Nivya Basheer 2018
Nivya Basheer 2018
Nivya Basheer 2018
Abstract. There are a number of offshore wind farms where the monopile is
socketed into rock layers. Since it is socketed into rock, it may behave different
from monopile embedded in soil. A numerical modelling of rock socketed
monopile is done using finite element (FE) software Abaqus. A stiffness
degradation method (SDM) is applied to FE model in order to predict the
behaviour under cyclic loading conditions. Parametric studies are carried out by
varying rock socketed depth (d), length of monopile below seabed (L), intensity
of horizontal loading (H) and subsoil conditions to evaluate the long-term
permanent deformation of offshore rock socketed monopile foundations. Sui-
table permanent deformation factors are proposed for offshore rock socketed
monopiles for the first time in the literature. It is observed from the results that
the deformation behaviour of the monopile changes from stiff to flexible with
increase in rock socketing and in turn the pile head deflection going down. From
the bending moment diagram, flexible and stiff behaviour of monopile can be
identified and is an indicator of curvature of the deflection line of pile.
1 Introduction
Monopiles are one of the common foundation options for offshore wind turbines. Often
the sea bed is made up of bed rock and rock socketing becomes necessary for installing
the monopiles. In many places, the monopiles are being successfully socketed into rock
for different wind farms, e.g., situated on the East coast of England in Northumberland,
Blyth offshore wind project installed monopiles of diameter 3.5 m into bedrock of
sandstone. At North Hoyle wind farm, installation of monopile (4.0 m diameter at
seabed) consists of driving through upper layers of sand and clay, and drilling and
driving through rock layer of sandstone and mudstone. In Bockstigen wind farm in
Sweden monopiles were socketed into rock (www.offshorecenter.dk, www.
subacoustech.com, www.technology.stfc.ac.uk).
Germanische Lloyd (GL) rules and regulations give the design procedure for
foundations of offshore wind energy converters in Germany (Achmus 2010). In this
regulation the p-y method defined by API code is recommended in order to estimate the
behaviour of piles under horizontal loading. But the use of p-y curves in estimating the
behaviour of monopiles may be misleading because of the fact that these curves are
formulated based on field testing of piles with number of cycles less than 200 and
applicable for piles with diameters up to 2 m.
Many studies have been conducted in the past in order to understand the behaviour
of laterally loaded and axially loaded monopiles (Achmus et al. 2008, 2009; Albiker
and Achmus 2012; Achmus and Albiker 2014; Arshi and Stone 2011; Kellezi and
Hansen 2003; Little and Briaud 1988; Schmoor and Achmus 2013) and some design
guide lines are also available in the literature (Achmus et al. 2008; Schmoor and
Achmus 2013; Thieken et al. 2014). But the rock-monopile interaction under cyclic
loading is less discussed. Arshi and Stone (2011) conducted a series of small scale
single gravity tests to investigate the performance of a monopile, combined monopile
and bearing plate foundation where the pile is socketed into a weak rock. In the model
studies, the weak rock layer is modelled using a weak sand and gypsum mix. The
results of the study provide an insight into the effect of the various foundation elements
(i.e. pile, plate and rock socket) and their contribution to the overall performance of the
foundation system. Wang et al. (2007) discusses the behaviour of large-diameter
rock-socketed CFST (concrete-filled steel tube) piles under lateral loads based on field
tests and numerical analysis. The horizontal capacity and deformation of large-diameter
rock-socketed piles are analyzed from the measured displacements and internal forces
of piles. The interactive behaviour of pile-rock and the influence of backfilled sand on
horizontal capacity are also discussed. Using the Finite Element Method (FEM) con-
sidering the properties of the pile-soil interface, the test results are simulated numeri-
cally. This result show that stress concentration effect in the region near the bottom of
the steel tube should be considered in the design, because the socketed part of piles
bears most of the lateral load.
From the literature review, it may be concluded that the offshore monopile foun-
dations are mainly subjected to wave and wind loading and these loads are cyclic in
nature. One of most important aspects of designing a monopile foundation is the
deformation under cyclic horizontal loading and accumulation of permanent defor-
mation with increasing number of cycles. In this paper, the deformation behaviour of
rock socketed monopile foundation under cyclic loading is studied by numerical
modelling of the pile-soil system using the FEM software Abaqus and by applying the
stiffness degradation method (SDM) with increasing number of load cycles. Permanent
deformation of the piles under cyclic loading is investigated for different rock socketing
depths, different pile lengths and horizontal load magnitudes. A deformation accu-
mulation parameter is proposed further.
176 N. Basheer et al.
This method was developed at the Institute for Geotechnical Engineering, Leibniz
University of Hannover, Germany (Achmus et al. 2009). In a cyclic triaxial test, the
degraded stiffness of soil after N cycles (Es,N) can be expressed in terms of stiffness
modulus after first cycle (Es,1) using the equation,
ES;N
¼ N b1 X
b2
ð1Þ
Es;1
Here b1 and b2 are soil parameters and X is the cyclic stress ratio defined by
r1;cyclic
X¼ ð2Þ
r1;f
where r1,cyclic is the maximum principal stress in a cycle and r1,f is the maximum
principal stress at failure subjected to static loading. From cyclic triaxial test results
documented in the literature, typical regression parameters b1 and b2 were found for
dense sand to be b1 = 0.12, b2 = 0.50 and for medium dense sand b1 = 0.15,
b2 = 0.50. Detailed numerical implementation of the SDM in the finite element code
Abaqus (Abaqus documentation version 6.11-3) is discussed in Achmus et al. (2008).
3 Numerical Modelling
The model considered in the present work consists of a monopile of diameter 7.5 m
and wall thickness of 9 cm. For simplicity the hollow cylindrical steel monopile
(modulus of elasticity E = 210 GPa and Poisson’s ratio m = 0.2) is replaced by a solid
cylindrical pile with same diameter such that bending stiffness of both piles remains the
same. The monopile is installed into a layered soil with upper sand and lower rock layer
as shown in Figs. 1 and 2. The sand is considered elasto-plastic with Mohr–Coulomb
failure criterion. The stiffness modulus of the soil varies with depth according to the
following equation,
k
rm
Es ¼ jrat ð3Þ
rat
Here Es is the oedometric stiffness modulus which varies with stress condition, rm
is the current mean principal stress in the considered soil element and rat= 100 kN/m2
is a reference (atmospheric) stress. The parameter j determines the soil stiffness at the
reference stress state and the parameter k rules the stress dependency of the soil
stiffness. The material parameters used to model different materials are listed in Table 1
(Das 2013; Jaeger et al. 2008; Achmus et al. 2009). At the contact of the pile soil
interface, interface friction coefficient tan d = 2/3 tan(/) is considered. At pile-rock
interface, tan d = tan(/) is taken. In the contact normal direction, hard contact is
Analysis of Offshore Rock Socketed Monopile … 177
V = 10 MN
H
h = 20 m
Sand
L
Rock d
D
Sand
Rock
Cylindrical outer
boundary U1 =U2= 0
U3, UR1, UR2 and
UR3 ≠0
considered at the pile-rock interface and soft contact is considered at the pile-soil
interface with assumed interface stiffness taken same as the soil stiffness.
The vertical load (V) acting on the monopile is assumed as 10 MN which is the
weight of the super structure. The variable horizontal load (H) is acting at a height
(20 m) above the seabed level. Three different horizontal load magnitudes are assumed-
10, 15 and 30 MN. Due to symmetry conditions, only the half of the pile-soil-model is
considered. The diameter of the numerical model considered for analysis is sixteen
178 N. Basheer et al.
times the pile diameter. The bottom boundary of the model is extended to 20 m below
the base of the monopile. The steps involved in modelling using stiffness degradation
method are shown in the Table 2.
4 Parametric Studies
A parametric study is conducted to estimate the effect of embedded length of the pile
below seabed (L), rock socketed depth (d), horizontal load intensity (H) and the subsoil
conditions on deformation behaviour of monopile foundation under long term cyclic
loading. Table 3 lists different loading conditions and pile geometries considered in this
study. The depths of rock socket is decided based on different literature (Choy et al.
2004; Srinivasamurthy and Pujar 2009; Liu et al. 2011). Three different subsoil con-
ditions are simulated (i) two layered soil with upper layer dense sand and lower layer
Analysis of Offshore Rock Socketed Monopile … 179
hard rock (case I), (ii) two layered soil with upper layer dense sand and lower layer soft
rock (case II), three layered soil with upper layer medium dense sand, middle layer
dense sand lower layer soft rock (case III). Hard rock is considered elastic (E = 70
GPa, m = 0.2) and soft rock elasto-plastic with Mohr-Coulomb criteria (E = 1 GPa,
m = 0.33, c = 50 kN/m2 and /′ = 25°) (Table 1).
0 0
Depth below seabed (m)
10 10
15 d=2m 15 d=4m
N=1 N=1
20 N = 10 20 N = 10
N = 100 N = 100
25 N = 1000 25 N = 1000
N = 10000 N = 10000
30 30
0 10 20 30 40 50 0 5 10 15 20 25
Lateral deflection of monopile (cm) Lateral deflection of monopile (cm)
0
Depth below seabed (m)
10
15 d = 10 m
N=1
20 N = 10
N = 100
25 N = 1000
N = 10000
30
0.0 2.5 5.0 7.5 10.0
Lateral deflection of monopile (cm)
Fig. 3. Stiff to flexible behaviour with increase in rock socketing depth (monopile socketed into
hard rock)
N = 10000
σ10000 ≠ σ1
N = 10000
E10000 ≠ E1
Stiff Pile
σ10000 ≠ σ1
E10000 ≠ E1 σ10000 ≈ σ1
E10000 ≈ E1
Flexible Pile
Static loading Cyclic loading
0 0
5 5
10 10
15 15
d=4m d=4m
20 H = 30 MN 20 H = 15 MN
N=1 N=1
25 N = 100 25 N = 100
N = 10000 N = 10000
30 30
-20 0 20 40 60 80 -2 0 2 4 6 8 10 12 14
Lateral deflection of monopile (cm) Lateral deflection of monopile (cm)
Fig. 5. Stiff and flexible behaviour of monopile socketed in soft rock at different load levels
ysN / ys1
ysN / ys1
3.0
ys10000 / ys1 ys10000 / ys1
2.5
2.0 1.5
1.5
1.0 1.0
0 2 4 6 8 10 0 2 4 6 8 10
Rock socketed depth (m) Rock socketed depth (m)
Fig. 6. Variation of accumulation rate of deformation with rock socketing depth for a monopile
subjected to a cyclic lateral load (H) of (i) 30 MN (ii) 15 MN
(i) 0 (ii) 0
Depth below seabed (m)
5
Depth below seabed (m)
10 10
15 15
Fig. 7. Monopile of L = 30 m, d = 10 m socketed in (i) hard rock (ii) soft rock subjected to
H = 30 MN
and the accumulation of head deformation for monopiles in weak rock is large com-
pared to the other case (about 4 times larger for 4 m rock socketed piles and 6 times
larger for 10 m rock socketed pile). In analogy to the statement given above when
different load levels were analysed, also here it can be concluded that a higher stiffness
of the rock leads to a reduced pile-soil system stiffness (flexible behaviour) and
therewith to a reduced cyclic deformation accumulation. Figure 8 presents the nor-
malized pile lateral deflection ysN =ys1 for different rock socketing depths in soft and
hard rock after increasing number of cycles. As the rock socketed depth increases, the
rate of accumulation under cyclic loading reduces.
(i) (ii)
2.4 Hard Rock 2.4 Soft Rock
d=0m d=0m
2.2 2.2
d=4m d=4m
2.0 d=8m 2.0 d=8m
ysN / ys1
ysN / ys1
1.8 1.8
1.6 1.6
1.4 1.4
1.2 1.2
1.0 1.0
1 10 100 1000 10000 1 10 100 1000 10000
Number of cycles N Number of cycles N
yN ¼ y1 N m ð4Þ
Here yN and y1 are the pile head deflection after N number of cycles and 1st cycle
respectively. The empirical parameter m in Eq. 4 is evaluated herein using the results
(deflection at seabed level after N cycles) obtained from numerical analysis using a
regression analysis. Figure 9 presents the variation in m for different rock socketed
depths. From the Fig. 9, it can be seen that the value of m depends on length of
monopile below seabed level (L), rock socketed depth (d), intensity of lateral loading
(H) and subsoil conditions. The m increases with load intensity and decreases with
increase of rock socketing and with increase of the pile length.
(i) (ii)
0.16 L = 30 m L = 40 m
0.14 H = 10 MN 0.08 H = 10 MN
H = 15 MN H = 15 MN
0.12 H = 30 MN H = 30 MN
0.10
0.06
m
0.08
m
0.06
0.04 0.04
0.02
0.00 0.02
0 2 4 6 8 10 0 2 4 6 8 10
Rock socketed depth (m) Rock socketed depth (m)
10
20
Depth (m)
N = 1, d = 0 m
30 N = 10000, d = 0 m
N = 1, d = 10 m
N = 10000, d = 10 m
40
50
60
0 200 400 600 800 1000
Bending Moment (MNm)
10
20
Depth (m)
N = 1, d = 0 m
N = 10000, d = 0 m
30 N = 1, d = 10 m
N = 10000, d = 10 m
40
50
0 50 100 150 200 250 300
Bending Moment (MNm)
6 Conclusions
The behaviour of laterally loaded piles for varying geometry, soil conditions, lateral
loading and rock socketing depths is analysed and studied numerically. From the
results it can be concluded that the behaviour of the monopile changes from stiff to
flexible with increase in rock socketed depth and that in turn the deformation accu-
mulation of the pile at the surface is going down. The other observation is that
accumulation rate of deflections follow the same trend. As the depth of the pile below
seabed level increases, rock socketing depth has no significant effect on the accumu-
lation of deformation at low load levels. For the evaluated cases, it is also concluded
that a ratio between rock socketed length d and total pile embedment length L of
approximately d/L = 1/6 might be seen as an orientation for sufficient cyclic stability.
From the bending moment diagram, flexible and stiff behaviour of monopile can be
identified and is an indicator of curvature of the deflection line of pile.
Acknowledgements. This work has been carried out with scholarship funded by DAAD for Ms.
Navy Basheer for a period of six months. This work is a result of collaboration between Institute
for Geotechnical Engineering, LUH, Germany and Department of Civil Engineering, IIT Delhi,
India.
References
Abaqus/Standard User’s Manual, Version 6.11-3: Dassault Systèmes Simulia Corporation,
Providence, Rhode Island, USA, 2014
Achmus, M.: Design of axially and laterally loaded piles for the support of offshore wind energy
converters. In: Indian Geotechnical Conference—2010, GEOtrendz, IGS Mumbai Chapter &
IIT Bombay, 16–18 Dec 2010
Achmus, M., Kuo, Y.-S., Abdel-Rahman, K.: Behaviour of monopile foundations under cyclic
lateral load. Comput. Geotech. 36(5), 725–735 (2009)
Achmus, M., Abdel-Rahman, K., Kuo, Y.-S.: Design of monopile foundations for offshore wind
energy plants. In: Proceedings of the 11th Baltic Geotechnical conference of Geotechnics in
Maritime Engineering, 15–18 Sept 2008, Gdansk, Poland
Achmus, M., Albiker, J.: Prediction of accumulated deformations of cyclic laterally loaded piles
in sand. In: Proceedings of the 8th European Conference on Numerical Methods in
Geotechnical Engineering (NUMGE), Delft, The Netherlands, 2014
Albiker, J., Achmus, M.: Cyclic performance of horizontally loaded piles in layered subsoil. In:
Proceedings of the 12th Baltic Sea Geotechnical Conference, Rostock, Germany, May 31st–
June 2nd, 2012
Arshi, H.S., Stone, K.J.L.: An investigation of a rock socketed pile with an integral bearing plate
founded over weak rock. In: Proceedings of the 15th European Conference of Soil Mechanics
and Geotechnical Engineering, 12–15 Sept 2011, Athens, Greece, 2011
Choy, K.K., Pang, T.C.P., Li, W.W., Tse, S.H.V., Lam, S.C., Kung, W.C.F., Lau, C.W.J.,
Pappin, J.W., Ng, H.K., Lee, W.H., Wong, N.K.P., Lee, K.K.P., Cheng, M.L.: Code of
Practice for Foundations. Technical Report. Buildings Department, Mongkok, Kowloon,
Hong Kong (2004)
http://www.offshorecenter.dk/log/bibliotek/Blyth%20Wind%20Farm.pdf as seen 24th June 2017
http://www.subacoustech.com/wp-content/uploads/544R0503.pdf as seen 24th June 2017
186 N. Basheer et al.