Ships and Offshore Structures

Hydrodynamics of an Unmanned Catamaran Ship with Fixed Hydrofoils in Tandem

--Manuscript Draft--

Manuscript Number: SAOS.1261

Full Title: Hydrodynamics of an Unmanned Catamaran Ship with Fixed Hydrofoils in Tandem

Article Type: Research Paper

Keywords: Tandem foils; thrust; unmanned ship

Abstract: Hydrofoils are approved having the ability of helping propulsion of ships. In this paper,
an unmanned catamaran ship, which is completely driven by two fixed hydrofoils at low
speeds, is proposed and systematically studied through numerical simulations. A CFD
model of this unmanned ship is established and analyzed, considering the interactions
between the ship and hydrofoils. The model is validated by a comparison with the
three-dimensional potential theory for the case of an unmanned ship without hydrofoils.
And the effects of the hydrofoil on ship motions with low forward speed are analyzed.
The results show that the horizontally fixed hydrofoils can significantly reduce the
ship’s heave and pitch motions within a certain encountered wavelength range.

Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation
 Click here to download Manuscript - with full author details
Manuscript - with author details.docx

1 Hydrodynamics of an Unmanned Catamaran Ship with Fixed

2 Hydrofoils in Tandem

3 Dongjiao Wanga, Ping Huoa, Kun Liua*, Shouqiang Qiua, Jiawei Yea, Fulin
4 Lianga

a
5 School of Civil Engineering and Transportation, South China University of Technology,
6 Guangzhou, China, 510640.

7 Correspondence：Kunliu_hit@hotmail.com

8
 9 Hydrodynamics of an Unmanned Catamaran Ship with Fixed
10 Hydrofoils in Tandem

11 Hydrofoils are approved having the ability of helping propulsion of ships. In this
12 paper, an unmanned catamaran ship, which is completely driven by two fixed
13 hydrofoils at low speeds, is proposed and systematically studied through numerical
14 simulations. A CFD model of this unmanned ship is established and analyzed,
15 considering the interactions between the ship and hydrofoils. The model is
16 validated by a comparison with the three-dimensional potential theory for the case
17 of an unmanned ship without hydrofoils. And the effects of the hydrofoil on ship
18 motions with low forward speed are analyzed. The results show that the
19 horizontally fixed hydrofoils can significantly reduce the ship’s heave and pitch
20 motions within a certain encountered wavelength range.

21 Keywords: Tandem foils; thrust; unmanned ship

22 1. Introduction
23 With the development of electronic and navigation technologies, unmanned ship

24 technology could play an important role in for a wide range of marine exploration,

25 development and monitoring. And the outstanding hydrodynamic characteristic of the

26 unmanned ship is the guarantee of its safety.

27 Compared to sailing in calm water, a ship advancing in waves experiences six

28 degrees of freedom motions and dissipates more energy. The extra induced loss of energy

29 is termed added resistance due to waves. The motions and added resistance in waves of a

30 ship can be analyzed by using potential theory. Based on the three-dimensional (3D)

31 Green’s function provided by Wehausen and Laitone[1], the velocity potential of

32 diffracted/radiated waves is expressed in terms of a system of pulsating sources,

33 distributed all over the wetted surface of the floating structure[2]. For zero forward speed

34 issue, the leading commercial 3D diffraction/radiation software AQWA[3], has already

35 been widely used in the offshore industry[4-6], which is also developed using the pulsating
 [7-8]
36 sources technique. And in the case of a ship advancing in waves, some authors

37 employed the translating-pulsating source Green function which the linearized free

38 surface condition with a forward speed was automatically satisfied. However, compared

39 with the issue of zero forward speed case, the translating-pulsating source Green’s

40 function under non-zero forward speed is always difficult to compute. In order to

41 overcome this difficulty, a correction for the forward speed is usually made to the zero

42 forward speed solution. This is in a manner similar to that used by Salvesen et al in their
[9]
43 frequency domain strip theory work . In cases where low to moderate speeds are

44 considered, the differences are negligible small by comparing the pulsating source

45 method with the translating-pulsating source method based on forwards speed Green’s

46 function[10]. Recently, RANS simulations based on computational fluid dynamics have

47 also been used to predict ship resistance and wave induced ship motions [11-13].

48 According to previously published work, it is noted here that suitably designed

49 flapping foils have the ability to extract energy from their generated vortices [14] as well

50 as free-surface waves[15-16] and uniform flows [17]. In their experimental studies, Anderson

51 et al[18], and Read et al [19]

have shown that propulsive performance of a harmonically

52 oscillating foil in uniform flow is dependent on the following parameters, which are the

53 heave amplitude, Strouhal number, angle of attack and the phase angle between heave

54 and pitch. According to the work published by Read and his associates [19], the optimum

55 efficiency was obtained at Strouhal number between 0.25 and 0.4. Also, in this regard,

56 by using computational fluid dynamics software FLUENT, De Silva and Yamaguchi[20]

57 performed a numerical analysis on a two-dimensional active oscillating hydrofoil in wavy

58 flow based on the Reynolds-average Navier–Stokes equation. It was found that when the

59 wave has a -90o phase difference with foil heave motion (when the wave elevation is at

60 the wave crest and the foil is at its bottommost position), the efficiency and thrust reached
61 their maximum values due to the high utilization of wave orbital velocity. An unsteady

62 boundary element method was developed by Filippas and Belibassakis[21] and applied to

63 the analysis of two-dimensional flapping hydrofoils in waves when operating beneath the

64 free surface. The hydrodynamic performance of tandem oscillating foils in regular head

65 waves has been investigated by Xie et al [22]. In this study, two fixed horizontal hydrofoils

66 were mounted beneath the keel of an unmanned ship, one at the bow and the other at the

67 stern. Two cases were considered, one case is the hydrofoils mounted to a fixed ship type

68 structure and another case mounted to a floating ship. It was concluded that the thrust

69 produced by the bow foil is higher than that of stern foil for the case of hydrofoils mounted

70 to a floating ship. And the thrust produced by the bow foil on a ship is much larger than

71 the case of foils mounted to a fixed offshore structure with no oscillating motions.

72 For the case of a ship with hydrofoils encounter a wave, the hydrofoils follow the

73 hull of the ship to perform heaving and pitching motions, and it could produce the forward

74 thrust on these hydrofoils. According to the published studies[23]-[26], employing a

75 hydrofoil at the bow of a ship, whether the foils are fixed or actively controlled, have the

76 effect of an auxiliary propulsion for the ship, and also reduce the vertical wave induced

77 motions of the ship. The experimental results obtained by Terao[27] show that a hydrofoil

78 installed under the catamaran hull of a floating wind turbine system has the effects on

79 reducing the wave drift forces and the pitch motion. And in theory, the motion and

80 resistance in waves of a ship with hydrofoils can be obtained by using linear seakeeping

81 analysis in conjunction with foil model based on quasi-steady lifting line approximation
[23, 25, 28]
82 .

83 In this paper, hydrodynamics of an unmanned ship with two fixed horizontal

84 hydrofoils are analyzed through numerical simulations. Firstly, the potential theory and

85 the CFD method based on FLUENT[29] for ship motion analysis are introduced. Next, the
 86 validation of the two numerical methods is carried out by comparing the results of heave

87 and pitch motions of the catamaran model without hydrofoils. Finally, the motions of the

88 catamaran type unmanned ship fitted with hydrofoils in head regular waves are analyzed

89 by using unsteady Reynolds-Average Navier-Stokes solver in FLUENT, and the effects

90 of hydrofoils on motion response of the unmanned ship are discussed.

91 2. Motions of unmanned ship without hydrofoils based on potential flow theory

92 As shown in Fig.1, three coordinate systems are introduced. The first one is a

93 space fixed right-handed reference axes OXYZ, with the origin O at the mean free surface

94 and Z-axis pointing vertically upwards. The second one is a moving reference axes system

95 oxyz, which is translating with the same velocity as that of the ship forward speed U. Its

96 x-axis points positively in the direction of the bow, and the z-axis points vertically

97 upwards through the centre of gravity of the ship. It is assumed that OXYZ coincide with

98 oxyz initially. The third axis system is adopted for the description of ship motions; here

99 it is convenient to use body-fixed axes Gxbybzb with the origin G at ship’s centre of

100 gravity. The wave direction  is defined in OXY plane of the fixed reference axes, the

101 angle between wave propagating direction and the positive X-axis measure anti-

102 clockwise.

Z
z,zb

X G xb x
O Wave o
Y direction
y,yb
 X x,xb
O o,G
103
104 Fig.1 The coordinate system of a ship
105 The linear sinusoidal waves are assumed, and the water is considered

106 incompressible and inviscid and flow is irrotational. The wave amplitude is assumed to

107 be small compared to both the wave length and water depth.

108 In the fixed reference axes, the water surface elevation at position X and Y can be

109 expressed as

110  ( X , Y , t )  A cos(kXcos  kYsin -t ) (1)

111 where, A is the wave amplitude,  is the wave frequency, k is the wave number, and

112 having wavelength =2/k.

113 As shown in Fig.1, the axis transformation between the fixed and moving

114 reference frame is

115 X  Ut  x , Y  y , Z  z (2)

116 Substituting Eq. (2) into Eq. (1), the wave elevation in the moving reference frame can

117 be expressed as

118  ( x, y, t )  A cos(kx cos   ky sin   et ) (3)

119 in which

120 e    kUcos (4)

121 where，e is the encounter frequency.

122 The wave surface elevation, Eq. (3) can also be expressed in complex form as

123  ( x, y, t )  Re  Aeik ( x cos   y sin  - t ) 

e
(5)

124 In the moving reference frame, the total unsteady fluid potential varies with the encounter

125 frequency and can be expressed as

126 T ( x, y, z, t )  Re T ( x, y, z )e i t  e

(6)

6
127 T ( x, y, z )  I ( x, y, z )  D ( x, y, z )   j ( x, y, z ) j 0 (7)
j 1
 -igA cosh k ( z  d ) ik ( x cos   y sin  )
128 I ( x, y, z )  e (8)
 cosh kd

129 where, I is the first order incident wave potential, D is the diffraction wave potential,

130 j0 is the radiation wave potential due to the j-th motion with unit motion amplitude, j0

131 is the complex amplitude of motion of the j-th degree of freedom.

132 The velocity potential should satisfy Laplace’s equation in the whole fluid domain

 2T  2T  2T

133  2  2 0 (9)
x 2 y z

134 If the disturbed steady flow is neglected, the linear free surface equation is satisfied, such

135 that

 2 
136 （-ie  U ）T  g T  0 on z=0 (10)
x z

137 If the forward speed U is considered to be small, then the following approximate free

138 surface boundary condition can be used

T
139 e2T  g  0 on z=0 (11)
z

140 The flow should also satisfy the boundary condition on the surface of the body surface SB

141 of the ship, expressed by the following equation

D ( x, y, z ))  ( x, y, z )
142 - I , on SB (12a)
n n

 j ( x, y, z )
143  -ie n j  Um j on SB (12b)
n

144 where
145 (n1 , n2 , n3 )  n (13a)

146 (n4 , n5 , n6 )  r  n (13b)

147 (m1 , m2 , m3 )  (0,0,0) (14a)

148 (m4 , m5 , m6 ) （0, n3 , n2） (14b)

149 in which, n is the outward unit normal vector on the body surface and r is the position

150 vector of a point with respect to the ship’s centre of gravity.

151 The boundary condition of the seabed should also be satisfied; expressed by the

152 equation

T
153  0, on z   d (15)
z

154 The radiation condition of the outgoing waves must also be satisfied so that as x 2  y 2

155  the generalized wave disturbance dies away.

156 In this case, the frequency domain pulsating Green's function can be employed

157 together with the body boundary condition given in Equation (12) to determine the

158 diffraction and radiation potential components. The amplitude Fj0 of wave exciting

159 forces, added mass Ajk, and wave damping Bjk can also be estimated from


160 Fj 0     (ie  U )(I  D )n j ds (16)
Sb
x


Fjk     (ie  U )k n j ds
161 Sb
x (17)
  Ajk  ie B jk
2
e

162 The unmanned ship is considered as a rigid body and the oscillating motion,

163 referred to the center of gravity, in the j-th mode in response to regular waves encountered

164 at frequency e may be expressed by the equation

165  j (t )   j 0ei t
e
(18)

166 where, 1, 2 and 3 are the translational displacements in x, y and z direction

167 respectively, while 4, 5 and 6 denote the rotational motions about the x, y and z

168 directions respectively.

169 The coupled six-degrees of freedom linear differential equations of motion can be

170 expressed by the following equation

6
171  [(M
k 1
jk  Ajk )k  Bk jk  K jkk ]  Fj 0e  iet (19)

172 where Mjk is the mass matrix, Kjk is the hydrostatic stiffness matrix.

173 3. Hydrodynamics of an unmanned ship based on FLUENT

174 The software AQWA[3] suite has the advantages of fast computational speed.

175 However, the effect of thrust and lift force caused by hydrofoils cannot be taken in

176 account by AQWA itself. In this case, the computational fluid dynamics software

177 FLUENT[29] is selected to calculate the motions of the unmanned ship. The turbulent flow

178 around the ship traveling in waves is simulated by solving the incompressible RANS

179 equations with the finite volume method. This is achieved by using Volume of Fluid

180 (VOF) formulation and the open channel boundary condition. The velocities of the ship

181 are calculated from the forces balance on the ship, as is done by the six degree of freedom

182 (6DOF) solver. The dynamic mesh model is used to update the volume mesh at each time

183 step based on the new position of the boundaries of the ship.

184 3.1 Volume Fraction Equation and Momentum Equation

185 VOF model is used to track the free surface by the solution of a continuity

186 equation for the volume fraction of one of the phases. Two phase incompressible flow

187 combining air and water are considered. The air is defined as the primary phase and the

188 water as the secondary phase. The phases are represented by the subscripts 1 and 2,

189 respectively. The volume fraction equation for secondary phase, water, has the following

190 form

 ( 2 )
191    ( 2V2 )  0 (20)
t
192 where, V2 is the velocity of water, 2 is the secondary phase volume fraction in the cell,

193 2=0 for air and 2=1 for water, and any intermediate value is a mixture of the two fluids.

194 The primary-phase volume fraction 1 is be computed from

195 1  1- 2 (21)

196 The density  and viscosity  in each cell are given by

197    2 2 （1- 2）1 (22)

198    2 2 （1- 2）1 (23)

199 where, q, q are the physical density, and viscosity of phase-q, respectively.

200 A single momentum equation is solved throughout the domain, and it is dependent

201 on the volume fraction of all phases through the properties  and the molecular viscosity

202 

( V )
203    ( VV )  p      (V  V T )    g (24)
t

204 Where, p is static pressure and  g is the gravitational force.

205 Reynolds averaging approach with Shear-Stress Transport (SST) k- model is

206 selected for the numerical calculation.

207 3.2 Open Channel Wave Boundary Conditions and Numerical Beach Treatment

208 Open channel wave boundary conditions in ANSYS Fluent allow us to simulate

209 the propagation of waves through velocity inlet boundary condition. In this work, the first

210 order Airy wave theory, which is applied to small amplitude waves in shallow to deep

211 water depth ranges, is applied. According to the relative motion theory, the effect of a

212 moving ship could be incorporated with the flow current when the flow is specified

213 relative to the ship.

214 To avoid wave reflection caused by outlet boundary for passing waves, a damping

215 sink term is added in the momentum equation for the cell zone in the vicinity of the
216 pressure outlet boundary. Numerical beach treatment in ANSYS Fluent uses linear

217 damping in vertical direction along gravity and quadratic damping in flow direction

218 3.3 Dynamic Meshing with 6DOF Solver

219 The tetrahedron unstructured grid based on spring-based smoothing, local re-

220 meshing and dynamic mesh updating techniques are used to model flow where the shape

221 of the domain is changing with the time due to motion of the ship. The 6DOF solver in

222 ANSYS Fluent is used to compute motions of the center of gravity of the ship. The

223 governing equation for the translational motion of the center of gravity is solved for in

224 the inertial coordinate system, and the angular motion is computed using body

225 coordinates. The angular and translational velocities are used in the dynamic mesh

226 calculations to update the rigid body position.

227 When a ship is moving thought head waves, the dominant motion responses are

228 heaving and pitching motions, while the surge, sway, roll and yaw motions are usually

229 neglected.

230 4. Results and discussion

231 The model of the catamaran with two fixed horizontal foils is illustrated in Fig.2.

232 And the principal particulars of the catamaran and the parameters of the foils are listed in

233 Table 1 and Table 2 respectively.

234
235 Fig.2 Geometry of the catamaran model with fixed foils

236
237 Table 1 Main particulars of catamaran model

Item symbol value unit

Length between perpendiculars L 2.1 m

Beam overall B 1.01 m

Draft d 0.095 m

Displacement  52 kg

Beam demi-hull B 0.185 m

Distance between centre of hulls K 0.825 m

Vertical centre of gravity above keel KG 0.095 m

Longitudinal centre of gravity from stern LCG 0.992 m

Pitch radius of gyration kyy 0.535 m

238

239 Table 2 Parameters of the foils

Item symbol value unit

Foil chord length c 0.15 m

Submergence h1 0.3 m

Pivot axis XR 0.25c m

240 4.1 Motion response of Unmanned Ship without Hydrofoils

241 Illustration of AQWA meshes for the diffraction/radiation calculations is shown

242 in Fig. 3. Total 17134 elements are automatically generated on the catamaran hull, and

243 8025 panels on the wetted hull. The water depth is 0.695m, wave amplitude A=0.05m.

244 The wave exciting forces on the catamaran is shown in Fig. 4.

245 The response motions, Eq. (18) can also be expressed as

246  j (t )   j 0 cos(et   j ) , where  j 0 is the motion amplitude of j-th mode, and j is the

247 phase angle relative to the incident wave at centre of gravity of the ship. The heave and

248 pitch motions of the catamaran model without hydrofoils moving at constant forward
249 speed U=0.1 m/s are shown Fig.5. It indicates that when the ratio of wavelength to ship

250 length /L>1.2, the phase angle of the heave motion relative to the incident wave 3 0o,

251 but the phase angle of the pitch lags behind the wave nearly about 90o. And the pitch

252 motion reaches its maximum value when /L1.5. The heave and pitch motions obtained

253 by FLUENT for the catamaran model without hydrofoils at speed U=0.2 m/s in regular

254 head waves are shown in Fig. 6, which are coincident with the results from AQWA.

255 For regular head waves with amplitude A=0.05m and wave length =3.5m, the

256 time histories of heave and pitch motions by FLUENT for the catamaran without foils

257 when travelling with no forward speed and low forward speeds (U= 0.1m/s and 0.2m/s)

258 are illustrated in Fig 7. It is shown that the amplitudes of both heave and pitch motions

259 are increase with the ship speed.

260
261 Fig.3 Hull mesh used in AQWA calculation
 50 5
Heave
40 Pitch 4

|F 30 |/( gVA /L)

| F50 |/( g VA)

30 3

20 2

10 1

0 0
0 1 2 3 4 5
262  /L
263 Fig.4 Wave exciting forces on the catamaran

1.2 120
Heave
1 Pitch 100

0.8 80

0.6 60

0.4 40

0.2 20

0 0
0 1 2 3 4 5
264  /L

180
Heave j =3
Pitch j =5
90

-90
0 1 2 3 4 5
265  /L

266 Fig.5 Heave and pitch motion responses of the catamaran without foils (=180o, U=0.1m/s)
 1.2
U =0.2m/s
1 AQWA
Fluent
0.8

0.6

0.4

0.2

0
0 1 2 3 4 5
 /L
267
268 (a) Heave response

100
U =0.2m/s
80 AQWA
Fluent
60

40

20

0
0 1 2 3 4 5
269  /L

270 (b) Pitch response

271 Fig.6 Motion responses of the examined catamaran against non-dimensional wave length /L

272 (=180o)
 0.04
U=0.0 m/s
0.03 U=0.1 m/s
U=0.2 m/s
0.02

0.01
3 ( m)

-0.01

-0.02

-0.03
0 2 4 6 8 10
273 t ( s)

274 (a) Heave

5
U=0.0 m/s
4
U=0.1 m/s
3
U=0.2 m/s
2
5 ( )

1
o

-1

-2

-3

-4
0 2 4 6 8 10
t ( s)
275
276 (b) Pitch

277 Fig.7 Time series of the motion responses for the examined catamaran model travelling at low speed

278 in head waves (=3.5m, A=0.05m)

279 4.2 The Effect of Hydrofoils on Motion responses of Unmanned Ship

280 The computational results from FLUENT for the catamaran model without foils

281 are validated by comparison with the data from AQWA, which is presented in Fig. 6. The

282 motions of a ship with foils can also be found by using Reynolds-Average Navier-Stokes

283 solver. In this case, the CFD software FLUENT is qualified for investigating the effect of

284 hydrofoils on the motions of the unmanned ship. For the catamaran model with foils, the
285 finer grids are needed in the process of volume mesh updating, which is handled

286 automatically by FLUENT at each time step. And the massive amount grid model takes

287 more computing time compared to the case without foils. As a result, only the comparison

288 of the ship motions for the catamaran model without hydrofoils and with fixed hydrofoils

289 at forward speed U=0.1 m/s in regular head waves are given, as shown in Fig.8. It is found

290 that when /L<2.2, the fixed hydrofoils reduces the pitch motion significantly, and when

291 /L>2.2, the pitch motion is slightly larger than that without hydrofoils. This is because

292 the fixed hydrofoils increase the inertia moment of the catamaran, and this which cause

293 the resonance period of the pitch motion to become larger than the catamaran without

294 hydrofoils.

295 The forces acting on the fixed structure in water waves are the wave exciting forces.

296 When the catamaran model is moving at speed U=0.1 m/s in regular head waves with

297 A=0.05m, =3.5m, the time series of the wave exciting forces in surge, heave and pitch

298 direction are shown in Fig 9~ Fig.11. It is found that the wave exciting force acting on

299 the catamaran with foils in surge direction is slightly larger than that without foils, but

300 the amplitudes in heave and pitch directions are become a bit smaller.

301 The time series of heave, pitch motions and the resistance for wave amplitude

302 A=0.05m, wave length =3.5m are shown in Fig 12~ Fig.14. The results show that the

303 fixed hydrofoils have not only reduced motions of the catamaran but also changed the

304 phase angle. Referring to Fig. 12 and Fig. 13, we observe that the phase angle of the heave

305 and pitch motions with foils lag nearly about 5o and 30o respectively, compared with the

306 case without foils. Compared with the case without foils, the amplitude of the added

307 resistance acting on the catamaran model has also become smaller.
 1.4
U =0.1m/s
1.2 Without foils-AQWA
Without foils-Fluent
1 With foils-Fluent
0.8
0.6
0.4
0.2
0
0 1 2 3 4
308  /L
309 a) Heave

120
U =0.1m/s Without foils-AQWA
100 Without foils-Fluent
With foils-Fluent
80

60

40

20

0
0 1 2 3 4
310  /L
311 b) Pitch
312 Fig.8 Heave and pitch response of the examined catamaran models in head waves at speed U=0.1m/s

30
catamaran without foils catamaran with foils
20
10
F 1 (N)

0
-10

-20
-30
0 2 4 6 8 10 12
t (s)
313
314 Fig. 9 Time series of wave exciting force in surge direction (=3.5m, A=0.05m, U=0.1m/s)
 300
catamaran without foils catamaran with foils
200
100
F 3 (N)

0
-100
-200
-300
0 2 4 6 8 10 12
t (s)
315
316 Fig. 10 Time series of wave exciting force in heave direction (=3.5m, A=0.05m, U=0.1m/s)

150
catamaran without foils catamaran with foils
100
F 5 (Nm)

50
0
-50
-100
-150
0 2 4 6 8 10 12
317 t (s)

318 Fig. 11 Time series of wave exciting moment in pitch direction (=3.5m, A=0.05m, U=0.1m/s)

0.04
Catamaran model without foils
0.03
Catamaran model with foils
0.02
0.01
0.00
-0.01
-0.02
-0.03
t ( s)
-0.04
0 1 2 3 4 5 6 7 8
319
320 Fig.12 Time series for the heave motion of the catamaran models in head waves (=3.5m, A=0.05m,

321 U=0.1m/s)
 5.0
4.0 Catamaran model without foils
3.0 Catamaran model with foils
2.0
1.0
0.0
-1.0
-2.0
-3.0
-4.0
t ( s)
-5.0
0 1 2 3 4 5 6 7 8
322
323 Fig.13 Time series for the pitch motion of the catamaran models in head waves (=3.5m, A=0.05m,

324 U=0.1m/s)

40
Catamaran model without foils
30 Catamaran model with foils
20
10
Fx (N)

0
-10
-20
-30
t ( s)
-40
0 1 2 3 4 5 6 7 8
325
326 Fig.14 Time series for the resistance of the catamaran models in head waves (=3.5m, A=0.05m,

327 U=0.1m/s)

328 5 Conclusions

329 The effect of fixed horizontal hydrofoils on the hydrodynamics of an unmanned

330 ship is investigated by using numerical method based on Reynolds-averaged Navier–

331 Stokes equation. The simulated results for the unmanned ship without hydrofoils in head

332 regular waves are validated by comparison with that of potential theory. In order to get

333 larger foil’s heaving motions, two fixed hydrofoils are mounted at the bow and stern

334 under the keel of an unmanned ship. The hydrofoils will follow the hull of the ship to

335 perform oscillating motions and then generate thrust. The fixed hydrofoils not only

336 change the heaving and pitching motions of the unmanned ship but also change their

337 phase angles.

338 Acknowledgements

339 This research is co-financed by Guangdong provincial department of science

340 and technology (Grant No. 2014A020217001), National Key R&D Program of China

341 (2016YFC1400202) and the special fund of Guangdong Provincial department of ocean

342 and fisheries (A201501D06).

343 References

344 1. Wehausen, J.V., Laitone, E.V. Surface Waves. Encyclopedia of Physics, Vol. IX/Fluid Dynamics
345 III. Springer-Verlag, Berlin, 1960, 446–778.
346 2. Garrison C.J. Hydrodynamic loading of large offshore structures: three-dimensional source
347 distribution methods, Numerical Methods in Offshore Engineering, John Wiley, 1978, 87-140.
348 3. ANSYS, 2010, AQWA-LINE Manual, ANSYS, Inc. United Kingdom.
349 4. Wang S., Wang X., Woo W.L. and Seow T.H. Study on green water prediction for FPSOs by a
350 practical numerical approach, Ocean Engineering, 2017, 143:88-96.
351 5. Hill J., Laycock S., Chai S., Balash C. and Morand H., Hydrodynamic loads and response of a
352 Mid Water Arch structure, Ocean Engineering, 2014, 83:76-86.
353 6. Geba K., Welaya Y., Leheta H. and Abdel-Nasser Y., The hydrodynamic performance of a novel
354 float-over installation, Ocean Engineering, 2017, 133:116-132.
355 7. Sun X.S., Cao C.B. and Ye Q., Numerical investigation on seakeeping performance of SWATH
356 with three dimensional translating-pulsating source Green function, Engineering Analysis with
357 Boundary Elements, 2016,73:215-225.
358 8. Hong L., Zhu R.C., Miao G.P., Fan J. and Li S., An investigation into added resistance of vessels
359 advancing in waves, Ocean Engineering, 2016, 123:238-248.
360 9. Salvesen, N., Tuck, E.O. and Faltinsen O.M. Ship motions and sea loads, Transactions - Society
361 of Naval Architects and Marine Engineers, 1970, 78:250-287.
362 10. Inglis R.B. and Price W.G. A three dimensional ship motion theory-comparison between
363 theoretical predictions and experimental data of hydrodynamic coefficients with forward speed.
364 Transactions of the Royal Institution Naval Architects, 1981, 124: 141–157.
365 11. Guo B.J., Deng, G.B., Steen s., Verification and validation of numerical calculation of ship
366 resistance and flow filed of a large tanker, Ships and Offshore Structures, 2013, 8(1):3-14.
367 12. Castiglione T., Stern F., Bova S., Kandasamy M., Numerical investigation of the seakeeping
368 behavior of a catamaran advancing in regular head waves, Ocean Engineering, 2011, 38:1806-
369 1822.
370 13. Tezdogan T., Incecik A., Turan,O., Full-scale unsteady RANS simulations of vertical ship
371 motions in shallow water. Ocean Engineering, 2016, 123:131–145.
372 14. Gopalkrishnan R., Triantafyllou , M,S., Triantafyllou, G.S. and Barrett D., Active vorticity control
373 in a shear flow using a flapping foil, Journal of Fluid Mechanics, 1994, 274:1-21.
374 15. Wu T.Y., Extraction of flow energy by a wing oscillating in waves, Journal of Ship Research, 1972,
375 16: 66–78.
376 16. Grue J., Mo A. and Plam E., Propulsion of a foil moving in water waves, Journal of fluid
377 mechanics, 1988, 186(1):393-417.
378 17. Zhu Q., Peng Z., Mode coupling and flow energy harvesting by a flapping foil, Physics of Fluids,
379 2009, 21(3):033601:1-10.
380 18. Anderson J.M., Streitlien K., Barrett D.S., Triantafyllou M.S., Oscillating foils of high propulsive
381 efficiency. Journal of Fluid Mechanics, 1998, 360:41–72.
382 19. Read D.A., Hover F.S., Triantafyllou M.S., Forces on oscillating foils for propulsion and
383 maneuvering, Journal of Fluids and Structures, 2003, 17: 163–183.
384 20. De Silva L.W.A., Yamaguchi H., Numerical study on active wave devouring propulsion, J Mar Sci
385 Technol, 2012, 17:261–275.
386 21. Filippas E.S., Belibassakis K.A., Hydrodynamic analysis of flapping-foil thrusters operating
387 beneath the free surface and in waves, Engineering Analysis with Boundary Elements, 2014, 41: 47-
388 59.
389 22. Xie H.M., Wang D.J., Lin Z.J., Qiu S.Q. and Ye J.W., Hydrodynamic performance of tandem
390 oscillating foils in waves, The 27th International Ocean and Polar Engineering Conference, San
391 Francisco, June 25-30, 2017, Volume III:865-870.
392 23. Naito S., Isshiki H., Effect of Bow Wings on Ship Propulsion and Motions, Applied Mechanics
393 Reviews, 2005, 58:253-268
394 24. Bøckmann E., Steen S., The Effect of a Fixed Foil on Ship Propulsion and Motions, Third
395 International Symposium on Marine Propulsors, Launceston, Tasmania, Australia, May 2013, 553-
396 561.
397 25. Belibassakis K. A., Politis G. K., Hydrodynamic performance of flapping wings for augmenting
398 ship propulsion in waves, Ocean Engineering, 2013, 72: 227-240.
399 26. Belibassakis K. A., Politis G. K., Model test and simulation of a ship with wavefoils, Applied Ocean
400 Research, 2016, 57: 8-18.
401 27. Terao Y., Sunahara S., Application of wave devouring propulsion system to ocean engineering, 31st
402 International Conference on Ocean, Offshore and Arctic Engineering, 2012, OMAE2012-83122,
403 Brazil, 1-8.
404 28. Belibassakis K. A., Politis G. K., Ship propulsion in waves by actively controlled flapping foils,
405 Applied Ocean Research, 2015, 52:1-11.
406 29. ANSYS Fluent theory guide, 2013, ANSYS, Inc. USA.