![turbine count (farm capacity [GW])](https://www.researchgate.net/publication/381328460/figure/tbl1/AS:11431281251696082@1718303456833/turbine-count-farm-capacity-GW_Q320.jpg)
Available via license: CC BY 4.0
Content may be subject to copyright.
Journal of Physics: Conference
Series
PAPER • OPEN ACCESS
Passive Mooring-based Turbine Repositioning
Technique for Wake Steering in Floating Offshore
Wind Farms
To cite this article: Yuksel R. Alkarem
et al
2024
J. Phys.: Conf. Ser.
2767 092056
View the article online for updates and enhancements.
You may also like
Experimental Analysis of the Wake
Meandering of a Floating Wind Turbine
under Imposed Surge Motion
L. Pardo Garcia, B. Conan, S. Aubrun et
al.
-
Wind turbine wakes can impact down-wind
vegetation greenness
Jay E Diffendorfer, Melanie K Vanderhoof
and Zach H Ancona
-
The potential role of airborne and floating
wind in the North Sea region
Hidde Vos, Francesco Lombardi,
Rishikesh Joshi et al.
-
This content was downloaded from IP address 216.19.199.129 on 12/06/2024 at 17:04
Content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd
The Science of Making Torque from Wind (TORQUE 2024)
Journal of Physics: Conference Series 2767 (2024) 092056
IOP Publishing
doi:10.1088/1742-6596/2767/9/092056
1
Passive Mooring-based Turbine Repositioning
Technique for Wake Steering in Floating Offshore
Wind Farms
Yuksel R. Alkarem1, Kimberly Huguenard1, Amrit S. Verma2,
Diederik van Binsbergen3, Erin Bachynski-Poli´c3, Amir R. Nejad3
1Civil and Environmental Engineering Department, University of Maine, 35 Flagstaff Road,
Orono, Maine 04469, USA
2Mechanical Engineering Department, University of Maine, 35 Flagstaff Road, Orono, Maine
04469, USA
3Department of Marine Technology, Norwegian University of Science and Technology
(NTNU), Jonsvannsveien 82, 7050 Trondheim, Norway
E-mail: yuksel.alkarem@maine.edu
Abstract. Power loss due to wake effects from upstream wind turbines is an important
factor in the design of floating wind farms. These wake disturbances also increase fatigue
loads on downstream units. Economic-driven wind farm layout optimization may culminate in
non-standard, irregular configurations of offshore wind farms that deviate from conventional
engineering practices. Consequently, wind farm developers are inclined towards a more
geometrically coherent layout that are sub-optimal. Floating wind turbines horizontal offsets
can be used to overcome wake losses. In this study, a novel method suggesting various
mooring orientation methods to passively reposition the wind turbines in the farm to maximize
annual energy yield. Preliminary parametric optimization demonstrates the effectiveness of the
proposed method for standard floating wind farm designs to reduce wake effects by as much
as 30% at rated wind speed compared to a baseline case, while preserving the farm’s overall
uniformity.
1. Introduction
Recent develpments in offshore wind technologies have focused on reducing costs and enhancing
their effeciency. As wind farms move towards deep waters [1], floating wind projects are growing
and more seabed leasing in deep waters will take place worldwide [2].
An important aspect in the design of offshore wind farms (OWF) is the annual energy
production (AEP) [3]. After selecting a site, the first step to optimize the AEP is by assessing
the optimal selection of each individual wind turbine (WT) location in the farm to mitigate wake
effects. The wake effects are exacerbated further the more the OWF is composed of clusters
of turbines. The optimal arrangement can be either symmetrical or irregular. Researches have
shown that symmetrically optimal configuration can be achieved with farm-level parametric
variation [4].
Even though stochastic arrangements tend to increase AEP, highly irregular arrangements
are practically difficult to implement and standardize. Therefore, the developer tends to design
OWFs with uniformly distributed WTs due to reasons such as visual impact restrictions, reduced
The Science of Making Torque from Wind (TORQUE 2024)
Journal of Physics: Conference Series 2767 (2024) 092056
IOP Publishing
doi:10.1088/1742-6596/2767/9/092056
2
maximum turbine loads due to turbulence (as spacing between WT tend to be larger for
symmetrical layouts [4]), or other factors imposed in certain locations. For example, New
England wind energy lease areas [5] require clearly delineated transit corridors, navigational
safety, and ease of search and rescue operations near or inside the farm [6].
A well documented approach to combat wake losses is wind farm control. The strategy
is implemented on the individual WTs to achieve an optimum wind-farm objective [7]. One
example of such systems is the misaligned yaw-based wake redirection approach [8] that depends
on active controls that sacrifice upstream turbine power to steer the downstream wake. However,
this requires control effort and derates upstream turbines.
Floating wind turbines (FWTs) are implemented when the water depth exceeds a certain
threshold (around 60m) where a fixed substructure is no longer feasible. The key advantage of
FWTs is that they can operate in areas with more consistent and faster winds, resulting in higher
capacity factors of the floating wind farm (FWF) compared to their fixed-bottom counterparts
as demonstrated by many projects around the world such as the Hywind Scotland pilot project
[9]. Certainly, wake mitigation approaches for onshore or fixed-bottom offshore WTs can also
be implemented on FWTs. However, due to the rigid-body motion allowed by the floating
platform, the dynamics of the system are altered. The mobility of FWTs can either deter the
wind farm control such as the case with yaw and induction-based turbine repositioning (YITuR)
[10] or can be used to mitigate the wake effects even further by the turbine repositioning (TR)
in real-time to mitigate wake effects [10, 11, 12, 13]. Figure 1(a & b) demonstrates the wake
effects on FWF systems. The literature illustrates the effectiveness of integrating TR technique
with active yaw-based wake redirection strategies to dynamically maximize energy yield [14, 15].
A FWT’s horizontal excursion is governed by the station-keeping system (typically mooring
lines). The mooring configuration determines the dynamic behavior of the floater under
the combination of various environmental forcing. The stiffer the system due to mooring
configuration the shorter the natural period of the system and the more the system’s offset
is constricted and vice versa. Therefore, it is of importance to consider macro-scale parameters
at the farm level during mooring design and explore their potential to mitigate losses due to wake
effects. Researchers have explored active mooring system controls for TR for wake mitigation
[16] while others have demonstrated that the passive relocation of WTs, by the static variation
of mooring lines, can boost wind farm efficiency [17, 18]. However, such optimal strategies may
lead to highly irregular and hard-to-maintain mooring setups. Characterizing a distinct set of
mooring-related parameters for each turbine can present significant logistical difficulties within
the context of supply chain management.
This paper proposes a TR technique for layout economics (TRTLE) through farm-level
mooring reconfiguration as illustrated in Figure 1(c). The proposed solution is constrained
to prevent turbines collision. Such limitations are discussed by Ref. [15] when they observed
movable ranges of the turbines overlap. Another issue is the power cable (also called umbilical).
It must be designed specifically to prevent high tensions and curvatures that come along with
high offsets. The design of suspended umbilical for FWT application in deep water is investigated
by Ref. [19], and an approach to the dynamic inter-array cable optimal design with a lazy-wave
configuration is provided by Ref. [20]. Extreme static and dynamic responses of the umbilical in
very deep waters must be considered [21]. Sufficient spacing under failure conditions is ensured
and the umbilical is designed to accommodate high excursions.
The remaining structure of this paper proceeds as follows: Section 2 gives a general
background on the methodology as well as the scope and limitations. Section 3 presents a brief
description of the problem and of the parametric study and optimization procedure. Results
regarding parametric analysis and umbilical design are provided in Section 4, while Section 5
concludes with remarks on how to proceed with future studies.
The Science of Making Torque from Wind (TORQUE 2024)
Journal of Physics: Conference Series 2767 (2024) 092056
IOP Publishing
doi:10.1088/1742-6596/2767/9/092056
3
BL
ASYMoor
turbine after repositioning
Wind direction
turbine before repositioning
(a) (b) (c)
wind direction
TR
TR
(c)(b)(a)
Figure 1. a) illustration of FWF and wake effects caused by upstream turbines, b) top view
highlighting velocity deficit in the wakes and c) the proposed TRTLE technique utilizing floater
displacement from the mooring design to avoid wake from upstream turbines (repositioning
shown is not to scale).
fairlead
anchor
𝑑
o𝑥𝑓
𝑧𝑓
𝑥𝑎
mooring
umbilical
seabed
Sea water level
ΔS
ℓ1
ℓ2
𝑥𝑢
𝑧𝑏𝑠
𝐷
(a)
turbine: turbine:
(b)
(a)
a) b)
Figure 2. a) illustration of stand-alone turbine parameters and b) farm level parameters.
2. Methodology
The problem is characterized by parameters of four categories: farm, environmental, mooring,
and umbilical. Each parameter is further classified into sub-categories depending on whether it
varies in an intra-array (IA) fashion (IA-variable) or not (IA-fixed) as summarized and illustrated
in Table 1 and Figure 2. The farm-level parameters determine the number of turbines to be
placed at the site. The main driving parameter for the mooring design is the catenary coefficient,
β. For a mooring line of length, Lm, the catenary coefficient ranges between zero and one and
can be described as
β=Lm−Lmmin
Lmmax −Lmmin
,(1)
where Lmmin and Lmmax are the minimum and maximum lengths of the mooring cable,
respectively;
Lmmin =q(zf−za)2+ (xf−xa)2
Lmmax = (zf−za)+(xf−xa).
(2)
2.1. Hydrodynamic and mooring model:
The model used for the floating platform, the mooring, and power cable is Orcina OrcaFlex
driven via a dynamic link library (DLL) through the python OrcFxAPI interface. This allows
optimization processes and parametric analyses to be conducted effortlessly while enabling
coupling with the wake model. The turbine model is the UMaine VolturnUS-S reference platform
[22] supporting the IEA 15-megawatt (MW) offshore reference wind turbine [23].
The Science of Making Torque from Wind (TORQUE 2024)
Journal of Physics: Conference Series 2767 (2024) 092056
IOP Publishing
doi:10.1088/1742-6596/2767/9/092056
4
Table 1. parametric description of the problem
Category parameter symbol sub-category
farm layout orientation θFIA-fixed
spacing ratio SIA-fixed
environmental water depth dIA-fixed
mooring
number of mooring lines NmIA-fixed
fairlead horizontal distance xfIA-fixed
fairlead vertical distance zfIA-fixed
anchor horizontal distance xaIA-fixed
catenary coefficient βIA-fixed
mooring segment length ∆SIA-fixed
mooring orientation ϕmIA-variable
umbilical
umbilical orientation ϕuIA-variable
bend stiffener range zbs IA-fixed
umbilical horizontal reach xuIA-fixed
umbilical sectional lengths ℓ1, ℓ2IA-fixed
2.2. Wake model:
The wake is simulated via the open-source FLOw Redirection and Inductive in Steady State
(FLORIS v3.0.0) software developed by the National Renewable Energy Laboratory (NREL)
and Delft University of Technology. The wake deflection and velocity models are configured as
Gaussian and the combination model is the sum of squares freestream superposition model
(SOSFS) [25]. A fixed turbulence intensity of I∞= 6% is assumed with wake expansion
coefficients ka= 0.10, kb= 0.004, transition point near-far wake coefficients αw= 0.58,
βw= 0.077, and no lateral wake deflection [26].
2.3. Scope of study and key assumptions
The investigation in this research is restricted to a singular site shape but can be easily extended
to more complex site boundaries. Additionally, this study only scrutinizes the effectiveness of the
proposed approach from the steady state perspective. In future work, the dynamic properties
of the system will be studied. This study only focuses on deep water applications. The water
depth is fixed at d= 1000m and the mooring type is the same chain properties described in
the definition of the VolturnUS-S. These properties might not be realistic for this preliminary
study. Future research aims to bridge these gaps by applying parametric study to investigate
water depth variation and synthetic mooring type application. Some key assumptions are 1)
the seabed slope is set to zero and bathymetric variations are not included in this study, 2) the
main driving force of the turbine’s repositioning is the thrust force applied at the hub height, 3)
wind speed across the wind rose is constant and is equal to the rated wind speed, and 4) wave
and current actions are not included in this study.
3. Problem analysis, and optimization schemes
3.1. Analysis flowchart
The flowchart in Figure 3 provides a brief description of the process conducted in this study.
The site boundaries are provided as input to the algorithm and the turbine locations inside the
boundaries are created based on the layout orientation θFand a given spacing ratio, S=x/D
The Science of Making Torque from Wind (TORQUE 2024)
Journal of Physics: Conference Series 2767 (2024) 092056
IOP Publishing
doi:10.1088/1742-6596/2767/9/092056
5
between the turbines, where xis the horizontal distance between two neighboring turbines in
either directions and Dis the rotor diameter of the turbine. The site in this study is arbitrarily
selected as a square. The AEP for the baseline case can be computed given a turbine model
and the energy resources at the site. The wind rose is selected from the case study given by the
international energy agency (IEA) Task 37 [27] with dominant wind direction from the west. As
for TRTLE analysis, the IA-fixed properties are used to determine the catenary configuration of
the mooring lines under two constraints: maximum excursion and maximum allowable horizontal
anchor spread. These two constraints address limited excursion due to the umbilical connection
and ensure turbines do not collide with each other in case of a mooring line failure.
The constraints are met via running an optimization problem through varying βuntil
excursion limits are met. An excursion error threshold of 5m is set. The framework used for
optimization is the open-source Optuna package [28] using a Tree-structured Parzen Estimator
(TPE) Sampler algorithm [29]. The ratio of maximum platform offset to water depth, emax/d,
is selected to be between 0.3 and 0.4. This ratio is chosen to demonstrate the potential of the
proposed design. The recommended practice in the literature for excursion to water depth ratio
is between 0.125 to 0.3 [30]. However, ratios up to 0.5 are reported in the literature [31]. Anchor
horizontal spread is limited to half the spacing between the turbines, xa=S×D
2.
To achieve fast computation, the platform excursion of the stand-alone turbine in the
horizontal plane as a function of wind speed and direction is computed with a radial discretization
value of ∆θ= 22.5◦. The generated excursion amplitude is identical for all turbines since they
share the same IA-fixed parameters. An example of excursion values as functions of wind speed,
u, and direction, θis illustrated in Figure 4(a).
The optimization process begins via adjusting the IA-variables, dictating how the line
headings are oriented in the farm (to be discussed in section 3.2). Once IA-variables are selected,
the algorithm loops over all wind speeds and wind directions. For each case, the flow field is
first computed to obtain thrust forces on all turbines. Subsequently, the platform excursion is
quantified and the turbines are relocated. Since the new turbine locations have different wind
velocities, a recursive loop takes place to find the flow field, relocating the turbines until their
locations converge. The average location of all turbines must meet a relocation tolerance of
0.25m (a fraction of the mean offset values). The method requires the system to reach a new
equilibrium in all degrees of freedom of the rigid body motion. Then, energy is computed for the
given wind speed and wind direction and the AEP is computed given the wind speed/direction
frequencies.
3.2. Mooring orientation methods
Preliminary analyses suggest a linear mooring arrangement (2-line configuration in opposing
directions) achieves high difference between the static equilibrium position and the excursion
caused by the thrust force. On the farm-level, two methods for mooring orientation are presented:
1) the chessboard-pattern, creating a zigzag mooring orientation throughout the farm, and 2)
plane variation, altering mooring orientation linearly in the x−yplane. Plan views of these
orientation methods are shown in Figure 4(b).
4. Results
4.1. Parametric analysis
To investigate TRTLE’s impact on farm performance relative to the baseline (defined here as
turbines having identical mooring orientation with equal offsets or equivalently zero offset), a
parametric analysis is conducted. Key variables included farm orientation, θF∈[0◦,45◦] at 5◦
intervals, and turbine spacing ratio, S∈[6,10] in increments of 0.5. The layout constraints
included a 15km square boundary without turbine quantity limitations, subject to the spacing
constraint.
The Science of Making Torque from Wind (TORQUE 2024)
Journal of Physics: Conference Series 2767 (2024) 092056
IOP Publishing
doi:10.1088/1742-6596/2767/9/092056
6
site boundaries
farm-level properties
turbine fixed location
turbine model
wind resources
compute baseline AEP
find mooring catenary
configuration
TRTLE
baseline
farm model
constraints
maximum
excursion
maximum anchor
spread IA-fixed properties
TRTLE optimization loop
IA-variables
orientation
method
loop over wind speeds/wind directions
find flow
field
compute
thrust at
turbine
location
compute
excursion
relocate
turbines
no
converged?
compute
TRTLE AEP
yes
Figure 3. Flowchart for the computation of AEP for the baseline and TRTLE cases.
b)
at rated at cut-in
at cut-off
𝑥(𝑢, 𝜃) 𝜃
𝑢
𝑦(𝑢, 𝜃) 𝜃
𝑢
a)
chessboard pattern
plane variation
b)
turbine
Figure 4. a) platform excursion functions example based on wind speed and direction, b)
various orientation methods used in farm-level optimization, with red lines showing mooring
line orientation and black dots marking turbine locations.
Subsequently, the TRTLE optimization is executed, proposing various mooring orientation
methods to maximize AEP. A consistent trial count of 500 is maintained for all configurations, a
number deemed sufficient for optimizer convergence based on preliminary sensitivity trials (data
not presented). Parametric analysis is depicted in Figure 5, highlighting wake loss avoidance
within the farm when using TRTLE relative to the baseline. Notably, power output enhancement
is achieved across-the-board. The extent of this enhancement varies relative to the extremity of
The Science of Making Torque from Wind (TORQUE 2024)
Journal of Physics: Conference Series 2767 (2024) 092056
IOP Publishing
doi:10.1088/1742-6596/2767/9/092056
7
F
case A
case B
Figure 5. Comparison of wake loss reductions across farm-level variations in a square
configuration relative to the baseline.
the wake effects. Minimal enhancement, with a mere 0.65% reduction in wake losses, is observed
at θF= 10◦and S= 8.0. In contrast, the maximum improvement, a significant 29.05% reduction
in wake losses, is noted at θF= 0◦and S= 10.0. Hence, TRTLE is a robust solution for sub-
optimally configured farms, particularly in wake-dominant scenarios. For example, Table 2
illustrates the number of turbines and the wind farm capacity when the two layouts share the
same wake loss percentage at farm orientations (θF= 0◦,20◦,and 25◦). The baseline requires a
spacing increase to 7.5 to achieve similar wake effects compared to the TRTLE-optimized layout
with a spacing of S= 6 with higher turbine count. This implies the feasibility of denser turbine
deployment within the same wind energy area and high farm capacity under a standard layout
while keeping wake losses at minimum, thereby enhancing energy production. Amongst the two
proposed mooring orientation methods depicted in Figure 4(b), the plane variation is selected
57% of the time. However, chessboard pattern has a higher potential for avoiding wake effects
with an average of 20% compared to mere 6% for the plane variation method. This indicates
that chessboard pattern is very effective for unfavorable farm properties that have high wake
losses but not as effective when wake losses are originally low.
In scenarios where optimal layout selection is unfeasible, TRTLE methodology enables
developers to attain heightened production under diverse constraints. Consider, for instance,
a developer constrained by financial limitations aiming to minimize the number of turbines
deployed. The two cases annotated in Figure 5 highlight these options which involve deploying
36 turbines, each spaced 10 rotor diameters apart, at a farm orientation of θF= 0◦,and 5◦.
Given the farm parameters, case A and B share the same design constraints that govern the
The Science of Making Torque from Wind (TORQUE 2024)
Journal of Physics: Conference Series 2767 (2024) 092056
IOP Publishing
doi:10.1088/1742-6596/2767/9/092056
8
Table 2. turbine count (farm capacity [GW])
layout θF= 0◦θF= 20◦θF= 25◦
TRTLE 100 (1.50) 107 (1.61) 107 (1.61)
baseline 64 (0.96) 68 (1.02) 67 (1.01)
ratio of anchor horizontal distance to water depth;
xa
d=S×D
2d= 1.211.(3)
The adapted mooring configuration to the 2-line design at water depth of 1000m has the
corresponding catenary coefficient βand mooring line length presented in Table 3. This ensures
the platform motion is constrained to the maximum allowable excursion, emax. The normalized
watch circles for the baseline and the optimized mooring configuration adapted from moderate
to deep water location are plotted in Figure 6. In addition to the baseline having smaller
excursions, the mooring orientation of the baseline design (3-line) is considered identical for all
wind turbines in the farm. This leads to similar offsets of all turbines and negligible effects on
the wake computation.
Case A coincides with the most optimistic results and the optimum design for case B achieves
a 17.62% reduction in wake effects. The TRTLE optimization technique selects the chessboard
pattern for case A and plane variation pattern for case B. Figure 7(a) and Figure 8(a) detail
TRTLE’s mooring line configurations and anchor points, alongside the watch circle (rendered
as an ellipse). Furthermore, in the event of mooring line failure, the affected turbine drifts
towards the alternate anchor, remaining within a defined failure circle. This circle, where the
turbine resides post-failure, assumes a mooring line with catenary shape β= 1 and Lm=Lmmax ,
indicating the touchdown point directly beneath the fairlead. Since it is highly unlikely for two
opposite mooring lines in two wind turbines fail at the same instant, line failure circles are
allowed to overlap. However, they must not overlap with neighboring turbines’ watch circles
to mitigate collision risks. Figure 7(b) and Figure 8(b) delineate the wake patterns induced
by prevailing wind in the baseline case, juxtaposed with the TRTLE wake. Notably, TRTLE’s
wake is broader, due to turbine repositioning, thereby augmenting overall farm energy output.
Note that the possibility of mooring lines to share anchors is likely especially for case A with
chessboard pattern mooring orientation. This is a topic for future studies.
Table 3. Example of mooring design adaptation to 1000m, 2-line mooring configuration under
excursion and anchor radius constraints (S= 10).
xa/d = 1.211 emax[m]xa[m]β[−]Lm[m]max (e/emax )[−]
2-line 3-line
d= 200m80 242.2 0.118 274.57 1.05 0.33
d= 1000m400 1211.2 0.296 1701.35 1.01 0.25
4.2. Umbilical design and analysis
The TRTLE mooring configuration allows high horizontal excursions of the platform. This
puts further constraints on the design of the power cable (umbilical). The umbilical faces
The Science of Making Torque from Wind (TORQUE 2024)
Journal of Physics: Conference Series 2767 (2024) 092056
IOP Publishing
doi:10.1088/1742-6596/2767/9/092056
9
e
e
max
x
y
z
a) b)
Figure 6. a) normalized watch circle of an optimized mooring design example of 2-line mooring
configuration compared to the baseline 3-line mooring configuration at moderate and deep water
depths for S= 10 and b) mooring lines profile for the 2-line and 3-line system.
a) b)
Figure 7. Case A: a) TRTLE farm-level configuration with visuals of anchor locations, mooring
lines, watch circle, and line failure circle for the studied case of lowest number of turbines and
b) wake visuals of baseline and TRTLE layered on top of each other for the prevailing wind
direction from the west
highest variations in tensions and curvature if it is perpendicular to the 2-line mooring setup
(i.e. the angle between ϕuand ϕmis a right angle). Therefore, to be conservative, all umbilical
analysis and design assume maximum excursion and perpendicular configuration between the
lines and the cable. Traditionally, the umbilical must be designed to prevent high tensions and
high curvatures in the power cable while keeping it suspended in the water column under all
conditions. Hence, two constraints are defined here:
•near constraint: when the platform offsets towards the umbilical cable, a clearance distance
between the lowest point of the hang-off catenary section (the sag portion) and the seabed
The Science of Making Torque from Wind (TORQUE 2024)
Journal of Physics: Conference Series 2767 (2024) 092056
IOP Publishing
doi:10.1088/1742-6596/2767/9/092056
10
a) b)
Figure 8. Case B: a) TRTLE farm-level configuration with visuals of anchor locations, mooring
lines, watch circle, and line failure circle for the studied case of lowest number of turbines and
b) wake visuals of baseline and TRTLE layered on top of each other for the prevailing wind
direction from the west
must be kept higher than 10% of the water depth [20],
•far constraint: when the platform offsets away from the umbilical cable, maximum tension
at the bend stiffener (the section where the umbilical and the platform meets) must be kept
below than 20% the tension value of the nominal position (zero offset).
Two different configurations for the umbilical are chosen for this investigation: a single lazy-
wave, and a double lazy-wave [32]. As the name suggests, the single lazy wave has one hang-off
section, one buoyancy section, and one last touchdown catenary section. Whereas the double
lazy-wave has two hang-off, two buoyancy, and one touchdown sections. The optimization
problem investigates the variation of umbilical-related non-dimensional parameters that would
satisfy the aforementioned constraints. These parameters are:
•the ratio of the total cable’s length, Lu, to the umbilical horizontal reach, (Lu
xu), varies
between 0.75 and 2,
•the ratio of the umbilical horizontal reach to the water depth, (xu
h), varies between 1.4 and
1.8,
•the ratio of the hang-off catenary section(s) length to the total cable length, ℓ1
Lu, varies
between 0.5<ℓ1
Lu<0.8 for a single lazy-wave and 0.15 <ℓ1
Lu<0.3 for double lazy-wave
configuration, and
•the ratio of the buoyancy catenary section(s) length to the total cable’s length, ℓ2
Lu, varies
between 0.05 <ℓ2
Lu<0.1 for both configurations.
The bend stiffener is modelled as a steel section for the first 1m with 120e3 kN.m2bend
stiffness and 5m as polymer section with a bending moment value of 103 kN.m under curvature
of 1.12 ×10−3rad/m. These values, as well as the umbilical cross-sectional properties, are
replicated from the work of Ottesen [21]. This resulted in two configurations successfully able
to meet the two constraints. Figure 9 compares the two solutions to one another at nominal,
near, and far platform excursions. It is clear that the double-wave is more favorable in terms
The Science of Making Torque from Wind (TORQUE 2024)
Journal of Physics: Conference Series 2767 (2024) 092056
IOP Publishing
doi:10.1088/1742-6596/2767/9/092056
11
a)
b) c)
Figure 9. a) umbilical cable profiles, b) tensions, and c) curvatures in nominal, near, and far
excursions
of seabed clearance (16% of water depth compared to 12% for a single lazy wave configuration)
and lower tensions in the bend stiffener under far excursions (a little above 150 kN compared
to over 350 kN). The double lazy-wave shows worse response than single lazy-wave in curvature
as expected. These are static results and the dynamic excitation of the umbilical cable must be
examined in future work, particularly at the bend stiffener.
5. Conclusions and remarks for future studies
This paper highlighted a new mooring conceptual design framework, TRTLE, that aims to
minimize wake losses by considering their variation with respect to the turbines layout in the
farm. It is shown that TRTLE is capable of limiting wake losses, hence advancing the energy
yield of the farm when compared to the baseline cases where platforms’ excursions are limited.
The level of improvement depends on the level of optimality of the baseline layout. Developers
can be faced with multitude of constraints that force them to select a sub-optimal layout as
far as wake effects are concerned. These restrictions vary from increasing the farm production,
which lead to higher number of turbines and higher wake losses, competing use of the sea for
military, nature, or shipping routes, and others. TRTLE provides a solution to reducing wake
effects by allowing each platform to relocate with limited excursion depending on the wind speed
and direction while reducing mooring complexity and cost. Parametric analysis of farm-level
parameters illustrated a reduction in wake effects between 1% and 30% when TRTLE is applied
at rated wind speeds. The umbilical cable is redesigned and analyzed to allow for such excursions
to take place without risking the cable touching the seabed in near excursions and avoiding high
tensions in far excursions. The proposed double lazy-wave configuration accommodates higher
The Science of Making Torque from Wind (TORQUE 2024)
Journal of Physics: Conference Series 2767 (2024) 092056
IOP Publishing
doi:10.1088/1742-6596/2767/9/092056
12
platform excursions at the expense of increased curvature in the hang-off and buoyancy sections.
For future research, exploring the impact of TRTLE on wake reduction in realistic wind farms,
under realistic wind conditions, is crucial. For instance, the analysis presented here are limited of
a single wind speed. This limitation overestimates both wake effects and annual energy yields as
the farm produces less power in lower wind speeds and velocity deficit due to wake effects become
irrelevant to the produced energy at high wind speeds. Furthermore, conducting dynamic
analyses of the mooring configuration and power cable is critical to assess the soft mooring’s
low natural frequencies in horizontal motions, investigate possible resonance with environmental
forces, and estimate the extreme excursions and line tensions. Additionally, it is imperative to
investigate the economical aspects of the proposed solution. The proposed design is assumed to
have multiple economical advantages in terms of increasing energy production, reducing wake
effects and wake-induced fatigue loads on downstream turbines, reducing the number of mooring
lines and anchors usually required to limit horizontal excursions, and reducing tension-induced
fatigue on the mooring systems due to its relaxed configuration. On the other hand, it might have
implications that might increase the overall cost such as those driven by designing specialized
mooring and umbilical cables and additional cost to introduce more redundancy to the system.
For instance, the umbilical might need to be increased in length, which might increase the
footprint on the seabed. Due to high excursions, the laid portion of the mooring line on the
seabed can cause more disturbances, which in turn can disturb aquatic life in its proximity.
These issues must be addressed in future work.
6. Acknowledgment
The authors would like to acknowledge the support from the Norwegian Directorate for Higher
Education and Skills (HK-dir) through the NUWind Project (Project number UTF-2021/10157).
References
[1] Campanile, A., Piscopo, V. and Scamardella, A., 2018. Mooring design and selection for floating offshore
wind turbines on intermediate and deep water depths. Ocean Engineering, 148, pp.349-360.
[2] Lee J and Zhao F 2022 Global wind report 2022.https://gwec.net/global-wind-report-2022/
[3] Gonz´alez, J.S., Pay´an, M.B., Santos, J.M.R. and Gonz´alez-Longatt, F., 2014. A review and recent
developments in the optimal wind-turbine micro-siting problem. Renewable and Sustainable Energy
Reviews, 30, pp.133-144.
[4] Neubert, A., Shah, A. and Schlez, W., 2010, November. Maximum yield from symmetrical wind farm layouts.
In Proceedings of DEWEK (Vol. 98).
[5] Uniform turbine layout touted for NE US offshore wind, Accessed 2 January 2020; 2019
https://renews.biz/56505/
[6] Gonzalez-Rodriguez, A.G., Serrano-Gonzalez, J., Burgos-Payan, M. and Riquelme-Santos, J., 2022. Multi-
objective optimization of a uniformly distributed offshore wind farm considering both economic factors
and visual impact. Sustainable Energy Technologies and Assessments, 52, p.102148.
[7] Kheirabadi, A.C. and Nagamune, R., 2019. A quantitative review of wind farm control with the objective of
wind farm power maximization. Journal of Wind Engineering and Industrial Aerodynamics, 192, pp.45-73.
[8] Jim´enez, ´
A., Crespo, A. and Migoya, E., 2010. Application of a LES technique to characterize the wake
deflection of a wind turbine in yaw. Wind energy, 13(6), pp.559-572.
[9] Equinor 2022 Floating Wind
[10] Kheirabadi, A.C. and Nagamune, R., 2020. Real-time relocation of floating offshore wind turbine platforms
for wind farm efficiency maximization: An assessment of feasibility and steady-state potential. Ocean
Engineering, 208, p.107445.
[11] Kheirabadi, A.C. and Nagamune, R., 2019, July. Modeling and power optimization of floating offshore wind
farms with yaw and induction-based turbine repositioning. In 2019 American Control Conference (ACC)
(pp. 5458-5463). IEEE.
[12] Han, C., Homer, J.R. and Nagamune, R., 2017, May. Movable range and position control of an offshore
wind turbine with a semi-submersible floating platform. In 2017 American Control Conference (ACC) (pp.
1389-1394). IEEE.
[13] Fleming, P., Gebraad, P.M., Lee, S., van Wingerden, J.W., Johnson, K., Churchfield, M., Michalakes, J.,
The Science of Making Torque from Wind (TORQUE 2024)
Journal of Physics: Conference Series 2767 (2024) 092056
IOP Publishing
doi:10.1088/1742-6596/2767/9/092056
13
Spalart, P. and Moriarty, P., 2015. Simulation comparison of wake mitigation control strategies for a
two-turbine case. Wind Energy, 18(12), pp.2135-2143.
[14] Ceriello, V., 2023. Wake Effect Mitigation of Floating Offshore Wind Farms: Combining Layout
Optimization, Turbine Repositioning and Yaw-based Wake Redirection. TUDelft Master’s thesis.
[15] Kilin¸c, U., 2022. Dynamic wind farm layout optimization: To find the optimal spots for movable floating
offshore wind turbines through dynamic repositioning.
[16] Rodrigues, S.F., Pinto, R.T., Soleimanzadeh, M., Bosman, P.A. and Bauer, P., 2015. Wake losses optimization
of offshore wind farms with moveable floating wind turbines. Energy conversion and management, 89,
pp.933-941.
[17] Mahfouz, M.Y. and Cheng, P.W., 2023. A passively self-adjusting floating wind farm layout to increase the
annual energy production. Wind Energy, 26(3), pp.251-265.
[18] Mahfouz, M.Y., Hall, M. and Cheng, P.W., 2022, May. A parametric study of the mooring system design
parameters to reduce wake losses in a floating wind farm. In Journal of Physics: Conference Series (Vol.
2265, No. 4, p. 042004). IOP Publishing.
[19] Schnepf, A., Lopez-Pavon, C., Devulder, A., Johnsen, Ø. and Ong, M.C., 2022, June. Suspended power
cable configurations for floating offshore wind turbines in deep water powering an FPSO. In International
Conference on Offshore Mechanics and Arctic Engineering (Vol. 85932, p. V008T09A049). American
Society of Mechanical Engineers.
[20] Rentschler, M.U., Adam, F. and Chainho, P., 2019. Design optimization of dynamic inter-array cable systems
for floating offshore wind turbines. Renewable and Sustainable Energy Reviews, 111, pp.622-635.
[21] Ottesen, T., 2010. Extreme response estimation of umbilical on very deep water (Master’s thesis, Norges
teknisk-naturvitenskapelige universitet, Fakultet for ingeniørvitenskap og teknologi, Institutt for marin
teknikk).
[22] Allen, C., Viscelli, A., Dagher, H., Goupee, A., Gaertner, E., Abbas, N., Hall, M. and Barter, G., 2020.
Definition of the UMaine VolturnUS-S reference platform developed for the IEA wind 15-megawatt offshore
reference wind turbine (No. NREL/TP-5000-76773). National Renewable Energy Lab.(NREL), Golden,
CO (United States); Univ. of Maine, Orono, ME (United States).
[23] Gaertner, E., Rinker, J., Sethuraman, L., Zahle, F., Anderson, B., Barter, G.E., Abbas, N.J., Meng, F.,
Bortolotti, P., Skrzypinski, W. and Scott, G.N., 2020. IEA wind TCP task 37: definition of the IEA
15-megawatt offshore reference wind turbine (No. NREL/TP-5000-75698). National Renewable Energy
Lab.(NREL), Golden, CO (United States).
[24] NREL. FLORIS. Version 2.4, 2021. Available online: https://github.com/nrel/floris
[25] Katic, I., Højstrup, J. and Jensen, N.O., 1986, October. A simple model for cluster efficiency. In European
wind energy association conference and exhibition (Vol. 1, pp. 407-410). Rome, Italy: A. Raguzzi.
[26] van Binsbergen, D., Daems, P.J., Verstraeten, T., Nejad, A. and Helsen, J., 2023. Hyperparameter tuning
framework for calibrating analytical wake models using SCADA data of an offshore wind farm applied on
FLORIS. Wind Energy Science Discussions, 2023, pp.1-28.
[27] Baker, N.F., Stanley, A.P., Thomas, J.J., Ning, A. and Dykes, K., 2019. Best practices for wake model and
optimization algorithm selection in wind farm layout optimization. In AIAA Scitech 2019 forum (p. 0540).
[28] Optuna: A hyperparameter optimization framework. Available online:
https://optuna.readthedocs.io/en/stable/index.html
[29] Watanabe, S., 2023. Tree-structured Parzen estimator: Understanding its algorithm components and their
roles for better empirical performance. arXiv preprint arXiv:2304.11127.
[30] Campanile, A., Piscopo, V. and Scamardella, A., 2018. Mooring design and selection for floating offshore
wind turbines on intermediate and deep water depths. Ocean Engineering, 148, pp.349-360.
[31] Brommundt, M., Krause, L., Merz, K. and Muskulus, M., 2012. Mooring system optimization for floating
wind turbines using frequency domain analysis. Energy Procedia, 24, pp.289-296.
[32] Zhao, S., Cheng, Y., Chen, P., Nie, Y. and Fan, K., 2021. A comparison of two dynamic power cable
configurations for a floating offshore wind turbine in shallow water. AIP Advances, 11(3).
