Available online at www.sciencedirect.

com

ScienceDirect
Procedia Engineering 194 (2017) 67 – 73

10th International Conference on Marine Technology, MARTEC 2016

Numerical and Experimental Study of Resistance, Sinkage and Trim

of a Container Ship
Saha G. K.a , Miazee M. A.b,∗
a Professor, Department of Naval Architecture and Marine Engineering, BUET, Dhaka 1000, Bangladesh
b Lecturer, Department of Naval Architecture and Marine Engineering, Sonargaon University (SU), 147/I Green Road, Dhaka 1215, Bangladesh

Abstract
This paper presents the effectiveness of the Computational Fluid Dynamics modeling software SHIPFLOW, in determining different
hydrodynamic values such as resistance, sinkage and trim for a container ship. This ship was simulated in deep water at different
Froude number including service speed. The viscous, wave and total resistance were calculated for speeds ranging from 8.0 to
10.0 knots. The model scale was simulated to obtain the resistance, sinkage and trim values for these corresponding speeds.
Finally, CFD values compared with the experimental values carried out in Towing Tank facilities. Experimental investigation
appears to provide agreement with computed values and showing that CFD methods may adequately deal with the problem of
wave interference produced by the container ship hull.

©c 2017
2017Published
The Authors. Published
by Elsevier by Elsevier
Ltd. This is an openLtd.
access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of the 10th International Conference on Marine Technology.
Peer-review under responsibility of the organizing committee of the 10th International Conference on Marine Technology.
Keywords: Containership; CFD; Ship resistance; Sinkage; Trim; SHIPFLOW

1. Introduction
A container ship can be loaded and unloaded in a few hours compared to days in a traditional cargo vessel. This,
besides cutting labor costs, has reduced shipping times between ports to a great extent. It has also resulted in less
breakage due to less handling; also, there is less danger of cargo shifting during a voyage. As containers are sealed
and only opened at the destination, pilferage and theft levels have been greatly reduced.
Container ships are constantly becoming larger and are required to carry more loads and travel at faster speeds.
Therefore, it is vital to the development of these huge structures that accurate methods of obtaining design data
are established. Model testing in towing tanks is the past and present way of doing this for container ships. With
computers reaching new levels and our understanding of CFD always improving, new and old programs require their
accuracy in hydrodynamics to be determined. In the field of ship hydrodynamics, the first theoretical solution for
the problem of wave resistance was given by Michell[1] for a thin ship moving on the surface of an inviscid fluid.
Later on, Kelvin [2] established the fundamental theory of ship waves. Since then many theoretical studies in ship
hydrodynamics have been undertaken. Havelock [3] studied the effects of shallow water on the wave resistance and
wave pattern for a point pressure impulse traveling over a free surface. Peng [4] presented a numerical method based
on Michell’s theory for the resistance in calm water.

∗ Corresponding author. Tel.:+88-01922524924

E-mail address: alamin.name@gmail.com

1877-7058 © 2017 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of the 10th International Conference on Marine Technology.
doi:10.1016/j.proeng.2017.08.118
68 G.K. Saha and M.A. Miazee / Procedia Engineering 194 (2017) 67 – 73

Nomenclature

CF Specific frictional resistance coefficient

CFD Computational Fluid Dynamics RR Residuary resistance
CPV Viscous pressure resistance coefficient RS Ship resistance
CT Total resistance coefficient RW Wave resistance
CV Specific viscous resistance coefficient Rn Reynolds number
CW Wave resistance coefficient t Time
Fn Froude number u Velocity component in x-direction
g Acceleration due to gravity U Free stream velocity
ITTC International Towing Tank Conference v Velocity component in y-direction
P Pressure ν Kinematic viscosity
RANS Reynolds Averaged Navier-Stokes w Velocity component in z-direction
RT Total resistance ρ Mass density
RV Viscous resistance φ Velocity Potential
RF Frictional resistance μ Dynamic viscosity

Suzuki [5] developed the Neumann-Kelvin problem as a method of calculating the effect of sinkage and trim on
the wave resistance of a ship. Yasukawa [6] predicted the wave-making resistance taking into account the effect of
sinkage and trim by the Rankine source method. The importance of the trim and sinkage of a ship are well indicated
by Subramani et al.[7]. The other researchers who have made important contributions in the field of wave-making
resistance of ship taking into account the effects of sinkage and trim are Tuck [8], Bessho and Sakuma [9], Gourlay
and Tuck [10], Azcueta [11] etc.
The solution of problems involving determination of resistances using Computational Fluid Dynamics (CFD) anal-
ysis is now becoming tractable due to enhanced accessibility to high performance computing. Determination of the
resistance characteristics of ships/ vessels is one of the most important topics in Naval Architecture, Offshore and
Ocean Engineering.
Today, several CFD tools play an important role in the design of the ship hull forms. CFD has been used for the
analysis of ship resistances, seakeeping, maneuvering and investigating the variation in resistance encountered due
to changes in the ship hull resulting from variation in its parameters. This paper is organized as follows: first, the
resistance components of ship are defined. Second, the calculation procedure of resistance is described. Third, a short
description of the SHIPFLOW software is given. Forth, calculated results are compared with experimental results.
Finally, a summary of the work is provided with future works.

2. Theoretical background

This section present the theory behind the work carried out in the project. The CFD software and its corresponding
theories behind the calculations of resistance, sinkage and trim that are described here.

2.1. Resistance

When a ship is moving through water there will be forces opposing the motion. The total resistance, RT , of a
ship is defined as the force needed to tow the ship at a constant speed and it can be divided into subcomponents in
different ways. One way is to divide it into frictional resistance, RF , and residuary resistance, RR , which includes
all components related to the three-dimensional form of the ship and wave-making resistance. It can also be divided
according to physical phenomena into viscous resistance, RV , and wave resistance, RW . The viscous effects are
excluded from the wave resistance and it is therefore considered as an inviscid phenomenon [12].
A boundary layer is created along the entire hull which grows downstream. The thickness of the boundary layer is
defined as the distance from the hull surface to the point where the velocity is 99% of the undisturbed flow velocity
[13].
 G.K. Saha and M.A. Miazee / Procedia Engineering 194 (2017) 67 – 73 69

2.2. Computation of resistance

There are three types of resistance components that can be computed in the present method, wave resistance,
frictional resistance and viscous pressure resistance.
Wave resistance is obtained from an integration of the potential flow pressure over the body surface. The pressure
integration is for first order panels carried out as a summation over flat panels where the pressure and the normal
direction are assumed to be constant over each panel.
Integration of the local skin friction coefficient over the hull surface is carried out to obtain the frictional resistance.
The local skin friction is computed as a part of the solution along the stream lines in the boundary layer method and
from the wall shear stress of the wall in the Navier-Stokes solution.
The potential-flow pressure over the stern part of the hull is changed due to viscous effects computed by the Navier-
Stokes solution. An integration of the pressure including viscous effects gives the viscous pressure resistance. A form
factor is computed from the viscous pressure resistance and the frictional resistance where ITTC-57 line is used as
reference.
The frictional resistance coefficient, CF , in accordance with the ITTC-57 formula is defined by:
0.075 RF
CF = = (1)
(logRn − 2) 2 1/2.ρ.S .V 2
Where the frictional resistance, RF , is sum of tangential stresses along the wetted surface in the direction of the motion
and Rn is the Reynolds number. The total resistance coefficient, CT of a ship can be defined by:
RT
CT = (2)
1/2.ρ.S .V 2
Where, RT is the total resistance.

2.2.1. Potential flow

A flow that is irrotational, inviscid and incompressible is called potential flow. If irrotational flow is assumed,
∇ × V = 0, the components of the velocity vector are no longer independent of each other and a velocity potential φ
exist as described by Eq. 3.
 ∂φ ∂φ ∂φ 
∇φ ≡ , , = V ≡ (u, v, w) (3)
∂x ∂y ∂z
∂u ∂v ∂w
∇.V ≡ + + =0 (4)
∂x ∂y ∂z
Eq. 3 together with the continuity equation, Eq. 4, yields Laplace’s equation Eq. 5, which is the equation that potential
flow methods are based upon [12].

∂u ∂v ∂w ∂2 φ ∂2 φ ∂2 φ
+ + = 0 → 2 + 2 + 2 ≡ ∇2 φ = 0 (5)
∂x ∂y ∂z ∂x ∂y ∂z
When the flow involves a free surface, the Bernoulli equation, Eq. 6, provides a relation between the flow velocity,
V, and the elevation of the free surface, z, according to Eq. 7[14]. Finally, the pressure coefficient can be determined
from Eq. 8.
∂φ p 1 2
+ + V + gz = constant (6)
∂t ρ 2
V 2 = |∇φ|2 = constant − 2gz (7)
2
V
Now the pressure coefficient, C P = 1 − (8)
U2 
Rw 1
And from [15] the wave making resistance coefficient, Cw = =− C p n x dS (9)
0.5ρS V 2 S
70 G.K. Saha and M.A. Miazee / Procedia Engineering 194 (2017) 67 – 73

2.2.2. Viscous flow, RANS

The Navier-Stokes equations, Eq. 10 together with the continuity equation, Eq. 4, would be enough to solve all
flow fields for ship flows. Unfortunately, it requires immense computational power. That is why a number of methods
exist today that deal with simplifications of the Navier-Stokes equations.
⎧  ∂2 u ∂2 v ∂2 w 
⎪
⎪
⎪ ∂u ∂u ∂v ∂w 1 ∂p
⎪
⎪
⎪ +u +v +w =− +ν 2 + 2 + 2 (10)
⎪
⎪
⎪ ∂t ∂x ∂y ∂z ρ ∂x ∂x ∂y ∂z
⎪
⎪
⎪
⎪
⎨ ∂v ∂v ∂v ∂v 1 ∂p  ∂2 v ∂2 v ∂2 v 
⎪
⎪ +u +v +w =− +ν 2 + 2 + 2 (11)
⎪
⎪
⎪ ∂t ∂x ∂y ∂z ρ ∂y ∂x ∂y ∂z
⎪
⎪
⎪
⎪
⎪
⎪ ∂w ∂w ∂w ∂w 1 ∂p  ∂ w ∂2 w ∂2 w 
2
⎪
⎩ ∂t + u ∂x + v ∂y + w ∂z = − ρ ∂z − g + ν ∂x2 + ∂y2 + ∂z2
⎪ (12)

By dividing the velocities and pressure into a time average and a fluctuating part it is possible to bring the Navier-
Stokes equations into a form which is possible to solve numerically with the computational power of today. This
time averaged method is called Reynolds Averaged Navier-Stokes, RANS, and is the viscous method used in Eq. 11.
The time averaged velocities are denoted according to ui = u¯1 + ui where u¯1 is the average and ui is the fluctuating
component in the x-, y- and z-direction. Time averaging Navier-Stokes equations eliminate the turbulent fluctuations
in all terms except the Reynolds stresses,−ρul u1 , see the last terms in Eq. 11. The Reynold stresses introduce six
additional unknowns into the equation system consisting of four equations. This is solved by using a turbulence
model together with RANS to couple the Reynold stresses with the average velocities [16, 17].
⎧
⎪
⎪
⎪  ∂u ∂u ∂v ∂w  ∂p  ∂2 u ∂2 v ∂2 w   ∂u2 ∂u v ∂w u 
⎪
⎪
⎪ ρ + + + = − + μ + + − ρ + +
⎪
⎪
⎪ ∂t
u
∂x
v
∂y
w
∂z ∂x ∂x 2 ∂y 2 ∂z 2 ∂x ∂y ∂z
(13)
⎪
⎪
⎪
⎪
⎪
⎪     
⎨ ∂v ∂v ∂v ∂v ∂p ∂ v ∂v ∂v
2 2 2 
∂u v  ∂v 2 ∂w v 

⎪
⎪
⎪ ρ + u + v + w = − + μ + + − ρ + + (14)
⎪
⎪
⎪ ∂t ∂x ∂y ∂z ∂y ∂x2 ∂y2 ∂z2 ∂x ∂y ∂z
⎪
⎪
⎪
⎪
⎪
⎪  ∂w ∂w ∂w ∂w  ∂p ∂ w ∂ w ∂ w
2 2 2  ∂u w ∂v w ∂w2 
⎪
⎪
⎩ρ ∂t + u ∂x + v ∂y + w ∂z = − ∂z − ρg + μ ∂x2 + ∂y2 + ∂z2 − ρ ∂x + ∂y + ∂z
⎪ (15)

RANS solver uses a finite volume method where the computational domain is divided into a number of cells. Govern-
ing equations are integrated over each cell before the investigated variable is approximated with a value at the centroid
of each cell. The finite volume method ensures that the quantity of variables such as mass, momentum and energy is
conserved since errors at the cell faces cancel with the errors of neighboring cells [18].

2.3. SHIPFLOW CFD software

The CFD software used for the investigation is called SHIPFLOW which is a powerful suite of CFD tools by
FLOWTECH International AB. SHIPFLOW uses panel methods to calculate the co-efficient of wave resistance, and
therefore it is necessary to define the grid of panels that will be used for the analysis. The module XMESH is used to
define the groups/surfaces that are to be considered. To define the grid of panels representing the body, the number of
stations to be used along the length of the hull and the number of points across each station must be specified.
The theoretical wave resistance coefficient, CW is calculated by splitting the flow into three regions where an
efficient approximation of the flow equations may be made and a complete flow calculation may be accomplished in
a few hours using the potential flow, as described by [19].

• Flow in Zone 1 is calculated using a higher order panel method with linear or non-linear free surface boundary
conditions.
• Flow in Zone 2 is calculated using momentum integral methods for laminar and turbulent boundary layers.
• Flow in Zone 3 is calculated using the Reynolds-Average Navier-Stokes method with a κ- turbulence model
and a numerically generated body fitted coordinate system.
 G.K. Saha and M.A. Miazee / Procedia Engineering 194 (2017) 67 – 73 71

Fig. 1: Boundary layer along the hull Fig. 2: Hull of the container ship

3. Result and discussion

The 3D model of the container ship is shown in Fig. 2 and the principle particulars are described in the Table 1.

Table 1: Principle particulars of the Container ship

Particulars Prototype Model

Length Over All (LOA) 76.00 m 4.750 m
Length between perpendiculars (LBP) 72.87 m 4.554 m
Breadth Overall 15.00 m 0.938 m
Draft 3.80 m 0.238 m
Depth 7.00 m 0.438 m

In SHIPFLOW the total number of cells used is 744372. The computed values of resistance coefficients are
tabulated in Table 2. There it is seen that the Viscous resistance coefficient (CV ) which is the combination of Frictional
resistance coefficient (CF ) and Viscous Pressure resistance coefficient (CPV ) decreases with the increase of Froude
number (Fn ). The Wave resistance coefficient (CW ) and finally the Total resistance coefficient (CT ) also decrease
smoothly with the increase of Froude number (Fn ). Fig. 3 shows the comparison between the computed (RT ) by ship

Table 2: Computed and experimental values of resistance coefficients

Experimental
Computed by SHIPFLOW
results
V (Knots) Fn CF CPV CF + CPV = CV CW CV + CW = CT RT(kN) RT(kN)
8 0.154 0.001671 0.000611 0.002282 0.001581 0.003862 48.2 49.7
8.5 0.164 0.001653 0.000610 0.002264 0.001393 0.003657 51.5 54.4
9 0.173 0.001642 0.000609 0.002252 0.001285 0.003537 55.9 62.6
9.5 0.183 0.001626 0.000606 0.002233 0.001201 0.003434 60.4 74.7
10 0.192 0.001601 0.000605 0.002206 0.001100 0.003306 64.5 83.9

SHIPFLOW and the experimental results. Therefore, it is observed that the difference between the computed values
and the experimental values increases with the increase of speed. It is shown that in the lower Froude number the
agreement is good but the difference is high in higher Froude number. Table 3 shows the changes of Sinkage and Trim
at centre of flotation with changes of speed. Negative sinkage indicates downwards directions from free surface and
negative trim indicates anticlockwise directions or trim by stern.
Results regarding sinkage and trim are presented in Fig.4. From the Figure, it is seen that the sinkage is less than
10% of the draft and the absolute trim angle remains small (between −0.090 to −0.20 ) [21]. Fig. 5 and Fig. 6 present
the Kelvin wave pattern (wave contours) and wave profiles at Froude numbers 0.173, and 0.192. The variation in the
wave pattern as the wave propagates away from the hull can be observed. The resolution of the free surface reduces
significantly and the wave cuts along the length of the ship increase with speed. The wave height increases with the
72 G.K. Saha and M.A. Miazee / Procedia Engineering 194 (2017) 67 – 73

Fig. 3: Resistance prediction of SHIPFLOW and Towing

Tank Fig. 4: Sinkage and trim calculated by SHIPFLOW

Table 3: Values of sinkage and trim with different Froude number

Speed (Knots) Froude Number Sinkage (m) Trim (Degree)

8 0.154 -0.0722142 -0.095
8.5 0.164 -0.0823431 -0.112
9 0.173 -0.0918162 -0.13
9.5 0.183 -0.1034754 -0.15
10 0.192 -0.1136772 -0.173

increase of ship speed. The wave profile near the bow starts with a trough at low Froude number and it starts with
crests as the Froude number increases. In SHIPFLOW results wave resistance of the model has been calculated based
on the regression equation developed by [20].

Fig. 5: Kelvin wave pattern and wave profiles at the speed of 9 knots

Fig. 6: Kelvin wave pattern and wave profiles at the speed of 10 knots
 G.K. Saha and M.A. Miazee / Procedia Engineering 194 (2017) 67 – 73 73

4. Conclusions
If finer meshes would have been used, if the number of panels per wavelength would be increased and if the hull
is optimized, we would have received more accurate values for this case. Resistance calculated in SHIPFLOW are
slightly lower than the experimental results in this speed range. It appears that the resistance result of the SHIPFLOW
software is very close, especially at 0.15 < Fn < 0.18, with experimental results for the container ship. It is seen that
if the number of discretization cells is increased that is, in case of fine meshe SHIPFLOW prediction is very much
close to the experimental one.
It is imperative to note that only the container ship have been checked in this instance, which implies that the range
could be further enlarged and more rigorous validation is required against experimental results for many other ships.

References
[1] Michell, J. H., The wave resistance of a ship, Philosophical Magazine, Series 5, Vol. 45, No. 272, London, 1898.
[2] Kelvin, L., Deep sea ship waves, Proceedings of Royal Society Edinburgh, No. 5, 1905.
[3] Havelock, T. H., The propagation of groups of waves in dispersive media with application to waves on water produced by a travelling distur-
bance, Proceedings of the Royal Society, London, Vol. 81, 1908.
[4] Peng, H., Numerical Computation of Multi-hull Ship Resistance and Motion, Ph. D. Thesis, Dalhousie University, Halifax, Nova Scotia, 2001.
[5] Suzuki K., Calculation of ship wave resistance with special reference to sinkage. Proceedings of the Workshop on Ship Wave Resistance
Computations, vol.2, pp.256-281, 1979.
[6] Yasukawa H., A Rankine panel method to calculate steady wave-making resistance of a ship taking the effect of sinkage and trim into account.
Transactions of the West Japan Society of Naval Architects, No.86, pp.27-35, 1993.
[7] Subramani A.K., Paterson E.G. and Stern F., CFD calculation of sinkage and trim. Journal of Ship Research, 44 1 (2000) pp.59-82.
[8] Tuck E.O., Sinkage and trim in shallow water of finite width. Schiffstechnik, vol.14, pp.92-94, 1967.
[9] Bessho M. and Sakuma S., Trim and sinkage of two-dimensional shallow draft ships. Journal of the Kansai Society of Naval Architects, Japan,
218 (1992).
[10] Gourlay T.P. and Tuck E.O., The maximum sinkage of a ship. Journal of Ship Research, 45 1 (2001) pp.50-58.
[11] Azcueta R., RANSE simulations for sailing yachts including dynamic sinkage and trim and unsteady motions in waves. High Performance
Yacht Design Conference, Auckland, 2002.
[12] Larsson, L. and Raven, H., Ship Resistance and Flow. Gothenburg and Wageningen: Chalmers University of Technology, Gothenburg, Sweden,
2004.
[13] Bertram, V., Practical Ship Hydrodynamics, Oxford: Butterworth-Heinemann, 2000.
[14] White, F., Fluid Mechanics, 6th Edition. New York: McGraw-Hill, 2008.
[15] Yu Lu, Xin Chang and An-kang Hu, A hydrodynamic optimization design methodology for a ship bulbous bow under multiple operating
conditions, Engineering Applications of Computational Fluid Mechanics, 10:1, 330-345, 2016.
[16] Flowtech Int., XCHAP Theoretical Manual. Gothenburg: FLOWTECH International AB, 2010.
[17] Flowtech Int., SHIPFLOW User Manual. Gothenburg: FLOWTECH International AB, 2010.
[18] Eliasson, S. and Olsson, D., Barge Stern Optimization: Analysis on a Straight Shaped Stern using CFD, MSc thesis, Chalmers University of
Technology, Gothenburg, Sweden, 2011.
[19] Larsson, L., Resistance and Flow Predictions Using SHIPFLOW Code, 19th WEGEMNT School, Nantes, France, 1993.
[20] Sahoo, P.K., Browne, N.A. and Salas, M., Experimental and CFD Study of Wave Resistance of High-Speed Round Bilge Catamaran Hull
Forms, Proc. of the 4th International Conference on High Performance Marine Vehicles, 2004.
[21] Kim, Y. H., Jenkins, D., Trim and Sinkage effects on wave resistance with series 60; Cb= 0.60, David W. Taylor Naval Ship Research and
Development Center, Bethesda, Maryland, 1981.