IMSC 2017 1

Numerical Analysis of the Breaking Wave Impact on the

Monopile Support Structure
Veić Duje 1, Sulisz Wojciech 1
1
Institute of Hydroengineering, Polish Academy of Sciences (IBW PAN)
email: [email protected]; [email protected]; [email protected]

ABSTRACT

Hydrodynamic loading on monopile foundations for an offshore wind turbine are numerically
analyzed using a fully 3D nonlinear solver. The numerical model is based on the decompositions
strategy developed at DTU between two open source codes, namely OceanWave 3D and
OpenFOAM. Special attention is given to the mesh grid dependency analysis, where wave
impact is analysed by taking into account different levels of the free-surface sharpness.
Numerical results are compared against the recommended analytical solution.

KEY WORDS

Plunging breaking wave, Incompressible Navier-Stokes/VOF, OpenFOAM, OceanWave3D

MW) and heavier, which requires larger

foundations (larger diameter of the
1. INTRODUCTION monopile). The wave loading is scaling with
the square of the pile diameter which
Projection for European Union energy implies that the hydrodynamic loading from
demand for year 2030 is 3.178 TWh, EWEA waves and breaking waves in particular may
(2015). It is expected that clean and become design drivers.
renewable wind energy will cover 24.4% of The impact forces from breaking waves
European Union electricity demand, out of on the vertical cylindrical structures have
which 16.7% involves onshore and 7.7% mostly been studied experimentally. The
offshore wind energy. limitation of laboratory tests is that
The majority of the support structures for information such as pressure is only
offshore wind turbines are monopiles measured at a limited number of locations
installed up to the 40m depth. In this selected prior to the tests. Experimental
relatively shallow water, the progressive procedures are in general very challenging
waves have high possibility of reaching and subjected to the range of measurement
breaking limit. Since breaking waves uncertainties. The existing formulas for
release a high amount of energy, the calculation of impact forces from breaking
structure could be excited with the extreme waves are empirical and semi-empirical,
impact (slamming) loads. and thus subjected to the experimental
The general trend in wind industry is that observations. Numerical computations,
wind turbines are getting larger (up to the 10 however, provide the results of high spatial
IMSC 2017 2

and temporal resolution in a controlled In the present study, the same numerical
environment. Therefore, numerical analysis technique is used for the analysis of an
could yield the new insight into the breaking extreme hydrodynamic loading on the
wave impact phenomena. monopile support structure, observed from
The numerical models regarding two- laboratory measurements during the
phase air-water modelling of undisturbed simulation of realistic 50-year storm
breaking wave phenomena could be found condition. The 50-year storm condition is
in the studies of Chen et al. (1999), idealized with the JONSWAP energy
Christensen (2006), Wang et al. (2009), spectrum. Parameters of the spectrum
Jacobsen et al. (2012), Chella et al. (model scale) are: 𝐻𝑠 = 0.22𝑚, 𝑇𝑝 = 1.9𝑠,
(2015a). The scope of numerical models on gamma = 3.3. The reference monopile
breaking waves includes analysis on: model corresponds to the commonly
breaking wave kinematics, overturning jet, installed monopile prototypes. The chosen
air entrapments and turbulence parameters water depth is 𝑑 = 30𝑚. The diameter of the
during the breaking wave process. The monopile is 𝐷 = 7𝑚. Numerical model is
numerical results show a good comparison solved assuming Froude model scaling
against the experimental data, in e.g. the according the scale factor 45. Special
experimental campaign on spilling breakers attention is given to the grid dependency
from Ting & Kirby (1994-1996). Similar analysis, where wave impact is analysed by
numerical model could be used for analysis taking into account different levels of the
of the fully 3D breaking wave impact on the free-surface sharpness. The results from
structures. In the recent studies of Kamath the numerical study are compared against
et al. (2015) and Bihs et al. (2015), the proposed analytical solution given in
evaluation of the breaking-wave forces on IEC 61400-3 (2009) standard.
the vertical cylinder is solved within the
framework of the open-source code
REEF3D (incompressible Navier-Stokes 2. ANALYTICAL APPROACH
/Level set method). However, their relatively
large computational domain (> 15 millions of The offshore structures are designed
computational cells) and long-time of according to the design standards, which
realization (>25s) is computationally too recommend formulations for calculation of
expensive. hydrodynamic loads. For slender structures,
In order to increase the computational such as jacket and monopile, hydrodynamic
efficiency, Paulsen, et al. (2014) proposed a loads are usually calculated according to the
domain decomposition strategy. The model well-known Morison’s equation, (Morison et
is based on one-way coupling between fully al. (1950)). The Morison’s equation is a sum
non-linear potential flow solver and fully of two terms; one being an empirical drag
non-linear Navier-Stokes/VOF solver. For term proportional to the square of the wave
the case of non-breaking wave loading, particle velocity and the other being an
Paulsen demonstrated a good comparison inertia term, derived from potential flow
between the numerical model solutions and theory and proportional to the wave particle
experimental results. However, some acceleration. The Morison’s equation is
discrepancies were observed in the peak of defined as follows:
the in-line force signal for the case where
breaking wave interacts with the structure. 𝐹𝑀 = 1⁄2 𝐶𝑑 𝜌𝐷|𝑢|𝑢 + 𝐶𝑚 𝜌𝐴|𝑢̇| (1)
Using the same decompositions strategy, in
his recent study, Ghadirian et al. (2016)
where empirical coefficients 𝐶𝑚 and 𝐶𝑑 ,are
studied impact forces from the phase-
inertia and drag coefficient respectively. The
focused breaking waves on vertical
Morison’s equation based on a stream
cylindrical structure. Discrepancies of the
function wave kinematics and empirically
line-force signal between the numerical
determined coefficients, predicts hydrodynamic
model and laboratory measurements could
be found as well.
IMSC 2017 3

loading of weakly non-linear waves with a to use the value of wave celerity 𝑐𝑏 for the
good engineering accuracy. calculation.
However, the impact forces from the Temporal development of the slamming
breaking waves are completely out of the coefficient is presented in Fig 1. The
scope of the Morison’s equation. When abscissa presents the azimuth angle of the
hydrodynamic loading is governed by the wetted area (α). At the instant of the impact
impulse force from the breaking wave, the (𝑡 = 0s) the magnitude of the line force is
total hydrodynamic force is calculated as: maximum, and when the cylinder wetted
area (see Fig 1) is around α=50˚ the line
𝐹𝑇𝑂𝑇𝐴𝐿 = 𝐹𝑀𝑂𝑅𝐼𝑆𝑂𝑁 + 𝐹𝐼𝑀𝑃𝐴𝐶𝑇(𝑆𝐿𝐴𝑀𝑀𝐼𝑁𝐺) (2) force falls to zero. The peak value of the
slamming coefficient is 𝐶𝑠 = 2𝜋.
The impact forces from the plunging wave
breakers are expected to induce the most
violent loads on the structure, and they are
usually calculated according to the
analytical proposal of Wienke (2001).
Wienke’s solution is based on the solution
developed by Wagner (1932), who
described the impact of infinitely long
cylinder hitting the calm water by a potential
flow theory, where the flow is assumed to be
incompressible, inviscid and irrotational.
Furthermore, the surface tension of the fluid
and forces due to gravity are neglected and
cylinder is assumed to be rigid. The potential Fig 1 Integration of pressure distribution for
flow around the cylinder is assumed to be different time steps - Wienke(2001) solution
equivalent to the potential flow around the
flat plate. According the definition of the
velocity potential (𝛷) (see Faltinsen 1993, To obtain the total breaking wave impact
page 286) and assumption of the constant force (𝐹𝐼𝑀𝑃𝐴𝐶𝑇 [N]), 2D line impact forces
velocity of the impact (c [m/s]), Wagner (𝑓𝑖 [N/m]) are integrated over the impact
solved only the temporal part of the area. Typically, this is done in a discrete
Bernoulli's equation, while Wienke included manner where the integral is replaced by a
solutions for spatial derivatives too. sum of discrete elements, as it is illustrated
Bernoulli's equation: in Fig 2, and hence expressed as:

𝜕𝛷 𝜌 𝜕𝛷 𝜕𝛷
𝑝 = −𝜌 𝜕𝑡 − 2 (( 𝜕𝑥 )2 + ( 𝜕𝑦 )2 ) + 𝑝(𝑡) (3) 𝐹𝐼𝑀𝑃𝐴𝐶𝑇 = ∑ 𝑓𝑖 (5)

The curling factor parameter (𝜆), indicates

By integrating the pressure along the the area of the impact, or in other words,
pressure plane, the so-called line force how much of the wave crest (𝜂𝑏 ) is
(𝑓𝑖 [N/m]) is determined: contributing to the impact (slamming) force,
Fig 2. The curling factor is usually estimated
𝑓𝑖 = 0.5 𝐶𝑠 (𝑡)𝜌𝐷𝑐 2 ; [N/m] (4) semi-empirically and in the case of the
plunging wave breaker the curling factor
where, 𝐶𝑠 is the slamming coefficient, and c might be approximated as 𝜆 = 0.4, e.g.
is the impact velocity, but for calculation of Goda (1966), Wienke and Oumeraci (2005).
the breaking wave impact it denotes the With the assumption of the rectangular
water particle velocity in the plunging wave distribution of the line forces along the
crest. As breaking wave occurs when water- impact area (Wienke (2001)) the slamming
particle velocities under the wave crest force is determined as:
exceed the wave celerity, Wienke proposed
IMSC 2017 4

𝐹𝑆𝐿𝐴𝑀𝑀𝐼𝑁𝐺 = 0.5 𝐶𝑠 (𝑡)𝜌𝐷𝑐𝑏 2 𝜆𝜂𝑏 (6)

The Navier-Stokes equations are solved
in combination with a volume of fluid (VOF)
surface capturing scheme. The governing
equations are solved within the framework
of the open-source computational fluid
dynamics toolbox OpenFoam®. The
equations are discretized using a finite
volume approximation on generally
unstructured grids. Wave generation and
absorption relaxation zones are applied at
the inlet and outlet boundary respectively
(OF relaxation zones, see Fig 3). The
solutions in relaxation zones are forced
towards the known solution from the “outer”
region, Fig 4. Wave propagation in “outer”
Fig 2 Schematic view - wave impact region is solved within the framework of
(inspired by Wienke (2001) figure) OceanWave3D (OCW3D) toolbox, developed
by Engsig-Karup et al. (2009). The OCW3D
solves fully nonlinear wave propagation up
3. NUMERICAL MODEL to the limit of the wave breaking. More
detailed description of numerical model and
The numerical model is based on the corresponding governing equations could
decomposition strategy proposed by be found in the study of Paulsen et al.
Paulsen et al. (2014), where incompressible (2014).
Navier-Stokes/VOF equations are solved in The seabed is modelled as flat and
a very small "inner" region of interest, while impermeable. At the atmosphere boundary,
wave propagation up to the "inner" region of an inlet/outlet boundary condition is applied.
interest is solved with a potential flow solver, This boundary condition allows air and water
as it is illustrated in Fig 3. This technique to escape the domain, however, only air is
significantly reduces the computational able to re-enter. The non-slamming part of
time. Simulations of the wave propagation in the force is governed by the low Keulegan-
a long Navier-Stokes/VOF domain could be Carpenter number, and the slamming part of
influenced by the numerical diffusion. This is the force is governed by the wave impact of
manifested by the artificial reduction of the high frequency, hence it is justified to
propagating wave height. Hence, the neglect the viscous effects on the forcing.
decomposition strategy which allows shorter So, the slip conditions were applied at all
numerical domain is favourable regarding solid surfaces. The time step size is
the numerical diffusion issue as well. controlled with adaptive time stepping
based on Courant-Friedrichs-Lewy (CFL)
criterion. For all computations in this study
CFL number is kept <0.2.
The Navier-Stokes/VOF numerical domain
used in this study is presented in Fig 4. The
domain is defined with the length ≈ 1 wave
length and with half-width ≈ 5 cylinder
diameter.

Fig 3 Sketch of decomposed domain

IMSC 2017 5

Fig 4 Sketch of the numerical domain

the hydrostatic pressure, 𝒈 is the acceleration

due to gravity, 𝒙 is the Cartesian coordinate
4 NUMERICAL ANALYSIS vector, μ is the dynamic molecular viscosity
(see e.g. Paulsen et. al (2014) for more
details). The density is calculated regarding
4.1. Grid dependency analysis the linear weighting:
The solution of numerical wave
propagation depends on the grid density 𝜌 = 𝛼𝜌𝑤 + (1 − 𝛼)𝜌𝑎 (8)
around the free surface area. The non-
breaking wave propagation is usually where α is water volume fraction, obtained
captured with 15-20 computational cells per from equation:
wave height (where the aspect ratio of the 𝜕𝛼
cell is best to be 1, but at least ¼). Moreover, 𝜕𝑡
+ 𝛻 ⋅ 𝒖𝛼 + 𝛻 ⋅ 𝒖𝒓 𝛼(1 − 𝛼) = 0 (9)
in order to capture the process of the local
wave breaking, additional grid refinement is The marker function 𝛼 is 1, when the
necessary. As part of this study, 44 computational cell is filled with water and 0,
computational cells per wave height have when it is empty. In the free surface zone the
been sufficient to capture the wave marker function will have a value in the
propagation and the local wave breaking in interval 𝛼 ∈ [0; 1] indicating the volume
the numerical domain. However, in order to fraction of water and air respectively.
analyse the breaking wave - monopile The numerically computed free-surface,
structure interaction, the grid has to be based on the VOF method is often smeared
further refined in the zone of the impact due to the numerical discretisation error, as
area. Next paragraphs explain the argument it is illustrated in Fig 4 and Fig 5.
why the grid resolution is an important Furthermore, because the coupling between
parameter in presented numerical model for dynamic pressure and density in the
the analysis this phenomena. momentum equation, spurious air velocities
The numerical solution in this study is are present. This is presented in Fig 6,
based on the incompressible Navier-Stokes where velocity distribution exist in both
equation for the two phase flow of water and phases: water and air. The Fig 6 presents
air: horizontal velocity distribution in the
numerical domain for the moment slightly
𝜕 1
𝜕𝑡
𝜌𝒖 + 𝛻 ⋅ 𝜌𝒖𝒖𝑇 = −𝛻𝑝∗ + 𝒈 ⋅ 𝒙𝛻𝜌 + 𝑛 𝛻 ⋅ (𝜇)𝛻𝒖 (7) prior to the wave impact on the monopile
structure. The figure is supported with the
where, 𝒖 = (𝑢, 𝑣, 𝑤) is the instantaneous wave profiles which correlate to the different
velocity in Cartesian coordinates and 𝑢, 𝑣 levels of water fraction in computational
and 𝑤 are the velocity components in the cells at the moment of impact, α = 0.01, 0.1,
three Cartesian coordinates 𝑥, 𝑦 and 𝑧; 𝜌 is 0.5, 0.95 respectively. This shows that at the
the density, 𝑝∗ is the pressure in excess of instant of the wave impact on the structure,
IMSC 2017 6

Fig 5 The free-surface smearing in the numerical domain, "alpha" presents percentage of
the water in the computational cells ; refinement zones used in presented study

the computed pressure is governed by free-surface and the impact area, as it is

the presence of the artificial velocity in the illustrated in Fig 5. The finest mesh in the
air and the small amount of water fraction in near boundary region is defined with the
the computational cells. Hence, according to grid size 𝑑𝑥 = 𝑑𝑦 = 𝑑𝑧 = 0.001m (level_4).
the equations Eq.(7) – Eq.(9), the computed This numerical domain corresponds to the near
impact pressures are artificially damped. 4.4 million cells in total. The total number of
Therefore, for the case of the presented nodes for the refinement levels defined
numerical model, the impact pressure according to the Fig 5 is presented in Fig 7.
signals around the structure are sensitive to
the size of the numerical grid.

Fig 7 Comparioson of the breaking wave

profiles for different levels of computational
grid refinement; total number of nodes for
Fig 6 The horizontal velocity field prior the the different level of the grid
wave impact; wave profiles which
coresopnds to the equal water fraction Small discrepancies in the shape of the
levels in computational cells, looknig from breaking wave profiles are observed for
up: α = 0.01, 0.1, 0.5, 0.95 different levels of the grid refinements, as it
is presented in Fig 7. However, significant
Hence, the grid dependency analysis is discrepancies are observed in both the
conducted. The unstructured hexahedral magnitude and the distribution of the impact
mesh is used to refine the zone around the pressure. Series of figures Fig 8 - Fig 10,
IMSC 2017 7

present dynamic pressure distribution on

the monopile structure for the different grid
resolution levels. The figures are supported
with the shape of the breaking wave profile
at the corresponding moment of impact. The
peak impact pressure observed for the case
“level_4” is even 300% higher than the peak
impact pressure observed for the case
“level_2”.

Fig 10 Dynamic pressure disribution and

corresponding shape of the wave profile for
grid resolution "level_4"

Fig 8 Dynamic pressure disribution and

corresponding shape of the wave profile for
grid resolution "level_2"

Fig 11 Comparison between the

hydrodynamic loading on the monopile
structure for three different grid resolution
levels

4.1. Results & comparison

The results of the study are referenced to
Fig 9 Dynamic pressure disribution and the numerical model presented in the
corresponding shape of the wave profile for section_3. The vertical line-force (𝑓𝑖 [N/m])
grid resolution "level_3" distribution is obtained by integrating the
pressure field around the circumference of
The comparison between the impact the monopile structure. The vertical line
force signals considering the different grid force distribution on the monopile structure
resolution levels are presented in Fig 11. It regarding the selected case “level_4” is
is found that impact forces are not yet presented in Fig 12 for the three different
converging to the final solution, although the time-steps. The vertical line force distribution
discrepancies between the peak impact before the impact (non-slamming force)
forces are smaller compared to the corresponds to the time slightly before 𝑡 <
discrepancies in the peak impact pressures. 𝑡1 . The vertical line force distribution which
contributes to the impact force is presented
as a blue shadowed area. It could be
approximated that the vertical impact line
IMSC 2017 8

force distribution follows triangular shape. The reason of the high discrepancies
This corresponds to the laboratory presented in Table 2 might be found in the
measurements on the breaking wave impact nature of the breaking wave–structure
pressures at the front line of the vertical interaction. The breaking wave impact on
cylinder, which follows the triangular the monopile support structure is strongly 3-
distribution as well, e.g. Zhou et. al (1991), dimensional phenomena. The solution
Chan et al. (1995). based on the theoretical 2-dimensional
model might provide with the too
conservative results. Discrepancies in
results found from laboratory analysis
indicates possible influence of the air
entrapments and the shape of the breaking
wave on the impact force measurements.
The results from presented numerical model
are however, found to be grid dependable
and this could be a source of high
discrepancies in the value of the slamming
coefficient.

Fig 12 The vertical line force 𝒇𝒊 [N/m]

distribution for the case "level_4" which
corresponds to the different time steps

The maximum obtained slamming

coefficient for the case “level_4” is 𝐶𝑠 =
0.5𝜋, which is significantly lower than the
maximum slamming coefficient proposed by
Wienke (2001). The comparison regarding
the laboratory studies of Tanimoto et
al.(1986) and Ros et al. (2011), as well as
regarding the numerical solution from
Hildebrand et al. (2011) are presented in
Table 1. Fig 13 Comparison between the analytical
and numerical solution for the:
hydrodynamic breaking wave force; - wave
Table 1 The slamming coefficient elevation
compariosn
The hydrodynamic force obtained from
Author Cs/π the proposed numerical model is compared
Wienke (2001), to the analytical solution given in IEC 61400-
=2 3 (2009) standard. The total hydrodynamic
theoretical
Tanimoto (1986). load on the monopile structure is calculated
≈1 according to the Eq. (2). The parameters for
experimental
Ros (2011), the Morison’s equations are: 𝐶𝑚 = 1.8, 𝐶𝑑 =
≈ 1.37 0.95, 𝐻 = 0.32𝑚, 𝑇 = 0.95𝑠, where the wave
experimental
Hildebrandt (2011), kinematics is solved by assuming the stream-
≈ 1.1 function wave theory. The parameters for
numerical
level_4, calculation of the slamming force according
≈ 0.5 Eq. (6) are: 𝑐𝑏 = 2.8 m/s, 𝑅 = 0.08 m, ρ =
numerical
1000 kg/m3 and 𝜂𝑏 = 0.25𝑚. The slamming
coefficient 𝐶𝑠 is related to the solution
presented in Fig 1. The slamming force is
IMSC 2017 9

calculated assuming the triangular force

distribution. Therefore, the slamming force  The artificial velocity in the air phase
formulation from Eq. (4) is multiplied by the and smeared water-air interface are
factor 0.5. source of artificial damping of the
The comparison between the analytically impact pressures.
calculated and numerically computed  The spatial and temporal distribution of
hydrodynamic force on the monopile the impact pressures are dependable
structure is presented in Fig 13. The non- on the size of computational grid.
slamming part of the force is found similar in Correspondingly, the characteristic of
the analytical and numerical solution. the total impact force is affected too. For
However, the slamming part of the force is the presented case, the impact force
significantly different comparing analytical to magnitude is significantly influenced by
numerical solution. The presented the computational grid size. However,
numerical model captures fully 3- the duration of the impact force is found
dimensional nature of the breaking wave- to be equal regardless the grid
structure interaction. As it is presented in Fig refinement.
11, the magnitude of the impact force is  In order to reach convergence of the
dependable on the sharpness of the free- impact force signal, a further grid
surface. However, the duration of the impact refinement is necessary. However, this
force does not depend on the grid size. would be too expensive solution. An
Therefore, it could be observed that the alternative approach is to reduce the
analytical solution proposed by Wienke effects of the artificial velocity in the air
(2001) estimates significantly shorter impact and effects of the smeared free-surface
force time (for the breaking wave case under interface in the computation.
the consideration). Moreover, the analytical  The vertical line-force distribution
solution proposed by Wienke (2001) follows the triangular shape.
provides much higher magnitude of the  For the breaking wave case under the
impact force. This could be correlated with consideration, the maximum obtained
the shape of the breaking wave under the slamming coefficient is four times lower
consideration. However, because the than analytical proposal by Wienek
computed impact force is dependable on the (2001). Moreover, the obtained
grid size, more validation cases are required slamming coefficient is significantly
before conclusions could be made. lower compared to studies presented in
Table 3. However, as presented model
is found to be grid dependable, a more
5. CONCLUSIONS validation cases are required before
conclusions could be made.
The numerical model based on  For the breaking wave case under the
incompressible Navier-Stokes/VOF solution consideration, the analytical solution
is established. The advantage of the proposed by Wienke (2001) significantly
selected numerical model is fully 3- underestimates the duration of the
dimensional solution of the breaking wave – impact force, which is an important
structure interaction, where the physics can parameter for estimation of the
be evaluated with a high spatial and monopile structure response.
temporal resolution in a controlled
 The future work includes further grid
environment. The presented numerical
sensitivity analysis of the proposed
model could lead to improved design
numerical model. The future work also
methodology and fundamental understanding
includes parametric study regarding the
of the wave slamming forces, which might
influence of the breaking wave shape
results with optimization of offshore wind
on the magnitude and duration of the
turbine foundations and lower costs.
impact force.
Regarding the analysis on selected case,
the conclusions are:
 IMSC 2017 10

REFERENCES Structure. In Proceedings of the Coastal

Engineering Conference (No. 33)
[1] Bihs, H., Kamath, A., Alagan Chella, M., & [12] IEC 61400-3 (2009), International
Arntsen, Ø. A. (2016). Breaking-Wave Electrotechnical Commission
Interaction with Tandem Cylinders under [13] Jacobsen, N. G., Fuhrman, D. R., &
Different Impact Scenarios. Journal of Fredsøe, J. (2012). A wave generation
Waterway, Port, Coastal, and Ocean toolbox for the open source CFD library:
Engineering, 04016005. OpenFoam®. International Journal for
[2] Chan, E. S., Cheong, H. F., & Tan, B. C. Numerical Methods in Fluids, 70(9), 1073-
(1995). Laboratory study of plunging wave 1088
impacts on vertical cylinders. Coastal [14] Kamath, A., Chella, M. A., Bihs, H., &
Engineering, 25(1), 87-107. Arntsen, Ø. A. (2015). Breaking Wave
[3] Chella, M. A., Bihs, H., Myrhaug, D., & Interaction with a Vertical Cylinder and the
Muskulus, M. (2015). Breaking Effect of Breaker Location. CFD based
characteristics and geometric properties of Investigation of Wave-Structure Interaction
spilling breakers over slopes. Coastal and Hydrodynamics of an Oscillating
Engineering, 95, 4-19. Water Column Device, 171.
[4] Chen, G., Kharif, C., Zaleski, S., & Li, J. [15] Morison, J. R., Johnson, J. W., & Schaaf,
(1999). Two-dimensional Navier–Stokes S. A. (1950). The force exerted by surface
simulation of breaking waves. Physics of waves on piles. Journal of Petroleum
fluids, 11(1), 121-133. Technology, 2(05), 149-154
[5] Christensen, E. D. (2006). Large eddy [16] Paulsen, B. T., Bredmose, H., & Bingham,
simulation of spilling and plunging H. B. (2014). An efficient domain
breakers. Coastal Engineering, 53(5), 463- decomposition strategy for wave loads on
485. surface piercing circular cylinders. Coastal
[6] Engsig-Karup, A. P., Bingham, H. B., & Engineering, 86, 57-76
Lindberg, O. (2009). An efficient flexible- [17] Paulsen, B. T., Bredmose, H., & Bingham,
order model for 3D nonlinear water H. B. (2014). An efficient domain
waves. Journal of computational physics, decomposition strategy for wave loads on
228(6), 2100-2118. surface piercing circular cylinders. Coastal
[7] EWEA (2015), Giorgio Corbetta, Andrew Engineering, 86, 57-76.
Ho, Iván Pineda, Wind energy scenarios [18] Ros Collados, X. (2011). Impact forces on
for, 2030. a vertical pile from plunging breaking
[8] Faltinsen, O. (1993). Sea loads on ships waves, Master Thesis
and offshore structures (Vol. 1). [19] Ting, F. C., & Kirby, J. T. (1994).
Cambridge university press. Observation of undertow and turbulence in
[9] Ghadirian, A., Bredmose, H., & Dixen, M. a laboratory surf zone. Coastal
(2016, September). Breaking phase Engineering, 24(1-2), 51-80.
focused wave group loads on offshore [20] Ting, F. C., & Kirby, J. T. (1995). Dynamics
wind turbine monopiles. In Journal of of surf-zone turbulence in a strong plunging
Physics: Conference Series (Vol. 753, No. breaker. Coastal Engineering, 24(3-4), 177-
9, p. 092004). IOP Publishing. 204.
[10] Goda, Y., Haranaka, S., & Kitahata, M. [21] Ting, F. C., & Kirby, J. T. (1996). Dynamics
(1966). Study of impulsive breaking wave of surf-zone turbulence in a spilling
forces on piles. Report Port and Harbour breaker. Coastal Engineering, 27(3-4), 131 -
Technical Research Institute, 6(5), 1-30 160.
[11] Hildebrandt, A., & Schlurmann, T. (2012). [22] Wagner, H. (1932). ÜberStoß‐und
Breaking Wave Kinematics, local Gleitvorgängean der Oberfläche von
Pressures and Forces on a Tripod Support Flüssigkeiten. ZAMM‐Journal of Applied
 IMSC 2017 11

Mathematics and Mechanics/ Zeitschriftfür

Angewandte Mathematik und
Mechanik, 12(4), 193-215.
[23] Wang, Z., Yang, J., Koo, B., & Stern, F.
(2009). A coupled level set and volume-of-
fluid method for sharp interface simulation
of plunging breaking waves. International
Journal of Multiphase Flow, 35(3), 227-
246.
[24] Wienke, J. (2001). Druckschlagbelastung
auf schlanke zylindrische Bauwerke durch
brechende Wellen. Technical University of
Braunschweig, Germany.
[25] Wienke, J., & Oumeraci, H. (2005).
Breaking wave impact force on a vertical
and inclined slender pile—theoretical and
large-scale model investigations.Coastal
Engineering, 52(5), 435-462.
[26] Zhou, D., Chan, E. S., & Melville, W. K.
(1991). Wave impact pressures on vertical
cylinders. Applied Ocean Research, 13(5),
220-23 Zylindrische Bauwerke durch
brechende Wellen. Technical University of
Braunschweig, Germany.