You are on page 1of 19

ITTC – Recommended 7.

5 – 02
07 – 03.8
Procedures and Guidelines Page 1 of 19
Model Tests for Offshore Wind Tur- Effective Date Revision
bines 2017 01

ITTC Quality System Manual

Recommended Procedures and Guidelines

Procedure

Model Tests for Offshore Wind Turbines

7.5 Process Control

7.5-02 Testing and Extrapolation Methods

7.5-02-07 Loads and Responses

7.5-02-07-03 Ocean Engineering

7.5-02-07-03.8 Model Tests for Offshore Wind Turbines

Updated / Edited Approved


Specialist Committee on Testing of Marine
28th ITTC 2017
Renewable Devices of the 28th ITTC
Date: 07/2017 Date: 09/2017
ITTC – Recommended 7.5 – 02
07 – 03.8
Procedures and Guidelines Page 2 of 19
Model Tests for Offshore Wind Tur- Effective Date Revision
bines 2017 01

Table of Contents

2.9 Test Case Parameters ...................... 10


1. PURPOSE OF GUIDELINE ................ 1
2.9.1 Serviceability Limit State tests ... 10
2. TEST PARAMETERS .......................... 3 2.9.2 Survivability Limit State tests .... 10
2.1 Experimental stages ........................... 3 2.9.3 Fatigue Limit State tests ............. 11
2.2 Types of offshore wind turbines ....... 3 2.9.4 Offshore Wind Turbine Arrays .. 11
2.2.1 Bottom-mounted offshore wind 3. DESCRIPTION OF TEST
turbines (BMOWT) ...................... 3 PROCEDURE ...................................... 11
2.2.2 Floating offshore wind turbines
(FOWT) ........................................ 4 3.1 Model & Installation ....................... 11
3.1.1 Platform Model .......................... 11
2.3 Types of facilities suitable for use .... 4
3.1.2 Rotor / Nacelle Assembly Model12
2.4 Model parameters and scale ............. 4 3.1.3 Moorings & Foundation ............. 13
2.5 Uncoupled hydrodynamic tests ........ 5 3.1.4 Installation .................................. 13
2.6 Coupled aero-hydrodynamic tests ... 6 3.2 Calibration of Environment ........... 13
2.6.1 Simplified Simulation of Rotor .... 6 3.3 Collection and analysis of data ....... 14
2.6.2 Physical modelling of the rotor in 3.3.1 Extrapolation to Full Scale ......... 14
fully coupled tests ......................... 7
3.3.2 Presentation of Results ............... 14
2.6.3 Combined real-time numerical
simulation and physical test ......... 7 3.4 Uncertainty Analysis ....................... 15
2.7 Environmental parameters ............... 9 4. REFERENCES .................................... 16
2.8 Mooring Systems ............................. 10
5. ACRONYMS USED IN THIS
DOCUMENT ....................................... 18
ITTC – Recommended 7.5 – 02
07 – 03.8
Procedures and Guidelines Page 1 of 19
Model Tests for Offshore Wind Tur- Effective Date Revision
bines 2017 01

Model Tests for Offshore Wind Turbines

Many aspects of the experiments for


1. PURPOSE OF GUIDELINE FOWT structures are covered by the ITTC
Recommended Procedure 7.5-02-07-03.1,
The purpose of this document is to offer "Floating Offshore Platform Experiments".
guidance to researchers to assist in perform-
ing model tests of offshore wind turbines However, there are some key differ-
(OWTs) according to the state of the art. ences between model tests of OWTs and
These tests may include: model tests of other offshore structures. The
main distinctive features of the tests of
• measurements of foundation loads for OWTs may include:
bottom-mounted (fixed) OWTs;
• measurement of hydro-elastic response • Requirement to simulate complex kine-
of OWT towers; matics and material properties driving
• measurements of global dynamic re- key interactions within the OWT system
sponse of OWTs, including responses to between aero-elastic response of blades,
specified design load cases, measure- dynamics of rotor and nacelle assembly
ment of natural period of motion, and (RNA), hydro-elastic response of tow-
additional damping for floating offshore ers and support structures, gyroscopic
wind turbines (FOWTs); loading, and hydrodynamic response of
• measurement of maximum offset of floating platforms;
moored OWTs, allowing selecting ap- • Special requirements for the model con-
propriate length of the power umbilical; struction, in particular related to the ro-
• investigation of the interaction between tor;
the rotor aerodynamics and the dynamic • Large size of full-scale structures: 7MW
response of the support structure; turbines are approaching 200m in height
• quantification of technical performance with rotor diameters up to 170m;
variables; validation of numerical mod- • Challenges related to dynamic similarity
els; and scaling for aerodynamic and hydro-
• investigation of operational and surviv- dynamic flows for model testing in
ability characteristics limits, including wind/wave tanks;
responses to impulsive loadings, such as • Challenges of accurate simulation of re-
slamming or collision; alistic aerodynamic environment over
• investigation of transportation and in- large areas of test tanks;
stallation methodology • Requirement for analyses of design load
cases from International Electro tech-
This guideline does not address the as- nical Commission (IEC) standards;
sessment of the aerodynamic performance • Rapid evolution of design of FOWTs:
of OWTs. diversity of innovative concepts, some
presenting challenges for scaled model
testing;
ITTC – Recommended 7.5 – 02
07 – 03.8
Procedures and Guidelines Page 3 of 19
Model Tests for Offshore Wind Tur- Effective Date Revision
bines 2017 01

• Requirement for testing throughout the The main objectives of tests in concept
various experimental stages for novel validation stages (TRL 1-3) are to validate
floating concepts: the concept validation the OWT concept, to investigate OWT var-
stage, the design validation stage, the iables and physical properties that affect the
system validation stage, and the proto- performance or energy capture, and to opti-
type and demonstration stage. mize the OWT for power production using
• Potential requirement for tests of multi- small-scale models.
ple scaled models corresponding to an
array of OWTs. The main objectives of tests in the con-
• Particular challenges in simulation of cept development stage (TRL 4-6) are to
unconventional, innovative mooring validate the OWT design, to develop con-
systems. The permanent mooring sys- trol strategies for improved performance,
tems often include synthetic fibre ropes, and to verify the mooring and anchor sys-
which require special treatment in tem using medium scale models. The
model testing. wind/wave spectra at a specific site should
be used. For TRL 4, small-scale models can
These features place particular demands be used; for TRL 5, large-scale model
upon the experiment design, model con- should be applied, while for TRL 6, the tests
struction; facility capability, and experi- can be carried out using scaled prototype in
ment procedure. open environment.

Tests in the prototype demonstration


2. TEST PARAMETERS stage (TRL 7-8) are carried out in the near-
full or full scale in open environment, while
tests in the commercial demonstration and
2.1 Experimental stages
system development stage (TRL 9) are car-
The development of an OWT from the ried out at full scale at sea.
original idea to a marketable product in-
volves a series of test stages including the 2.2 Types of offshore wind turbines
concept validation stage, the design valida-
tion stage, the system validation stage, and The vast majority of OWTs are currently
the prototype and demonstration stage. horizontal axis devices, although some ver-
tical axis devices are also under develop-
These stages are commonly described in ment. OWTs may also be categorised de-
the renewable industry in terms of Technol- pending on the nature of the foundation that
ogy Readiness Levels (TRLs) (e.g. U.S. supports the Rotor-Nacelle Assembly
Department of Energy (2011)); TRL 1-3 (RNA) and the tower.
correspond to research stages up to and in-
cluding proof of concept, TRL 4-6 corre- 2.2.1 Bottom-mounted offshore wind tur-
spond to concept development and scale bines (BMOWT)
testing, TRL 7-8 correspond to prototype
demonstration. TRL 9 correspond to com- Bottom mounted turbines are currently
mercial demonstration and system develop- utilised in water depths up to around 50m.
ment. Bottom-mounted turbine foundations in-
clude the following:
ITTC – Recommended 7.5 – 02
07 – 03.8
Procedures and Guidelines Page 4 of 19
Model Tests for Offshore Wind Tur- Effective Date Revision
bines 2017 01

• Monopile: simple foundation design • Ocean basins capable of generating both


constructed from tubular steel structure, long- and short-crested waves; (includ-
typically used in water up to around ing facilities with wind generation)
25m. • Ocean basins with wind, wave and cur-
• Multipile: foundation based on multiple rent facilities
piles; tripod structures are also a possi- • Shallow water wave tanks.
ble alternative, when the guyed mono-
pile is not feasible; It should be noted that the scale models
• Jacket: braced lattice-frame structure required for OWT testing can place substan-
typically used in deeper water; tial demands on wave-making in terms of
• Gravity base: used when installation of both wave heights, wave periods and run
piles in seabed is difficult. durations and on wind-generating in terms
of wind speed, turbulence intensity and run
2.2.2 Floating offshore wind turbines durations.
(FOWT)
Particular care must be taken to mini-
Floating offshore wind turbine types mise build-up of reflected waves and to
continue to evolve rapidly, but currently in- maintain the quality of wind/wave field dur-
clude the following (Koo et al. (2012)): ing long duration realisations of large waves.

• Barge 2.4 Model parameters and scale


• Spar
• Tension Leg Platform (TLP) The choice of scale ratio will be based
on the OWT size, the goal of the tests, the
• Semi-submersible
target wind/wave conditions, the water
depth and the test stage. It may be necessary
In the FOWT literature (IPCC (2011)),
to build models at different scales to assess
platforms are often categorised as ballast-
performance in operational conditions and
stabilised (e.g. spars), mooring-stabilised
survivability in extreme external conditions.
(e.g. TLP) or buoyancy-stabilised (e.g.
The scale factor will be limited by the
barges or semi-submersibles).
model basin dimensions, its wind/wave
generation capability, the mooring system
2.3 Types of facilities suitable for use to be employed, and the simulation ap-
proach for control strategies.
Different facilities can be used at differ-
ent stages of the design process. These may In small-scale model tests, viscous hy-
include: drodynamic damping and, in particular,
damping associated with vortex shedding
• Wave flumes/Towing tanks with wave- from sharp edges, cannot be scaled appro-
makers (including facilities with wind priately with Froude similarity and may be
generation) overestimated.
• Circulating water channel with wave-
makers (including facilities with wind For an operational FOWT, the gyro-
generation) scopic effect of the combination of the rota-
ITTC – Recommended 7.5 – 02
07 – 03.8
Procedures and Guidelines Page 5 of 19
Model Tests for Offshore Wind Tur- Effective Date Revision
bines 2017 01

tion of the rotor and the pitching of the plat- and the wave/current load as these loads
form in waves results in a yaw moment act- may be considered independently. This may
ing on the structure. The response of the tur- be particularly relevant for simulation of
bine to this yaw moment depends on the dynamic responses of the tower and support
number of degrees of freedom of the model. structure in extreme weather for which the
If the nacelle is fixed to the tower axis (6- turbine will typically be shut down. Hence,
DOF), the yaw moment will result in a yaw in these cases, hydrodynamic tests can be
motion whose magnitude will depend on the carried out without the rotor as long as the
mooring and restoring system in yaw. influence of the rotor mass is correctly rep-
Hence, correct simulation of the yaw stiff- resented (e.g. de Ridder et al. (2011)).
ness of the mooring system as well as the
moments of inertia of the rotor will be im- Decoupling measurements of wave /
portant in order to achieve similarity. current loads from measurement of wind
loads may be advantageous. This is because
On the other hand, when the nacelle part a larger model can be used for the measure-
rotates separately from the tower (7-DOF), ments of wave/current loads, facilities with-
the observed yaw motion will depend on the out wind generation may be employed, and
correct simulation of the moments gener- measurement uncertainty may be reduced
ated between the nacelle and the tower since the entire range of the load cells can
(Wang & Sweetman (2012)). be used for measuring the wave/current
loads. However, where tests are aimed at in-
Some particularly complex phenomena, vestigating the coupled dynamic response
which represent a challenging research area, of the structure in operational conditions,
are involved with the interaction of including realistic modelling of flexibility
wind/wave/current flow and a FOWT. and aerodynamic damping, then inclusion
Mean offsets, including trim, list and azi- of the aerodynamic coupling due to the ro-
muth angles in rotational modes and drifts tor is necessary. Moreover, investigation of
in translational modes, may be caused by the dynamic response requires modelling of
the second-order effect of waves, waves the soil-structure interaction, as it influ-
trapped between columns and pontoons, as ences the loads and especially the natural
well as mean component of wind speed and frequencies of the structure (Abhinav, Saha,
current. The mean offsets may be detri- 2015).
mental to power generation performance
and seakeeping performance, including the Model tests of FOWTs can be carried
stability in waves and wind. out without the rotor at preliminary stage of
the tests or for special purposes, e.g. com-
The offsets can be adjusted by the water paring different support structures in re-
ballast system and/or the mooring system spect of response to waves, validation of nu-
with delta connections in the mooring lines. merical models etc. However, final tests
aiming in evaluating the global response of
2.5 Uncoupled hydrodynamic tests the system from the concept validation
stage to the prototype and demonstration
In tests of BMOWTs focused on deter- stage should include at least simplified
mination of loads, it may not be necessary
to simulate the coupled effects of wind load
ITTC – Recommended 7.5 – 02
07 – 03.8
Procedures and Guidelines Page 6 of 19
Model Tests for Offshore Wind Tur- Effective Date Revision
bines 2017 01

modelling of the rotor due to strong cou- The disc should be sized to generate a drag
pling between rotor and platform dynamics, load corresponding to the thrust on the tur-
and in particular the gyroscopic effect. bine. If a rotating disc or a separate rotating
arm is employed with the correct rotary mo-
In facilities without current generation, ment of inertia, it is possible with the correct
using a set of linear springs inserted into a moment of inertia, to capture the coupled
mooring cable and a pulley system at the response of the structure taking into account
fairlead point for simulation of steady cur- the gyroscopic coupling between the rotor
rent load may be employed (Chakrabarti and the platform (see Cermelli et al. (2009)).
(1998)). The initial fairlead angles and pre-
tension at top end of mooring cable are ad- This approach neglects the aerodynamic
justed to match the calculated ones. Then, torque and other smaller aerodynamic loads
steady current loads can be calculated by the in yaw, sway and roll exerted by the rotor
empirical formulas in RCS rules including on the platform as well as blade / tower in-
API-RP2SK (2005) and depend on the ge- teractions; problems may result due to the
ometry of the exposed structural compo- unsteadiness of the flow around the disc. In
nents and types of floaters. However, this addition, this may not be able to capture the
method will not account for the wave-cur- slope of the thrust curve (i.e. the thrust co-
rent interaction. The pulley system will also efficient vs. tip speed ratio curve), which is
lead to incorrect changes of the static con- important for aerodynamic damping.
figuration of mooring lines in the catenary
spread mooring system A further possibility, which may be suit-
able for small-scale tests in the concept val-
idation stage, is to use the rotor as a fan ro-
2.6 Coupled aero-hydrodynamic tests
tating in otherwise stationary air (e.g.
Kraskowski (2012)). This offers a rather
2.6.1 Simplified Simulation of Rotor simplified approach to the investigation of
response of FOWTs in facilities that do not
A number of methods may be employed have wind generation capabilities. In this
to simulate the presence of the rotor without case, separate measurements are required to
using an accurate representation of the rotor calibrate the system, i.e. to identify the force
aerodynamics, although none captures all of vs. rpm characteristics.
the physics of the fully coupled system.
This method of modelling the rotor is
Simulating the steady wind load using a quite simple and allows for easy adjustment
wire attached to a weight will lead to the in- of the mean wind load. However, it is diffi-
correct inertia of the system during testing cult with this approach to control the blade
(Chakrabarti (2005)). It can only be justified pass frequency and wind load simultane-
for rough estimation of the maximum moor- ously to achieve the correct mean thrust and
ing offset (e.g. Chujo, et al. (2011)). How- torque whilst capturing tower interaction ef-
ever, using wires attached to dynamic fects. Further challenges of this approach
winches is recommendable (ISSC (2012)) include the correct simulation of orientation
and an example is shown in 2.6.3. of gyroscopic moments in relation to steady
moments, and the difficulty in realistically
A solid disc may be used in place of the
simulating the behaviour of the magnitude
rotor in conjunction with a battery of fans.
ITTC – Recommended 7.5 – 02
07 – 03.8
Procedures and Guidelines Page 7 of 19
Model Tests for Offshore Wind Tur- Effective Date Revision
bines 2017 01

and direction of the thrust vector as the tur- duction of the mean wind thrust load in or-
bine pitches. It is also difficult to model the der to generate correct aerodynamic over-
correct torque and other smaller aerody- turning moments and mooring offsets. The
namic loads, as well as the correct slope of impact of rotor aerodynamics on pitch
the thrust curve. damping is also of great importance. Main-
taining the Reynolds similarity is in general
The simplified methods of modelling not possible for typical sizes of basin mod-
the rotor described above are reasonably els, and thus detailed modelling of aerody-
well suited for preliminary tests in steady namics, including stall phenomena, is usu-
wind; however, particular care is required in ally not possible. Variations in wind speed
the interpretation of results from these types caused by motions of a floating platform
of tests for turbulent wind and extreme will be driven by wave effects, which are
events. A more sophisticated approach, uti- governed by Froude similarity.
lising the actively controlled fan for simu-
lating the rotor thrust obtained from simul- Depending on the required outcome of
taneous numerical simulations, is described the tests, modelling the rotor will usually
in 2.6.3. also require maintaining the Froude similar-
ity for the rotor RPM to generate the correct
2.6.2 Physical modelling of the rotor in representation of the gyroscopic effect of
fully coupled tests the rotor as well to allow more accurate rep-
resentation of the aerodynamic interaction
Direct modelling of an OWT rotor is between the rotor and the support structure.
usually realized by exposing a working, but This will also involve realistic representa-
not necessarily geometrically scaled, rotor tion of the mass distribution and possibly
to a wind field generated by a battery of fans the elasticity of supporting structure and ro-
(see for example Chujo, et al. (2011) for tor blades.
spar OWT, Shin, et al. (2013) for semi-sub-
mersible OWT and Goupee, et al. (2012), Performance models of OWTs will
for spar, semi-submersible and TLP). The therefore normally be scaled using Froude
rotor rpm and the spatial variation of wind similitude. However some key parameters
speed should be carefully calibrated prior to related to wind loading will not scale in this
the main experiments. manner, leading to scale effects when ex-
trapolating to full-scale, particularly for
Particular challenges in this approach FOWTs. Approaches to address this
with respect to the wind generation include through redesign of the rotor model are dis-
the representation of wind gradients, the cussed in more detail in 3.1.2.
wind turbulence and the difficulty of gener-
ating wind in a wave tank close to a wavy 2.6.3 Combined real-time numerical sim-
water surface, particularly in tests with large ulation and physical test
waves.
This class of tests refers to methods that
The minimum aerodynamic requirement combine, in real-time and interactively, nu-
for modelling the presence of rotor in a fully merical simulation of a virtual substructure
coupled test of a FOWT is the correct repro- with a physical substructure tested experi-
mentally in model scale. These hybrid
ITTC – Recommended 7.5 – 02
07 – 03.8
Procedures and Guidelines Page 8 of 19
Model Tests for Offshore Wind Tur- Effective Date Revision
bines 2017 01

methods often go by the terms real-time hy- moments, etc.) and the frequencies of inter-
brid model testing, hardware-in-the-loop, est, i.e. the frequency range of the quantities
or software-in-the-loop. For model testing that have to be captured correctly by the ex-
of offshore wind turbines in hydrodynamic periment. This, together with the magnitude
laboratories, this means that the platform re- of the loads and responses, will govern the
sponses (motions, etc.) are measured exper- selection of actuators and the control strat-
imentally and passed into the numerical egy of the hybrid setup. Further, a numeri-
simulations, whereas actuators, or other cal sensitivity analysis should be performed
means, apply the appropriate aerody- to identify which components of the aero-
namic/generator loads according to simulta- dynamic load vector that has insignificant
neous simulations of the wind turbine. effect on the quantities of interest (see
Bachynski et al. (2015) and Hall (2014)).
One important advantage with this The complexity of the hybrid setup may be
method, adapting a "virtual turbine", is that reduced by removing load components that
it solves Froude-Reynolds scaling conflict, induce insignificant responses. When the fi-
i.e. the aerodynamic loads are calculated in nal setup is developed, a numerical tool
full-scale and are then scaled down using modelling the entire experimental setup, in-
Froude scaling. A well-functioning hybrid cluding the actuators and control system,
approach allows for investigating the re- should carry out a virtual testing of the sys-
sponses of OWT in operational and survival tem, to verify that the system is performing
conditions, as well as in fault conditions as designed. A suggested summary of the
such as transient responses in emergency procedure can be found in Sauder et al.
shutdown of the generator in large wind (2016). There are different ways of applying
speeds or for blade seize (e.g. loss of pitch the aerodynamic loads on the physical sub-
control for one blade). Also, it can be used structure, and a couple of approaches are
to perform detailed sensitivity studies on the described in the following.
wind modelling, e.g. including the effect of
large waves on the wind field - which is dif- A simple approach that may be adequate
ficult to setup accurately in a hydrodynamic for some concepts is to replace the rotor
laboratory. The aerodynamic simulation with a ducted fan driven by an electronic
tool inherently captures aerodynamic damp- motor (Azcona et al. (2014)). The fan can
ing. The hybrid approach has also the ad- generate a force representing the thrust
vantage that the aerodynamic loads in the force on the turbine, obtained from simulta-
tests are known, and hence, the uncertain- neous numerical simulations. The simula-
ties related to physical modelling of the tions can include effects such as turbulence,
wind and the turbine are eliminated, which gusts, aerodynamic damping, and wind tur-
makes it suitable for calibration and valida- bine control. This approach focuses on the
tion of hydrodynamic coefficients in nu- application of the thrust force only. It
merical tools (Berthelsen et al. (2016)). should therefore be investigated whether
the neglected aerodynamic load compo-
When preparing the hybrid setup it is nents have negligible influence on the quan-
particularly important to identify the quan- tities of interests.
tities of interest for the tests (e.g. motions,
mooring lines tensions, tower-base bending It may be necessary for some concepts
to apply a larger number of components of
ITTC – Recommended 7.5 – 02
07 – 03.8
Procedures and Guidelines Page 9 of 19
Model Tests for Offshore Wind Tur- Effective Date Revision
bines 2017 01

the calculated aerodynamic loads, in order 2.7 Environmental parameters


to obtain correct estimates of the quantities
of interest. The physical substructure may A discussion of key parameters related
then be connected to several land-based dy- to environmental properties such as water
namic winches (actuators) as described in depth, basin dimensions, calibration of
e.g. Sauder et al. (2016). Bachynski et al. wave characteristics (and current and wind
(2016) provides examples of tests that are where relevant), and combined environ-
feasible with this approach. The drawback ment characteristics can be found in ITTC
of this method is the need for more hard- Procedure 7.5-02-07-03.1.
ware, and advanced control and allocation
strategies. In OWTs wind/wave tank testing, par-
ticular attentions should be paid to the im-
One important challenge with real-time pact of open air and wave blockage, since
hybrid testing is related to time-delays from OWTs naturally affect the wind and wave
numerical simulation, data transfer and the field in a more complex manner than con-
actuator response. Typically, the time-delay ventional offshore floating structures.
may introduce additional damping or spuri-
ous energy that may cause instabilities. A Testing in long-crested waves and/or
delay compensation strategy is important to uniform wind is commonly adopted at the
prevent that any unphysical damping or in- concept validation stage, for comparative
stability appears in the experiment. Another studies, and for component testing where
challenge is related to the physical limita- appropriate. This process may include tests
tion of the actuators to emulate high fre- with the OWT oriented at different angles to
quency loads that may be important for cer- the direction of wave propagation. Concept
tain types of OWT (e.g. TLP's and mono- validation testing may involve regular wave
piles). tests with/without uniform wind to charac-
terise frequency response as well as testing
The accuracy of the aerodynamic loads in irregular sea states with/without turbulent
is limited by the simplifications and uncer- wind appropriate to the intended deploy-
tainties in the numerical model. Uncertain- ment site in order to estimate performance
ties can be related to both the input wind including dynamic responses.
field description (e.g. lack of proper full-
scale data) as well as the aerodynamic load At later stages of the design process,
modelling (e.g. simplifications and assump- when accurate estimates of performance in
tions, accuracy of numerical solvers). On combined external conditions are required,
the other hand, hybrid testing allows for full since the performance of OWTs depends
control of the aerodynamic loads, i.e. the ac- greatly on both incident wave direction and
tual wind loads applied in the experiments misalignment in directions of wind and
are known, and uncertainties in the numeri- wave, tests in short-crested irregular waves
cal modelling can to some extent be evalu- considering the azimuth of the principal
ated qualitatively and quantitatively by nu- wind/wave direction should be conducted.
merical sensitivity studies.
For these tests, the directional wave
spectral density function can be used to de-
scribe the short-crested waves. It is defined
ITTC – Recommended 7.5 – 02
07 – 03.8
Procedures and Guidelines Page 10 of 19
Model Tests for Offshore Wind Tur- Effective Date Revision
bines 2017 01

as the product of the wave frequency spec- Guidance on the use of a hybrid mooring
trum and the directional spreading function. system may be found in the ITTC Recom-
The most popular model for the directional mended Procedure 7.5-02-07-03.4, "Sta-
spreading is a cosine squared (cos2s) func- tionary Floating Systems Hybrid Mooring
tion originally proposed by Longuet-Hig- Simulation".
gins et al. (1963). It is recommended (Goda
1985) to use s = 10 for wind driven seas, 2.9 Test Case Parameters
s = 25 for swell with short decay and s = 75
for swell with long decay distance.
2.9.1 Serviceability Limit State tests
Alternatively, the wave spectrum of an
actual site may also be used in the tests. Site In the tests of serviceability limit state
data could suggest that sea states composed performance (normally limits on operating
of multiple wave systems are common at condition), the ability of the OWT to cap-
that particular location. When device per- ture and convert the wind energy is regarded
formance can be compromised by multi di- as the most important criterion.
rectionality, testing in sea states with multi-
Tests on the serviceability limit state
ple wave systems should be carried out.
performance should be carried out in both
regular and irregular waves with/without
2.8 Mooring Systems wind considering turbulence. The test pro-
grammes should aim at investigating the ef-
FOWT concepts utilise a range of moor- fect of OWT design variables on limit state
ing systems including single point and performance. Details of the design load
spread moorings as well as catenary, taut, cases under combined environmental condi-
TLP and multi-element systems. Where de- tions can be found in IEC Standards 61400-
tailed design information is available it is 3 for OWT and IEC Technical Specifica-
important to simulate moorings accurately, tions IEC 61400-3-2 for FOWT.
since mooring behaviour can affect both
power capture and extreme behaviour. This Model tests in irregular waves
is especially relevant where taut moorings with/without wind considering turbulence
are employed since these can have a signif- should normally be carried out for a dura-
icant impact on FOWT motions. tion corresponding to at least 60 minutes at
full scale in order to gain statistically valid
Guidance on mooring installation and results. Details of procedures for simulation
calibration can be found in ITTC Proce- and measurement of irregular short-crested
dures 7.5-02-07-03.1. In the case of FOWTs seas can be found in the ITTC Recom-
using catenary moorings, the footprint size mended Procedure 7.5-02-07-01.1, "Labor-
at the scale resulting from the maximum ca- atory Modelling of Multidirectional Irregu-
pability of the wave-makers may exceed the lar Wave Spectra".
size of the tank. Where the limitations on
the physical size of a testing basin do not
allow a full model of a mooring to be ac- 2.9.2 Survivability Limit State tests
commodated at a reasonable scale within
Before undertaking sea trials, it is im-
the basin, truncated systems or hybrid mod-
portant to conduct survivability (normally,
elling is required (e.g. Kraskowski (2012)).
ITTC – Recommended 7.5 – 02
07 – 03.8
Procedures and Guidelines Page 11 of 19
Model Tests for Offshore Wind Tur- Effective Date Revision
bines 2017 01

extreme condition) tests in model basins, to to the significant effect it may have on aer-
evaluate the seaworthiness of an OWT in- odynamic damping in the wave direction.
cluding hull structure and mooring system. The reduced aerodynamic damping may in-
The survivability tests should be conducted crease the wave induced fatigue damage.
in long and short crested irregular waves Sensitivity to the turbulence model and tur-
with extreme wind considering both gust bulence intensity should also be investi-
and turbulence. These tests must provide gated as this may have an impact on the fa-
extreme motions, extreme loads exerted on tigue damage, e.g. on mooring system for
the hull structure, shutdown and mooring FOWTs.
line loads under the design conditions cor-
responding to the metocean data of the in- Structures with high natural frequencies
stallation site. Tests should follow the prin- should also be investigated in operational
ciples set out in the ITTC Recommended sea states with wave periods near the struc-
Procedure 7.5-02-07-02.3, "Experiments on tural natural periods to investigate possible
Rarely Occurring Events". resonance responses.

Survivability tests are typically carried 2.9.4 Offshore Wind Turbine Arrays
out for a duration corresponding to three
hours at full scale. A series of wind and Testing of arrays can present substantial
wave angles should be used to evaluate their challenges for many OWT types especially
effect on OWT motion and mooring forces. when realistic mooring systems are de-
Tests involving failure modes with one or ployed, due to the footprint required, and
more mooring lines disconnected should be the potential importance of interactions on
carried out to simulate line-breaking scenar- mooring and foundation loads.
ios. The test matrix can be considerably re-
duced in cases where the most dangerous For an array with many OWTs installed,
wave direction in respect of mooring loads the interaction of OWTs can be determined
can be reliably identified. through tests involving a limited number of
systems. Due to the cost and scale con-
Where appropriate the OWT should be straints, the behaviour of arrays involving a
tested in design situations other than power large number of OWTs may be evaluated by
production (i.e. occurrence of fault, stand- numerical modelling.
ing still, idling, start up and shut down) to
simulate typical scenarios which could re-
sult in excessive body motions. 3. DESCRIPTION OF TEST
PROCEDURE
2.9.3 Fatigue Limit State tests
3.1 Model & Installation
Data from regular/irregular wave tests
with wind may be used to inform the esti- 3.1.1 Platform Model
mation of fatigue limit states.
Guidance on preparing the model of a
Wave and wind misalignment in opera- FOWT platform including model geometry,
tional conditions should be investigated due ballasting and loading can be found in ITTC
Procedure 7.5-02-07-03.1.
ITTC – Recommended 7.5 – 02
07 – 03.8
Procedures and Guidelines Page 12 of 19
Model Tests for Offshore Wind Tur- Effective Date Revision
bines 2017 01

In case of ballast-stabilised floating Manufacture of a rotor with the correct


structures, the design and manufacturing of mass properties and adequate stiffness can
the model of OWT may be more demanding prove very challenging. Martin et al. (2012)
than the corresponding process for models describe a test of a 1/50 scale model of a
of vessels or offshore platforms due to the 5MW turbine involving construction of a
extreme sensitivity of draught to accuracy blade 1.23m in length with a mass of 140g.
of ballasting and the limited possibility of Muthanna et al. (2013) discuss challenges
adjusting the mass distribution for correct in manufacture of small-scale rotor models.
reproduction of the moments of inertia.
Maintaining the correct aerodynamic
It should be also noticed that small water behaviour of the rotor is a substantial chal-
absorption or deformation due to hydro- lenge in fully coupled model tests. If Froude
static pressure can influence the model scaling is adopted for the rotor rpm, in order
draught and mass distribution. For that rea- to maintain the gyroscopic moments, then
son, it is recommended that the total mass use of Froude-scaled wind speed in con-
and mass distribution of the model are taken junction with an accurate geometric model
into account as parameters at the model de- of the rotor will generally lead to unrealisti-
sign stage, so as to minimize the need of cally low rotor torque and thrust due to the
ballasting the finished model. reduced Reynolds number. This is because
the foil sections typically utilised in OWTs
3.1.2 Rotor / Nacelle Assembly Model will exhibit substantially reduced lift and in-
creased drag compared to the full-scale foils
The Rotor-Nacelle Assembly (RNA) at the low Reynolds numbers generated by
and associate instrumentation must be care- Froude-scaled wind.
fully considered. Special care should be
taken where flexible models of components Martin et al. (2012) discuss three possi-
such as blades are constructed; for flexible ble approaches to address this challenge. In
models, it is important to scale the material the first approach, the wind speed is in-
properties correctly to achieve the correct creased beyond the Froude-scaled value to
elastic behaviour at model scale. compensate for the low thrust coefficient. If
rotor speed is maintained at Froude-scaled
During model tests with a working rotor, values, to retain correct gyroscopic mo-
the rotation of the rotor can sometimes re- ments, then the tip-speed ratio will be incor-
sult in vibration. The mass and stiffness rect, resulting in incorrect torque. However,
characteristics of the components of the this may be justified as an approximation
model change the degree and the position of since the overturning moment due to thrust
the vibration. Hence where the model con- is typically very much greater than that due
struction does not allow all aspects of simi- to torque. The ratio of unsteady velocity
larity to be maintained simultaneously, the (caused by platform motions) to mean ve-
priority of similarity, (i.e. the mass distribu- locity will be reduced leading to incorrect
tion, inertia distribution or distribution of modelling of effects of unsteady inflow on
the elasticity of supporting structure and the rotor. However, results show that the
blades) should be selected depending on the aerodynamic damping of the platform pitch
main purpose of the test. generated by the turbine is modelled with a
reasonable degree of accuracy.
ITTC – Recommended 7.5 – 02
07 – 03.8
Procedures and Guidelines Page 13 of 19
Model Tests for Offshore Wind Tur- Effective Date Revision
bines 2017 01

A second approach addressing low 3.1.3 Moorings & Foundation


Reynolds number effects is the placement
of studs or other roughened materials as a In the measurements of foundation loads
turbulence stimulator along the leading due to waves for bottom-mounted OWTs, it
edge of a blade. However, this is unlikely to is important to pay attention to the stiffness
improve the turbine performance ade- of the measurement devices; unrealistically
quately on its own to yield comparable per- flexible foundation of the model can influ-
formance with the full-scale device, and can ence the resulting wave loads.
yield unrealistic results if laminar separa-
tion occurs, as well as unrealistic unsteady In case of mooring systems utilising
aerodynamic loads during flow re-attach- synthetic fibre ropes, special care should be
ment. taken with correct modelling of their stiff-
ness during the tests. Viscoelastic properties
A third possible approach is to redesign of the material result in increased stiffness
the rotor blade sections to account for Reyn- of the ropes under dynamic loads, which
olds number effects, or even more radical should be taken into account in model tests
solutions such as changing the number of (Falkenberg (2011)). For ropes character-
blades and the rotor diameter. This can in- ized by increased stiffness under dynamic
volve choice of laminar flow sections for loads, it is usually not possible to reproduce
the model scale rotor so that the model rotor correctly both the maximum mooring loads
design can simulate as closely as possible and the mean offset of the structure.
the correct full-scale mean thrust and torque
coefficients at the model-scale Reynolds 3.1.4 Installation
Number (based on blade chord), whilst still
maintaining the correct mass properties. Model preparation and installation
Martin et al. (2012) demonstrate an example should follow the principles set out in ITTC
showing blade redesign leading to broadly Procedure 7.5-02-07-03.1. The installation
correct values of scaled thrust and aerody- of small-scale testing of moored structures
namic damping using Froude-scaled wind or free-floating structures should be clearly
speed. documented as installation approach could
impact motions or loading.
In order to minimize these errors, tests
with large scales are recommended where
3.2 Calibration of Environment
possible.
Details of the calibration of environment
Complete modelling of the RNA with
parameters can be found in the ITTC Proce-
respect to its influence on global response
dure 7.5-02-07-03.1. In testing offshore
of the FOWT includes actual representation
wind turbines including direct modelling of
of the blade pitch control system (e.g. Chujo,
the rotor, particular attention must be paid
et al. (2013)). Neglecting the influence of
to correct representation of the wind field,
blade pitch control can result in underesti-
which should be measured prior to main ex-
mation of the pitch angle of the floater
periments and documented.
(Wang & Sweetman (2011)). Simplifica-
tions assumed for the tests should then be
carefully studied and documented.
ITTC – Recommended 7.5 – 02
07 – 03.8
Procedures and Guidelines Page 14 of 19
Model Tests for Offshore Wind Tur- Effective Date Revision
bines 2017 01

3.3 Collection and analysis of data 3.3.2 Presentation of Results

General guidance on collection and The following provides a recommended


analysis of data can be found in the ITTC outline of a generic test procedure and re-
Procedure 7.5-02-07-03.1. In case of OWTs, port. An actual test procedure is likely to
the accelerations at key locations in full consist of a sub-set of these elements, and
scale are critically important parameters in may vary dependent on the test purpose and
the operation and maintenance of the sys- device type.
tems, so particular care should be taken with
the collection and analysis of these data. a) Purpose of the Test

b) Facility Characterization
3.3.1 Extrapolation to Full Scale i) No-model baseline performance
ii) Facility dimensions and model size
Model values of forces and motions are
capacity
scaled to full scale by applying Froude’s si-
iii) Operating ranges and test capabili-
militude law. Special treatment may be re-
ties
quired to address the challenges posed by
difficulties in reproducing the vertical wind c) Model & Installation
speed distribution correctly in model tests. i) Model Scale
ii) Model Complexity – simplified,
Particular care must be taken to account
system, component
for the relationship between the mean
iii) Model function/operation
torque and thrust and the dynamic forces
iv) Model installation: Mooring,
and moments, with regard to the impact of
Foundation and constraints
gyroscopic effects.
v) Model Measurements / calibration
Extrapolation of model test results to
d) Measurement Systems
the full-scale will require some empirical
i) Purpose of the measurements and
correction methods taking into considera-
required performance/accuracy
tion the correlation allowance in order to
ii) Instrumentation Type: Invasive /
make up for the modelling deficiencies of
non-invasive; embedded / free-
the small scale model test. Due to relatively
field; Steady / dynamic; Opera-
early stage of development of the method-
tional characteristics and require-
ology for scale model tests of offshore wind
ments
turbines and different testing methods being
iii) Resolution – Spatial and temporal
in use, feedback from full scale devices
iv) Calibration requirements
should be used to elaborate empirical cor-
rections specific for device type and testing e) Types of Measurements
method in order to make up for the model- i) Model motion and deformation
ling deficiencies in the small scale model ii) Flow field measurements
test. iii) RNA measurements
iv) Environmental measurements

f) Test Matrix
ITTC – Recommended 7.5 – 02
07 – 03.8
Procedures and Guidelines Page 15 of 19
Model Tests for Offshore Wind Tur- Effective Date Revision
bines 2017 01

i) Test parameters and conditions – (mass, moment of inertia, RPM, and blade
Scaling parameters (Fr, Re, etc.), pitch angles) should be documented. Other
operating conditions potential sources of uncertainty specific to
ii) Measurement Locations FOWTs are:
iii) Recommended practices
iv) Design and off-design testing Sensitivity of the motion response char-
(specify what is meant by off-de- acteristics to mass distribution and, on the
sign testing – e.g. is platform / ro- other hand, limited possibility of adjusting
tor yaw regarded as off-design). the mass distribution;
v) Steady vs. unsteady performance
vi) Shutdown conditions Sensitivity of the motion response char-
vii) Testing in waves for floating de- acteristics to additional inertia of a bundle
vices of instrument cables hung from RNA, on
viii) Component & Sub-component the other hand, limited possibility of reduc-
testing: Component and system ing the weight of instrument cables;
loading; Sub-component function
ix) Test repeatability and required Sensitivity of the response characteris-
number of repeat conditions for de- tics to the accurate installation of FOWTs’
sired accuracy scaled models including the mooring sys-
x) Installation & Recovery tests tem with anchors;

g) Data Acquisition Sensitivity of the motion response char-


i) System performance – rates, reso- acteristics to viscous damping of FOWTs
lution, sequential or simultaneous around the resonance range, on the other
sampling, number of channels, and hand, limited possibility of matching the
noise levels/floors damping at low Reynolds number;

h) Data Analysis In the case where it is required to model


i) Data corrections – bias errors, a mooring system consisting of synthetic fi-
blockage corrections, Zeroes or bre ropes - large sensitivity of extreme
Tares mooring loads to correct reproducing the
ii) Normalizations stress-strain characteristics of the material.
iii) Statistical Analyses; static vs. dy- Dynamic characteristics of the synthetic
namic studies ropes and effects of the delta connection on
iv) Uncertainty analyses yaw should be taken into account.

Standard aspects of valuation and ex-


3.4 Uncertainty Analysis pression of uncertainty can be found in the
ITTC Recommended Procedure 7.5-02-01-
The most important potential source of 01, "Guide to the Expression of Uncertainty
uncertainty in model tests of the OWTs is in Experimental Hydrodynamics".
the accuracy of modelling the rotor, as dis-
cussed in 2.5. Reproduction of its damping
characteristics, inertia and angular momen-
tum is recommended whenever possible;
the characteristics of rotor at model scale
ITTC – Recommended 7.5 – 02
07 – 03.8
Procedures and Guidelines Page 16 of 19
Model Tests for Offshore Wind Tur- Effective Date Revision
bines 2017 01

4. REFERENCES Cermelli, C., Roddier, D. and Aubault, A.,


2009, "Windfloat: A Floating Founda-
Abhinav K.A., Saha N., 2015, "Coupled hy- tion for Offshore Wind Turbines Part II:
drodynamic and geotechnical analysis Hydrodynamic Analysis", Proc. 28th Int.
of jacket offshore wind turbine", Soil Conf. on Ocean, Offshore and Arctic
Dynamics and Earthquake Engineering Engineering, OMAE2009-79231 Hono-
73, 66–79 lulu, Hawaii, USA .

API-RP2SK, 2005, "Design and Analysis of Chakrabarti, S.K., 2005, "Handbook of Off-
Stationkeeping Systems for Floating shore Engineering", Elsevier, pp. 1024.
Structures", Third Edition, American
Petroleum Institute Chakrabarti, S.K., 1998, "Physical Model
Testing of Floating Offshore Structures",
Azcona J., Bredmose H., Campagnolo F., Dynamic Positioning Conference, pp.
Manjock A., Pereira R., and Sander F., 16-17
2014, "Methods for performing scale-
tests for method and model validation", Chakrabarti, S.K., 1987, "Hydrodynamics
Technical report, INNWIND.EU Deliv- of Offshore Structures", Computational
erable D4.22. Mechanics Publications.

Bachynski, E. E., Chabaud, V., and Sauder, Chujo T., Ishida S., Minami Y., Nimura T.,
T., 2015, “Real-time Hybrid Model and Shunji Inoue S., 2011, "Model Ex-
Testing of Floating Wind Turbines: Sen- periments on the Motion of a SPAR
sitivity to Limited Actuation”, In En- Type Floating Wind Turbine in Wind
ergy Procedia, 80, pp. 2–12. and Waves", Proc. 30th Int. Conf. on
Ocean Offshore & Arctic Eng., OMAE
Bachynski E.E., Thys M., Sauder T., 2011-49793 Rotterdam, The Nether-
Chabaud V., and Sæther L.O., 2016, lands.
"Real-time hybrid model testing of a
braceless semi-submersible wind tur- Chujo T., Ishida S., Minami Y., Nimura T.,
bine. Part II: Experimental results", Proc. 2013, "Experimental Study for Floating
35th Int. Conf. on Ocean, Offshore and Offshore Wind Turbine with Blade
Arctic Engineering OMAE2016, Busan, Pitch Control", Proceedings of the Inter-
Korea. national Symposium on Marine and Off-
shore Renewable Energy. Oct 28-30,
Berthelsen P.A., Bachynski E.E., Karimirad Tokyo, Japan.
M., and Thys M., 2016, "Real-time hy-
brid model testing of a braceless semi- De Ridder, E., Aalberts, P. van den Berg, J.
submersible wind turbine. Part III: Cali- Buchner, B, Peeringa, J. 2011, "The Dy-
bration of a numerical model", Proc. namic Response of an Offshore Wind
35th Int. Conf. on Ocean, Offshore and Turbine with Realistic Flexibility to
Arctic Engineering OMAE2016, Busan, Breaking Wave Impact", Proc. 30th Int.
Korea. Conf. on Ocean Offshore & Arctic Eng.
OMAE 2011-49563 Rotterdam, The
Netherlands.
ITTC – Recommended 7.5 – 02
07 – 03.8
Procedures and Guidelines Page 17 of 19
Model Tests for Offshore Wind Tur- Effective Date Revision
bines 2017 01

Goda, Y., 2010, "Random Seas and Design Kim, H. C., 2008, "Nonlinear Water Waves
of Maritime Structures", World Scien- and Offshore Structures", World Scien-
tific. tific.

Falkenberg E. et al, 2011, "Global perfor- Koo B., Goupee A.J., Lambrakos K.F., and
mance of synthetic rope mooring sys- Kimball R.W., 2012, "Model Tests for a
tems – frequency domain analysis", Floating Wind Turbine on Three Differ-
Proc. 30th Int. Conf. on Ocean Offshore ent Floaters", Proc. 31st Int. Conf. on
& Arctic Eng., OMAE 2011-49723 Rot- Ocean Offshore & Arctic Eng.,
terdam, The Netherlands OMAE2012-83642 83645 Rio de
Janeiro, Brazil.
Goupee A.J., Koo B., Kimball R.W., Lam-
brakos K.F., and Dagher H.J. 2012, "Ex- Kraskowski, M., Zawadzki, K., and Rylke,
perimental Comparison of Three Float- A., 2012, "A Method for Computational
ing Wind Turbine Concepts", Proc. of and Experimental Analysis of the
the 31th Int. Conf. on Ocean Offshore & Moored Wind Turbine Sea-keeping",
Arctic Eng., OMAE2012-83645 Rio de Proc. 18th Australasian Fluid Mechan-
Janeiro, Brazil. ics Conference, Launceston, Australia.

Hall M., Moreno J., and Thiagarajan K., Longuet-Higgins, M.S., Cartwright, D.E.,
2014, "Performance specifications for Smith, N.D., 1963, "Observations of the
real-time hybrid testing of 1:50-scale directional spectrum of sea waves using
floating wind turbine models", Proc. motions of a floating buoy", In Ocean
33rd Int. Conf. on Ocean, Offshore and Wave Spectra. Prentice Hall, New York,
Arctic Engineering OMAE2014, San pp.111–136.
Francisco, California, USA.
Mankins, John C., 1995, "Technology
IEC 61400-3, 2009, "Wind Turbines – Part Readiness Levels: A White Paper".
3: Design requirements for offshore NASA, Office of Space Access and
wind turbines", International Electro- Technology.
technical Commission.
Martin H.R., Kimball R.W., Viselli A.M.,
IEC 61400-3-2, 2014, "Wind Turbines – and Goupee A.J., 2012, "Methodology
Part 3-2: Technical specifications for for Wind/Wave Basin Testing of Float-
floating offshore wind turbines", Inter- ing Offshore Wind Turbines", Proc. 31st
national Electrotechnical Commission Int. Conf. on Ocean Offshore & Arctic
Eng., OMAE2012-83627 Rio de Janeiro,
IPCC, 2011, "Special Report on renewable Brazil.
Energy Sources and Climate Change
Mitigation", Intergovernmental panel on Muthanna, C. Visscher, J. H., and Egeberg,
Climate Change, pp.101. T. F., 2013, "An Experimental Investi-
gation of Scale Effects on Offshore
ISSC, 2012, "Committee 1.2: Loads", Pro- Wind Turbine Blades", Proc. 32nd Int.
ceedings of 18th International Ship and Conf. on Ocean, Offshore and Arctic
Offshore Structures congress, Vol.1 Engineering OMAE2013, Nantes,
Rostock, Germany, pp.94. France.
ITTC – Recommended 7.5 – 02
07 – 03.8
Procedures and Guidelines Page 18 of 19
Model Tests for Offshore Wind Tur- Effective Date Revision
bines 2017 01

Nielsen, K., 2003, "Development of Rec- Wang, L., Sweetman, B., 2011, "Concep-
ommended Practices for Testing & tual Design of Floating Wind Turbines
Evaluating Ocean Energy Systems", with Large-Amplitude Motion",
IEA Ocean Energy Systems Annex Ⅱ SNAME.
Report, IEA-OES Annex Ⅱ.

Sauder T., Chabaud V., Thys M., Bachynski 5. ACRONYMS USED IN THIS
E.E., and Sæther L.O., 2016, "Real-time DOCUMENT
hybrid model testing of a braceless BMOWT Bottom-mounted offshore
semi-submersible wind turbine. Part I: wind turbine
The hybrid approach", Proc. 35th Int. FOWT Floating offshore wind tur-
Conf. on Ocean, Offshore and Arctic bine
Engineering OMAE2016, Busan, Korea. OWT Offshore wind turbine
RNA Rotor-nacelle assembly
Shin H., Kim B., Dam P.T., and Jung K., IEC International Electrotechnical Com-
2013, "Motion of OC4 5MW Semi-sub- mission
mersible Offshore Wind Turbine in ir- TRL Technology Readiness Level
regular waves", Proc. of the 32nd Int. IPCC Intergovernmental Panel on Climate
Conf. on Ocean Offshore & Arctic Eng., Change
OMAE2013-10463 Nantes, France. TLP Tension leg platform
DOF degree of freedom
Tucker, M.J., 2001, "Waves in Ocean Engi- RCS Recognized Certificate Society
neering", Elsevier. ISSC The International Ship and Offshore
Structures Congress
United States Department of Energy, 2011,
API-RP2SK American Petroleum Insti-
"Technology Readiness Assessment
tute – Recommended Practice: De-
Guide (DOE G 413.3-4)", Washington,
sign and Analysis of Stationkeeping
D.C.
Systems for Floating Structures
Wang, L., Sweetman, B., 2012, "Simulation
of large-amplitude motion of floating
wind turbines using conservation of mo-
mentum", Ocean Engineering 42, pp.
155–164.

You might also like