International Journal of Naval Architecture and Ocean Engineering 14 (2022) 100451

Contents lists available at ScienceDirect

International Journal of Naval Architecture and Ocean Engineering

journal homepage: http://www.journals.elsevier.com/
international-journal-of-naval-architecture-and-ocean-engineering/

Development of jacket substructure systems supporting 3MW

offshore wind turbine for deep water sites in South Korea
Thanh-Tuan Tran a, b, Daeyong Lee a, *
a
Institute of Offshore Wind Energy, Kunsan National University, Jeollabuk-do, Republic of Korea
b
Faculty of Engineering and Technology, Quy Nhon University, Binh Dinh, Viet Nam

a r t i c l e i n f o a b s t r a c t

Article history: This study aims to develop jacket substructures supporting a 3MW Offshore Wind Turbine (OWT) in
Received 3 October 2021 water depth ranges of 25e40 m. A simplified structural design process is introduced and influences of
Received in revised form jacket topological forms and brace systems on their dynamic performances are assessed. The results
6 April 2022
conclude that the topological configuration of the jacket substructure has a significant influence on the
Accepted 7 April 2022
Available online 13 April 2022
dynamic characteristics of the whole system, indicating a high priority when designing a jacket sub-
structure for OWTs. Furthermore, two jacket substructures having the lowest material costs and the best
performances under the Korean ocean environmental conditions are suggested. Additionally investigated
Keywords:
Offshore wind turbines
in this study is a sensitivity of environmental loading directions on the structural performances of the
Jacket substructures jacket substructures. All the findings in this study might be very useful to assist the structural engineers
Water depth when they design the support structures for large-scaled OWTs in the near future in South Korea.
Topological forms © 2022 Society of Naval Architects of Korea. Production and hosting by Elsevier B.V. This is an open
Brace systems access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction the water depth and distance from the shore, can be calculated
through the adjustment factors. For example, the total costs for
In recent years, Korea has been increasing embracing renewable foundations at water depths ranging from 40 to 50 m are 1.32 and
forms of energy. According to the “Renewable Energy 3020” plan 1.13 times greater than the cost for water depths of 20e30 m and
(Kim et al., 2020), the government will focus attention on solar and 30e40 m, respectively (Kang et al., 2011). Thus, a suitable sub-
wind energy, with the ambitious goal of generating about 16.5 GW structure plays a significant role since it can reduce the total cost of
using wind power by 2030. With regards to this energy, the future wind projects efficiently.
lies in developing offshore wind farms to take advantage of higher Currently, various substructures (i.e., monopile, suction caisson,
offshore wind speeds and less turbulent airflow. tripod, or jacket) are widely installed in offshore wind farms
At present, South Korea has more than 90 Offshore Wind Farm worldwide (Kim et al., 2014, 2016; Kim and Lee, 2015; Sandal et al.,
Projects (OWFPs). Among them, five OWFPs are currently oper- 2018; Tran et al., 2020). A typical monopile comprises a steel tube
ating, and the other projects have been confirmed or are in the pile that is popularly installed in shallow waters. In contrast, the
planning stage (4C Offshore, 2021). As known, almost all of the jacket and tripod systems are mainly deployed for use in deeper
current Offshore Wind Turbines (OWTs) are deployed in the tran- water (Kim et al., 2016; Tran et al., 2020, 2022). The tripod system is
sitional water depth. Thus, in order to meet the rising energy de- composed of three steel piles arranged in an equilateral triangle,
mands, it is necessary to develop OWFPs in deeper waters for the while the jacket substructures are assembled from steel tubular
upcoming projects (Oh et al., 2018; Shi et al., 2015). As stated by members comprised of a space frame structure (Wu et al., 2019).
Refs. (EEA, 2009; Kang et al., 2011; Oh et al., 2018), increasing water For the Korean environmental conditions, the jacket substructures
depth leads to increasing overall costs of the offshore project. The are recommended considering their feasibilities and availability of
cost of an offshore structure at a specific site, which is a function of installing equipment (Kim and Lee, 2015; Shi et al., 2013a, 2013b;
Tran et al., 2022). Consequently, the current work only studies the
jacket substructures.
* Corresponding author. This study focuses on the development of jacket substructures
E-mail addresses: [email protected] (T.-T. Tran), daeyong.lee@ supporting an existing 3 MW wind turbine at different water
kunsan.ac.kr (D. Lee).
depths (i.e., 25 m, and 40 m) under Korean environmental
Peer review under responsibility of The Society of Naval Architects of Korea.

https://doi.org/10.1016/j.ijnaoe.2022.100451
2092-6782/© 2022 Society of Naval Architects of Korea. Production and hosting by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://
creativecommons.org/licenses/by-nc-nd/4.0/).
T.-T. Tran and D. Lee International Journal of Naval Architecture and Ocean Engineering 14 (2022) 100451

conditions. The jacket substructures are designed based on the linearly between 4.5 m (bottom) to 3.07 m (top), and their corre-
main design requirements. First, the dynamic characteristics of the sponding thicknesses vary linearly from 34 mm (bottom) to 18 mm
whole system are evaluated to fix the target frequencies, followed (top), respectively.
by strength check under the possible situations according to codes
(DNV-OS-J101, 2014; DNVGL-ST-0126, 2018; IEC 61400-1, 2005; IEC
3. Design process for jacket substructures
61400-3, 2009). Moreover, the material consumption is also
compared to select the most cost-effective design. A significant
The simplified design procedure for the OWT jacket substruc-
factor here is that the reasonable jacket substructure for each water
ture is introduced in this section. The input data (i.e., turbine ge-
depth will be selected through a feasibility analysis. The objectives
ometries, site-specific environmental data, and geotechnical data)
of this study are summarized as follows:
for the design process have been collected from the specific project
(POSCO, 2017). The initial configurations (i.e., jacket height, leg

Providing the design procedure to develop jacket substructures
spacing, jacket slope, and number of jacket layers) are selected. The
supporting offshore wind turbines
jacket modeling is then completed with the main design re-

Proposing reasonable jacket substructures for upcoming
quirements in terms of natural frequency and structural integrity
offshore wind farms in South Korea
(DNVGL-ST-0126, 2018; IEC 61400-1, 2005). The design process for

Identifying the critical response-critical directions of jacket
the jacket substructure includes the three main steps, which is
substructures when being installed
depicted in Fig. 2. The process is considered to be complete when
the structural integrity satisfies the design requirements (DNVGL-
ST-0126, 2018; IEC 61400-1, 2005).
2. Description of reference wind turbine

The reference wind turbine used in this study is adopted from 3.1. Step 1: selecting initial jacket dimensions
the POSCO (2017). The main characteristics of the turbine are
shown in Fig. 1. It is a three-bladed upwind design with 3 MW rated The main parameters of the jacket are depicted in Fig. 3 and the
power. The rotor consists of the blades and hub with a total mass of relevant formula are summarized in Fig. 4. It is assumed that the
64.6 ton. The nacelle houses the gearbox, generator, shaft, brake jacket substructure does not have to be limited to a particular
disc, etc., having a mass of 128.0 ton. These components are con- number of legs or bays. These parameters are derived as follows:
nected together, forming the Rotor-Nacelle-Assembly (RNA). The
tower is in a steel tubular shape with a height of 56.77 m, providing
Jacket height ðhJ Þ is defined as the distance from the seabed to
support to the RNA. The diameters of the tower are assumed to vary the bottom of the transition piece (TP). This parameter is chosen

Fig. 1. Reference 3 MW OWT (POSCO, 2017).

2
T.-T. Tran and D. Lee International Journal of Naval Architecture and Ocean Engineering 14 (2022) 100451

Fig. 2. Design process for a jacket substructure.

support the existing 3 MW OWT; the configuration of TP is taken

from the previous references (POSCO, 2017; Tran et al., 2022).

The leg top spacing ðLtop Þ is then determined from the config-
uration of the transition piece.

Next, the initial slope ðmÞ of the jacket substructure is defined.
According to Refs. (El-Reedy, 2014; Jalbi and Bhattacharya,
2020), m varies from 1:6 to 1:10.

Spacing of the leg bottom ðLbottom Þ is a function of hJ ; Ltop ; and m
parameters, and it is calculated using Eq. (2).

Lbottom ¼ Ltop þ mhJ (2)

A number of jacket layers ðnÞ are designed based on NORSOK-

N004 (2004). It states that the angle between brace and leg
should be in the range from 30  to 90  .

Finally, the bay lengths and heights of each layer are
determined.

3.2. Step 2: natural frequency analysis

Fig. 3. Main parameters of jacket substructure.
In this step, the support structure will be checked with the
allowable frequency range (POSCO, 2017). This is important to avoid
such that there is no effect of the splash zone on the transition
the resonance effects caused by the rotor or blade passing.
piece. And hJ is calculated as follows:

3.3. Step 3: extreme event analysis

hJ ¼ HSWL þ g (1)
This step aims to evaluate the structural integrity of the jacket
where, HSWL is the high still water level in the considered envi- substructure. The loads acting on the jacket substructures will be
ronmental conditions; and g is the air gap, which is at least 20% of determined first, followed by static numerical analysis and strength
the 50-year significant wave height Hm;50 and a minimum value of check in accordance with standard designs (DNVGL-ST-0126, 2018;
1.0 m (DNVGL-ST-0126, 2018). NORSOK, 2004).

The transition piece is a complex component, which aims to 3.3.1. Loading acting on the jacket substructures
transfer loads from the turbine to the jacket. It is difficult to
design and there is no general process for selecting the type of 3.3.1.1. Environmental loads (Env.)
TP. Configuration of TP depends on the tower geometry, number In this study, the jacket substructures are designed based on
of legs, DLCs, and the required spacing for the working platform, environmental conditions in the Korean Southwest Sea. The wind
etc. In this research, the jacket substructures are designed to and wave parameters are selected from metocean analysis
3
T.-T. Tran and D. Lee International Journal of Naval Architecture and Ocean Engineering 14 (2022) 100451

Fig. 4. Relevant formula for selecting jacket dimensions.

provided in POSCO (2017). The wind load of 50-year conditions is Moreover, the current load acting on the jacket substructure is
selected for evaluation of the extreme aerodynamic loads. The 10- also considered. The current loading model is assumed to be con-
min mean wind speed at the hub height (42.5 m/s) is selected for stant along the sea level with a speed of 1.04 m/s over the 50-year
designing the substructure. return period. The impact of marine growth is adopted with
For the hydrodynamic effects, wave data is considered in terms thickness ranging between 5 and 10 cm.
of wave height, wave period, and wave direction. The significant With regards to the loading directionality, different angles
wave height of the 50-year return period ðHs;50 Þ is 5.97 m, which is ranging from 0 to 360 at an interval of 15 are applied to the
taken from POSCO (2017). The corresponding zero-crossing period jacket substructure, as presented in Fig. 5. For the numerical anal-
ðTÞ are calculated as (DNVGL-ST-0126, 2018): ysis that processes the critical environmental loading conditions,
sffiffiffiffiffiffiffiffiffiffiffi sffiffiffiffiffiffiffiffiffiffiffi the environmental loads (wind, wave, and current) acting on the
Hs;50 Hs;50 jacket substructure are assumed to be coincident.
Tmin ¼ 11:1  T  14:3 ¼ Tmax (3)
g g
3.3.1.2. Design load cases (DLCs)
The 50-year maximum wave height is the function of 50-year
significant wave height and can be calculated using Eq. (4):
For the Ultimate Limit State (ULS) analysis, Design Load Cases
Hm;50 z1:86Hs;50 (4) (DLCs) at the tower base are usually generated from integrated
time-domain simulations using GH-Bladed (Bossanyi, 2010). These

Fig. 5. Directionality of environmental loading.

4
T.-T. Tran and D. Lee International Journal of Naval Architecture and Ocean Engineering 14 (2022) 100451

load cases are established to cover all possible operation situations 4. Modeling of jacket structures
occurring throughout the lifetime of the system, and they are
defined according to Refs. (Bo €ker, 2010; DNV-OS-J101, 2014; 4.1. General descriptions
DNVGL-ST-0126, 2018; IEC 61400-1, 2005; IEC 61400-3, 2009). The
Design Load Cases considered for the ULS analysis in this study can Using the design process described in Section 3, the jacket
be found in Tran et al. (2022). substructures are built to support the existing 3 MW wind turbine.
Different jacket configurations, categorized into two groups (i.e.,
4LJ-25 and 4LJ-40) at different water depths (i.e., 25 m and 40 m)
3.3.2. Structural integrity are developed (Fig. 6). The first group has a total height of 46 m,
installed at a 25 m sea level (Fig. 6a). Three brace levels that comply
Static numerical analysis is performed under DLCs using SACS with the criteria of a minimum angle of 30  between leg and brace
software (Bentley, 2019), and strength check will be done in are implemented. In the case of 40 m sea level, the jacket sub-
accordance with c. In order to measure the strength capacity of the structures have a total height of 61 m (Fig. 6b) and four brace levels.
jacket members and joints, the Unity Check ðUCÞ is utilized. The UC For each sea level, this study considered various topological
is defined as follows: forms (i.e., m ¼ 0 and m ¼ 10) and bracing systems (i.e., Pratt and
sm X-brace). The jacket substructures are developed with these four
UC ¼ (5) configurations (Fig. 6). Except for the topological form, all of the
sRd
other properties (material, jacket height, height of each layer) are
In which, sm and sRd are the actual and allowable stresses. The the same. The structural parameters of the developed models are
UC formulations of the tubular member and joint can be found in summarized in Fig. 6.
section 6.3 and 6.4 of NORSOK (2004).

Fig. 6. Modeling of jacket substructure at different water depths.

5
T.-T. Tran and D. Lee International Journal of Naval Architecture and Ocean Engineering 14 (2022) 100451

4.2. Finite element model 5. Feasibility analysis

The numerical modelings of the jacket support structures are This section aims to compare and evaluate the feasibility of
developed using the software program SACS (Structural Analysis jacket substructures developed for each sea level. In this regard, the
Computer System). The main components of the structure are the numerical analysis will cover the main requirements given in
tower, transition piece, and jacket structure. Section 3.
A detailed example of the jacket support structure (i.e., 4LJ-40-
2P) is shown in Fig. 7. The Euler-Bernoulli beam elements are
implemented to model the jacket substructure. The use of this 5.1. Eigen analysis
element allows the linear behavior for axial, shear, torsion, and the
cubic behavior for moment (Bentley, 2019). Cross-sections of the Eigen analysis is first performed to compute the natural fre-
jacket legs and brace members are modeled as tubular sections quencies of the offshore wind turbine system. This is necessary to
with the corresponding parameters as shown in Fig. 6. The leg avoid the resonance phenomenon caused by the vibration of the
member is divided into three segments corresponding to its ge- rotor (1P) and blade-passing (3P). The 1P and 3P ranges are
ometries, as depicted in Fig. 7d. The influences of the overlap and determined from the cut-in and rated rotor speed of the 3 MW
joint-can are also considered (Fig. 7e). The former is applied to OWT specification provided by Doosan Heavy Industry. In this
avoid the duplication of brace members. The latter is simulated by study, the soft-stiff design, which is known as the most common
increasing the size of the leg members at the intersection points. It design of the current support structures, is selected to optimize the
is assumed that the jacket substructures considered in this study economy. The comparative modal analyses of the jacket sub-
are installed using the pre-piling method; skirt pile connections at structures are reported for the first four natural frequencies, as
the jacket base are also considered in the analysis model (Fig. 7f). shown in Fig. 8.
In this research, structural properties of the tower and transition
piece are the same as the existing jacket substructure installed at
14 m water depth, and their detailed geometry information can be
5.1.1. Effects of the topological forms
found in Tran et al. (2021). The tower is composed of a combination
As seen in Fig. 8, the topological forms have an important effect
of 22 elements, with a total height of 56.77 m. The Rotor Nacelle
on the dynamic characteristics of the jacket. In Fig. 8a, corre-
Assembly (RNA) is assumed as a lumped mass, and this is assigned
sponding to the sea level of 25 m, there is a significant difference
at the tower top. In addition, four lumped masses are assigned
between the two basic configurations (i.e., m ¼ 0 and m ¼ 10).
along the tower, representing the masses of tower flanges, as
Particularly, natural frequencies of the second topological form
depicted in Fig. 7a. Regarding the transition piece, shell elements
ðm ¼ 10Þ are found to be higher compared to those of the first to-
are used for modeling the H-section beams, while the beam ele-
pological form ðm ¼ 0Þ. A maximum difference of 22.8% is found for
ments are used for modeling the tubular tube and support mem-
the Pratt bracing system. This is due to the higher stiffness when
bers (Fig. 7b). In this analysis, the access door in the cylinder will
considering the slope ðm s0Þ, which is explained in Fig. 10. For the
not be considered as using the beam element simulates the
sea level of 40 m, there are slight changes in natural frequencies,
cylinder.
with a maximum difference of up to 9.7%.

Fig. 7. Modeling of support structure.

6
T.-T. Tran and D. Lee International Journal of Naval Architecture and Ocean Engineering 14 (2022) 100451

Fig. 8. Comparison of natural frequencies: (a) 4LJ-25 and (b) 4LJ-40.

sffiffiffiffiffiffiffiffiffi
1 Keq
f0 ¼ (6)
2p Meq

In which, Keq and Meq are equivalent stiffness and lumped mass
of the system, respectively (Fig. 9).
The fundamental natural frequency ðf0 Þ is taken from the nu-
merical simulation and the equivalent lumped mass Meq is given by
Eq. (7).

Meq ¼ 0:243meq htotal þ MRNA (7)

In which, meq is the equivalent mass of the support structure,
and it is taken from Jalbi and Bhattacharya (2018).
ð hJacket ð hTotal
mJacket 42 dz þ mTower 42 dz
0 hJacket
meq ¼ ð hTotal (8)
Fig. 9. Equivalent stiffness and mass of the support structure.
42 dz
hJacket

5.1.2. Effects of the bracing systems The mode shape function 4 is evaluated using the following
Considering the bracing system, the X-type provides a higher equation:
stiffness than the Pratt type as expected. At the sea level of 25 m
 
(Fig. 8a), the differences in natural frequencies between X- and l l l l
4 ¼ b sin z  sinh z þ cosh z  cos z (9)
Pratt braces are 6.0% and 3.1% for the first ðm ¼ 0Þ and second L L L L
ðm ¼ 10Þ topological forms, respectively. At the higher water depth
of 40 m (Fig. 8b), an average difference of 4% is found between X- where l ¼ 1:8751 and b ¼  cos lþcosh l
sin lþsinh l
and Pratt type braces. Fig. 10 shows a comparison of the structural properties of
The variations in the natural frequencies of the designed sub- different jacket configurations. At the sea level of 25 m (Fig. 10a),
structures can be explained based on the influences of their mass the second configuration provides a higher stiffness to the OWT,
and stiffness. As stated by Jalbi and Bhattacharya (2018), the with an average difference of about 55.4%, while the difference of
dominant frequencies ðf0 Þ of the jacket system can be calculated as effective mass is about 5.6%. In the case of the deeper sea level
follows: (Fig. 10b), there are slight differences in effective stiffness and mass,

Fig. 10. Structural properties of the support structures: (a) 4LJ-25 and (b) 4LJ-40.

7
T.-T. Tran and D. Lee International Journal of Naval Architecture and Ocean Engineering 14 (2022) 100451

Fig. 11. Stiffness-to-mass ratio: (a) 4LJ-25 and (b) 4LJ-40.

Fig. 12. Allowable frequency ranges: (a) 4LJ-25 and (b) 4LJ-40.

with values of 16.3% and 1.7%, respectively. The stiffness-to-mass average differences of 9% and 7% for the Pratt and X-bracing sys-
ratios of all jacket substructures are presented in Fig. 11. As ex- tems, respectively.
pected, a higher ratio correlates with a higher natural frequency.
The ratio of X-brace is higher at 1.12 times compared to the Pratt 5.2. Ultimate limit state design
brace in both cases.
To evaluate the dynamic performance of jacket substructures
5.1.3. Efficiencies of the jacket substructures under the ultimate limit state, jacket substructures are analyzed
The efficiencies of the developed models are evaluated graphi- with twelve DLCs explained in Section 3.3. The maximum lateral
cally in Fig. 12. As seen in Fig. 12a, the first topological form (4LJ-25- displacement at the transition piece and the maximum stress of leg
1) shows the soft-stiff design property, which is known to be the members are analyzed and compared. The maximum lateral dis-
most common for the current offshore development. On the other placements are the sum of displacement in X- and Y- directions and
hand, the dominant frequencies of the second topological form they are observed at the top of TP, whereas the maximum stresses
(4LJ-25-2) lie in the resonance area, which can lead to the structural are observed at the lower jacket legs. Comparisons of total
resonance problem as well as fatigue damage. displacement and stress response are presented in Fig. 13 and
Considering the deeper water (40 m) level, the natural fre- Fig. 14, respectively.
quencies of the designed structures lie in the soft-stiff range, and
hence cross deployment does not lead to resonance (Fig. 12b). 5.2.1. Effects of the topological forms
However, the second topological form (4LJ-40-2) shows smaller At the water depth of 25 m (Fig. 13a), there is a large difference
values compared to the first topological form (4LJ-40-1), with between the two configurations considered (4LJ-25-1 and 4LJ-25-
8
T.-T. Tran and D. Lee International Journal of Naval Architecture and Ocean Engineering 14 (2022) 100451

Fig. 13. Histograms of total displacement: (a) 4LJ-25 and (b) 4LJ-40.

Fig. 14. Histograms of maximum stress: (a) 4LJ-25 and (b) 4LJ-40.

2). The larger response is found in the case of the first configura- 25 m and 40 m, respectively.
tion; it is about 2 times compared to the second form. At the water
depth of 40 m (Fig. 13b), the maximum response of the jackets is 5.2.3. Design check
around 20 cm, and it belongs to the second configuration. The strength checks for all jacket substructures are depicted in
Compared to the second configuration, the first form shows smaller Fig. 15. All jacket members and joints are checked with NORSOK N-
differences, with an average difference of 50%. 004 (2004). It shows that the UC factor is smaller than one,
Analysis of stress responses is performed to check the capacity implying that the simulation results are within the safety zone.
of the selected jacket members. The results are shown in Fig. 14,
where the dashed line is the yield strength of the material. In 5.3. Mass of jacket substructure
general, the leg members satisfy the strength criteria. At the sea
level of 25 m, the first topological form reaches its ultimate strength Fig. 16 presents a comparison of the masses of the jacket sub-
earlier than the second form. However, in the case of 40 m, the structures for different sea levels. The results show that the average
stress distribution has a slight change. The stresses of each jacket masses of the X-bracing system are about 10% higher than those of
leg member under twelve DLCs are graphically displayed in the Pratt system for both 4LJ-25 and 4LJ-40. At the water depth of
Appendix. 25 m (Fig. 16a), the mass of 4LJ-25-1P is 211 ton, showing a slightly
smaller mass compared to 4LJ-25-1X (225 ton). Further, a differ-
5.2.2. Effects of the bracing systems ence of about 32 ton between Pratt and X-brace systems is found
According to Fig. 13, there are minor variations in lateral dis- for the second configuration. At the deeper sea level 40 m (Fig. 16b),
placements between the bracing models. The largest differences the jacket masses of all configurations are close to each other, with
are 16.1% and 4.7% at the sea water levels of 25 m and 40 m, the X-bracing system heavier (around 30 ton) than the Pratt sys-
respectively. In Fig. 13, the most significant finding from all models tem. Since the transition piece and tower are the same, the trend of
is that the response obtained from the X-brace system is smaller difference is the same for total weight of the support structure.
than that of the Pratt system. This is owing to the higher stiffness of
X-bracing, which is explained in Section 5.1. Moreover, the struc- 5.4. Selection of the optimal jacket substructure for wind turbine
tural response is primarily caused by the moments and forces in X-
and Y- directions. Based on the outcomes, the following observations are drawn:
Under the ultimate design loads (Fig. 14), the flexural capacity of
the X-type jacket substructure is higher than that of the Pratt brace
At the water depth of 25 m (4LJ-25): The results from Eigen
type. Reduction values of 12% and 6.7% are found at the sea levels of analysis indicate that the first group is in non-resonance
9
T.-T. Tran and D. Lee International Journal of Naval Architecture and Ocean Engineering 14 (2022) 100451

Fig. 15. Results of strength check: (a) 4LJ-25 and (b) 4LJ-40.

Fig. 16. Comparison of mass: (a) 4LJ-25 and (b) 4LJ-40.

10
T.-T. Tran and D. Lee International Journal of Naval Architecture and Ocean Engineering 14 (2022) 100451

structures, showing greater effectiveness compared with reso-

nance substructures (second group). Under DLCs, all sub-
structures satisfy the strength criteria; however, the Pratt brace
form reaches its ultimate strength capacity earlier than the X-
brace system. Thus, the first topological form with X-brace
system (4LJ-25-1X) is selected as an optimal jacket substructure
for a wind turbine at the water depth of 25 m (Fig. 17).

At the water depth of 40 m (4LJ-40): Natural frequencies of all
designed substructures are in the non-resonance area; notably,
the first group is closer to the lower bound of the 3P range.
Furthermore, these systems are within the safety zone. Relating
to material costs, the second topological form with the Pratt
brace system (4LJ-40-2P) is the most cost-effective substructure.
Therefore, this concept is chosen as the optimal model for the
sea level of 40 m (Fig. 17).

6. Critical directionality for installing jacket substructures

In the previous section, feasibility analyses were performed for

all jacket substructures, aiming to select the reasonable jacket
substructures for each sea level. In this section, the sensitivity of the
selected jacket substructures to the environmental loading direc-
tionality is investigated. A schematic representation of loads acting
on the jackets is provided in Fig. 18.

6.1. Structural performances under environmental loads (env)

Fig. 18. Loads acting on the jacket substructures.
Numerical simulations were conducted to find the critical
bending direction of each jacket substructure. As described in
Section 3.3.1, the environmental loading angles vary from 0  to 360 discernible trend to the loading orientation. At the water depth of
 in steps of 15  . The comparisons of lateral displacement at the TP 25 m (Fig. 20a), stress responses have the same trend for all jacket
are depicted by polar diagrams in Fig. 19. As can be seen, there is a legs, with the maximum stress of 79 MPa. With regards to the
distinct difference between the two basic configurations. At the deeper sea level (Fig. 20b), the responses show a great difference
water depth of 25 m (Fig. 19a), the total displacement of the jacket between leg members. The maximum responses are 77 MPa, and
substructure (3.5 cm) is almost independent of the loading di- belong to leg 2 and leg 3 of the jacket.
rections. Thus, their responses are visualized as a full circle in the
polar diagram. For the deeper water (Fig. 19b), the critical dis- 6.2. Structural performances under design load cases
placements are found at the angles of 135  and 315  , with a value (DLCs) þ environmental loads (env)
of about 3.4 cm.
The comparative stresses of the jacket substructures are re- The main objective is to identify the critical directionality under
ported in Fig. 20. In general, the individual values vary with a various loading conditions. With this aim, each selected jacket

Fig. 17. Suggested jacket substructures at different water depths.

11
T.-T. Tran and D. Lee International Journal of Naval Architecture and Ocean Engineering 14 (2022) 100451

Fig. 19. Polar diagrams of the lateral displacements: (a) 4LJ-25-1X and (b) 4LJ-40-2P [cm].

Fig. 20. Polar diagrams of the stresses: (a) 4LJ-25-1X and (b) 4LJ-40-2P [MPa].

Fig. 21. Lateral displacement under combined loads: (a) 4LJ-25-1X and (b) 4LJ-40-2P [cm].

substructure was analyzed with a total of 288 analysis cases, Fig. 21 illustrates the total displacements of the jacket structure
comprising the various combinations of the environmental (24 against the combined loads (CBs). As explained in Section 5.2,
cases) and design loads (12 cases) (see Fig. 18). structural responses are primarily caused by forces and moments in

12
T.-T. Tran and D. Lee International Journal of Naval Architecture and Ocean Engineering 14 (2022) 100451

Fig. 22. Stresses under combined loads: (a) 4LJ-25-1X and (b) 4LJ-40-2P [MPa].

13
T.-T. Tran and D. Lee International Journal of Naval Architecture and Ocean Engineering 14 (2022) 100451

the X- and Y- directions. Thus, only seven CBs (i.e., CB1, CB2, CB3, configurations and brace systems were discussed. To explore
CB4, CB7, CB9, CB10), which show larger responses compared to the reasonable jacket substructures, this study considered jackets that
other cases, are provided herein. The outcomes from the CBs are fulfilled all main design requirements in terms of the target fre-
much larger compared to the environmental load (dash-dotted quency and dynamic performances under ultimate loads. A com-
black line). At the water depth of 25 m (Fig. 21a), the maximum parison of structural weights is also reported. The results are
responses (24 cm) for CB1 and CB2 occur at the angles of 270  and summarized as follows:
90  , respectively. In the case of deeper water (Fig. 21b), the
maximum displacements of the jacket substructure occur at the
The topological configurations of jackets have a significant in-
angles of 285  and 120  . Based on the obtained outcomes, the fluence on the natural frequencies of the whole system, whereas
critical regions (dash red rectangular) are defined. The angles in the brace systems have small impacts on the dynamic charac-
these ranges will cause the highest responses for the jacket teristics of the jacket substructures. Thus, when designing a
substructure. jacket substructure, the highest priority is to select an appro-
Like displacement, critical regions for stresses are also defined. priate topological configuration.
This can be seen in Fig. 22, where the stress responses for each leg
With regards to strength check, all jacket members and joints
member are depicted. In general, under the CBs, the maximum are within the safety zone. The high-stress distribution of the
stresses of all leg members are less than the yield strength of the jacket substructure mainly occurs at the top layers due to the
material (355 MPa). In the cases of 4LJ-25-1X (Fig. 22a), the average bending effect. Under ultimate loads, the Pratt type reaches its
stress is 313.0 MPa, and the highest stress occurs to leg 4 ultimate strength earlier than X-type.
(339.1 MPa). As seen, the critical directions are different for each leg
By comparing the effectiveness of different jacket substructures,
member due to influences of the design load cases. The critical two reasonable jacket substructures corresponding to different
responses of leg 1 and leg 2 are mainly caused by the CB1. On the sea levels (i.e., 25 m and 40 m) are suggested. These jacket
other hand, the critical regions of leg 3 and leg 4 are mainly pro- substructures will yield the best performance under Korean
duced by CB2 and CB7, respectively. From the observation, the environmental conditions. Furthermore, they are feasible in
critical regions of the jacket substructure are defined and caused by terms of the availability of installing equipment for the Korean
the moments (CB1 and CB2) and forces (CB7) in the X-direction. offshore wind farm.
However, there is a distinct difference in the stress distribution
Furthermore, the critical bending directions of the selected
of 4LJ-40-2P, as displayed in Fig. 22b. As shown, leg 3 is subjected to jacket substructure are determined. At the water depth of 40 m,
the largest stress (266.9 MPa), with a maximum difference of up to the critical bending directions vary between 105 and 150 ,
21.1%, compared to other leg members. The critical bending di- while the jacket substructure installed at 25 m is almost inde-
rections are found to occur at angles varying between 105  and 150 pendent of the environmental loads.
 under CB9, which is caused by the moment in the Y-direction.

The results of this study are significant, as they provide a

7. Limitations framework for the development of jacket substructures with a
representative turbine. Two reasonable jacket concepts are sug-
It is noted that the jacket substructures are developed in relation gested and can be used as references for large-scale deployments in
to the given wind turbine configuration and water depth. For a South Korea.
larger wind turbine, a classical similarity rule with the upscaling
coefficients is recommended. The design approach for the larger
wind turbines corresponds to the approach of the 3 MW reference Declaration of competing interest
wind turbine.
Even though this limitation, this study provides applicable The authors declare that they have no known competing
knowledge of the design assessment of jacket substructures, which financial interests or personal relationships that could have
might be extended to the study of larger wind turbine cases. The appeared to influence the work reported in this paper.
feasible configuration should be examined the sensitivities of to-
pological forms or bracing systems, and environmental loading
directionality. Acknowledgments

8. Conclusions This work was supported by the National Research Foundation

of Korea (NRF) grant funded by the Korean government (Ministry of
With the aim of developing offshore wind farms in deep water Science and ICT) (No. 2021 R1F1A1046912) and by Basic Science
sites, this study conducted a feasibility analysis for jacket sub- Research Program through the National Research Foundation of
structures at different water depths (i.e., 25 m and 40 m). For each Korea (NRF) funded by the Ministry of Education (NRF
sea level, four configurations with different topological 2021R1A6A1A0304518511).

14
T.-T. Tran and D. Lee International Journal of Naval Architecture and Ocean Engineering 14 (2022) 100451

Appendix

Histograms of Stress: (a) 4LJ-25 and (b) 4LJ-40 [MPa].

15
T.-T. Tran and D. Lee International Journal of Naval Architecture and Ocean Engineering 14 (2022) 100451

Kim, S.Y., Kim, K.M., Park, J.C., Jeon, G.M., Chun, H.H., 2016. Numerical simulation of
wave and current interaction with A fixed offshore substructure. Int. J. Nav.
Archit. Ocean Eng. 8, 188e197. https://doi.org/10.1016/J.IJNAOE.2016.02.002.
References Kim, J.H., Kim, S.Y., Yoo, S.H., 2020. Public acceptance of the “renewable energy 3020
plan”: evidence from A contingent valuation study in South Korea. Sustain.
4 C Offshore, 2021. Offshore Wind Farms in South Korea [WWW Document]. Times 12. https://doi.org/10.3390/SU12083151.
https://www.4coffshore.com/windfarms/south-korea/. NORSOK, 2004. Standard Design of Steel Structure. DNV.
Bentley, 2019. Structural analysis computer system (SACS). User’s Manual, Release Oh, K.Y., Nam, W., Ryu, M.S., Kim, J.Y., Epureanu, B.I., 2018. A review of foundations
14 version 0, Engineering Dynamics. of offshore wind energy convertors: current status and future perspectives.
Bo€ker, C., 2010. Load Simulation and Local Dynamics of Support Structures for Renew. Sustain. Energy Rev. https://doi.org/10.1016/j.rser.2018.02.005.
Offshore Wind Turbines. Shaker. POSCO, 2017. 4-Leg Jacket Substructure for 3MW Offshore Wind Turbine, Structural
Bossanyi, E., 2010. GH-bladed Version 4.0 User Manual. Design Report.
DNV-OS-J101, 2014. Design of Offshore Wind Turbine Structures. Sandal, K., Latini, C., Zania, V., Stolpe, M., 2018. Integrated optimal design of jackets
DNVGL-ST-0126, 2018. Support Structures for Wind Turbines. and foundations. Mar. Struct. 61, 398e418. https://doi.org/10.1016/
EEA, 2009. European Environment Agency. Europe's Onshore and Offshore Wind J.MARSTRUC.2018.06.012.
Energy Potential an Assessment of Environmental and Economic Constraints. Shi, W., Park, H., Chung, C., Baek, J., Kim, Y., Kim, C., 2013a. Load analysis and
https://doi.org/10.2800/11373. comparison of different jacket foundations. Renew. Energy 54, 201e210.
El-Reedy, M.A., 2014. Marine Structural Design Calculations, Marine Structural https://doi.org/10.1016/j.renene.2012.08.008.
Design Calculations. Elsevier Inc. https://doi.org/10.1016/C2012-0-07922-7. Shi, W., Park, H., Han, J., Na, S., Kim, C., 2013b. A study on the effect of different
IEC 61400-1, 2005. Wind Turbines-Part 1: Design Requirements. modeling parameters on the dynamic response of A jacket-type offshore wind
IEC 61400-3, 2009. Wind Turbines - Part 3: Design Requirements for Offshore. Wind turbine in the Korean southwest sea. Renew. Energy 58, 50e59. https://doi.org/
Turbines, Switzerland. 10.1016/j.renene.2013.03.010.
Jalbi, S., Bhattacharya, S., 2018. Closed form solution for the first natural frequency Shi, W., Han, J., Kim, C., Lee, D., Shin, H., Park, H., 2015. Feasibility study of offshore
of offshore wind turbine jackets supported on multiple foundations incorpo- wind turbine substructures for southwest offshore wind farm project in Korea.
rating soil-structure interaction. Soil Dynam. Earthq. Eng. 113, 593e613. https:// Renew. Energy 74, 406e413. https://doi.org/10.1016/j.renene.2014.08.039.
doi.org/10.1016/j.soildyn.2018.06.011. Tran, T.-T., Hussan, M., Kim, D., Nguyen, P.-C., 2020. Distributed plasticity approach
Jalbi, S., Bhattacharya, S., 2020. Concept design of jacket foundations for offshore for the nonlinear structural assessment of offshore wind turbine. Int. J. Nav.
wind turbines in 10 steps. Soil Dynam. Earthq. Eng. 139, 106357. https://doi.org/ Archit. Ocean Eng. 12, 743e754.
10.1016/j.soildyn.2020.106357. Tran, T.-T., Kang, S., Lee, J.-H., Lee, D., 2021. Directional bending performance of 4-leg
Kang, K.-S., Lee, J.-S., Kim, J.-Y., Ryu, M.-S., 2011. Economic analysis of offshore wind jacket substructure supporting a 3MW offshore wind turbine, 2021 Energies 14,
farm considering domestic development conditions of Korea. Korea Wind En- 2725. https://doi.org/10.3390/EN14092725. Page 2725 14.
ergy Jounal 2, 37e43. Tran, T.T., Kim, E., Lee, D., 2022. Development of a 3-legged jacket substructure for
Kim, D.H., Lee, S.G., 2015. Reliability analysis of offshore wind turbine support installation in the southwest offshore wind farm in South Korea. Ocean Eng.
structures under extreme ocean environmental loads. Renew. Energy 79, 246, 110643. https://doi.org/10.1016/J.OCEANENG.2022.110643.
161e166. https://doi.org/10.1016/j.renene.2014.11.052. Wu, X., Hu, Y., Li, Y., Yang, J., Duan, L., Wang, T., Adcock, T., Jiang, Z., Gao, Z., Lin, Z.,
Kim, D.H., Lee, S.G., Lee, I.K., 2014. Seismic fragility analysis of 5MW offshore wind Borthwick, A., Liao, S., 2019. Foundations of offshore wind turbines: a review.
turbine. Renew. Energy 65, 250e256. https://doi.org/10.1016/ Renew. Sustain. Energy Rev. 104, 379e393.
j.renene.2013.09.023.

16