10th International Congress on Civil Engineering, 5-7 May 2015

University of Tabriz, Tabriz, Iran

Numerical Study on the Effect of Mooring Line Stiffness on

Hydrodynamic Performance of Pontoon-Type Floating Breakwater

Yashar Rafati1, Masoud Montazeri Namin2

1- M.Sc. Student of Marine Structures, University of Tehran, yashar_rafati90@ut.ac.ir
2- Assistant Professor, University of Tehran, mnamin@ut.ac.ir

Abstract
In this study a 2D numerical wave tank is built to simulate the wave-body interactions based
Reynolds-averaged Navier-Stokes equations. The equations are discretized based on the finite volume
method. In order to simulate the turbulence of flow in the wave tank the k   model is used. The
numerical model has been provided based on the open source C++ CFD toolbox named open field
operation and manipulation (OpenFOAM). For applying the mooring line forces to the floating
structure, the elastic catenary equations have been employed. In order to verify the model, calculated
results of simulation are compared with the experimental data. Moreover, the mooring lines with
different values of stiffness have been examined in different wave periods and the transmission
coefficient of the associated data is illustrated.
Keywords: floating breakwater, mooring lines, transmission coefficient, motion response.

1. INTRODUCTION

In the recent decades floating breakwaters (FBs) have been in great demand as an alternative to traditional
bottom-founded breakwaters in small marinas and recreational harbours in order to provide a sheltered area.
They usually cost less than bottom-founded breakwaters and can be built in areas that the bottom foundation
is poor or the areas in which the water depth is in a range that construction of the traditional breakwaters is
not economical. They also can be conveniently removed and rearranged into a new layout [1]. In spite of
their benefits, FBs are not able to fully attenuate the wave energy and a portion of the energy is always
transmitted. The most determinant parameter in evaluating the hydrodynamic performance of FB is the
transmission coefficient, defined by Ct  H t / H i , where H i is the incident wave height and H t represents
the transmitted wave height. Among different layouts introduced for FBs, pontoon type is more common as a
result of good wave attenuation and simple layout [2]. One of the effective factors in the performance of FB
is the characteristics of mooring lines. In this paper a 2D numerical wave tank is built based Reynolds-
averaged Navier-Stokes equations. The equations are discretized based on finite volume method. In order to
simulate the turbulence of flow in the wave tank the k   model is used. The numerical model has been
provided based on the open source C++ CFD toolbox named open field operation and manipulation
(OpenFOAM). For applying the mooring line forces on the floating body, the elastic catenary equations have
been employed [3]. In order to verify the model, the calculated results of simulation are compared with the
experimental data [2]. Moreover, the mooring lines with different values of stiffness have been examined in
different wave periods and the transmission coefficient and the three modes of motion including sway, heave,
and roll of corresponded data are illustrated. The responses are presented in terms of response amplitude
operator, RAO (response due to unite wave amplitude).

2. GOVERNING EQUATIONS

Reynolds-averaged Navier-Stokes equations are the governing equations, are as follows [4]:

1
 10th International Congress on Civil Engineering, 5-7 May 2015
University of Tabriz, Tabriz, Iran


(u j )  0, (1)
x j
  1 p 1  u j ui
(ui )  (u j ui )    [  eff (  )]  Di ui  g i (2)
t x j  xi  x j xi x j
Where x j ( j  1,2) represents coordinate component, u j is the fluid velocity, p is the pressure,  is the
water density, g is the acceleration of gravity, Di is the damping coefficient. eff     f ,  is the
molecular viscosity,  f is the turbulent eddy viscosity,
 f  C k /  , k is the turbulent kinetic energy and  is the turbulent energy dissipation rate. The k  
2

two-equation model is adopted to estimate the turbulence, as follows [4]:

   k
( k )  ( u j k )  [ k  eff ]  Pk   , (3)
t x j x j x j
     2
(  )  ( u j  )  [  eff ]  C1* Pk   f C2 (4)
t x j x j x j k k
Where
ui u j ui *  (1   / 0 )
Pk   f (  ) , C1  C1  ,
x j xi x j 1   3
k 1 u u
  (2Sij Sij ) 0.5 , Sij  ( i  j ).
 2 x j xi
The values of constants are shown in table 1.

Table 1- Values of constants in turbulent model

Constant C k  C1 C 2 0 
Value 0.0845 1.39 1.39 1.42 1.68 4.38 0.012

In order to capture the water-air free surface, volume of fluid (VOF) Eulerian method is adopted [5]. The
equation for the volume fraction is as follows:
 
 (u j )  0 (5)
t x j
Where  is the volume fraction of water and 1   represents the fraction of air. The volume fraction of
each phase is used to get mixture density and viscosity
  w  (1 -  )  a , (6)
  w  (1   ) a ( 7)
Where  w and  a represent the density of water and air, respectively.  w and  a are molecule viscosity
coefficient of water and air respectively.
In order to analyze the mooring cables, the elastic catenary equations [3] are solved using Newton-Raphson
method. The equations are as follows:
HL0 HL0 V V W
l  {sinh 1 ( )  sinh 1 ( )}, (8)
EA W H H
WL0 V 1 HL0 V V  W 2 1/ 2
h (  ) [{1  ( ) 2 }1/ 2  {1  ( )} ] (9)
EA W 2 W H H
Where E is Young’s modulus, A is cross section area of the cable, L0 is the unstrained length of the cable,
W is the submerged weight of the cable, H and V are the horizontal and vertical components of the reaction
forces respectively, l and h are the value of the horizontal and vertical components of the vector from
anchor to attachment point.

2
 10th International Congress on Civil Engineering, 5-7 May 2015
University of Tabriz, Tabriz, Iran

In order to estimate the transmission coefficient of FB, three gauges method is used [6].

3. NUMERICAL MODEL

The entire computational domain is shown in Fig.1. There are five types of boundary related to the equations
including inlet, outlet, bottom, floating structure and atmosphere. Two damping zones are placed at the inlet
and the outlet to avoid rereflection at the inlet and absorb transmitted wave energy and avoid reflection at the
outlet. Two mooring lines are attached to the floating structure to anchor the structure to the bottom. In order
to measure the water surface elevations, three gauges are placed on the onshore side of the floating structure.

Figure 1. Sketch of the FB with mooring lines in the wave tank

4. MODEL VALIDATION

In order to validate the numerical model, an experimental model [2] is simulated and a comparison of the
experimental results and numerical prediction of transmission coefficient are shown in Figure 2.

Figure 2. Comparison of transmission coefficient

It can be seen that numerical estimation of transmission coefficient agrees well with the experimental results.

5. ANALYSIS AND RESULTS

A floating breakwater with dimensions (B=1.6m, Dr=0.4m, D=0.7m) is placed in water depth d=2m in a
wave tank. The length of the wave tank is 30m and all the numerical domain has a thickness of 0.1m in the z


3
 10th International Congress on Civil Engineering, 5-7 May 2015
University of Tabriz, Tabriz, Iran

(transverse) direction. Six wave frequencies are examined where non-dimensional normalized frequency
 2 B / 2 g varies from 0.25 to 1.5, the wave height is chosen as H=0.2m for all cases.
Mooring line inclination h / l is chosen as 1/2, and the length of mooring line is chosen as 3.62m. The
stiffness per unit length of the mooring lines (EA) is chosen in relationship with the weight of the floating
structure (Ws), as EA/Ws varies from 500 to 3000. The weight per unit length of the mooring lines is chosen
based on IWRC1 characteristics [7].
The response amplitude operators in sway, heave, and roll are defined as follows:

Sway RAO: sway amplitude/wave amplitude;

Heave RAO: heave amplitude/wave amplitude;
Roll RAO: roll amplitude/wave amplitude.

Natural modes of oscillations in 2D are shown in Figure 3.

Figure 3. Sketch of the oscillation modes in 2D

Figure 4 depicts the trend of transmission coefficient with normalized frequency for six different values of
mooring line stiffness.

Figure 4. Comparison of transmission coefficient for different values of mooring lines stiffness

It can be seen that trends of transmission coefficient with normalized frequency is almost the same for
breakwaters with different values of mooring line stiffness. For all values of stiffness except EA/Ws=3000,
transmission coefficient reaches the maximum peak at  2 B / 2 g  0.5 . And then decreases with non-
dimensional normalized frequency to a minimum point at  2 B / 2 g  1.75 , and then increases slightly.

Figures 5, 6, and 7 depict the trends of sway, heave, and roll RAO with normalized frequency for six
different values of mooring line stiffness, respectively.

4
 10th International Congress on Civil Engineering, 5-7 May 2015
University of Tabriz, Tabriz, Iran

Figure 5. Variation of Sway RAO for different values of mooring lines stiffness

Figure 6. Variation of Heave RAO for different values of mooring lines stiffness

Figure 7. Variation of Roll RAO for different values of mooring lines stiffness

5
 10th International Congress on Civil Engineering, 5-7 May 2015
University of Tabriz, Tabriz, Iran

It can be seen that for waves with longer periods (value of  2 B / 2 g less than 0.5), the values of sway,
heave, and roll RAO are almost the same. However the average value of RAO for the stiffness of
EA/Ws=2500 is lower in comparison with other values of stiffness.

6. CONCLUSIONS

In this paper, wave-body interactions for a floating breakwater with different values of mooring line stiffness
under waves with different periods are simulated in a 2D numerical wave tank. The tank is based on RANS
equations and k   model. The simulated results are compared with experimental data. The numerical
results are shown to be in good agreement with experimental results. Then hydrodynamic performance of a
floating breakwater with different values of mooring line stiffness is evaluated through calculating the
transmission coefficient and response amplitude motions of corresponded data.

The salient conclusions drawn from the present study are detailed below.

1. Trends of transmission coefficient with normalized frequency are nearly the same for breakwaters
with different values of mooring line stiffness, with transmission coefficient reaching the maximum
peak at  2 B / 2 g  0.5 and the minimum point at  2 B / 2 g  1.75 .
2. The normalized motions of the floating breakwater in 2D are almost the same for waves with longer
periods. However the average value of RAO associated with the stiffness of EA/Ws=2500 is lower
in comparison with other values of mooring line stiffness, and it has a better hydrodynamic
performance.

7. REFERENCES

1. McCartney, L. (1985), “floating breakwater design”. Journal of Waterway, Port, Coastal and Ocean
Engineering, Volume 111.
2. Sannasiraj, S. A., Sundar, V. & Sundaravadivelu, R. (1998), “Mooring Forces and Motion Responses of
Pontoon-type Floating Breakwaters”. Ocean Enginnerig, Volume 25, pp. 27-48.
3. Irvine, M. (1981), “Cable Structures”. s.l.:The Massachusetts Institude of Technology.
4. Li, Y. & Lin, M. (2012), “Regular and Irregular Wave Impacts on Floating Body”. Ocean Engineering,
Volume 42, pp. 93-101.
5. Zhang, C. et al. (2013), “A Two-phase Flow Model Coupling with Volume of Fluid and Immersed
Boundary Methods for Free Surface and Moving Structure Problems”. Ocean Engineering, Volume 74,
pp. 107-124.
6. Mansard, E. & Funke, E. (1980), “The Measurement of Incident and Reflected Spectra Using a Least
Squares Method”. Coastal Engineering, pp. 155-172.
7. Barltrop, N. D. P. (1998), “Floating Structures”: A Guide for Design and Analysis. s.l.:Oilfield
Publications Limited.