabilisation costs. Dynamic analysis involves full dynamic up to 3,400 kg/m3 by using iron ore as an aggregate material.
simulations of a pipeline on the seabed subjected to the relevant storm Concrete coating is applied to pipe joints except at the ends, typically
conditions, including modelling of soil resistance, hydrodynamic forces, 300–350 mm of each pipe end is left without concrete coating, pipes are
boundary conditions, and dynamic response. The results of the dynamic then joined together by butt welding their ends which is known as field
analysis are pipeline displacements and stresses. The on-bottom stability of joints (FJs) (Fig. 2). The installation contractor applies a heat shrink
the pipeline is provided by specifying the concrete coating thickness to sleeve around the girth weld to provide corrosion protection to the
achieve the submerged weight for the required level of on-bottom stability exposed section of the steel pipe. However, this sleeve is not shown in

Fig. 1. Stacking of concrete coated cteel pipes at a fabrication yard/storage facility.

2
M. Elgazzar et al. Journal of Pipeline Science and Engineering 4 (2024) 100196

Fig. 2. Schematic diagram for pipe field joint showing the concrete coating cutback and the corrosion coating cutback.

Fig. 2 for simplification purposes. Where εfj represents the strain in the field joint; εg represents the global
strain.
1.4. Strain concentration factor Hook’s Law writes as below.

The stress concentration factor (SCF) or strain concentration factor σ

E= (2)
(SNCF) are applied to the calculated load to reflect different aspects not ε
captured directly in the calculation (local geometries, coating, varia- Where E represents Young’s Modulus; σ represents stress and ε repre-
tions in properties). SCF is mainly used in the linear material range and sents strain.
SNCF is used in the linear and the non-linear material range. A Fig. 3 shows a bending test which is simulating subsea pipeline
distinction should be made between global and local SCF/SNCF. Local bending during installation or while resting on the seabed (Sultan et al.,
stress concentration affects the pipe locally and is accounted for in fa- 2013). The bending test presents strains and stresses of the pipe due to
tigue fracture evaluations, and it is usually caused by welded attach- installation, environmental loads (wave & currents), or operational
ments, the weld itself, or very local discontinuities. Global stress/strain loads (such as global buckling). A testing rig as shown in Fig. 4 is used to
concentration, which is addressed in this article, affects the pipe globally simulate the different bending scenarios for concrete coated pipes. When
and is accounted for in the bending and buckling evaluations as well as bending starts until the concrete reaches its tensile limit, the entire
fatigue and fracture evaluations. concrete coating section resists the bending application (i.e., both tensile
The pipeline design involves the use of thick concrete and/or insu- and compression sides) utilising maximum concrete coating stiffness.
lation coatings, the mismatch in the bending stiffness between the During this period the low curvature associated with the low bending
coated pipe sections and the field joint results in stress/strain amplifi- moment relatively reduces the global stress/strain concentration. This
cation in the field joints, hence a global strain concentration in the field occurs while the stiffness mismatch between the coated pipe section and
joint (Bai, 2001). The amplified strain will directly affect the unity the field joint is in its highest values.
checks associated with local buckling, fatigue, and fracture. When the concrete coating reaches its tensile capacity (at approxi-
According to the beam theory, SNCF is defined as the ratio between mately 7 %− 10 % of the compression capacity), it cracks on the tension
the maximum longitudinal strain in steel at the field joint and the global side (as shown in Fig. 5). Consequently, the concrete contributes to the
longitudinal strain along the entire pipe (Igland and Moan, 2000). pipe stiffness solely through the compression side. During this period,
εfj while the concrete pipe stiffness decreases due to the lack of tensile
SNCF = (1)
εg capacity, the relatively higher curvature resulting from a higher bending

Fig. 3. Bending application to a bare steel pipe.

3
M. Elgazzar et al. Journal of Pipeline Science and Engineering 4 (2024) 100196

Fig. 4. Testing rig showing bending application to concrete coated pipe.

Fig. 5. Testing rig showing cracks observed at the tensile side of the concrete coating.

moment application along with the increased stiffness from the The longitudinal strain in bare steel is governed by pipe size, steel
compression capacity, leads to a relatively high global strain/stress grade, and the applied loading. Furthermore, the longitudinal strain in
concentration. However, full-scale tests and numerical modelling have concrete-coated pipes is dependent on the thickness and strength of the
shown that there is a relative sliding between the concrete coating and concrete coating, as well as and the thickness and strength of the
the steel pipe. This sliding leads to a reduction in the global stress/strain corrosion coating. For concrete-coated pipelines, modelling the field
concentration. Furthermore, it results from the shear strength which joints along the pipeline can be time and resource-consuming. Hence,
exceeded between the concrete coating and the corrosion coating under the pipeline is modelled considering bare steel sections where the stress-
specific loading. strain curve of the bare steel is used. The maximum strain in the field
When the concrete coating reaches its compression capacity, which joint is calculated by multiplying the SNCF by the maximum strain ob-
typically occurs at approximately 0.2 % strain, it ceases to function tained from either full-scale-test, analytical, or numerical models.
effectively on both the compression and tensile sides (as illustrated in Alternatively, the SNCF can be applied to the results of the pipelay
Fig. 6). Consequently, the contribution of the concrete coating’s stiffness analysis performed using a global moment-curvature model as indicated
becomes negligible. Therefore, the strain concentration tends to reach by Ness and Verley (1996). Underestimation of the SNCF can cause
its lowest values or may entirely vanish. significant failure in the pipeline during installation (ZEE Engineering,

4
M. Elgazzar et al. Journal of Pipeline Science and Engineering 4 (2024) 100196

Fig. 6. Testing rig showing cracks observed at both tensile and compression sides of the concrete coating.

2014). Conversely, the overestimation of the strain concentration can coated pipes, the objectives of these tests were to investigate the
lead to undesirable conservatism and high material and installation contribution of the concrete coating and the corrosion coating to the pipe
costs. The increase in strain at the field joint for concrete-coated pipe- stiffness and to understand the pipeline behaviour during pipe laying. The
lines, due to reduced pipe stiffness is also considered significant during strain concentration factor is calculated as the ratio between the strain in
pipelay analysis and in-service pipeline analysis as indicated by Reda the bare steel pipe (field joint) and the global (nominal) strain of the pipe
et al. (2018, 2019). The objective of this article is to present a literature joint. Livingston (1965) carried out full-scale tests on concrete-coated
review of the work conducted to investigate the increased stiffness of pipes to verify the results of the analytical model that the author created
concrete-coated pipes and the work performed to calculate the global within the same study. The tested pipes were coated with asphalt material
strain concentration factor for concrete-coated subsea pipelines. before the concrete jacket was applied. Livingston used chicken wire
reinforcement inside the concrete coating to avoid the concrete spalling.
2. Accomplished research The tested pipe is mounted vertically in the test equipment to avoid
bending moment due to its own weight, the bending process is achieved by
This section presents the literature performed to investigate the ef- employing a loading linkage that applies a pure torque at both ends of the
fect of the reduced pipeline stiffness at the field joint and the estimation test member (Fig. 7). The principal elements are the test member in the
of the SNCF. The effect of the reduced stiffness at the field joints has been centre while both the tension and compression members are at the sides.
addressed in previous projects, reports, and technical papers. The work The load is measured by strain gauges mounted on the tension member.
featured in the literature was based on full-scale laboratory testing, Test member deflections are measured by a low-power microscope on the
analytical/semi-analytical approach, numerical approach as well as stationary angle and a scale attached to the member. The used loading
other resources indicated empirical equations as presented by design linkages are arranged so that the tension and compression members can be
standards DNV-RP-F105 (2021). interchanged, so the moment can be applied to the test member in both
Researchers estimated the SNCF for concrete-coated pipelines by directions. Results have shown a 32 % increase in the flexural rigidity of
different methods as categorised below: the concrete-coated pipes compared to bare steel pipes.
Another full-scale test was performed by Mogbo et al. (1971) to
1) Full-scale test. verify the results of the analytical model which the author created
2) Analytical methods. within the same study, Mogbo performed eight full-scale bending tests
3) Numerical methods by finite element analysis (FEA) computer and several push-off tests. The test rig that Mogbo used for the bending
models. test consists of four-point loading as shown in Fig. 8, the loading
arrangement produces a region of pure bending between the two middle
Table 1 shows a descriptive comparison between the literature supports, whilst ensuring no associate axial forces. The test frame is
introduced in the literature, however, more details are explained in the mounted in a way that allows axial movement or rotation. The field joint
below sections. is placed within the two middle supports to generate results of a pure
bending process. The author measured the strains using strain gauges
2.1. Estimating the SNCF using full-scale test mounted on the steel and concrete, while the deflection of the pipe is
measured using a high precision level and scales hung from the cen-
Several researchers performed full-scale bending tests on concrete- trelines of the pipe.

5
M. Elgazzar et al. Journal of Pipeline Science and Engineering 4 (2024) 100196

75 %–85 % for pipes with field joints. The computed stiffness values
compared to the measured stiffness showed considerable agreement for
the limited number of tests performed. The study also presented several
push-off tests on short pipe sections to estimate the properties of the
corrosion coating material. To find an economical and reliable method
to prevent concrete coating slippage over thin film epoxy (TFE), Archer
and Adams (1983) implemented three methods to increase the shear
strength between the concrete and the TFE coating: Epoxy Spiral, Barrier
Coat, and Rough-Band. A series of push-off tests, bending tests, and
simulated laying tests were conducted. Bending tests were performed on
a few specimens after the lay barge test to simulate the actual pipeline
laying scenario. The push-off tests were repeated after the bending tests
to reassess the shear strength of the corrosion coating post-installation
(Fig. 9). The rough band has proven best option for controlling concrete
coating slippage during the pipelay, in addition, the added stiffness of
the concrete coating was found lower than using a continuous barrier
coat. Other full-scale tests were performed during the Gas Trans-
portation System project for Statpipe in the North Sea as Akten et al.
(1985) indicated in his literature, several push-off tests were carried out
to investigate the actual shear strength of the corrosion coating asphalt
material, the results of push-off tests showed shear strengths higher than
the minimum required for the project (0.104 MPa), the literature
included a description of field bending tests performed on three pipeline
strings each consisting of eight 12 m concrete coated pipes welded
together. Bending of the pipes was achieved by lifting one end of each
string off the ground at incremental heights. The test was then repeated
by lifting the second end of the string. The pipe strings were then rolled
through 180◦ and the lifts at each end are repeated to simulate the pipe
going from over bend to sag bend. A photo system was used to determine
the shape of each string at various stages of the lifting operation. The
literature concluded that slotting of concrete would be beneficial for the
Fig. 7. Pure bending test apparatus (Livingston, 1965). thickest concrete coatings, and maximum strains will be within the
project requirements. Figs. 9 and 10 show schematic diagrams for the
The results indicated that field joints allow for concrete coating corrosion coating shear strength (push-off) and photo of a similar pure
slippage, thereby, reducing the stiffness contribution of the concrete bending tests performed in the literature, respectively. It should be
coating. The stiffness of the concrete-coated pipes increased by 10 % – noted that slotting of the concrete coating is no longer used in the
34 % compared to the bare pipe, aligning with Livingston (1965). The pipelines, the current industry practice is to apply a continuous concrete
measured stiffness values were less than the computed stiffness, spe- coating on the pipe joints, only 350 mm at both pipe joint ends are left
cifically, 85 %–93 % of the computed stiffness for fully coated pipes and bare for welding purposes. Lund et al. (1993) shows the results of the

Fig. 8. Schematic diagram of the pure bending test setup (Mogbo et al., 1971).

6
M. Elgazzar et al. Journal of Pipeline Science and Engineering 4 (2024) 100196

Fig. 9. Schematic diagram of the concrete coating to corrosion coating interface shear test (push-off test) (Akten et al., 1985).

Fig. 10. Photo of a pure pipe bending test similar to the pure test (Akten et al., 1985).

7
M. Elgazzar et al. Journal of Pipeline Science and Engineering 4 (2024) 100196

Fig. 11. Strain distribution during lifting of 28″ string (Lund et al., 1993).

bending test indicated by Akten et al. (1985), and shows the strain this review, especially when laboratory tests are available, Elchalakani
measurement for the 28″ pipe bending test (Fig. 11). et al. (2001) has carried out experimental tests to investigate the
Ness and Verley (1995) performed full-scale bending tests on behaviour of the cold-formed steel tubes filled with concrete and sub-
concrete-coated pipes with field joints (Fig. 12). Ness performed four jected to large bending moment, the study determined the plastic limit
laboratory tests using different pipes and coating parameters. The of the diameter to thickness ratio (d/t), and presented a comparison
author presented the strain concentration factor distribution along against test results of hollow sections of similar pipes published later in
global strain until the concrete crushing. The maximum recorded global Elchalakani et al. (2016). The study indicated that there is more
strain is 0.33 % and 0.30 % for Test 2 and Test 3 as shown in Fig. 13. Ness enhancement in ductility due to concrete filling for thinner sections, the
also presented the behaviour of the corrosion coating under rapid and strong bond between the concrete and steel where no slippage is noticed
slow loading and the impact on the resulting SNCF. The full-scale tests during the test has also contributed towards the enhanced stiffness. This
are considered an effective method of validating the analytical and test proves that slippage of concrete over the steel pipe is a significant
numerical models for subsequent studies that are studying factor in determining the pipe stiffness and therefore the stress/strain
concrete-coated pipes under pure bending. Push-off full-scale tests were concentration within the field joint. The study has been extended by Hu
performed to estimate the shear strength of the asphalt corrosion et al. (2010) by investigating the bending capacity using a numerical
coating, to assist in investigating the feasibility of using high tensioner approach where results were found to be in good agreement with the
capacity during the pipelay operation. laboratory tests.
The contribution of the concrete filling to the stiffness of the steel While the full-scale test presents realistic measurements of the strain
pipes subjected to pure bending moment is found worth listing within concentration factor, the cost associated with test setup and test pipe is

Fig. 12. Schematic diagram of the full-scale pure bending test performed by Ness and Verley (1995).

8
M. Elgazzar et al. Journal of Pipeline Science and Engineering 4 (2024) 100196

analytical model results. The calculated increase in the stiffness of

concrete-coated pipe ranged between 19 % and 50 % compared to un-
coated pipes, significantly higher than the stiffness resulting from the
laboratory test (10 %–34 %), the average calculated stiffness is found in
agreement with the analytical model by Livingston (1965), the ratio of
the measured to calculated stiffnesses for pipes with continuous concrete
coating and pipes with field joints is 0.85–0.93 and 0.75–0.85 respec-
tively, indicating more accuracy in calculating the stiffness of the
continuously concrete-coated pipes than the pipes with field joints. A
reliability analysis by Igland and Moan (1993) was presented using a
developed limit state equation for steel pipelines subjected to pressure,
bending, and tension, the paper demonstrated that higher safety factors
should be applied for strain-controlled compared to stress-controlled
design criteria. Hence, the strain formulation yields a more economical
design than the stress formulation. The outcome of the paper contributes
towards the theory of allowing a portion of plastic strain to the steel pipe
during laying analysis, hence reducing the requirement of unnecessary
usage of thicker pipes. Lund et al. (1993) introduced an analytical model
predicting the equivalent bending stiffness of concrete-coated pipes when
subjected to bending and axial loads, the analytical model considered
linear steel behaviour, however, the model neglects the concrete slippage
and the concrete tensile capacity, moreover, the analytical model as-
sumes an equivalent moment curvature curve for the pipe, which is
chosen based on the stiffness of the concrete coated composite section
when the concrete strain reaches 0.2 %. As shown in Fig. 14, the calcu-
lated SNCF is underestimated by approximately 16 %, while significantly
overestimated post concrete crushing. Results are compared to field
bending tests conducted during the Statepipe project in North Sea (Akten
et al., 1985) and a good match was found. The work is followed by a
Fig. 13. SCF comparison between test (measured) by Ness and Verley (1995), parametric study estimating the SNCF at field joints during installation.
and analytical by Lund et al. (1993). As shown in Fig. 15, the literature indicates a remarkable effect of the
concrete coating thickness and strength on the pipe stiffness.
considered not insignificant. Furthermore, full-scale tests are performed Bruschi et al. (1995) continued the study that Lund started, he
to limited number of pipe sizes and coating properties. It is also worth investigated the stiffness of the concrete-coated pipelines when
noting that whilst most tests cover pure bending loading, subsea pipe- exceeding the elasticity limit, and concluded that any relaxation criteria
lines are always subjected to internal pressure, however laboratory tests to the strains generated in the pipeline during pipe lay shall be recon-
generally do not cover pipes subject to internal pressure due to required sidered since the concrete coating adds a significant contribution to the
higher safety measures and precautions. strains at weak sections of the pipeline (i.e., field joint) and can jeopardize
the integrity of the pipeline. Apart from studying the added stiffness due
to concrete coating application, Endal et al. (1995) presented a simplified
2.2. Estimating the SNCF using the analytical method analytical solution for the prediction of pipeline roll during installation.
When the pipeline exceeds the elasticity limit while at the overbend, the
As indicated in the previous section, Livingston (1965) introduced an
analytical method to calculate the increased flexural rigidity of the pipe
due to the application of concrete coating, the study is followed by
experimental (full-scale) tests to verify the analytical method results. The
analytical model assumed that the pipe is subjected to pure bending, with
no tensile capacity for concrete, and no slippage between the concrete
coating and steel pipe. The analytical model had shown a 17.8 % increase
in the flexural rigidity of the pipe compared to an increase of 32 %
measured from the laboratory test, indicating a shortfall of approximately
15 % using the analytical approach. When Livingston modified the model
to account for the concrete tensile capacity a flexural rigidity of 38 % was
computed, indicating higher rigidity than the laboratory test. To let the
analytical model account for the test results, 37 % of the test member was
considered to follow zero tension theory, and 63 % was supported by
equal compressive and tensile stresses. Since many parameters, such as
actual concrete tensile strength and concrete slippage cannot be accu-
rately predicted, the author recommended disregarding the tensile stress
in concrete. The results expressed only one full-scale test which is not
sufficient to validate the approach; therefore, additional tests with
varying pipe and coating properties are recommended. Another analyt-
ical model was developed by Mogbo et al. (1971) to calculate the effective Fig. 14. Comparison of the resulting SNCF, analytical by Ness and Verley
stiffness of concrete-coated pipes followed by a full-scale test to verify the (1995), numerical by Endal (1994), and analytical by Lund et al. (1993).

9
M. Elgazzar et al. Journal of Pipeline Science and Engineering 4 (2024) 100196

pipeline leaves the stinger with a residual curvature at the inflection point
where the pipeline attempts to overcome the residual curvature through
bending and twisting which occurs at the sag bend. The paper concluded
that if plastic deformation occurs on the stinger, for certain conditions, a
roll is likely to occur at the sag bend with minimal out-of-plane
displacement, a roll in the order of 180◦ is expected. Following the
analytical model introduced by Lund et al. (1993), Ness and Verley
(1995) presented an analytical model to calculate the SNCF in
concrete-coated pipes, the analytical model accounts for the nonlinear
behaviour of the steel, concrete, and corrosion coating. However, the
model did not account for the concrete tensile capacity. The validity of the
analytical model is envisaged at the concrete crushing limit (maximum
strain in the concrete is 0.2 %) where the SNCF value is held constant
beyond this point. Fig. 13 shows a comparison between the SNCF that
Ness obtained from both the analytical and the full-scale tests which is
indicated earlier in this review for test 2 and test 3 respectively. As
noticed, the results show the average SNCF resulting from the analytical
model is slightly higher than the laboratory test (in the order of 2 %− 4 %),
especially past 0.1 %–0.15 % strains, the analytical model shows constant
SNCF past 0.2 % strain, while the SNCF resulting from the laboratory test
decreases due to concrete crushing. Ness carried out parametric studies to
calculate the SCF amongst different concrete coating thicknesses and
strengths as shown in Fig. 16.
Eltaher et al. (2011) presented a semi-analytical approach to calcu-
late the SNCF on concrete-coated pipelines, unlike the approaches pre-
sented by Lund and Ness, El-Taher presented the results of the study
based on the load-controlled and the displacement-controlled ap-
proaches. Although the semi-analytical model does not distinguish be-
tween either side of the pipe (tension and compression), the author
compared the results against the 3D FEA outcome which was carried out
within the same study and calculated the SNCF for both pipe sides. As
indicated in Fig. 17, results show that the numerically calculated SNCF
at the compression side of the pipe is always larger than the tension side,
for both the displacement-controlled and load-controlled criterion, the
difference between compression and tensile sides becomes bigger when
Fig. 15. Results of pipe stiffness parametric study by Lund et al. (1993): (a) the global strain exceeds 0.2 %. When displacement-controlled criteria
concrete coating thickness, (b) concrete coating strength.

Table 1
Descriptive comparison between the literature.
Author Full scale test Analytical Numerical FE

Livingston (1965) 32 % flexural rigidity increase No concrete tensile capacity, 17.8 % flexural rigidity increase. N/A
Concrete tensile capacity, 38 % flexural rigidity increase.
Mogbo et al. (1971) 10 %− 34 % stiffness increase 19 %− 50 % stiffness increase. N/A
Fully Coated pipes, 17 %− 49 % stiffness increase.
Pipes with FJ, 25 %− 59 % stiffness increase.
Lund et al. (1993) N/A Used equivalent moment-curvature curve for the pipe. N/A
Deviates from Ness’s test/analytical, Endal’s numerical.
Endal (1994) N/A N/A Concrete tensile capacity is considered.
Concrete crushing is not accounted.
Acceptable match to Ness’s analytical
model.
Ness and Verley (1996) Test 2, average SNCF 1.37 Test 2, average SNCF 1.42 N/A
Test 3, average SNCF 1.18 Test 2, average SNCF 1.16
DNV-F105 (2021) N/A Matches Ness’s Test N/A
Pre-concrete crushing, Over-predicted (about 6 %)
Post-concrete crushing is underpredicted (about 7 %).
Bai (2001) N/A Underestimated by 15 % compared to Ness’s test N/A
Nourpanah and Taheri (2009a) N/A N/A Matches Ness’s Test
Nourpanah and Taheri (2009b) N/A Eq. (8) matches the SNCF compared to the numerical model N/A
and Ness’s test. (6 %− 11 %)
Eq. (14) (simplified) overestimates the SNCF compared to the
numerical model and Ness’s test (7 % to 35 %).
Eltaher et al. (2011) N/A Compression is higher than Tensile for both displacement and Displacement controlled, 10 %–15 %
load controlled. higher than the analytical model.
Load controlled, good match with the
analytical model.

10
M. Elgazzar et al. Journal of Pipeline Science and Engineering 4 (2024) 100196

Fig. 17. Comparison between the resulting SNCF from FEA and semi-analytical
Fig. 16. Resulting SCF from parametric studies by Ness and Verley (1995): (a) by Eltaher et al. (2011): (a) displacement-controlled; (b) load-controlled.
varying concrete thickness on a 20 inches pipe; (b) varying concrete strength
for 60 mm concrete thickness. the overbend and the sag bend, compared to 0.2 %/0.25 % (Static/dy-
namic) for the over bend and 0.18 % / 0.25 % (static/dynamic) for the
is used, the analytically calculated SNCF is found to be aligning with the sag bend proposed by DNV (1981) (Table 2). The proposed relaxation
SNCF calculated at the tension side of the pipe, with an overall values criteria permit this installation of heavy pipelines (due to on-bottom
around 10 %− 15 % less than the average between the compression and stability purposes) without the requirement of undertaking modifica-
tension sides of the numerical calculations. Results also shows an tions to the installation barge (e.g., increasing the lay tension capacity,
agreement betweeen the analytically calculated SNCF and the average or modifying the stinger), thus significantly, reducing the cost of pipe-
SNCF calculated by the 3D model when load-controlled criteria is fol- line installation. While the modified criteria demonstrated the structural
lowed, indicating more accurate outcome. integrity of the pipeline from the local buckling and collapse perspec-
In summary, authors utilised analytical method in predicting the tive, the paper does not provide evidence for the impact of the modified
strain concentration factors due to feasibility of applying the theoretical criteria on pipeline fatigue.
concepts related to pipe bending, in addition, the economy of con- Endal (1994) performed a numerical model using FEA software
structing analytical model compared to other methods of calculations. DIANA (1992) (Fig. 18), followed by parametric studies.The study
On the other hand, results are found to be limited to steel yield and aimed to investigate the bending stiffness of the discontinuously
concrete crushing limit (about 0.2 % strain). Moreover, while insignif- concrete-coated offshore pipelines and, hence calculate the SNCF. The
icant, neglecting the concrete tensile strength has shown inaccurate model considered shell elements for the pipe and coating and accounted
results at the low bending moments, it is also noted that analytical for the tensile capacity of the concrete coating, but ignored the peak
models indicated in the literature do not account for the effect of the bending moment due to the point load at stinger rollers.
concrete coating sliding as well as the internal pressure.

Table 2
2.3. Estimating the SNCF using the numerical approach
Allowable strains relaxation criteria by Sriskandarajah and Mahendran (1992).

This section introduces research carried out to calculate the SNCF for Location Load case Sriskandarajah and Mahendran (1992) DNV (1981)

concrete-coated pipelines using numerical approaches. Sriskandarajah Overbend Static 0.225 % 0.20 %
and Mahendran (1992) suggested allowing for higher steel strain during Dynamic 0.275 % 0.25 %
Sagbend Static 0.225 % 0.18 %
pipeline installation opposed to the criteria indicated in DNV (1981); the
Dynamic 0.275 % 0.25 %
proposed values are 0.225 % / 0.275 % strains (static/dynamic) for both

11
M. Elgazzar et al. Journal of Pipeline Science and Engineering 4 (2024) 100196

Fig. 18. Pure bending FE model by Endal (1994), using FE Software DIANA (1992).

As shown in Fig. 14, while the resulting SNCF was found in a good Where CT represents concrete coating (mm). D0 /t represents the ratio of
agreement with the analytical model performed by Ness and Verley diameter to wall thickness. εn represents global strain. τ represents
(1995), the latter shows less SNCF at very low and very high strains. This corrosion coating shear strength (MPa).
discrepancy is due two factors: Ness’s analytical model does not account The recommended practice DNV-RP-F105 (2021) introduces a
for concrete tensile capacity, whereas Endal’s numerical model accounts design equation to calculate the SNCF for the concrete-coated pipelines
for concrete tensile capacity, but does not consider concrete crushing. with either polypropylene or asphalt enamel corrosion coating, in
Lund’s analytical model deviates from the other two models since the contrast to the numerical, analytical, and test approaches, the equation
model assumes an equivalent moment curvature curve for the pipe. This introduces a constant value of the SNCF despite the strain level of the
dimensions are chosen based on the stiffness of the concrete coated pipe:
composite section when the concrete strain reaches 0.2 %. Using a 3D
model (ABAQUS, 2011), Endal et al. (1995) demonstrated that the sag SNCF = 1 + CSF (8)
bend roll phenomenon happens when a pipeline over-bend experiences ( )0.75
plastic deformation. Moreover, the sag bend roll is found to be depen- EIc
CSF = kc (9)
dent on several factors such as water depth, pipe tension, pipe weight, EIs
and the bending stiffness of the pipeline. Although the study did not
focus on the concrete coating behaviour during installation, it is sug- Where CSF represents the concrete stiffness factor. EIc represents the
gested that the pipeline stiffness contributes to the sag bend rolling concrete coating bending stiffness. EIs represents the steel bending
phenomena. Igland and Moan (2000) introduced a numerical model stiffness.
using the FEA tool (DIANA, 1992). However, Igland considered a con- The empirical parameter kc accounts for the deformation/slippage in
stant SNCF post-concrete crushing similar to the analytical approach the corrosion coating and the cracking of the concrete coating. The
that Ness and Verley (1995) introduced. Igland compared results of the design practice recommends a value of 0.33 of kc for asphalt coating and
numerical model with the full-scale tests performed by Ness and Verley 0.25 for PP/PE coating.
(1995), carried out parametric studies, and eventually introduced the As shown in Fig. 19, when compared to Ness’s analytical results
SNCF equation as follows: (Ness and Verley, 1995), the SNCF calculated by DNV-RP-F105 (2021) is
overpredicted prior concrete crushing (about 6 %), and underpredicted
εmax
fj post concrete crushing (about 7 %), while it generally agrees with
SNCF = (3)
εn full-scale test results (Ness and Verley, 1995). It should be noted that the
( )( ) full-scale test is performed on a limited range of pipe and coating
εn
SNCF = a− b + c (4) properties, therefore the proposed equation is required to be validated
0.0015
against a wider range of pipe sizes, material grades and coating prop-
( )
( ) D0 erties. Similar to the DNV approach (DNV-RP-F105, 2021), Bai (2001)
a = 1 + 0.0031CT1.3 0.687 + 0.011 (0.934 + 0.220τ) (5) has considered a constant value of 1.2 for the SNCF for concrete-coated
t
pipelines, referencing the analytical and full-scale tests conducted by
b = 0.138CT0.6 τ0.2 (6) Ness and Verley (1995), which shows an underprediction of approxi-
mately 15 %. The proposed value is associated with an allowable strain
( )
D0 0.3 of 0.4 % obtained from the fracture criterion and the technical infor-
c = 0.144CT0.5 0.75 + 0.009 τ (7)
t mation by Ness and Verley (1996). It is evident that a full-scale test by
Ness and Verley (1995) has resulted in a value of approximately 1.4 for a

12
M. Elgazzar et al. Journal of Pipeline Science and Engineering 4 (2024) 100196

Fig. 19. Comparison of the resulting SNCF from Analytical model (Ness and Verley, 1995), Full Scale Test (Ness and Verley, 1995), and DNV-RP-F105 (2021).

Fig. 20. Typical FE model by Nourpanah and Taheri (2009a).

particular pipe property, therefore the SNCF proposed by Bai should be

subjected to further validation. The journey of estimating the SNCF for
the concrete-coated pipes was continued by Nourpanah and Taheri
(2009a), who introduced a numerical model using FEA tool (ABAQUS,
2011). The developed FEA model consisted of 8-node reduced integra-
tion continuum elements (C3D3R), with finer meshing near the FJ. The
author used four layers of elements to model each of the concrete
coating thickness and steel pipe thickness, to fully capture their bending
stiffness. Both the steel and the corrosion coating materials were
modelled using the available plasticity consecutive material model,
while the concrete damaged plasticity model is used to model the con-
crete coating. Due to symmetry, the author modelled only one-quarter of
a 12 m pipe joint, with consideration of utilising appropriate Z-plane and
X-plane boundary conditions as shown in Fig. 20.
The author validated his approach by comparing the results of the
FEA model to the results obtained by Ness and Verley (1995) from the
full-scale test. Results indicated in Nourpanah and Taheri (2009a) and Fig. 21. Comparison of the resulting SNCF at different global strain levels as
Fig. 21 show an agreement between both results. The FEA model is then depicted by Nourpanah and Taheri (2009a), Design Equation (Nourpanah and
used in performing parametric studies that are published later from Taheri, 2009b), and Ness Test (Ness and Verley, 1995).

13
M. Elgazzar et al. Journal of Pipeline Science and Engineering 4 (2024) 100196

Nourpanah and Taheri (2012). The author introduced theoretical

equations in Nourpanah and Taheri (2009b). The equations are intended
to calculate the SNCF for X65 concrete-coated pipelines under bending,
the introduced equations are as follows:
( )C
̂ = β1 + β2 tc
SCF (10)
ts

tc β εg
C = β3 + β4 + β5 fʹc + 6 + β7 (11)
D τy 0.0015

Where tc is the concrete coating thicknessts and ts is the pipe steel wall
thickness. D is the pipeline diameter. fʹc refers to the compressive
strength of the concrete. τy refers to the shear strength of the Anti Fig. 23. Conceptual sketch and FE modelling of CFT columns (Hu et al., 2010).
Corrosion Laye, β1 to β7 are coefficients that can be extracted from the
referenced paper Nourpanah and Taheri (2009b). SNCF at both sides of the pipe (compression and tension) considering the
The following simplified equation were also: load-controlled and the displacement-controlled criterion. Results were
compared to the semi-analytical model created within the same study
7 (Fig. 17), as indicated previously, the average SNCF shown by the FEA
SCFUL = 1.35 + (12)
λ1.7 demonstrates good agreement to the analytical model for the load
controlled criteria, while it shows a 10 %–15 % increase for the
Dts
λ= (13) displacement-controlled criteria. The 3D FEA was also used to verify the
εg tc2
accuracy of a simpler elastic-plastic concrete material model in pre-
Fig. 21 shows a comparison between the SNCF calculated using the dicting SNCF. However, the elastic-plastic concrete model did not pre-
equations proposed by Nourpanah and Taheri (2009b) and the SNCF sent accurate results compared to the response of the concrete softening
value measured by Ness and Verley (1995). The figure shows a good plasticity model. In 2012,Eltaher carried out an FEA analysis to inves-
agreement of SNCF resulting from FEA and the test outcome. The tigate the effect of sliding of the concrete coating on the SNCF for pipes
resulting SNCF using Eq. (8) as referred in Fig. 21 gives comparable subjected to bending moment (Eltaher et al., 2012). The analysis
results to the testing results (within 6 %− 11 %), while the SNCF comprised an FEA model with a defined limited shear capacity between
resulting from the simplified Eq. (12) as referred in Fig. 21 shows less the concrete coating and the corrosion coating (3LPE), and another
accurate values (within 7 %− 35 %) model that does not allow concrete slippage (a very high shear stiffness).
Hu et al. (2010) has studied the effect of a confined concrete core of The author presented the outcome of the comparison based on the
steel tube casing when the section is subjected to pure bending moment. load-controlled and displacement-controlled approaches. As shown in
Hu used ABAQUS for the FEA model and compared the results to the Figs. 24 and 25, the limited capacity of the interface facilitated the
experiment carried out by Elchalakani et al. (2001) as shown in Fig. 22. concrete coating slippage over the corrosion coating layer, resulting in a
Due to symmetry, Hu modelled only a quarter of the pipe and significant reduction in the SNCF when considering the
enforced a symmetry boundary condition on the symmetric planes as displacement-controlled case. However, a minimal reduction is found in
shown in Fig. 23. The model considers U3=0 on the left surface while the load-controlled case. It was also noted that while there was no sig-
U1=0 on the right surface. To simulate the wheel support, the nificant reduction in the maximum strain when reducing the shear from
displacement U2 for the nodes at the mid-depth of the right edge of the 0.5 MPa to 0.1 MPa, the damage spreads faster when weaker capacity is
element mesh are all set to zero. Both the steel pipe and concrete core are used.
modelled using solid elements with reduced integration rule, with a pair Turnbull and Crocker (2014) conducted a computational FEA by
of contact surfaces representing the interface between the two materials application of four bending tests on curved specimens, and the results of
with a consideration of 0.25 friction coefficient. the stress and the strain across the specimen have been compared to
Results show that the behaviour of the concrete-filled tubes (CFT) is previous FEA work performed on flat specimens Results indicate that the
similar to the empty steel tubes for d/t< 20. An FEA model (ABAQUS, curved specimens experienced relatively higher stresses and strain by
2011) is introduced by Eltaher et al. (2011) to verify the semi-analytical moving to both edges across the specimen when the bending moment is
model created within the same study, the 3D FEA model calculated the within the elastic range. When the specimen enters the plastic region,
the strain distribution follows the same trend as elastic. However, the
largest stresses are experienced midway between the mid and edge of
the specimen, reflecting stress relaxation and redistribution after
yielding. The contribution of the pipeline stiffness and the wave crest
length on the overall on-bottom stability of the pipeline was also studied
by Robertson et al. (2015). Observations of dynamic stability analysis on
pipelines subject to the same metocean conditions show that the pipeline
stiffness contributes to the lateral response of the pipeline on the seabed.
Results also indicated that the increased bending stiffness and axial
stiffness significantly reduce the lateral stability for quasi-static pipes
(<0.5 D) and dynamically stable pipes (>2 D) respectively. The effect of
added stiffness due to concrete coating will lead to a better estimation of
pipeline lateral displacement. Ravirala and Jukes (2017) introduced an
investigation into Auxetic composite materials for offshore and onshore
pipeline applications. The study introduced results of FEA performed on
Fig. 22. Experimental set-up for CFT column subjected to pure bending Auxetic material on a pipe form, the results are then compared to carbon
moment (Elchalakani et al., 2001).

14
M. Elgazzar et al. Journal of Pipeline Science and Engineering 4 (2024) 100196

Fig. 24. Strain distribution for interface strength 0.5 MPa—load controlled (Eltaher et al., 2012).

steel pipe of the same size and thickness. FEA results have shown a (2011, 2012).
promising effect of the negative Poisson’s ratio of the Auxetic materials Utilising FE analysis is considered an efficient method in calculating
as a result produces significantly low stresses on pipe material when the SNCF for the concrete coated pipelines producing comparable results
subject to axial, internal, and external loads. The study recommends to full-scale bending tests. Results confirmed the ability of FEA to cap-
further investigations into these composites including full-scale tests. ture the behaviour of the concrete tensile strength and the concrete
Battistini et al. (2017) performed 3D FEA using ABAQUS to assess the coating sliding over the corrosion coating. The FEA also has proven
pipeline stiffness under bending moment at plastic deformation. Results capability in capturing the steel non-linearity, post concrete crushing
were presented for analysis performed using ABAQUS standard, and a behaviour and different case loadings (i.e., inclusion of the internal
quasi-static analysis using ABAQUS Explicit. A simplified concrete pressure). Nevertheless, FEA requires careful consideration specially
model has been utilized for the standard analysis, while a when selecting the suitable meshing criteria, boundary conditions,
concrete-damaged model was utilized for the explicit analysis. The suitable material properties, interface between the concrete coating and
explicit analysis with a damaged concrete model has shown less onerous the corrosion coating, and benchmarking the results with the full-scale
pipeline strain at the field joint compared to the standard concrete tests is also crucial. Additionally, the cost associated with FE software,
model, this concludes that a damaged concrete model is more suitable in required skills, and processing time should be taken into consideration.
modelling concrete during pipe bending as indicated by Eltaher et al.

15
M. Elgazzar et al. Journal of Pipeline Science and Engineering 4 (2024) 100196

Fig. 25. Strain distribution for interface strength 0.5 MPa—displacement controlled (Eltaher et al., 2012).

3. Conclusion and further work cover a broad range of pipe and coating properties, strengths, and ability
to model various cases of loading. However, the numerical method re-
In this article, the author carried out a comprehensive literature re- quires careful selection of the meshing criteria, boundary conditions,
view showing work and literature conducted to calculate the strain material properties, concrete coating to corrosion coating, interface, and
concentration factor for the concrete-coated pipelines. The literature benchmarking the results with the full-scale tests. Additionally, the
review presents methods that are mostly used in calculating the stiffness associated cost of using the suitable software and producing reliable re-
of the concrete coated pipes and the strain concentration factor in the sults shall be taken into consideration. The available literature does not
concrete coated pipes. The methods presented in the article are full-scale cover a wide range of pipe and coating properties that are widely used in
tests, analytical models, and numerical models. The values of the SNCF the industry. Also, majority of the published literature is focused on
resulting from the full-scale tests are found to be the most reliable. calculating the strain concentration factor during pipeline installation (i.
However, this method is not able to cover a broad range of pipe prop- e., zero internal pressure). Even though this is considered one of the main
erties (i.e., pipe outside diameter and wall thickness, concrete coating aspects in subsea pipeline design, the strain concentration factor for
strength and thickness) due to cost impact associate with the pipe ma- subsea pipelines under operational conditions is deemed equally impor-
terial, coating material, test setup and equipment. Therefore, the authors tant. The likeliness of pipeline bending is considered significant while in
recommend applicability to perform a full-scale test for projects which the operational phase (i.e., lateral displacement due to hydrodynamic
require high safety considerations. The analytical method is incapable of effects or global buckling). For concrete coated pipelines prone to global
capturing the complexities associated with modelling the concrete buckling, the stress ranges should also consider the stress concentrations
coating sliding and the behaviour of the composite section beyond the due to discontinuity of the concrete coating in the field joint. Although
concrete crushing and steel yield, which are shown to be significantly the studies show a significant reduction in the strain concentration
affecting the results. Therefore,the analytical method is recommended beyond steel yield and concrete compression crushing, further investi-
until concrete compression crashing or steel yield, additionally, the gation using wider range of pipe properties is recommended to confirm
analytical method may be adopted where high safety precautions are not pipeline integrity. Further work is currently under progress through
deemed necessary. The author foresees using the numerical approach in creating mathematical models using a newly developed numerical
determining the strain concentration factor due to ability to modelling approach and parametric studies to obtain more accurate predictions of
the main contributors to the strain concentration, moreover, the ability to the strain concentration factor for the concrete-coated pipelines.

16
M. Elgazzar et al. Journal of Pipeline Science and Engineering 4 (2024) 100196

CRediT authorship contribution statement Endal, G., 1994. Extreme bending of concrete coated offshore pipelines: a numerical
study. International DIANA Conference on Computational Mechanics. Delft,
Netherlands.
Mohamed Elgazzar: Writing – review & editing, Writing – original Endal, G., Ness, O.B., Verley, R., Holthe, K., Remseth, S., 1995. The behaviour of offshore
draft, Visualization, Validation, Supervision, Software, Resources, pipelines subjected to residual curvature during laying. OMAE.
Project administration, Methodology, Investigation, Funding acquisi- Hu, H.T., Su, F.C., Elchalakani, M., 2010. Finite element analysis of CFT columns
subjected to pure bending moment. Steel Compos. Struct. 415–428.
tion, Formal analysis, Data curation, Conceptualization. Ahmed Reda: Igland, R.T., Moan, T., 1993. Reliability Analysis of Deep-Water Pipelines During Laying,
Writing – review & editing, Writing – original draft, Visualization, For Combined Pressure, Tension and Bending Loads. International Offshore and
Validation, Supervision, Software, Resources, Project administration, Polar Engineerin. International Society of Offshore and Polar Engineers, Singapore,
pp. 613–621.
Methodology, Investigation, Funding acquisition, Formal analysis, Data Igland, R.T., Moan, T., 2000. Reliability analysis of pipelines during laying,
curation, Conceptualization. Ibrahim Sultan: Writing – review & edit- considering ultimate strength under combined loads. Offshore Mech. Arctic. Eng.
ing, Writing – original draft, Visualization, Validation, Supervision, 40–46.
Livingston, L.E., 1965. Flexural Rigidity of Concrete-Coated Steel Pipe. Rice University,
Software, Resources, Project administration, Methodology, Investiga- Houston.
tion, Funding acquisition, Formal analysis, Data curation, Conceptuali- Lund, S., Bruschi, R., Montesi, M., Sintini, L., 1993. Laying criteria versus strain
zation. Truong Phung: Writing – review & editing, Writing – original concentration at field joints for heavily coated pipelines. Proceedings of the
International Conference on Offshore Mechanics and Arctic Engineering 41–56.
draft, Visualization, Validation, Supervision, Formal analysis, Data Mogbo, N.C., Jirsa, J.O., Wilhoit, J.C., 1971. Effective stiffness of concrete coated line
curation, Conceptualization. pipe. Proceedings of the Petroleum Mechanical Engineering Conference. Houston:
ASME Paper 7l-Pet-26.
Ness, O.B., Verley, R., 1995. Strain concentrations in pipelines with concrete coating: an
Declaration of competing interest
analytical model. Offshore Mechanics and Arctic Engineering – OMAE. Copenhagen,
Denmark, pp. 507–512.
The authors declare that they have no known competing financial Ness, O.B., Verley, R., 1996. Strain concentrations in pipeline with concrete coating.
interests or personal relationships that could have appeared to influence J. Offshore Mech. Arctic. Eng. 118/225, 118–225.
Nourpanah, N., Taheri, F, 2009a. Finite element analysis of strain concentration in field
the work reported in this paper. joint of concrete coated pipelines. International Conference on Offshore Mechanics
and Arctic Engineering. Honolulu, Hawaii.
References Nourpanah, N., Taheri, F., 2009b. A design equation for evaluation of strain
concentration factor in concrete coated X65 pipelines. J. Marine Struct. 22,
758–769.
ABAQUS, 2011. Version 6.11 documentation. SIMULIA. Nourpanah, N., Taheri, F., 2012. A Comprehensive parametric finite element study on
Archer, G.L., Adams, A.J., 1983. The behaviour of concrete over thin film epoxy coatings the development of strain concentration in concrete coated offshore pipelines.
on offshore pipelines. Offshore Technology Conference, OTC Paper No 4453, 85–89. J. Press. Vessel Technol.
Akten, H.T., Lund, S., Miller, D.M., 1985. “On the design and construction of Statpipe Ravirala, N., Jukes, P., 2017. Advanced composite structures for subsea pipeline design
pipeline system” OTC paper no. 4922. Annual Offshore Technology Conference. development. International Society of Offshore and Polar Engineers (ISOPE), San
Houston, USA. Francisco, CA, USA, pp. 53–60.
Bai, Y., Bai, Q., 2001. Pipelines and risers Elsevier ocean engineering book series. Reda, A., Sultan, I.A., Howard, I.M., Forbes, G.L., McKee, K.K., 2018. Pipeline walking
Battistini, A., Catena, L., Mancini, A., Marchionni, L., Parrella, A., Vitali, L., 2017. and anchoring considerations in the presence of riser motion and inclined seabed.
Further advances on concrete coating impact on pipeline strength. OMAE. Int. J. Press. Vessels Piping 71–85.
Trondheim. ASME, Norway. Reda, A., McKee, K.K., Howard, I.M., Sultan, I.A., 2019. When is a subsea anchor
Bruschi, R., Malacari, L.E., Torselletti, E., Vitali, L., 1995. Concrete coated submarine required for a short pipeline/SCR system? Int. J. Press. Vessels Piping 278–298.
pipelines: further advances in strain concentration at field joint and relevant implications Reda, A., Rawlinson, A., Sultan, I.A., Elgazzar, M.A., Howard, I.M., 2020. Guidelines for
on strain based design. Offshore Technology Conference, pp. 491–501. Italy. safe cable crossing over a pipeline. Appl. Ocean Res.
DIANA, 1992. User’s Manual 5.0, TNO Building and Construction Research. Department Reda, A., Thiedeman, J., Elgazzar, M.A., Shahin, M.A., Sultan, I.A., McKee, K.K., 2021a.
of Computational Mechanics, Delft. Design of subsea cables/umbilicals for in-service abrasion-part 1: case studies. Ocean
DNV, 1981. Rules For Submarine Pipeline Systems. DNV, Norway. Engineering.
DNV-ST-F101, 2021. Submarine Pipeline Systems. DNV, Norway. Reda, A., Elgazzar, M.A., Sultan, I.A., Shahin, M.A., McKee, K.K., 2021b. Failure analysis
DNV-RP-F105, 2021. Free Spanning Pipelines. DNV, Norway. of articulated paddings at crossing interface between crossing cable and crossed
DNV-RP-F109, 2021. On-Bottom Stability Design of Submarine Pipelines. DNV, Norway. pipeline. Appl. Ocean Res.
Elchalakani, M., Zhao, X.L., Grzebieta, R.H., 2001. Concrete-filled circular steel tube Robertson, M., Terry, G., Graham, V., Stuart, O., Peter, M., Aiman, A.S., et al., 2015. The
subjected to pure bending. J. Constr. Steel. Res. 1141–1168. influence of pipeline bending stiffness on 3D dynamic on-bottom stability and
Elchalakani, M., Zhao, X.L., Grzebieta, R., 2016. Plastic slenderness limit for cold-formed importance for flexible flowlines, cables and umbilicals. In: International Conference
circular hollow sections plastic bending tests of tubular steel members. Aust. J. on Offshore Mechanics and Arctic Engineering, 56512.
Struct. Eng. Sriskandarajah, T., Mahendran, I.K., 1992. Critique of offshore pipelay criteria and its
Eltaher, A., Al-Showaiter, A., Sourabh, A., Banneyake, R., Jules, P., 2011. Strain effect on pipeline design. Offshore Technology Conference OTC 533–542.
intensification in concrete coated pipelines under S-lay conditions a 3D FEA Sultan, I.A., Reda, A.M., Forbes, G.L., 2013. Evaluation of slug flow-induced flexural
investigation. Proceedings of Rio Pipeline Conference & Exposition. Rio de Janeiro, loading in pipelines using a surrogate model. J. Offshore Mech. Arctic Eng.
Brazil. Turnbull, A., Crocker, L.E., 2014. Four-point Bend Testing – Finite element Analysis of
Eltaher, A., Al-Showaiter, A., Jukes, P., Abdalla, B., 2012. Effect of modelling of concrete the Stress and Strain Distribution Accounting For Lateral Specimen Curvature.
weight coating on assessment of strain concentration in pipeline field joints. In: Teddington, Middlesex. NPL Report MAT 66.
International Offshore and Polar Engineering Conference. International Society of ZEE Engineering Sdn. Bhd., 2014. Labuan pipeline installation engineering.
Offshore and Polar Engineers (ISOPE, Greece, pp. 519–528.

17