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

net/publication/383008936

Optimal Mooring Pattern for a Semi-Submersible FOWT in a Typhoon

Environment

Conference Paper · August 2024

DOI: 10.1115/OMAE2024-122874

CITATIONS READS

0 114

7 authors, including:

Glib Ivanov Zhao-Yu Lai

National Taiwan University National Taiwan University
19 PUBLICATIONS 34 CITATIONS 3 PUBLICATIONS 9 CITATIONS

SEE PROFILE SEE PROFILE

Kai-Tung Ma
National Taiwan University
63 PUBLICATIONS 837 CITATIONS

SEE PROFILE

All content following this page was uploaded by Glib Ivanov on 19 September 2024.

The user has requested enhancement of the downloaded file.

 Proceedings of the ASME 2024 43rd International
Conference on Ocean, Offshore and Arctic Engineering
OMAE2024
June 9-14, 2024, Singapore

OMAE2024-122874

OPTIMAL MOORING PATTERN FOR A SEMI-SUBMERSIBLE FOWT IN A TYPHOON ENVIRONMENT

Glib Ivanov Yongyan Wu DongHui Chen

National Taiwan University Aker Solutions Genesis Engineering
Taipei, Taiwan Houston, Texas, USA Houston, Texas, USA
Zhao-Yu Lai Jui-Chen Chen Nikolai Gladkov
National Taiwan University National Taiwan University Google
Taipei, Taiwan Taipei, Taiwan New Taipei, Taiwan
Kai-Tung Ma
National Taiwan University
Taipei, Taiwan

ABSTRACT In conclusion, a 3x2 mooring pattern for a 15MW semi-

The scarcity of suitable shallow waters for fixed-bottom submersible FOWT in typhoon regions might be theoretically
offshore wind turbines has prompted developers to explore achieved on paper. However, it may not be practically achieved,
deeper waters, albeit with caution due to the significant capital if the size limit of the manufactured chain is considered.
expenditure (CAPEX) associated with Floating Offshore Wind
Turbines (FOWTs). A major cost component in FOWTs is the Keywords: Floating Offshore Wind Turbines, FOWT, Mooring
mooring system, a concern exacerbated in regions with the System, Cost Optimisation, Minimum Break Load, Floating
presence of typhoons, necessitating a more robust and therefore Wind, V-Share Mooring, Mooring Chain.
expensive 3x3 mooring solution compared to 3x1 in areas such
as the North Sea. This study tried to propose a 3x2 mooring 1. INTRODUCTION
arrangement tailored for FOWTs in the typhoon region, offering As pressure to de-carbonise intensifies, offshore wind power
a potential cost reduction of up to one-third compared to the emerges as one of the most efficient solutions to go green. As
original 3x3 configuration. sites suitable for fixed OWT are already scarce, Floating
To achieve cost savings, an in-depth analysis of spreading Offshore Wind Turbines (FOWT) emerge as the technology to
angles from 10° to the widest angle of 126° is performed using save the day [2]. However, they are still 3-6 times more
the pre-tension-diameter vs offset-tension 3D response surfaces expensive than fixed OWT, which doesn’t allow for mass
technique. The research reveals that, theoretically, a 60° commercial application. There was a similar situation with fixed
mooring angle minimizes floater offset with the least tension. turbines until their capital cost per MW went down by more than
However, the discrepancy in safety factors between intact and 50% in less than a decade [2].
damaged conditions makes grouped mooring a more cost- This became possible due to 1) Economies of scale and
effective choice. Utilizing the American Bureau of Shipping supply chain development, 2) technological advancements in all
(ABS) safety factor, a 10° spread angle mooring system's chain parts of the turbines, including jacket foundations, tower
cost emerges as roughly one-tenth cheaper than alternative optimisation, nacelle and blades rotation control and components
configurations. made for easier installation procedures.
Additionally, the study explores an innovative V-Share For Floating Wind, some of the systems are similar, but
mooring [1], wherein a single pile anchor connects to two there are unique parts, notably the mooring system that holds the
different columns via two mooring lines. Depending on the floater in place, the dynamic power export cable and port-side
anchoring conditions, it can be the most cost-effective. integration. Also, FOWTs are generally much bigger, most now
designed to support 15 MW turbines. Big turbines add another

1 © 2023 by ASME
layer of complexity when manufacturing or assembling them; [7], this floater is used for in this paper, shown in Figure 1.2.
this was already discussed in the author’s earlier research [3].

FIGURE 1.2: TaidaFloat semi-submersible with a 3x2 mooring

system that is at 10° spread angle as the base case.
FIGURE 1.1: Schematic representations of different mooring patterns
that were examined in the study. (Not to scale) About half of the mooring cost comes not from the line’s
material but from the charter of installation vessels. The
S. Verde and E. N. Lages [3] have shown that FOWT installation time often takes more than estimated due to rough
mooring line tension is governed mostly by wave loads in weather or unforeseen circumstances [8]. Big Anchor Handling
operational and parked conditions, and there is a noticeable Vessels (AHV) [9] are chartered for about 250,000 USD/day and
impact of wind load in operational conditions. Xu et al [4] show their supply around the world is sparse and they are
the cost-benefit of taut fibre rope mooring in shallow water. disproportionally located in Europe. If a developer wants to
However, in the current industry practice, fibre ropes remain charter a ship, it usually needs to pay for a whole year, while the
susceptible to potential harm from marine growth, UV exposure installation work can only be done during a 5-month weather
near the sea surface, and soil ingress close to the sea bottom. window in spring and summer (e.g. for Taiwan Strait).
Consequently, in shallow waters, only a short part of the mooring Offshore Renewable Energy Catapult analysed different mooring
line can be fibre rope, so the elastic benefit of stretchy rope is systems’ carbon footprint [10]. It was found that steel chain production
limited, while chain-rope connectors add installation cost and and transportation was by far the most significant emission source. They
fatigue susceptibility. Therefore, a full-chain catenary system is also compare 3x1, 3x2 and 3x3 catenary moorings, finding that
explored in this paper due to its proven robustness with extensive switching from 3x1 to 3x2 significantly increased carbon footprint but
track records in the offshore industry. switching from 3x2 to 3x3 was only a slight increase. The monetary cost
was not reported.
This paper focuses on the most prominent, and most costly
A Floating Wind Farm cost items are shown in Figure 1.3.
difference of a FOWT – the mooring system. FOWT moorings
As mooring and dynamic cable capital, installation and
come in different shapes and sizes depending on the
decommissioning costs for FOWT contribute around 11% of the
environmental conditions and their cost varies, accordingly; they
total wind farm cost according to data from European Projects
are shown in Figure 1.1. For the most recently launched deep
by BVG Associates[11], there is a big incentive to make FOWT
water FOWT Guan Lan, it amounted to 7% of the project cost
moorings non-redundant, especially as it is allowed by class rules
[5]. However, many FOWTs find themselves in shallower water,
such as ABS rules for FOWT [12]. While it is a straightforward
where it is generally more challenging to design a mooring
way to reduce mooring costs, in that case, the prescribed safety
system and more chains will be used [6]. For example, for
factor (2) is higher resulting in much thicker and heavier chains,
TaidaFloat, which is designed to be moored at 70 m depth, the
that need to be operated by bigger vessels, which can offset some
mooring system weight is getting close to the hull steel weight
or much of the benefits.

2 © 2023 by ASME
 The other approach is a redundant mooring system, which conditions investigated in this paper. Actual data from a site in
is designed with a failure in mind – an accident that cuts off one Taiwan strait is used from authors’ previous research, shown in
of the mooring lines, the so-called one-line damaged case. There Table 2.2. [13]. Please note, that for this floater and conditions,
are different safety factors applicable for all lines intact and one- max total wind force on floater, tower and turbine occurs when
line damaged conditions. Due to lower safety factors and the the turbine is parked. [14]
environmental force distributed over several lines, redundant The wind, wave and current’s magnitude are constant with
systems have considerably thinner and sometimes shorter chains. direction and they are collinear for a fair comparison. The mesh
Because of that, the governing design condition is the damaged with coordinate system is shown in Figure 2.1 and its principal
condition, usually resulting in much higher line tension and dimensions are in Table 2.1, a simulation snapshot from
platform offset (excursion) than intact. OrcaFlex is shown in Figure 2.2. In OrcaFlex environmental
Dev force’s 0° direction means wind/wave is propagating from 180°
OSS towards 0°, contrary to wind data, where 0° means wind is
Decom 2% 3%
3% coming from 0° towards 180°, as seen in Figure 1.2. In the
Floater figures, 0° means OrcaFlex wind/wave/current direction.
O&M 18%
25y
33%

Turbine
Installation 25%
7% Mooring Cable
3% 6%
FIGURE 1.3: Floating Wind Farm cost items. Installation cost in the
figure includes sub-station. When re-arranged, mooring and dynamic
cable capital, installation and decommissioning costs account for 11%.
[11]. FIGURE 2.1: 2.5 meter calculation mesh model of TaidaFloat .
However, this is not true for all mooring configurations. As TABLE 2.1: TaidaFloat’s principal dimensions
seen in Figure 1.1, when an outside force acts in a certain Length 81.6 m
direction, only lines making an angle of more than 90° with the Breadth 94.2 m
load direction are tensioned – while others are slack, because a Height 34.75 m
chain or rope can’t transfer compression loads. The amount of Draught 20 m
mooring reaction the line provides is proportional to the cosine Total displacement 20 300 t
of this angle’s supplement. Literature review shows many Hull weight 4142 t
designs fix on an angle of 10° between mooring lines without Fairlead 3.75 m above keel
much consideration of the consequences. This is fine in intact Pontoon centroid at
Origin
condition, but in the case that the second most tensioned mooring base plane
line is damaged, the highest tension line will have to bear the
whole disturbing force alone and will break if not made overly TABLE 2.2: Site environmental conditions for DLC 6.1
thick. Wind/Wave/Current 50-Year return period
A proposed way to retain the benefit of low cost is to allow Wind speed at hub height
57 m/s
more mooring lines to counteract the outside force. Shown in (10-min average)
Figure 1.1 are some possible configurations with mooring spread Wind spectrum Full-field IEC Kaimal
angles between 10° and the widest angle, 126°[1]. Note here that Significant wave height (Hs) 12.7 m
at the 126°, there are two mooring lines connecting two platform Peak enhancement factor (γ) 2.08
columns to the one anchor. This is the innovative mooring design Wave period (Tp) 11.8 sec
known as the V-Share mooring [1]. Wave spectrum JONSWAP
The current speed at the
1.59m/s
2. FLOATER AND MOORING MODEL surface
To put the configurations to a test, their maximum tensions Current profile Power law method profile
and offsets are calculated in a dynamic coupled hydro-aero- Tides 3.13m to -2.8m
servo-elastic simulation using the time domain simulation Water depth 70 m
software OrcaFlex.
The moorings are attached to TaidaFloat – an open-source
floater specifically designed for environmental (typhoons) and
industrial conditions of Taiwan Strait and East Asia [3][5][14],
as other open-source floaters would not be feasible in storm

3 © 2023 by ASME
 Seed sensitivity study was performed after the completion of the
study. The chosen number of seeds (6) is slightly lower than
optimal (7-8), but still acceptable, as seen in Figure 2.5

FIGURE 2.2: OrcaFlex dynamic simulation of case 126°. White

dots represent contact with the seabed.

RAOs were calculated in OrcaWave with a 2.5 m mesh and

an example is shown in Figure 2.3; natural periods for heave,
pitch and roll are coupled due to 2 different column sizes [15],
and the pitch eigenperiod is 23 s, much higher than the
environmental waves peak period Tp 12 s. The natural period of
the system is reasonably far from the wave period range.
Damping based on tank test results was added in the OrcaFlex
directly. FIGURE 2.5 Seed sensitivity in OrcaFlex
Pitch Heave
1.00 3. MOORING ANALYSIS TO COMPARE DIFFERENT
23 s
RAO (m/m or 100*rad/m)

PATTERNS
0.80 3.1 Mooring Patterns of Various Spreading Angles
As the base case, 160 mm stud chain is used with anchor radius
0.60 840 m. In the first phase, where the most critical wave direction
is sought, a 160 mm diameter chain is used for all simulations,
0.40
as the absolute value of tension does not matter in this case. The
0.20 mooring analysis results for the 30°, 45°, 60°S, and 120° were
between the critical cases of 10°, 60°, and 126°, so only the latter
15 s Period, s
0.00 are shown for graphic clarity.
0 20 40 60 80 The purpose of the mooring system is to limit the offset,
the results for offset are shown in Figure 3.1.1, offset is defined
FIGURE 2.3: Damped RAOs of TaidaFloat platform
as the geometric sum of the platform’s X and Y axes
For all the simulations, at least 3 seeds were calculated for displacement from the initial position. The offset limit is
each parameter set and their results were averaged. During the assumed at 20 m, i.e. 28 % of water depth, exceeding this value
cost comparison, 5 seeds per set were used. might damage the dynamic power export cable [16].
The simulation duration was 3600 s (increased to 10800 s The offset is directly connected to the pre-tension. In all
for response surfaces simulations), and finally, the chain element the simulations, all the lines were pre-tensioned to 2000 ± 100
length was set to 10 m to save processing time without kN i.e. 8% of MBL. First, the offset is limited by the catenary
compromising accuracy. A mesh sensitivity study was done on shape of the chain, as the floater has to counteract the catenary
the chain element length, results are shown in Figure 2.4 segment’s weight to move. The higher the pre-tension, the faster
the mooring line becomes taut, at this point, any further offset is
Tension, MN Time, h restrained, while the tension is increasing rapidly.
Space, 100 MB Limiting the offset in the intact condition is important for
30 protecting the dynamic cable and preventing its fatigue, offset
results are presented in Figure 3.1.1. The offset varies a lot
20 depending on the wave direction, also, case 10°’s peak offset
happens at wave direction for which case 126°’s offset is the
10 lowest and vice versa. This implies different cases would need to
be rotated when considering a wind farm arrangement with a
prevailing wind direction. The maximum offsets from 10°, 60°,
0
and 126° are 16.9m, 14m, and 15.7 m, respectively. Among these,
0 5 10 15 20 the 60° separate angle configuration exhibits the lowest
Segment length, m
maximum offset while the 10° case results in the highest
FIGURE 2.4: Mesh sensitivity of mooring line with OrcaFlex

4 © 2023 by ASME
maximum offset. While keeping the offset under the limit, the mooring
system should be as economical as possible. The smaller the
tension, the cheaper the system will be. The maximum tension is
also subject to a safety factor according to ABS rules [12]. In the
results for maximum tension shown in Figure 3.1.3, the safety
factor of 1.67 was applied to the intact condition tension and a
factor of 1.05 was applied to the damaged condition. The
maximum tensions from 10°, 60°, and 126° configurations are
22719kN, 25619kN, and 26397kN, respectively. The 10°
FIGURE 3.2.1: Safety factors for mooring lines from ABS separate angle configuration gives the lowest maximum tension.
[12] FOWT Rules, 8-3/3 During the first stage of the numerical analysis, it was
found that damaged tension results seem more unpredictable, so
In the previous simulations, the same chain diameter (160 simulation time was increased from 1h to 3h per seed to ensure
mm) was used, and the resulting offset varied among mooring reliability. It was shown that about 8.1 hours of total simulation
patterns. To compare the cost fairly, the mooring systems must time is enough for mooring tension result being independent
achieve the same performance criteria – an offset limit from seed [18]. The cases, where the lines are grouped (10°, 126°)
Also, in Figure 3.1.1, the tilt angle which is the geometric sum are characterised by lower intact tension, and higher damaged
of pitch and roll is shown. Generally, all the heel results are quite tension as compared to spread mooring (60°), because in
similar among configurations – this is because the heel is most damaged condition, they only have 1 active line, while others are
impacted by the stability of the platform and the volume of added slack. Damaged and intact tension SF graphs look similar, as
buoyancy as the floater inclines. This volume is determined by seen in Figure 3.1.3. Damaged tension results are not further
the hull configuration of the platform and the inclination discussed, as for redundant systems, intact tension multiplied by
direction, which are only slightly influenced by the mooring safety factor is always higher than damaged condition tension.
system, however, case 10° does have a considerably higher The tension values are shown in Figure 3.1.4.
average tilt. However, this paper focuses on the mooring system design
Yaw motion is an important factor for the offshore wind and cost, and therefore only intact offset and damaged tension
platform design. It is a leading contributor to wind turbine will be considered in the cost comparison later.
system reliability failure, as well as wind tower design [17]. Case 60°S is superior in offset, but having the chains stem
The yaw motions results are shown in Figure 3.1.2. The results out of the middles of pontoons would make the dynamic power
indicate a considerable reduction in yaw from the 126° cable design quite challenging, as it would have even less room
configuration, while the 10° case exhibits the largest yaw motion. to move around so it was discarded.
The maximum yaw angles from 10°, 60°, and 126° configuration
are 13.3°, 7.9°, and 6.5°, respectively. The 126° separate angle
configuration reduce the yaw motion by 52% compared with the
10° case. However, the added stiffness in Yaw also produces
tension to stop the yaw rotation, which is one of the reasons case
126° has higher tension.

FIGURE 3.1.3: Damaged and Intact Tension actual safety factors for
different spreading angles. The wave directions corresponding to
minimum values are used for tension response surfaces.

FIGURE 3.1.2: Intact condition Yaw max values for different

spreading angles. The bow is facing the 0-degree direction.

5 © 2023 by ASME
 ° ° °
° ° °

° °

FIGURE 3.1.4: Damaged and Intact maximum line tension.

These cases can also be viewed through the lenses of
offset-stiffness relationship, shown in Figure 3.1.5. Due to the
platform’s triangle shape and the difference in spreading angles,
their stiffness in surge and sway direction can be different. The FIGURE 3.1.5: Mean environmental forces acting on the floater.
graph indicates that the 10° and 126° patterns have a big gap While the theoretical mean force in a regular wave is
between their maximum and minimum stiffness, while for 60° assumed 0, in real and time-domain simulations of FOWT,
both are almost the same. 10° is slightly stiffer than 126°, but observing a non-zero mean first-order wave force can result from
have the same average stiffness. several factors. Hydrodynamic damping and added mass effects
influence the dynamic response of the structure, potentially
leading to a non-zero mean due to phase differences.
Additionally, wave directionality and spread, if not perfectly
aligned with the FOWT's principal axes, may not allow wave
forces to symmetrically cancel out. Mooring and restoring forces
contribute to the structure's mean position, and inaccuracies in
wave modelling or spectrum representation, along with the
specific methods used for force analysis in the software, might
also contribute to discrepancies. OrcaFlex also employs
techniques like kinematic stretching to extend the linear wave
theory beyond its basic assumptions, further complicating the
direct application of the zero-mean expectation from linear
theory.
It can be seen that the 15 MW FOWT is not dominated by
wave, wind or current forces, wave force magnitude is
FIGURE 3.1.5: Restoring force of mooring patterns, large difference comparable to combined current and wind forces. 2nd order (drift)
between max and min implies non-linear stiffness. force varies nonlinearly with environmental direction. The mean
Another perspective of load difference between mooring forces are functions of the floater only and the mooring pattern
patterns is provided by the scale of each environmental force’s doesn’t affect them.
impact. Even as the loads are omnidirectional, environmental When the peak of environmental force sum coincides with
forces (2nd order wave forces, current and wind) contributions the mooring line direction – largest tension occurs. That’s the
change with load’s direction. Unlike a usual simplified approach second reason for tension difference between different mooring
which takes force values from static analysis, here force values patterns. The first and main reason is that the added stiffness in
are taken from dynamic simulation, allowing to compare means, Yaw also produces tension to stop the yaw rotation.
shown in Figure 3.1.5. Worth noting, mean wind force distribution curve is very
smooth and magnitude almost always slightly larger than current
force. However, at angles where current flows directly into
column’s face, the current force can exceed the wind force; the
distribution curve is much more uneven, too.

6 © 2023 by ASME
3.2 Offset as Function of Chain Size and Tension
In practice, the mooring lines are subject to a safety factor Line tension, kN
set by the class society rules, such as ABS FOWT rules, shown 30000
in Figure 3.2.1. As the designed 3x2 systems are redundant, the 27000
tension results of all simulations, are multiplied by a safety factor 24000
of 1.67 for intact condition and 1.05 for damaged condition. 21000
18000
Other class societies have different safety factors [19], and this
15000 190
value is critical as the intact condition becomes the governing 12000
design case, as opposed to the damaged condition being 155

753
1400
2000
2600
governing in Oil & Gas.

3200
120

3800
4400
5000

5600
Having a complete set of tension and offset results for all
wave headings for different spreading angles, for each spreading 12000-15000 15000-18000 18000-21000
angle configuration two wave headings were used: one resulting
21000-24000 24000-27000 27000-30000
in maximum tension in intact or damaged condition and one with
maximum offset in intact condition. These wave headings are FIGURE 3.2.3: Tension response surface for 10° case. (pre-
used in cost comparison as the chain diameter is chosen based on tension subtracted)
the minimum breaking load (MBL). R4S chain from Vicinay
catalogue is chosen[20]. For the 160 mm specimen, this chain’s
MBL is 24 281 kN as per ABS rules. Offset, m
40
35
30
25
20
15
10
5

753
120

1600
140

2400
160

3200
4000
180
FIGURE 3.2.1: Safety factors for mooring lines from ABS

4800
200
[12] FOWT Rules, 8-3/3

5600
220
In the previous simulations, the same chain diameter (160
mm) was used, and the resulting offset varied among mooring 5-10 10-15 15-20 20-25 25-30 30-35 35-40
patterns. To compare the cost fairly, the mooring systems must FIGURE 3.2.4: Offset response surface for 60° case.
achieve the same performance criteria – an offset limit.

Offset, m
Line tension, kN
40
35 33000
30
25 30000
20
15 27000
10 24000
5
753

21000
120

1600
140

2400

18000
160

3200
4000
180

15000
4800
200

190
5600
220

12000
155
753
1400
2000
2600

5-10 10-15 15-20 20-25 25-30 30-35 35-40

3200

120
3800

4400

5000

5600

FIGURE 3.2.2: Offset response surface for 10° case.

12000-15000 15000-18000 18000-21000 21000-24000
24000-27000 27000-30000 30000-33000
FIGURE 3.2.5: Tension response surface for 60° case. (pre-
tension subtracted)

7 © 2023 by ASME
 Offset, m 𝑍 = 𝑏0 + 𝑏1 𝑋 + 𝑏2 𝑌 + 𝑏3 𝑋 2 + 𝑏4 𝑋𝑌 + 𝑏5 𝑌 2 (3.2.1)
40
To find the minimal chain weight for a given configuration,
35 a nonlinear solver was used, varying pre-tension and diameter as
30 long as the offset is 23 m and the tension factored by the safety
25 factor is less than MBL.
20 A very interesting finding is that no matter the spreading
15 angle, the lowest required line weight is achieved when the
10 pre-tension is 4-5% of MBL, so this value can be recommended
5 for future design.

753
The fact that the four corners of each response surface are
120

1400
135

2000
all located at different height shows that a significant
150

2600
3200
165

interaction effect is present, so that linear prediction

3800
180

4400
195

approaches (when chain diameter and pre-tension are varied

5000
210

5600

separately) are not suitable to predict the performance of a

mooring system.
5-10 10-15 15-20 20-25 25-30 30-35 35-40
TABLE 3.2.1: Numerical analysis setup for cost comparison
FIGURE 3.2.6: Offset response surface for 126° case. P - h ,
A ly P - , kN
%
Line tension, kN Bb
B
36000
B
33000 Bx .
30000 Mb
M
27000
M
24000 xb .
21000 b
18000
15000 190
12000
155
753
1400
2000
2600
3200

120
3800
4400

5000

5600

12000-15000 15000-18000 18000-21000 21000-24000

24000-27000 27000-30000 30000-33000 33000-36000
FIGURE 3.2.7: Tension response surface for 126° case. (pre-
tension subtracted)
To make all the configurations have the same offset, pre- TABLE 3.2.2: Response surface coefficients
tension was changed. For every case, simulations were run to Cas b b b b b b
find the tension and offset with parameter combinations shown °O . - . - . . E- - . E- .

in Table 3.2.1. Bxs and Sxb cases were added to ease the °T - . - . . . . - .
response surface computation. A brute-force evolutionary solver
°O . - . - . . E- - . E- .
was used to fit the data to Equation 3.2.1[21] and draw response
surfaces for offset and tension for each case, as shown in Figures °T . - . . . . - .

3.2.2-7, coefficients shown in Table 3.2.2. The tensions in . - . - . . E- - . E- .

Figures are shown with pre-tension not included, as the pre- °O
tension’s is an input variable. This part of tension is re-added at °T - . . . - . E- . - .

the final step when chain cost is assessed.

While offset response surfaces are rather similar, the
3.3 Preliminary Cost Comparison of Mooring System
tension surfaces’ shape differs. Mooring line’s stiffness consists
The comparison of different system’s costs is shown in
of elastic part which depends on the chain length and diameter
Figure 3.3.1, also compared to the 3x3 mooring system [7] cost.
and of geometric stiffness, which depends on the pre-tension and
As the 3x3 system had offset of 23 m, 3x2 systems in this cost
spreading angle. Spreading angle also impacts the elastic
compare also designed for maximum offset of 23 m, the mooring
stiffness by increasing line length, but only up to 2.5%, while the
chain was since updated due to floater modification, and the 3x3
geometric stiffness is changing considerably, that’s why the
system uses 171 mm chain.
shapes of tension response surfaces can be rather different.

8 © 2023 by ASME
 First step is calculating the chain cost with Eq. 3.3.1. For configuration overall.
3x2 systems, chain cost is the main contributor to the overall cost.
Mooring cost, mil. USD
𝐶𝑜𝑠𝑡𝑐ℎ𝑎𝑖𝑛 = 𝑤𝑒𝑖𝑔ℎ𝑡 × 𝑢𝑛𝑖𝑡 𝑝𝑟𝑖𝑐𝑒 (3.3.1) 35
𝑤𝑒𝑖𝑔ℎ𝑡 = 𝑙𝑖𝑛𝑒 𝑙𝑒𝑛𝑔𝑡ℎ × 𝑢𝑛𝑖𝑡 𝑤𝑒𝑖𝑔ℎ𝑡 × 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑙𝑖𝑛𝑒𝑠
30
3x3 3x2
Then, the overall cost of the mooring system including the
installation and anchor cost are added in Eq. 3.3.2.
25
s
𝐶𝑜𝑠𝑡𝑎𝑛𝑐ℎ𝑜𝑟 𝑐𝑎𝑝𝑖𝑡𝑎𝑙 = 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑎𝑛𝑐ℎ𝑜𝑟𝑠 × 𝑢𝑛𝑖𝑡 𝑝𝑟𝑖𝑐𝑒 ×
𝑠ℎ𝑎𝑟𝑖𝑛𝑔 𝑐𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 (3.3.2) 20
𝐶𝑜𝑠𝑡𝑎𝑛𝑐ℎ𝑜𝑟 𝑖𝑛𝑠𝑡𝑎𝑙𝑙𝑡𝑖𝑜𝑛
= 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑎𝑛𝑐ℎ𝑜𝑟𝑠 15
𝑖𝑛𝑠𝑡𝑎𝑙𝑙𝑎𝑡𝑖𝑜𝑛 𝑡𝑖𝑚𝑒
×( 10
2
+ 𝑡𝑒𝑠𝑡 𝑡𝑖𝑚𝑒) × 𝑣𝑒𝑠𝑠𝑒𝑙 𝑟𝑒𝑛𝑡
, where installation time is half of the mooring unit pre-lay time 5
(3.9 days) and test time is anchor test time (2.5 days) from [8]
0
A shared anchor might need to withstand larger load. To 10°3x3 10° 60° 126° 126°S
account for that, sharing coefficient is used, the anchor is
assumed to be larger, and the unit anchor cost is multiplied by Chain cost Anchor capital cost
1.5 for anchor shared by two lines (126S and 10S), and by 2.0
Anchor installation cost Mooring installation cost
for anchor shared by three lines (10S 3x3).
The installation cost is calculated in Eq. 3.3.3 with time Soil test cost
data from Japanese projects compiled by T. Hasumi [8]. It did
not consider anchor and line installation time separately, so it is FIGURE 3.3.1: Mooring cost comparison.
split in half here. It was found that the total mooring system cost’s largest
contributor is the chain cost in the case of spread or V-Share
𝐶𝑜𝑠𝑡𝑖𝑛𝑠𝑡𝑎𝑙𝑙𝑎𝑡𝑖𝑜𝑛 = 𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑚𝑜𝑜𝑟𝑖𝑛𝑔 𝑙𝑖𝑛𝑒𝑠 × mooring (60° or 126°), and vessel cost in the case of clustered
𝑖𝑛𝑠𝑡𝑎𝑙𝑙𝑎𝑡𝑖𝑜𝑛 𝑡𝑖𝑚𝑒 mooring (10°). Anchor’s capital cost also contributes a large part,
( + ℎ𝑜𝑜𝑘 𝑢𝑝 𝑡𝑖𝑚𝑒) × 𝑣𝑒𝑠𝑠𝑒𝑙 𝑟𝑒𝑛𝑡 (3.3.3) more than the vessel cost in some anchor cases.
2
,where hook-up time (0.5 days) is from [8] When compared to the reference 3x3 system, 10° 3x2
configuration is a fifth cheaper. Overall, 126° (V-Share)
Finally, as two-three nearby anchors only require one mooring system was found to be the cheapest. However, there
geotechnical survey, soil test cost is number of anchor clusters may be some potential challenges when sharing one anchor
times unit cost of 100 000$. It is only different for case 60° as all between different lines, such as out-of-plane bending of the
the anchors are far from each other. anchor loops or the anchor’s insufficient moment holding
The vessel cost is estimated at 0.25 mil. USD/day for a big capacity.
Anchor Handling Vessel and the mooring chain price as 1500 TABLE 3.3.1: Preliminary cost comparison of different
USD/t based on industry experience. The cost data is shown in mooring patterns. Reduction for the maximum in each cost
Table 3.3.1. Tension is shown in percent from the average among category.
all the configurations for easier comparison. 10° 10° 60° 126° 126°S
Pattern
A companion paper [22] employing a different method
looked at the different spreading angles, but also different No. of lines 3x3 3x2 3x2 3x2 3x2
mooring radii. For the same mooring radius, the results are Line 171 234 360 274 274
almost the same, however, it was found that the mooring diameter mm mm mm mm mm
No. of
pattern’s response to different mooring radii varies somewhat. So, anchors 9 6 6 6 3
the cost analysis here might only be precise for a mooring radius
Chain cost -61% -34% -7% 0% 0%
of 840 m, which can be considered a medium value [22].
Anchor
The results show that the 10° configuration is the cheapest capital 0% -33% -33% -33% -50%
in terms of chain material costs. Only when anchor sharing is Anchors
used for 126°, it is cheaper than 10°. A 10° with a shared anchor install. 0% -33% -33% -33% -67%
would be even cheaper, but it’s chains would clash and get Mooring
damaged if they were attached to a single column. install. 0% -33% -33% -33% -33%
The 60° spread mooring configuration is the least cost- Soil test -50% -50% 0% -50% -50%
effective. This is also the more troublesome case to arrange on- Total 0% -17% -5% -3% -20%
site and connect the dynamic cable due to the mooring lines all
over the place. These factors make it the less-desired mooring

9 © 2023 by ASME
4. DISCUSSION Marit France R3 - 200 mm R4 -
The results show, that for FOWT in a typhoon region, a 180 mm
Tullyn China R4 up to 162 mm
grouped 10° system is the most cost-efficient. However, in the
Oil & Gas (O&G) Industry, spread 60° moorings are widely used
for a 6-line mooring system. The difference comes down to the It shall be noted here that the mooring technologies for
required safety factors between the two industries, FOWT and floating wind are evolving quickly, including new mooring
O&G. As there is no danger of accidents involving human lives configurations (V-Share mooring, shared mooring lines), new
or oil spills, safety factors required by class societies for mooring components such as polymer spring load reduction
damaged conditions are smaller for FOWT (1.05) than for O&G devices, new mooring materials such as HMPE rope [6, 19]. New
(1.25), making the intact condition the governing case for mooring anchors for floating wind are also emerging quickly, e.g.
mooring design, as opposed to the damaged condition being the multiline ring anchor, helical anchor. The direction these
governing for O&G platforms. The safety factors change as rules new mooring technologies take will influence the overall
are updated, and are different among class societies, therefore the mooring system cost.
governing case and best mooring patterns are also subject to In this study, the 126° case, i.e. the V-Share mooring [1],
change. demonstrates over a 50% reduction in maximum yaw motion,
Mooring radius is another main design factor for mooring and 16% higher tension than the 10° case. The main reason for
systems. Companion paper showed that different mooring the higher tension is the much-increased yaw stiffness. It's
patterns have different sensitivity to it, so this cost comparison’s expected that the performance of the V-Share mooring might
limitation is its applicability to medium mooring radius only vary across different floaters, especially considering variations
(840 m). in pontoon size (smaller or deeper) which directly impact the
When it comes to anchor sharing between lines, there is a dominant wave loads and platform length, impacting yaw
lot of uncertainty on its feasibility. Potential problems include stiffness.
anchor fatigue and difficulty of installation. The way these Given the substantial chain size used in this study,
problems are solved will impact the cost of the mooring system. significantly increasing the chain size becomes necessary to
Also, the shared anchor will most probably be a pile anchor achieve the targeted tension safety factor and offset in the 126°
[4][13], while drag anchors can be used for unshared systems. In case. Yet, enlarging the chain size might be not the most cost-
the present paper, pile anchors were used for all cases, as the efficient approach for this objective. Consequently, it is
site’s soil condition might not allow drag anchors. It follows that anticipated that the V-Share mooring will showcase more
whether shared anchors are economically viable depends on the advantages in mooring systems that allow for more parameter
site’s soil condition and anchor design. Installation vessel cost is adjustments, including variations in component size and lengths,
the largest part of the 3x3 mooring system cost, and it is prone to azimuth angles, and material grades.
much more severe changes than material costs. In Taiwan, vessel A shared anchor connecting two or three mooring lines
hire costs rose more than 2 times after more wind farm projects among different floaters is a field-proven design, such as the
were announced, due to a spike in demand with slowly growing Hywind Tampen project. However, connecting two mooring
supply. In case the vessel prices fall by a large margin, the 3x3 lines from a single floater to one anchor is novel. The V-Share
system might become less expensive than 3x2 due to its lower mooring system also provides an opportunity to optimize the
chain cost, which is less subjected to a volatile market. The mooring installation procedure [1]. The evolution of V-Share
bottom line is, based on supply constraints, the available chain mooring technology is expected to influence the cost of the
size is limited, so a 3x3 system must be used even if 3x2 could mooring system for floating wind.
be cheaper theoretically. At the time of writing, only one Finally, this paper primarily focuses on assessing the
company on Earth – Vicinay – claimed to have made chain with mooring system. The influence of the mooring system on hull
diameter of 220 mm, as seen in Table 4.1, and no bigger chain and turbine size remains unexplored. This study employs an all-
could be used. Chains so big had never been tested yet, so their chain mooring system in shallow water with a specified radius.
use for a large-scale project soon is unlikely. It was found that the optimal pre-tension for such system is 4-
TABLE 4.1: Worldwide Offshore Mooring Chain 5%, making chain size the sole adjustable parameter to attain the
Manufacturing Capability. R3-6 refer to metal grade. desired performance. Concerning performance, the focus is on
American Petroleum Institute Country of Production tension and offsets. Recent engineering observations show that
Approved Vendor List: supplier capability fatigue might govern the mooring design, especially because of
Asian Star China R6 up to 200 mm the impact of dynamic wind turbine loads on top of wave
Zheng Mao China R6 up to 208 mm frequency fatigue damage. These are all topics that deserve to be
Vicinay Spain R6 up to 220 mm, further studied.
capacity 40 000
t/year
Hamanaka Chain Japan R3 up to 132 mm 5. CONCLUSION
In summary, the economic viability of a mooring system is
Ramnas Sweden R5 up to 177 mm
Other suppliers:
intricately tied to both installation expenses and material costs.
Among the various components, the mooring chain plays a major
Daihan Korea R3 up to 162 mm
DeYuan Marine China Up to 162 mm
role alongside anchors, clump weight, shackles, and connectors.
Interestingly, the cost dynamics reveal a counterintuitive
SOTRA Norway R4 up to 178 mm
relationship — as installation vessel prices soar, employing

10 © 2023 by ASME
fewer mooring chains proves more cost-effective. This E g g, . 11( ): p. .DOI:
phenomenon stems from the environmental load being h p: . g . j
distributed thinly over larger chains, even as the individual chain . X , K., l., Design and comparative analysis of
cost increases. alternative mooring systems for floating wind turbines
The research findings underscore the advantages of a in shallow water with emphasis on ultimate limit state
clustered mooring system featuring a 10° spreading angle design. Oc E g g, . 219: p. .DOI:
between lines. This configuration emerges as the most cost- h p: . g . j. c g. .
efficient choice when compared to traditional alternatives such
. J g N , H., New Energy Generation Equipment
as the spread 60° mooring, which shows a small offset in both
Business Study. , l k c . A l bl
damaged and intact conditions. Note that the 60° spread is more
common for applications like offloading buoys and small f :
production vessels in the oil and gas industry. Moreover, the h p : p f. fcf .c p f H _AP
prospect of sharing anchors between two neighbouring lines in _ .p f? .p f.
126° configuration (V-Share) holds promise for even more . M , K.-T., l., Mooring System Engineering for
substantial cost savings. This strategy becomes particularly Offshore Structures. : El . A l bl f :
relevant in cases where site soil conditions present challenges to h p: .c c c .c b k
the use of drag anchors. g- y - g g-f - ff h -
It is imperative to address the feasibility of a 3x2 mooring c .
pattern for a 15MW semi-submersible FOWT in typhoon-prone . h , .-A., l., Design Of Mooring System For a
regions. Regrettably, this configuration currently proves 15mw Semi-Submersible, Taidafloat,inTaiwan Strait,
unfeasible for the typhoon condition of the site, when 42nd International Conference on Ocean, Offshore and
considering the 180 mm practical size limit of mooring chain. Arctic Engineering. , A ME: M lb ,A l .
However, a strategic shift to a 3x2 pattern would yield A l bl f :
considerable advantages, reducing the total cost by utilizing h p: . chg . p bl c
three fewer mooring lines compared to a 3x3 pattern in typhoon _OMAE -
regions.
_DE IGN_OF_MOORING_ Y TEM_FOR_A_
In conclusion, the optimal mooring pattern for a 15MW
M _ EMI-
semi-submersible FOWT in typhoon-prone regions involves a
clustered arrangement with either a 10° or 126° spreading angle. UBMER IBLE_TAIDAFLOAT_IN_TAI AN_ TRAIT_D
These configurations emerge as the top choices, outperforming g _L _ _E M_Ex _ _p _M l_
alternatives like 45° and 60° options. The findings may help E _Ex _ _ _E M_Ex _ _M
economic viability of mooring systems for floating wind, but l_E LR_Ex .
also contribute to the advancement of sustainable and cost- . H , T., l., ESTIMATION FOR EFFICIENCY OF
effective offshore wind energy solutions. OFFSHORE INSTALLATION PROCESS OF FLOATING
OFFSHORE WIND TURBINES IN JAPAN, 42nd
ACKNOWLEDGEMENTS International Conference on Ocean, Offshore and Arctic
The authors highly appreciate the funding support from the Engineering. , A ME: M lb ,A l .
government agency, the Taiwan (ROC) Ministry of Science and . M , K.-T. Y. F , Harvest Wind - Offshore Engineering.
Technology, for the FOWT research, and thank the Taiwan (ROC) , h p & Oc T ch l gy R ch ,
Ministry of Education’s Yushan fellowship programme. N l T U y: T p . A l bl f :
h p: b g.c pc l
REFERENCES ?l c l = .
. , Y., L ch, P., P ., Næ E., A New
. , L., l., WP4 Innovation in Low Carbon
Mooring System for Floating Offshore Wind Turbines:
Design and Manufacturability
Theory, Design and Industrialization, Offshore
Task 4 – Mooring and Anchoring Systems, CORNWALL FLOW
Technology Conference. :H , U A. A l bl
ACCELERATOR. , Off h R bl E gy
f : h p: p . g OT ONF p c g-
p l. A l bl f :
b c OT - OT D R .
h p: .c p l . g. k p-
. h g, H.- ., l., Prospects of offshore wind power in
c pl -RPT- FAR-O - -
Taiwan and the development of semi-submersible
- P -T k- -A ch g- -M g.p f.
floating platforms, Taiwan Wind Energy Symposium.
. BVG A c , Guide to a Floating Offshore Wind
: T , T . A l bl f :
Farm. , Th E E
h p: . chg . p bl c
c l : Off h R bl E gy p l.
_ l f g zh q j gy b q hf zh f
A l bl f :
zh .
h p: g fl g ff h .c p-
. I , G., I.-J. H , K.-T. M , Design Considerations
c pl BVGA- -Fl g-
on Semi-Submersible Columns, Bracings and Pontoons
G - .p f.
for Floating Wind. J l f M c c
. AB , Guide for Building and Classing Floating Offshore

11 © 2023 by ASME
 Wind Turbine Installations, The American Bureau of . 25( ): p. - .DOI: . - -
Shipping, Houston (TX), USA. . A l bl f : -
h p: . gl . g c gl l - - . L , Z.-Y., l., Mooring Anchor radius and Spread-
g c ff h _f f -g - angle Optimization for a 2 MW Semi-Submersible
j ly .p f. Floating Wind Turbine in Taiwan Strait, International
. H , I.-J., l. Optimization of Semi-Submersible Hull Conference on Ocean, Offshore and Arctic Engineering
Design for Floating Offshore Wind Turbines. 41st (OMAE 2024). ,A c c y f M ch c l
International Conference on Offshore Mechanics and E g : g p .
Arctic Engineering (OMAE 2022). . H b g,
G y: A c c y f M ch c l
E g .DOI: h p : . g . OMAE -
. A l bl f :
h p: . g . OMAE - .
. I , G., I.-J. H , K.-T. M , Overview of FOWT
Demo Projects Cost and Analyses of Hull Design
Features, 3rd World Conference on Floating
Solutions (WCFS2023). , El : T ky , J p .
A l bl f :
h p: . chg . p bl c
_O _ f_FO T_D _P j c _ _ _A
ly _ f_H ll_D g _F .
. I , G., Investigation report. Column size influence
on RAO peaks. , N l T U y:
R ch G . A l bl f :
h p: x. . g . RG. . . . .
. Ikh ch , M., l., Review of the state of the art of
mooring and anchoring designs, technical challenges
and identification of relevant DLCs. , ORE IND.
A l bl f : h p: c . p-
c pl f l p bl c ORE IND-D . -
R - f- h - - f- h - - f- g- -
ch g- g .p f.
. K , M.G. P.H. D lh ff, Yaw Systems for wind
turbines – Overview of concepts, current challenges
and design methods. J l f Phy c : f c
, . 524( ): p. .DOI: . -

. l y, ., l., Sensitivity analysis of numerical

modeling input parameters on floating offshore wind
turbine loads. E g. c ., . 8( ): p. -
.DOI: . - - -
. M , K.-T., l., Mooring Designs for Floating Offshore
Wind Turbines Leveraging Experience From the Oil &
Gas Industry, 41st International Conference on
Offshore Mechanics and Arctic Engineering (OMAE
2021). , A c c y f M ch c l
E g . A l bl f :
h p: . g . OMAE - .
. Vc y , The Future of Mooring (Brochure).
A l bl f :
h p : c yc . b ch bl x.h
l#p= .
. h ck, L.R., l., Polynomial Regression with
Response Surface Analysis: A Powerful Approach for
Examining Moderation and Overcoming Limitations of
Difference Scores. J l fB P ych l gy,

12 © 2023 by ASME

View publication stats