See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/321504574

ANALYSIS OF DESIGN METHODOLOGIES OF SUBSEA STRUCTURE

FOUNDATIONS

Conference Paper · January 2017

DOI: 10.20906/CPS/CILAMCE2017-0078

CITATIONS READS
0 262

3 authors, including:

João Paulo Dandoulakis Paulo Santa Maria

Universidade Federal Fluminense COPPE UFRJ
3 PUBLICATIONS   4 CITATIONS    7 PUBLICATIONS   7 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Behaviour of shallow foundations on clayey seabed View project

All content following this page was uploaded by Paulo Santa Maria on 22 May 2019.

The user has requested enhancement of the downloaded file.

 ANALYSIS OF DESIGN METHODOLOGIES OF SUBSEA
STRUCTURE FOUNDATIONS

João Paulo Emmanuel Godinho Dandoulakis

jgodinho@id.uff.br
Fluminense Federal University
Rua Passos da Pátria, 156, Niterói, 24210-240, R.J., Brazil
Paulo Eduardo Lima de Santa Maria
paulosm@smengenheiros.com.br
Rua Fernando Nogueira de Sousa, 183/101, Rio de Janeiro, 22620-380, R.J., Brazil
Elson Antônio do Nascimento
elsonn@vm.uff.br
Fluminense Federal University
Rua Passos da Pátria, 156, Niterói, 24210-240, R.J., Brazil

Abstract.
The aim of this paper is to show that analytical methods tend to yield conservative
results for geotechnical mudmat design. This is consistent with the structure of these methods,
which are based on exact solutions obtained from the Theory of Plasticity to simplified basic
cases of geometry and loading conditions. In order to adapt this basic configuration to real
cases, a few corrections factors need to be applied. As should be expected, the cumulative
effect of these corrections produces a final conservative result. On the other hand, the finite
element methods are able to model almost any situation/problem without requiring correction
factors. Since they are based on equations of Elasticity and Plasticity Theories, finite element
methods can produce highly accurate solutions to highly complex geometry, material
behavior and boundary problems. This paper suggests how the analytical and numerical
methodologies can be used to bridge this gap and ensure the effective design of shallow
foundation of subsea structures. The comparison of the finite element case results, performed
with the commercial finite element program, with the analytical methodology, recommended
by the API RP 2GEO, has shown that the prediction of the safety factors and soil short- and
long-term settlements are in good agreement and of the same order of magnitude; however,
the former methodology provides results closer to the real physical values.
Keywords: shallow foundations, subsea manifold, finite element analysis, offshore
engineering.

CILAMCE 2017
Proceedings of the XXXVIII Iberian Latin-American Congress on Computational Methods in Engineering
P.O. Faria, R.H. Lopez, L.F.F. Miguel, W.J.S. Gomes, M. Noronha (Editores), ABMEC, Florianópolis, SC,
Brazil, November 5-8, 2017.
Analysis of design methodologies of subsea structure foundations

INTRODUCTION
The search for oil has extended the frontiers of prospecting and exploration to ever
deeper waters. The accelerated pace in this direction requires compatible technological
development for the construction of today's complex subsea systems. In the current offshore
engineering scenario, the area of structural engineering has sought to design in an increasingly
sophisticated way. Specific computational tools for geotechnical analysis are becoming
increasingly more accessible and powerful. As a result, they started to have a key role in the
day-to-day of the engineering professionals who look for greater efficiency in the execution of
their projects; however, the bodies that define the standards that regulate the use and adoption
of new upcoming technologies in this area hardly keep pace with the market, becoming
outdated as a result. Emphasis has been placed on analytical methods that were initially
developed for very simple mechanical and geometric configurations.
The design of foundations is one of the most common and enduring problems in
geotechnical engineering. Shallow foundation design utilizes rational design methods, based
on Soil Mechanics principles established over 60 years ago. The analytical solutions utilized
contain idealizations, either of the problem geometry or of the soil response. They still
provide, however, a framework that links the outcome to the various input parameters. The
bearing capacity theory relates the vertical, horizontal, and moment loads with the foundation
footprint and soil properties. The shear strength of a soil can be defined as the maximum
shear stress that the soil can withstand without failing or the shear stress in soil in the plane
that the failure is occurring (Pinto, 1998).
More recently, general evaluation of bearing capacity of shallow foundations has been
addressed with both analytical and numerical methods (e.g. Bransby and Randolph, 1998;
Randolph and Puzrin, 2003; Gourvenec and Randolph, 2003). The interaction of complex
loads is still largely undefined though. Design guidelines such as API (API RP 2 GEO
(2014)), ISO (ISO 19901-4 (2016)) and DNV (DNVGL-OS-C-101 (2016)) have been slow to
evolve and in many places rely on field proven, but somewhat dated recommendations, either
empirical or quasi-analytical design methodologies.
In an era where virtually any geotechnical application can be modeled numerically, it
is tempting to wonder whether true analytical solutions still have a role (Randolph, 2013). The
fact is that although the analytical and numerical approaches come from different theoretical
backgrounds, there is nothing that prevents their combination. Actually, the inherent strengths
and limitations of each methodology suit them to complement one another.
Our aim in this paper is to show that analytical methods tend to be conservative when
applied to geotechnical mudmat design. This is consistent with the structure of these methods,
which are based on exact solutions obtained from the Theory of Plasticity (i.e. the basic case
of a rigid strip footing on a horizontal surface of a weightless soil, subjected to a central
vertical load). In order to adapt this basic configuration to real cases, a few correction factors
need to be applied. As should be expected, the cumulative effect of these corrections produces
a final conservative result. On the other hand, the finite element methods are able to model
almost any situation/problem without requiring correction factors. Since they are based on
equations of Elasticity and Plasticity Theories, finite element methods can produce highly
accurate solutions to highly complex geometry, material behavior and boundary problems.

CILAMCE 2017
Proceedings of the XXXVIII Iberian Latin-American Congress on Computational Methods in Engineering
P.O. Faria, R.H. Lopez, L.F.F. Miguel, W.J.S. Gomes, M. Noronha (Editores), ABMEC, Florianópolis, SC,
Brazil, November 5-8, 2017.
 Dandoulakis, J.P., Santa Maria, P.E., Nascimento, E.A.

1 SUBSEA MANIFOLD WITH SKIRTED MUDMAT FOUNDATIONS

Subsea systems are composed of heavy pieces of equipment (e.g. manifolds, UTAs,
PLEMs, PLETs, etc.) that are installed on the seabed. These pieces of equipment are
subjected to forces, moments, and torsions that sometimes exceed the bearing capacity of the
soil beneath them. Mudmat is a type of shallow foundation, which is composed of thin
reinforced steel plates installed on the base of the subsea structures. They “enlarge” the
equipment footprint in contact with the soil, which provides additional bearing support by
redistributing the stresses to the soil, also reducing the settlements to tolerable values.
Skirted mudmat is another type of shallow foundation, in which the structure loads are
transferred to deeper soil layers by vertical slender elements that confine the soil beneath the
foundation footprint (skirts). They also increase the foundation bearing capacity to withstand
horizontal loads (Fig. 1).

Figure 1. Skirted mudmat (Aker Solutions, 2016)

2 SHALLOW FOUNDATION DESIGN

Despite the huge volume of studies performed, many oil industry regulating bodies
still recommend the traditional field-proven design methods based on Terzaghi’s theory
proposed in 1943. Since then, several researchers have made improvements to adapt his
theory to actual day-to-day loading situations (Fig. 2).

CILAMCE 2017
Proceedings of the XXXVIII Iberian Latin-American Congress on Computational Methods in Engineering
P.O. Faria, R.H. Lopez, L.F.F. Miguel, W.J.S. Gomes, M. Noronha (Editores), ABMEC, Florianópolis, SC,
Brazil, November 5-8, 2017.
Analysis of design methodologies of subsea structure foundations

1940
Keverling Buisman 1951 1963 1970 1975 1997
1921
included the Meyerhof
Ludwig Prandtl published Meyerhof Brinch Hansen Chen Michalowski
a solution to load capacity soil weight, γ

1969 - 1977 1979 - 2000 2011 - 2017

1a - 9a API RP 2A 10a - 21a API RP 2A API RP 2GEO

1924 1943 1953 1973 2005

Reissner included Terzaghi γ Meyerhof Vesic Zhu & Michalowski
a surrounding surcharge, q

Figure 2. Soil Mechanic development timeline

The Analytical methods were initially developed for very simple mechanical and
geometric configurations. The actual cases of foundations, for the most part, depart
considerably from those configurations, requiring the introduction of various correction
factors. These factors, in turn, introduce errors that will cumulatively affect the accuracy of
the results.
Several oil and gas industry normative documents, DNV CN 30.4 (1992), DNVGL-
OS-C101 (2016), ISO 19901-4 (2016), API RP 2A-WSD (2014) and more recently, API RP
2GEO (2014), recommend expressions and methodologies to calculate the bearing capacity of
shallow foundations; however, more advanced computational tools for geotechnical design
are only superficially described in those documents. The lack of more comprehensive
regulation on the utilization of advanced numerical methodologies goes against the current
trend, which shows that computational tools started to have a key role in the day-to-day of the
engineering professionals, who look for greater efficiency in the execution of their projects.
All in all, the design of shallow foundations can be carried out by using two different
methodologies (Fig. 3).

Design
Methodology

Analitical Numerical

API RP 2 DNVGL- ISO PLAXIS 2

API RP 2A ABAQUS ANSYS DIANA
GEO OS-C101 19901-4 e 3D

Figure 3. Shallow foundation design methodologies

As of 2011, API RP 2 GEO, Geotechnical and Foundation Design Considerations, has

become the API reference on shallow foundation design, superseding API RP 2A in this
regard.

CILAMCE 2017
Proceedings of the XXXVIII Iberian Latin-American Congress on Computational Methods in Engineering
P.O. Faria, R.H. Lopez, L.F.F. Miguel, W.J.S. Gomes, M. Noronha (Editores), ABMEC, Florianópolis, SC,
Brazil, November 5-8, 2017.
 Dandoulakis, J.P., Santa Maria, P.E., Nascimento, E.A.

2.1 Analysis Procedure

The analysis of the manifold foundation was split into 15 cases (Tab. 1). The initial
case corresponds to its self-weight as well as the all MCV cap weights. The first 12 cases
address the retrieval of the individual MCV caps and the subsequent installation of the
respective MCV and jumper. The installation of the MTUs does not require hub cap removals.
The last 3 cases address the removal of each of one of the modules (i.e. choke and crossover)
from the manifold. This is necessary to simulate maintenance activities during the project
lifespan.

Table 1. Analysis Cases

Analytical Numerical Loading

Initial 1 Self-weight and caps
7 3 Self-weight + MCV #7
6 5 Self-weight + MCV #7 to #6
3 7 Self-weight + MCV #7 to #3
4 9 Self-weight + MCV #7 to #4
2 11 Self-weight + MCV #7 to #2
9 13 Self-weight + MCV #7 to #9
8 15 Self-weight + MCV #7 to #8
11 17 Self-weight + MCV #7 to #11
14 19 Self-weight + MCV #7 to #14
10 21 Self-weight + MCV #7 to #10
13 23 Self-weight + MCV #7 to #13
1 25 Self-weight + MCV #7 to #13 - #1
5 27 Self-weight + MCV #7 to #13 - #1 and #5
12 29 Self-weight + MCV #7 to #13 - #1, #5 and #12

The case sequence above has been defined to unbalance the manifold by applying
eccentric loads, which create moments, and thus reduce its bearing capacity. Table 2 provides
the manifold main features.

CILAMCE 2017
Proceedings of the XXXVIII Iberian Latin-American Congress on Computational Methods in Engineering
P.O. Faria, R.H. Lopez, L.F.F. Miguel, W.J.S. Gomes, M. Noronha (Editores), ABMEC, Florianópolis, SC,
Brazil, November 5-8, 2017.
Analysis of design methodologies of subsea structure foundations

Table 2. Manifold features

Parameter Value

Weight in water (kN) 2151.2

Length (m) 16

Width (m) 13.5

Skirt length (m) 16

Skirt width (m) 8

Skirt height (m) 0.5

Figure 4. Skirted Manifold (Aker Solutions, 2016)

The resulting forces and moments due to the hoisting / installation of retrievable
components are reported in Table 3 below.

Table 3. Forces and moments applied to the manifold retrievable components

Case Pos. Force on local axes (kN) Moment on local axes

(kNm)

X Y Z X Y Z
Initial NA 12.9 0 -164.6 -23.6 123.1 -0.49
7 SB 23.6 0 -109.7 -29.9 226.8 -2.62
6 SB 23.6 0 -109.7 -29.9 226.8 -2.62
3 SB 0 0 -10.6 0 0 0
4 SB 0 0 -10.6 0 0 0
2 SB 12.9 0 -164.6 -23.6 123.1 -0.49
9 PS 0 0 -12.9 0 0 0
8 Bow 23.6 0 -109.7 -29.9 226.8 -2.62
11 PS 23.6 0 -109.7 -29.9 226.8 -2.62
14 PS 0 0 -10.6 0 0 0
10 PS 0 0 -10.6 0 0 0
13 PS 0 0 -200.8 0 0 0
1 Stern 0 0 -270.7 0 0 0
5 SB 0 0 -270.7 0 0 0

CILAMCE 2017
Proceedings of the XXXVIII Iberian Latin-American Congress on Computational Methods in Engineering
P.O. Faria, R.H. Lopez, L.F.F. Miguel, W.J.S. Gomes, M. Noronha (Editores), ABMEC, Florianópolis, SC,
Brazil, November 5-8, 2017.
 Dandoulakis, J.P., Santa Maria, P.E., Nascimento, E.A.

Case Pos. Force on local axes (kN) Moment on local axes

(kNm)

X Y Z X Y Z
12 PS 12.9 0 -164.6 -23.6 123.1 -0.49

(a) local axes (b) moment signs

Figure 5. Main retrievable component axes

3 ANALYTICAL SOLUTION

The Analytical calculation was performed in accordance with the API RP 2GEO
methodology. The soil parameters were obtained from Fagundes et al. (2012) paper. The soil
properties were inferred from the reconstituted soil sample properties utilized in his research
to obtain the soil parameters (Tab. 4).

Table 4. Soil Parameters (Inferred from Fagundes et al. (2012))

Parameter Nomenclature Value

Submerged unit weight of the soil γsat (kN/m3) 6.1
Undrained shear strength su (kPa) 𝑠𝑢(ℎ) = (2 + 1.5 × ℎ) ∗
Young’s modulus E (kPa) 4339
Elastic shear modulus G (kPa) 1446
Poisson’s ratio ν 0.5
Ground inclination angle β(ͦ) 0
Obs.: *h = depth of soil
CILAMCE 2017
Proceedings of the XXXVIII Iberian Latin-American Congress on Computational Methods in Engineering
P.O. Faria, R.H. Lopez, L.F.F. Miguel, W.J.S. Gomes, M. Noronha (Editores), ABMEC, Florianópolis, SC,
Brazil, November 5-8, 2017.
Analysis of design methodologies of subsea structure foundations

3.1 Results
The load capacity and safety factor calculations were performed according to the
following sequence of installation of MCVs and MTUs: 7, 6, 3, 4, 2, 9, 8, 11, 14, 10 and 13
(Tab. 4). The sequence has been defined so as to overload one manifold side (i.e. starboard)
and cause moments that will reduce the effectiveness of the foundation design. The purpose
of doing so is to check the foundation's settlements and safety factors in a completely
unfavorable installation and/or operating conditions. It should be noted that the eccentricities
due to the layout of the removable components in the manifold contributed to a small
reduction of the effective area. The results of soil bearing capacity calculations for undrained
loading are summarized below (Tab. 5):

Table 5. Analytical Results (API RP 2GEO)

Case A’ (m2) Qult + Qu (kN) SF uv (mm) uh (mm) θr (rad) θt (rad)

Initial 212.9 5062 2.35 22.4 0.0 0 0
7 197.1 4591 2.00 23.9 0.2 0 0
6 186.7 4280 1.78 25 0.4 0 0
3 182.4 4140 1.65 26.2 0.4 0 0
4 181.9 4189 1.66 26.3 0.0 0 0
2 182.0 4188 1.65 26.4 0.0 0 0
9 181.2 4172 1.56 27.8 0.2 0 0
8 180.6 4193 1.56 28 0.0 0 0
11 187.8 4345 1.56 29.1 0.4 0 0
14 199.3 4653 1.60 30.2 0.4 0 0
10 199.5 4723 1.62 30.4 0.0 0 0
13 199.4 4718 1.61 30.5 0.0 0 0
1 184.6 4370 1.58 28.9 0 0 0
5 179.9 4239 1.53 29 0 0 0
12 182.3 4319 1.64 27.4 0 0 0
Obs.: the order of magnitude of the manifold rotation angles was so low (i.e. 10 -4) that
they were considered zero.

The envelopes of the foundation bearing capacity for the highest and lowest safety
factor cases are shown below (Fig. 6).

CILAMCE 2017
Proceedings of the XXXVIII Iberian Latin-American Congress on Computational Methods in Engineering
P.O. Faria, R.H. Lopez, L.F.F. Miguel, W.J.S. Gomes, M. Noronha (Editores), ABMEC, Florianópolis, SC,
Brazil, November 5-8, 2017.
 Dandoulakis, J.P., Santa Maria, P.E., Nascimento, E.A.

(a) Initial case (b) Case #12

Figure 6. Envelope of bearing / sliding capacity

The long-term displacement result for drained loading condition, when pore water
pressure is in equilibrium with the in situ conditions, shows a settlement of 0.31 m over an
infinite time.
As can be seen in Table 5, the analytical calculation results obtained through the API
RP 2GEO methodology alone suggest that the manifold foundation was designed to a safety
factor of less than 2; however, this alone cannot be confirmed without the knowledge of the
actual soil properties at the exact location of the manifold installation and with the execution
of more refined analyses.

4 NUMERICAL SOLUTION

The mechanical behavior of the soil can be simulated with varying degrees of
accuracy. Over the years, numerous soil constitutive models have been developed for analysis
in EF (Lade, 2005); however, few of them have the capacity to reproduce the mechanical
behavior of the material over any three-dimensional stress trajectory. Each model has
different capacities and requires different input (experimental) data for its calibration. It is
essential to model the behavior as realistically as possible, for reasons of economy, time, and
safety. If the input parameters are sufficiently accurate, then the finite element model results
are reliable. Three-dimensional models are computationally intensive. Because of this, it is
important to select the simplest computational model that can accurately capture the behavior
of the soil in question.

CILAMCE 2017
Proceedings of the XXXVIII Iberian Latin-American Congress on Computational Methods in Engineering
P.O. Faria, R.H. Lopez, L.F.F. Miguel, W.J.S. Gomes, M. Noronha (Editores), ABMEC, Florianópolis, SC,
Brazil, November 5-8, 2017.
Analysis of design methodologies of subsea structure foundations

4.1 Finite Element Modeling

PLAXIS is a commercial finite element computer program developed by the
University of Delft (Brinkgreve and Vermeer, 1998). It is used for simulation purposes in
different geotechnical applications, mainly for stress-strain analysis and stability.
The numerical model created in PLAXIS 3D utilized specific constitutive models for
each one of its parts (i.e. manifold structure, soil, and soil-structure interface) as described in
Figure 7, below:

Numerical Model

Soil Structure Interface

(Hardening Soil) (Elastic-linear) (Mohr-Coulomb)

Figure 7. Numerical model structure

4.2 Soil Constitutive Model and Parameters

The PLAXIS 3D Hardening Soil (HS) is an advanced model that has been developed
according to the Theory of Plasticity. The main feature of the Hardening Soil model is its
ability to simulate hardening behavior (Schanz and Vermeer, 1999). The total strains are
calculated by using a stress-dependent stiffness, which is unique for each loading and
unloading/reloading. The hardening is assumed to be isotropic and dependent on the plastic
shear and volumetric strains (Fig. 8).

Figure 8. Behavior of the Hardening Soil model (Ehsan, 2013)

The soil parameters for the HS model were obtained from Fagundes et al. (2012) and
are summarized in Table 6 below:

CILAMCE 2017
Proceedings of the XXXVIII Iberian Latin-American Congress on Computational Methods in Engineering
P.O. Faria, R.H. Lopez, L.F.F. Miguel, W.J.S. Gomes, M. Noronha (Editores), ABMEC, Florianópolis, SC,
Brazil, November 5-8, 2017.
 Dandoulakis, J.P., Santa Maria, P.E., Nascimento, E.A.

Table 6. HS Constitutive Model Parameters

Parameter Nomenclature Value

3
Specific weight γsat (kN/m ) 16.1
Failure ratio Rf 0.9
Dilatancy Ψ (degrees) 0
Secant stiffness module 𝐸 (kPa) 1731b,c

Oedometric stiffness module 𝐸 (kPa) 1384b,c

Unload-reload stiffness module 𝐸 (kPa) 5834b,c

Stiffness stress level dependency m 1a

Unload-reload Poisson coefficient νur 0.2
Undrained shear strength at suo (kPa) 2
foundation base level
Rate of increase with depth of initial k (kPa/m) 1.5
undrained shear strength
Coefficient of lateral pressure for 𝐾 1
normal consolidation
Obs.: a - default PLAXIS 3D values
b – values inferred from parameters Cc = 0.515, Cs = 0.11 e eo = 2.1
c – PLAXIS reference pressure 100 kPa (1 atm)

4.3 Geometry and Mesh Definition

The three-dimensional finite-element mesh is illustrated in figure 9a, below. Several
sensitive analyses were carried out to define the mesh and model boundaries. This was
necessary to prevent interference with the results as well as identify a mesh optimum density
to reduce computational demand without compromising the model accuracy; therefore, the
modeling soil was delimited between points xmin = -18 m, xmax = 90 m, ymin = -15 m, and ymax
= 75 m, totaling an area of 90 m wide by 108 m long and 38 m deep.
The Average element size (Ie) parameter was set to 5.097. The relative element size
factor (re) was 0.7. In order to improve the model accuracy, the mesh beneath and surrounding
the manifold was further refined by creating a volume of 16 m x 13.5 m x 0.5 m, in which the
refinement factor is 0.10, and other volume of 26 m x 24 m x 9 m, in which the refinement
factor is 0.16. The horizontal displacements were prevented at the vertical limits of the model
as well as all the vertical and horizontal displacements referring to the base of the model (Fig.
9a). The mesh is comprised of 137,161 10-node tetrahedral quadratic elements (Fig. 9b).

CILAMCE 2017
Proceedings of the XXXVIII Iberian Latin-American Congress on Computational Methods in Engineering
P.O. Faria, R.H. Lopez, L.F.F. Miguel, W.J.S. Gomes, M. Noronha (Editores), ABMEC, Florianópolis, SC,
Brazil, November 5-8, 2017.
Analysis of design methodologies of subsea structure foundations

(a) (b)

Figure 9. Finite element model boundaries and element type

The contact between the soil and the skirted mudmat is modeled with Mohr-Coulomb
interface elements. The PLAXIS interface strength reduction factor (Rinter) was to 0.9 to
define the adherence between soil-structure interface contact.
The soil behavior is considered undrained for all analysis cases. The HS option
Undrained B was utilized. This option prevents volumetric changes and utilizes the soil
undrained shear strength. The Poisson ratio adopted by default was 0.495 to prevent numeric
problems during the stiffness matrix calculation (POTTZ and ZDRAVKOVIC, 1999).

Figure 10. Model mesh refinement

CILAMCE 2017
Proceedings of the XXXVIII Iberian Latin-American Congress on Computational Methods in Engineering
P.O. Faria, R.H. Lopez, L.F.F. Miguel, W.J.S. Gomes, M. Noronha (Editores), ABMEC, Florianópolis, SC,
Brazil, November 5-8, 2017.
 Dandoulakis, J.P., Santa Maria, P.E., Nascimento, E.A.

4.4 Numerical Analysis Results

PLAXIS calculates the safety factors by using its φ-c reduction function. The program
considers the shear strength of the soil and compares it with the minimum shear strength
corresponding to the equilibrium limit (Brinkgreve at al., 2016). Table 7 below shows the
calculated values:

Table 7. Numerical Results (PLAXIS 3D)

Case SF ux (mm) uy (mm) uz (mm) θr (rad) θt (rad)

1 3.34 0 0 19.4 0 0
3 2.79 1 0 21.6 0 0
5 2.70 1 -1 23.7 0 0
7 2.51 1 -1 25.1 0 0
9 2.46 1 0 25.4 0 0
11 2.46 1 0 25.3 0 0
13 2.30 1 0 27.2 0 0
15 2.34 1.3 0 27.6 0 0
17 2.22 1.3 1 28.7 0 0
19 2.20 1 2 30.3 0 0
21 2.23 1 2 30.7 0 0
23 2.22 1 2 30.8 0 0
25 2.21 1.3 2 29.8 0 0
27 2.43 1.3 2 29.9 0 0
29 2.68 1.3 2 28.0 0 0
Obs.: the order of magnitude of the manifold rotation angles were so low (i.e.
10-4) that they were considered zero.

The short-term displacement values were obtained by calculating the arithmetic mean
of the displacement values in 9 main points of the manifold (Fig. 11). The rotations around X
and Y axis were obtained through the arithmetic mean of the vertical displacements along
each side of the manifold.

CILAMCE 2017
Proceedings of the XXXVIII Iberian Latin-American Congress on Computational Methods in Engineering
P.O. Faria, R.H. Lopez, L.F.F. Miguel, W.J.S. Gomes, M. Noronha (Editores), ABMEC, Florianópolis, SC,
Brazil, November 5-8, 2017.
Analysis of design methodologies of subsea structure foundations

1 2 3

4 5 Bow
6 X

7 8 9

Figure 11. Short-term displacement measurement points

The figure below (Fig. 12) shows the displacement vectors, which indicate the soil short-term
rupture type (general rupture).

Manifold

General failure

Figure 12. Displacement vectors during short-term settlement

Figures 13 and 14, below, show the foundation displacements when subjected to
MCVs’ loads:

CILAMCE 2017
Proceedings of the XXXVIII Iberian Latin-American Congress on Computational Methods in Engineering
P.O. Faria, R.H. Lopez, L.F.F. Miguel, W.J.S. Gomes, M. Noronha (Editores), ABMEC, Florianópolis, SC,
Brazil, November 5-8, 2017.
 Dandoulakis, J.P., Santa Maria, P.E., Nascimento, E.A.

Figure 13. Short-term displacements due to MCV’s and MTU’s loads.

Manifold

Figure 14. Short-term displacement

Figure 15 shows a graph of the long-term displacement (99.9% of the final settlement)
calculated by PLAXIS 3D:

CILAMCE 2017
Proceedings of the XXXVIII Iberian Latin-American Congress on Computational Methods in Engineering
P.O. Faria, R.H. Lopez, L.F.F. Miguel, W.J.S. Gomes, M. Noronha (Editores), ABMEC, Florianópolis, SC,
Brazil, November 5-8, 2017.
Analysis of design methodologies of subsea structure foundations

Figure 15. Long-term displacement (drained conditions)

5 CONCLUSION

The choice of the most adequate methodology to solve a foundation problem is of the
paramount importance to achieve a safe and economical foundation design.
The comparison of the finite element case results, performed with the commercial
finite element program, with the analytical methodology, recommended by the API RP
2GEO, has shown that the prediction of the safety factors and soil short- and long-term
settlements are in good agreement and of the same order of magnitude; however, the former
methodology provides less conservative results than the latter. The analytical method
proposed by API RP 2GEO resulted in safety factors corresponding to 78% of those obtained
from the PLAXIS 3D analyses. Similar results were obtained by Lai (2009) and Schmid
(2009). The immediate displacements evaluated using formulae recommended by API RP
2GEO were almost the same as those obtained by PLAXIS 3D. The safety factor against
general failure of a superficial foundation recommended by Standards for offshore
applications is 2.0 (API RP 2GEO, 2014, p. 16). Thus, the design carried out by analytical
methods will result in a heavier and larger footprint mudmat than one designed with
numerical methods. This difference implies not only in additional material, project time, and
labor costs, but will also reflect in higher installation cost. This fact suggests that the payback
period of the investment in a finite element modeling software as well as in the corresponding
user training is certainly short.

CILAMCE 2017
Proceedings of the XXXVIII Iberian Latin-American Congress on Computational Methods in Engineering
P.O. Faria, R.H. Lopez, L.F.F. Miguel, W.J.S. Gomes, M. Noronha (Editores), ABMEC, Florianópolis, SC,
Brazil, November 5-8, 2017.
 Dandoulakis, J.P., Santa Maria, P.E., Nascimento, E.A.

REFERENCES
Aker Solutions. 2016. Manual de instalação, operação e manutenção do MSIAG. Rev. 00. São José
dos Pinhais: Aker Solutions Brazil.
American Petroleum Institute. 2014. Recommended practice 2-GEO –Geotechnical and foundation
design considerations, API-RP-2GEO, 2nd Edition
American Petroleum Institute. 2014. Recommended Practice for Planning, Designing and
Constructing Fixed Offshore Platforms – Working Stress Design, API-RP-2A – WSD, 21 st Edition
Bransby, M. F. & Randolph, M. F., 1999. The effects of skirted foundation shape on behaviour under
combined V-M-H loading. Journal for Offshore and Polar Engineering, Vol. 9, No. 3, pp.214-218
Brinkgreve, R. B. J. & Vermeer, P.A. 1998. PLAXIS Version 7. A.A. Balkema, Rotterdam,
Netherlands.
Brinkgreve, R. B. J., Kumarswamy, & S., Swolfs, W.M. 2016. PLAXIS 2016. Plaxis bv, Delft,
Netherlands.
Det Norske Veritas. 1992. Offshore Standard DNV-CN-30.4, Rules for the Design, Construction and
Inspection of Offshore Structures - Appendix F - Foundations, Oslo: Det Norske Veritas
Det Norske Veritas. 2016. Offshore Standard DNVGL-OS-C101, Design of offshore steel structures,
general - LRFD method. Oslo: Det Norske Veritas.
Ehsan, R. 2013. A Study of Geotechnical Constitutive Models Used in PLAXIS 2D. [s.l.]: ICE G&S
Papers Competition.
Fagundes, D. F.’ Rammah, K. I.; Pequeno, J. ; Almeida, M. S. S.; Oliveira, J. R. M. S.; & Borges, R.
G. 2012. Strength behavior analysis of an offshore Brazilian marine clay. Rio de Janeiro: OMAE
Gourvenec, S. & Randolph, M. F. 2003. Effect of strength nonhomogeneity on the shape and failure
envelopes for combined loading of strip and circular foundations on clay. Geotechnique, 53, No. 6, pp.
575–586.
ISO 19901-4 .2016. Petroleum and natural gas industries-specific requirements for offshore structures-
Part4: Geotechnical and Foundation Design Considerations, 2nd ed.
Lade, P.V. 2005. Overview of constitutive models for soils. [s.l.]: ASCE Geotechnical special
publication #128, Soil Constitutive Models: Evaluation, Selection, and Calibration, Edited by J.A.
Yamamuro and V.N. Kaliakin,. pp. 1-34.
Lai, H. S. 2009. Análise Estática de Fundação Rasa para Manifolds Submarinos. 2009. 117f.
Dissertation (Master’s degree in Oceanic Engineering) – Programa de Pós-graduação em Engenharia
Oceânica, Universidade Federal do Rio de Janeiro, Rio de Janeiro.
Pinto, C. S. 2000. Curso Básico de Mecânica dos Solos. São Paulo: Oficina de Textos.
Potts, D.M. & Zdravkovic, L. 1999. Finite element analysis in geotechnical engineering: theory. Pub
Thomas Telfod, London, 440p.
Randolph, M. F. & Puzrin, A. M. 2003. Upper bound limit analysis of circular foundations on clay
under general loading. Geotechnique, Vol. 53, No. 9, pp. 785-796.
Randolph, M. F. 2013. Analytical contributions to offshore geotechnical engineering. Paris:
Proceedings of the 18th International Conference on Soil Mechanics and Geotechnical Engineering.
Schanz, T., Vermeer, P. A., & Bonnier, P.G. 1999. The hardening soil model: formulation and
verification. Beyond 2000 in computational geotechnics. Rotterdam: Balkema. pp. 281-296.

CILAMCE 2017
Proceedings of the XXXVIII Iberian Latin-American Congress on Computational Methods in Engineering
P.O. Faria, R.H. Lopez, L.F.F. Miguel, W.J.S. Gomes, M. Noronha (Editores), ABMEC, Florianópolis, SC,
Brazil, November 5-8, 2017.
View publication stats

Analysis of design methodologies of subsea structure foundations

Schmid, M. D. 2009. Análise de Fundações Rasas em Argila para PLETs. 2009. 129f. Dissertation
(Master’s degree in Oceanic Engineering) – Programa de Pós-graduação em Engenharia Oceânica,
Universidade Federal do Rio de Janeiro, Rio de Janeiro.
Terzaghi, K. 1943. Theoretical Soil Mechanics. New York: John Wiley & Sons.

CILAMCE 2017
Proceedings of the XXXVIII Iberian Latin-American Congress on Computational Methods in Engineering
P.O. Faria, R.H. Lopez, L.F.F. Miguel, W.J.S. Gomes, M. Noronha (Editores), ABMEC, Florianópolis, SC,
Brazil, November 5-8, 2017.