Hindawi

Advances in Civil Engineering

Volume 2023, Article ID 8262901, 18 pages
https://doi.org/10.1155/2023/8262901

Research Article
Simplified Calculation Method for the Natural
Frequencies of the Offshore Wind Turbine Structures with
Monopile Foundations Based on Euler–Bernoulli Beam Theory

Yanpei Jiang , Xiaomin Zhou , and Yong Liu

School of Civil and Resource Engineering, University of Science and Technology Beijing, Beijing 100083, China

Correspondence should be addressed to Xiaomin Zhou; [email protected]

Received 14 February 2023; Revised 3 June 2023; Accepted 6 June 2023; Published 28 July 2023

Academic Editor: Agathoklis Giaralis

Copyright © 2023 Yanpei Jiang et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

The wind and wave loads on the offshore wind-turbine (OWT) structures with monopile foundations occur at specific frequencies.
When these excitation frequencies are close to the natural frequencies of OWT structures, resonance can disturb the proper
operation of the power-generation equipment and shorten the service life of the structural system. Therefore, to ensure safe and
efficient operation, the natural frequencies of the OWT structures with monopile foundations must be determined. To this end, a
calculation method for the natural frequencies of the OWT structures with monopile foundations is proposed. This method, which
considers the mass ratio (the ratio of the lumped mass on the top of the tower to the total mass of the tower, the transition piece,
and the monopile above the mudline) as well as the nonuniform moment of inertia of the structures above the mudline and the
soil–structure interaction (SSI), is derived using the Euler–Bernoulli beam differential equation and slope-displacement equation of
a monopile according to the flexibility matrix, stiffness equivalent principle, and virtual work principle. Finally, the natural
frequency calculation method is compared with finite element simulation and other calculation methods.

1. Introduction power plants, and it can become competitive to natural gas

and other energy sources in the near future [6].
At the Paris Climate Conference in 2015, more than Among offshore wind power support structures, the mono-
190 countries negotiated an agreement to curtail climate pile option is the most prevalent, accounting for approximately
change, hoping to replace fossil fuels with green energy 80% of the installed support structures [7]. The monopile foun-
worldwide [1]. Moreover, sustainable development policies dation is an economical option for offshore wind power [8].
promote the transition from traditional energy to new A monopile is a simple pipe segment driven into the seabed,
energy [2]. In addition, the gradual depletion of hydrocarbon and its advantages include easy production and installation as
reserves is pushing the energy market toward a clean and well as low-construction cost and risk [9]. The main loads on
sustainable path [3]. As a potential clean energy alternative, offshore wind turbines (OWTs) are of dynamic or cyclic
wind-energy technology is gradually shifting from onshore to nature, which makes the support structure highly sensitive to
offshore. Compared with similar onshore technologies, off- dynamic loads [10]. Therefore, it is important to fully under-
shore wind energy technology has certain advantages [4], stand the structural frequency and adjust the natural frequency
including higher wind velocity, larger wind turbines, and of the structure and its components during the design stage
broader installation areas. In recent years, as the costs have [11]. By adjusting the natural frequency, the resonance caused
decreased and generator dimensions and power have by the wind, waves, and wind turbine operation can be avoided,
increased, offshore wind power consumption has increased and the service life of the structure can be prolonged. The wind,
significantly [5]. Based on the price in 2017, the cost per MW- wave, and operating frequencies of the wind turbines are shown
hour of offshore wind farms is lower than that of nuclear in Figure 1.
2 Advances in Civil Engineering

a simplified calculation method for the frequency of OWTs

1P: rotor frequency considering the monopile. Arany et al. [37] studied the
3P: blade passing frequency frequency of the OWTs with monopile foundations using
soft–soft
mechanical and mathematical models and provided an approx-
Power spectral density

soft–stiff stiff–stiff
imation calculation formula for the natural frequency.
Similarly, Arany et al. [10] proposed a simple calculation
method for the natural frequency of the OWTs with monopile
foundations based on the dimensions of the tower and mono-
pile as well as the soil properties. Darvishi-Alamouti et al. [1]
1P 3P
established a mathematical model for the soil–pile interaction
based on Winkler’s method and the concept of elastic founda-
tion beams derived a simplified calculation method for the
natural frequency of the OWTs with monopile foundations
0.0 0.2 0.4 0.6 0.8 1.0 using the Rayleigh method of total energy conservation of
Frequency (Hz) the system. Ko [38] derived a closed-form solution for the
Wind natural frequency of the wind turbine structures with tapered
Wave towers based on Rayleigh’s method.
There are also deficiencies in the methodologies reported
FIGURE 1: Wind, wave, and operation frequencies of an OWT.
in the literature [1, 10, 36–38]. Many researchers have used a
simplified treatment of the upper tower [10, 36, 37], together
Currently, the structural designs are either soft–soft, with a large number of parameters and a narrow application
soft–stiff, or stiff–stiff. The soft–soft design is used for fre- scope. Darvishi-Alamouti et al.’s methodology [1] is applica-
quencies below 1P, the soft–stiff design is used for frequen- ble to cohesionless soils. However, Ko’s methodology [38] is
only applicable to tapered towers. To avoid such oversights, in
cies between 1 and 3P, and the stiff–stiff design is used for
this study, a natural frequency calculation method is estab-
frequencies above 3P [11]. With the soft–soft design, the
lished that reasonably considers the variation of upper tower
structure tends to be excessively deformed, which can affect
diameter and wall thickness, has a wide application range and
the operation of the OWT, and the natural frequency of the
clearer dynamic principles, and is applicable to different soils.
structure can become close to the frequency of the external
wind and wave loads, resulting in resonance. With the stiff–
stiff design, more structural steel is used, which increases the
2. Calculation Method for the Natural
costs. Most of monopile OWTs adopt soft–stiff design to Frequency of the OWT Structures with
reduce the construction costs [12]. When monopiles are Monopile Foundations
used for large-size turbines, it is difficult for the designer to
2.1. Natural Frequency for a Structure with a Fixed Support.
avoid the soft–soft design [13]. Another design strategy for
First, a natural-frequency calculation method that does not
OWTs, which are becoming increasingly larger in size (with a
consider monopiles and soil below the mudline is proposed.
consequent shift of the natural frequency toward the reso-
The ideal dynamic model of OWTs with monopile founda-
nance range), is the implementation of vibration control tions is shown in Figure 2.
devices [14–35]. The undamped free vibration equation for the motion of
The frequency corresponding to the first-order bending Euler beams [39] is as follows:
vibration type is closer to the 1P frequency than to the 3P
frequency. Moreover, under the action of wind, wave, and
∂2 vðz; t Þ ∂4 vðz; t Þ
tide loads, offshore wind turbine structures swing back and m þ Et It ¼ 0; ð1Þ
forth and sideways, exhibiting a swinging pattern similar to ∂t 2
∂z 4
that of the first-order bending vibration type. That is, the
wind, wave, and tide loads stimulate the first-order bending where m is the mass per unit length, Et is the elastic modulus
vibration of offshore wind turbine structures. Therefore, of the structure, It is the moment of inertia, t is time, and z is
in this study, the frequency corresponding to the first-order the space coordinate.
bending vibration of the structure was considered. Suppose that the solution of the above equation has the
Natural frequency calculation methods include numerical following form:
and simplified calculation methods. Owing to their overall
complexity, numerical calculations require computer program- vðz; t Þ ¼ X ðz ÞY ðt Þ: ð2Þ
ing. However, in terms of a simplified calculation method for
the natural frequency, a calculator or spreadsheet program can
be used. Substituting Equation (2) into Equation (1) gives
Using an elastically supported Euler–Bernoulli beam,
Adhikari et al. [36] derived the characteristic equation that m
X ðz ÞY 00 ðt Þ þ X ð4Þ ðz ÞY ðt Þ ¼ 0: ð3Þ
controls the natural frequency of the structure and proposed Et It
Advances in Civil Engineering 3

m1 cos2 ⁡ðaLt Þ þ cosh2 ⁡ðaLt Þ þ sin2 ⁡ðaLt Þ − sinh2 ⁡ðaLt Þ

þ 2 cos ðaLt Þ cosh ðaLt Þ
m
þ 2a 1 cos ðaLt Þ sinh ðaLt Þ
m
m1
− 2a cosh ðaLt Þ sin ðaLt Þ ¼ 0:
m
m
ð10Þ

By substituting m1 =m ¼ μLt and aLt ¼ J into Equation (10),

Lt the following simplified equation is derived as follows:

cos2 ⁡ðJ Þ þ cosh2 ⁡ðJ Þ þ sin2 ⁡ðJ Þ − sinh2 ⁡ðJ Þ þ 2 cos ðJ Þ coshðJ Þ
EtIt þ 2Jμ cos ðJ Þ sinh ðJ Þ − 2Jμ cosh ðJ Þ sin ðJ Þ ¼ 0;
ð11Þ
z

where μ ¼ m1 =mLt ; μ is the ratio of the lumped mass on the

top of the tower (total mass of the nacelle, hub, and blade) to
X (z) the total mass of the tower, transition piece, and monopile
above the mudline. Substituting the aLt ¼ J into Equation
FIGURE 2: Dynamic model of an OWT. (8) yields
rffiffiffiffiffiffiffiffiffi
Separating the variables in the above equation yields Et It
ω ¼ J2 : ð12Þ
mL4t
m Y 00 ðt Þ X ð4Þ ðz Þ
þ ¼ 0: ð4Þ
E t I t Y ðt Þ X ðz Þ The natural frequency of OWT structures with monopile
foundations can be expressed as follows:
To make the above equation valid, the following equation sffiffiffiffiffiffiffiffiffi
must be satisfied as follows: ω J2 E t It
f ¼ ¼ : ð13Þ
2π 2π mL4t
X ð4Þ ðz Þ m Y 00 ðt Þ
¼− ¼ a4 : ð5Þ
X ðz Þ E t I t Y ðt Þ
Here, the J value can be solved with the coefficient μ cor-
responding to different OWT structures using Equation (11).
Using the above equation, two ordinary differential equa- To obtain the natural frequency, the J value must be the first
tions and circular frequency expressions are obtained as fol- solution to the transcendental Equation (11). By substituting
lows: the J value into Equation (13), the structural natural frequency
without considering the soil–structure interaction (SSI) can
Y 00 ðt Þ þ ω2 Y ðt Þ ¼ 0; ð6Þ be derived. The J values corresponding to μ are shown in
Figure 3.
X ð4Þ ðz Þ − a4 X ðz Þ ¼ 0; ð7Þ A monopile-supported OWT structure usually has a
nonuniform moment of inertia due to variations in tower
a4 Et It diameter and tower wall thickness, the presence of the tran-
ω2 ¼ : ð8Þ sition piece, and the monopile wall being thicker than the
m
tower wall. As Equation (13) is applicable only for a uniform
moment of inertia, it must be corrected.
To determine the circular frequency from Equation (8),
the differential Equation (7) must be solved, and the solution (1) Moment of inertia
can be expressed as follows:
The structure shown in Figure 4 is clearly divided into
X ðz Þ ¼ A1 cos az þ A2 sin az þ A3 cosh az þ A4 sinh az: three segments: 0∼l1 ; l1 ∼l2 , and l2 ∼l3 . There is unnoticeable
change in the diameter and wall thickness of segment l1 ∼l2 ,
ð9Þ
whereas there is a noticeable change in the diameter and wall
thickness of segment l2 ∼l3 . Therefore, it is necessary to con-
According to the boundary conditions of the displacement, sider the changes in the diameter and wall thickness of the
shear force, bending moment, and slope in Figure 2, the fol- tower based on the two sections when calculating the
lowing transcendental equation can be derived as follows: moment of inertia of the tower.
4 Advances in Civil Engineering

1.7 It ¼ f ðz Þ: ð16Þ
1.6
1.5
1.4 The weight number of the microelements in segment
1.3
l1 ∼l2 in Figure 4 is given by
1.2
dz
Z ;
J

1.1
1.0
l2 ð17Þ
dz
0.9 l1
0.8
0.7 and the proportion of the moment of inertia of the microel-
0.6 ement in segment l1 ∼l2 is expressed as follows:
0.5
0 2 4 6 8 10 12 14 16 18 20 22
μ I dz
Z tl :
2 ð18Þ
FIGURE 3: J value corresponding to μ. dz
l1

z The moment of inertia of segment l1 ∼l2 corresponds to

the above integral and is expressed as follows:
l3 Z l2
Z l2
L1 I1
It dz f ðz Þdz
I2 ¼ Z Z
l1 l1
l2
¼ l2
: ð19Þ
l2 dz dz
l1 l1
Tower

L2 I2
When segment l1 ∼l2 is discrete, as is segment q, its
moment of inertia is given by
dz
Z  Z 
li li
q q
∑i¼1 fi ðz Þdz ∑i¼1 fi ðz Þdz
l1 li−1 li−1 ð20Þ
I2 ¼ ¼ ;
L3 I3
Transition piece l2 − l1 L2
and monopile

o x
where fi ðz Þ is the inertia function of segment, i; li−1 is the
dn D lower limit value of segment, i in the coordinate system, li is
2 2
the upper limit value of segment i in the coordinate system,
and L2 is the length of segment l1 ∼l2 .
In the above formula, the analytical expression of the
FIGURE 4: Structural profile. moment of inertia of segment l1 ∼l2 contains an integral for-
mula, which is not conducive to application; therefore, it can
be simplified to the following expression:
From Figure 4, the outer diameter function can be q
∑i¼1 zi Ii
expressed as follows: I2 ¼ q ; ð21Þ
∑i¼1 zi

D ¼ F1 ðzÞ; ð14Þ
where q is the discrete number of segment l1 ∼l2 of the tower,
Ii is the moment of inertia of segment i, and zi is the length of
and the inner diameter function can be expressed as segment i.
The equations for I1 and I3 have the same form as Equa-
tion (21). In practical engineering, the diameter and thick-
dn ¼ F2 ðz Þ: ð15Þ ness of offshore wind towers change from bottom to top. The
effect of this change on the structural stiffness through the
moment of inertia is now considered.
D and dn are functions of z; therefore, the moment of
inertia It can be expressed as a function of z: (2) Solution to equivalent natural frequency problem
Advances in Civil Engineering 5

Blade

Hub m1 m1
Fb Nacelle
Fb Fb

L1 E1I1 z L1

L2 E2I2 L2 Lt
Tower

Transition piece
EtIe

Mudline L3 E3I3 L3

Monopile

FIGURE 5: Calculation model for natural frequency without considering SSI.

Z Z
In Figure 2, the moment of inertia It is uniform, which ∂ ½M ðz ފ2
L1 þL2 þL3 ∂ L1 ðFb z Þ2
is an ideal model. The moment of inertia varies among the DF ¼ dz ¼ dz
∂F 0 2EI ðz Þ ∂F 0 2E1 I1
diameter-varying segment of the tower (the tapering seg- Z L þL Z
∂ 1 2 ðFb z Þ
2
∂ L1 þL2 þL3 ðFb z Þ2
ment), the segment with constant tower diameter or the þ dz þ dz
segment with mildly varying tower diameter, the transition ∂F L1 2E2 I2 ∂F L1 þL2 2E3 I3
piece, and the monopile. To apply Equation (13) for the Fb L31 Fb ðL1 þ L2 Þ3 − Fb L31
calculation of the natural frequency of the OWT structures ¼ þ
3E1 I1 3E2 I2
with monopile foundations, the moment of inertia of the
F ðL þ L2 þ L3 Þ3 − Fb ðL1 þ L2 Þ3
diameter-varying segment of the tower, the segment with þ b 1 ;
constant tower diameter or the segment with mildly vary- 3E3 I3
ing tower diameter, the transition piece, and the monopile, ð23Þ
must be transformed into equivalent measures. The calcu-
lation model for the natural frequency without considering
the SSI is shown in Figure 5. where M ðz Þ is Fb z; L1 is the length of the diameter-varying
According to Crotti–Engesser’s theorem, the elastic dis- segment of the tower, L2 is the length of the segment with
placement can be expressed as follows: constant tower diameter or the segment with mildly vary-
ing tower diameter, L3 is the length from the mudline to
∂VC the tower bottom, E1 I1 is the flexural stiffness of the
DF ¼ ; ð22Þ
∂F diameter-varying segment of the tower, E2 I2 is the flexural
stiffness of the segment with constant tower diameter or
where VC is the complementary energy of strain and F repre- the segment with mildly varying tower diameter, and E3 I3
sents forces. is the flexural stiffness from the mudline to the tower
Regarding the OWTs, when the wind turbine load Fb is bottom.
applied on the hub on the tower top, the displacement at Substituting LL1t ¼ β1 ; LL2t ¼ β2 , and LL3t ¼ β3 into the above
the tower top can be expressed as follows equation yields
6 Advances in Civil Engineering

Fb β31 L3t Fb ½ðβ1 þ β2 Þ3 − β31 ŠL3t Fb ½1 − ðβ1 þ β2 Þ3 ŠL3t m1

DF ¼ þ þ ; μ¼ ; ð31Þ
3E1 I1 3E2 I2 3E3 I3 0:228mLt
ð24Þ
and Equation (13) can be corrected to
and the structural displacement under unit load can be sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
expressed as follows J2 E t Ie
ft ¼ : ð32Þ
2π 0:228mL4t
β31 L3t ½ ðβ 1 þ β 2 −
Þ3 β31 ŠL3t ½ 1 − ðβ 1 þ β2 Þ3 ŠL3t
D1 ¼ þ þ :
3E1 I1 3E2 I2 3E3 I3
ð25Þ
2.2. Natural Frequency with SSI. The following analysis
focuses on the eigen solutions of a linear system; therefore,
According to the definition of stiffness, the structural the nonlinear soil behavior is not modeled. The foundation
stiffness is the reciprocal of Equation (25). systems generally do not go into a nonlinear regime, and there-
fore, a linear approximation is considered acceptable [10]. An
1 OWT structure with a monopile foundation includes a nacelle,
K¼ : ð26Þ
D1 hub, blade, tower, transition piece, and monopile. The interac-
tion between the monopile and soil below the mudline can
be represented by three flexibility dimensions, as shown in
Another structural stiffness can be expressed in the fol- Figure 6: horizontal flexibility δL , rotational flexibility δR , and
lowing equation: coupling flexibility δLR . Generally, the settlement at the mono-
pile foundation root is not considered; therefore, the inclined
3Et Ie spring model is used, as shown in Figure 6.
Ke ¼ : ð27Þ
L3t The horizontal force and bending moment on the mud-
line as well as the horizontal displacement and slope on the
Now, we equate Equations (26) and (27). For an OWT mudline of a monopile can be expressed using the following
with a monopile foundation made of steel, the elastic modu- equations:
lus of steel can be considered as the elastic modulus at all
locations, and all the moduli are equal (E1 ¼ E2 ¼ E3 ¼ Et ). ( ) " #( )
The elastic moduli at both sides of the equal sign can be y0 δL δLR F0
¼ ; ð33Þ
reduced, yielding the following simplified equation: ψ0 δLR δR M0

I1 I2 I3
Ie ¼ :
β31 I2 I3 þ ½ðβ1 þ β2 Þ − β31 ŠI1 I3 þ ½1 − ðβ1 þ β2 Þ3 ŠI1 I2
3 where F0 is the horizontal force on the mudline, M0 is the
bending moment on the mudline, y0 is the monopile dis-
ð28Þ
placement on the mudline, and ψ 0 is the monopile slope
on the mudline.
If the structure above the mudline is a linearly tapered With a load Fb on the tower top, the support displace-
tower, Ie can be expressed as follows: ment is CR . To calculate the displacement yt on the tower
top, a virtual force system must be provided based on the
Ie ¼ I1 : ð29Þ virtual work principle. With a nonzero-force increment δFb
imposed on the tower top, the nonzero-force increment at
the support is δFR , as shown in Figure 7.
The equivalent moment of inertia is introduced through The increments of the bending moment M and end reac-
the stiffness, which is approximately equal to the generalized tion FR are expressed as follows:
stiffness. Therefore, the equivalent moment of inertia can be
introduced through the generalized stiffness, and the equiv- ∂M
alent mass can be introduced using the generalized mass. For δM ¼ δF ; ð34Þ
∂Fb b
a structure with one fixed support and one free end, the
generalized mass [39] is expressed as follows:
∂FR
δFR ¼ δF : ð35Þ
∗
m ¼ 0:228mLt : ð30Þ ∂Fb b

With the equivalent mass introduced through the gener-

alized mass, the coefficient μ in Equation (11) can be cor- The virtual force system performs virtual work in the defor-
rected to mation state, and its virtual force equation is given as follows:
Advances in Civil Engineering 7

Blade

Hub
m1
Fb = 1
Nacelle Fb = 1

L1

δR

m
L2 Lt
Tower

Transition piece EtIe

δLR
L3
Mudline
δL

Monopile

FIGURE 6: Model of an OWT with a monopile foundation.

Z Lt M therefore, Equation (39) can be expressed as follows:

yt δFb þ CR δFR ¼ δMdz: ð36Þ
0 Et Ie
∂VC ∂ðCR FR Þ ∂C
yt ¼ − þ FR R : ð40Þ
Substituting Equations (34) and (35) into Equation (36) ∂Fb ∂Fb ∂Fb
gives
The support displacement CR is given by y0 and ψ 0 ,
Z L whereas the support reaction FR is given by −Fb and −FbLt.
t M ∂M ∂F
yt ¼ dz − CR R : ð37Þ Using Fb ¼ F0 and M0 ¼ Fb Lt , the support reaction FR can be
0 Et Ie ∂Fb ∂Fb
expressed as −F0 and −M0 . Substituting Equation (33) into
Equation (40) yields
Since the support displacement CR and support reaction
FR are functions of Fb , this yields ∂VC
yt ¼ þ y 0 þ ψ 0 Lt : ð41Þ
∂Fb
∂ðCR FR Þ ∂F ∂C
¼ C R R þ FR R : ð38Þ
∂Fb ∂Fb ∂Fb
Figure 8 shows the deformation of the loaded tower top.
According to the deformation diagram of the loaded
By substituting Equation (38) into Equation (37), the tower top, the total displacement of the tower top can be
following simplified equation is obtained as follows: expressed as
Z
∂ Lt M2 ∂ðCR FR Þ ∂C ∂VC
yt ¼ dz − þ FR R : ð39Þ yt ¼ y0 þ Lt sin ðψ 0 Þ þ : ð42Þ
∂Fb 0 2Et Ie ∂Fb ∂Fb ∂Fb

The first integrand to the right of the equal sign in Equa- Equations (41) and (42) are derived from the virtual
tion (39) is the complementary strain energy density; work equation and geometric deformation conditions,
8 Advances in Civil Engineering

yt

Fb m1 δFb m1

ψ0

Lt

EtIe

CR

y0

δFR

FIGURE 7: Virtual work.

respectively. With a smaller ψ 0 , Equations (41) and (42) With M0 ¼ Fb Lt and F0 ¼ Fb , Equation (43) simplifies to
become equal according to the concept of the equivalent
infinitesimal, and therefore, the two equations share the same
expression. Therefore, with a smaller ψ 0 , either Equations (41) F0 L3t
yt ¼ δL F0 þ δLR F0 Lt þ Lt sin ðδLR F0 þ δR F0 Lt Þ þ :
or (42) can be used to represent the total displacement of the 3Et Ie
tower top after it is stressed. Equation (42) is selected to repre- ð44Þ
sent the total displacement at the tower top after the structural
tower top is stressed.
Substituting Equation (33) into Equation (42) yields Considering the influence of the SSI on the natural fre-
quency of the structure, an equivalent structure can be intro-
yt ¼ δL F0 þ δLR M0 þ Lt sin ðδLR F0 þ δR M0 Þ duced to make the displacements at the top of the two
Z L
∂ t ðF b z Þ
2 ð43Þ structures equal when they are subjected to the same load.
þ dz:
∂F0 0 2Et Ie These two structures are illustrated in Figure 9.
Advances in Civil Engineering 9

∂VC sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Lt sin (ψ0) + J2 Et Is
∂Fb
fs ¼ : ð49Þ
Fb m1 2π 0:228mL4t

ψ0
The symbol explanations are given in Table 1. Elements δL ;
δLR and δR of the flexibility matrix are given in Appendix A.
The natural frequency of an OWT structure with a
m monopile foundation considering the SSI is then calculated
using the following four steps:

(1) Calculate the moment of inertia Ie above the mudline

Lt for the structure.
EtIe (2) Determine whether the monopile is rigid or slender.
Depending on the soil properties, determine whether
the monopile is rigid or slender using Equations (A3),
(A4), (A8), and (A9). For a multilayer soil, this can be
determined based on the percentage of the total
thickness of the cohesive soil and the total thickness
of the cohesionless soil in the embedded length of the
monopile.
(3) Calculate the flexibility matrix. According to the soil
properties and depending on whether the monopile
y0 is rigid or slender, calculate the flexibility matrix
using Equations (A5), (A6), (A10), and (A11).
FIGURE 8: Structural deformation of the loaded tower top. (4) Determine the natural frequency of the structure.
First, substitute the elements in the flexibility matrix
The displacement of the tower top under stress is thus into Equation (48) to calculate the equivalent
expressed as follows: moment of inertia Is . Second, substitute the equiva-
lent moment of inertia into Equation (49) to obtain
Z
∂V ∂ Lt ðF b z Þ2 F L3 the natural frequency of the structure.
yt ¼ C ¼ dz ¼ b t : ð45Þ
∂Fb ∂Fb 0 2Et Is 3Et Is

Under the same load, the top displacements of the two 3. Simulation Comparison
structures are equal: 3.1. Parameter Selection for a Finite Element Model. Three
finite element models are employed to enable differentiation
Fb L3t F0 L3t
¼ δL F0 þ δLR F0 Lt þ Lt sin ðδLR F0 þ δR F0 Lt Þ þ : and diversity of the study objects. In each case, the following
3Et Is 3Et Ie assumptions are made: the total mass m1 of the blade, hub,
ð46Þ and nacelle is 243,000 kg. The values of L1 ; L2 , and L3 are
35.877, 39.833, and L3 19 m, respectively. The monopile
Under a unit load, the displacements of both structures length is 74 m, and the monopile embedded length is 55 m.
are equal and can be expressed as follows: Poisson’s ratio is 0.3, Young’s modulus is 206 Gpa, and the
density is 7,850 kg/m3. For Model 1, the outer diameter of the
L3t L3 monopile is 5 m, and the wall thickness of the monopile is
¼ δL þ δLR Lt þ Lt sin ðδLR þ δR Lt Þ þ t : ð47Þ 0.06 m. For Model 2, the outer diameter of the monopile is
3Et Is 3Et Ie
5.5 m, and the wall thickness of the monopile is 0.06 m. For
Model 3, the outer diameter of the monopile is 5 m, and the
After converting the numerator and denominator at both wall thickness of the monopile is 0.08 m. The parameters of
sides of the equal sign in Equation (47), the equation remains the upper tower section of the OWT are shown in Figure 10.
valid. At this time, both sides of the equal sign represent the Finite element models are shown in Figure 11. For the soil
structural stiffness and can be further simplified as model, a hardened soil with a small-strain stiffness is
employed. The parameters for the hardened soil with
L3t small-strain stiffness are given in Table 2.
Et Is ¼ h i: ð48Þ
L3
3 δL þ δLR Lt þ Lt sin ðδLR þ δR Lt Þ þ 3EttIe In the finite element models, it is necessary to increase
the elastic modulus and gravity density of the soil to reduce
the horizontal displacement of the structural mudline. Using
The calculation equation of the structural natural fre- Model 1 as an example, the calculation results are shown in
quency after considering the SSI is Figure 12.
10 Advances in Civil Engineering

∂VC yt
Lt sin (ψ0) +
∂Fb

Fb m1 Fb m1

ψ0

m
m

Lt Lt

EtIe EtIs

y0

FIGURE 9: Equivalent structure model.

TABLE 1: List of symbols.

Symbol Explanation Symbol Explanation
m Mass per unit length Fb Wind turbine load
Et Elastic modulus of the structure K Structural stiffness
It Moment of inertia Ke Equivalent stiffness of the structure
Equivalent moment of inertia after considering the
t Time Ie
nonuniform moment of inertia
z Space coordinate m∗ Generalized mass
Natural Frequency after considering the nonuniform
a Constant ft
moment of inertia
ω Circular frequency δL Horizontal flexibility
vðz; tÞ Displacement function δR Rotational flexibility
X ðz Þ Shape function δLR Coupling flexibility
Y ðt Þ Amplitude function F0 Horizontal force on the mudline
A1 A2 A3 A4 Real constant M0 Bending moment on the mudline
Lt Length from the mudline to the tower top y0 Monopile displacement on the mudline
μ Mass ratio ψ0 Monopile slope on the mudline
J Constant a times Lt CR Support displacement
Natural frequency of OWT structures with monopile
f yt Displacement of the tower top
foundations
D Outer diameter FR End reaction
Advances in Civil Engineering 11

TABLE 1: Continued.
Symbol Explanation Symbol Explanation
dn Inner diameter M Bending moment
L1 Length of the diameter-varying segment of the tower IS Equivalent moment of inertia after considering the SSI
Length of the segment with constant tower diameter or
L2 fs Natural Frequency after considering the SSI
the segment with mildly varying tower diameter
L3 Length from the mudline to the tower bottom kh Modulus of the subgrade reaction
I1 Moment of inertia within a length range of L1 DP Pile diameter
I2 Moment of inertia within a length range of L2 E Elastic modulus of the soil
I3 Moment of inertia within a length range of L3 IP Moment of inertia of the pile
VC Complementary strain energy υS Poisson’s ratio of the soil
E1 E2 E3 Elastic modulus β Slenderness parameter
F Forces LP Embedded length of the pile
β1 L1 divided by Lt nh Coefficient of the subgrade reaction
β2 L2 divided by Lt zs Depth below the mudline
β3 L3 divided by Lt EP Elastic modulus of the pile

0.0 0.0

0.1 0.1

0.2 0.2

0.3 0.3

0.4 0.4
z (L1 + L2)

z (L1 + L2)

0.5 0.5

0.6 0.6

0.7 0.7

0.8 0.8

0.9 0.9

1.0 1.0
3.0 3.2 3.4 3.6 3.8 4.0 4.2 4.4 4.6 4.8 5.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Diameter (m) Moment of inertia (m4)
Outer diameter
Inner diameter

FIGURE 10: Parameters for the upper tower section of a 4 MW OWT.

5m

m1 m1 m1
10 m
L1 L1 L1

L3
L2 L2 L3 L2
L3 9m

5.5 m

110 m 110 m 110 m

0m
0m
0m

11
11
11

132 m 132 m 132 m

Model 1 Model 2 Model 3

FIGURE 11: Finite element models.

12 Advances in Civil Engineering

TABLE 2: Parameters for hardened soil with small-strain stiffness.

Soil Thickness/m c/kPa φ/∘ E/MPa γ 0:7 e m ref
E50 /MPa ref
Eoed /MPa ref
Eur /MPa Gref
0 /MPa
Clay 3.6 15 8.4 2.74 0.0001 1.23 0.80 4.56 3.04 24.32 44.80
Clay 8.6 15 8.4 2.68 0.0001 1.26 0.80 4.47 2.98 23.84 42.70
Silt 8.1 18 26.4 10.59 0.0001 0.72 0.60 11.80 11.80 50.03 96.84
Sand 9.2 7 33.4 16.34 0.0001 0.63 0.55 18.20 18.20 54.60 111.00
Silt 9.2 22 12.1 4.03 0.0001 0.97 0.80 5.60 4.48 29.21 67.00
Clay 15 37 17.8 5.48 0.0001 0.77 0.80 6.09 6.09 30.45 90.24
Sand 56.3 7 33.6 14.84 0.0001 0.66 0.55 16.50 16.50 54.45 106.08
γ0.7 is generally 0.0001 as the threshold shear strain. e is the void ratio. m is the power for stress-level dependency of stiffness, with a value between 0.5 and 1.
ref ref
Sand and silt are near 0.5, and soft clay is near 1. E50 is the secant stiffness in standard drained triaxial test. Eoed is the tangent stiffness for primary oedometer
ref is the unloading/reloading stiffness. Gref is the shear modulus at the small strain. The un-/reloading Poisson’s ratio is 0.2.
loading. Eur 0

1,000
The elastic modulus is indicated by E
The gravity density of soil is indicated by γ
Structural mudline displacement (m)

10

0.1

1E-3

1E-5

1E-7

1E-9

10 100 1,000 10,000 100,000

Horizontal load on the top of tower (kN)
E = 2.06E8 kPa γ = 7.7E1 kN/m3
E = 2.06E9 kPa γ = 7.7E2 kN/m3
E = 2.06E12 kPa γ = 7.7E5 kN/m3

FIGURE 12: Displacement under different elastic moduli and gravity densities of soil.

0.10 0.10 0.10

Tower top displacement (m)

Tower top displacement (m)

Tower top displacement (m)

0.08 Model 1 0.08 Model 2 0.08 Model 3

0.06 0.06 0.06
0.04 0.04 0.04
0.02 0.02 0.02
0.00 0.00 0.00
–0.02 –0.02 –0.02
–0.04 –0.04 –0.04
–0.06 –0.06 –0.06
–0.08 –0.08 –0.08
–0.10 –0.10 –0.10
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
Time (s) Time (s) Time (s)

Without SSI
With SSI

FIGURE 13: Cycles of the models.

As shown in Figure 12, with a gravity density of soil of 3.2. Comparative Analysis of Calculation Results. A horizon-
7.7E5 kN=m3 and elastic modulus of 2.06E12 kPa, the horizon- tal load is applied to the tower top in the model, which is later
tal displacement of the structural mudline is quite small, close to removed to allow the structure to vibrate freely. The calcula-
zero. At that time, the position of the mudline is similar to that of tion results are shown Figure 13. The natural frequency of
a fixed support; therefore, the SSI is ignored. In the following the structure can be obtained by evaluating the time required
finite element calculation, the elastic modulus and gravity den- for each cycle.
sity of the soil are 2.06E12 kPa and 7.7E5 kN/m3, respectively, According to Table 2, the total thickness of the sandy soil
without considering the SSI. layer is 10.5 m, accounting for only 19.091% of the embedded
Advances in Civil Engineering 13

TABLE 3: Natural frequencies of the models.

Model Method Natural frequency without SSI (Hz) Deviation (%) Natural frequency with SSI (Hz) Deviation (%)
Ko [38] 0.295 11.940 0.286 6.230
Yang [40] 0.258 22.985 0.257 15.738
1 Arany et al. [10] 0.248 25.970 0.226 25.902
Method herein 0.323 3.582 0.304 0.328
Finite element method 0.335 – 0.305 –
Ko [38] 0.305 12.104 0.295 2.318
Yang [40] 0.291 16.138 0.290 3.974
2 Arany et al. [10] 0.253 27.089 0.231 23.510
Method herein 0.331 4.611 0.310 −2.649
Finite element method 0.347 – 0.302 –
Ko [38] 0.304 12.392 0.294 6.369
Yang [40] 0.272 21.614 0.271 13.694
3 Arany et al. [10] 0.253 27.089 0.230 26.752
Method herein 0.332 4.323 0.311 0.955
Finite element method 0.347 – 0.314 –

length of the monopile (55 m), whereas the total thickness of both approaches. Compared with the calculation results from
the cohesive soil layer is 44.5 m, accounting for 80.909% of the finite element model, the calculation results for the three
the embedded length of the monopile (55 m). Therefore, models using the method proposed by Ko [38] are different,
whether the monopile is rigid or slender is determined by the with deviations ranging from 2.318% to 12.392%.
cohesive soil. According to Equations (A3) and (A4), the mono- To investigate the effect of μ on the frequency, only the
pile embedded lengths in the three models are closer to the mass of m1 in Models 1, 2, and 3 is changed. In Figures 14–17,
determination value of the rigid pile in Equation (A4); therefore, the coefficient μ in Equation (11) is employed.
the monopiles in the three models are considered as rigid piles. The natural frequency decreases as μ increases. The devi-
For the layered soil, the element value in the flexibility matrix in ation between the finite element calculation results and the
Equation (48) is the weighted average value. To determine this, calculation results of the simplified method is small, indicat-
the elements in the flexibility matrix of each layer are first ing that the simplified method used in this study meets the
calculated, then multiplied by the thickness of the soil layer, engineering accuracy requirement.
and finally superimposed and divided by the monopile embed-
ded length. The calculation results are shown in Table 3. 3.3. Other Simulation Comparison. The OWT parameters are
Different methods are used to calculate the natural fre- obtained from relevant literature [41]. The bottom of the
quency of the finite element model, which will be compared tower is fixed. The structure above the mudline is only a
with the finite element calculation results. The calculation tapered tower; thus, Equation (29) is used to calculate Ie .
results are listed in Table 3. The calculation results are listed in Table 4.
According to Table 3, the calculation method for the According to the comparison of the calculation results
natural frequency of the OWT structures with monopile between the four natural-frequency simplified calculation
foundations considering the SSI can consider the nonuni- methods and the numerical calculation, the difference
form moment of inertia above the mudline of the structure between the Arany method and the numerical calculation
and the SSI into account when calculating the natural fre- is large, indicating that the model with such geometric fea-
quency of the structure. The natural frequencies of the three tures is beyond the scope of application of the Arany method.
models are calculated and compared with the finite element The OWT parameters are obtained from relevant litera-
results to validate this method. As shown in Table 3, the ture [42]. It should be noted that the 1P range for the 10 MW
natural frequencies of the models are reduced after consid- DTU corresponds to 0.1–0.16 Hz, whereas the 3P range cor-
ering the SSI. responds to 0.3–0.48 Hz [43]. Considering a safety margin of
Compared with the calculation results of the finite ele- 10%, the “allowable” frequency range for the OWT is thus
ments model, the calculation results for the three models using 0.176–0.273 Hz [42]. The calculation results are listed in
the method proposed by Arany [10] are significantly different, Table 5.
with deviations ranging from 23.510% to 27.089%. The calcu- Without considering the SSI, the calculation results of the
lation results for the three models using the method proposed Yang and Arany methods are significantly different from the
by Yang [40] are also significantly different, with deviations finite element calculation results. This difference is signifi-
ranging from 3.974% to 22.985%. The reason for the signifi- cantly reduced when the SSI is considered. This is because
cant deviation in the calculation results of the two methods is the finite element calculation results considering the SSI are
that the three models are beyond the scope of application of reduced by 11.106% compared with the finite element
14 Advances in Civil Engineering

Tower top displacement (m)

Tower top displacement (m)

Tower top displacement (m)

0.10
0.08 Model 1 0.08 Model 2 0.08 Model 3
0.06 0.06 0.06
0.04 0.04 0.04
0.02 0.02 0.02
0.00 0.00 0.00
–0.02 –0.02 –0.02
–0.04 –0.04 –0.04
–0.06 –0.06 –0.06
–0.08 –0.08 –0.08
–0.10
0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10
Time (s) Time (s) Time (s)

μ = 0.456 μ = 4.001
μ = 0.912 μ = 4.004
μ = 2.002 μ = 5.005
μ = 3.003

FIGURE 14: Cycles of the models without SSI.

Tower top displacement (m)

Tower top displacement (m)

Tower top displacement (m)

0.10 Model 1 0.10 Model 2 0.10 Model 3

0.05 0.05 0.05

0.00 0.00 0.00

–0.05 –0.05 –0.05

–0.10 –0.10 –0.10

0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10
Time (s) Time (s) Time (s)

μ = 0.456 μ = 4.001
μ = 0.912 μ = 4.004
μ = 2.002 μ = 5.005
μ = 3.003

FIGURE 15: Cycles of the models with SSI.

0.6 8 0.6 8 0.6 8

Model 1 Model 2 Model 3
0.5 0.5 0.5 6
Frequency (Hz)

Frequency (Hz)

Frequency (Hz)

6 6
Deviation (%)

Deviation (%)

Deviation (%)
0.4 0.4 0.4 4
4 4
0.3 0.3 0.3 2
2 2
0.2 0.2 0.2 0

0.1 0 0.1 0 0.1 –2

0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5
μ μ μ

Finite element method

Simplified calculation method
Percentage difference

FIGURE 16: Natural frequency comparison without considering the SSI.

calculation results without the SSI, and the single-digit per- that the method can only be used after modifying the formu-
centage is reduced by other methods. The results of the four las. The Ko calculation method is suitable for models with
simplified calculation methods are within the allowable tapered towers and substructures above the mudline and for
range. models with only a tapered tower above the mudline. This
Whether or not the SSI is considered, the simplified cal- method considers the changes in the moment of inertia and
culation method proposed in this paper performs well for mass of the tapered tower.
tapered and multisegment towers. Therefore, this simplified In the Yang calculation method, it is assumed that the
calculation method has a wide range of applications. tower wall thickness is constant and the structure above
When the SSI is not considered, Ko’s method performs the mudline is a tapered tower. Therefore, this method is
well for tapered towers. It is worth noting that there are suitable for models with only a tapered tower above the
errors in Equations (11) and (15) in Ko’s paper [38] and mudline.
Advances in Civil Engineering 15

0.6 4 0.6 2 0.6 4

Model 1 Model 2 Model 3
0.5 0.5 0.5
Frequency (Hz)

Frequency (Hz)

Frequency (Hz)
0

Deviation (%)

Deviation (%)

Deviation (%)
2 2
0.4 0.4 0.4
–2
0.3 0 0.3 0.3 0
–4
0.2 0.2 0.2
–2 –2
0.1 0.1 –6 0.1
0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5
μ μ μ

Finite element method

Simplified calculation method
Percentage difference

FIGURE 17: Natural frequency comparison considering the SSI.

TABLE 4: Comparison of natural frequencies calculated using various methods.

Method Natural frequency (Hz) Deviation (%) Deviation (%)
Ko [38] 0.3243 −0.0926 −1.502
Yang [40] 0.3321 −2.500 −3.944
Arany et al. [10] 0.2788 13.951 12.739
Method herein 0.3115 3.858 2.504
ADAMS Jonkman et al. [41] 0.3195 1.389 –
FAST Jonkman et al. [41] 0.3240 – −1.408

TABLE 5: Comparison of natural frequencies calculated using various methods.

Natural frequency Natural frequency
Method Deviation (%) Deviation (%) Deviation (%)
without SSI (Hz) with SSI (Hz)
Ko [38] 0.228 −0.885 0.2271 −13.041 0.395
Yang [40] 0.192 15.044 0.1874 6.720 2.396
Arany et al. [10] 0.206 8.850 0.1981 1.394 3.835
Method herein 0.221 2.212 0.2176 −8.313 1.538
Alkhoury et al. [42] 0.226 – 0.2009 – 11.106

In the Arany calculation method, it is assumed that a foundations, the algorithm can consider not only the impact
tapered tower with a variable wall thickness is ideally equiv- of the structural wall thickness changes on the natural fre-
alent to a tower with a constant diameter and wall thickness quency but also the impact of more complex cross-sectional
and that the structure above the mudline comprises a tapered changes; thus, the algorithm provides more accurate calcula-
tower and substructure (monopile and transition piece). tions and can be applied more widely.
Therefore, this method is suitable for models with tapered Moreover, a coefficient μ was introduced to consider the
towers and substructures above the mudline. influence of the changes in the lumped mass on the tower top
and in the total mass of the tower, the transition piece, and the
4. Conclusion monopile above the mudline of the structure on the natural
frequency. By solving the transcendental Equation (11), the
Based on the Euler–Bernoulli beam theory, an approxima- relationship between μ and J can be established.
tion algorithm was proposed to determine the natural fre- In addition, two finite element calculation models were
quencies of the OWT structures with monopile foundations. established: without and with consideration of the SSI.
The algorithm considers the effects of changes in the diame- According to the finite element calculations, the natural
ter and wall thickness of the structure above the mudline on frequency calculated using the former model is higher
the natural frequency, using segmentation and a weighted than that calculated using the latter. In the calculation
average. The effect of the soil around the monopile on the using the model without considering the SSI, it was found
natural frequency is also considered. Unlike existing numer- that the natural frequency of the OWT structure is over-
ical methods, this algorithm does not require programing. estimated because the model does not adequately include
Compared with other approximation algorithms for the nat- the displacement and slope of the position of the monopile
ural frequencies of the OWT structures with monopile mudline.
16 Advances in Civil Engineering

Appendix The displacement and slope equations for rigid piles

under the condition of constant modulus of the subgrade
A. Flexibility matrix reaction are expressed in Equation (A6), as proposed by
Poulos and Davis [44].
The calculation method for the flexibility matrix of rigid and
slender piles is provided by Poulos and Davis [44] based on 2 4 6 3(
the constant or linear modulus of the subgrade reaction kh . ( ) )
y0 6 kh D P LP kh DP L2P 7 F0
¼6
4 6
7 : ðA6Þ
(1) Cohesive soil ψ0 12 5 M 0
kh DP L2P kh DP L3P
The modulus of the subgrade reaction kh of the cohesive
soil is a constant, which is calculated as follows [45]:
sffiffiffiffiffiffiffiffiffi (4) Noncohesive soil

0:65 12 ED4P E
kh ¼ ; ðA1Þ The noncohesive soil modulus of subgrade reaction kh is
DP EP IP 1 − υ2S linear, and it is calculated as follows [44]:

zs
where DP is the pile diameter, E is the elastic modulus of the k h ¼ nh ⋅ ; ðA7Þ
soil, EP is the elastic modulus of the pile, IP is the moment of Dp
inertia of the pile, and υS is Poisson’s ratio of the soil.
The calculation method for the slenderness parameter β, where nh is the coefficient of the subgrade reaction and zs is
is provided by Poulos and Davis [44]. the depth below the mudline.
rffiffiffiffiffiffiffiffiffiffiffi
4 k h DP (5) Slender and rigid piles in noncohesive soil
β¼ : ðA2Þ
4EP IP
Poulos and Davis [44] proposed a determination method
for slender and rigid piles under the condition of linear
(2) Slender and rigid piles in cohesive soils modulus of subgrade reaction.
Poulos and Davis [44] proposed a determination method (a) Slender pile
for slender and rigid piles under the condition of constant
modulus of subgrade reaction.  1=5
LP >4:0 EnPhIP : ðA8Þ
(a) Slender piles
 1=4
LP >2:5 4E P IP
; ðA3Þ (b) Rigid pile
kh D P
 1=5
LP <2:0 EnPhIP : ðA9Þ
where LP is the embedded length of the pile.

(b) Rigid piles

(6) Stiffness matrix in noncohesive soil
 1=4
LP <1:5 4E P IP
kh D P : ðA4Þ The displacement and slope equations of a slender pile
under the condition of linear modulus of subgrade reaction
are expressed in Equation (A10), as proposed by Poulos and
(3) Stiffness matrix of cohesive soil Davis [44].

The displacement and slope equations of slender piles 2 3

2:40 1:60
under the condition of constant modulus of subgrade reac- ( ) ( )
6 n3=5 ðE I Þ2=5 n2=5 3=5 7
tion are expressed in Equation (A5), as proposed by Poulos y0 6 P P h ðE P I P Þ 7 F0
¼6 h 7 :
and Davis [44]. ψ0 4 1:60 1:74 5 M0
2 3 n2=5
h ðE P I P Þ
3=5 n1=5
h ðE P I P Þ
4=5

( ) 2β 2β2 ( )
6 ðA10Þ
y0 6k D k h DP 7
7 F0
¼ 6 h 2P 7 ; ðA5Þ
ψ0 4 2β 4β3 5 M0
The displacement and slope equations of a rigid pile
k h DP k h DP
under the condition of linear modulus of the subgrade reac-
tion are expressed in Equation (A11), as proposed by Poulos
where β is the slenderness parameter. and Davis [44].
Advances in Civil Engineering 17

2 18 23:94 3(
( ) ) power generation, onshore and offshore,” Wind Engineering,
y0 6 L2P nh L3P nh 7 F0 vol. 26, no. 4, pp. 211–222, 2002.
¼6
4 24
7 : ðA11Þ
36 5 M
[12] S. Bhattacharya, J. A. Cox, D. Lombardi, and D. Muir Wood,
ψ0 0 “Dynamics of offshore wind turbines supported on two
L3P nh L4P nh foundations,” Proceedings of the Institution of Civil Engineers -
Geotechnical Engineering, vol. 166, no. 2, pp. 159–169, 2013.
[13] F. Petrini, H. Li, and F. Bontempi, “Basis of design and
numerical modeling of offshore wind turbines,” Structural
Data Availability Engineering and Mechanics, vol. 36, no. 5, pp. 599–624, 2010.
[14] G. Bertollucci Colherinhas, F. Petrini, M. V. G. de Morais, and
The authors confirm that the data supporting the findings of F. Bontempi, “Optimal design of passive-adaptive pendulum
this study are available within the article. tuned mass damper for the global vibration control of offshore
wind turbines,” Wind Energy, vol. 24, no. 6, pp. 573–595,
2021.
Conflicts of Interest [15] V. Laface, G. Alotta, G. Failla, C. Ruzzo, and F. Arena, “A two-
The authors declare that they have no conflicts of interest. degree-of-freedom tuned mass damper for offshore wind
turbines on floating spar supports,” Marine Structures, vol. 83,
Article ID 103146, 2022.
Acknowledgments [16] G. Alotta, C. Biondo, A. Giaralis, and G. Failla, “Seismic
protection of land-based wind turbine towers using the tuned
National Natural Science Foundation of China funded project inerter damper,” Structures, vol. 51, pp. 640–656, 2023.
(5167082378). [17] V. Jahangiri and C. Sun, “A novel three dimensional nonlinear
tuned mass damper and its application in floating offshore
References wind turbines,” Ocean Engineering, vol. 250, Article ID
110703, 2022.
[1] S. Darvishi-Alamouti, M.-R. Bahaari, and M. Moradi, “Natural [18] S. Xie, X. Jin, J. He, J. Gao, C. Zhang, and Y. Yan, “Applying
frequency of offshore wind turbines on rigid and flexible multiple tuned mass dampers to control structural loads of
monopiles in cohesionless soils with linear stiffness distribu- bottom-fixed offshore wind turbines with inclusion of
tion,” Applied Ocean Research, vol. 68, pp. 91–102, 2017. soil–structure interaction,” Ocean Engineering, vol. 205,
[2] I. Depina, T. M. Hue Le, G. Eiksund, and T. Benz, “Behavior of Article ID 107289, 2020.
cyclically loaded monopile foundations for offshore wind [19] C. Sun and V. Jahangiri, “Bi-directional vibration control of
turbines in heterogeneous sands,” Computers and Geotechnics, offshore wind turbines using a 3D pendulum tuned mass
vol. 65, pp. 266–277, 2015. damper,” Mechanical Systems and Signal Processing, vol. 105,
[3] S. Corciulo, O. Zanoli, and F. Pisanò, “Transient response of pp. 338–360, 2018.
offshore wind turbines on monopiles in sand: role of cyclic [20] J. Yang, E. M. He, and Y. Q. Hu, “Dynamic modeling and
hydro–mechanical soil behaviour,” Computers and Geotech- vibration suppression for an offshore wind turbine with a
nics, vol. 83, pp. 221–238, 2017. tuned mass damper in floating platform,” Applied Ocean
[4] S.-P. Breton and G. Moe, “Status, plans and technologies for Research, vol. 83, pp. 21–29, 2019.
offshore wind turbines in Europe and North America,” [21] B. Fitzgerald, J. McAuliffe, S. Baisthakur, and S. Sarkar,
Renewable Energy, vol. 34, no. 3, pp. 646–654, 2009. “Enhancing the reliability of floating offshore wind turbine
[5] G. M. Álamo, J. J. Aznárez, L. A. Padrón, A. E. Martínez- towers subjected to misaligned wind–wave loading using
Castro, R. Gallego, and O. Maeso, “Dynamic soil–structure tuned mass damper inerters (TMDIs),” Renewable Energy,
interaction in offshore wind turbines on monopiles in layered vol. 211, pp. 522–538, 2023.
seabed based on real data,” Ocean Engineering, vol. 156, [22] Z. Zhang and C. Høeg, “Inerter-enhanced tuned mass damper
pp. 14–24, 2018. for vibration damping of floating offshore wind turbines,”
[6] R. Harrabin, Offshore Wind Power Cheaper than New Nuclear Ocean Engineering, vol. 223, Article ID 108663, 2021.
2017, BBC, 2017. [23] H. Ding, O. Altay, and J.-T. Wang, “Lateral vibration control
[7] Wind Europe, “The european offshore wind industry,” in Key of monopile supported offshore wind turbines with toroidal
Trends and Statistics 2017, p. 37, Wind Europe, Brussels, tuned liquid column dampers,” Engineering Structures,
Belgium, 2018. vol. 286, Article ID 116107, 2023.
[8] D. Lombardi, S. Bhattacharya, and D. Muir Wood, “Dynamic [24] X. Liu, J. Xu, G. He, and C. Chen, “Lateral vibration mitigation
soil–structure interaction of monopile supported wind of monopile offshore wind turbines with a spring pendulum
turbines in cohesive soil,” Soil Dynamics and Earthquake pounding tuned mass damper,” Ocean Engineering, vol. 266,
Engineering, vol. 49, pp. 165–180, 2013. Part 4, Article ID 112954, 2022.
[9] M. Karimirad, Offshore Energy Structures For Wind Power, [25] M. Hussan, M. S. Rahman, F. Sharmin, D. Kim, and J. Do,
Wave Energy and Hybrid Marine Platforms, Springer “Multiple tuned mass damper for multi-mode vibration
International Publishing, Switzerland, 2014. reduction of offshore wind turbine under seismic excitation,”
[10] L. Arany, S. Bhattacharya, J. H. G. Macdonald, and Ocean Engineering, vol. 160, pp. 449–460, 2018.
S. J. Hogan, “Closed form solution of eigen frequency of [26] D. Chen, S. Huang, C. Huang, R. Liu, and F. Ouyang, “Passive
monopile supported offshore wind turbines in deeper waters control of jacket-type offshore wind turbine vibrations by
incorporating stiffness of substructure and SSI,” Soil Dynamics single and multiple tuned mass dampers,” Marine Structures,
and Earthquake Engineering, vol. 83, pp. 18–32, 2016. vol. 77, Article ID 102938, 2021.
[11] J. van der Tempel and D.-P. Molenaar, “Wind turbine [27] G. M. Stewart and M. A. Lackner, “The impact of passive
structural dynamics—a review of the principles for modern tuned mass dampers and wind–wave misalignment on
18 Advances in Civil Engineering

offshore wind turbine loads,” Engineering Structures, vol. 73, [44] H. G. Poulos and E. H. Davis, Pile Foundation Analysis and
pp. 54–61, 2014. Design, Hongqiao Bookstore Co. Ltd, Taipei, 1981.
[28] J. W. Zhang, X. Liang, L. Z. Wang, B. X. Wang, and [45] A. B. Vesić, “Bending of beams resting on isotropic elastic
L. L. Wang, “The influence of tuned mass dampers on solid,” Journal of the Engineering Mechanics Division, vol. 87,
vibration control of monopile offshore wind turbines under no. 2, pp. 35–53, 1961.
wind–wave loadings,” Ocean Engineering, vol. 278, Article ID
114394, 2023.
[29] V. Jahangiri and C. Sun, “Three-dimensional vibration control
of offshore floating wind turbines using multiple tuned mass
dampers,” Ocean Engineering, vol. 206, Article ID 107196,
2020.
[30] Z. Lei, G. Liu, and M. Wen, “Vibration attenuation for offshore
wind turbine by a 3D prestressed tuned mass damper
considering the variable pitch and yaw behaviors,” Ocean
Engineering, vol. 281, Article ID 114741, 2023.
[31] W. Wang, X. Li, H. Zhao, B. Wang, and Y. Li, “Vibration
control of a pentapod offshore wind turbine under combined
seismic wind and wave loads using multiple tuned mass
damper,” Applied Ocean Research, vol. 103, Article ID 102254,
2020.
[32] D. Leng, Y. Yang, K. Xu et al., “Vibration control of offshore
wind turbine under multiple hazards using single variable-
stiffness tuned mass damper,” Ocean Engineering, vol. 236,
Article ID 109473, 2021.
[33] Z. Zhang, “Vibration suppression of floating offshore wind
turbines using electromagnetic shunt tuned mass damper,”
Renewable Energy, vol. 198, pp. 1279–1295, 2022.
[34] A. Hemmati, E. Oterkus, and M. Khorasanchi, “Vibration
suppression of offshore wind turbine foundations using tuned
liquid column dampers and tuned mass dampers,” Ocean
Engineering, vol. 172, pp. 286–295, 2019.
[35] H. Zuo, K. Bi, and H. Hao, “Using multiple tuned mass
dampers to control offshore wind turbine vibrations under
multiple hazards,” Engineering Structures, vol. 141, pp. 303–
315, 2017.
[36] S. Adhikari and S. Bhattacharya, “Dynamic analysis of wind
turbine towers on flexible foundations,” Shock and Vibration,
vol. 19, p. 20, Article ID 408493, 2012.
[37] L. Arany, S. Bhattacharya, S. J. Hogan, and J. Macdonald,
“Dynamic soil-structure interaction issues of offshore wind
turbines,” in Proceedings of the 9th International Conference
on Structural Dynamics, EURODYN 2014, A. Cunha,
E. Caetano, P. Ribeiro, and G. Müller, Eds., pp. 3611–3617,
Porto, Portugal, 2014.
[38] Y.-Y. Ko, “A simplified structural model for monopile-
supported offshore wind turbines with tapered towers,”
Renewable Energy, vol. 156, pp. 777–790, 2020.
[39] R. W. Clough and J. Penzien, Dynamics of Structures, Higher
Education Press, Beijing, 2006.
[40] C. B. Yang, “The Regulation, control and evaluation of dynamic
response for offshore wind turbine foundations,” Ph.D. Thesis,
Tsinghua University, Beijing, China, 2017.
[41] J. Jonkman, S. Butterfield, W. Musial, and G. Scott, “Definition
of a 5-MW reference wind turbine for offshore system
development,” Technical Report. NREL/TP-500-38060,
National Renewable Energy Laboratory, USA, 2009.
[42] P. Alkhoury, A.-H. Soubra, V. Rey, and M. Aït-Ahmed, “A full
three-dimensional model for the estimation of the natural
frequencies of an offshore wind turbine in sand,” Wind Energy,
vol. 24, no. 7, pp. 699–719, 2021.
[43] C. Bak, F. Zahle, R. Bitsche et al., “Description of the DTU 10 MW
reference wind turbine,” pp. 1–138, 2013, DTU Wind energy
report-I-0092, https://dtu-10mw-rwt.vindenergi.dtu.dk.