Seaworthiness of Resonance-Free SWATH

as an oceangoing fast ship

M. Yoshida
Kyushu University, Japan

H. Iwashita
Hiroshima University, Japan

H. Kihara
National Defense Academy of Japan, Japan

T. Kinoshita
The University of Tokyo, Japan

ABSTRACT: The speed reduction, additional resistance or slamming, which are caused by the large
amplitude of ship motions, should be restricted completely for oceangoing large fast ship, because of the
strict time-punctuality and high value of the cargo. A Resonance-Free SWATH (RFS) as the oceangoing
large fast ship has the negative restoring moments, which leads to resonance-free in the motion responses,
because of the extraordinary small water plane area. The RFS is designed to cross 4,800 nautical miles of
Pacific Ocean in 5 days punctually at a high speed of 40 knots, with the good seaworthiness such as no speed
reduction or absolutely no slamming even in the rough sea. To verify the seaworthiness of the RFS,
experiments in model basin and theoretical predictions are carried out to examine the lowest limit of motion
responses in waves. The results in regular head waves are compared with those of various hull forms, such as
mono-hull, ordinary SWATH or trimaran. The predominance of the RFS regarding seaworthiness will be
pointed out in the conclusion. For example, by using PD control action, heave motion responses of the RFS is
reduced to about 1/40 compared with those of mono-hull or trimaran, and pitch motion responses of the RFS
become about 1/8.

INTRODUCTION

The developments of fast ships in various hull forms

such as mono-hull, catamaran and trimaran are very
active worldwide nowadays. Above all, the research
and development of the oceangoing large fast ship is
an important subject.
It is supposed that the perfect accuracy of navigation
time schedule and delicate handling are required for
the fast ships to transport the high-valued cargo even
in the rough sea. Accordingly, the speed reduction,
additional resistance or slamming, which are caused
by the large amplitude of ship motions, should be
restricted completely. The objective of this study is
IX HSMV Naples 25 - 27 May 2011

the conceptual design of oceangoing large fast ship,

which has 40 knots speed, 5,000-10,000 tons
payload, especially has the good sea-keeping quality
such as no speed reduction and absolutely no
slamming in the waves of sea state 7 (with
significant wave height of 6-9 meters).
In this study, a Resonance-Free SWATH (RFS)
ship is introduced as an example of the oceangoing
large fast ship. Some results of experiments and
theoretical predictions regarding the motion
responses of RFS in waves using small controllable
fins are presented. First, theoretical estimation and
experiments of PD control for RFSs motion are
1

discussed. Where P indicates the proportional

control action and D denotes the derivative control
action. Secondly, the results of motion responses of
RFS in regular head waves are compared with other
hull forms such as mono-hull, ordinary SWATH or
trimaran.

K a

Comparing the motion amplitude between ships with

and without restoring force or moment, it is
recognized that the latter has no resonant peak and
its response amplitude is smaller than the former as
shown in Figure 1. The ship without restoring
moments, especially in the case of pitching, can be
realized by means of extremely small water plane
area compared with ordinary SWATH as shown in
Figure 2. Consequently, the ship has no resonance in
pitch response, which is called as a Resonance-Free
SWATH (RFS) in this study.

K : wave number

a : wave amplitude
RM : restoring moment

RM 0

CONCEPTUAL DESIGN OF RFS

2.1 Design policy of the ship form

RM > 0

: wave frequency
: motion amplitude

Figure 1 Resonance Amplitude Operator (RAO)

Extremely small water plane area

Velocity

Figure 2 Water plane area of RFS

2.2 Summary of conceptual design for RFS

The overall appearance of the conceptual design1), 2)
for RFS is shown in Figure 3.
The resistance components of the ship are estimated
as follows: Frictional resistance is determined by
using Schoenherrs coefficient for equivalent plate.
Wave making resistance is estimated by means of
Michells thin ship theory for the strut and
singularity distribution method for the lower hull.
Viscous pressure resistance is considered as
correction term from real ship data. As a result, total
resistance of the ship equals 810 tf.
The calculation concerning the structural strength is
principally carried out under the condition of regular
wave with a wave height of 10.8 m. This is
equivalent to the 1/1000 maximum expected value
of the sea state with a significant wave height of 6 m.
Head and beam seas are selected as the wave
directions. Normal and shear stress acting on the
three parts, i.e. the strut end of the lower hull (the
root of the overhang portion of the lower hull), the
upper deck connection (central cross section of the
upper deck) and the connecting part between the
upper deck and the strut, are calculated respectively.
Wave loads such as pitch moment, yaw moment,
split force and split force moment are considered as
the external load condition in the strength
calculation. According to the calculation,the
maximum principal stress is 24.4 kgf/mm2, which is
well within the acceptable limits for a high strength
steel of 70 kgf/mm2 yield strength, where the
IX HSMV Naples 25 - 27 May 2011

Figure 3 Overall appearance of RFS

Table 1 Principal particulars of RFS

Displacement tonnage24,000 t
Light weight :10,367 t
Power plants : 3,157 t
Dead weight : 13,633 t
Lighter : 1,000 t
Payload5,400 t, 540 containers (40 ft )
Fuel : 6,833 t
Upper hull : length:200 m, breadth:55 m
Lower hull : length:230 m, maximum diameter:8.85 m
Strut : length:90 m, maximum breadth:4.425 m
Draft : 12.85 m
Speed : 40 knots
Resistance : 810 tf
Main engine : 8 Gas turbines (44,000 Ps), Total 352,000 Ps
Propulsion : 8 Contra-rotating propellers
Cruising distance : 4,800 nautical miles (Pacific Ocean)
Controlling fin : 8 Fins, Total fin area 160 m2

thickness of plate at the lower hull, the upper deck or

the strut is determined as 40 and 20 mm, 16 mm and
19 mm, respectively.
Four pairs of controlling fins are installed near the
bow and stern of lower hulls. Each fin has an area of
20 m2 respectively. To maintain the stability and
superior sea-keeping quality of RFS even effectively
in the rough sea, these fins should operate at one
meter below the wave surface.
Consequently, the conceptual design of RFS is as
shown in Table 1 synthetically. RFS has the
capability of crossing 4,800 nautical miles of Pacific
Ocean within 5 days with a payload of 5,400 tons at
a high speed of 40 knots, with total engine power of
352,000 Ps.

F33 = F33 f + F33 a

F53 = F33 f l f F33a l a
F55 = F55 f + F55 a
F35 =

F55 f
lf

(2)

F
+ 55 a
la

and
1
F33 f ,a = A f ,aC L 3 f ,a C3 f ,a ( )U 2
2
1
F55 f ,a = A f ,aCL 5 f ,aC5 f ,a ( )U 2l f ,a
2

(3)

where,
C jf ,a ( ) = Ccjf ,a ( ) + i Csjf ,a ( )

3 THEORY OF PD CONTROL

(4)

3.1 Motion equations of ship

A Cartesian coordinate system O-xyz that follows
ship forward speed U is adopted to describe the
problem. The O-xy plane coincides with the
undisturbed free surface while the z axis is pointing
upward and passes through the gravity center G of
the ship model.
The ship model has 8 fins. Each fin has the
configuration such as plane area A= 0.001518 m2,
chord length c=0.0357 m (base side), 0.0278 m (tip
side), span s=0.0478 m, aspect ratio s2/A=1.51 and
symmetry wing profile NACA0012.
Equations of coupled motion in heave z and
pitch directions including controlling forces and
time lag of control system are shown as follows:
( m + A33 ) &z& +B33 z& + C33 z + A35 && + B35 & + C35
= E3 + F33 c 3 + F35 c 5
&
&
&
( I + A55 ) +B55 + C55 + A53 &z& + B53 z& + C53 z
= E5 + F55 c 5 + F53 c 3

(1)

T1 &&c 3 + T2 & c 3 + c 3 = ( K D 3 z& + K P 3 z )

T && + T & + = ( K & + K )
1

c5

c5

c5

D 5

P 5

where, controlling targets equal z=0, =0, m

indicates mass of model, I denotes inertia moment,
Aij, Bij and Cij are added mass, damping coefficient
and restoring force or moment, Ei is wave exciting
force or moment, Fij describes lift characteristics, cj
indicates attack angle of fin, T1, T2 describe dynamic
characteristics of second order time lag in fin
control system and KPj or KDj denotes P or D
control gain constant which is reduced to attack
angle base.

IX HSMV Naples 25 - 27 May 2011

and subscript f or a indicates fore or aft fin,

l f ,a

denotes moment lever of fin, is density of water,

Af,a is total fin area of fore or aft fin, U indicates ship
speed. Also, CLjf,a denotes quasi-steady lift-curve
slope of fore or aft fin in heave or pitch motion
control and Cjf,a describes interaction among fins and
lower hulls, and unsteady characteristics (time lag of
lift generation).
The condition of fin control is that attack angle of
fore and aft fin is the same in the same direction for
heave motion control while is the same in the
inverse direction for pitch motion control.
As explained previously, it is summarized that time
lag and interaction among fins and lower hulls
regarding fin lift generation, controlling force due to
fin lift and time lag of control system are considered
in the motion equations.
In equation (1), hydrodynamic forces based on
potential flow are calculated by strip method or
various 3D method, hydrodynamic forces due to
viscosity are obtained by application of Lee et al.s
method 3) and hydrodynamic forces due to fin lift are
obtained by the method shown in section 3.2.
3.2 Unsteady characteristics of fin lift
It is supposed that wave damping due to fins can be
neglected as fin depth is larger than chord length of
fin and Froude number (=3.42) calculated by the use
of chord length is large.
When AFij or BFij indicates hydrodynamic force
coefficient due to fin, the imaginary part of damping
coefficient BFij is transferred to AFij, because of time
lag of fin lift as shown in equation (4). Accordingly,
AFij and BFij are expressed as follows:
In the case of heave motion,
3

AF 33 = AF 33 f + AF 33a
+

1 1
1
( Af UCL 3 f Cs 3 f + AaUCL 3a Cs 3a )
2
2
= A33 wif A33 wof

BF 33 =

1
1
Af UCL 3 f Cc 3 f + AaUCL 3a Cc 3a
2
2
= B33 wif B33 wof

AF 53 = AF 33 f l f AF 33a l a

(a) Mono-hull

(5)

1 1
1
( Af U l f CL 3 f Cs 3 f + AaU l aCL 3 a Cs 3a )
2
2
= A53 wif A53 wof
1
1
BF 53 = Af U l f CL 3 f Cc 3 f AaU l aCL 3a Cc 3 a
2
2
= B53 wif B53 wof

CL3fCc3f, CL3fCs3f and CL3aCc3a, CL3aCs3a can be

obtained from equation (5). On the other hand, in the
case of pitch motion,
AF 55 = AF 33 f l f + AF 33a l a
2

(b) Ordinary SWATH

1 1
1
2
2
( Af Ul f CL 5 f Cs 5 f + AaUl a CL 5 a Cs 5 a )
2
2
1 1
1
+ 2 ( Af U 2l f CL 5 f Cc 5 f + AaU 2l aCL 5 a Cc 5 a )
2
2
= A55 wif A55 wof
+

1
1
Af Ul f 2CL 5 f Cc 5 f + AaUl a 2CL 5 a Cc 5 a
2
2
1 1
1
( Af U 2l f CL 5 f Cs 5 f + AaU 2l aCL 5 a Cs 5 a )
2
2
= B55 wif B55 wof

BF 55 =

(6)

(c) Trimaran

AF 35 = AF 33 f l f AF 33a l a
1 1
1
( Af U l f CL 5 f Cs 5 f + AaU l aCL 5 a Cs 5 a )
2
1 1
1
2
2 ( Af U CL 5 f Cc 5 f + AaU 2CL 5 a Cc 5 a )
2
2
= A35 wif A35 wof

1
1
BF 35 = Af U l f CL 5 f Cc 5 f AaU l aCL 5 a Cc 5 a
2
2
1 1
1
2
+ ( Af U CL 5 f Cs 5 f + AaU 2CL 5 a Cs 5 a )
2
2
= B35 wif B35 wof

In a similar way, CL5fCc5f, CL5fCs5f and CL5aCc5a,

CL5aCs5a are obtained. Here mass of fin and viscous
force due to fin are neglected. Aijwif, Bijwif or Aijwof,
Bijwof denote added mass, damping coefficient of
RFS with fins or without fins, respectively. Aijwif,
Bijwif or Aijwof, Bijwof can be calculated by using 3D
Green function method or Rankine panel method,
and also can be measured by experiments.

IX HSMV Naples 25 - 27 May 2011

(d) RFS
Figure 4 General view of four kinds of hull forms

4 EXPERIMENTS
4.1 Models of four hull forms

Table 2 Principal particulars of four models

restoring moments because of the extremely small

water plane area.
4.2 Experiments of controllable fins
Assembling drawing of movable fins is shown in
Figure 5. Four controlling equipments of fins are
installed in the bow and stern ends of both lower
hulls, with a diameter of about 40 mm. Controlling
equipment consists of DC servomotor, worm gear,
fin axes and potentiometer principally. The attack
angle of four pairs of movable fin equipment can be
controlled independently. Maximum amplitude of
attack angle of each fin is designed as 20 deg,
maximum frequency of fin oscillation equals 3.0 Hz.
The instructed value cf,a of fin attack angle is
calculated according to equation (7).
cf ,a = c 3 f ,a m c 5 f ,a =

K D 3 z& + K P 3 z K D 5& + K P 5
m
4 ACLU 2
4l 0 ACLU 2

(7)

where the right-hand side of equation (7) is - at fore

fin (abbreviated as f) or + at aft fin (abbreviated as a),
CL indicates lift-curve slope (=3.12 1/rad3), 4)) and
l 0 denotes moment lever of fin (=0.8333 m).
Phase lag of fin control system between the output
value of potentiometer (feedback heave or pitch
signal) and the output value of fin actuator (actual
attack angle) are measured. The result in the case of
1.105 Hz (corresponds to /L=2.00) equals about 26
deg. Phase lag of 26 deg is equivalent to time
constant of about 0.07 sec in control system.
Figure 5 Plan of controllable fins installed in RFS model

Experiments to compare motion responses in waves

among four hull forms, i.e. mono-hull, ordinary
SWATH, trimaran and RFS, are carried out in model
basins.
The general view of four ship models is shown in
Figure 4.
Also principal particulars of these four models are
shown in Table 2 with items of length L, breadth B,
draught d, longitudinal center of buoyancy l cb , water
plane area Aw, height of gravity center KG,
longitudinal metacentric height GML, radius of
gyration yy/L, mass V, advancing speed U or
Froude number Fn where indicates density of
water and V denotes displaced volume of the model.
The detail of experiments is as discussed in the
previous study5). Ordinary SWATH and RFS have
the same lower hulls but different strut length and
consequently different parameters of Aw, GML, V
and so on. The strut length of RFS is equal to 0.783
m, approximately one third of the strut length 2.0 m
of ordinary SWATH. So, RFS model has negative
IX HSMV Naples 25 - 27 May 2011

4.3 Experiments of PD control

According to findings from previous study2),
minimum P gain constant and maximum D gain
constant should be adopted for the PD control of
RFS motions.
Impulse response experiments of RFS model with
controllable fins running at Fn=0.43 are carried out
in still water to obtain the maximum stable gain.
The discriminant for the maximum stable D gain is
performed systematically. The model starts to run at
a pitch attitude of =+ or -2 deg. During the tests, it
is checked whether the model can be controlled well
and the horizontal attitude can be kept when running
at Fn=0.43. In practice, the failure of control system
is not only the problem of divergence but also
includes the hunting problem, i.e. a phenomenon
relating to the high oscillation of control fins.
Consequently, the maximum stable D gain is usually
decided at the turning point of hunting.
4.4 The maximum stable gain for ordinary SWATH
and RFS

5 RESULTS AND DISCUSSION

5.1 Effect of D gain on motion responses

unstable
stable

Figure 6 Maximum stable gain rate for ordinary SWATH

and RFS

Table 3 Gain constants for ordinary SWATH and RFS

Results of experiments and theoretical estimations of

discriminant6) are shown in Figure 6. The abscissa of
gain rate indicates magnification ratio of
fundamental D gain constant as shown in Table 3,
which is defined as gain GA, while the ordinate
denotes control stability as described above and the
value greater than 1.0 means that control system is
unstable. From Figure 6, it can be seen that
experimental results agree well with theoretical
estimation and the maximum stable gain rate is
determined as 1.0, i.e. gain GA, for RFS.
Gain rate 1.0 for RFS is determined as described
above, on the other hand, maximum stable gain rate
0.6 for ordinary SWATH is selected as shown in
Figure 6 as the result of the same experiments as
RFS. The reason to choose different values of gain
rate, i.e. 1.0 and 0.6 for these two hull forms, arises
from that the energy accumulated by the attack angle
of fin is dissipated easily in the case of RFS because
of the large damping coefficients of ship hull
compared with those of ordinary SWATH. In the
experiments of motion responses, fundamental PD
gain constants adopted for ordinary SWATH or RFS
are as shown in Table 3. In the table, OGA, in which
D gain rate equals 0.6, indicates the gain constants
for ordinary SWATH while GA, in which D gain
rate is 1.0, denotes the gain constants for RFS,
where P gain constant of pitch motion for ordinary
SWATH is adopted as KP5=0 because ordinary
SWATH has enough positive restoring moments.
IX HSMV Naples 25 - 27 May 2011

First, the experimental results regarding the effect of

D gain constant on motion responses of ordinary
SWATH are shown in Figure 7. The gain constants
tested in experiments are listed in Table 4. In the
table OGA, OGC2 and OGC indicate gain rate 0.6,
0.8 and 1.0 respectively. The experiments in the case
of OGC2 or OGC are carried out in spite of attack
angle fluctuations. The abbreviation Ord stands for
ordinary SWATH with fixed fins which is not
controlled by PD and the symbol Cal denotes the
calculated results by means of strip method in the
figure. It is observed that the heave and pitch motion
responses of ordinary SWATH is reduced
considerably by using D control gain and the effect
of that is saturating near gain OGA, and also phase
differences of each experiment agree well with each
other.
Secondly, the experimental results regarding the
effect of D gain constants on motion responses of
RFS are shown in Figure 8 and the gain constants
tested are listed in Table 5. In the table, GB, GA and
GC denote gain rate 0.5, 1.0 and 1.2 respectively.
Also the experiments in the case of GC are carried
out in spite of attack angle fluctuations. It can be
seen that there exists obviously the effect of D gain
on the reduction of heave and pitch motion
responses, and the effect of that is saturating near
gain GA. Moreover, the result that phase differences
of each experiment agree well with each other shows
that experiments are carried out with high accuracy.
5.2 Effect of P gain on motion responses
The experimental results regarding the effect of P
gain constants on motion responses of RFS are
shown in Figure 9 and the gain constants tested are
listed in Table 6 where GD2, GDF or GD3 has the P
gain constant of 2.0, 2.4 or 3.0 fold the P gain
constant of GA. It can be seen that there is no effects
of P gain constant on the heave and pitch motion
responses in the figure.
Accordingly, it is understood that the policy to adopt
the PD gain constants described previously, i.e.
minimum P gain and maximum D gain should be
selected, is correct.
5.3 Motion responses in regular head waves
Experimental results of heave and pitch motion
responses of four hull forms advancing in regular
head waves are presented in Figure 10. Gains of
OGA for ordinary SWATH and GA for RFS are

Figure 7 Effect of D gain constant on motion responses of ordinary SWATH in regular head waves (Fn=0.43)

Figure 8 Effect of D gain constant on motion responses of RFS in regular head waves (Fn=0.43)

IX HSMV Naples 25 - 27 May 2011

Figure 9 Effect of P gain constant on motion responses of RFS in regular head waves (Fn=0.43)

adopted in experiments. Calculated results for monohull and theoretical results for RFS by the use of
RNM are also plotted in the same figure, where
RNM indicates that the hydrodynamic coefficients
and wave exciting forces used in the motion
equations of theoretical study are renormalized from
the experimental measurement. In addition, the
typical wave spectrum at North Atlantic Ocean in
winter is plotted, where the ordinate of spectrum
indicates the density of occurrence probability of
waves with significant wave height more than 6 m.
First, it can be seen that calculated or theoretical
results and experimental results agree very well with
each other.
Secondly, in comparison among mono-hull, trimaran
and RFS, heave and pitch motion responses of RFS
are significantly smaller than those of mono-hull or
trimaran. Although Froude numbers for these hull
forms are a little different from each other, the
tendency of the magnitude of motion responses can
be compared qualitatively. It can be observed that
resonant peaks obviously exist in motion responses
of mono-hull, trimaran or ordinary SWATH while
there is no resonance in the case of RFS.
Thirdly, in comparison between ordinary SWATH
and RFS, i.e. the same SWATH models with
different strut length, it is observed that motion
responses of RFS are much smaller in heave motion
while are smaller in pitch motion than those of
IX HSMV Naples 25 - 27 May 2011

Table 4 D gain constant for ordinary SWATH

Heave

Pitch

KP3 (kg/s2) KD3 (kg/s) KP5 (kgm2/s2) KD5 (kgm2/s)

Ord+mf-OGA

128.4

67.2

Ord+mf-OGC2

171.2

89.6

Ord+mf-OGC

214

112

Table 5 D gain constant for RFS

Heave

Pitch

KP3 (kg/s2) KD3 (kg/s) KP5 (kgm2/s2) KD5 (kgm2/s)

RFS+mf-GB

107

99

56

RFS+mf-GA

214

99

112

RFS+mf-GC

256.8

99

134.4

Table 6 P gain constant for RFS

Heave

Pitch

KP3 (kg/s2) KD3 (kg/s) KP5 (kgm2/s2) KD5 (kgm2 /s)

RFS+mf-GA

214

99

112

RFS+mf-GD2

214

198

112

RFS+mf-GDF

214

235

112

RFS+mf-GD3

214

297

112

ordinary SWATH. The difference between the

motions of ordinary SWATH and RFS may attribute
to the advantages of RFS such as no resonant
8

Figure 10 Motion responses of four kinds of hull form in regular head waves

peak, small wave exciting forces, large damping

coefficients of hull and large D gain can be
adopted for RFS.
5.4 Seakeeping properties of RFS
Theoretical estimation of seaworthiness properties
for real RFS running at 40 knots in regular head
waves with 8 m wave height is presented in Figure
11. The figures are numbered from the top. The
first figure shows amplitude of attack angle of fore
or aft fin. The results of those are less than 10 deg
i.e. less than stall angle except for the case of
/L=1.0. The second or third figure shows relative
motion between fin or bow and wave surface. It
can be seen that relative motion is ensured
sufficiently small in big waves with 8 m height so
that no slamming or propeller racing happens. The
fourth figure shows vertical acceleration of bow. It
is observed that the acceleration is less than 0.1 G.

6 CONCLUSIONS
Figure 11 Seakeeping properties of RFS running at 40
knots in regular head waves with 8 m wave
height

IX HSMV Naples 25 - 27 May 2011

First, theory of PD control to reduce motion

responses of RFS significantly is devised.
Regarding PD control of RFSs motions by the use
of fin lift, motion equations are formulated to
9

include controlling forces due to fin lift, unsteady

characteristics of fin lift generation, interaction
among fins and lower hulls and time lag of control
system. Theoretical calculations and experiments
to measure motion responses of RFS running at
Fn=0.43 in regular head waves using proper
control gain constants are carried out. Theoretical
and experimental results agree well with each
other. Accordingly, theoretical method to predict
the stability of control system and motion
responses is reliable.
Secondly, the comparison of motion responses in
regular head waves among four hull forms such as
mono-hull, trimaran, ordinary SWATH and RFS is
carried out. Where the motions of RFS and
ordinary SWATH are controlled by the use of PD
control. As the result, motion responses of RFS are
significantly reduced compared with those of
mono-hull, trimaran or ordinary SWATH. Then, it
can be concluded from the results of theoretical
estimations and the experiments that RFS shows
very good seaworthiness.

7 ACNOWLEDGEMENTS
This study is performed by the assist of grant-inaid for scientific research (No. 19206093), Japan
Society for the Promotion of Science.

8 REFERENCES
1) Yoshida, T., Fujita, Y. and Fujino, M. (2000) A
proposal of the CS-Swath as a Transocean High Speed Ship,
Transactions of RINA, Vol.142 (B), 136-149, 2000.
2) Yoshida, M., Kihara, H., Iwasita, H., Itakura, H., Bao,
W. and Kinoshita, T. (2009) On the resonance-MotionFree SWATH (RMFS) as an oceangoing large fast ship,
OMAE2009-79103.
3) Lee, C. M. and Curphey, M. (1977) Prediction of
motion, stability, and wave load of Small-Waterplane-Area
Twin-Hull Ships, SNAME Transactions, vol. 85, 94-130.
4) Pitts, W. C., Nielsen, J. N. and Kaattari, G. E. (1959)
Lift and center of pressure of wing-body-tail combinations at
subsonic, transonic, and supersonic speeds, National
Advisory Committee for Aeronautics, Report 1307.
5) Yoshida, M., Kihara, H., Iwashita and Kinoshita, T.
(2009) On the resonance-Motion-Free SWATH as an
oceangoing large fast ship, International conference on
innovation in high speed marine vessels, 31-38.
6) Yoshida, M., Kihara, H., Iwashita and Kinoshita, T.
(2011) Motion control of Resonance-Free SWATH using
small movable fins, International Conference High Speed
Marine Vessels, 106-118.
IX HSMV Naples 25 - 27 May 2011

10