Palestine Journal of Mathematics
Vol. 11(1)(2022) , 316–325 © Palestine Polytechnic University-PPU 2022
ANALYTICAL STUDY OF NONLINEAR ROLL MOTION OF
SHIPS: A HOMOTOPY PERTURBATION APPROACH
R. Umadevi, K. Venugopal, P. Jeyabarathi, L. Rajendran and M. Abukhaled
Communicated by Suheel Khoury
MSC 2010 Classifications: Primary 34A25, 65L05.
Keywords and phrases: Mathematical modelling, Nonlinear rolling, Damping and restoring moments, Homotopy pertur-
bation method, Analytical solution.
Abstract In this article, a mathematical model for a nonlinear roll motion of a ship is dis-
cussed and analytically solved. The model is a second-order nonlinear differential equation con-
taining nonlinear terms of damping and restoring moments. Under normal or irregular waves,
analytical expressions of roll angle, roll velocity, and moments of damping and restoring are
derived using a modified version of the homotopy perturbation method. The obtained analyti-
cal expressions are shown to be in satisfactory agreements with MATLAB generated numerical
simulations.
1 Introduction
For every ship, protection from capsizing is of great importance. One of the leading causes of
a ship capsizing in waves is the loss of stability due to nonlinear roll motion. Over the past few
decades, a significant effort has been made to develop a basic ship safety evaluation methodology
that is applicable during the design process and ship operation.
The roll motion of a ship is described by a second order differential equation, where damp-
ing coefficients can be linear or nonlinear. Typically, the linear model is used for low rolling
amplitudes, where the spectral analysis describes the problem in a frequency domain. However,
a large-amplitude ship movement is expected to result in a strongly nonlinear motion or even a
chaotic behavior, and in this case, the nonlinear rolling formula, which could be set for regular
or irregular waves, aims at foreseeing the ship’s nonlinear response. The ship-rolling problem
can be investigated in the frequency or time domains. However, the analysis in the time domain
is more feasible than that in the frequency domain, because in the latter case, there is a need for
conducting a significant number of realizations to determine the probability of capsizing. But
with the advancements in numerical simulations and probabilistic analysis, the development of
instructions for a ship handling based on risk has become possible [1].
Various viscous roll damping prediction methods have been applied by researchers over the
past few decades. Numerical methods for the simulation of ship motions are built in parallel
with the growing computing power. Mathematical techniques vary in complexity and ability to
compensate for different phenomena of flow, such as the two-phase flow, turbulence, and ship
motion [2]. Flazarano et al. [3] presented an overview of roll damping and viscous roll damping
prediction methods. Graham et al. [4] discussed the viscous damping prediction of large floating
bodies in waves. Seah et al. [5] have coupled a vortex method for simulating the roll decline of
a floating cylinder with a rigid body design. Several scientists have also studied the effects of a
moving ship’s nonlinear recovery moments. For example, Koskinen [6] discussed the numerical
simulation of ship motion due to waves and maneuvering. Ibrahim [7] presented the modelling
of a ship roll dynamics and its coupling with a heave and pitch. The oscillation of the roll as
one of the most important movements that can result in the boat being capsized has also been
investigated by researchers. Ibrahim [7] articulated the ships’ response through a linear formula
for small angles of roll motions. Unneland et al. [8] carried out theoretical studies on low-order
potential damping models for surface vessels. Perez et al. [9] presented a detailed simulation
model of the naval coastal patrol vessel. Wu and Sheu [10] obtained an exact solution for the
design of a ship hull’s heave and pitch movements, and Thuhad [11] investigated the coupled
NONLINEAR ROLL MOTION OF SHIPS 317
heave and pitch motions of a non-uniform hull on a still surface.
In this paper, a nonlinear model of roll motion of a ship in a time domain is discussed and an-
alytically solved by using a modified version of the homotopy perturbation method. In addition,
approximate analytical expressions for roll angle and velocity and the moments of damping and
restoring are also derived as they are essential parameters of ship dynamics. Numerical simula-
tions are performed to investigate the characteristics of the movements of ships and to validate
the derived analytical results.
2 Mathematical formulation of the problem
Using Newton’s law, the roll motion of a ship can be described by the following second-order
ordinary differential equation
Aẍ + B ẋ + Cx + f (t) = 0, (2.1)
where A, B and C represent the coefficients of inertia, damping and restoring terms, respectively.
The first term represents the force of inertia, and the second and third terms describe the moments
of damping and regeneration. Eq. (2.1) can be reduced to a single differential equation with one
degree of freedom as follows:
I θ̈ + B θ̇ + Cθ = Mθ (t), (2.2)
where the roll angle θ is the independent variable, I is the moment of inertia and B and C
are nonlinear functions representing the roll damping and restoring moments of the roll angle.
Figure 1 shows a general roll damping prediction model [2].
Figure 1: Design of the roll damping prediction system development.
3 Analytical expression of roll angle
As shown in Section 2, the governing equation of the roll motion is a nonlinear boundary value
problem for which no exact solution is known. There are many numerical schemes that can be
employed to find an approximate solution of the equation of roll motion such as, finite difference
methods [12], spline collocation methods [13], and wavelet-based methods [14]. But although
numerical solutions can be efficiently obtained, there are some pitfalls that make analytical so-
lutions more desirable by researchers. Of the most serious pitfalls that come with numerical
solutions, we mention instability and the difficulty of adjusting parameters to match the numer-
ical data. In addition, understanding the nonlinear behavior vessel rolling analysis can not be
achieved without analytical methods. Of the analytical methods that have proved to be effective
in solving nonlinear differential systems, we mention the variational iteration method [15, 16],
homotopy analysis method [17], Green’s function based method [18, 19], and differential trans-
formation method [20].
The homotopy perturbation method (HPM), which was first introduced by He [21] in 1999,
is a powerful and efficient method for finding analytical solutions of nonlinear differential equa-
318 R. Umadevi et al
tions. Of the many advantages of the HPM, we emphasize that HPM does not involve discretiza-
tion of the variables involved, and hence it is free from rounding off errors. In addition, the
HPM is applicable to various kinds of complicated nonlinear, stiff, delay, and fractional-order
differential equations. Accurate results can be obtained with just a few iterations, although more
iterations may be required in cases like strong oscillation models, which may be regarded as
a possible limitation of the method. For the past two decades, HPM has undergone various
modifications and has been successfully employed for finding reliable solutions for many non-
linear equations arising in physical, chemical, and engineering sciences [22–28]. In the next two
subsections, we show how a modified version of the HPM is employed to derive approximate
analytical expressions for roll angle in the case of linear damping and restoring moment with
and without wave exciting moment.
3.1 Linear damping and nonlinear restoring moments with wave exciting moment
The ship roll in a regular wave is described by the second order nonlinear equation [11]
θ̈(t) + d1 θ̇(t) + k1 θ(t) + k3 θ3 (t) + k5 θ5 (t) − a cos(ωt) = 0, (3.1)
where d1 is a relative damping coefficient, k1 , k3 and k5 are relative restoring coefficients, a is
the excitation amplitude and ω is the frequency. The initial conditions are given by
θ|t=0 = l, θ̇|t=0 = 0. (3.2)
We construct the homotopy for Eq. (3.1) as follows (see Appendix A)
(1 − p)[θ̈(t) + d1 θ̇(t) + k1 θ(t)]
(3.3)
+p[θ̈(t) + d1 θ̇(t) + k1 θ(t) + k3 θ3 (t) + k5 θ5 (t) − a cos(ωt)] = 0
where p ∈ [0, 1] is an embedding parameter. The approximate homotopy solution is expressed
in the series form
θ = θ0 + pθ1 + p2 θ2 + · · · (3.4)
Substituting (3.4) into (3.3) and equating like powers of p leads to the linear system
p0 : θ̈0 + d1 θ̇0 + k1 θ0 (t) = 0 (3.5)
p1 : θ̈1 + d1 θ̇1 + k1 θ1 (t) + k3 θ03 (t = 0) + k5 θ05 (t = 0) − a cos(ωt) = 0 (3.6)
..
.
Subject to initial boundary conditions
θ0 (0) = l, θ̇0 (0) = 0, (3.7)
θ1 (0) = l, θ̇1 (0) = 0, (3.8)
..
.
Solving system (3.5)-(3.8) gives
d1 t
n γt ld γt o
1
θ0 (t) = e− 2 l cos + sin , (3.9)
2 γ 2
d1 t
n γt d γt o
1
θ1 (t) = e− 2 (α − β (k1 − ω 2 )) cos + (α − β (k1 + ω 2 )) sin (3.10)
2 γ 2
−α + β (d1 ω sin(ωt) + (k1 − ω 2 ) cos(ωt)),
..
.
The two-term HPM solution for the roll angle is now obtained from Eq. (3.4) by taking the limit
as p → 1. That is,
d1 t
n γt d γt o
1
θ(t) =e− 2 (l + α − β (k1 − ω 2 ) cos + (l + α − β (k1 + ω 2 )) sin
2 γ 2 (3.11)
2
− α + β (d1 ω sin(ωt) + (k1 − ω ) cos(ωt)).
NONLINEAR ROLL MOTION OF SHIPS 319
By direct differentiation with respect to time, we obtain the explicit velocity expression
( )
2
dθ 2 d1 2 γ
d1 t
γt
=e 2 − (l + α − β (k1 + ω )) + (l + α − β (k1 − ω )) sin
dt 2γ 2 2
(3.12)
d1 t
γt
− βd1 ω 2 e 2 cos + β (d1 ω 2 cos(ωt) − ω (k1 − ω 2 ) sin(ωt)),
2
and the explicit acceleration expression
(
d2 θ d1 t βd21 ω 2
d2 γ 2 γt
=e 2 − − ((l + α − β(k1 + ω 2 )) 1 − ((l + α − β(k1 − ω 2 )) cos
dt2 2 4 4 2
)
βd ω 2 γ d3 d γ γt
1 1
+ + ((l + α − β(k1 + ω 2 )) 1 + ((l + α − β(k1 ω 2 )) sin
2 4γ 4 2
− βω 2 (d1 ω sin(ωt) + (k1 − ω 2 ) cos(ωt)),
(3.13)
where
k3 l3 + k5 l5 a
q
α= , β= 2 2 , γ= 4k1 − d21 . (3.14)
k1 d1 ω + (k1 − ω 2 )2
Now the analytical expressions for the restoring and damping moments are, respectively,
given by
Restoring moment = cθ = k1 θ(t) + k3 θ3 (t) + k5 θ5 (t), (3.15)
and
Damping moment = B θ̇(t) = d1 θ̇(t), (3.16)
where θ(t) and θ̇(t) are given by Eqs. (3.11) and (3.12).
3.2 Linear damping and nonlinear restoring moments without wave exciting moment
The nonlinear equation representing a ship without wave exciting moment is given by
θ̈(t) + d1 θ̇(t) + k1 θ(t) + k3 θ3 (t) + k5 θ5 (t) = 0, (3.17)
where d1 , k1 , k3 and k5 are as given in Eq. (3.1) subject to initial conditions (3.2). Following same
procedure as in section 3.1, the homotopy perturbation method gives the following analytical
expression for the roll angle
d1 t
n γt d γt o
1
θ(t) = e− 2 (l + α) cos + (l + α) sin − α, (3.18)
2 γ 2
where α and γ are as given in Eq.(3.14). The roll velocity and acceleration are, respectively
given by
dθ d1 t
n d2 γ γt o
= e− 2 (l + α) − 1 − sin , (3.19)
dt 2γ 2 2
d2 θ d t
n −d2 γ d1 γt γ γt o
− 21 1
= e (l + α ) − − sin + cos . (3.20)
dt2 2γ 2 2 2 2 2
4 Results and discussion
The following experimental values of parameters [1] will be used to test the accuracy of the
derived analytical expressions for the roll angle, velocity and acceleration with and without wave
exciting moments.
d1 = 0.0126513, k1 = 0.67199703, k3 = −0.5392039, k5 = −0.086792, a = 0.1,
(4.1)
ω = 0.1, l = 0.3.
320 R. Umadevi et al
Using these parameters, the simplified roll angle, velocity and acceleration in the presence of
wave exciting moment are, respectively, given by
θ(t) = e−0.00633t (0.127 cos(0.81973t) + 0.00094 sin(0.81973t))
(4.2)
+ 0.00029 sin(0.1t) + 0.1511 cos(0.1t) + 0.02198,
dθ
= e−0.00633t (−0.00029 cos(0.81973t) − 0.10408 sin(0.81973t))
dt (4.3)
+ 0.000029 cos(0.1t) − 0.01511 sin(0.1t),
d2 θ(t)
= (−0.08532 cos(0.81973t) + 0.000682 sin(0.81973t))e−0.00633t
dt2 (4.4)
−6
− 0.001511 cos(0.1t) − 2.88683 × 10 sin(0.1t).
In the absence of wave exciting moments, the roll angle, velocity and acceleration are, re-
spectively
θ(t) = 0.021979 + (0.002145 sin(0.81973t) + 0.27802 cos(0.81973t))e−0.0063t (4.5)
dθ
= −0.22792e−0.00632565t sin(0.81973t) (4.6)
dt
d2 θ(t)
= (0.001442 sin(081973t8062t) − 0.18683 cos(0.81973t))e−0.006326t (4.7)
dt2
Tables 1 and 2 show that the analytical and numerical solutions for the roll angle and the
velocity are in a strong agreement when t is sufficiently close to the initial value. Figures 2-5
show that although as time increases the analytical curves for the roll angle and velocity deviate
from the numerical curves but they are still satisfactorily close to them. It is also noted that all
analytical curves maintain stability over a very large time domain.
Table 1: Comparison between numerical and analytical results for roll angle and velocity with
wave exciting moment
Roll angle θ (rad) Velocity θ̇ (cm/s)
Time (s) analytical numerical abs. error analytical numerical abs. error
0.0 0.30000 0.30000 0 0 0 0
0.1 0.29957 0.29957 0 -0.00867 -0.00867 0
0.2 0.29827 0.29827 0 -0.01727 -0.01729 0.00002
0.3 0.29612 0.29611 0 -0.02574 -0.02580 0.00006
0.4 0.29312 0.29311 0.00001 -0.03404 -0.03417 0.00013
0.5 0.28932 0.28928 0.00003 -0.04210 -0.04236 0.00026
0.6 0.28472 0.28465 0.00006 -0.04987 -0.05031 0.00044
0.7 0.27935 0.27923 0.00012 -0.05730 -0.05799 0.00069
0.8 0.27327 0.27306 0.00021 -0.06435 -0.06535 0.00100
0.9 0.26650 0.26617 0.00033 -0.07096 -0.07235 0.00139
1.0 0.25909 0.25861 0.00049 -0.07710 -0.07895 0.00185
NONLINEAR ROLL MOTION OF SHIPS 321
(a) With wave exciting moment (Eq. (4.2)) (b) Without wave exciting moment (Eq. (4.5))
Figure 2: Analytical and numerical curves of roll angle in the case of linear damping.
(a) With wave exciting moment (Eq. (4.3)) (b) Without wave exciting moment (Eq. (4.6))
Figure 3: Analytical and numerical curves of velocity in the case of linear damping.
Table 2: Comparison between numerical and analytical results for roll angle and velocity without
wave exciting moment
Roll angle θ (rad) Velocity θ̇ (cm/s)
Time (s) analytical numerical abs. error analytical numerical abs. error
0.0 0.30000 0.30000 0 0 0 0
0.1 0.29907 0.29907 0 -0.01865 -0.01865 0
0.2 0.29627 0.29627 0 -0.03715 -0.03719 0.00004
0.3 0.29165 0.29163 0.00002 -0.05538 -0.05550 0.00012
0.4 0.28521 0.28518 0.00003 -0.07321 -0.07350 0.00028
0.5 0.27702 0.27695 0.00007 -0.09053 -0.09108 0.00054
0.6 0.26713 0.26698 0.00014 -0.10722 -0.10814 0.00091
0.7 0.25560 0.25534 0.00026 -0.12317 -0.12458 0.00140
0.8 0.24252 0.24209 0.00043 -0.13828 -0.14029 0.00201
0.9 0.22798 0.22731 0.00066 -0.15243 -0.15518 0.00275
1.0 0.21207 0.21109 0.00098 -0.16555 -0.16914 0.00359
Figure 2 shows the roll decay of a ship motion from an initial angle together with the main
properties of damped roll motion. As a ship performs harmonic roll oscillation from rest with
an initial roll angle and an initial velocity, then the roll damping can be measured from the
decrement of each successive roll angle maxima [2].
Figure 3 depicts the analytical curves of velocity with and without wave exciting moments.
It is inferred from this figure that the amplitude velocity of linear damping is smaller in the
presence of exciting wave moments. Figure 4, which portrayed the acceleration curves versus
time, shows that the amplitudes of both curves decrease as time increases.
322 R. Umadevi et al
(a) With wave exciting moment (Eq. (4.4)) (b) Without wave exciting moment (Eq. (4.7))
Figure 4: Analytical acceleration curve in the case of linear damping.
(a) With wave exciting moment (b) Without wave exciting moment
Figure 5: Analytical curve of restoring moment in the case of linear damping (Eq. (3.15)).
(a) With wave exciting moment (b) Without wave exciting moment
Figure 6: Analytical curve of damping moment in the case of linear damping (Eq. (3.16)).
NONLINEAR ROLL MOTION OF SHIPS 323
The analytical curves of the restoring and damping moments, which are used to measure the
bilge keel height, are given in Figures 5 and 6, respectively. From these figures, it is noted that
the period of oscillation is not constant but is slowly increasing with time. Using the envelope
d1 t
function f = ±θ(t = 0)e− 2 , the damping coefficient can be obtained.
5 Conclusions
In this paper, a theoretical model describing the nonlinear roll motion of a ship was discussed.
A modified version of the homotopy perturbation method was employed to derive an analytical
solution of the model that is represented by a nonlinear time-dependent differential equation.
Approximate analytical expressions of the roll angle, velocity, acceleration, and resorting and
damping moments for all possible values of parameters were obtained. The derived analytical
results were shown to maintain a satisfactory agreement with numerical simulation throughout
the time domain. These analytical results can play an essential role in validating the experimental
and numerical results, such as finding a roll decay test and improve the level of design and safety
of a ship. The methodology presented in this paper can be intrinsically extended to solve other
nonlinear systems in various fields of science and engineering.
Appendix A The basic idea of the homotopy perturbation methd
Consider the nonlinear differential equation
A(u) − f (r) = 0, r ∈ Ω, (Appendix A.1)
with the boundary condition
du
B u, = 0 , r ∈ Γ, (Appendix A.2)
dr
where A, B, f (r) and Γ are a general differential operator, a boundary operator, a known analyt-
ical function and the boundary of the domain Ω , respectively. Expressing A(u) as the sum of
linear (L) and nonlinear (N ) parts, Eq. (Appendix A.1) becomes
L(u) + N (u) − f (r) = 0. (Appendix A.3)
The homotopy technique begins by defining v (r, p) : Ω × [0, 1] → R, such that
H (v, p) = (1 − p)[L(v ) − L(u0 )] + p[A(v ) − f (r)] = 0, (Appendix A.4)
where p ∈ [0, 1] is an embedding parameter and u0 is an initial approximation of Eq. (Appendix
A.1) that satisfies boundary conditions (Appendix A.2). Evidently, Eq.(Appendix A.4) implies
that
H (v, 0) = L(v ) − L(u0 ) = 0, (Appendix A.5)
H (v, 1) = A(v ) − f (r) = 0. (Appendix A.6)
As p changes from 0 to 1, v (r, p) changes from u0 to ur , a process known as a homotopy. The
solution of Eq. (Appendix A.4) may be expressed in terms of a power series in the form:
v = v0 + pv1 + p2 v2 + · · · . (Appendix A.7)
With p = 1, an approximate solution to Eq. (Appendix A.4) is given by:
u = lim v = v0 + v1 + v2 + · · · . (Appendix A.8)
p→1
324 R. Umadevi et al
References
[1] M.P. Buča, I. Senjanović, Nonlinear ship rolling and capsizing, Polish Maritime Research 57, 321–331
(2006).
[2] H.P. Piehl, Ship Roll Damping Analysis, PhD Thesis, University of Duisburg-Essen (2016).
[3] J. Falzarano, A. Somayajula, R. Seah, An overview of the prediction methods for roll damping of ship,
Ocean Systems Engineering 5, 55–76 (2015).
[4] J. Graham, S. Sherwin, T. Kendon, The prediction of viscous damping of large floating bodies in waves,
20th International Workshop on Water Waves and Floating Bodies Spitzbergen (2005).
[5] R. Seah, T. Celano, R. Yeung, The roll decay of floating cylinders with bilge keels, Mechanical Engineer-
ing and Ocean Engineering University of California at Berkeley (2002).
[6] K. Koskinen, Numerical simulation of ship motion due to waves and manoeuvring, Degree project in
Naval Architecture Second cycle Stockholm , Sweden (2012).
[7] R. Ibrahim, I. Grace, Modelling of ship roll dynamics and its coupling with heave and pitch, Mathematical
Problems in Engineering 2010 Article ID 934714, (2010), https://doi.org/10.1155/2010/934714
[8] K. Unneland, T. Fossen, P. Van Dooren, O. Egeland, Low order potential damping models for surface
vessels, In the Seventh IFAC Conference on Maneuvering and Control of Marine Craft, September 20-22,
Lisbon (2006).
[9] T. Perez, A. Ross, T.I. Fossen, A 4-dof Simulink model of a coastal patrol vessel for Maneuvering in
waves, In the Seventh IFAC Conference on Maneuvering and Control of Marine Craft, September 20-22,
Lisbon (2006).
[10] J. Wu, J. Sheu, An exact solution for a simplified model of the heave and pitch motions of a ship hull due
to a moving load and a comparison with some experimental results, Journal of Sound and Vibration 192,
495–520 (1996).
[11] A.M. Thu, E.E. Htwe, H.H. Win, Mathematical modeling of a ship motion in waves under coupled mo-
tions, Int. J. of Eng. Appl. Sci. 2, 97–102 (2015).
[12] I. Tirmizi, E. Twizell, Higher-order finite-difference methods for nonlinear second-order two-point
boundary-value problems, Appl Math Lett. 5, 897–902 (2002).
[13] M Abukhaled, S Khuri, A Sayfy, Spline-based numerical treatments of Bratu-type equations, Palestine J.
Math 1, 63–70 (2012).
[14] T Abualrub, M Abukhaled, B Jamal, Wavelets approach for the optimal control of vibrating plates by
piezoelectric patches, J. of Vibration and Control 24, 1101–1108 (2018).
[15] J.H. He, Variational iteration method-some recent results and new interpretations, J Comput Appl Math
207, 3–17 (2007).
[16] M. Abukhaled, Variational iteration method for nonlinear singular two-point boundary
value problems arising in human physiology, J. of Mathematics 2013 ID 720134 (2013),
https://doi.org/10.1155/2013/720134
[17] S.J. Liao, Homotopy analysis method in non-linear differential equations, Springer and Higher Education
Press, Heidelberg (2012).
[18] M. Abukhaled, S. Khuri, An efficient semi-analytical solution of a one-dimensional curvature equation
that describes the human corneal shape, Mathematical and Computational Applications 24, 8 (2019).
[19] M. Abukhaled, Green’s function iterative method for Solving a class of boundary value problems arising
in heat transfer, Applied Mathematics and Information Sciences 11, 229–234 (2017).
[20] I. Hassan, Application to differential transformation method for solving systems of differential equations,
Applied Mathematical modelling 32, 2552–2559 (2008).
[21] J.H. He, Homotopy perturbation technique, Comp. Meth. Appl. Mech. Eng. 178, 257–262 (1999).
[22] J.H. He, A coupling method of a homotopy technique and a perturbation technique for non-linear prob-
lems, Int. J. Non-Linear Mech. 35, 37–43 (2000).
[23] J.H. He, The homotopy perturbation method for nonlinear oscillators with discontinuities, Appl. Math.
Comp. 156, 591–596 (2004).
[24] J. Biazar, H. Ghazvini, Homotopy perturbation method for solving hyperbolic partial differential equa-
tions, Comput. Math. Appl. 56, 453–458 (2008).
[25] R. Swaminathana , K. Venugopalb, M. Rasic , M. Abukhaled, L. Rajendrane, Analytical expressions
for the concentration and current in the reduction of hydrogen peroxide at a metal-dispersed conducting
polymer film, Quim. Nova 43, 58–65 (2020).
[26] R. Saravanakumar, P. Pirabaharan, M. Abukhaled, L. Rajendran, Theoretical analysis of voltammetry at a
rotating disk electrode in the absence of supporting electrolyte, Phys. Chem. B 124, 443–450 (2020).
NONLINEAR ROLL MOTION OF SHIPS 325
[27] L. Cveticanin, Homotopy perturbation method for pure nonlinear differential equations, Chaos, Solitons
and Fractals 30, 1221–1230 (2006).
[28] Sven Wassermann, Dag-FrederikFeder, Moustafa Abdel-Maksoud , Estimation of ship roll damping-A
comparison of the decay and the harmonic excited roll motion technique for a post panamax container
ship, Ocean Engineering 120, 371–382 (2016).
Author information
R. Umadevi, K. Venugopal, P.G. Research Department of Mathematics, Government Arts College (Affiliated
to Bharathidasan University), Tiruchirappali, India.
E-mail: [email protected], [email protected]
P. Jeyabarathi, L. Rajendran, Department of Mathematics, AMET (Deemed to be University), Tamil Nadu,
India.
E-mail: [email protected], [email protected]
M. Abukhaled, Department of Mathematics, American University of Sharjah, PO Box 26666, Sharjah, U.A.E..
E-mail: [email protected]
Received: July 22, 2020.
Accepted: November 19, 2020.