POLISH MARITIME RESEARCH 2 (106) 2020 Vol. 27; pp.

48-57
10.2478/pomr-2020-0026

PREDICTION OF VORTEX-INDUCED VIBRATION RESPONSE

OF DEEP SEA TOP-TENSIONED RISER IN SHEARED FLOW
CONSIDERING PARAMETRIC EXCITATIONS
Guanghai Gao
Yunjing Cui
Xingqi Qiu
China University of Petroleum (East China), China

ABSTRACT

It is widely accepted that vortex-induced vibration (VIV) is a major concern in the design of deep sea top-tensioned
risers, especially when the riser is subjected to axial parametric excitations. An improved time domain prediction
model was proposed in this paper. The prediction model was based on classical van der Pol wake oscillator models,
and the impacts of the riser in-line vibration and vessel heave motion were considered. The finite element, Newmark-β
and Newton‒Raphson methods were adopted to solve the coupled nonlinear partial differential equations. The entire
numerical solution process was realised by a self-developed program based on MATLAB. Comparisons between the
numerical calculation and the published experimental test were conducted in this paper. The in-line and cross-flow
VIV responses of a real size top-tensioned riser in linear sheared flow were analysed. The effects of the vessel heave
amplitude and frequency on the riser VIV were also studied. The results show that the vibration displacements of the
riser are larger than the case without vessel heave motion. The vibration modes and frequencies of the riser are also
changed due to the vessel heave motion

Keywords: top-tensioned riser, vortex-induced vibration, wake oscillator model, time-varying axial tension force, sheared flow

INTRODUCTION oscillator model was improved by Balasubramanian and Skop

and Krenk and Nielsen [3-4]. Facchinetti et al. created three
Since the vortex-induced vibrations (VIV) of deep sea expressions for the wake oscillator model. They found that the
top-tensioned risers can result in large amplitude responses acceleration coupling term was the best expression to predict
in both the in-line (IL) and cross-flow (CF) directions, the cylinder VIV response [5].
accurate prediction of the risers’ VIV response has attracted So far, many researchers have focused their attention on
wide attention among researchers. Meanwhile, a riser’s axial the study of the coupled in-line and cross-flow VIV response
tension force will change constantly with time due to the vessel of different riser models [6-11]. However, the axial parametric
heave motion. The time-varying axial tension force will cause excitations have not been considered in these studies. Spanos
transverse vibration of the riser, which is called parametric et al. and Lei et al. established frequency domain analysis
excitation vibration. models to predict the riser’s transverse vibration considering
Wake oscillator models have been widely used to predict the time-dependent axial tension force [12-13]. Considering
riser VIV responses in recent years. Nonlinear dynamic the time-varying axial tension force, time domain analysis
oscillators are adopted to predict the main features of vortex models to predict the riser transverse vibration have also been
shedding in the wake region of the risers. Bishop and Hassan proposed in some works [14-18]. Some studies have investigated
first put forward that the oscillator satisfied the van der Pol the effect of variable axial tension force on the vertical risers’
equation [1]. Hartlen and Currie first established the specific cross-flow VIV when subjected to uniform and sheared flows
expression of the wake oscillator mode [2]. Then the wake [19-21].

48 POLISH MARITIME RESEARCH, No 2/2020

 From the above review, it can be seen that a number of In Eq. (1):
works in the literature have analysed the effect of the time- EI ‒ the riser bending stiffness;
varying axial tension force on the transverse vibration response 7 ( ] , W ) ‒ the effective axial tension force of the riser;
and the VIV response which are induced by the sea current Z ‒ the wet weight of riser per unit length;
and/or waves. However, there are few studies on the coupled P ‒ the equivalent mass of riser per unit length;
IL and CF VIV of deep sea top-tensioned risers in sheared flow F ‒ the damping coefficient;
considering time-varying axial tension force. In this paper, an I x ( ] , W ) , I y ( ] , W ) ‒ the IL and CF fluctuating hydrodynamic
improved three-dimensional time domain prediction model forces acting on the riser.
was proposed. The effects of the riser’s in-line vibration and The vessel heave motion is modelled by a harmonic
the vessel heave motion were considered in the prediction fluctuation in the present study. The time-varying effective
model. The coupled IL and CF VIV response of a real size axial tension force of the riser can be written as [18, 24-25]:
top-tensioned riser was studied in this paper. The impact of H
the vessel heave amplitude and frequency on the riser’s VIV 7 ( ], W ) I top Z+  ³ Zd]  .$ sin(: W ) (2)
]
response has also been discussed.
In Eq. (2):
I top ‒ the static top tension force coefficient;
MODEL DESCRIPTION H ‒ the water depth;
‒ the equivalent spring stiffness of the heave compensator
( . Z+ Dc ), Dc is the critical amplitude associated with
STRUCTURE MODEL AND EQUATIONS OF MOTION the heave compensator (usually set as 10 m);
A , : ‒ the heave motion amplitude and frequency of the
A two-end pinned Euler‒Bernoulli beam model is used to vessel.
simulate deep sea top-tensioned risers. A Cartesian coordinate The wet weight and mass of riser per unit length are:
system is defined, in which the x-axis is parallel to the direction
of seawater flow and the ]-axis is parallel to the axial direction (3)
of the un-deformed riser. Fig. 1 shows the schematic model of
a deep sea top-tensioned riser.
In Eq. (3):
J ‒ the gravity acceleration;
U r , Uf , U w ‒ the density of the riser, internal fluid and sea
water;
D , d ‒ the riser’s outer and inner diameters;
Ca ‒ the added mass coefficient.
The damping coefficients are as follows [9]:
­°F Fs  Ff
® (4)
°̄Ff J: f U w '
2

In Eq. (4):
Fs ‒the structure damping coefficient;
Ff ‒ the fluid damping coefficient;
J ‒ the related coefficient
: f ‒ the vortex shedding angular frequency;
CD ‒ the mean drag force coefficient;
Fig. 1. Schematic model of deep sea top tensioned riser
St ‒ the Strouhal number.
Considering the effect of the riser’s in-line vibration,
the modified vortex shedding frequency can be expressed
The equations of the riser IL and CF vibrations can be as follows:
written as [22-23]:
(5)
­ w 4 [( ] , W ) w 2 [( ] , W ) w[( ] , W )
° (,  7 ( ], W ) Z
° w] 4
w ] 2
w] In Eq. (5):
° w 2 [( ] , W ) w[( ] , W ) X ( ] ) ‒ the sea current velocity;
° P F I x ( ], W )
° wt 2 wt w[( ] , W ) wW ‒ the riser’s IL vibration velocity.
® Considering the effect of the riser’s in-line vibration, the
° (, w \ ( ] , W )  7 ( ] , W ) w \ ( ] , W )  Z w\ ( ] , W ) (1)
4 2

° w] 4
w] 2 w] fluctuating hydrodynamic forces can be expressed as follows:
°
° w \( ], W )
2
w\ ( ] , W )
° P wt 2  F wt I y ( ], W )
¯

POLISH MARITIME RESEARCH, No 2/2020 49

 ­ I x ( ] , W ) 0.5&D U w 'X ( ] ) 2  0.5&D U w '(X ( ] )  w[( ] , W ) wW ) 2 NUMERICAL SOLUTION METHOD
°
°° -0.5&L U w ' (X ( ] )  w[( ] , W ) wW )2 w\ ( ] , W ) wW
® (6) The coupled form of Eq. (1) with Eq. (7) is nonlinear
° I y ( ] , W ) 0.5&L U w '(X ( ] )  w[( ] , W ) wW )
2
partial differential equations with variable coefficients. Some
°
°̄ +0.5&D U w '(X ( ] )  w[( ] , W ) wW ) 2 w\ ( ] , W ) wW coefficients in the coupled equations change continuously along
In Eq. (6): the spatial position, and some coefficients change continuously
CD ‒ the IL VIV drag force coefficient; with time. The coupled equations are very complicated and
CL ‒ the CF VIV lift force coefficient; the analytical solution cannot be obtained. So, numerical
w\ ( ] , W ) wW ‒ the riser’s CF vibration velocity. solution methods are used in the present study to solve the
coupled equations.
WAKE OSCILLATOR MODEL AND EQUATIONS OF The Galerkin-type finite element method is used to
MOTION conduct the spatial discretisation of the coupled equations.
Based on the parameters of the riser and seawater, as well
According to the wake oscillator models [12, 16], CD and as the boundary and initial conditions, the Hermite cubic
CL can be simulated by the time-varying fluid vortex variables interpolation functions are used to obtain the system’s flexural
p and q : stiffness matrix, geometric stiffness matrix, mass matrix, and
­ $x w 2 [( ] , W ) damping matrix. Due to the uncertainty of the riser’s damping
p  2H x : f ( p 2  1) p  4: f2 p
°° 
D wt 2 (7) coefficient, the Rayleigh damping theory is adopted in this
®
°q  H : (q 2  1)q  : 2 q Ay w \ ( ] , W )
2
paper to obtain the structural damping matrix. It is assumed
¯° D wt 2
y f f
that the vibration shape functions of wake oscillators are
In Eq. (7): trigonometric functions. Then the wake oscillators are used
p , q ‒ the IL and CF wake fluid variable parameter to obtain the damping matrix of the fluid and the dynamic
( p 2 CD CD0 , q 2 CL CL0 ); load matrix at the initial moment.
CD0 , CL0 ‒ the IL VIV drag force coefficient and CF VIV lift The Newmark-β method is used to solve the coupled
force coefficient of the fixed riser; equations in the time domain. In each time step, the Newton‒
Ax , Ay , H x , H y ‒ the non-dimensional coefficients. Raphson iterative method is used to solve the nonlinear
equations. The system’s geometric stiffness matrix, mass
BOUNDARY AND INITIAL CONDITIONS matrix, damping matrix, dynamic load matrix and natural
vibration frequency are all constantly updated at each time
The boundary conditions are expressed as follows: step. The entire numerical solution process is realised by a
­ x(0, t ) y (0, t ) 0 ; x( H , t ) y ( H , t ) 0 self-developed program based on MATLAB.
° 2
®w x(0, t ) w 2 y (0, t ) w 2 x( H , t ) w 2 y ( H , t ) (8)
° 0; 0 VALIDATION OF SIMULATION
¯ w] 2
w] 2
w] 2 w] 2

At the initial moment, the vibration displacement and In order to test the validity of the prediction models,
velocity of the riser are assumed as zero. The values of the fluid comparisons with experimental results are carried out in this
variables p and q are set as 2.0, and their first derivatives with paper. The experiment was carried out at the Deep Sea Basin
respect to time are set to zero. of the National Maritime Research Institute in Japan [26]. It
should be pointed out that the water depth of the fluid flow
MODEL PARAMETERS in the experiment ranges from 0.4 m to 5 m, while the water
depth of the fluid flow in the numerical calculation ranges
According to [12, 13, 15, 16, 17], the values of the empirical from 0 to 5 m. The flow is stable and the flow velocity is set
coefficients are given in Table 1. to 0.3 m/s in the water depth of 0 to 0.4 m in the numerical
Tab. 1. Values of the empirical coefficients calculation. When the dynamic variation of the top tension
is considered, two cases are calculated using the prediction
model proposed in this paper. The added-tension/self-weight in
Empirical coefficients Values the experiment is varied from 0.58 to 1.08 in a certain period.
St 0.17 It should be pointed out that according to Eq. (2) and the
CD0 equivalent principle, the top tension coefficient ( Itop) is set to
1.2
1.83 and the .$ is set to 0.257top in the numerical calculation.
CD0 0.1 Fig. 2 illustrates the experimental and numerical results
CL0 0.3 of the model’s cross-flow maximum VIV amplitude at different
top tension variation periods. It can be seen that the maximum
Ax Ay 12 VIV amplitudes calculated in this paper are a little larger
,
Hx ,Hy 0.3 than the experimental results, especially in the upper part of
the riser model. Comparing the experimental and numerical
results, it can be seen that the vibration mode of the upper part

50 POLISH MARITIME RESEARCH, No 2/2020

 Tab. 2. Main parameters of the riser and vessel
of the riser is comparatively consistent, but the vibration mode
of the lower part of the riser is somewhat different. However,
in general, the relative errors between the numerical and Parameters Values
experimental results are acceptable in engineering applications. Water depth (m) 2000
Overall, the prediction model proposed in this paper is valid
in predicting the main features of the VIV response of large Outer diameter (m) 0.5334
aspect (length/diameter) ratio top-tensioned risers. Inner diameter (m) 0.4826
2
Elastic modulus (N/m ) 2.1×1011
Riser density (kg/m3) 7850
Internal fluid density (kg/m3) 1200
Seawater density (kg/m3) 1030
Length/diameter ratio 3750
Static top tension coefficient 1.1~1.6
Sea surface current velocity (m/s) 0.5~1.5
Sea bottom current velocity (m/s) 0.2
Structural damping coefficient (%) 0.3
Vessel heave motion amplitude (m) 0.5~2.0
Vessel heave motion frequency (rad/s) 0.4~1.6

(a) Period=2.27s
NATURAL FREQUENCIES OF THE RISER

Some key features of the riser’s VIV response can be

illustrated by inspecting the transverse vibration natural
frequencies of the riser. Hence, the subspace iteration method
is first applied to calculate the riser’s natural frequencies. In
this section, the time-varying dynamic component of axial
tension is not considered. Fig. 3 shows the natural frequencies
of the riser with the top tension coefficient of 1.6. As shown in
Fig. 3, the natural frequencies of the riser are low and dense,
which is a unique feature of the transverse vibration of deep
sea top-tensioned risers.

(b) Period=4.47s
Fig. 2. VIV amplitude of the riser model

CASE STUDY

In the present study, the VIV response of a real size top- Fig. 3. Natural frequencies of the riser
tensioned riser is studied. It is assumed that the current is
a linear sheared flow. The main parameters of the riser and
vessel are given in Table 2.

POLISH MARITIME RESEARCH, No 2/2020 51

DYNAMIC RESPONSE OF THE RISER

According to the element precision analysis, the riser is

divided equally into 400 elements in the calculation. According
to the maximum vortex shedding period and the calculation
accuracy of the Newmark-ȕ method, the interval time in the
calculation is 0.01 s, and the total calculation time is 500 s.
The top tension force coefficient is chosen to be 1.6 and the sea
surface current velocity is chosen to be 1.5 m/s in the present
study. In order to study the effect of the parametric excitation
on the riser’s VIV response, two typical cases are calculated
and they are specified as follows. Case A: A=2.0 m, ȍ=1.6
rad/s; Case B: A=0 m, ȍ=0 rad/s. It is conformed such that the
lock-in and parametric resonance in both cases does not occur.
Fig. 4 shows the root-mean square (RMS) of IL and CF
displacements of the riser. It should be noted that the RMS of
IL displacements of the riser were obtained after removing the
mean displacement. It can be seen from Fig. 4 that whether or Fig. 4. RMS of IL and CF VIV displacements of the riser
not the vessel heave motion is taken into account, the riser’s
VIV modes are all a mixture of travelling and standing waves.
Meanwhile, the dominant modes of the riser in the in-line According to the calculation results, when the heave
direction are larger than the cross-flow direction in both cases. motion of the vessel is not considered, the IL and CF maximum
In each case, the riser’s VIV responses are all characterised by VIV amplitudes of the riser are 0.0754 m and 0.5184 m,
high order multi-mode vibrations in the non-locking region. respectively. In addition, when the heave motion of the vessel
As shown in Fig. 4, when considering the heave motion of is considered, the IL and CF maximum VIV amplitudes of the
the vessel, both the IL and CF vibration modes of the riser riser are 0.0892 m and 0.5615 m, respectively. The IL and CF
have changed remarkably. When the parametric excitation is maximum VIV amplitudes of the riser are increased by 18.3%
considered, the fluctuation of the riser VIV is more intense. It and 8.31%, respectively. It can be seen that the vessel heave
can be seen that the VIV mode of the riser in linear sheared flow motion has a greater effect on the riser IL VIV displacements.
is very complicated when considering parametric excitations. Overall, the heave motion of the vessel makes the riser VIV
more severe.
Fig. 5 shows the amplitude spectrum of the riser vibration
displacements at different water depths. It can be seen that,
whether or not the vessel heave motion is considered, the bulks
of the IL response are all between 0.021 Hz and 1.178 Hz, and
the bulks of the CF response are all between 0.021 Hz and 0.594
Hz. According to Fig. 3, the 1st, 26th and 44th modes’ natural
frequencies are 0.024 Hz, 0.597 Hz and 1.177 Hz, respectively.
It can be seen that the IL VIV participating modes reach the
44th, and the CF VIV participating modes reach the 26th. At
the same time, the riser IL and CF VIV response are all shown
to be a high order multi-mode response at each position. This
is because even at the same position, the frequency of the
vortex shedding is constantly changing due to the motion
of the riser in the in-line direction, which causes the riser’s
vibration frequency to be constantly changing. As can be seen
from Fig. 5, the highest peak frequencies at different positions
of the riser are different, and the participating modes in the
vibration are also different. This is because the sea current
velocity and the axial tension force at different positions are all
different, which makes the vortex shedding frequency different
in different positions, thereby causing the above phenomenon.
Meanwhile, the richness in frequency content indicates an
irregular character of the VIV response of the riser. The IL
response is much more irregular than the CF response, as can
be seen from the width of the spectrum and the high number of
participating modes. In addition, it can be seen from Fig. 5 that

52 POLISH MARITIME RESEARCH, No 2/2020

when the heave motion of the vessel is considered, the IL and the riser IL and CF RMS displacements with different heave
CF VIV dominant frequencies and participating frequencies amplitudes.
at different positions of the riser are all changed. The heave
motion of the vessel has different effects on the riser IL and
CF VIV frequencies at different positions of the riser. It can be
seen that the impact of the vessel heave motion on the IL and
CF VIV response of the riser is very complicated.

Fig. 6. RMS of VIV displacements of the riser with different heave amplitudes

As is shown in Fig. 6, when the vessel heave frequency is

1.6 rad/s, the effect of the variation of the vessel heave amplitude
Fig. 5. Amplitude spectrum of the riser VIV displacements
on the riser’s VIV mode is complicated. The riser IL and CF
VIV modes are different at different heave amplitudes. It can
also be seen that the variation of the heave amplitude has a
EFFECT OF VESSEL HEAVE AMPLITUDE more significant effect on the riser’s IL VIV mode than the
CF. According to Eq. (2), the maximum axial tension force
According to the environmental condition in the South of the riser increases gradually with the increase of the heave
China Sea, the heave period of the vessel is commonly between amplitude, which causes the natural frequencies of the riser
4 s and 15 s. So, the frequency of the axial parametric excitation to change constantly. Therefore, at the same static top tension
falls into the range from 0.4 rad/s to 1.6 rad/s. The heave force and sea current velocity, the riser VIV mode changes
amplitude of the vessel is usually between 0.5 m and 2.0 m. gradually with the variation of the vessel heave amplitude.
In this section, four cases are studied to investigate how the According to the calculation results, the IL maximum
vessel heave amplitude affects the riser’s VIV response. The VIV amplitudes of the riser in the four cases are 0.0744 m,
vessel heave frequency is chosen to be 1.6 rad/s. Fig. 6 shows 0.0763 m, 0.0805 m and 0.0892 m, respectively. Meanwhile,
the CF maximum VIV amplitudes of the riser in the four cases

POLISH MARITIME RESEARCH, No 2/2020 53

are 0.5517 m, 0.5539 m, 0.5573 m and 0.5615 m, respectively. According to the calculation results, the IL maximum
It can be seen that with the increase of the heave amplitude, VIV amplitudes of the riser in the four cases are 0.0813 m,
the IL and CF maximum VIV amplitudes of the riser increase 0.0812 m, 0.0826 m and 0.0892 m, respectively. Meanwhile, the
continuously. CF maximum VIV amplitudes of the riser in the four cases are
The displacements of the riser midpoint under different 0.5541 m, 0.5543 m, 0.5597 m and 0.5615 m, respectively. It can
heave amplitudes are analysed by spectrum analysis, and the be seen that when the heave frequency changes from 0.4 rad/s
spectrum curves are shown in Fig. 7. As shown in Fig. 7, the to 0.8 rad/s, the IL and CF maximum VIV amplitudes of the
dominant frequencies and the participating frequencies of riser basically remain unchanged. When the heave frequency
the riser VIV change continuously when the heave amplitude is large, the IL and CF maximum VIV amplitudes of the riser
changes. It can be concluded that the variation of the heave increase gradually with the increase of the heave frequency.
amplitude has a great influence on the riser VIV frequency. The displacements of the riser midpoint under different
heave amplitudes are analysed by spectrum analysis, and the
spectrum curves are shown in Fig. 9. It can be seen from Fig.
9 that the variation of the vessel heave frequency does not
change the components of the riser VIV frequency. However,
the proportions of the VIV dominant frequency of the riser
increase gradually with the increase of the vessel heave
frequency. It can be concluded that the dominant frequency
of the riser VIV is strengthened by the larger vessel heave
frequency.

Fig. 7. Amplitude spectrum of the riser’s VIV displacements with different

heave amplitudes

EFFECT OF VESSEL HEAVE FREQUENCY

In this section, four cases are studied to investigate how

the vessel heave frequency affects the riser VIV response. The
heave amplitude of the vessel is chosen to be 2.0 m. Fig. 8 shows
the riser’s IL and CF RMS displacements with different heave
frequencies. As shown in Fig. 8, when the heave frequency is
small, the variation of the heave frequency has little effect on
the riser’s IL and CF VIV modes. However, when the heave
frequency is large, the variation of the heave frequency has a
relatively large influence on the riser’s VIV mode, especially Fig. 8. RMS of VIV displacements of the riser with different heave frequencies
on its IL VIV mode.

54 POLISH MARITIME RESEARCH, No 2/2020

 prediction model is reasonable to predict some main features
of the riser’s VIV response. It is assumed that the sea current
is linear sheared flow, and the attention is focused on the study
of the riser’s VIV response when the lock-in and parametric
resonance phenomenon does not occur. The time-domain
simulation results show that the periodic time-varying axial
tension force has a large effect on the IL and CF VIV response
of the riser. At the same static top tension force and current
velocity, when the periodic time-varying tension is considered,
the VIV displacements of the riser increase and the VIV’s
dominant frequency and the participating frequencies of the
riser have changed. Meanwhile, the riser’s VIV mode has
also changed when the periodic time-varying axial tension
is considered. When the lock-in and parametric resonance
phenomenon does not occur, the impact of the vessel heave
amplitude and frequency on the riser’s VIV response is
complicated. The impacts of the vessel heave amplitude and
frequency on the IL and CF VIV responses of the riser are
different. So, it requires a detailed analysis of the impact at
different operating conditions.
It is well known that the lock-in phenomenon easily occurs
when the riser is subjected to VIV. The lock-in phenomenon is
very complicated when the riser is in sheared flow. Meanwhile,
parametric resonance of the riser is very likely to occur when
the heave motion of the vessel is considered. To investigate
the coupling between the VIV and parametric excitation will
therefore be one of our research topics when the lock-in and
parametric resonance occur simultaneously.
Fig. 9. Amplitude spectrum of the riser’s VIV displacements with different
heave frequencies ACKNOWLEDGEMENTS

From the above analysis, it can be concluded that when the The authors gratefully acknowledge the financial
lock-in and parametric resonance of the riser does not occur, support from the National High Technology Research and
the effect of the variation of the vessel heave amplitude and Development Program of China (863 Program, Grant No.
frequency on the VIV response of the riser is complicated. The 2013AA092602), and the Fundamental Research Funds for
impact of the variation of the heave amplitude and frequency the Central Universities (Grant No. 16CX06015A).
on the IL and CF VIV response of the riser is different. It
should be pointed out that the above analysis is based on the
same static top tension force and current velocity. However, REFERENCES
the impact of the variation of the vessel heave amplitude and
frequency on the riser’s VIV response is different at different 1. Bishop R .E. D., Hassan A. Y. (1964): The lift and drag
static top tension forces and current velocities. Hence, in forces on a circular cylinder oscillating in a flowing fluid.
practical operation, the relevant parameters should be adjusted Proceedings of the Royal Society of London, Series A:
in time to avoid serious VIV, which will lead to the damage Mathematical and Physical Sciences, Vol. 277(1368), 51-75.
of the riser.
2. Hartlen R. T., Currie I. G. (1970): Lift-oscillator model
of vortex-induced vibration. Journal of the Engineering
CONCLUSIONS Mechanics Division, Vol. 96(5), 577-591.

An improved three-dimensional time domain prediction 3. Balasubramanian S., Skop R. A. (1996): A nonlinear
model of long flexible risers has been proposed in the present oscillator model for vortex shedding from cylinders and
study to predict the IL and CF VIV response of top-tensioned cones in uniform and shear flows. Journal of Fluids and
risers. The impact of the riser’s in-line vibration is considered in Structures, Vol. 10(3), 197-214.
the prediction model. The periodic time-varying axial tension
force caused by the vessel heave motion is also considered in 4. Krenk S., Nielsen S. R. K. (1999): Energy balanced double
the prediction model. Comparisons between the numerical oscillator model for vortex-induced vibrations. Journal of
calculation and experimental test results have shown that the Engineering Mechanics, Vol. 125(3), 263-271.

POLISH MARITIME RESEARCH, No 2/2020 55

5. Facchinetti M. L., Langre E. D., Biolley F. (2004): Coupling 17. Wang Y. B., Gao D. L., Fang J. (2015): Coupled dynamic
of structure and wake oscillators in vortex-induced analysis of deep water drilling riser under combined forcing
vibrations. Journal of Fluids and Structures, Vol. 19(2), and parametric excitation. Journal of Natural Gas Science
123-140. and Engineering, Vol. 27(3), 1739-1747.

6. Furnes G. K., Sørensen K. (2007): Flow induced vibrations 18. Lei S., Zheng X. Y., Kennedy D. (2017): Dynamic response
modeled by coupled non-linear oscillators. In: Proceedings of a deepwater riser subjected to combined axial and
of the 17th International Offshore and Polar Engineering transverse excitation by the nonlinear coupled model.
Conference, Lisbon, Portugal, 2007. International Journal of Non-Linear Mechanics, Vol. 97,
68-77.
7. Ge F., Long X., Wang L., Hong Y. S. (2009): Flow-induced
vibrations of long circular cylinders modeled by coupled 19. da Silveira L. M. Y., Martins C. D. A., Cunha L. D., Pesce C.
nonlinear oscillators. Science in China Series G: Physics, P. (2007): An investigation on the effect of tension variation
Mechanics and Astronomy, Vol. 52(7), 1086-1093. on VIV of risers. In: Proceedings of 26th International
Conference on Offshore Mechanics and Arctic Engineering,
8. Wu X. D., Ge F., Hong, Y. S. (2010): Effect of travelling California, USA, 2007.
wave on vortex-induced vibrations of submerged floating
tunnel tethers. Procedia Engineering, Vol. 4(6), 153-160. 20. Chen W. M., Li M., Guo S. X., Gan K. (2014): Dynamic
analysis of coupling between floating top-end heave and
9. Srinil N., Zanganeh H. (2012): Modelling of coupled riser’s vortex-induced vibration by using finite element
cross-flow /in-line vortex-induced vibrations using double simulations. Applied Ocean Research, Vol. 48, 1-9.
Duffing and van der Pol oscillators. Ocean Engineering,
Vol. 53(3), 83-97. 21. Thorsen M. J., Sævik S. (2017): Vortex-induced vibrations
of a vertical riser with time-varying tension. Procedia
10. Gu J. J., An C., Levi C., Su J. (2012): Prediction of vortex- Engineering, Vol. 199, 1326-1331.
induced vibration of long flexible cylinders modeled by a
coupled nonlinear oscillator: integral transform solution. 22. Burke B. G. (1974): An analysis of marine risers for deep
Journal of Hydrodynamics, Series B, Vol. 24(6), 888-898. water. Journal of Petroleum Technology, Vol. 26(4), 455-465.

11. Xue H. X., Wang K. P., Tang W. Y. (2015): A practical 23. Wang Y. B., Gao D. L., Fang J. (2014): Static analysis of
approach to predicting cross-flow and in-line VIV response deep-water marine riser subjected to both axial and lateral
for deepwater risers. Applied Ocean Research, Vol. 52, forces in its installation. Journal of Natural Gas Science
92-101. and Engineering, Vol. 19(7), 84-90.

12. Spanos P. D., Tein W. Y., Ghanem R. (1990): Frequency 24. Chucheepsakul S., Monprapussorn T., Huang T. (2003):
domain analysis of marine risers with time dependent Large strain formulations of extensible flexible marine
tension. Applied Ocean Research, Vol. 12(4), 200-210. pipes transporting fluid. Journal of Fluids and Structures,
Vol. 17(2), 185-224.
13. Lei S., Zhang W. S., Lin J. H., Yue Q. J., Kennedy D.,
Williams, F. W. (2014): Frequency domain response of a 25. Kuiper G. L., Brugmans J., Metrikine A. V. (2008):
parametrically excited riser under random wave forces. Destabilization of deep-water risers by a heaving platform.
Journal of Sound and Vibration, Vol. 333(2), 485-498. Journal of Sound and Vibration, Vol. 310, 541-557.

14. Patel M., Park H. I. (1995): Combined axial and lateral 26. Fujiwara T., Uto S., Kanada S. (2011): An experimental
responses of tensioned buoyant platform tethers. study of the effects that change the vibration mode of riser
Engineering Structures, Vol. 17(10), 687-695. VIV. In: Proceedings of 30th International Conference
on Ocean, Offshore and Arctic Engineering, Rotterdam,
15. Park H. I., Jung D. H. (2002): A finite element method Netherlands, 2011.
for dynamic analysis of long slender marine structures
under combined parametric and forcing excitations. Ocean
Engineering, Vol. 29(11), 1313-1325.

16. Chatjigeorgiou I. K., Ioannis K., Mavrakos S. A. (2005):

Nonlinear resonances of parametrically excited risers ‒
numerical and analytic investigation for Ω=2ω1. Computers
& Structures, Vol. 83(8-9), 560-573.

56 POLISH MARITIME RESEARCH, No 2/2020

 CONTACT WITH THE AUTHORS

Guanghai Gao
e-mail: [email protected]

Yunjing Cui
e-mail: [email protected]

Xingqi Qiu
e-mail: [email protected]

China University of Petroleum (East China)

No. 66 , West Changjiang Road
Huangdao District
266580 Qingdao
CHINA

POLISH MARITIME RESEARCH, No 2/2020 57