THEORETICAL SHIP WAVE PATTERN RESISTANCE EVALUATION USING KOCHIN

WAVE AMPLITUDE FUNCTION

Chavali Revathi 1 and Md. Kareem Khan2
1
Scientist , Indian Maritime University (Visakhapatnam) , India
2
Scientist , Naval Science & Technological Laboratory , India

ABSTRACT

Ship wave pattern resistance evaluation has been an important event in the optimization process of variety of
hull forms ranging from small crafts to fast displacement hull forms including multihull vessels. Several experimental
methods for wave pattern analysis like Hogben Matrix Element Method (MEM), Longitudinal & transverse wave cuts
involving Landweber Fourier Transform (LFT) and Matrix solution method, theoretical methods like Michell's thin ship
theory, Rankine source-sink and Kochin-Fourier method exist today with their merits and limitations. Ship wave pattern
resistance is mostly a derived quantity from experiments either as a subset of residuary resistance or when total viscous
component is known. In general applying Froude's hypothesis, the residuary resistance coefficient is not subjected to
scaling effects from model to prototype and evaluation of total viscous component for surface ships by model
experiments is seldom done. Optimization of hull forms through wave pattern studies is a powerful tool and in-house
potential flow solvers/experimental evaluation techniques for free surface flows have been developed in many
hydrodynamic test facilities. The present paper focuses on ship wave pattern evaluation using Michell's theory (using
Kochin wave amplitude function) and its comparison with Shipflow® & Longitudinal wave cut method (using LFT) on
bench mark test model R/V Athena. Matlab® code development and ship wave pattern resistance evaluation sensitivity
for the method chosen is also a part of present discussion.

NOMENCLATURE 1. INTRODUCTION

Nomenclature of the terms used is as listed below The demand for high performance ships that
CW Wave Resistance Coefficient develops better speed at lower power demand and better
RWP Wave Resistance (N) ride comfort in a seaway has increased in the recent past.
L, LWL Length on waterline (m) To develop optimal hull forms for such vessels, thorough
LBP Length Between Perpendiculars (m) investigation into resistance and other hydrodynamic
B Breadth/ Beam (m) characteristics of the designs have to be performed.
D Depth (m) Wave making resistance is an important component of
T Draft (m) total hydrodynamic drag for surface ships and deals with
N Number of free modes the energy required to maintain the wave system. Several
Fn Froude number wave pattern analysis methods, both analytical [4, 19]
U Velocity (m/s) and experimental [1, 20], have been developed in order
W Width of tank (m) to assess the wave making resistance qualitatively and
y Distance of wave probe from centreline (m) quantitatively and also as a tool for the hull form
ρ Density of water (kg/m3) optimization process.
S Wetted Surface area (m2)
t Time interval for measuring the wave profile Determination of the resistance to the motion of a
ζ Height of waves at given section/waterlines (m) body in fluid (liquids) is a fundamental problem in
Δ Displacement of vessel (tonnes) hydrodynamics. This problem, difficult enough for
g Acceleration due to gravity (m/s2) arbitrary shaped bodies which are completely immersed,
k wave number is even more complex for surface ships which are only
k0 basic wave number partially immersed (multi-phase flow phenomenon).
θ wave incidence angle Hydrodynamics for surface ships comprises of free
kx Longitudinal Wave Number = k cos θ surface effects and viscous effects. The physical
ky Transverse Wave Number = k sin θ modelling of free surface effects is complex. The wave
A(θ) The wave amplitude spectrum = A(k(θ),θ) making resistance (mainly residuary resistance) is
AS (θ) sine (odd) component of the wave spectrum conventionally determined through hydrodynamic model
AC (θ) cosine (even) component of the wave spectrum tests in a towing tank using ITTC method [7].
Cw(θ) wave pattern resistance coefficient with respect
to θ Perform analytical and numerical analysis in tandem
x wave number with experiments for estimating the wave making
wn frequency (sec-1) resistance of ships will surely optimise the model testing
X Distance along x direction (m) matrix. A thorough validation of these non-experimental
methods using published bench mark data is very
important in further application of such methods [1]. The
present paper makes an attempt to determine the wave
making resistance using analytical method – Kochin
amplitude wave function using Michell’s thin ship theory
and its validation with existing published results on R/V
Athena [26] & commercially available software
Shipflow®
2. TECHNIQUES/METHODS FOR ESTIMATION
OF WAVE MAKING RESISTANCE
Wave making resistance has been obtained using
analytical/numerical and experimental techniques [1] as
listed in table below.
Table 1
S.No Techniques Methods
1 Experimental -Fourier Transform
(wave cuts) -Matrix solution
-Spectral analysis
2 Analytical -Michell's thin ship theory
-Rankine source distribution
3 Numerical -CFD methods
The wave making resistance estimation through
experimentation is normally conducted in a towing tank.
Wave probes connected to a data acquisition system are
used to measure longitudinal wave traces. Capacitance
type wave probes are positioned on one side of the tank
in transverse direction in order to conduct the multiple
longitudinal cut wave pattern analysis. The wave pattern
of a ship model, symmetric relative to the tank centreline,
can be expressed as a summation of a number of free
waves,
∞ specified, we can use this to determine the actual wave
2πny
𝜁 = ∑(𝜉𝑛 cos 𝑥𝑤𝑛 + 𝜂𝑛 sin 𝑥𝑤𝑛 ) cos ( )
W A(θ) is a property only of the body’s hull geometry,
𝑛=0
The wave pattern analysis aims to derive the wave
coefficients ξn and ηn from the wave traces taken either
across the tank or along the tank by employing analysis
methods. Four main analysis techniques are available.
2.1 Fourier transformation:
The free-wave spectrum can be determined by
Fourier transformation of either longitudinal or
transverse wave cut data for the steady-ship wave system
in the time domain. Standard procedure is described in
detail by Eggers et al.
2.2 Matrix Solution Method
The method uses simultaneous equations for solving
the wave coefficients .The errors due to truncation of
wave trace length and due to non periodic wave trace can
be prevented by applying this method [1].
2.3 Spectral density method:
Wave profiles at a particular probe distances is
utilised for estimating wave spectrum and thus analysed
to obtain the wave energy /wave making resistance.
3. KOCHIN WAVE AMPLITUDE FUNTION
METHODOLOGY
The steady surface wave pattern z = ζ(x, y) of any
moving body, as seen at a point (x, y) sufficiently far
from the ship, is of the form of a sum of plane waves
travelling at various angles θ of propagation relative to
the direction of motion (negative x-axis) of the body.
Fig 1 Coordinate system definition
+𝜋/2
ζ(x, y) = ℜ ∫−𝜋/2 𝐴(𝜃) e−iΩ(θ) 𝑑𝜃
𝛺(𝜃) = 𝑘(𝜃)[𝑥𝑐𝑜𝑠𝜃 + 𝑦𝑠𝑖𝑛𝜃]
𝛺(𝜃) is a phase function and A(θ) is the amplitude and
k(θ) the wave number of the wave component travelling
at angle θ. Note that the contributions to this integral
from positive angles θ correspond to waves being
propagated to the left or portside of the body. In the
present report the body and flow are assumed to have
lateral symmetry, and then A(θ) is an even function and
need only be computed for θ≥ 0. Once A(θ) and k(θ) are
pattern. At any given speed U, the amplitude function
whereas k(θ) is fully determined from the dispersion
relation for plane waves. In infinite water depth, which
we shall assume in the present report, we have simply
𝑘(𝜃) = k 0 sec 2 θ
Where k0 = g/U2 = k(0) is the wave number of pure
transverse waves at θ = 0. The complex amplitude
function A(θ), sometimes also called the free wave
spectrum or Kochin function, can be computed by
various means, e.g. Michell’s thin-ship theory, a
complete nonlinear near-field computation or by
experimental measurement . For a thin body with offsets
y = Y (x, z) in water of infinite depth, Michell’s theory
indicates that
𝐴(𝜃) = −
2𝑖
𝑘02 𝑠𝑒𝑐 4 𝜃 ∬ 𝑌(𝑥, 𝑧)exp(𝑘0 𝑧𝑠𝑒𝑐 2 𝜃 + 𝑤2𝑖+1 = 4(𝑠𝑖𝑛𝐾 − 𝐾𝑐𝑜𝑠𝐾)/𝐾 3
𝜋
𝑖𝑘0 𝑥𝑠𝑒𝑐𝜃)𝑑𝑥𝑑𝑧
where K = k0secθΔx.
The total energy left behind in the wave field can easily 𝑁𝑧
be computed once the amplitude function A(θ) is known,
which leads to the famous Michell's integral for the wave 𝐹(𝑥, 𝜃) ≈ ∑(𝜔j 𝑌(𝑥, 𝑧𝑗 ) exp( k 0 zj sec 2 θ)∆z)
resistance R for a ship, namely 𝑗=0
+𝜋/2
𝜋 2 where the section is divided into Nz segments each of
𝑅𝑊𝑃 = 𝜌𝑈 ∫ |𝐴(𝜃)|2 𝑐𝑜𝑠 3 (𝜃)𝑑𝜃 height Δz, with end-points zj; j = 0, 1, 2, …,Nz, and the
2
−𝜋/2 weights are given by ωj
where
The thin-ship theory of Michell [] represents the body by 𝜔0 = (𝑒 𝐾 − 1 − 𝐾)/𝐾 2
a centre plane source distribution proportional to its 𝜔𝑁𝑧 = (𝑒 −𝐾 − 1 + 𝐾)/𝐾 2
longitudinal rate of change of thickness (local beam).
The only requirement for its validity is that that quantity and for j ≠ 0, Nz, 𝜔𝑗 = (𝑒 𝐾 + 𝑒 −𝐾 − 2)/𝐾 2
be small. Hence the theory applies to submerged as well
as surface-piercing bodies. The double integral over the where K = k0 sec2θΔz.
body’s centre plane, determining A(θ) for each fixed
angle θ from the offsets Y(x, z), can be reduced to a pair wave making resistance coefficient can be determined as
of separate integrals over depth and length as follows. follows
Evaluation of 𝐹(𝑥, 𝜃) for all stations x is attempted in +𝜋/2
the first phase. 𝑅𝑊𝑃
𝐶𝑊 = = ∫ 𝐶𝑊𝑃 (𝜃)𝑑𝜃
1 2
𝜌𝑈 𝑆 −𝜋/2
𝐹(𝑥, 𝜃) = ∫ 𝑌(𝑥, 𝑧) exp(𝑘0 𝑧𝑠𝑒𝑐 2 𝜃) 𝑑𝑧 2

4. PRESENTATION OF RESULTS
The integral is in the vertical z-direction, from the lowest
point of the section in its actual position with z <0, up to Matlab® code using Kochin amplitude wave
the waterline z = 0. Next, evaluation of pair of integrals function method was generated for the evaluation of
𝑃(𝜃), 𝑄(𝜃) along the body from bow to stern is wave making resistance and the CFD analysis was
performed. performed using Shipflow®. A bench mark ship (R/V
Athena) has been selected for the application of the
𝑃(𝜃) = ∫ 𝐹(𝑥, 𝜃) cos(𝑘0 𝑥𝑠𝑒𝑐𝜃) 𝑑𝑥 present code and its validation with the published
𝑄(𝜃) = ∫ 𝐹(𝑥, 𝜃) sin(𝑘0 𝑥𝑠𝑒𝑐𝜃) 𝑑𝑥 experimental data as well with Shipflow®. The
experiments conducted at NSWC [26] and NSTL [27]
Amplitude function can be finally represented as has been utilised for validation. Main particulars of R/V
Athena are given in Table 2.
2𝑖 2 4
𝐴(𝜃) = − 𝑘 𝑠𝑒𝑐 𝜃[𝑃(𝜃) + 𝑖𝑄(𝜃)] Table 2
𝜋 0
S.No. Parameter
There are a number of special features, especially the
1 LBP 46.87 m
singularity as θ→ π/2. In that limit (which corresponds to
the diverging part of the ship wave pattern), the effective 2 B 6.68 m
wave number k0secθ becomes infinite, and the x integral 3 Δ 260 tonnes
has a highly-oscillatory integrand. This is a well-
4 U 12 kn - 35 kn
understood problem, but it does require some care,
especially in data specification near the bow and stern,
which contribute most to the integrals in that limit. It is The evaluation of the code was done in a Froude number
moderated, especially for submerged or semi submerged range of 0.25 to 0.7 for a fixed sinkage and trim and
comparison of the wave profiles at y/L = 0.2 for Fn 0.257
hulls, by the exponential decay factor from the z integral
& 0.44 is given in Fig 3 & 4.
that damps out contributions from near θ= π/2 [19] . If
there is no contribution from the extreme bow and stern
(i.e. Y = 0 and hence F = 0 there), the algorithm used is
simply

𝑥 𝑁 −1
𝑃(𝜃) ≈ ∑𝑖=1 (𝜔i 𝐹(𝑥𝑖 , 𝜃) (cos k 0 xi secθ)∆x)

𝑤2𝑖 = (3𝐾 + 𝐾𝑐𝑜𝑠2𝐾 − 2𝑠𝑖𝑛2𝐾)/𝐾 3


Fig 2 Wave contour for Fn = 0.44
Fig 3 Wave heights at y/L = 0.2 for Fn = 0.257
Fig 4 Wave heights at y/L = 0.2 for Fn = 0.44
Fig 5 Coefficient of wave resistance vs Froude number
5. CONCLUSIONS
The present paper successfully made an attempt through
Matlab code for evaluation of wave making resistance of
a surface ship using Kochin amplitude wave function.
Fig 3 & 4 shows a comparison of the present analytical
method. Fig 5 shows a comparison of wave making
resistance coefficient with respect to Froude number. The
wave profile comparison fro Froude number less than 0.3
shows more number of wave oscillation than with
experiments, but the amplitudes after 1.2L are nearly
same and the same is the behaviour with the phase
angles. For Fn > 0.3 the wave amplitudes & phase angles
have a very fair match with experiments. Shipflow
results too behaved in a similar manner.
Wave contour plot in Fig 2 shows the wave pattern for
the present ship at Fn 0.44, which seems to be simulating
the actual physics. One can observe wave crest at bow
region (red in colour) and trough at the after location
(slightly blue in colour). Fig 5 plot shows the reliability
of the present method in evaluating the wave making
resistance. Cw plot is in close agreement with both
experiments and Shipflow® results. Except at lower
Froude numbers (Fn<0.3) where the Cw from present
method as well as from Shipflow® diverged, the rest of
the curve has a trend. Table 3 gives the values of wave
making resistance comparison at Fn 0.257 & 0.44
Table 3:
Fn = 0.257 Fn = 0.44
RW 103 CW RW 103 CW
NSWC 2.98 0.39 36.96 1.70
Shipflow® 13.6 1.75 41.6 1.91
Kochin wave
amplitude function 12.54 1.69 40.64 1.86
method
The mathematical code generated for the evaluation of wave
making resistance have given the results very close to the
experimental and the Shipflow results. Hence the code can be
used for the preliminary estimation of wave making
resistance and optimization the hull form. Further the code
can be improved by incorporating free sinkage & trim to
simulate the dynamic motion conditions and also its
extension to multihulls.
6. REFERENCES
1. Mustafa Insel (1990) An investigation into the
resistance components of high speed displacement
catamarans, PhD thesis, University of
Southampton, Department of Ship Science
2. Douglas S. Jenkins (1984) Resistance
characteristics of the high speed transom stern
ship R/V Athena in the bare hull condition,
represented by DTNSRDC model 5365 by
Douglas S. Jenkins, 1984.
3. Justus Heimann (2005) CFD based optimization
of the Wave Making Characteristics of Ship Hulls
by, Friendship systems GmbH.
4. Hoyte C, Raven and Henk J. Prins (1997) Wave
pattern analysis applied to nonlinear ship wave
calculations
5. Dimtris E Nakos, (1991) Transverse wave cut
analysis by a Rankene panel method, MIT
Cambridge, Massachusetts.
6. D. Bülent Danışman, Ömer Gören, Mustafa
Insel, Mehmet Atlar (2000) An Optimization
Study for the Bow Form of High Speed
Displacement Catamarans.
7. Donald C. Wyatt, Thomas C. Fu, Genevieve L.
Taylor, Eric J. Terrill, Tao Xing, Shanti
Bhushan, Thomas T. O’Shea, and Douglas G.
Dommermuth (2008) A comparison of full-scale
experimental measurements and computational
predictions of the transom-stern wave of the R/V
Athena I.
8. John P. Hackett, Jessica Calix St. Pierre,
Christopher Bigler, Todd J. Peltzer, Frans
Quadvlieg, Frans van Walree (2007)
Computational Predictions vs. Model Testing for a
High Speed Vessel with Lifting Bodies.
9. McLean, VA (1998) Evaluation of Surface-Ship
Resistance and Propulsion Model-Scale Database
for CFD Validation by Fred Stern and Joe Longo,
Moustafa Abdel-Maksoud, Toshio Suzuki.
Proceedings 1st Symposium on Marine
Application of Computational Fluid Dynamics.
10. Mr P.R. Couser, Dr J.F. Wellicome and Dr
A.F. Molland (1996) An improved method for the
theoretical prediction of the wave resistance of
transom-stern hulls using a slender body approach.
11. Thomas Fu, Anna Karion, Anne Pence, James
Rice, Don Walker, Toby Ratcliffe (2005)
Characterization of the Steady Wave Field of the
High Speed Transom Stern Ship - Model 5365
Hull Form at Naval Surface Warfare Center
Carderock Division (NSWCCD), September 2005.
12. E.O. Tuck and D.C. Scullen (2002) A
Comparison of Linear and Nonlinear
Computations of Waves made by Slender
Submerged Bodies
13. E. O. Tuck, D. C. Scullen, and L. Lazauskas
(2002) Wave Patterns and Minimum Wave
Resistance for High-Speed Vessels 24th
Symposium on Naval Hydrodynamics Fukuoka,
JAPAN, 8-13 July 2002
14. S. W. W. Shor (1963) Mathematical methods for
calculating ship hull forms of decreased wave-
making resistance
15. David W. Taylor Naval Ship Research and
Development Center (1979) Proceedings of the
workshop on Ship wave-resistance computations.
16. Edwin Frank George van Daalen (1965)
Numerical and Theoretical Studies of Water
Waves and Floating Bodies.
17. David D. Moran (1971) Experimental procedure
for the determination of wave resistance in a
towing tank.
18. E.O. Tuck, D.C. Scullen and L. Lazauskas
(2001) Ship-wave patterns in the spirit of Michell.
19. E.O. Tuck, L. Lazauskas and D.C. Scullen
(1999) Sea wave pattern evaluation Part 1 Report:
Primary Code and Test Results (Surface Vessels)
20. Nastia Degiuli, Andreja Werner, Zdravko
Doliner (2003) Experimental determination of
wave pattern resistance of a trimaran
21. RonaldW. Yeung (2005) Interference Resistance
of Multi-Hulls per Thin-Ship Theory.
22. Francis Noblesse (2000) Velocity Representation
of Free-Surface Flows and Fourier-Kochin
Representation of Waves.
23. Justus Heimann, Bruce L. Hutchison, P.E.,
Benjamin J. Racine (2008) Application of the
Free-Wave Spectrum to Minimize and Control
Wake Wash.
24. A.W. Hughes, S.R. Turnock (1997)
Computational Fluid Dynamic investigation of
Hull-Waterjet flow interaction.
25. P.R. Couser, A.F. Molland, N.A. Armstrong,
I.K.A.P. Utama (1997) Calm water Powering
Predictions for high speed Catamarans.
26. NSWC Report, Wesley et al (2006), Comparison
of predicted and measured wave fields for model
5365 hull form.
27. NSTL Report, 2012, Wave pattern resistance
experiments on benchmark R/V Athena model.
AUTHORS BIOGRAPHY
Revathi Chavali holds the current position of Scientist C
at Indian Maritime University, Visakhapatnam and
involved in the R&D activities in the field of
hydrodynamics.
Md. Kareem Khan holds the current position of
Scientist C at Naval Science & Technological
Laboratory, Visakhapatnam and involved in the field of
Experimental Hydrodynamics.