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

net/publication/320749558

Performance Evaluation of Submerged Floating Tunnel Subjected to

Hydrodynamic and Seismic Excitations

Article  in  Applied Sciences · October 2017

DOI: 10.3390/app7111122

CITATIONS READS

7 521

3 authors, including:

Naik Muhammad Zahid Ullah

Balochistan University of Information Technology, Engineering and Management … Hanyang University
13 PUBLICATIONS   16 CITATIONS    13 PUBLICATIONS   18 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Dynamic Analysis of Submerged Floating Tunnel View project

All content following this page was uploaded by Naik Muhammad on 01 November 2017.

The user has requested enhancement of the downloaded file.

 applied
sciences
Article
Performance Evaluation of Submerged Floating
Tunnel Subjected to Hydrodynamic and
Seismic Excitations
Naik Muhammad ID
, Zahid Ullah and Dong-Ho Choi *
Department of Civil and Environmental Engineering, Hanyang University, 222 Wangshimni-ro,
Seoul 04763, Korea; naik2954@hanyang.ac.kr (N.M.); engrzahid@hanyang.ac.kr (Z.U.)
* Correspondence: samga@hanyang.ac.kr; Tel.: +82-222-20-4154

Received: 24 September 2017; Accepted: 27 October 2017; Published: 31 October 2017

Abstract: Submerged floating tunnels (SFTs) are innovative structural solutions to waterway crossings,
such as sea-straits, fjords and lakes. As the width and depth of straits increase, the conventional
structures such as cable-supported bridges, underground tunnels or immersed tunnels become
uneconomical alternatives. For the realization of SFT, the structural response under extreme
environmental conditions needs to be evaluated properly. This study evaluates the displacements and
internal forces of SFT under hydrodynamic and three-dimensional seismic excitations to check the
global performance of an SFT in order to conclude on the optimum design. The formulations
incorporate modeling of ocean waves, currents and mooring cables. The SFT responses were
evaluated using three different mooring cable arrangements to determine the stability of the mooring
configuration, and the most promising configuration was then used for further investigations.
A comparison of static, hydrodynamic and seismic response envelope curves of the SFT is provided
to determine the dominant structural response. The study produces useful conclusions regarding the
structural behavior of the SFT using a three-dimensional numerical model.

Keywords: submerged floating tunnel (SFT); mooring cable configuration; dynamics of SFT;
hydrodynamic response; seismic response

1. Introduction
Submerged floating tunnels (SFTs) are tubular structures that float at a specified depth and are
supported by mooring cables. The balance between pre-tensions in the mooring cables and net positive
buoyancy (or residual buoyancy) maintains the structural stability of the SFT. The cost of an SFT per unit
length remains almost constant with increasing length [1,2]. Therefore, it is considered an economical
alternative for waterway crossings in comparison to cable-supported bridges, underground tunnels or
immersed tunnels, especially for deep and wide crossings. In addition, SFT is more advantageous,
because it is not bound to geographic terrain and does not interfere with water surface traffic.
The research efforts for development of the SFT began in 1923 when the first Norwegian patent
was issued for it. Since then, many case studies and proposals have been made for different proposed
projects, such as Hogsfjord in Norway [3], Funka Bay in Japan [4,5], Messina Strait in Italy [6,7],
Qiandao Lake in China [8,9] and Mokpo-Jeju SFT in South Korea [10].
The lack of experimental and model test results may be one of the reasons why an SFT has not yet
been constructed. The work in [11] calculated the wave loads on a circular SFT using the boundary
element method (BEM) and Morison’s equation and compared the results with model tests. Wave
loads on an elliptical SFT were calculated similarly by [12] using Morison’s equation and compared
with the test results. These studies indicated that both BEM and Morison’s equation gave reliable
estimates of wave loads on SFT.

Appl. Sci. 2017, 7, 1122; doi:10.3390/app7111122 www.mdpi.com/journal/applsci

 Appl. Sci. 2017, 7, 1122 2 of 17

with the test results. These studies indicated that both BEM and Morison’s equation gave reliable
estimates
Appl. Sci. 2017,of
7, wave
1122loads on SFT. 2 of 17
Researchers have focused the research developments on the behavior of SFT for more than three
decades. A formulation for the dynamic analysis of the SFT under seismic and sea waves effects was
Researchers have focused the research developments on the behavior of SFT for more than three
presented by [13]. A primary attempt to analyze the fluid-structure interaction (FSI) of the SFT using
decades. A formulation for the dynamic analysis of the SFT under seismic and sea waves effects was
a two-dimensional model was made by [14], followed by a more comprehensive presentation of FSI
presented by [13]. A primary attempt to analyze the fluid-structure interaction (FSI) of the SFT using
by [15] using the finite element implementation of the Navier–Stokes equation. The soil-structure
a two-dimensional model was made by [14], followed by a more comprehensive presentation of FSI
interaction of SFT is another issue that needs to be investigated properly. Some SFT numerical models
by [15] using the finite element implementation of the Navier–Stokes equation. The soil-structure
have idealized the soil-structure interaction as horizontal, vertical and rotational springs at the
interaction of SFT is another issue that needs to be investigated properly. Some SFT numerical
connecting points [16–18]. The vortex-induced vibrations of mooring cables caused by water currents
models have idealized the soil-structure interaction as horizontal, vertical and rotational springs at the
can produce fatigue damage to the cables [19–21]. Since SFT is considered an alternative solution for
connecting points [16–18]. The vortex-induced vibrations of mooring cables caused by water currents
transportation, moving load analysis or dynamic analysis under traffic loading is a key demand of
can produce fatigue damage to the cables [19–21]. Since SFT is considered an alternative solution for
the structure, but the research efforts are very limited regarding this problem. However, a primary
transportation, moving load analysis or dynamic analysis under traffic loading is a key demand of
attempt was made by [22], who investigated the effects of two- and three-dimensional added mass
the structure, but the research efforts are very limited regarding this problem. However, a primary
models on the response of SFT under single moving loads.
attempt was made by [22], who investigated the effects of two- and three-dimensional added mass
The numerical models described above focused on the dynamic response of SFT, including some
models on the response of SFT under single moving loads.
recent developments [10,23,24]. However, the global performance of the SFT under complex
The numerical models described above focused on the dynamic response of SFT, including
environmental conditions is not yet well understood. The determination of stable mooring cable
some recent developments [10,23,24]. However, the global performance of the SFT under complex
configurations for the SFT is one of the challenges for its realization [12,13]. Especially, the excessive
environmental conditions is not yet well understood. The determination of stable mooring cable
lateral displacement of the SFT due to waves and currents make it questionable for safe
configurations for the SFT is one of the challenges for its realization [12,13]. Especially, the excessive lateral
transportation. Three different cable configurations were analyzed numerically by
displacement of the SFT due to waves and currents make it questionable for safe transportation. Three
Mazzolani et al. [8] using the commercial software package ABAQUS/Aqua for the hydrodynamic
different cable configurations were analyzed numerically by Mazzolani et al. [8] using the commercial
conditions of Qiandao Lake (People’s Republic of China). Similar configurations are evaluated
software package ABAQUS/Aqua for the hydrodynamic conditions of Qiandao Lake (People’s Republic
further in the present study for an SFT with a cable inclination of 45°, in order to check the
of China). Similar configurations are evaluated further in the present study for an SFT with a cable
performance of SFT. The mooring cables with this inclination are more stable [21,25]. In addition, a
inclination of 45◦ , in order to check the performance of SFT. The mooring cables with this inclination are
comparison of the SFT responses for static, hydrodynamic and seismic loadings needs to be provided
more stable [21,25]. In addition, a comparison of the SFT responses for static, hydrodynamic and seismic
for SFT performance quantification under different conditions.
loadings needs to be provided for SFT performance quantification under different conditions.
More explicitly, this study evaluates the global performance of an SFT taking into account
More explicitly, this study evaluates the global performance of an SFT taking into account
hydrodynamic and seismic excitations. The formulation of the problem includes the modeling of
hydrodynamic and seismic excitations. The formulation of the problem includes the modeling of
cable element, hydrodynamics, ground motions and the SFT structure itself. A numerical model of
cable element, hydrodynamics, ground motions and the SFT structure itself. A numerical model
the SFT prototype to be built in Qindao Lake is used to evaluate the global performance under
of the SFT prototype to be built in Qindao Lake is used to evaluate the global performance under
different loadings, and some useful conclusions are drawn that can provide an initial step towards
different loadings, and some useful conclusions are drawn that can provide an initial step towards the
the realization of this innovative structural solution for waterway crossings.
realization of this innovative structural solution for waterway crossings.
2. Governing Equations of Motion
2. Governing Equations of Motion

2.1.Modeling
2.1. ModelingofofWaves
Wavesand
andCurrents
Currents

AAschematic
schematiclayout
layoutofofthe
thesubmerged
submergedfloating
floating tunnel
tunnel (SFT)
(SFT) is is shown
shown inin Figure
Figure 1. 1.
TheThe X-,Y-Y-and
X-, and
Z-axes are pointed along the axis of the tunnel, traverse and vertical to the tunnel axis, respectively.
Z-axes are pointed along the axis of the tunnel, traverse and vertical to the tunnel axis, respectively.
The
The tunnel
tunnel is is supported
supported byby vertical
vertical and
and inclined
inclined mooring
mooring cables.
cables.

Free water surface Z

Wave
X Y
h1
Tunnel

Tunnel Mooring cable

Mooring cable
h

Seabed

(a) (b)
Figure 1. Schematic layout of the submerged floating tunnel (SFT): (a) side view of the SFT and
Figure 1. Schematic layout of the submerged floating tunnel (SFT): (a) side view of the SFT and (b)
(b) cross-sectional
cross-sectional view view
of theof the SFT.
SFT.
Appl. Sci. 2017, 7, 1122 3 of 17

Most of the wave theories used for the evaluation of wave forces on offshore structures are based
on three parameters, i.e., water depth, wave height and wave period. The Airy wave theory, which
holds for wave heights up to 10–15 m, covers a wide range and can consider significant waves of any
actual sea states [13]. As the SFT is located at locations where the significant wave height is smaller
than the water depth, the Airy wave theory can be used to model the sea states. According to the Airy
wave theory, the first-order velocity potential Φ is given as follows [26]:

gH cosh[k ( Z + h)]
Φ= sin α (1)
2ω cosh(kh)

where α = k(Y ± ct) = kY ± ωt; k (= 2π/λ) is the wave number; ω (= 2π/T ) is the wave frequency;
λ (= gT 2 /2π ) is the wavelength selected on the basis of deep-water criteria; H is the wave height; c is
the wave speed; Y and Z are the vertical and transverse coordinates, respectively, having their origin
at the free water surface; and h is the water depth. From the first-order velocity potential Φ, the water
particle velocities in the transverse (Y) and vertical (Z) directions can be obtained as follows:

. dΦ gkH cosh[k ( Z + h)]

u= = cos α (2)
dY 2ω cosh(kh)

. dΦ gkH sinh[k ( Z + h)]

v= = sin α (3)
dZ 2ω cosh(kh)
Differentiating the above equations with respect to time, the water particle accelerations in the
transverse and vertical directions can be given as follows:

.. gkH cosh[k( Z + h)]

u= sin α (4)
2 cosh(kh)

.. gkH sinh[k( Z + h)]

v=− cos α (5)
2 cosh(kh)
The calculation of hydrodynamic forces on the SFT is one of the primary tasks in the design of
the structure. It is also one of the most challenging tasks during numerical simulations due to the
involvement of the fluid-structure interaction phenomena [27]. Morison’s equation can be used to
evaluate the hydrodynamic forces on the SFT and gives reliable estimates as compared to experimental
data [11,12]. The hydrodynamic forces from ocean waves and currents, acting per unit length of the
SFT, are given by the modified Morison’s equation as follows [26].
n . . .  .
 . .
o 2  .. 2  ..
{ f (qi , t)} = 12 CD ρw D wi ± W i − qi  wi ± W i − qi + CM ρw πD wi − C A ρw πD qi (i = 1, 2) (6)

4 4

. . . ..
where subscript i denotes the Y or Z direction; wi is the water particle velocity (u or v); wi is the water
.. .. . . .
particle acceleration (u or v); W i is the velocity (U or V) of water currents acting at the centerline of the
. . . . . .
tunnel and is obtained by the linear equation: U = [(h − Z )/h]U c or V = [(h − Z )/h]V c ; U c and V c
. ..
are the velocities of surface currents in the Y and Z directions, respectively. qi and qi are the structural
velocity and acceleration, respectively; ρw is the density of water; D is the external diameter of SFT; CD
is the drag coefficient; C M is the inertia coefficient; and C A = C M − 1 is the added mass coefficient.
In Equation (6), the first term on the right-hand side denotes the drag force; the second term
denotes the inertia force; and the third term denotes the added mass effect. The drag force is much
smaller than the inertia [12], so the drag force term is linearized by removing the structural velocity
term for numerical simplification. Substituting Equations (2)–(5) into Equation (6), the horizontal and
vertical components of hydrodynamic forces in the Y-direction and the Z-direction can be obtained
as follows:
Appl. Sci. 2017, 7, 1122 4 of 17

. .   . . 
FYD = 12 CD ρw D u ± U  u ± U

2 ..
FYI = C M ρw πD
4 u
. .   . .  (7)
1
FZD = 2 CD ρw D v ± V  v ± V
2 ..
FZI = C M ρw πD
4 v

where the subscripts YD and YI represent the fluid drag and inertia forces in the Y-direction
respectively; the subscripts ZD and ZI represent the fluid drag and inertia forces in the
Z-direction, respectively.

2.2. Equations of Motion

The equation of motion for the SFT moored by the mooring cables and subjected to gravity,
hydrodynamic and seismic loads can be written in matrix form as follows:
 .. . n .. o
[ M + Ma ] q + [C ] q + [Ke + Km ]{q} = { f } + { f (q, t)} − [ M + Ma ]{ I } q g (8)

where [ M] is the structural mass matrix of SFT; [ Ma ] is the added mass matrix, which is the mass
contribution by water and is obtained from the third term on the right hand side of Equation (6). [C ] is
the system damping matrix. [Ke ] and [Km ] are the elastic stiffness of the SFT and the mooring stiffness
matrices, respectively. { f (q, t)} is the vector of time-dependent hydrodynamic forces. { f } is the vector
..
n o
showing hydrostatic and live loads acting on the SFT. q g is the vector of ground motions acceleration.
{ I } is the influence coefficient vector, having one for elements corresponding to the degrees of freedom
 ..  .
in the direction of the applied ground motion and zero for the other degrees of freedom. q , q
and {q} are the vectors representing structural acceleration, velocity and displacement, respectively.
The vector {q} for an element is given as follows:
h iT
{q} = q X1 qY1 q Z1 θY1 θ Z1 q X2 qY2 q Z2 θY2 θ Z2 (9)

where q X , qY and q Z are the displacements along the X, Y and Z directions, respectively; while θY and
θ Z are the rotations about the Y-axis and Z-axis, respectively. The subscripts 1 and 2 represent the first
and second nodes of an element, respectively.
The structural mass [ M] and the added mass [ Ma ] are lumped at the nodes. The added mass is
applied in the traverse and vertical directions only. The elastic stiffness [Ke ] is calculated using the
3D beam element, and the nonlinear mooring stiffness [Km ] is calculated using the truss element. We
used both cable elements and truss elements for modeling the mooring cables, and there was almost
no difference in the response of the SFT due the fact that mooring cables always remain in tension
under the input conditions of the present study. Therefore, we adopted the truss element. However,
cable elements should be used in general when severe environmental conditions exist. In modeling the
mooring cables, the residual buoyancy, which is the net hydrostatic force, was used as pretensions in
the cable elements. The damping matrix [C ] is calculated using the Rayleigh damping model as follows:

[C ] = β 1 [ M + Ma ] + β 2 [ Ke + Km ] (10)

where:
2ωi ω j
β1 = ζ (11)
ωi + ω j
2
β2 = ζ (12)
ωi + ω j
where β1 and β2 are the Rayleigh damping coefficients; ζ is the modal damping ratio; ωi and ω j are the
ith and jth modal natural frequencies of the structure, respectively. The mass and stiffness proportional
 Appl.
Appl.Sci.
Appl. Sci.2017
Sci. 2017
2017, 7, 7,1122
,, 7, 1122
1122 Appl. Sci. 2017, 7, 1122Appl. Sci. 2017, 7, 1122 6 66ofof
of1717
17
Appl. Sci. 2017, 7, 1122 Appl. Sci. 2017, 7, 1122 6 of 17
Appl. Sci. 2017, 7, 1122 Z ZAppl. Sci. 2017, 7, 1122
Z 5 of 17
Z
Appl. Sci. 2017, 7, 1122Z 6 of 17
YZY XX
Y X Y X YZ X
YZ X Z Xtwo 2A22mm B modal2 m C 2
Rayleigh damping coefficients are chosen Abased on the
AA BBB Yfirst A Cmodes
C
C of
B SFT, m
so that
C the
Y
damping ratio is 2.5%. X same damping
The A ratios haveBbeen Y used X forC the structure2Am and for the B cables C 2
due to the fact that SFT is a tubular structure
A and resembles B a pipeC[28]. 2Am B C 2
The numerical procedure for the static and dynamic analysis is briefly described here. In the
static analysis of SFT, the maximum wave and current25.8 m forces were applied
25.8mm
25.8 25.8 m as uniformly-distributed
25.8 m
static loads. In the dynamic analysis, the equations of
25.8 m motion were solved by Newmark’s 25.8 direct
m time
integration method. A computer program was 25.8 developed
m
to perform the dynamic analysis
25.8 m
of SFT
subjected to net residual buoyancy and time-dependent hydrodynamic and seismic forces. The tunnel
was modeled using FEM with30 30
30m3D-beam
m elements,
2020 30 and
20mm
m m the torsional
2020
mm
20 m 20 effect 30
30 m was
30mm20
m neglected
m 20 m because 20 it
30 m
m 30 m
30 m 30
was very small. 30 m 20 m 20 m 30 m 30 m 20 m 20 m 30 m
(a)
(a)
(a) (a) (a)
30 m 20 m (a) 20 m 30 m 30 m 20 m (a) 20 m 30 m
3. Numerical Example
(a) (a)
The performance evaluation of SFT needs to be carried out using a realistic model. Therefore,
the prototype of the proposed SFT model, as shown in Figure 2 [8,9], which is planned to be built in
Qiandao Lake, China,
1 11 is evaluated in this
1 2 2study. The factored
1 2load combinations 2 used by [8] were used
2
for both static and hydrodynamic analyses in the present study. 3 33 3 3
1 2 1 2
Figure 2a shows the side view of the SFT model;1the SFT is supported at 3 Locations A, B and C 3
1 2 2
by mooring cables. Figure 2b shows three different types of mooring cables arrangements. 3 Figure 2c 3
(b)
(b)
(b) (b) (b)
shows the material cross-section of the SFT. Three different SFT cases are analyzed based on three
Aluminum (b) (b)(t=0.1 m)
different mooring cable arrangements defined in Table 1,(t=0.1
Aluminum (t=0.1m)
where Aluminum
m)
the (t=0.1Aluminum
circled numbers m)
represent the cable
arrangement shown in Figure 2b. The input Aluminum (b)(t=0.1
parameters for SFTm) tunnel, hydrodynamic and(b)
Aluminum (t=0.1 m)
mooring
Aluminum
cables are given in Table 2; the hydrodynamic (t=0.1
parameters arem) Aluminum
those measured at Qindao Lake(t=0.1
[8]. m)
The assumptions made in this study and criteria are given as follows:
m m

3.55 m

m m
m
3.55 m
m m

4.39 m

m m
m
4.39 m
m 3.55

m 3.55
• The cross-section of the SFT is hollow circular, consisting of an external aluminum layer followed
3.55
m 4.39

m 4.39
4.39

by a sandwiched concrete layer and internal steel layer. The material’s thicknesses and tunnel
3.553.55

3.553.55
4.394.39

4.394.39

diameters are shown in Figure 2c. A uniformly-distributed load of 125 kN/m, per unit length,
was used for the tunnel. The Archimedes buoyancy was 160 kN/m (ρw gATunnel = 1050 × 10 ×
π × 4.42/4 = 160 kN/m), and the net residual buoyancy was 35 kN/m (160 − 125 = 35 kN/m).
Concrete
Concrete (t(t= =0.3
0.3 Concrete (t = 0.3 Concrete (t = 0.3
A uniformly-distributed live load Steel (t(t=kN/m
of 10
Steel =0.02
0.02m)
was Steel (t =to0.02
m) assumed m)Steel
be acting on(tthe
= 0.02 m)
tunnel;
m)
Concretem) (t = 0.3 m) Concrete m) (t = 0.3
• The mooring cable connections at Steel the (tanchor
= 0.02point
m) (seabed) and tunnelSteel were(ttreated
= 0.02 m)
as pins; and
m) (t = 0.3 (c) m) (t = 0.3
• Steel (t = 0.02
The tunnel displacements were restrained at one (c)Concrete
(c)(c)
m)end, Steel (t = 0.02
while the axial displacement was (c) Concrete
m)left free at
(c) m) (c) m)
the other
Figureend.
Figure
Figure2.2. The flexural
2.Schematic
Schematic
Schematic rotations
layout
Figure
layout
layoutofof
of2.the
the
the were
Schematic
submergedallowed
Figure
submerged
submerged layout at
floating
2. of
floating
floatingboth
Schematic
the ends
tunnel
submerged
tunnel
tunnel of the
layout
(SFT):
(SFT):
(SFT):(a)
of tunnel.
floating
(a)
(a)the
sidesubmerged
side
sideview
tunnel
viewofof
view (SFT):
ofthefloating
the
the SFT(a) and
SFT
SFT side
tunnel
and
and view(SFT):
of the
(a)SFT
sidea
(b)
(b)mooring
mooring
Figure cables
cablesarrangements;
2. Schematic (b) mooring
of theand
arrangements;
layout cables
and(c)(c)
(c)material
(b)
submerged (c)
arrangements;
mooring
2.cross-section
material
Figure cables
andarrangements;
(c)
cross-section
floating
Schematic
tunnel ofmaterial
ofthe
layout
(SFT):theSFT.cross-section
SFT.
(a)
ofSFT.
theand
side (c) material
of of
submerged
view the (c)
the cross-section
SFT.
floating
SFT and of the SFT.
tunnel (SFT): (a) side
(b) mooring cables arrangements; and material cross-section of the
(b) mooring
Figure cables arrangements;
Table
2. Schematic 1. Three
layout theand
of cable (c) material
(b)
submerged mooring
Figure 2.cross-section
configurations
floatingcables
used
Schematic
tunnelarrangements;
of the
to support
layouttheSFT.
(SFT): ofSFT.
(a) theand
side (c) material
submerged
view of the cross-section
floating
SFT and of the SFT.
tunnel (SFT): (a) side
Table
Table
Table1.1.
Three
1. Three
Threecable
cable
cableconfigurations
Table 1. Three
configurations
configurationsusedcable
used
used to
Table
to
tosupport
configurations
1. Three
support
support the SFT.
the
the cable
used
SFT.
SFT. configurations
to support theused
SFT.to support the SFT
(b) mooring cables arrangements; and (c) material
(b) mooring
cross-section
cables arrangements;
of the SFT. and (c) material cross-section of the SFT.
Table 1. Three cable
Mooring configuration configurations
Section A usedSection
to
Table
support
B1. Three
the SFT.
cable configurations
Section C used to support the SFT
Mooring
Mooringconfiguration
configuration MooringSection
configuration
SectionMooring
AA configuration
Section
Section
SectionBA B Section
Section
Section
A B CC
Section Section Section
B C
Table 1. Three cable configurations
1 11 1Mooring used to
Tablesupport
11 111(C1) 1. Three
the SFT.
cable configurations
1Section
1A1 11 C used to
1 support the SFT
Configuration
Mooring
Configuration 1 1(C1)
configuration
1 (C1)
Configuration (C1)Configuration
Section A(C1)
Configuration
configuration
Section B Section Section B1
Configuration
Mooring 21 2(C2)
configuration
Configuration 2 (C2)
Configuration (C2)Configuration
(C1) 11 11 2Mooring
Section A(C2)
Configuration 112 122(C2)
2
configuration
Section (C1)
B 11Section
Section 2 11 11
A C 12
Section B1
213 3(C2)
(C3)
(C3)Configuration 1
11 11 3 (C3) 3 1 1 3 1 3 1
Configuration
Configuration
Configuration 3 (C3) (C1) 1312 33(C2)
Configuration 2 (C3)
(C1) 11 11 11 12
Configuration 23 (C2)
(C3) 11 Configuration 2332 (C2)
(C3) 11 11 32
Table 2.
Table 2.
Configuration
Table Parameters
2. of the
Table
Parameters of
3 (C3)
Parameters SFT
of the 2.
the 1SFT tunnel,
Parameters
SFT tunnel, mooring
tunnel, Table
of the
cables
2. SFT
Parameters
mooring cables
Configuration
mooring cablestunnel,
and hydrodynamics.
mooring
of the SFT cables
and hydrodynamics.
33 (C3)and hydrodynamics.
1 tunnel,
1 andmooring
hydrodynamics.
cables
3 and hydrod
Table 2. Parameters of the SFT tunnel, mooring
Table cables
2. Parameters
and hydrodynamics.
of the SFT tunnel, mooring cables and hydrod
Element
Element Element Parameter Element
Parameter Parameter Unit
Unit Parameter
Value
Value Unit ValueUn
Table 2. Parameters of the SFT tunnel, mooring
Table cables
2. Parameters
and hydrodynamics.
of the SFT tunnel, mooring cables and hydrod
Element Tunnel equivalent
Parameter density
ElementTunnel
Tunnel equivalent density equivalent
kg/m
Tunnel
density
Unit
kg/m 3 2451
3 3 equivalent
Parameter
Value
2451 density
kg/m 3 2451 kgUn
Element Elastic
TunnelElastic modulus
equivalent
Parameter density
Element
modulus Elastic modulus
N/m
kg/m
Tunnel
Unit 23Elastic
2 3
N/m2 Parameter×modulus
10
2451
equivalent1010 N/m
3 × 1010 density
Value 2 3 × 10 10N/
kg
Un
Tunnel
Tunnel Tunnel Area
Elastic
Tunnel Tunnel
Area
modulus
equivalent density Area kg/m
mm
N/m2 2
Tunnel 3Area
23Elastic 5.1
2451 m2
5.110 density
×modulus
10
equivalent 5.1 kg
N/ m
Tunnel Moment ofof
Area
Elastic
Moment inertia
inertia Moment of inertia
Tunnel
modulus mm
N/m24 Moment
3Area
442Elastic of
10inertia
12.3
5.1
×modulus
12.310 m4 12.3 N/ m
Tunnel Length
Moment of
of
Area
Length tunnel
tunnel Length of tunnel
inertia
ofTunnel mmm Length
2 Moment
4 100
of
Areaof
12.3
5.1
100tunnel
inertia m 100 m m
Mooring
Mooringcables
cables Mooring cables
Elastic
Length
Moment Mooring
modulus
of
Elastic oftunnelcables
inertia
modulus Elastic modulus
N/m
m 1.4
2Elastic
Length
m Moment
N/m4 2
2 ×
100
1.4of 10
modulus
1111 N/m2
tunnel
of× inertia
12.3 1011 1.4 × 1011
N/ m
m
Mooring cables ElasticMooring
Length moduluscables
of tunnel N/m 1.4100
Elastic
m 2Length × 10
modulus
of 11
tunnel N/m
Mooring cables ElasticMooring
moduluscables 1.4modulus
N/m2Elastic × 1011 N/
Appl. Sci. 2017, 7, 1122 6 of 17
Appl. Sci. 2017, 7, 1122 6 of 17

Z
Y X

A B C 2m

25.8 m

30 m 20 m 20 m 30 m

(a)

1 2
3

(b)

Aluminum (t=0.1 m)
3.55 m
4.39 m

Concrete (t = 0.3
Steel (t = 0.02 m)
m)
(c)

Figure 2. Schematic layout of the submerged floating tunnel (SFT): (a) side view of the SFT and
Figure 2. Schematic layout of the submerged floating tunnel (SFT): (a) side view of the SFT and (b)
(b) mooring cables arrangements; and (c) material cross-section of the SFT.
mooring cables arrangements; and (c) material cross-section of the SFT.

Table 1. Three cable configurations used to support the SFT.

Table 2. Parameters of the SFT tunnel, mooring cables and hydrodynamics.
Mooring configuration Section A Section B Section C
Element
Configuration 1 (C1) Parameter
1 1 Unit 1 Value
Configuration 2 (C2) 1
Tunnel equivalent density 2 kg/m3
1 2451
Configuration 3 (C3) Elastic
1 modulus 3 N/m2 1 3 × 10
10

Tunnel Area m2 5.1

Moment of inertia m4 12.3
Table 2. Parameters of the SFT tunnel,
Length of tunnel mooring cables and hydrodynamics.
m 100
2 1.4 × 1011
Element ElasticParameter
modulus N/mUnit Value
Diameter of cable density
Tunnel equivalent m kg/m3 0.06
2451
Mooring cables −7
Moment of inertia m4 6 × 10
Elastic modulus N/m2 3 × 7850
1010
Cable density kg/m3
Tunnel Area m2 5.1
Moment of inertia m4 12.3
Length of tunnel m 100
Mooring cables Elastic modulus N/m2 1.4 × 1011
Appl. Sci. 2017, 7, 1122 7 of 17
Appl. Sci. 2017, 7, 1122 7 of 17
Diameter of cable m 0.06
Moment of inertia m4 6 × 10−7
Table 2. Cont.
Cable density kg/m3 7850
Element Wave height (H)
Parameter Unit m 1
Value
Time period (T) s 2.3
Wave height (H) m 1
Surface
Timecurrent velocity ( U c )
period (T) s m/s 0.12.3
.
Surface Depth
currentof water(U
velocity (h)c ) m/s m 300.1
Hydrodynamics Depth of water (h) m
Hydrodynamics Distance of SFT from free surface (h1) m 2 30
Distance of SFT from free surface (h1 ) m 2
Density of water
Density of water (ρw ) (ρ w) kg/m
kg/m3
3 1050
1050
Drag
Drag coefficient
coefficient (CD )(CD) – -- 11
Inertia
Inertia coefficient
coefficient (CM )(CM) – -- 22

4.
4. Results
Results

Hydrodynamic Response
4.1. Hydrodynamic
comparativestudy
A comparative studyof ofthe
theSFT
SFTresponse
responseusing
usingthethethree
threedifferent
differentcable
cable configurations
configurations defined
defined in
in Table
Table 1 is1 is presented.
presented. TheThe configuration
configuration thatthat proves
proves totobebethe
themost
mostpromising
promisingisis then
then used
used for
further investigations.
The wave and current forces are obtained using the hydrodynamic conditions of Qiandao Lake
(Table 2).
(Table 2).These
Theseforces acting
forces on the
acting on SFT
the are
SFTcalculated using Morison’s
are calculated equation.equation.
using Morison’s During the numerical
During the
simulations,
numerical the drag forces
simulations, the and
draginertia
forcesforces were considered
and inertia forces were distributed loadings
considered and were
distributed converted
loadings and
to equivalent
were converted nodal forces using
to equivalent theforces
nodal work-equivalence method [29]. The
using the work-equivalence time history
method oftime
[29]. The thesehistory
forces
perthese
of unit forces
lengthper
of SFT
unitislength
shown ofin Figure
SFT 3. The
is shown intransverse andtransverse
Figure 3. The vertical inertia forces are
and vertical dominant
inertia forces
hydrodynamic
are forces, while the
dominant hydrodynamic dragwhile
forces, forcesthe
aredrag
veryforces
small.are very small.

Figure
Figure 3.
3. Hydrodynamic
Hydrodynamic force
force time
time history
history per
per unit
unit length
length of
of the
the SFT.
SFT.

4.1.1.
4.1.1. Verification
Verification ofof the
the Numerical
Numerical Model
Model
In
In order
order toto verify the presented
verify the presented formulations
formulations and
and the
the numerical
numerical procedure,
procedure, the
the static
static and
and
dynamic
dynamic responses
responses of of the
the SFT are compared
SFT are compared with
with finite
finite element
element results
results from
from ABAQUS/Aqua,
ABAQUS/Aqua,
Mazzolani
Mazzolani et et al.
al. [8].
[8]. For
For comparison
comparison purposes,
purposes, the
the absolute
absolute displacement
displacement is
is defined as follows:
defined as follows:

DABS  qqY 2  qZ 2 (13)

D ABS = qY 2 + q Z 2 (13)
where DABS is the absolute displacement. The same definition is valid for both static and dynamic
where D ABS is In
displacement. thethe
absolute displacement.
dynamic Themaximum
analysis, the same definition
Y and is valid for both static
Z displacements and used
were dynamic
for
displacement.
calculating theIn the dynamic
absolute analysis, the maximum Y and Z displacements were used for calculating
displacements.
the absolute displacements.
Both static and dynamic absolute displacements based on the same input properties and cable
Both static
configurations areand dynamicwith
compared absolute displacements
[8]. The comparisonsbased onand
of static the dynamic
same input properties
absolute and cable
displacements
configurations are compared with [8]. The comparisons of static and dynamic absolute displacements
using mooring configurations C2 and C3 are shown in Figure 4a,b, respectively. The close agreement
using
of the mooring configurations
results shows C2 of
the accuracy and
theC3presented
are shown in Figure 4a,b, respectively. The close agreement
model.
of the results shows the accuracy of the presented model.
Appl. Sci.Sci.
Appl. 2017, 7,, 1122
2017 7, 1122 8 of 17 17
8 of
Appl. Sci. 2017, 7, 1122 8 of 17

DABS (m)
DABS (m)
(a) (b)
(a) (b)
Figure 4. Comparison of absolute displacements with ABAQUS/Aqua [8]: (a) absolute static
Figure 4.4. Comparison
Comparison
displacement
Figure of absolute
absolute
and (b) absolute
of displacements
dynamic with ABAQUS/Aqua
displacement.
displacements with ABAQUS/Aqua [8]:
[8]: (a)
(a) absolute
absolute static
static
displacementand
displacement and(b)
(b)absolute
absolutedynamic
dynamicdisplacement.
displacement.
4.1.2. Static Response
4.1.2. Static Response
4.1.2. Static Response hydrodynamic forces in the transverse and vertical directions calculated from
The maximum
The maximum
Equation hydrodynamic
(6) or (7) hydrodynamic
were applied asforces forces
equivalentin the transverse and vertical directions calculated from
The maximum in thestatic nodal forces
transverse in the static
and vertical analysis.
directions The intensities
calculated from
Equation (6)
of these(6) or
forces (7) were applied as equivalent static nodal forces in the static analysis. The intensities
Equation or (7)inwere the applied
transverse (Y) and vertical
as equivalent (Z) directions
static nodal forces in were 4.875
the static kN/m and
analysis. The 4.858 kN/m,
intensities
ofrespectively.
these forcesThe in the transverse
static analysis (of
Y)theand vertical
SFT is (Z) for
useful directions
assessing werethe 4.875
performancekN/m andof 4.858
the SFT kN/m,
for the
of these forces in the transverse (Y) and vertical (Z) directions were 4.875 kN/m and 4.858 kN/m,
respectively.
preliminaryThe The
design.static analysis of the SFT is useful for assessing the performance of the SFT for the
respectively. static analysis of the SFT is useful for assessing the performance of the SFT for the
preliminary
Static design.
displacements and internal forces envelope curves of the SFT are shown in Figure 5. The
preliminary design.
Static displacements
transverse response of and
the internal
SFT, forces envelope
i.e., rotations curves of the(SFT
( Z ), displacements ), are
qY SFT shown
bending in Figure(M
moments 5.Z)The
and
Static displacements and internal forces envelope curves of the are shown in Figure 5.
transverse response of the SFT, i.e., rotations ( Z ), displacements ( qY ), bending moments (MZ) and
The transverse
shear forces (SF response of the when
Y) decreased SFT, i.e., therotations
mooring(θcable Z ), displacements
configuration (q was
Y ), bending
changed moments
from C1 (M
to Z)
C3.
shear
and forces
shear forces
However, (SF ) decreased
(SFY ) decreased
the vertical
Y when
response,when the mooring
the rotations
i.e., the cable
mooring cable configuration
( Y ), displacements
configuration was was changed
( qZ ),changed from
bendingfrom C1 to
C1 to C3.
moments C3.
(MY)
However, the
However, the vertical
vertical response,
response, i.e.,i.e., the
the rotations
rotations (θ ( YY),),displacements
displacements (q ( qZZ),),bending
bendingmoments
moments(M (MYY))
and shear forces (SFZ), have different trends, where the response for configuration C2 is the greatest,
and
and shear
shearforces
followed forces (SF
(SFZZ), have different
by configuration C3, and trends,
different trends, where
where the
configuration C1 response
the response for
for configuration
has the minimum configuration
response.C2 is
is the
the greatest,
C2 Comparisons
greatest,of
followed
followed byby configuration
configuration C3, and
C3, and configuration
configuration C1 has
C1 the
has minimum
the minimum
static displacements and internal forces of the SFT for three mooring configurations at Sections response. Comparisons
response. Comparisons of static ofB
A,
displacements
static displacementsand internal
and forces
internal of the
forces SFT
of for
the three
SFT for mooring
three
and C (Figure 2a) are summarized in Table 3 to clearly show the trends described above. configurations
mooring configurationsat Sections
at A, B
Sections andA, CB
(Figure
and C The 2a) are
(Figureshapesummarized
2a) areof summarized
bending in Table in3 Table
moment to clearly
3 to show
gives clearly
direct the trends
show
insightthedescribed
trends
into the above.
described above.
cable configurations and
The
The shape
shape of bending
of bendingmoment moment gives direct
gives insight
direct into the
insight
performances of mooring systems and reflects the uniqueness of the adopted mooring cable cable
into configurations
the cable and performances
configurations and
of mooring systems
performances
configurations, of as and
canreflects
mooring the
systems
be seen inuniqueness 5c. of
and reflects
Figure thethe
From adopted
these mooring
uniqueness
results, cable
itofcan thebeconfigurations,
adopted
concluded thatasthe
mooring can
cablebe
static
seen in Figureprovides
configurations,
response 5c.asFrom
can anthese
be seenresults, it can
in Figure
important be
5c.concluded
benchmarkFrom thesefor that the static
results, response
it can
decision-making providesmooring
be regarding
concluded an important
that the static
cable
benchmark
response for decision-making regarding mooring cable arrangements
arrangements and configuration selection for the preliminary design of SFT. However, thecable
provides an important benchmark for decision-making and configuration
regarding selection
mooring for
static
the preliminary
arrangements
response is very design
andsmall of SFT.
configuration However,
and couldselection the
not be used static
for forresponse
thethepreliminary is very small and
designofofSFT,
practical design could
SFT. not be
as However, used
shown in Figure for the
the static4, as
practical
response design
is very of SFT,
small as
and shown
could innotFigure
will be demonstrated further in the following sections.be 4,
used as
forwill
the be demonstrated
practical design further
of SFT, in
asthe following
shown in sections.
Figure 4, as
will be demonstrated further in the following sections.

(a) (b)
(a) (b)
Figure 5. Cont.
Appl. Sci. 2017, 7, 1122 9 of 17
Appl. Sci. 2017, 7, 1122 9 of 17

(c) (d)
Figure
Figure 5. 5. Comparisonofof
Comparison static
static displacementand
displacement andinternal
internalforce
forceenvelope
envelopecurves
curvesofofthe
theSFT
SFTfor
for three
three
mooring configurations: (a) vertical and transverse rotation; (b) vertical and transverse
mooring configurations: (a) vertical and transverse rotation; (b) vertical and transverse displacement;displacement;
(c)(c) vertical
vertical and
and transverse
transverse bending
bending moment;
moment; and
and (d)(d) vertical
vertical andand transverse
transverse shear
shear force.
force.

Table
Table 3. 3.Comparison
Comparisonofofstatic
staticdisplacements
displacementsand
andinternal
internalforces
forcesofofthe
theSFT
SFTfor
forthree
threemooring
mooring
configurations
configurations atat Sections
Sections A,A, B and
B and C.C.
A B C
C1 A C2 C3 C1 B C2 C3 C1 CC2 C3
C1 C2 C3 C1 C2 C3 C1 C2 C3
θY (×10−3 rad) −0.441 −0.585 −0.512 0.022 0.027 0.024 0.452 0.599 0.524
θY (×10−3 rad) −0.441 −0.585 −0.512 0.022 0.027 0.024 0.452 0.599 0.524
θZ θ(×Z10(×10
−3 rad)
−3 rad) 0.312 0.312 0.1960.196 0.1340.134 −4 ×−4 10×−13
10−13 −4.9−4.9
× 10×−10 −2.2
14 −14 −2.2 × ×1010
−13
−13 −0.312
−0.312 −0.196 −0.134
−0.196 −0.134
qY (×10−2 m) 1.395 0.918 0.663 1.716 1.114 0.791 1.395 0.918 0.663
q (×10−2 m) 2.264 1.395 2.7660.918 2.5110.663 2.6841.716
q Z (×Y10−2 m) 1.114
3.360 0.791
3.016 1.395
2.198 0.918 2.437
2.685 0.663
MY (× 10 3 kN-m) 10.230 11.382 10.796 10.368 17.044 13.648 8.500 9.269 8.878
q (×10 m) −2
2.264 2.766 2.511 2.684 3.360 3.016 2.198 2.685 2.437
MZ (×Z103 kN-m) 7.640 5.263 3.978 9.095 5.128 2.982 7.638 5.262 3.977
× 2 kN) − × −12 − − −1.459 −0.659 −
MY Y (×10 kN-m) 10.230 11.382 10.796
SF ( 10 3 1.459 0.659 0.231 6.2 10
10.368 0.800
17.044 1.228
13.648 8.500 9.269 0.230
8.878
SFZ (×102 kN) −5.028 −4.645 −4.840 −5.552 −2.799 −4.200 −6.555 −9.500 −8.002
M Z (×103 kN-m) 7.640 5.263 3.978 9.095 5.128 2.982 7.638 5.262 3.977

4.1.3.SF (×10 kN)

Dynamic
Y
2
1.459
Response 0.659 0.231 −6.2 × 10−12 −0.800 −1.228 −1.459 −0.659 −0.230
SFZ (×10 kN)
2
−5.028 −4.645 −4.840 −5.552 −2.799 −4.200 −6.555 −9.500 −8.002
The modal analysis was performed, and the natural frequencies of the first 20 natural modes are
listed in Table 4. The comparison of the natural frequencies of the SFT supported by different mooring
4.1.3.configurations
cable Dynamic Response
gives a direct measure of the cable system stiffness. The natural frequencies of
the SFT increase
The modalwhen the was
analysis cable configuration
performed, and is
thechanged
natural from C1 to C3.
frequencies Configuration
of the C3 provides
first 20 natural modes are
more stiffness to the SFT than configurations C1 and C2, and similarly, configuration
listed in Table 4. The comparison of the natural frequencies of the SFT supported by different mooring C2 provides
more
cablethan configuration
configurations C1. aHowever,
gives the natural
direct measure of the frequencies
cable systemof the SFT forThe
stiffness. the natural
three mooring cableof
frequencies
configurations
the SFT increaseare when
very close to each
the cable other for higher
configuration modes.from C1 to C3. Configuration C3 provides
is changed
more stiffness to the SFT than configurations C1 and C2, and similarly, configuration C2 provides
Table 4. Modal analysis of the SFT and comparison of the first 20 natural frequencies (Hz) for each
more than configuration C1. However, the natural frequencies of the SFT for the three mooring cable
cable configuration.
configurations are very close to each other for higher modes.
Mode C1 C2 C3 Mode C1 C2 C3
Table 4. Modal analysis of the SFT and comparison of the first 20 natural frequencies (Hz) for each
1 configuration.
cable 0.5660 0.6997 0.824 11 14.0467 14.0359 14.0408
2 1.2825 1.1542 1.2144 12 20.0309 20.0309 20.0309
Mode
3 C1
2.2606 C2
2.2606 C3
2.2606 Mode
13 C1
20.0417 C2
20.0417 C3
20.0417
41 0.5660
2.5041 0.6997
2.5041 0.824
2.5041 11
14 14.0467
26.2123 14.0359
26.2123 14.0408
26.2123
52 1.2825
5.0739 1.1542
5.0907 1.2144
5.1103 12
15 20.0309
27.0944 20.0309
27.0975 20.0309
27.1010
63 2.2606
5.1324 2.2606
5.1025 2.2606
5.1160 13
16 20.0417
27.1051 20.0417
27.0996 20.0417
27.1021
74 2.5041
8.7454 2.5041
8.7454 2.5041
8.7454 14
17 26.2123
35.1364 26.2123
35.1364 26.2123
35.1364
85 5.0739
8.9895 5.0907
8.9895 5.1103
8.9895 15
18 27.0944
35.1522 27.0975
35.1522 27.1010
35.1522
96 5.1324
9.0143 5.1025
9.0143 5.1160
9.0143 16
19 27.1051
43.6074 27.0996
43.6074 27.1021
43.6074
10 7 8.7454
13.985 8.7454
13.991 8.7454
13.998 17
20 35.1364
44.1137 35.1364
44.1155 35.1364
44.1176
8 8.9895 8.9895 8.9895 18 35.1522 35.1522 35.1522
9 9.0143 9.0143 9.0143 19 43.6074 43.6074 43.6074
10 13.985 13.991 13.998 20 44.1137 44.1155 44.1176
Appl. Sci. 2017, 7, 1122 10 of 17
Appl. Sci. 2017, 7, 1122 10 of 17

Figure
Figure66represents
representsthethecomparison
comparisonofofthe thehydrodynamic
hydrodynamicresponse
responseofofthe
theSFT
SFTatatSection
SectionA,A,using
using
three
three mooring cable configurations. Figure 6a shows the comparison of the SFT displacementsininthe
mooring cable configurations. Figure 6a shows the comparison of the SFT displacements the
transverse direction: configuration C1 has the largest displacement, followed by
transverse direction: configuration C1 has the largest displacement, followed by configuration C2, configuration C2,
and
andconfiguration
configurationC3 C3has
hasthe
theminimum
minimumdisplacement.
displacement.The Themaximum
maximumdisplacements
displacementsof ofthe
theSFT
SFTfor
for
configuration C1, C2 and C3 are 0.046 m, 0.021 m and 0.011 m, respectively. The
configuration C1, C2 and C3 are 0.046 m, 0.021 m and 0.011 m, respectively. The same trend occurs same trend occurs
for
forthe
thetransverse
transverse bending
bending moment
moment (M (Mzz),), as
as shown
shownin inFigure
Figure6b.
6b.The
Themaximum
maximumbendingbendingmoments
moments of
of
SFT for configurations C1, C2 and C3 are 25,066.628 kN-m, 11,943.442 kN-m and 6838.661 kN-m,
SFT for configurations C1, C2 and C3 are 25,066.628 kN-m, 11,943.442 kN-m and 6838.661 kN-m,
respectively.
respectively. Figure
Figure 6c6c shows
shows thethe comparison
comparison of of the
the SFT
SFT displacement
displacement in in the
the vertical
vertical direction;
direction;
configuration C2 has the largest displacement, followed by configuration C3, and
configuration C2 has the largest displacement, followed by configuration C3, and configuration configuration C1 has
C1
the minimum displacement. The maximum displacements of SFT for configurations
has the minimum displacement. The maximum displacements of SFT for configurations C1, C2 C1, C2 and C3 and
are
0.029
C3 are m,0.029
0.035m,m and
0.0350.032
m andm, 0.032
respectively. The sameThe
m, respectively. trend occurs
same foroccurs
trend the vertical
for thebending
verticalmoment
bending
(M Y ),
moment as shown in Figure 6d. The maximum bending moments of SFT for configurations
(MY), as shown in Figure 6d. The maximum bending moments of SFT for configurations C1, C2 andC1,
C3 are 12,202.859 kN-m, 13,694.792 kN-m and 12,934.200 kN-m, respectively.
C2 and C3 are 12,202.859 kN-m, 13,694.792 kN-m and 12,934.200 kN-m, respectively.

5 C1 C2 C3 3 C1 C2 C3
4
Y displacement (x10 -2 m)

3 2
2 1
1
0 0
-1
-2 -1
-3 -2
-4 C1 C2 C3 C1 C2 C3
-5 -3
0 5 10 15 20 25 30 0 5 10 15 20 25 30
Time (s) Time (s)
(a) (b)
3 C1 C2 C3
Z displacement (x10-2 m)

2
C3 C2
MY (x104 kN-m)

1
C1
0

-1

-2

-3
0 5 10 15 20 25 30
Time (s)
(c) (d)
Figure6.6. Comparison
Figure Comparison of
of the
the hydrodynamic
hydrodynamic response
response of
of the
the SFT
SFT at
atSection
SectionA:
A:(a)
(a)Y-displacement;
Y-displacement;
(b) bending moment M Z; (c) Z-displacement; and (d) bending moment MY.
(b) bending moment MZ ; (c) Z-displacement; and (d) bending moment MY .

Figure 7 represents the comparison of the hydrodynamic response of SFT at Section B. Figure 7
Figure 7 represents the comparison of the hydrodynamic response of SFT at Section B. Figure 7
shows the same trend as described to that of Section A. However, the general response at Section B
shows the same trend as described to that of Section A. However, the general response at Section B
is larger than that of Section A. In addition, the response shown by configuration C3 here is further
is larger than that of Section A. In addition, the response shown by configuration C3 here is further
reduced as compared to Section A, because Section B is supported by double inclined mooring cables.
reduced as compared to Section A, because Section B is supported by double inclined mooring cables.
 Appl. Sci. 2017, 7, 1122 11 of 17
Appl. Sci. 2017, 7, 1122 11 of 17

4 C1 C2 C3

3
2

MZ (x104 kN-m)
1
0
-1
-2
-3 C1 C2 C3

-4
0 5 10 15 20 25 30
Time (s)
(a) (b)

(c) (d)
Figure
Figure7.7. Comparison
Comparisonofofthe
thehydrodynamic
hydrodynamicresponse
responseofofthe
theSFT
SFTatatSection
SectionB:B:(a)
(a)Y-displacement;
Y-displacement;
(b) bending moment M Z; (c) Z-displacement; and (d) bending moment MY.
(b) bending moment MZ ; (c) Z-displacement; and (d) bending moment MY .

Both Figures 6 and 7 show that the transient motions of the SFT are small and decay quickly
Both Figures 6 and 7 show that the transient motions of the SFT are small and decay quickly with
with time, and the steady-state motions are more pronounced, especially in the traverse direction, as
time, and the steady-state motions are more pronounced, especially in the traverse direction, as is
is clear from the Y-displacement and bending moment MZ. In the vertical direction, the SFT vibrated
clear from the Y-displacement and bending moment M . In the vertical direction, the SFT vibrated
under the effect of inertial force and buoyancy force likeZ a beam on elastic supports, as shown by
under the effect of inertial force and buoyancy force like a beam on elastic supports, as shown by the
the Z-displacement and bending moment MY in Figures 6c,d and 7c,d, respectively. The SFT moored
Z-displacement and bending moment MY in Figure 6c,d and Figure 7c,d, respectively. The SFT moored
by tension leg mooring cables (C1) showed high peaks in the displacements and bending moments.
by tension leg mooring cables (C1) showed high peaks in the displacements and bending moments.
The SFT moored by a tension leg and a single inclined mooring cable at the center (C2) provided
The SFT moored by a tension leg and a single inclined mooring cable at the center (C2) provided better
better stiffness in the transverse direction, but lesser stiffness in the vertical direction. The SFT moored
stiffness in the transverse direction, but lesser stiffness in the vertical direction. The SFT moored by
by double inclined mooring cables at the center (C3) showed the minimum response in the transverse
double inclined mooring cables at the center (C3) showed the minimum response in the transverse
direction. The response of SFT moored by the C1 and C2 configurations showed large peaks, due to
direction. The response of SFT moored by the C1 and C2 configurations showed large peaks, due to
very small restraining in the transverse direction. Therefore, C3 type configurations should be used
very small restraining in the transverse direction. Therefore, C3 type configurations should be used at
at one or more locations at least, depending on the overall length of the SFT. It can be concluded that
one or more locations at least, depending on the overall length of the SFT. It can be concluded that the
the C3 type cable configuration is more stable.
C3 type cable configuration is more stable.
The SFT moored by mooring cable configuration C3 is used for the analysis hereafter.
The SFT moored by mooring cable configuration C3 is used for the analysis hereafter.
Three-dimensional visualization of the SFT response in both time and space gives a very clear
Three-dimensional visualization of the SFT response in both time and space gives a very clear
understating of the structural behavior of the SFT subjected to hydrodynamic waves and currents.
understating
The of the structural
three-dimensional behavior
time histories ofofthe
thedisplacements
SFT subjectedand to hydrodynamic
bending momentswavesareand currents.
shown in
The three-dimensional time histories of the displacements and bending moments
Figures 8 and 9. The response of the SFT is plotted along the Z-axis; the tunnel longitudinal are shown in
Figures 8 and
coordinates (x)9.are
The response
plotted of the
along the SFT is plotted
X-axis; along
and time the Z-axis;
is plotted thethe
along tunnel longitudinal coordinates
Y-axis.
(x) are plotted along the X-axis; and time is plotted along the Y-axis.
Figure 8 show the displacements along the length of the tunnel with time. Figure 8a illustrates
Figure
that the 8 show the
SFT vibrates displacements
harmonically along
in the the length
Y-direction bothofin
the
thetunnel with
positive time.
and Figureextremities.
negative 8a illustrates
that the SFT vibrates harmonically in the Y-direction both in the positive and negative extremities.
Appl. Sci. 2017, 7, 1122 12 of 17

Appl. Sci. 2017, 7, 1122 12 of 17

Appl. Sci. 2017, 7, 1122 12 of 17

(a)

(a)

(b)
Figure 8. Hydrodynamic displacements along the length of the SFT with time: (a) Y-displacement and
Figure 8. Hydrodynamic displacements along the(b
length
) of the SFT with time: (a) Y-displacement and
(b) Z-displacement.
(b) Z-displacement.
Figure 8. Hydrodynamic displacements along the length of the SFT with time: (a) Y-displacement and
Figure
(b) 8b also shows a harmonic motion pattern, but only in the positive limit (this upward
Z-displacement.
Figure
positive8b also shows
displacement is a harmonic
caused by themotion
residual pattern,
buoyancybut only
acting oninthethe
SFTpositive limit (this
in this direction); upward
which
means
positive that the
displacement
Figure 8b cables were ainharmonic
also isshows
caused taut conditions
by the residual
motion and did notbut
buoyancy
pattern, permit
only slackness
acting onthe
in in this
thepositive
SFT direction.
inlimit
this direction);
(this upward which
means Figure
positive
that the 9a shows
displacement
cables were the bending
is in
caused
taut by moment
the residual
conditions Mand
Y distribution
buoyancy alongon
did not acting
permit the length
the SFT in
slackness ofinthis
the tunnel withwhich
direction);
this direction. time.
The shape
means
Figure that of
the
9a shows bending
cables moments
thewere
bending gives
in taut a clear
conditions
moment MYpicture
and didof the
not
distribution mooring
permit
along cables
slackness
the configuration
in
length this
ofdirection. (C3) with
the tunnel used. time.
The inclined
Figure mooring
9a shows cables
the at themoment
bending central location
M provided
Y distribution flexible
along the elastic of
length bilateral
the restraint
tunnel with to the
time.
The shape of bending moments gives a clear picture of the mooring cables configuration (C3) used.
transverse
The shape and almost moments
of bending unilateral gives
restraint to the
a clear vertical
picture motion.
of the mooring cables configuration (C3) used.
The inclined mooring cables at the central location provided flexible elastic bilateral restraint to the
Figure 9bmooring
The inclined shows the bending
cables at themoment MZ distribution
central location provided along the length
flexible of the tunnel
elastic bilateral withto
restraint time,
the
transverse and and
transverse
almost
and it is similar almost
unilateral
to the one
restraint
relatedrestraint
unilateral
totothe
to a uniform the
vertical
beam
motion.
vibrating
vertical motion.in the first natural mode. It can be seen
Figureonly9binclined
that Figure shows
9b shows the bending
cables,
the located
bending moment MMZZ distribution
at the center,
moment along
provided along
distribution thethe
significant length
restraining
length of the
of the tunnel
in the
tunnel with with
time, time,
transverse
and itand
is similar
direction
it is similarto
(i.e., M the
toZ), one
theand related
onethe to a
othertotwo
related uniform
groupsbeam
a uniform beam vibrating
of vertical cables
vibrating in
in thefromthe first
first left natural
andmode.
natural right It mode.
ofcan
thebe It can be
center
seen
seen that onlyalmost
provided
that only inclined
inclined cables,
nocables,
restraininglocated atatthe
in the
located thecenter,
center,
transverse provided
direction.
provided significant
significant restraining
restraining in thein transverse
the transverse
direction
direction (i.e., M(i.e.,
Z ), Mand
Z ), and
the the
other other
two two
groups groups
of of vertical
vertical cablescables
from from
left left
and and
right right
of theof the
centercenter
provided
provided almost no restraining
almost no restraining in the transverse direction. in the transverse direction.

(a)

(a)

Figure 9. Cont.
Appl. Sci. 2017, 7, 1122 13 of 17
Appl. Sci. 2017, 7, 1122 13 of 17
Appl. Sci. 2017, 7, 1122 13 of 17

(b)
(b)
Figure 9. Hydrodynamic bending moments along the length of the SFT with time: (a) bending
Figure 9. Hydrodynamic bending moments along the length of the SFT with time: (a) bending
Figure
moment Hydrodynamic
9. M bending
Y and (b) bending moments
moment along the length of the SFT with time: (a) bending moment
moment MY and (b) bending momentM
MZZ..
MY and (b) bending moment MZ .
4.2. Seismic Response
4.2. Seismic Response
4.2. Seismic Response
As stated
As stated before,
before, thethe
SFTSFT
is isa amassive
massivetubular
tubular structure,
structure,and itsits
and structural response
structural response dramatically
dramatically
As
increases stated
increases before,
withwith the
external
external SFT is a
forcing.To
forcing. massive
Tocheck tubular
check the structure,
the maximum and
maximum orordominant its
dominant structural
structural response
response
structural dramatically
of
response SFT,
of aSFT, a
comparison
increases
comparison with of static,
external
of static, hydrodynamic
forcing. To and
hydrodynamic andseismic
checkseismic response
responseneeds
the maximum totobebe
or dominant
needs provided [16,23].
structural
provided The El-Centro
response
[16,23]. of SFT,
The El-Centro
a(1940)(1940) ground
comparison of motions
static, are assumedand
hydrodynamic for the SFT numerical
seismic response example
needs totobe
evaluate the [16,23].
provided dynamicThe behavior
El-Centro
ground motions are assumed for the SFT numerical example to evaluate the dynamic behavior
(1940) of the SFTmotions
for the three-dimensional ground motion input.example
Three components of the
El-Centro ground
of the ground are assumed forground
SFT for the three-dimensional the SFTmotion
numerical to evaluate
input. Three components of dynamic
El-Centrobehavior
ground
motions used as input motions are shown in Figure 10. The effective seismic forces using these
of the
motions SFT for the
usedmotionsthree-dimensional
as input ground motion input. Three components of El-Centro ground
ground aremotions
applied are shown in Figure
simultaneously at the10. The points
nodal effective in seismic forces using these
the three-dimensional
motions
ground used as input
motions are motions
applied aresimultaneously
shown in Figureat 10. the
The nodal
effectivepoints
seismicinforces
the using these ground
three-dimensional
numerical model.
motions
numerical aremodel.
applied simultaneously at the nodal points in the three-dimensional numerical model.

(a) (b)

(a) (b)

(c)
Figure 10. Ground motion input (El-Centro 1940 site Imperial Valley irrigation district): (a) S00E
component; (b) S90W component; and (c) vertical component.
(c)
For the evaluation of the ground motion’s effects on the structural behavior of the SFT, two cases
Figure 10.
Figure 10. Ground
Ground motion
motion input
input (El-Centro
(El-Centro 1940
1940 site
site Imperial
Imperial Valley
Valley irrigation
irrigation district):
district): (a)
(a) S00E
S00E
are defined: Case (1): The ground motion component S00E was applied in the longitudinal direction
component;
component; (b) S90W component;
(b)component
S90W component; and (c) vertical
and applied
(c) vertical component.
(X-direction); S90W was in component.
the transverse; and the vertical component was
applied in the vertical direction of the SFT, respectively; this case will be referred to as Load
For the evaluation of the ground motion’s effects on the structural behavior of the SFT, two cases
For the evaluation of the ground motion’s effects on the structural behavior of the SFT, two
are defined: Case (1): The ground motion component S00E was applied in the longitudinal direction
cases are defined: Case (1): The ground motion component S00E was applied in the longitudinal
(X-direction); component S90W was applied in the transverse; and the vertical component was
direction (X-direction); component S90W was applied in the transverse; and the vertical component
applied in the vertical direction of the SFT, respectively; this case will be referred to as Load
was applied in the vertical direction of the SFT, respectively; this case will be referred to as Load
Appl. Sci. 2017, 7, 1122 14 of 17
Appl. Sci. 2017, 7, 1122 14 of 17

Combination
Combination 1 (LC1).
1 (LC1). Case(2):
Case (2):The
Theground
groundmotion
motion component
component S90W S90Wwas wasapplied
appliedininthe
thelongitudinal
longitudinal
direction; component S00E was applied in the transverse direction; the vertical componentwas
direction; component S00E was applied in the transverse direction; the vertical component wasapplied
applied
in the vertical direction of the SFT, respectively; this case will be referred to as
in the vertical direction of the SFT, respectively; this case will be referred to as Load Combination Load Combination 2 2
(LC2). The ground motion component S00W is the maximum ground motion input having a peak
(LC2). The ground motion component S00W is the maximum ground motion input having a peak
ground acceleration (PGA) of 0.35 g. In contrast with the previous sections, the unfactored static and
ground acceleration (PGA) of 0.35 g. In contrast with the previous sections, the unfactored static and
hydrodynamic loads are used here for evaluating the structural behavior of the SFT.
hydrodynamic loads are used here for evaluating the structural behavior of the SFT.
A comparison of the static, hydrodynamic and seismic response envelope curves of the SFT is
A comparison of the static, hydrodynamic and seismic response envelope curves of the SFT is
shown in Figure 11. The maximum response of the tunnel is used to produce the envelope curves.
shown in Figure 11. The maximum response of the tunnel is used to produce the envelope curves.
The SFT response envelope curves of transverse and vertical displacements are shown in Figure 11a,b
Therespectively;
SFT response theenvelope curves
longitudinal of transverse
response of the SFTand is vertical
very smalldisplacements are shown in Figure 11a,b
and is not shown.
respectively; the longitudinal response of the SFT is very small and is not shown.
The transverse displacements of SFT in Figure 11a show a large difference between the static
andThe transverse
dynamic displacements
analysis. of SFTrelative
The approximate in Figure 11a show
differences in amaximum
large difference between
transverse the static
displacements
andofdynamic
the SFT from analysis cases LC2, LC1 and hydrodynamics to that of the static case are 95%, 94% of
analysis. The approximate relative differences in maximum transverse displacements
theand
SFT42%,
fromrespectively.
analysis cases LC2, LC1 and hydrodynamics to that of the static case are 95%, 94% and
42%, respectively.
The vertical displacements of the SFT are nearly the same for static and hydrodynamic analysis.
The vertical
Similarly, for LC1displacements
and LC2, theofvertical
the SFTdisplacements
are nearly theofsame the SFTfor are
static and the
almost hydrodynamic
same, as shownanalysis.
in
Figure 11c. The approximate relative differences in maximum vertical displacements
Similarly, for LC1 and LC2, the vertical displacements of the SFT are almost the same, as shown in of the SFT from
analysis
Figure casesapproximate
11c. The LC2, LC1 relative
and hydrodynamics
differences in to that of vertical
maximum the static case are 28%,
displacements 28%
of the SFTandfrom
1.7%, cases
analysis respectively.
LC2, LC1 and hydrodynamics to that of the static case are 28%, 28% and 1.7%, respectively.
TheThe transverse
transverse responseisismore
response moreaffected
affectedby by the
the ground
ground motions,
motions,while
whilethethevertical
verticalresponse
response is is
more affected by gravity and hydrodynamic (waves and currents) actions.
more affected by gravity and hydrodynamic (waves and currents) actions. The transverse response is The transverse response
is extremely
extremely large large
for theforseismic
the seismic analysis
analysis as compared
as compared to theto the vertical
vertical response;
response; for instance,
for instance, the
the relative
relative difference in transverse and vertical displacements for LC1 and LC2 are 61% and
difference in transverse and vertical displacements for LC1 and LC2 are 61% and 65%, respectively.
65%, respectively.
Figure 11b,d represents the SFT bending moment MZ and bending moment MY envelope curves,
Figure 11b,d represents the SFT bending moment MZ and bending moment MY envelope curves,
respectively. Similar trends prevail for bending moments as described for displacements. The shape of
respectively. Similar trends prevail for bending moments as described for displacements. The shape
bending moments clearly demonstrates the effectiveness of the adopted mooring configuration.
of bending moments clearly demonstrates the effectiveness of the adopted mooring configuration.

(a) (b)

(c) (d)
Figure 11. Comparison of static, hydrodynamic and seismic response envelope curves of the SFT: (a)
Figure 11. Comparison of static, hydrodynamic and seismic response envelope curves of the SFT:
Y-displacement; (b) bending moment MZ; (c) Z-displacement; and (d) bending moment MY. LC1, Load
(a) Y-displacement; (b) bending moment MZ ; (c) Z-displacement; and (d) bending moment MY . LC1,
Combination 1.
Load Combination 1.
Appl. Sci.Sci.
Appl. 2017 , 7, 1122
2017, 7, 1122 15 15
of of
17 17

To show a clear picture of the dynamic response of the SFT under seismic loadings, the
To show
comparison of amid-span
clear picture of the dynamic
displacements response
and bending of the SFT under
moments forseismic loadings, the
hydrodynamic andcomparison
seismic
of mid-span displacements and bending moments for hydrodynamic and seismic
analysis (LC2) are shown in Figure 12. The hydrodynamic response is very small as compared to analysis (LC2) are
shown
when theinseismic
Figureload
12. The hydrodynamic
is considered. The response
comparison is very small only
is shown as compared to whenscenario
for one seismic the seismic load
(LC2);
is considered. The comparison is shown only for one seismic scenario (LC2); due
due to that, both seismic load combinations (LC1, LC2) produce very close responses in comparison to that, both seismic
toload combinations
hydrodynamic (LC1, LC2)
analysis. produce very close
The hydrodynamic responses
response of theinSFT
comparison
resemblestothehydrodynamic
shape of theanalysis.
wave
Themodel
force hydrodynamic
used, whileresponse of theloading
the seismic SFT resembles
effect thethe shape of
transient the wave
motion force
of the SFT.model used, while
In addition, the
the seismic loading effect the transient motion of the SFT. In addition, the transverse
transverse displacements (Figure 12a) and transverse bending moment (MZ, Figure 12b) are very large displacements
(Figure
and 12a) andaffected
significantly transverse bendingloading
by seismic momentas(M Z , Figure to
compared 12b) are very
vertical large and significantly
displacements (Figure 12c) affected
and
by seismic
vertical loading
bending as compared
moment to vertical
(MY, Figure displacements
12d), and (Figureto12c)
this is attributed theand vertical
small cable bending
stiffness moment
in the
(MY , Figure
transverse 12d), and this is attributed to the small cable stiffness in the transverse direction.
direction.

(a) (b)

(c) (d)
Figure
Figure12.
12.SFT
SFTdynamic responsecomparison
dynamic response comparisonat at mid-span
mid-span (Section
(Section B):Y-displacements;
B): (a) (a) Y-displacements; (b)
(b) bending
bending
moment moment
M ; (c)M Z; (c) Z-displacements; and (d) bending moment MY.
Z-displacements; and (d) bending moment M .
Z Y

5. Conclusions
5. Conclusions
The submerged floating tunnel (SFT) is a new structural solution to waterway crossings. Despite
The submerged floating tunnel (SFT) is a new structural solution to waterway crossings. Despite
some valuable research having been accomplished, no SFT has been constructed in the world up to
some valuable research having been accomplished, no SFT has been constructed in the world up to
now. For the realization of SFT projects, the global performance in extreme environmental conditions
now. For the realization of SFT projects, the global performance in extreme environmental conditions
needs to be properly evaluated. This study evaluates the displacements and internal forces of an SFT
needs to be properly evaluated. This study evaluates the displacements and internal forces of an SFT
subjected to hydrodynamic and seismic loadings. The SFT tunnel was modeled by 3D beam elements;
subjected to hydrodynamic and seismic loadings. The SFT tunnel was modeled by 3D beam elements;
the mooring cables were modeled by truss elements; the ocean waves and currents were modeled by
the mooring cables were modeled by truss elements; the ocean waves and currents were modeled by
Airy wave theory. The SFT response subjected to hydrodynamic loading was evaluated using three
Airy wave theory. The SFT response subjected to hydrodynamic loading was evaluated using three
different mooring cable configurations; the suitable mooring cable configuration was then used for
different mooring cable configurations; the suitable mooring cable configuration was then used for
further evaluation. A comparison of response envelope curves for static, hydrodynamic and seismic
further evaluation. A comparison of response envelope curves for static, hydrodynamic and seismic
loadings was presented to evaluate the most effective structural response of SFT.
loadings was presented to evaluate the most effective structural response of SFT.
Based on the present study, the following conclusions can be drawn:
Based on the present study, the following conclusions can be drawn:
Appl. Sci. 2017, 7, 1122 16 of 17

• The static response can provide a benchmark for decision-making on mooring cable arrangements
and configurations for the preliminary design. However, the static analysis underestimates the
responses in comparison to those of dynamic analysis
• The SFT moored by tension leg mooring cables was least effective, because of lesser horizontal
stiffness, and the SFT moored by such mooring cables undergoes extreme displacements in the
transverse direction as compared to the one moored by inclined mooring cables. The SFT moored
by a combination of tension legs and single inclined mooring cables was effective for moderate
environmental conditions. For severe environmental conditions, the SFT should be moored either
by a combination of tension legs and double inclined mooring cables or only by double inclined
mooring cables.
• For hydrodynamic analysis, the transient motions of SFT were small and decayed quickly, and the
steady-state motions mainly governed the structural response of SFT under the prescribed
damping conditions. In the case of the seismic analysis, the transient motions are more pronounced
as compared to the hydrodynamic response.
• In the hydrodynamic analysis, the vertical response of SFT was very small and mainly attributed
to gravity and hydrodynamic forces; while the transverse response was very large for seismic
analysis. The transverse displacements and internal forces were found to be larger than those
of the vertical direction, which showed less restraining and lesser mooring cable stiffness in the
transverse direction.

Acknowledgments: The Basic Science Research Program through the National Research Foundation of Korea
(NRF), funded by the Ministry of Education, Science and Technology (NRF-2015R1D1A1A09060113), supported
this research. The authors wish to express their gratitude for this financial support.
Author Contributions: Dong-Ho Choi supervised the research. Naik Muhammad contributed the idea, performed
the numerical simulations and wrote the manuscript. Zahid Ullah helped with the simulations and the writing of
the manuscript.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Ahrens, D. Chapter 10 submerged floating tunnels—A concept whose time has arrived. Tunn. Undergr.
Space Technol. 1997, 12, 317–336. [CrossRef]
2. Faggiano, B.; Landolfo, R.; Mazzolani, F. The SFT: An innovative solution for waterway strait crossings.
IABSE Symp. Rep. 2005, 90, 36–42. [CrossRef]
3. Skorpa, L. Innovative Norwegian fjord crossing. How to cross the HØGSJORD, alternative methods.
In Proceedings of the 2nd Congress AIOM (Marine and Offshore Engineering Association), Naples, Italy,
15–17 November 1989.
4. Maeda, N.; Morikawa, M.; Ishikawa, K.; Kakuta, Y. Study on structural characteristics of support systems for
submerged floating tunnel. In Proceedings of the 3rd Symposium on Strait Crossings, Ålesund, Norway,
12–15 June 1994; pp. 579–674.
5. Lu, W.; Ge, F.; Wang, L.; Wu, X.; Hong, Y. On the slack phenomena and snap force in tethers of submerged
floating tunnels under wave conditions. Mar. Struct. 2011, 24, 358–376. [CrossRef]
6. Bruschi, R.; Giardinieri, V.; Marazza, R.; Merletti, T. Submerged Buoyant Anchored Tunnels: Technical Solution
for the Fixed Link across the Strait of Messina Strait Crossings; Balkema: Rotterdam, Dutch, 1990.
7. Faggiano, B.; Landolfo, R.; Mazzolani, F. Design and modelling aspects concerning the submerged floating
tunnels: An application to the Messina strait crossing. In Proceedings of the Fourth Symposium on Strait
Crossings, Bergen, Norway, 2–5 September 2001; pp. 1–5.
8. Mazzolani, F.; Landolfo, R.; Faggiano, B.; Esposto, M.; Perotti, F.; Barbella, G. Structural analyses of the
submerged floating tunnel prototype in Qiandao Lake (PR of China). Adv. Struct. Eng. 2008, 11, 439–454.
[CrossRef]
9. Martinelli, L.; Barbella, G.; Feriani, A. A numerical procedure for simulating the multi-support seismic
response of submerged floating tunnels anchored by cables. Eng. Struct. 2011, 33, 2850–2860. [CrossRef]
 Appl. Sci. 2017, 7, 1122 17 of 17

10. Han, J.S.; Won, B.; Park, W.-S.; Ko, J.H. Transient response analysis by model order reduction of a Mokpo-Jeju
submerged floating tunnel under seismic excitations. Struct. Eng. Mech. 2016, 57, 921–936. [CrossRef]
11. Kunisu, H. Evaluation of wave force acting on submerged floating tunnels. Procedia Eng. 2010, 4, 99–105.
[CrossRef]
12. Seo, S.I.; Mun, H.S.; Lee, J.H.; Kim, J.H. Simplified analysis for estimation of the behavior of a submerged
floating tunnel in waves and experimental verification. Mar. Struct. 2015, 44, 142–158. [CrossRef]
13. Brancaleoni, F.; Castellani, A.; D’Asdia, P. The response of submerged tunnels to their environment.
Eng. Struct. 1989, 11, 47–56. [CrossRef]
14. Okstad, K.M.; Haukas, T.; Remseth, S.; Mathisen, K.M. Fluid-structure interaction simulation of submerged
floating tunnels. In Computational Mechanics New Treds and Applications; Centro Internacional de Métodos
Numéricos en Ingeniería: Barcelona, Spain, 1998; pp. 1–13.
15. Remseth, S.; Leira, B.J.; Okstad, K.M.; Mathisen, K.M.; Haukås, T. Dynamic response and fluid/structure
interaction of submerged floating tunnels. Comput. Struct. 1999, 72, 659–685. [CrossRef]
16. Di Pilato, M.; Perotti, F.; Fogazzi, P. 3D dynamic response of submerged floating tunnels under seismic and
hydrodynamic excitation. Eng. Struct. 2008, 30, 268–281. [CrossRef]
17. Di Pilato, M.; Feriani, A.; Perotti, F. Numerical models for the dynamic response of submerged floating
tunnels under seismic loading. Earthq. Eng. Struct. Dyn. 2008, 37, 1203–1222. [CrossRef]
18. Fogazzi, P.; Perotti, F. The dynamic response of seabed anchored floating tunnels under seismic excitation.
Earthq. Eng. Struct. Dyn. 2000, 29, 273–295. [CrossRef]
19. Li, J.; Li, Y. Analytical solution to the vortex-excited vibration of tether in the submerged floating tunnel.
Am. Soc. Civ. Eng. 2006, 164–169. [CrossRef]
20. Luoa, G.; Chen, J.; Zhou, X. Effects of various factors on the viv-induced fatigue damage in the cable of
submerged floating tunnel. Pol. Marit. Res. 2015, 22, 76–83. [CrossRef]
21. Yiqiang, X.; Chunfeng, C. Vortex-induced dynamic response analysis for the submerged floating tunnel
system under the effect of currents. J. Waterw. Port Coast. Ocean Eng. 2012, 139, 183–189. [CrossRef]
22. Tariverdilo, S.; Mirzapour, J.; Shahmardani, M.; Shabani, R.; Gheyretmand, C. Vibration of submerged
floating tunnels due to moving loads. Appl. Math. Model. 2011, 35, 5413–5425. [CrossRef]
23. Lee, J.H.; Seo, S.I.; Mun, H.S. Seismic behaviors of a floating submerged tunnel with a rectangular
cross-section. Ocean Eng. 2016, 127, 32–47. [CrossRef]
24. Mirzapour, J.; Shahmardani, M.; Tariverdilo, S. Seismic response of submerged floating tunnel under support
excitation. Ships Offshore Struct. 2016, 1–8. [CrossRef]
25. Xiang, Y.; Yang, Y. Spatial dynamic response of submerged floating tunnel under impact load. Mar. Struct.
2017, 53, 20–31. [CrossRef]
26. Chakrabarti, S.K. Hydrodynamics of Offshore Structures; WIT Press: Southampton, UK, 1987.
27. Sarpkaya, T.; Isaacson, M. Mechanics of Wave Forces on Offshore Structures; Van Nostrand Reinhold: New York,
NY, USA, 1981.
28. Veritas, D.N. Free Spanning Pipelines; Recommended Practice DNV-RPF105; Det Norske Veritas Germanischer
Lloyd (DNV GL): Oslo, Norway, 2006. Available online: http://rules.dnvgl.com/docs/pdf/DNV/codes/
docs/2006-02/RP-F105.pdf (accessed on 24 September 2017).
29. Logan, D.L. A First Course in the Finite Element Method; Cengage Learning: Boston, MA, USA, 2011.

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).

View publication stats