This action might not be possible to undo. Are you sure you want to continue?

]

On: 2 April 2008

Access Details: [subscription number 778557058]

Publisher: Taylor & Francis

Informa Ltd Registered in England and Wales Registered Number: 1072954

Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Ships and Offshore Structures

Publication details, including instructions for authors and subscription information:

http://www.informaworld.com/smpp/title~content=t778188387

Practical evaluation of resistance of high-speed

catamaran hull forms - Part I

Prasanta K. Sahoo

a

; Marcos Salas

b

; Adam Schwetz

c

a

Australian Maritime College, Launceston, Australia

b

University of Austral, Chile

c

Schwetz Design, Australia

Online Publication Date: 01 December 2007

To cite this Article: Sahoo, Prasanta K., Salas, Marcos and Schwetz, Adam (2007)

'Practical evaluation of resistance of high-speed catamaran hull forms - Part I', Ships

and Offshore Structures, 2:4, 307 - 324

To link to this article: DOI: 10.1080/17445300701594237

URL: http://dx.doi.org/10.1080/17445300701594237

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article maybe used for research, teaching and private study purposes. Any substantial or systematic reproduction,

re-distribution, re-selling, loan or sub-licensing, systematic supply or distribution in any form to anyone is expressly

forbidden.

The publisher does not give any warranty express or implied or make any representation that the contents will be

complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should be

independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings,

demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or

arising out of the use of this material.

D

o

w

n

l

o

a

d

e

d

B

y

:

[

A

u

s

t

r

a

l

i

a

M

a

r

i

t

i

m

e

C

o

l

l

e

g

e

]

A

t

:

0

4

:

5

2

2

A

p

r

i

l

2

0

0

8

Practical evaluation of resistance of high-speed

catamaran hull forms—Part I

doi:10.1080/17445300701594237

Prasanta K Sahoo

1

, Marcos Salas

2

and Adam Schwetz

3

1

Australian Maritime College, Launceston, Australia

2

University of Austral, Chile

3

Schwetz Design, Australia

Abstract: Although catamaran conﬁguration has been around for a longtime, it is only in the recent

past that such hull forms have seen unprecedented growth in the high-speed ferry industry. One of

the design challenges faced by naval architects is accurate prediction of the hydrodynamic charac-

teristics of such vessels primarily in the areas of resistance, propulsion and seakeeping. Even though

considerable amount of research has been carried out in this area, there remains a degree of uncertainty

in the prediction of calm water resistance of catamaran hull forms. This paper attempts to present

the research work carried out so far and what needs to be undertaken in future for a reasonably accurate

prediction of catamaran resistance characteristics. The authors have examined the deep water wave

resistance characteristics of a series of transom stern, semi-displacement slender catamaran hull forms

of round bilge as well as single chine hull forms, which are of utmost importance to the high-speed ferry

industry. The accuracy of the established regression equation has been seen to deviate appreciably by

various sources of uncertainties. Veriﬁcation of the equation with experimental database is also lacking

to a certain extent. Further research is, therefore, needed to reﬁne the accuracy as well as to complete

the selection of crucial parameters employed. However, the results obtained have shown considerable

promise, and a regression equation for predicting wave resistance of catamarans in calm water can be

seen as achievable.

Key words: Catamaran, resistance, wave resistance, computational ﬂuid dynamics.

ABBREVIATIONS

CFD Computational Fluid Dynamics

LCB Longitudinal center of buoyancy, reference from

the transom

LCF Longitudinal center of ﬂoatation, reference from

the transom

DWL Design waterline

ITTC International Towing Tank Conference

WSA Wetted Surface Area

Corresponding Author:

Prasanta K. Sahoo

Maritime Engineering

Australian Maritime College

Newnham Drive

Launceston

TAS 7250, Australia

Tel.: +61 363354822

Fax: +61 363354720

Email: p.sahoo@amc.edu.au

NOTATION

A

AP

Area at aft perpendicular

A

FP

Area at forward perpendicular

A

X

Area at maximum transverse section

B

X

Breadth at maximum transverse section

B Demi-hull beam at the waterline

B/T Beam-Draught ratio

C

A

incremental resistance coefﬁcient

C

B

Block Coefﬁcient C

B

=

∇

L.B.T

C

F

ITTC ‘57 ship-model correlation line

C

F

=

0.075

(Log

10

Rn−2)

2

C

P

Prismatic coefﬁcient

C

M

Midship coefﬁcient

C

Wc at,

C

Wdemi

Wave resistance coefﬁcient

C

R

Residuary resistance coefﬁcient

C

T

Total resistance coefﬁcient

C

∇

Volumetric coefﬁcient

C

W

Wave resistance coefﬁcient correction

F

n

Froude number (based on length)

F

n∇/2

Froude number based on volumetric

displacement of demi-hull

Copyright C 2007 Taylor & Francis 307 SAOS 2007 Vol. 2 No. 4 pp. 307–324

D

o

w

n

l

o

a

d

e

d

B

y

:

[

A

u

s

t

r

a

l

i

a

M

a

r

i

t

i

m

e

C

o

l

l

e

g

e

]

A

t

:

0

4

:

5

2

2

A

p

r

i

l

2

0

0

8

P. K. Sahoo, M. Salas and A. Schwetz

g Acceleration due to gravity 9.81 m/s

2

i

E

Half waterline entry angle

L or L

WL

Waterline length

L/B or L

WL

/ Length–beam ratio (demi-hull)

B

XDH

L/∇

1/3

Slenderness ratio

LCF Longitudinal center of ﬂotation from

transom

LCB Longitudinal center of buoyancy from transom

P

E

Effective power

R

W

Wave resistance

R

F

Frictional resistance

R

T

Total resistance

s Separation (measured between demi-hull centre

planes)

s/L Separation ratio (between demi-hulls)

S Wetted surface area

T

AP

Draft at aft perpendicular

T

X

Draft at maximum transverse section

1+k Form factor

1+γ k Viscous form factor for catamarans

β

M

Deadrise angle at amidships

δ

W

Transom wedge angle

ε

R

Residual drag–weight ratio

φ Factor for pressure ﬁeld change

σ Velocity augmentation factor

γ Viscous interference factor

ρ Fluid density

τ Wave resistance interference factor

υ Kinematic viscosity of ﬂuid

Displacement

Volumetric displacement

INTRODUCTION

Catamarans account for 43%of the ﬂeet by vessel numbers

as given by the report of Drewry Shipping Consultants

(1997). Slender hull forms and higher speed capabilities

provoked the need of technological evolution in predicting

their preliminary characteristics of resistance. Calm water

resistance of catamarans is in general attributed to two ma-

jor components, namely, viscous resistance and calmwater

wave resistance. The former has been acceptably deter-

mined from ITTC-1957 line using a form factor compo-

nent whilst the latter still remains a stimulating question to

researchers. It is understood that the solutions cannot be

generalized by one simple formula but varied in accordance

with speciﬁc conﬁgurations of catamarans.

With the advent of computational ﬂuid dynamics

(CFD), there is hope for further development. In this

paper a computational package, SHIPFLOW, is used to

generate data of wave making resistance of catamaran hull

forms, and the regression equations were developed based

on the data. In the end, credibility of these equations has

been compared with several other theoretical and exper-

imental methods presently available. This paper concen-

Table 1 Catamaran geometric parameters (Insel and

Molland, 1992)

Geometric parameters L/∇

1/3

L/B B/T C

B

Range of Application 6 to 9 6 to 12 1 to 3 0.33 to 0.45

Table 2 Model particulars (Insel and Molland, 1992)

LCB/Lfrom

Models L/∇

1/3

L/B B/T C

B

transom

C2 7.1 10 1.6 0.44 50%

C3 6.3 7 2 0.397 43.6%

C4 7.4 9 2 0.397 43.6%

C5 8.5 11 2 0.397 43.6%

trates on both single hard chine as well as round bilge hull

forms with transom stern. The model parameters have

been based on data of modern catamarans found from the

literature survey.

Prediction of total resistance—background

The background of the work has been based on some of

the important modern methods in application so far. These

methods have been brieﬂy explained below.

Insel and Molland’s method (1992)

The paper by Insel and Molland (1992) summarizes a

calm water resistance investigation into high-speed semi-

displacement catamarans, with symmetrical hull forms

based on experimental work carried out at the University

of Southampton.

Two interference effects contributing to the total resis-

tance effect were established, being viscous interference,

caused by asymmetric ﬂow around the demi-hulls, which

effects the boundary layer formation, and wave interfer-

ence, due to the interaction of the wave systems produced

by each demi-hull. The particulars of models tested by

Insel and Molland (1992) are presented in Table 1. The

particulars of the models used in the investigation are pre-

sented in Table 2.

Models C3, C4 and C5 were of round bilge hull form

derived from the NPL series and model C2 was of the

parabolic Wigley hull form. Body plans of these models are

shown in Figure 1. All models were tested over a range of

Froude numbers of 0.1to1.0inthe demi-hull conﬁguration

and catamaran conﬁguration with separation ratios (s/L)

of 0.2, 0.3, 0.4 and 0.5. Calm water resistance, running

trim, sinkage and wave pattern analysis experiments were

carried out.

The authors proposed that the total resistance of a cata-

maran should be expressed by Equation (1):

C

TCAT

= (1 +φk)σC

F

+τC

w

(1)

Here φ has been introduced to take account of pressure

ﬁeld change around the demi-hull and σ takes account of

308 SAOS 2007 Vol. 2 No. 4 Copyright C 2007 Taylor & Francis

D

o

w

n

l

o

a

d

e

d

B

y

:

[

A

u

s

t

r

a

l

i

a

M

a

r

i

t

i

m

e

C

o

l

l

e

g

e

]

A

t

:

0

4

:

5

2

2

A

p

r

i

l

2

0

0

8

Practical evaluation of resistance of high-speed catamaran hull forms—Part I

Figure 1 Body plan of models C2, C3, C4 and C5 (Insel and Molland, 1992).

Table 3 Derived form factors (Insel and Molland,

1992)

Mono-hull C2 C3 C4 C5

(1 +k) 1.10 1.45 1.30 1.17

the velocity augmentation between the hulls and would be

calculated from an integration of local frictional resistance

over the wetted surface and (1 +k) is the form factor for

the demi-hull in isolation. For practical purposes, φ and σ

can be combined into a viscous interference factor γ where

(1 +φk)σ = (1 +γ k), whence:

C

TCAT

= (1 +γ k)C

F

+τC

W

(2)

Noting that for demi-hull in isolation, γ =1 and τ =1,

and for a catamaran, τ can be caculated from Equation (3).

τ =

C

WCAT

C

WDEMI

=

[C

T

−(1 +γ k)C

F

]

CAT

[C

T

−(1 +k)C

F

]

DEMI

(3)

The authors conclude that the form factor, for practi-

cal purposes, is independent of speed and should, thus,

be kept constant over the speed range. This was a good

practical solution to a complex engineering problemat that

point in time. The derived form factors for the mono-hull

conﬁguration are shown in Table 3. The authors also con-

clude that the viscous interference factor γ is effectively

independent of speed and should be kept constant across

the speed range and it depends primarily on L/B ratio.

The authors further conclude that:

r

The vessels tested have an appreciable viscous form effect,

and are higher for catamarans where viscous interference

takes place between the hulls.

r

Viscous resistance interference was found to be indepen-

dent of speed and hull separation, and rather is dependent

on demi-hull length to beam ratio.

r

Generally higher hull separation ratios result in smaller

wave interference, with beneﬁcial wave interference be-

tween Froude numbers of 0.35 and 0.42.

r

Catamarans display higher trim angles than mono-hulls,

and that the trim angle is reduced with increasing hull

separation ratios.

r

A ship to model correlation exercise is required for the

extrapolation techniques presented to be validated.

Figure 2 Effect of hull separation on catamaran resistance

(Millward, 1992).

Millward’s method (1992)

In his investigation, Millward (1992) has reported his test

results on a series of catamarans characterised by hull

length–beam ratio (L/B) of 10 and a beam–draft ratio

(B/T) of 2. Millward (1992) in fact intended to adhere

to the common parameter range as suggested by Insel

and Molland (1992). Figure 2 reproduced from the article

demonstrates the effect of separation ratio on resistance.

He introduced a new wave resistance coefﬁcient,

C

∗

W

=

R

∗

Fn

2

(4)

in which R

∗

=

R

W

8

π

ρg

B

2

T

2

L

and R

W

is the wave resistance.

The frictional resistance is calculated using ITTC 1957

line. From this, the total resistance (R

T

) of catamaran can

be found by:

R

T

= 2[(1 +k)R

F

+ R

W

] (5)

Molland et al. method (1994)

The paper by Molland et al. (1994) is an extension of the

work conducted by Insel and Molland (1992). Additional

models are tested with the particulars listed in Tables 4.

309 Copyright C 2007 Taylor & Francis SAOS 2007 Vol. 2 No. 4

D

o

w

n

l

o

a

d

e

d

B

y

:

[

A

u

s

t

r

a

l

i

a

M

a

r

i

t

i

m

e

C

o

l

l

e

g

e

]

A

t

:

0

4

:

5

2

2

A

p

r

i

l

2

0

0

8

P. K. Sahoo, M. Salas and A. Schwetz

Table 4 Particulars of models (Molland et al ., 1994)

Models 3b 4a 4b 4c 5a 5b 5c 6a 6b 6c

L/B 7.0 10.4 9.0 8.0 12.8 11.0 9.9 15.1 13.1 11.7

L/∇

1/3

6.27 7.40 7.41 7.39 8.51 8.50 8.49 9.50 9.50 9.50

B/T 2.0 1.5 2.0 2.5 1.5 2.0 2.5 1.5 2.0 2.5

C

B

0.397 0.397 0.397 0.397 0.397 0.397 0.397 0.397 0.397 0.397

LCB/L (%) 43.6 43.6 43.6 43.6 43.6 43.6 43.6 43.6 43.6 43.6

Table 5 Model form factors (Molland et al., 1994)

s/L= 0.2 s /L = 0.3 s /L = 0.4 s /L = 0.5

Model Mono-hull

Form factors (1+k) 1+γ k γ 1+γ k γ 1+γ k γ 1+γ k γ

3b 1.45 1.6 1.33 1.65 1.44 1.55 1.2 1.60 1.3

4a 1.30 1.43 1.43 1.43 1.43 1.46 1.5 1.44 1.5

4b 1.30 1.47 1.57 1.43 1.43 1.45 1.5 1.45 1.5

4c 1.30 1.41 1.37 1.39 1.30 1.48 1.6 1.44 1.5

5a 1.28 1.44 1.57 1.43 1.54 1.44 1.6 1.47 1.7

5b 1.26 1.41 1.58 1.45 1.73 1.40 1.5 1.38 1.5

5c 1.26 1.41 1.58 1.43 1.65 1.42 1.6 1.44 1.7

6a 1.22 1.48 2.18 1.44 2.0 1.46 2.1 1.48 2.2

6b 1.22 1.42 1.91 1.40 1.82 1.47 2.1 1.44 2.0

6c 1.23 1.40 1.74 1.40 1.74 1.45 2.0 1.44 1.9

Table 6 Viscous form factors of catamarans (Molland

et al. 1994)

s/L= 0.2 s/L = 0.3 s/L = 0.4 s/L = 0.5

L/∇

1/3

B/T 1+γ k 1+γ k 1+γ k 1+γ k

8.5 1.5 1.44 1.43 1.44 1.47

8.5 2.0 1.41 1.45 1.4 1.38

8.5 2.5 1.41 1.43 1.42 1.44

Average — 1.42 1.44 1.42 1.43

9.5 1.5 1.48 1.44 1.46 1.48

9.5 2.0 1.42 1.40 1.47 1.44

9.5 2.5 1.40 1.40 1.45 1.44

Average — 1.43 1.41 1.46 1.45

The research and results are also detailed in Molland et al.

(1994). Formfactors as per Molland et al. (1994) are shown

in Table 5. In addition to this, Molland et al. (1994) gives

the experimental data of a systematic series of high-speed

displacement catamaran forms in which the viscous form

factors are shown as in Table 6. For further details on the

resistance data, readers are referred to the above report.

Armstrong’s thesis entitled ‘AThesis on the Viscous Resis-

tance and Form Factor of High-speed Catamaran Ferry Hull

Forms,’ (Armstrong, 2000), examines the current methods

for predicting the resistance of recently designed high-

speed catamarans. Current literature suggests large form

factors are needed for correlation between model scale and

full scale which, Armstrong (2000) claims, contradicts the

expectation that long slender hull forms would have low

values. Armstrong proposals on form factors are presented

in Appendix I.

Table 7 Regression coefficients for Equation 7

a

0

0 a

8

−2.506

a

1

0.258 a

9

−2.432

a

2

2.505 a

10

100.173

a

3

−150.791 a

11

−1.636

a

4

4.932 a

12

1.417

a

5

−1.446 a

13

−43.355

a

6

68.628 a

14

−2.927

a

7

6.549

Form factor of catamarans

As there is no equation giving the form factor of catama-

rans (1+γ k) directly from their particulars, a regression

model had to be developed from the data of Table 6. For

each Froude number studied, the catamaran form factor

(1+γ k) will be a function of the different geometrical par-

ticulars and the mono-hull form factor (1+k) of the hull

forms. Using multiple regression analysis, a general equa-

tionhas beenfoundas showninEquation(7) andregression

coefﬁcients shown in Table 7.

(1 +γ k) = f

B

T

;

L

∇

1/3

;

s

L

; (1 +k)

(6)

(1 +γ k) = a

0

+a

1

B

T

+a

2

L

∇

1/3

+a

3

s

L

+a

4

(1 +k)

+ a

5

B

T

L

∇

1/3

+a

6

B

T

s

L

+a

7

L

∇

1/3

s

L

+ a

8

B

T

(1 +k) +a

9

L

∇

1/3

(1 +k)

310 SAOS 2007 Vol. 2 No. 4 Copyright C 2007 Taylor & Francis

D

o

w

n

l

o

a

d

e

d

B

y

:

[

A

u

s

t

r

a

l

i

a

M

a

r

i

t

i

m

e

C

o

l

l

e

g

e

]

A

t

:

0

4

:

5

2

2

A

p

r

i

l

2

0

0

8

Practical evaluation of resistance of high-speed catamaran hull forms—Part I

+ a

10

s

L

(1 +k) +a

11

B

T

L

∇

1/3

s

L

+ a

12

B

T

L

∇

1/3

(1 +k) +a

13

B

T

s

L

(1 +k)

+ a

14

L

∇

1/3

s

L

(1 +k) (7)

Regression following Molland et al. (1994) method

Basing on the NPL data of the studies conducted by

Mollandet al. (1994), a speedindependent regressionequa-

tion has been developed so as to obtain the wave resistance

coefﬁcient directly from the particulars of the catamaran.

The C

WC AT

is given by Equation (8) and regression

coefﬁcients as shown in Table 8.

C

WCAT

= b

0

+b

1

L

B

+a

2

B

T

+b

3

L

∇

1/3

+ b

4

s

L

+b

5

L

B

B

T

+b

6

L

B

L

∇

1/3

+ b

7

B

T

L

∇

1/3

+b

8

L

B

s

L

+b

9

B

T

s

L

+ b

10

L

∇

1/3

s

L

+b

11

L

B

B

T

L

∇

1/3

+ b

12

L

B

B

T

s

L

+b

13

L

B

L

∇

1/3

s

L

+ b

14

B

T

L

∇

1/3

s

L

+b

15

L

B

B

T

L

∇

1/3

s

L

(8)

VWS Hard Chine Series ‘89 regression methodology (1995)

This method was proposed by Zips (1995) using multiple

regression analysis of test data intended to predict the re-

sistance of hard chine catamarans with hull parameters in

the scope of the VWS Hard Chine Catamaran Hull Series

‘89. This series is valid for the ranges shown in Table 9.

The total resistance is given by:

R

T

= [R

F

+(ε

R

×∇ ×ρ × g)] (9)

The details of this methodology are illustrated in Ap-

pendix II.

Hanhirova, Rintala and Karppinen method (1995)

The authors have proposed a prediction method of esti-

mating the resistance of high-speed mono- and multihull

vessels based on Michell’s integral along with a regres-

sion correction. The regression method is based on the

resistance predicted by Michell’s integral and model ex-

periments carried out on 30 different hull shapes, several

of which were catamarans and trimarans. A signiﬁcant as-

pect of this method is that it can be applied to both mono-

and multihull vessels in the preliminary design stage. It

may be noted that the regression coefﬁcients correction to

C

W

have not been published. Regression correction was

carried out as follows:

C

T

= C

F

+C

R

+C

A

C

F

=

0.075

(Log

10

R

n

−2)

2

C

A

= 0

(10)

The experimental residual coefﬁcient was given by:

C

R

= C

T

−C

F

(11)

which was used to calculate the required correction to wave

resistance coefﬁcient C

W

as predictedbyMichell’s integral.

The required correction to wave resistance coefﬁcient was

given by:

C

W

= C

R

−C

W

(12)

The required regression equations for C

W

have been

reproduced in Appendix III.

Single chine series regression of Pham, Kantimahanthi

and Sahoo (2001)

A systematic series of 18 hard-chine demi-hulls were gen-

erated, and their wave resistance in calm water was deter-

mined using SHIPFLOW. The recorded data were then

statistically analysed to determine an accurate regression

equation. The achieved regression equation has been com-

pared with three empirical methods that have commonly

been used so far. The accuracy of the established regression

equation has been seen to deviate appreciably by various

sources of uncertainties. Veriﬁcation of the equation with

experimental database is also lacking. Further research

is, therefore, needed to reﬁne the accuracy as well as to

complete the selection of crucial parameters employed.

However, the results obtained have shown considerable

promise, and a regression equation for predicting wave re-

sistance of single chine catamaran hull forms in calm water

can be seen as achievable.

The result of the literature survey on 50 contemporary

catamaran conﬁgurations when integrated with the results

shown by Doctor et al. (1994) have led to the parameters

shown in Table 10 (Appendix IV). A parent hull form was

developed with C

B

= 0.55, L/B = 15.6 and B/T = 2.0.

Based on this hull form, a total of 18 models were devel-

oped (total including the parent hull form). The details of

the models are shown in the table above. Only the demi-

hulls were considered during hull form generation, which

were later extended to twin hulls, with demi-hulls being

symmetrical with respect to each other and with respect to

their individual centre-line planes.

With C

WC AT

as the dependent variable, and the target

vessel type being catamaran where s/L could be a sig-

niﬁcant parameter, the equation shown has been assumed

for wave resistance coefﬁcient along with the regression

311 Copyright C 2007 Taylor & Francis SAOS 2007 Vol. 2 No. 4

D

o

w

n

l

o

a

d

e

d

B

y

:

[

A

u

s

t

r

a

l

i

a

M

a

r

i

t

i

m

e

C

o

l

l

e

g

e

]

A

t

:

0

4

:

5

2

2

A

p

r

i

l

2

0

0

8

P. K. Sahoo, M. Salas and A. Schwetz

T

a

b

l

e

8

R

e

g

r

e

s

s

i

o

n

c

o

e

f

f

i

c

i

e

n

t

s

f

o

r

E

q

u

a

t

i

o

n

8

F

n

b

0

b

1

b

2

b

3

b

4

b

5

b

6

b

7

b

8

b

9

b

1

0

b

1

1

b

1

2

b

1

3

b

1

4

b

1

5

0

.

2

0

0

−

1

.

0

1

0

−

3

.

4

8

2

2

.

9

3

6

0

.

4

3

4

−

1

.

7

8

2

0

.

0

5

3

1

.

1

6

9

2

.

4

3

7

8

.

2

4

2

−

6

.

9

6

2

0

.

0

4

1

4

.

2

3

0

−

0

.

1

3

7

−

2

.

8

1

6

−

0

.

0

9

2

0

.

2

5

0

−

0

.

6

2

4

−

1

.

4

3

0

0

.

9

1

1

1

4

.

4

9

4

−

0

.

5

7

6

0

.

0

6

3

0

.

5

7

3

1

.

0

5

1

−

3

.

9

2

7

−

3

.

7

6

3

−

0

.

0

0

4

0

.

9

0

5

−

0

.

0

4

3

0

.

1

2

0

−

0

.

0

2

1

0

.

3

0

0

−

0

.

1

3

5

−

1

.

1

7

1

0

.

9

2

8

−

2

1

.

6

1

0

−

0

.

7

3

0

0

.

0

1

2

0

.

4

8

7

3

.

2

3

5

1

2

.

6

2

8

0

.

7

5

4

0

.

0

1

7

−

0

.

3

5

1

−

0

.

3

2

8

−

1

.

7

2

8

0

.

1

1

9

0

.

3

5

0

−

1

.

8

7

0

−

8

.

1

7

7

5

.

0

5

9

4

.

1

4

4

−

2

.

3

4

6

0

.

0

9

5

2

.

1

0

6

4

.

9

1

8

2

1

.

6

0

4

−

1

2

.

5

0

3

0

.

0

1

2

5

.

0

8

2

−

0

.

2

7

0

−

5

.

1

5

3

0

.

0

2

1

0

.

4

0

0

−

1

.

4

3

7

−

2

.

6

2

0

2

.

1

5

0

3

2

.

4

8

9

−

0

.

3

0

6

0

.

0

8

2

0

.

4

1

3

2

.

8

5

4

−

2

.

5

1

9

−

1

1

.

6

1

5

−

0

.

0

2

2

2

.

8

3

0

−

0

.

0

3

5

−

0

.

2

8

0

−

0

.

0

7

5

0

.

4

5

0

2

.

5

0

4

1

0

.

9

0

0

−

6

.

9

7

9

1

0

.

9

7

9

4

.

1

3

2

−

0

.

1

1

4

−

2

.

9

0

1

−

0

.

9

7

0

−

7

.

7

9

8

1

0

.

7

3

0

−

0

.

1

0

0

−

1

1

.

1

5

0

−

0

.

0

4

6

5

.

2

8

8

0

.

4

4

4

0

.

5

0

0

5

.

9

2

1

1

9

.

5

6

5

−

8

.

4

3

1

3

9

.

2

2

6

1

.

2

7

0

−

0

.

3

2

4

−

2

.

4

3

0

−

1

1

.

3

7

8

−

2

4

.

0

6

7

1

3

.

5

0

0

0

.

0

7

6

−

6

.

2

4

2

0

.

6

1

4

3

.

8

1

6

0

.

1

7

1

0

.

5

5

0

3

.

1

4

9

2

6

.

8

2

6

−

2

.

5

8

9

1

1

1

.

7

0

5

−

1

.

6

1

9

−

0

.

2

3

7

−

2

.

9

8

9

−

1

3

.

6

3

8

−

7

7

.

9

8

6

−

5

.

4

3

9

0

.

2

4

4

3

.

3

8

8

1

.

2

5

4

9

.

1

3

4

−

0

.

5

8

6

0

.

6

0

0

1

.

3

9

8

2

9

.

0

9

6

0

.

6

1

4

1

0

7

.

2

9

6

−

2

.

6

5

5

−

0

.

1

9

1

−

3

.

5

0

3

−

9

.

3

2

6

−

8

3

.

3

2

7

−

1

3

.

3

2

6

0

.

3

2

0

6

.

5

8

4

1

.

1

3

4

1

0

.

0

5

1

−

0

.

7

9

7

0

.

6

5

0

0

.

7

0

1

1

7

.

1

2

0

−

0

.

5

1

4

6

5

.

4

5

7

−

0

.

3

6

8

−

0

.

0

9

2

−

2

.

4

0

9

−

4

.

0

1

5

−

3

8

.

3

2

5

−

7

.

1

1

9

0

.

1

2

0

0

.

6

2

0

0

.

5

1

2

5

.

2

5

3

−

0

.

1

9

9

0

.

7

0

0

1

.

6

0

2

1

3

.

7

8

9

−

2

.

2

1

8

5

9

.

0

3

1

0

.

2

9

1

−

0

.

1

2

2

−

2

.

0

2

4

−

5

.

9

4

9

−

3

3

.

1

3

0

−

2

.

2

8

3

0

.

0

7

9

−

0

.

8

7

6

0

.

5

7

2

4

.

8

2

2

−

0

.

1

3

6

0

.

7

5

0

2

.

1

7

3

9

.

9

4

1

−

2

.

5

8

5

5

3

.

3

7

6

0

.

1

9

2

−

0

.

1

5

1

−

1

.

3

5

6

−

7

.

2

3

3

−

2

6

.

7

1

2

−

1

.

2

4

2

0

.

0

6

7

0

.

1

8

4

0

.

6

3

7

3

.

3

8

0

−

0

.

1

6

9

0

.

8

0

0

1

.

8

1

1

9

.

8

3

3

−

2

.

2

1

6

4

4

.

5

6

1

0

.

2

1

7

−

0

.

1

3

9

−

1

.

4

9

7

−

6

.

2

7

8

−

2

7

.

2

5

9

−

1

.

1

4

1

0

.

0

7

4

0

.

6

1

8

0

.

5

9

1

3

.

7

0

2

−

0

.

2

3

0

0

.

8

5

0

1

.

8

4

1

1

0

.

5

4

0

−

1

.

9

7

1

3

8

.

8

3

3

−

0

.

3

1

5

−

0

.

1

3

8

−

1

.

4

0

9

−

5

.

9

7

6

−

2

8

.

4

3

0

−

1

.

7

5

8

0

.

1

0

6

2

.

4

7

6

0

.

5

5

6

3

.

2

1

4

−

0

.

3

3

0

0

.

9

0

0

2

.

1

4

8

1

3

.

2

6

5

−

2

.

7

4

9

3

3

.

4

7

4

0

.

1

0

4

−

0

.

1

6

8

−

2

.

0

8

0

−

5

.

9

3

4

−

3

2

.

6

7

5

0

.

1

7

1

0

.

1

2

0

1

.

4

0

1

0

.

5

5

9

4

.

5

1

5

−

0

.

3

3

8

0

.

9

5

0

2

.

4

4

8

1

3

.

9

2

3

−

3

.

4

7

2

3

1

.

4

1

2

0

.

4

0

2

−

0

.

1

8

3

−

2

.

2

6

6

−

6

.

8

4

5

−

3

6

.

4

4

8

3

.

3

4

9

0

.

1

1

7

−

0

.

1

6

9

0

.

5

9

9

5

.

6

3

8

−

0

.

3

1

5

1

.

0

0

0

2

.

8

8

2

1

4

.

5

8

6

−

4

.

7

8

1

7

.

1

8

4

0

.

9

6

1

−

0

.

1

8

9

−

2

.

4

8

0

−

5

.

6

2

3

−

2

7

.

4

0

6

9

.

5

1

8

0

.

1

0

0

−

2

.

9

7

2

0

.

3

6

3

5

.

1

2

9

−

0

.

1

4

0

312 SAOS 2007 Vol. 2 No. 4 Copyright C 2007 Taylor & Francis

D

o

w

n

l

o

a

d

e

d

B

y

:

[

A

u

s

t

r

a

l

i

a

M

a

r

i

t

i

m

e

C

o

l

l

e

g

e

]

A

t

:

0

4

:

5

2

2

A

p

r

i

l

2

0

0

8

Practical evaluation of resistance of high-speed catamaran hull forms—Part I

Table 9 Parameter range (Zips, 1995)

Parameter Range

Length (m) 20–80

Displacement (tonnes) 25–1000

F

n

0.8–1.4

Length-Beam ratio of demi-hull L

WL

/B

XDH

7.55–13.55

Deadrise amidships β

M

16

◦

–38

o

Transom Wedge δ

W

0

o

–12

o

Table 10 Model particulars (Pham, Kantimahanthi

and Sahoo, 2001)

Models C

B

L/B B/T L/∇

1/3

WSA/L

2

β

M

M1 0.50 10.40 1.50 6.69 0.16 23.14

M2 0.50 10.40 2.50 7.93 0.12 23.20

M 3 0.50 15.60 2.00 9.67 0.09 26.68

M 4 0.50 20.80 1.50 10.62 0.08 22.96

M 5 0.50 20.80 2.50 12.58 0.06 23.25

M 6 0.55 10.40 2.00 7.13 0.14 23.80

M 7 0.55 15.60 1.50 8.49 0.11 26.43

M 8 0.55 15.60 2.00 9.35 0.09 23.80

M 9 0.55 15.60 2.50 10.08 0.08 19.15

M 10 0.55 20.60 2.00 11.33 0.07 23.80

M 11 0.60 10.40 1.50 6.30 0.17 24.53

M 12 0.60 10.40 2.50 7.47 0.13 16.21

M 13 0.60 15.60 1.50 8.24 0.11 24.02

M 14 0.60 15.60 2.00 9.09 0.09 20.58

M 15 0.60 20.80 1.50 9.98 0.08 24.02

M 16 0.60 20.80 2.50 11.86 0.06 16.21

M17 0.55 13.00 1.86 8.07 0.11 24.53

M18 0.59 17.20 1.60 9.12 0.10 24.66

Table 11 Summary of regression coefficients for

Equation 13

F

n

α β

1

β

2

β

3

β

4

0.4 2.507751 −2.255878 −1.819332 0.921796 −0.026670

0.5 2.448887 −2.424720 −1.582805 0.861936 −0.278595

0.6 2.231476 −2.442478 −1.528469 0.931836 −0.232555

0.7 1.898569 −2.402987 −1.489982 0.961013 −0.129839

0.8 1.543052 −2.351095 −1.442334 0.965683 −0.046904

0.9 1.208420 −2.308691 −1.384697 0.966650 −0.004858

1.0 0.911271 −2.279982 −1.317368 0.979194 0.004593

1.1 0.063404 −2.257688 −1.240560 0.995197 −0.004378

1.2 0.391235 −2.242743 −1.155136 1.021166 −0.017454

1.3 0.162273 −2.233282 −1.050167 1.036256 −0.027712

1.4 0.002700 −2.235047 −0.908676 1.119485 −0.031137

1.5 −0.028588 −2.268397 −0.692935 1.326583 −0.035505

coefﬁcients shown in Table 11. The body plans of demi-

hulls have been shown in Appendix IV as Figure A1.

C

WCAT

= α

L

B

β

1

B

T

β

2

C

β

3

s

L

β

4

(13)

Round bilge catamaran series of Schwetz and Sahoo (2002)

The research program was devised to:

r

examine variations in C

W

using CFD, while modifying ba-

sic hull parameters and maintaining the same displacement

and LCB position.

r

examine variations in C

W

using CFD, while modifying ba-

sic hull parameters, including the displacement and LCB.

r

compare C

W

results of CFD with results from towing tank

tests and develop regression model.

The series of symmetrical hull shapes used in this study

were generated by the authors, and are believed to closely

represent the hull forms being used in industry at the

moment. The models are not mathematical in nature, and

do not form part of any published systematic series. The

body plans of models 1 to 10 are presented in Figure 3 and

a summary of the particulars are presented in Table 12.

Following a review of current vessel dimensions, hull

model M1-RB was created, which has a round bilge

hull form and was designed to have an overall length of

50 meters, with a transom stern to accommodate two sets

of Kamewa 71 series water jets on each demi-hull. The dis-

placement was 255 tonnes, with the LCBlocated at around

42% and 44% of the waterline length, referenced from

the transom. An amount of semi-swathness was added to

model M1-RB to create models M2-SS–M4-SS, where

the amount of semi-swathness increases from models M2

to M4. Model M5-CH was generated from model M1-RB

by replacing the round bilge with a single chine, while

maintaining the same displacement and LCB. Hull model

M7-CH contains a hard chine and hull model M6-CH was

generated from model M7-CH by rounding or ﬁlleting the

hard chine, while maintaining the same displacement and

LCB as models M1–M5. Models M8-CH–M10-CH were

includedtoexamine the general effects of reducingL/∇

1/3

.

The displacements and LCB vary for models M8-CH–

M10-CH. Further details on regression equations have

been provided in Appendix V.

Round bilge catamaran series of Sahoo, Browne

and Salas (2004)

The authors have expanded on the work carried out by

Schwetz and Sahoo (2002) by conducting further work on

a systematic series of round bilge catamaran hull forms

and subjecting these to CFD analysis. The systematic se-

ries that was used for this analysis is based on typical hull

forms used by the high-speed ferry industry in Australia.

A parametric transformation procedure was used to pro-

duce the desired demi-hull series. Table 13 illustrates the

geometrical parameters of the demi-hull series developed.

For each model, hydrostatic information was extracted as

presented in Table 14, containing parameters relevant to

the regression analysis.

It may be noted that LCB and LCF locations are with

respect to the transom. The systematic series of demi-

hulls thus produced was conﬁned to s/L ratio between 0.2

and 0.4 while speed range (Fn) was constrained between

313 Copyright C 2007 Taylor & Francis SAOS 2007 Vol. 2 No. 4

D

o

w

n

l

o

a

d

e

d

B

y

:

[

A

u

s

t

r

a

l

i

a

M

a

r

i

t

i

m

e

C

o

l

l

e

g

e

]

A

t

:

0

4

:

5

2

2

A

p

r

i

l

2

0

0

8

P. K. Sahoo, M. Salas and A. Schwetz

Figure 3 Body plans of models (Schwetz and Sahoo, 2002).

Table 12 Model parameters (demi-hull)

LCB/L

from

transom LCB/

Models L/B B/T C

B

L/∇

1/3

i

E

(deg) (% ) LCF

M1-RB 15.00 2.11 0.55 9.56 8.68 42.00 1.00

M2-SS 15.20 1.79 0.49 9.56 8.66 45.00 1.12

M3-SS 15.10 1.73 0.46 9.54 3.15 42.00 1.10

M4-SS 15.00 1.71 0.46 9.53 2.10 42.00 1.20

M5-CH 15.30 2.07 0.54 9.55 9.16 42.00 0.98

M6-CH 14.80 2.31 0.66 9.18 16.60 44.00 0.96

M7-CH 14.90 2.31 0.65 9.20 13.60 44.00 0.97

M8-CH 8.80 1.47 0.52 6.30 38.00 49.00 0.99

M9-CH 10.40 1.73 0.61 7.08 15.00 47.00 1.03

M10-CH 13.00 1.77 0.68 7.60 15.00 40.00 0.92

Table 13 Systematic series of catamarans [Sahoo,

Browne and Salas (2004)]

Model 1 2 3 4 5 6 7

L/B 15 15 15 15 12.5 12.5 10

B/T 1.5 1.5 2.5 2.5 1.5 2.5 2.5

C

B

0.40 0.45 0.50 0.40 0.45 0.45 0.45

L/∇

1/3

9.45 9.08 10.40 11.20 8.04 9.54 8.22

0.2 and 1.0. The body plans of models developed during

this research study are illustrated in Figure 4. The re-

gression equations to determine the C

W

are presented in

Appendix VI.

Resistance estimation method of Subramanian

and Joy (2004)

The authors have illustrated a procedure for rapid de-

velopment of hull form and preliminary prediction of

resistance of high-speed catamarans with slender demi-

hulls. They have made use of Michell’s integral for

slender vessels to estimate the wave resistance of demi-

hulls, which combined with the average form factor

value of 1.42 (1+γ k) for ITTC ‘57 friction line would

provide the total resistance. The theoretical details of

models generated and tested numerically are shown in

Table 15.

Calculation of wave resistance coefﬁcient

The wave resistance coefﬁcients were calculated for each

hull model using SHIPFLOW, a CFDprogramdeveloped

by FLOWTECH International of Sweden.

SHIPFLOWwas developed as a pioneering effort to ad-

dress the complication of ﬂuid ﬂow characteristics around

moving objects both in fully submerged situation and in

free surface situation. Even though SHIPFLOW is in-

tended specially for marine applications, it has also been

extended to sufﬁciently solve closely related problems such

as highly turbulent ﬂow around automobiles.

The theoretical wave resistance coefﬁcient, C

W

, is

calculated by splitting the ﬂow into three regions where an

efﬁcient approximation of the ﬂow equations may be made

and a complete ﬂow calculation may be accomplished

in a few hours using the potential ﬂow, as described by

Larsson (1993). Figure 2 represents the zonal approach or

314 SAOS 2007 Vol. 2 No. 4 Copyright C 2007 Taylor & Francis

D

o

w

n

l

o

a

d

e

d

B

y

:

[

A

u

s

t

r

a

l

i

a

M

a

r

i

t

i

m

e

C

o

l

l

e

g

e

]

A

t

:

0

4

:

5

2

2

A

p

r

i

l

2

0

0

8

Practical evaluation of resistance of high-speed catamaran hull forms—Part I

Table 14 Hydrostatics of systematic series [Sahoo, Browne and Salas (2004)]

Model Length (m) Beam (m) Draught (m) Displacement () (Max.) i

E

(deg) β(deg) WSA (m

2

) LCB (m) LCF (m)

1 50.00 3.33 2.22 151.93 5.43 42.99 246.10 22.30 20.70

2 50.00 3.33 2.22 170.91 7.18 44.32 256.20 22.27 21.54

3 50.00 3.33 1.33 113.90 7.03 24.94 195.89 22.34 21.41

4 50.00 3.33 1.33 91.08 4.00 23.32 181.97 22.29 19.99

5 50.00 4.00 2.67 246.10 8.60 44.11 307.57 22.27 21.54

6 50.00 4.00 1.60 147.69 8.60 30.37 231.71 22.27 21.54

7 50.00 5.00 2.00 230.77 10.71 30.37 289.80 22.27 21.54

Figure 4 Body plans of model 1 to 7 (Sahoo, Browne and

Salas, 2004).

Table 15 Geometric parameters of vessels

[Subramanian and Joy (2004)]

Parameter L/B B/T L/V

1/3

C

B

C

P

S(m

2

)

Form 1 9.37 1.79 6.69 0.57 0.60 130.40

Form 2 8.33 1.93 6.69 0.49 0.52 131.10

Form 3 7.69 2.29 6.69 0.49 0.67 125.20

Form 4 12.50 1.88 8.92 0.48 0.67 155.50

Form 5 11.76 2.06 8.92 0.47 0.65 152.40

Form 6 1.11 2.25 8.92 0.46 0.62 151.10

Form 7 15.62 2.13 11.15 0.45 0.94 173.00

Form 8 13.89 2.57 11.15 0.43 0.92 169.20

Form 9 12.50 3.20 11.15 0.43 0.90 166.53

regions used by SHIPFLOW to maximize computational

efﬁciency.

Major areas in which SHIPFLOW has been found to

be highly applicable include calculation of ship hull resis-

tance both viscous and wave-related, development of wave

proﬁles and sequential matters consisting of trim and sink-

age characteristics, changes in velocities and pressure ﬁeld

Figure 5 Zonal distribution for ﬂuid ﬂow computation in

SHIPFLOW.

around objects such as propellers. Some of these prob-

lems remain a challenge to researchers in order to produce

more sophisticated CFD program to handle the complex

phenomenon of ﬂuid and object interactions.

According to Larsson (1993), the development of

SHIPFLOW is based on three major methods each ap-

plied in its most efﬁcient zone of ﬂuid condition:

i. Zone 1: Potential ﬂow method.

ii. Zone 2: Boundary layer method.

iii. Zone 3: Navier–Stokes method.

The laminar ﬂow starts from the stagnation point, di-

verge gradually as it moves downstream, and when they

reach the transition point where the viscous force is insuf-

ﬁciently strong to bond the streamlines, it breaks down and

becomes turbulent.

Potential ﬂow method is used to analyse the ﬂuid-ﬂow

in the outermost area of the free surface designated as

Zone 1 in Figure 5. In this zone the ﬂuid-ﬂow is treated as

continuous streamlines starting from fore end of the ship,

and extending up to the aft end. The region of free surface

that describes the thin boundary layers along the ship hull

is deﬁned as Zone 2. The nature of ﬂuid ﬂow change as

the ﬂuid moves along the hull in this region. The boundary

layer theory is used to compute the ﬂuid characteristics in

zone 2.

The remaining region of the free surface is fully tur-

bulent and will have wakes. It is speciﬁed as zone 3 and

extending far aft fromthe transition point, which is usually

about amidships. Navier–Stokes theory is applied in this

zone to calculate the energy and hence the corresponding

resistance incurred.

315 Copyright C 2007 Taylor & Francis SAOS 2007 Vol. 2 No. 4

D

o

w

n

l

o

a

d

e

d

B

y

:

[

A

u

s

t

r

a

l

i

a

M

a

r

i

t

i

m

e

C

o

l

l

e

g

e

]

A

t

:

0

4

:

5

2

2

A

p

r

i

l

2

0

0

8

P. K. Sahoo, M. Salas and A. Schwetz

The free group represents the free surface. The free sur-

face can be considered as the water surrounding the body

group, which is used to model the waves created. There-

fore, it is necessary to create a free surface that extends

forward of the bow, well aft of the model (approximately

two wave lengths), and a considerable distance abeam of

the vessel. After the limits of the free surface have been

introduced, it is necessary to once again deﬁne the num-

ber of stations along the length of the free surface and the

number of points across each station to create the grid.

The transom group represents a part of the free surface,

which extends directly aft of the transom. This group is,

therefore, quite long and only as wide as the vessel. As in

the previous section, it is necessary to deﬁne the number

of stations and points required to produce a grid. For con-

sistency, the number of stations aft of the body must be the

same for the free surface as it is for the transom group so

that the panels are aligned.

The module XPAN is the solver that iteratively con-

verges on the value of coefﬁcient of wave resistance. It is,

therefore, necessary to input the maximum number of it-

erations that are to be used. In addition to this, the type of

solver that will be used must be speciﬁed. The non-linear

solver will generally produce a more accurate result than

the linear solver; however, it is more unstable particularly

at high speeds and the solution may not converge. If ref-

erence is not made to the type of solver, then the linear

solver is used as the default. The other important feature

of XPANis whether the model is enabled to freely sink and

trim. It is important to note that SHIPFLOW undergoes

its analysis by non-dimensionalising the vessel down to a

model of unit length. Therefore, all of the co-ordinates are

non-dimensionalised by the length between perpendicu-

lars L

PP

. As mentioned XMESH module enables the user

of SHIPFLOWto construct a gridof panels to illustrate the

scenario to be tested. Due to the ﬂexibility of SHIPFLOW

to be applied to many different applications, it can produce

varying results, which will not match model testing or full-

scale data. The program will produce an accurate result of

coefﬁcient of wave resistance based on the grid supplied;

however, if the grid is not well set-up the result does not

have much validity.

One of the major limitations of SHIPFLOW is its in-

ability to model spray and wave-breaking phenomena at

high speeds with a Froude number of 1.0 considered as

the upper limit. Therefore, the investigation has been re-

stricted to this speed. When considering the validity of

results there are twokeyaspects, the precisionandthe accu-

racy. If SHIPFLOW is used correctly very precise results

may be obtained, however, these results cannot be con-

sidered as accurate until they have been scaled according

to some model testing or full-scale data. Therefore, when

constructing the grid in SHIPFLOW the aim is to achieve

precise results, which can then be altered for accuracy.

At low Froude numbers the transom wave has a small

wavelength and a large wave height. Conversely, at high

Froude numbers the transom wave has a large wavelength

with small wave amplitude. Therefore, if a constant grid

is applied to all of the models at the full range of speeds

the degree of precision varies. Therefore, caution must

be taken when comparing results at different speeds. To

overcome this problem, the grid must be systematically

altered as the speed is increased to take into account the

larger wavelength. This was achieved by increasing the free

and transom surfaces further aft until two wavelengths are

included as a guideline. On the other hand, at lower speeds

it is not necessary to extend the free and transom surfaces

further aft of the body group, but it will be necessary

to include smaller panels in the grid to account for the

signiﬁcant changes in wave height.

If the grid is not altered it can be expected that as the

Froude number is increased the results can be considered

as becoming increasingly precise. However, as previously

mentioned, when the speed is increased SHIPFLOW be-

comes increasingly unstable in its ability to model spray

and wave breaking phenomena. Therefore, using this soft-

ware is a balance of stability and precision and to produce

valid results an extensive amount of time is required to

analyse the different scenarios. The change in grid den-

sity was applied to this analysis to account for changes in

Froude numbers.

RESULTS

Only three vessels from the series tested by Molland et al.

(1994) coincided with the series produced for this analysis.

It is observed that above a Froude number of 0.5 the corre-

lation of results is extremely good. The NPL series tested

by Molland et al. (1994) is based on a mono-hull series that

has been put into a catamaran conﬁguration. The other

problem with using this work is the insufﬁcient informa-

tion available in their paper on the hydrostatics of each

model. The value especially for dead-rise amidships and

half angle of entrance have beenkept constant at 7

◦

and30

◦

,

respectively, for NPL models and regression analysis for

comparison purposes, and therefore, a closer correlation

would hopefully exist when the input is more accurate.

It is interesting to note the variables that have been in-

cluded in the different equations for each Froude number

for a catamaran and demi-hull. The main variable that be-

comes apparent as having the most inﬂuence on resistance

is the wet volume slenderness ratio (L/∇

1/3

). Half angle of

entrance, dead-rise and separation ratio are also signiﬁcant

throughout the speed range. The breadth to draft ratio

becomes signiﬁcant only at the higher speeds.

It is interesting to note the variables that have been in-

cluded in the different equations for each Froude number

for a catamaran and demi-hull. The main variable that be-

comes apparent as having the most inﬂuence on resistance

is the slenderness ratio (L/∇

1/3

).

The form factor due to viscous resistance interference

factor is another aspect of catamaran resistance that could

be further analysed. The work by Armstrong (2000) is

limited to the applicable range of lowFroude numbers that

316 SAOS 2007 Vol. 2 No. 4 Copyright C 2007 Taylor & Francis

D

o

w

n

l

o

a

d

e

d

B

y

:

[

A

u

s

t

r

a

l

i

a

M

a

r

i

t

i

m

e

C

o

l

l

e

g

e

]

A

t

:

0

4

:

5

2

2

A

p

r

i

l

2

0

0

8

Practical evaluation of resistance of high-speed catamaran hull forms—Part I

Table 16 Parameters of a chine hull catamaran model

vessel (tonne) 1.369 × 10

−2

demi (tonne) 6.844 × 10

−3

∇demi (m

3

) 6.678 × 10

−3

L(m) 1.600

T(m) 0.073

B(m) 0.145

Separation, s (m) 0.48

Half angle of entrance, i

E

(

◦

) 8

Dead rise angle β

M

(

◦

) 30

LCB (m) 0.92

LCF (m) 0.95

C

B

0.397

Gravitational constant, g (m/s

2

) 9.81

υ

SW

(m

2

/s) 1.1881 × 10

−6

ρ

SW

(t/m

3

) 1.025

WSA demi (m

2

) 0.276

WSA cat (m

2

) 0.553

C

A

4.00 ×10

−4

Transom wedge 10

◦

L/∇

1/3

8.497

B/T 2.000

LCB/WL 10.00

LCB/LCF 0.97

s /L 0.300

L/B 11.0

can be used. Therefore, if a similar analysis was undertaken

with carefully monitored SHIPFLOW and model testing

results, an equation for form factor of catamarans could be

produced. This seems to be the least researched aspect of

determining catamaran resistance.

We have attempted to demonstrate here an example of

the results obtained so as to summarise and compare the

different methods. For different values of the catamaran

parameters, the program allows us to plot the graphs of

the predicted wave resistance coefﬁcients versus Froude

number andof the power predictedversus velocity. Results

Figure 6 Comparison of C

W

against C

R

values of Zips (1995).

are split into two, so as to differentiate the single hard chine

hulls and the round bilge hulls.

All the methods explained here have been integrated

into the program, except the ones for which data were not

available (Millward’s method, 1992, Hanhirova, Rintala

and Karppinen method, 1995 and Subramanian and Joy

method, 2004).

The results for a model catamaran with chine hull (par-

ticulars shown in Table 16) have been shown in Figure 6.

The results obtained from the different methods are

very close and reliable. So these methods can easily be used

for a predimensioning. The only limit of these methods is

the range of the catamaran parameters. Indeed they are

different for each method, and if they are not respected

the results can lose their numerical reliability. Figure 7

represents a comparative analysis between the round bilge

catamaran resistance undertaken by the authors and the

NPL-hull form catamaron series investigated by Molland

et al. (1994).

Final remarks

The variation of grid density must be very precise in or-

der to obtain accurate results over the whole speed range.

Therefore, time must be spent before analysis to set-up a

grid for each Froude number so that the results are con-

sistently precise. In order to perform suitable analysis on

round bilge catamaran hull forms the constraints as shown

in Table 10 should be strictly adhered to.

In order to calculate the total resistance, if actual data is

unavailable, the following empirical formulae may be used:

Wetted Surface Area:

S = 1.7LT +

∇

T

m

2

as per Mumford (14)

S =

∇

B

¸

1.7

C

B

−0.2(C

B

−0.65)

+

B

T

¸

m

2

(15)

317 Copyright C 2007 Taylor & Francis SAOS 2007 Vol. 2 No. 4

D

o

w

n

l

o

a

d

e

d

B

y

:

[

A

u

s

t

r

a

l

i

a

M

a

r

i

t

i

m

e

C

o

l

l

e

g

e

]

A

t

:

0

4

:

5

2

2

A

p

r

i

l

2

0

0

8

P. K. Sahoo, M. Salas and A. Schwetz

Figure 7 Comparison of round bilge catamaran hull forms against results of catamaran hull forms of Molland et al. (1994).

Figure 8 C

B

against L/B as used by various authors.

Finally the Figure 8 has been illustrated to show the

range of C

B

values against L/B ratios used by various

authors to arrive at the regression equations. It may be

noted that the catamaran hull forms investigated by Insel

and Molland (1992) and that of Molland et al. (1994) have

a constant C

B

value of 0.397. It can be seen from the

ﬁgure that the range of applicability have been signiﬁcantly

enhanced by inclusion of higher values of C

B

.

CONCLUSIONS AND DISCUSSION

In viewof the analysis and validation process undertaken in

this research work the following conclusions can be drawn:

r

A systematic series of round bilge catamaran hull forms

have been designed and wave resistance coefﬁcients for

demi-hull as well as catamaran hull form conﬁgurations

have been determined using CFD (SHIPFLOW) for a

range of Froude numbers.

r

A regression analysis has been performed based on CFD

results and has been compared with experimental results

of NPL series hull forms as conducted by Molland et al.

(1994).

r

It appears that the regression equation is robust enough

as it compares favourably, especially at 0.5 < F

n

< 1.0,

with experimental results for the three NPLmodels whose

geometrical parameters closely matchthe constraints of the

regression analysis.

318 SAOS 2007 Vol. 2 No. 4 Copyright C 2007 Taylor & Francis

D

o

w

n

l

o

a

d

e

d

B

y

:

[

A

u

s

t

r

a

l

i

a

M

a

r

i

t

i

m

e

C

o

l

l

e

g

e

]

A

t

:

0

4

:

5

2

2

A

p

r

i

l

2

0

0

8

Practical evaluation of resistance of high-speed catamaran hull forms—Part I

r

It is expected that in Part II, the paper will attempt

to validate the theoretical results against experimen-

tal results from a random selection of catamaran hull

forms.

It is imperative to note that a limited number of models

have been created in this instance, which implies that the

range could be further enlarged and more rigorous valida-

tion is required against experimental results.

ACKNOWLEDGMENTS

The authors would like to express their sincere gratitude to

the Australian Maritime College, Australia and the Uni-

versity Austral, Chile for their support, encouragement

throughout the course of this research work. They also

want to extend sincere gratitude to Nick Browne of BMT

Defence Services, Australia and David Pontaut from the

ENSIETA Engineering School (Brest, France), who have

done some of the regressions (catamaran form factor and

NPL series) and developed the spread sheets which calcu-

late the wave resistance coefﬁcient and the power predic-

tion for several methods (NPL Series, VWS Hard Chine,

Single Chine, 2001 and Round Bilge, 2002). Without their

valuable contributions this paper would not have seen the

light of the day.

REFERENCES

Armstrong NA. 2000. On the Viscous Resistance and Form Factor of

High-Speed Catamaran-Ferry Hull Forms, Ph.D Thesis, The

University of New South Wales, Sydney, Australia, pp. 1–

227.

Drewry Shipping Consultants. 1997. FAST FERRIES: Shaping

the Ferry Market for the 21st Century. Drewry Shipping

Consultants, London, UK.

Hanhirova K, Rintala S, Karppinen T. 1995. Preliminary resistance

prediction method for fast mono- and multihull vessels, In

Proceeding of High-Speed Vessels for Transport and

Defence,Royal Society of Arts, London, UK.

Insel M, Molland AF. 1992. An investigation into resistance

components of high-speed displacement catamarans, Trans. of

Royal Institute of Naval Architects, 134:1–20.

Larsson, L. 1993. Resistance and ﬂow predictions using

SHIPFLOW code, 19th WEGEMNT school, Nantes,

France.

Molland AF, Wellicome JF, Couser PR. 1994. Resistance

experiments on a systematic series of high speed displacement

catamaran forms: Variation of length-displacement ratio and

breadth-draft ratio, Ship Science Report No.71, University of

Southampton, UK.

Pham XP, Kantimahanthi K, Sahoo PK. 2001. Wave resistance

prediction of hard-chine catamarans through regression

analysis, In Proceedings of 2nd International Euro Conference

on High Performance Marine Vehicles (HIPER’01),

Hamburg, Germany, pp. 382–394.

Schwetz A, Sahoo PK. 2002. Wave resistance of semi-displacement

high speed catamarans through CFD and regression analysis,

In Proceedings of 3rd International Euro Conference on High

Performance Marine Vehicles (HIPER’02), Bergen, Norway,

pp 355–368.

Subramanian VA, Joy P. 2004. A method for rapid hull form

development and resistance estimation of catamarans, Trans.

of Marine Technology, Spring, 38(1):5–11.

SHIPFLOW User’s Manual, Flowtech International, Edition 1,

December 2003.

Zips JM. 1995. Numerical resistance prediction based on results of

the vws hard chine catamaran hull series ‘89, In Proceedings

of 4th International Conference on Fast Sea Transportation

(FAST ‘95), Luebeck, Germany, Vol 1, pp. 67–74.

319 Copyright C 2007 Taylor & Francis SAOS 2007 Vol. 2 No. 4

D

o

w

n

l

o

a

d

e

d

B

y

:

[

A

u

s

t

r

a

l

i

a

M

a

r

i

t

i

m

e

C

o

l

l

e

g

e

]

A

t

:

0

4

:

5

2

2

A

p

r

i

l

2

0

0

8

P. K. Sahoo, M. Salas and A. Schwetz

APPENDIX I

As per Armstrong (2000) the form factors are given by

Parameter Equation Validity

1+k 0.65 + 350(F

n

)

−0.67

(B/T)

0.15

(L/B)

−1.5

(L/∇

1/3

)

−1.5

0.5 < F

n

< 1.0

6 < L/∇

1/3

< 8

12 < L/B < 15

1.5 < B/T < 2.5

(1+k)

model

1.45 − 0.139 (L/∇

1/3

)

0.6

(B/T)

−0.1

3 ×10

6

< R

n

< 5×10

6

6.5 < L/∇

1/3

< 9.5

1.5 <B/T < 2.5

0.6 < F

n

< 1.0

(1+k)ship 1.72 − f(L/∇

1/3

)

g

(B/T)

−0.1

10

9

< R

n

< 2×10

9

f −2.25Fn

2

+ 4.47F

n

− 1.61 F

n

< 1.0

f 0.61 F

n

> 1.0

g 0.76 − 1.09f

APPENDIX II

Detailed regression analysis had been performed on this se-

ries (Zips 1995)) and in order to carry out resistance predic-

tion some of the parameters have been reproduced below:

r

Variables deﬁned as X

1

=(L

WL

/B

XDH

−10.55)/3, X

2

=

(β

M

− 27

◦

)/11 and X

3

= δ

W

/12

r

Length–displacement ratio of demi-hull is given by:

L

WL

(∇/2)

1/3

= 7.651877 +1.694413 ∗ X1

+0.282139 ∗ X1

2

−0.052496 ∗ X1

2

∗ X2

r

Wetted surface area coefﬁcient is given by:

S

W(V=0)

∇

2/3

= CS∗

XS∗ 10

Table 17 Matrix of regression coefficients C

R

(Zips, 1995)

F

n∇/2

1.00 1.25 1.50 1.75 2.00 2.50 3 3.50

2.348312 4.629531 5.635988 5.627470 5.690865 6.209794 7.243674 7.555179

−0.706875 −2.708625 −2.371713 −2.266895 −2.500808 −2.900769 −3.246017 −2.647421

0.272668 −0.447266 −0.328047 −0.428999 −0.422339 −0.391296 0.000000 0.453125

0.586730 0.000000 0.000000 0.000000 −0.288437 −0.447655 0.000000 0.000000

0.256967 0.701719 0.349687 0.416250 0.571875 0.832031 0.554213 0.332042

0.000000 0.000000 0.165938 0.342187 0.496875 0.656719 1.196250 1.884844

0.000000 0.148359 0.000000 0.000000 0.000000 0.000000 0.000000 −0.276875

0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 −1.877430 0.000000

C

R

= −0.152163 0.000000 −0.251026 −0.429128 −0.450245 −0.866917 0.000000 0.000000

0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 −1.036289

0.000000 0.149062 0.090188 0.089625 0.076125 0.000000 −0.332250 −0.767250

−0.151312 −0.090188 −0.135563 −0.194250 −0.190125 −0.225938 −0.211125 0.000000

−0.059200 −0.322734 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000

0.000000 −0.148359 −0.096328 0.000000 0.000000 0.000000 0.000000 0.000000

0.000000 0.409500 0.484800 0.000000 0.817200 1.189350 1.007700 0.000000

0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.588758 0.683505

0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 −0.241426

0.000000 0.000000 0.000000 0.704463 0.000000 0.000000 0.000000 0.000000

−0.083789 0.000000 0.000000 0.120516 0.137585 0.257507 0.000000 0.000000

CS =

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

1.103767

0.151489

0.00983

−0.009085

0.008195

−0.029385

0.041762

XS =

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

1

X1

X2

2

X1

2

X1

2

X2

X1X2

2

X1

3

X2

r

Residual drag-weight ratio is given by: ε

R

(F

n∇/2

) =

(XR

T

∗CR)

100

r

Vector of regression parameter XR is given by:

320 SAOS 2007 Vol. 2 No. 4 Copyright C 2007 Taylor & Francis

D

o

w

n

l

o

a

d

e

d

B

y

:

[

A

u

s

t

r

a

l

i

a

M

a

r

i

t

i

m

e

C

o

l

l

e

g

e

]

A

t

:

0

4

:

5

2

2

A

p

r

i

l

2

0

0

8

Practical evaluation of resistance of high-speed catamaran hull forms—Part I

XR =

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

X0 = 1

X1 = f (L

WL

/B

XDH

)

X2 = f (β

M

)

X3 = f (δ

W

)

X4 = X1

2

X5 = X1

3

X6 = X2

2

X7 = X3

1/2

X8 = X3

1/3

XR =

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

¸

X9 = X3

1/4

X10 = X1 ∗ X2

X11 = X1 ∗ X6

X12 = X4 ∗ X6

X13 = X4 ∗ X2

X14 = X1 ∗ X3

X15 = X4 ∗ X7

X16 = X1 ∗ X9

X17 = x1 ∗ X7

3

X18 = X2 ∗ X8

r

Procedure for calculation is as follows:

• Given the following parameters L

WL

/B

XDH

trans-

formed to X1, β

M

transformed to X2 and δ

W

trans-

formed to X3, ∇, C

A

, ρ and ν

• Determine

L

WL

(∇/2)

1/3

,

S

W(V=0)

∇

2/3

and ε

R

(F

n∇/2

)

• R

n

=

L

WL

(∇/2)

1/3

× F

n∇/2

×(

∇

2

g)

1/2

ν

• C

F

=

0.075

(log R

n

−2)

2

• R

F

=

(C

F

+C

A

) ρF

2

n∇/2

×

S

W(V=0)

∇

2/3

×∇ × g ×2

2/3

4

• R

T

= R

F

+(ε

R

×∇ ×ρ × g)

APPENDIX III

The wave resistance coefﬁcient correction formula for low

L/B mono-hulls is:

C

W

= C

R

−C

W

= C

0

+C

1

cos

λF

e

n

B

AP

B

X

+C

2

F

e

n

C

∇

exp

a

F

2

n

+C

3

C

2

∇

cos

λF

e

n

+C

4

C

∇

C

P

+C

5

F

e

n

cos

λF

e

n

log (90 −i

E

)

+C

6

¸

cos

λF

e

n

¸

2

C

∇

+C

7

F

e

n

C

X

exp

a

F

2

n

+C

8

(C

∇

)

2

B

X

T

X

+C

9

[log (90 −i

E

)]

3

+C

10

F

e

n

cos

λF

e

n

+C

11

F

e

n

cos

λF

e

n

C

P

+C

12

¸

exp

a

F

2

n

¸

2

T

AP

T

X

+ C

13

(C

X

)

2

T

AP

T

X

where λ = a

1

C

P

+a

2

C

∇

.

The wave resistance coefﬁcient correction formula for

high L/B mono-hulls is:

C

W

= C

R

−C

W

= C

0

+C

1

F

n

sin

λF

e

n

+C

2

¸

cos

λF

e

n

¸

2

A

AP

A

X

+C

3

¸

sin

λF

e

n

¸

2

C

P

+C

4

¸

sin

λF

e

n

¸

2

T

AP

T

X

+C

5

F

3

n

+C

6

(C

∇

)

2

cos

λF

e

n

+C

7

C

2

P

F

n

+C

8

A

AP

A

X

2

F

n

+C

9

sin

λF

e

n

T

AP

T

X

2

+C

10

cos

λF

e

n

T

AP

T

X

2

+C

11

T

AP

T

X

1

F

2

n

+C

12

[log (90 −i

E

)]

3

+C

13

cos

λF

e

n

C

∇

A

AP

A

X

The wave resistance coefﬁcient correction formula for

catamaran is:

C

W

= C

R

−C

W

= C

0

+C

1

(C

∇

)

2

1

F

n

+C

2

C

P

[log (90 −i

E

)] +C

3

¸

cos

λF

e

n

¸

2

B

AP

B

X

+C

4

C

2

P

log (90 −i

E

) +C

5

C

3

P

+C

6

T

AP

T

X

2

sin

λF

e

n

+C

7

B

AP

B

X

2

sin

λF

e

n

+C

8

C

2

P

+C

9

C

2

∇

+C

10

¸

sin

λF

e

n

¸

2

cos

λF

e

n

+C

11

cos

λF

e

n

sin

λF

e

n

exp

−

w

B

+C

12

exp

−

w

B

2

C

∇

+C

13

¸

cos

λF

e

n

¸

2

T

AP

T

X

+C

14

¸

cos

λF

e

n

¸

2

sin

λF

e

n

+C

15

C

2

∇

sin

λF

e

n

+C

16

sin

λF

e

n

1

F

n

+C

17

C

2

∇

exp

a

F

2

n

+C

18

C

2

∇

exp

−

w

B

+C

19

cos

λF

e

n

exp

−

w

B

T

AP

T

X

+C

20

¸

exp

−

w

B

¸

2

321 Copyright C 2007 Taylor & Francis SAOS 2007 Vol. 2 No. 4

D

o

w

n

l

o

a

d

e

d

B

y

:

[

A

u

s

t

r

a

l

i

a

M

a

r

i

t

i

m

e

C

o

l

l

e

g

e

]

A

t

:

0

4

:

5

2

2

A

p

r

i

l

2

0

0

8

P. K. Sahoo, M. Salas and A. Schwetz

APPENDIX IV

Figure A1 Body plans of demi-hull models as per Pham, Kantimahanthi and Sahoo (2001).

322 SAOS 2007 Vol. 2 No. 4 Copyright C 2007 Taylor & Francis

D

o

w

n

l

o

a

d

e

d

B

y

:

[

A

u

s

t

r

a

l

i

a

M

a

r

i

t

i

m

e

C

o

l

l

e

g

e

]

A

t

:

0

4

:

5

2

2

A

p

r

i

l

2

0

0

8

Practical evaluation of resistance of high-speed catamaran hull forms—Part I

Table 18 Range of parameters for Equation 16

Geometric

Parameters L/∇

1/3

LCB/LCF C

B

F

n

Range of 9.2 to 9.6 0.97 to 1.2 0.46 to 0.66 0.4 to 1.4

Application

Table 19 Coefficients for Equation 16

F

n

C

1

C

2

C

3

C

4

R

2

0.4 1.54 ×10

2

−5.058 −0.305 0.000 0.96

0.5 3.00 ×10

2

−5.519 −0.466 −0.1339 0.98

0.6 3.61 ×10

2

−5.715 −0.488 −0.1154 0.99

0.7 6.87 ×10

2

−6.113 −0.591 −0.108 0.99

0.8 1.81 ×10

3

−6.637 −0.648 −0.0981 0.99

0.9 4.83 ×10

3

−7.155 −0.775 −0.0933 0.99

1 2.99 ×10

4

−8.064 −0.982 −0.1907 0.99

1.1 1.97 ×10

5

−8.995 −1.191 −0.292 0.99

1.2 1.39 ×10

6

−9.932 −1.309 −0.3178 0.99

1.3 5.03 ×10

6

−10.551 −1.392 −0.2913 0.99

1.4 5.50 ×10

7

−11.69 −1.543 −0.3903 0.98

APPENDIX V

Some of the details of calculatingthe demi-hull andcatama-

ran resistance (Schwetz and Sahoo, 2002) are reproduced

below. The wave resistance coefﬁcient for a demi-hull can

be predicted from Equation (16), whose validity range is

shown in Table 18, using the constants C

1

to C

4

fromTable

19.

C

Wdemi

= C

1

L/∇

1/3

C

2

(LCB/LCF)

C

3

C

C

4

B

(16)

The wave resistance coefﬁcient for a catamaran can be

predicted from Equation (17), using the constants C

1

to C

7

Table 20 Coefficients for Equation 17

F

n

C

1

C

2

C

3

C

4

C

5

C

6

C

7

R

2

0.5 2.151 × 10

7

−12.180 −0.195 −0.042 0.018 2.818 −3.398 0.95

0.6 8.509 × 10

3

−8.207 −0.235 0.000 0.000 1.942 −2.035 0.97

0.7 2.194 × 10

4

−8.840 −0.180 −0.073 0.027 1.992 −2.215 0.92

0.8 5.508 × 10

4

−9.388 −0.106 −0.182 0.042 2.012 −2.327 0.92

0.9 1.488 × 10

5

−9.938 −0.046 −0.285 0.050 2.029 −2.399 0.96

1.0 1.303 × 10

4

−8.590 −0.016 −0.422 0.026 1.583 −1.757 0.98

1.1 5.438 × 10

3

−8.002 0.023 −0.403 −0.012 1.303 −1.264 0.96

1.2 8.261 × 10

6

−12.005 0.015 −0.164 0.020 2.302 −2.473 0.97

1.3 1.440 ×10

10

−16.090 0.004 0.072 0.102 3.230 −3.690 0.99

1.4 1.965 ×10

12

−18.571 0.003 0.348 0.116 3.649 −4.052 0.99

Table 21 Range of parameters for Equation 18

Geometric Parameters L/∇

1/3

LCB/LCF B/T C

B

F

n

Range of Application 6.3 to 9.6 0.92 to 1.2 1.47 to 2.3 0.46 to 0.68 0.4 to 1.4

Table 22 Coefficients for Equation 18

F

n

C

1

C

2

C

3

C

4

C

5

R

2

0.5 0.30 −1.2168 −2.2795 −2.5075 1.4337 0.96

0.6 0.41 −1.4599 −1.9655 −2.4304 1.5754 0.98

0.7 0.68 −2.1421 −1.6111 −1.6934 1.1637 0.99

0.8 0.78 −2.4272 −1.5211 −1.4089 1.0263 0.99

0.9 0.87 −2.6947 −1.5148 −1.1202 0.8731 0.98

1.0 0.93 −2.9213 −1.5536 −0.8650 0.7080 0.98

1.1 1.00 −3.1409 −1.5821 −0.6142 0.5526 0.98

1.2 1.16 −3.3948 −1.5593 −0.3228 0.4110 0.97

1.3 1.38 −3.6728 −1.5278 0.000 0.2509 0.97

1.4 1.65 −3.9787 −1.5547 0.35234 0.000 0.97

from Table 20.

C

Wc at

= C

1

L/∇

1/3

C

2

(s /L)

C

3

(LCB/LCF)

C

4

i

C

5

E

(B/T)

C

6

C

C

7

B

(17)

The generalized wave resistance coefﬁcient for a demi-

hull can be predicted from Equation (18), whose validity

range is shown in Table 21, using the constants C

1

to C

5

from Table 22.

C

Wdemi

= C

1

L/∇

1/3

C

2

(LCB/LCF)

C

3

(B/T)

C

4

C

C

5

B

(18)

The generalized wave resistance coefﬁcient for a cata-

maran can be predicted from Equation (19), using the con-

stants C

1

to C

7

from Table 23.

C

Wc at

= C

1

L/∇

1/3

C

2

(s /L)

C

3

(LCB/LCF)

C

4

i

C

5

E

·C

C

6

B

(B/T)

C

7

(19)

323 Copyright C 2007 Taylor & Francis SAOS 2007 Vol. 2 No. 4

D

o

w

n

l

o

a

d

e

d

B

y

:

[

A

u

s

t

r

a

l

i

a

M

a

r

i

t

i

m

e

C

o

l

l

e

g

e

]

A

t

:

0

4

:

5

2

2

A

p

r

i

l

2

0

0

8

P. K. Sahoo, M. Salas and A. Schwetz

Table 23 Coefficients for Equation 19

F

n

C

1

C

2

C

3

C

4

C

5

C

6

C

7

R

2

0.5 1.501 −2.632 −0.201 −1.554 −0.132 1.070 −1.460 0.99

0.6 1.122 −2.817 −0.305 −1.265 −0.090 0.971 −1.259 0.99

0.7 0.613 −2.734 −0.278 −1.290 −0.064 0.988 −1.317 0.99

0.8 0.282 −2.652 −0.195 −1.472 −0.052 0.996 −1.395 0.99

0.9 0.209 −2.668 −0.111 −1.645 −0.048 1.002 −1.422 0.99

1.0 0.356 −2.820 −0.056 −1.756 −0.052 0.964 −1.339 0.99

1.1 0.878 −3.129 0.000 −1.640 −0.068 0.974 −1.171 0.99

1.2 1.455 −3.476 0.000 −1.365 −0.092 1.051 −0.962 0.99

1.3 1.594 −3.615 0.000 −1.105 −0.069 1.179 −0.873 0.99

1.4 2.337 −4.056 −0.032 −0.658 −0.072 1.338 −0.614 0.99

APPENDIX VI

C

Wdemi

= e

C

1

L

B

C

2

(C

B

)

C

3

L

∇

1/3

C

4

(i

E

)

C

5

(β)

C

6

(20)

C

Wc at

= e

C

1

L

B

C

2

B

T

C

3

(C

B

)

C

4

L

∇

1/3

C

5

(i

E

)

C

6

(β)

C

7

s

L

C

8

(21)

Table 24 Regression coefficients and R

2

for demi-hull configuration for Equation 20

Fn C

1

C

2

C

3

C

4

C

5

C

6

R

2

0.2 3.001 −0.159 0.515 −3.666 −0.194 0.000 0.967

0.3 1.221 0.000 0.815 −3.445 0.218 0.000 0.985

0.4 3.180 −0.702 0.377 −3.114 −0.390 0.000 1.000

0.5 2.519 0.396 −0.775 −4.175 0.000 −0.410 0.999

0.6 2.031 −0.239 0.000 −3.402 −0.138 −0.091 0.999

0.7 1.130 −0.220 0.000 −3.221 −0.043 −0.081 0.999

0.8 0.600 −0.272 0.000 −3.079 0.000 −0.063 0.999

0.9 −0.216 0.000 −0.228 −3.158 0.173 −0.178 0.999

1.0 −1.086 0.000 −0.396 −2.965 0.300 −0.203 0.998

Table 25 Regression coefficients and R

2

for catamaran configuration for Equation 21

Fn C

1

C

2

C

3

C

4

C

5

C

6

C

7

C

8

R

2

0.2 2.571 0.436 0.000 0.000 −4.124 −0.039 −0.199 0.037 0.995

0.3 0.585 0.000 0.000 0.945 −3.282 0.246 0.087 −0.089 0.989

0.4 3.324 0.000 −0.471 −0.963 −3.523 0.000 −0.688 −0.035 0.984

0.5 2.439 0.379 0.000 −0.600 −4.262 0.000 −0.337 −0.368 0.999

0.6 1.809 −0.110 0.000 0.000 −3.625 −0.061 −0.095 −0.314 0.997

0.7 1.055 0.000 0.082 −0.025 −3.617 0.000 −0.064 −0.181 0.997

0.8 0.603 0.222 0.266 0.000 −3.869 0.000 0.000 −0.069 0.998

0.9 −0.466 0.049 0.162 0.000 −3.322 0.128 0.000 −0.006 0.999

1.0 −1.221 0.000 0.117 0.000 −3.046 0.264 0.000 0.075 0.995

324 SAOS 2007 Vol. 2 No. 4 Copyright C 2007 Taylor & Francis

needed to reﬁne the accuracy as well as to complete the selection of crucial parameters employed. ABBREVIATIONS NOTATION CFD LCB LCF DWL ITTC WSA Computational Fluid Dynamics Longitudinal center of buoyancy. propulsion and seakeeping. Further research is.Downloaded By: [Australia Maritime College] At: 04:52 2 April 2008 Practical evaluation of resistance of high-speed catamaran hull forms—Part I doi:10. wave resistance. and a regression equation for predicting wave resistance of catamarans in calm water can be seen as achievable. Australia Tel. However. Sahoo Maritime Engineering Australian Maritime College Newnham Drive Launceston TAS 7250.T ITTC ‘57 ship-model correlation line C F = (Log0. 4 pp. Even though considerable amount of research has been carried out in this area. computational ﬂuid dynamics.sahoo@amc. 2 No. semi-displacement slender catamaran hull forms of round bilge as well as single chine hull forms. Veriﬁcation of the equation with experimental database is also lacking to a certain extent. The accuracy of the established regression equation has been seen to deviate appreciably by various sources of uncertainties. Australia Abstract: Although catamaran conﬁguration has been around for a longtime.B. Launceston. the results obtained have shown considerable promise. Australia University of Austral. reference from the transom Longitudinal center of ﬂoatation. Chile 3 Schwetz Design. resistance. there remains a degree of uncertainty in the prediction of calm water resistance of catamaran hull forms. therefore. This paper attempts to present the research work carried out so far and what needs to be undertaken in future for a reasonably accurate prediction of catamaran resistance characteristics.: +61 363354822 Fax: +61 363354720 Email: p. reference from the transom Design waterline International Towing Tank Conference Wetted Surface Area A AP AF P AX BX B B/T CA CB CF CP CM CWc at.075 2 10 Rn−2) Prismatic coefﬁcient Midship coefﬁcient Wave resistance coefﬁcient Residuary resistance coefﬁcient Total resistance coefﬁcient Volumetric coefﬁcient Wave resistance coefﬁcient correction Froude number (based on length) Froude number based on volumetric displacement of demi-hull SAOS 2007 Vol. CWd emi CR CT C∇ CW Fn Fn∇/2 Corresponding Author: Prasanta K. The authors have examined the deep water wave resistance characteristics of a series of transom stern. 307–324 Copyright C 2007 Taylor & Francis 307 . One of the design challenges faced by naval architects is accurate prediction of the hydrodynamic characteristics of such vessels primarily in the areas of resistance.edu. it is only in the recent past that such hull forms have seen unprecedented growth in the high-speed ferry industry. Marcos Salas2 and Adam Schwetz3 1 2 Australian Maritime College.1080/17445300701594237 Prasanta K Sahoo1. which are of utmost importance to the high-speed ferry industry.au Area at aft perpendicular Area at forward perpendicular Area at maximum transverse section Breadth at maximum transverse section Demi-hull beam at the waterline Beam-Draught ratio incremental resistance coefﬁcient ∇ Block Coefﬁcient C B = L. Key words: Catamaran.

Schwetz Downloaded By: [Australia Maritime College] At: 04:52 2 April 2008 g iE L or LW L L/B or LW L / B XD H L/∇ 1/3 LCF LCB PE RW RF RT s s/L S T AP TX 1+k 1+γ k βM δW εR φ σ γ ρ τ υ Acceleration due to gravity 9. This paper concenSAOS 2007 Vol. Two interference effects contributing to the total resistance effect were established. The particulars of the models used in the investigation are presented in Table 2.45 Slenderness ratio Longitudinal center of ﬂotation from transom Longitudinal center of buoyancy from transom Effective power Wave resistance Frictional resistance Total resistance Separation (measured between demi-hull centre planes) Separation ratio (between demi-hulls) Wetted surface area Draft at aft perpendicular Draft at maximum transverse section Form factor Viscous form factor for catamarans Deadrise angle at amidships Transom wedge angle Residual drag–weight ratio Factor for pressure ﬁeld change Velocity augmentation factor Viscous interference factor Fluid density Wave resistance interference factor Kinematic viscosity of ﬂuid Displacement Volumetric displacement Table 2 Model particulars (Insel and Molland.4 and 0.2. there is hope for further development.6% trates on both single hard chine as well as round bilge hull forms with transom stern. Body plans of these models are shown in Figure 1. namely.3 7. The particulars of models tested by Insel and Molland (1992) are presented in Table 1. SHIPFLOW. Prediction of total resistance—background The background of the work has been based on some of the important modern methods in application so far.33 to 0. The model parameters have been based on data of modern catamarans found from the literature survey. Salas and A.6% 43. 4 The paper by Insel and Molland (1992) summarizes a calm water resistance investigation into high-speed semidisplacement catamarans.397 0. 1992) Geometric parameters Range of Application L/∇ 1/3 6 to 9 L/B 6 to 12 B/T 1 to 3 CB 0. C4 and C5 were of round bilge hull form derived from the NPL series and model C2 was of the parabolic Wigley hull form.44 0. The authors proposed that the total resistance of a catamaran should be expressed by Equation (1): C TCAT = (1 + φk)σ C F + τ C w (1) Here φ has been introduced to take account of pressure ﬁeld change around the demi-hull and σ takes account of 308 Copyright C 2007 Taylor & Francis .4 8. caused by asymmetric ﬂow around the demi-hulls. 0. Slender hull forms and higher speed capabilities provoked the need of technological evolution in predicting their preliminary characteristics of resistance. and wave interference.1 6.6% 43.0 in the demi-hull conﬁguration and catamaran conﬁguration with separation ratios (s/L) of 0.5.397 LCB/Lfrom transom 50% 43. running trim. K. It is understood that the solutions cannot be generalized by one simple formula but varied in accordance with speciﬁc conﬁgurations of catamarans.1 to 1. The former has been acceptably determined from ITTC-1957 line using a form factor component whilst the latter still remains a stimulating question to researchers.3. M. With the advent of computational ﬂuid dynamics (CFD). 2 No. sinkage and wave pattern analysis experiments were carried out. due to the interaction of the wave systems produced by each demi-hull. with symmetrical hull forms based on experimental work carried out at the University of Southampton. 1992) Models C2 C3 C4 C5 L/∇ 1/3 7.397 0. and the regression equations were developed based on the data.81 m/s2 Half waterline entry angle Waterline length Length–beam ratio (demi-hull) Table 1 Catamaran geometric parameters (Insel and Molland. Insel and Molland’s method (1992) INTRODUCTION Catamarans account for 43% of the ﬂeet by vessel numbers as given by the report of Drewry Shipping Consultants (1997). Calm water resistance of catamarans is in general attributed to two major components.6 2 2 2 CB 0. Calm water resistance.P.5 L/B 10 7 9 11 B/T 1. being viscous interference. These methods have been brieﬂy explained below. is used to generate data of wave making resistance of catamaran hull forms. 0. In this paper a computational package. credibility of these equations has been compared with several other theoretical and experimental methods presently available. which effects the boundary layer formation. In the end. Sahoo. Models C3. All models were tested over a range of Froude numbers of 0. viscous resistance and calm water wave resistance.

The authors further conclude that: Millward’s method (1992) In his investigation. The derived form factors for the mono-hull conﬁguration are shown in Table 3. and that the trim angle is reduced with increasing hull separation ratios. r r r r and are higher for catamarans where viscous interference takes place between the hulls. C4 and C5 (Insel and Molland. Viscous resistance interference was found to be independent of speed and hull separation. 1992) Mono-hull (1 + k) C2 1. 1992). thus. 1992). A ship to model correlation exercise is required for the extrapolation techniques presented to be validated. Additional models are tested with the particulars listed in Tables 4. method (1994) 8 π RW 2 T2 ρg B L and RW is the wave resistance. τ= [C T − (1 + γ k)C F ]CAT C WCAT = C WDEMI [C T − (1 + k)C F ]DEMI (3) The authors conclude that the form factor.17 the velocity augmentation between the hulls and would be calculated from an integration of local frictional resistance over the wetted surface and (1 + k) is the form factor for the demi-hull in isolation. with beneﬁcial wave interference between Froude numbers of 0. Generally higher hull separation ratios result in smaller wave interference. C3. whence: C TCAT = (1 + γ k)C F + τ C W (2) Figure 2 Effect of hull separation on catamaran resistance (Millward. Noting that for demi-hull in isolation. From this. Figure 2 reproduced from the article demonstrates the effect of separation ratio on resistance. Catamarans display higher trim angles than mono-hulls.35 and 0. be kept constant over the speed range.Practical evaluation of resistance of high-speed catamaran hull forms—Part I Downloaded By: [Australia Maritime College] At: 04:52 2 April 2008 Figure 1 Body plan of models C2. the total resistance (RT ) of catamaran can be found by: RT = 2[(1 + k)RF + RW ] Molland et al.30 C5 1. C R∗ Fn 2 (4) in which R∗ = The frictional resistance is calculated using ITTC 1957 line. Millward (1992) in fact intended to adhere to the common parameter range as suggested by Insel and Molland (1992). He introduced a new wave resistance coefﬁcient. Millward (1992) has reported his test results on a series of catamarans characterised by hull length–beam ratio (L/B) of 10 and a beam–draft ratio (B/T) of 2. 309 SAOS 2007 Vol. φ and σ can be combined into a viscous interference factor γ where (1 + φk)σ = (1 + γ k). 2 No. The authors also conclude that the viscous interference factor γ is effectively independent of speed and should be kept constant across the speed range and it depends primarily on L/B ratio. For practical purposes. 4 Copyright 2007 Taylor & Francis . ∗ CW = r The vessels tested have an appreciable viscous form effect. (5) The paper by Molland et al. Table 3 Derived form factors (Insel and Molland. and for a catamaran. for practical purposes. This was a good practical solution to a complex engineering problem at that point in time. γ = 1 and τ = 1. is independent of speed and should. and rather is dependent on demi-hull length to beam ratio. (1994) is an extension of the work conducted by Insel and Molland (1992). τ can be caculated from Equation (3).42.10 C3 1.45 C4 1.

6 6c 11. 2 No.43 1.48 1.45 1.41 1.0 1.44 1.50 2.7 2.5 1.0 7.397 43. Salas and A. Armstrong proposals on form factors are presented in Appendix I.6 1.41 1.6 5a 12.5 8.30 1.5 Average 9.1 9.6 5c 9.58 2.41 1.0 2.42 1.636 1.P.6 1.5 — 1.43 1.42 1. Molland et al.417 −43.51 1.44 1.45 1.397 43.39 2.173 −1.18 1.6 4b 9.5 1.0 7. (1994) are shown in Table 5.44 1.22 1.0 0. 1994) L/∇ 1/3 8.6 2.5 2.43 1.44 1.41 1.47 1.30 1.44 1.5 8.432 100.82 1.397 43.47 1.6 4a 10.40 γ 1.5 1.26 1.5 0.397 43.47 1.9 Table 6 Viscous form factors of catamarans (Molland et al.4 7. (1 + γ k) = f B L s .355 −2.’ (Armstrong.43 1.43 1.5 1.397 43.49 2. M.2 s/L = 0.44 1.2 2.2 1. (1 + k) T ∇ 1/3 L (6) (1 + γ k) = a 0 + a 1 B L s + a 2 1/3 + a 3 + a 4 (1 + k) T ∇ L B L Bs L s + a5 + a6 + a 7 1/3 1/3 T∇ TL ∇ L B L + a 8 (1 + k) + a 9 1/3 (1 + k) T ∇ Copyright C 310 2007 Taylor & Francis .44 1.46 1.47 1.44 1.397 43.43 1.50 2.26 1.4 1+γ k 1.41 2.74 s /L = 0.8 8. (1994) gives the experimental data of a systematic series of high-speed displacement catamaran forms in which the viscous form factors are shown as in Table 6.74 s /L = 0.48 1.43 1.549 a8 a9 a 10 a 11 a 12 a 13 a 14 −2.33 1.2 1+γ k 1.397 43.0 6.54 1.58 1.40 1.57 1.44 1.4 1.5 2.5 0.44 1.48 1.5 0.6 4c 8. .40 γ 1.505 −150.23 s/L= 0. 4 As there is no equation giving the form factor of catamarans (1+γ k) directly from their particulars. K.6 6b 13.65 2. 1994) Model Form factors 3b 4a 4b 4c 5a 5b 5c 6a 6b 6c Mono-hull (1+k) 1.4 s/L = 0.3 1.628 6.397 43.0 2.48 1..6 5b 11.37 1.791 4.50 1.43 1. readers are referred to the above report.5 0.3 1+γ k 1.0 s /L = 0. Current literature suggests large form factors are needed for correlation between model scale and full scale which.506 −2. examines the current methods for predicting the resistance of recently designed highspeed catamarans. 2000).44 1.27 2.40 1.7 1.42 1. a regression model had to be developed from the data of Table 6. .1 2.44 1.50 2.5 0.5 1.45 1.5 1+γ k 1.65 1.40 1.397 43.3 s/L = 0. For further details on the resistance data.0 0.38 1.73 1. 1994) Downloaded By: [Australia Maritime College] At: 04:52 2 April 2008 Models L/B L/∇ 1/3 B/T CB LCB/L (%) 3b 7.1 9. Using multiple regression analysis.42 1.42 1..5 1.39 1.43 1.42 1.7 9.41 1.6 Table 5 Model form factors (Molland et al. Armstrong’s thesis entitled ‘A Thesis on the Viscous Resistance and Form Factor of High-speed Catamaran Ferry Hull Forms.45 1.46 1.60 1.40 1.43 1.57 1. In addition to this.9 8.55 1.28 1.0 8.47 1.446 68.40 1. Form factors as per Molland et al.43 1. Sahoo.30 1.5 — 1.397 43.45 1.43 1. SAOS 2007 Vol.46 1.5 9.44 1.5 Average s/L= 0.46 1.44 1.5 0. a general equation has been found as shown in Equation (7) and regression coefﬁcients shown in Table 7. (1994). Schwetz Table 4 Particulars of models (Molland et al .0 0.0 1.48 1.5 B/T 1+γ k 1+γ k 1+γ k 1+γ k 1.38 1.45 Table 7 Regression coefficients for Equation 7 a0 a1 a2 a3 a4 a5 a6 a7 0 0.45 γ 1.40 1.44 1. For each Froude number studied.258 2. Armstrong (2000) claims.1 2.44 γ 1.41 1.30 1. the catamaran form factor (1+γ k) will be a function of the different geometrical particulars and the mono-hull form factor (1+k) of the hull forms.45 1.91 1.6 6a 15.22 1.927 Form factor of catamarans The research and results are also detailed in Molland et al.5 9. contradicts the expectation that long slender hull forms would have low values.6 1.932 −1.0 0.5 1.

Based on this hull form. It may be noted that the regression coefﬁcients correction to CW have not been published.0. Further research is. This series is valid for the ranges shown in Table 9. The required correction to wave resistance coefﬁcient was given by: CW = C R − CW The required regression equations for reproduced in Appendix III. The total resistance is given by: RT = [RF + (ε R × ∇ × ρ × g )] (9) The details of this methodology are illustrated in Appendix II. The recorded data were then statistically analysed to determine an accurate regression equation. 2 No. therefore.6 and B/T = 2. C WCAT = b 0 + b 1 + + + + + L B L + a 2 + b 3 1/3 B T ∇ s LB L L b4 + b5 + b6 L BT B ∇ 1/3 B L Ls Bs + b9 b7 + b8 1/3 T∇ BL TL L s LB L b 10 1/3 + b 11 ∇ L B T ∇ 1/3 LBs L L s b 12 + b 13 BTL B ∇ 1/3 L B L s LB L s + b 15 b 14 1/3 L T∇ B T ∇ 1/3 L which was used to calculate the required correction to wave resistance coefﬁcient CW as predicted by Michell’s integral. A signiﬁcant aspect of this method is that it can be applied to both monoand multihull vessels in the preliminary design stage. and their wave resistance in calm water was determined using SHIPFLOW. a total of 18 models were developed (total including the parent hull form). several of which were catamarans and trimarans. The regression method is based on the resistance predicted by Michell’s integral and model experiments carried out on 30 different hull shapes.075−2)2 10 Rn CA = 0 (10) The experimental residual coefﬁcient was given by: C R = CT − C F (11) Basing on the NPL data of the studies conducted by Molland et al. Veriﬁcation of the equation with experimental database is also lacking. Kantimahanthi and Sahoo (2001) (8) VWS Hard Chine Series ‘89 regression methodology (1995) This method was proposed by Zips (1995) using multiple regression analysis of test data intended to predict the resistance of hard chine catamarans with hull parameters in the scope of the VWS Hard Chine Catamaran Hull Series ‘89.55. The achieved regression equation has been compared with three empirical methods that have commonly been used so far. Only the demihulls were considered during hull form generation. Rintala and Karppinen method (1995) The authors have proposed a prediction method of estimating the resistance of high-speed mono. However.Practical evaluation of resistance of high-speed catamaran hull forms—Part I Downloaded By: [Australia Maritime College] At: 04:52 2 April 2008 s B L s (1 + k) + a 11 L T ∇ 1/3 L B L Bs (1 + k) (1 + k) + a 13 + a 12 T ∇ 1/3 TL L s (7) + a 14 1/3 (1 + k) ∇ L + a 10 Regression following Molland et al. (12) CW have been Single chine series regression of Pham. The C WC AT is given by Equation (8) and regression coefﬁcients as shown in Table 8. (1994). the results obtained have shown considerable promise. 4 2007 Taylor & Francis . L/B = 15. and the target vessel type being catamaran where s/L could be a signiﬁcant parameter. needed to reﬁne the accuracy as well as to complete the selection of crucial parameters employed.and multihull vessels based on Michell’s integral along with a regression correction. with demi-hulls being symmetrical with respect to each other and with respect to their individual centre-line planes. With C WC AT as the dependent variable. The details of the models are shown in the table above. Hanhirova. a speed independent regression equation has been developed so as to obtain the wave resistance coefﬁcient directly from the particulars of the catamaran. The result of the literature survey on 50 contemporary catamaran conﬁgurations when integrated with the results shown by Doctor et al. Regression correction was Copyright C A systematic series of 18 hard-chine demi-hulls were generated. The accuracy of the established regression equation has been seen to deviate appreciably by various sources of uncertainties. and a regression equation for predicting wave resistance of single chine catamaran hull forms in calm water can be seen as achievable. (1994) method carried out as follows: CT = C F + C R + C A C F = (Log0. (1994) have led to the parameters shown in Table 10 (Appendix IV). the equation shown has been assumed for wave resistance coefﬁcient along with the regression 311 SAOS 2007 Vol. which were later extended to twin hulls. A parent hull form was developed with C B = 0.

376 0.540 13.962 0.90 0.503 −9.324 −2.782 0.106 4.216 44.259 −1.448 2.979 10.063 0.437 2.585 53.637 3.25 0.439 0.789 9.320 6.015 −38.758 0.378 −24.270 −5.150 −0.280 −0.30 0.402 −0.120 0.448 3.586 2.85 0.004 0.233 −26.041 4.514 65.226 1.173 1.150 32.136 −2.169 0.021 2.095 2.100 −11.986 −5. Sahoo.80 0.189 −2.781 7.70 0. 2 No.242 0.901 −0.619 −0.398 0.35 0.092 0.60 0.971 38.480 −5.00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 −1.051 −3.949 −33.701 1. K.573 1.833 10.346 0.171 −2.749 33.265 13.046 5.168 −2.556 3.431 39.638 −0.051 −0.315 −4.230 −1.119 5.043 0.199 −2.437 8.326 0.876 0.797 −0.20 0.140 Copyright C 2007 Taylor & Francis .961 −0.75 0.614 107.970 −7.184 0.024 −5.082 −0.589 111. 4 Table 8 Regression coefficients for Equation 8 b3 b4 b5 b6 b7 b8 b9 b10 b11 b12 b13 b14 b15 Fn b0 b1 b2 312 0.035 −0.900 19. M.730 −0.010 −0.816 −0.082 0.730 0.430 −1.401 0.905 −0.572 4.139 −1.503 0.474 0.712 −1.675 0.067 13.388 1.816 0.134 10.184 0.628 0.351 −0.114 −2.022 2.584 1.610 −0.291 −0.494 −0.830 −0.833 −0.244 3.979 4.283 0.059 4.586 0.798 10.012 0.989 −13.235 12.315 −0.363 5.841 2.565 26.515 −0.476 0.130 −2.217 −0.705 −1.135 −1.214 −0.270 −0.406 9.936 0.623 −27.921 3.45 0.620 10.95 1. Schwetz SAOS 2007 Vol.882 −3.177 −2.500 0.618 0.075 −6.100 −2.923 14.080 −5.512 5.138 −1.655 −0.430 −11.325 −7.65 0.457 −0.144 −2.242 −6.149 1.120 1.349 0.591 3.409 −4.191 −3.326 −83.559 4.576 0.Downloaded By: [Australia Maritime College] At: 04:52 2 April 2008 P.826 29.338 −3.141 0.288 0.134 −0.702 −0.50 0.169 2.927 −3.278 −27.169 −2.117 −0.602 2.306 0.120 13.122 −2.854 −2.148 2.518 0.021 0.409 −5.934 −32.330 −2.254 9.472 31.941 9.504 5.067 0.976 −28.811 1.763 −0.434 −1.561 0.638 −77.171 0.137 −2. Salas and A.918 21.604 −12.487 3.253 −0.171 −8.754 0.845 −36.728 0.489 −0.153 0.104 −0.380 −0.599 5.822 −0.368 −0.519 −11.031 0.55 0.192 −0.017 −0.296 −2.074 0.327 −13.129 −0.972 0.183 −2.482 −1.928 −21.012 5.092 −2.096 17.614 3.413 2.356 −7.40 0.132 −0.079 −0.911 14.430 −1.151 −1.624 −0.620 0.497 −6.119 0.615 −0.242 0.870 −1.266 −6.106 2.053 1.237 −2.412 0.076 −6.230 −0.120 −0.218 59.444 −8.328 −1.

2 No.80 13. and are believed to closely represent the hull forms being used in industry at the moment.257688 −2.60 0.60 20.02 16.40 15. Browne and Salas (2004) coefﬁcients shown in Table 11.208420 1. Further details on regression equations have been provided in Appendix V.017454 −0.50 2. where the amount of semi-swathness increases from models M2 to M4.50 2.026670 −0.09 0.1 0. and do not form part of any published systematic series.002700 1.16 0.00 1.55 0. 4 C 2007 Taylor & Francis . Table 10 Model particulars (Pham.543052 0.966650 0.58 24.55 0.036256 1.231476 0.50 2.9 1. Hull model M7-CH contains a hard chine and hull model M6-CH was generated from model M7-CH by rounding or ﬁlleting the hard chine.55 0.11 0. 2001) Models M1 M2 M3 M4 M5 M6 M7 M8 M9 M 10 M 11 M 12 M 13 M 14 M 15 M 16 M17 M18 CB 0.2 and 0. The displacements and LCB vary for models M8-CH– M10-CH.442478 −2.391235 1.12 WSA/L2 0.21 24.55 0.80 24. The body plans of models 1 to 10 are presented in Figure 3 and a summary of the particulars are presented in Table 12.50 2.43 23.00 1.11 0.53 16.7 1.60 0. Model M5-CH was generated from model M1-RB by replacing the round bilge with a single chine. including the displacement and LCB.5 2. while maintaining the same displacement and LCB.58 7.046904 −0.07 0.5 −0.424720 −2.268397 β2 −1.4 while speed range (Fn) was constrained between SAOS 2007 Vol.00 1. Models M8-CH–M10-CH were included to examine the general effects of reducing L/∇ 1/3 . It may be noted that LCB and LCF locations are with respect to the transom.69 7.40 10.326583 β4 −0.232555 −0.60 15.93 9.119485 1.4 2.063404 1. r compare CW results of CFD with results from towing tank tests and develop regression model.129839 −0.06 0.20 B/T 1.861936 0.55 0.50 0.66 Table 11 Summary of regression coefficients for Equation 13 Fn α β1 −2.06 0.20 26.12 0.49 9.60 20.02 20.59 L/B 10. with a transom stern to accommodate two sets of Kamewa 71 series water jets on each demi-hull.80 20.30 7.60 0.14 0. Round bilge catamaran series of Sahoo. referenced from the transom.235047 −2.60 L/∇ 1/3 6.40 15. The systematic series of demihulls thus produced was conﬁned to s/L ratio between 0.242743 −2.50 1.2 0.0 0.50 2.979194 0.50 0.279982 −2.50 2.317368 −1.55 16◦ –38o 0o –12o The research program was devised to: r examine variations in CW using CFD.931836 0. For each model. A parametric transformation procedure was used to produce the desired demi-hull series.004593 −0.028588 The series of symmetrical hull shapes used in this study were generated by the authors.80 26. The systematic series that was used for this analysis is based on typical hull forms used by the high-speed ferry industry in Australia.40 15. which has a round bilge hull form and was designed to have an overall length of 50 meters.96 23.448887 0.14 23.60 0.692935 β3 0. 1995) Round bilge catamaran series of Schwetz and Sahoo (2002) Downloaded By: [Australia Maritime College] At: 04:52 2 April 2008 Parameter Length (m) Displacement (tonnes) Fn Length-Beam ratio of demi-hull LW L /B XD H Deadrise amidships β M Transom Wedge δ W Range 20–80 25–1000 0.80 10.384697 −1.68 22. The models are not mathematical in nature. r examine variations in CW using CFD.80 20.08 0.00 2.10 βM 23. An amount of semi-swathness was added to model M1-RB to create models M2-SS–M4-SS. Kantimahanthi and Sahoo.50 1.155136 −1.53 24.4 0. C WCAT = α Copyright L B β1 B T β2 C β3 s L β4 (13) 313 The authors have expanded on the work carried out by Schwetz and Sahoo (2002) by conducting further work on a systematic series of round bilge catamaran hull forms and subjecting these to CFD analysis.60 0.24 9.489982 −1.60 15.995197 1.50 0.908676 −0.50 2.09 0.582805 −1.8–1. hull model M1-RB was created.00 1.55–13.60 0.050167 −0.13 0. containing parameters relevant to the regression analysis.004378 −0.86 1.911271 1.255878 −2.6 2.528469 −1.21 24.67 10.50 0.08 0.3 0.961013 0.240560 −1.33 6.4 7.308691 −2.50 0.278595 −0.08 11.031137 −0.62 12.442334 −1. hydrostatic information was extracted as presented in Table 14.60 20. The body plans of demihulls have been shown in Appendix IV as Figure A1.13 8.898569 0.819332 −1. Following a review of current vessel dimensions.004858 0.507751 0.15 23.027712 −0.40 10.55 0.60 10.921796 0.00 17.Practical evaluation of resistance of high-speed catamaran hull forms—Part I Table 9 Parameter range (Zips.8 1. while modifying basic hull parameters and maintaining the same displacement and LCB position.402987 −2.09 9.80 19.021166 1.98 11.50 2.233282 −2.25 23. while maintaining the same displacement and LCB as models M1–M5.17 0.50 2.08 0.162273 1.07 9.351095 −2. with the LCB located at around 42% and 44% of the waterline length. Table 13 illustrates the geometrical parameters of the demi-hull series developed.09 0. The displacement was 255 tonnes.965683 0.11 0.035505 0.86 8.47 8.60 15. while modifying basic hull parameters.35 10.

20 0.10 1.68 9.30 14.98 0.46 0.08 3 15 2.31 1.45 8.11 1.46 0.45 8.22 0.0. SAOS 2007 Vol.03 0. The regression equations to determine the CW are presented in Appendix VI.50 10.60 The authors have illustrated a procedure for rapid development of hull form and preliminary prediction of resistance of high-speed catamarans with slender demihulls. SHIPFLOW was developed as a pioneering effort to address the complication of ﬂuid ﬂow characteristics around moving objects both in fully submerged situation and in free surface situation.2 and 1.47 1.52 0.80 14.45 9.40 4 15 2. M.53 9.04 6 12.00 1.66 0.00 42. 4 The wave resistance coefﬁcients were calculated for each hull model using SHIPFLOW.56 9. which combined with the average form factor value of 1.61 0. The theoretical wave resistance coefﬁcient.00 15. CW . Sahoo.45 2 15 1.00 40.31 2.40 11.5 0.54 9.56 9.00 44.12 1.00 42.55 0.42 (1+γ k) for ITTC ‘57 friction line would provide the total resistance.60 38.00 45.99 1.55 9.20 15. Calculation of wave resistance coefﬁcient Table 13 Systematic series of catamarans [Sahoo.00 47. They have made use of Michell’s integral for slender vessels to estimate the wave resistance of demihulls.00 15.P.71 2. as described by Larsson (1993).00 42.65 0.40 9.5 0.66 3.92 Resistance estimation method of Subramanian and Joy (2004) Models M1-RB M2-SS M3-SS M4-SS M5-CH M6-CH M7-CH M8-CH M9-CH M10-CH L/B B/T C B L/∇ 1/3 15.5 1.18 9.00 15.15 2.00 1.16 16.5 0. 2 No. The body plans of models developed during this research study are illustrated in Figure 4.5 2.20 5 12. Figure 2 represents the zonal approach or 314 Copyright C 2007 Taylor & Francis .40 13. Even though SHIPFLOW is intended specially for marine applications.54 7 10 2.00 42.10 9.54 0.5 0.60 13.00 44.00 49.97 0.20 6.5 0.08 7.96 0.45 9.10 15. K.73 1.68 8. Table 12 Model parameters (demi-hull) LCB/L from transom LCB/ i E (deg) (% ) LCF 8.00 2.00 15. Schwetz Downloaded By: [Australia Maritime College] At: 04:52 2 April 2008 Figure 3 Body plans of models (Schwetz and Sahoo.79 1. Browne and Salas (2004)] Model L/B B/T CB L/∇ 1/3 1 15 1. it has also been extended to sufﬁciently solve closely related problems such as highly turbulent ﬂow around automobiles. The theoretical details of models generated and tested numerically are shown in Table 15.5 0. is calculated by splitting the ﬂow into three regions where an efﬁcient approximation of the ﬂow equations may be made and a complete ﬂow calculation may be accomplished in a few hours using the potential ﬂow. a CFD program developed by FLOWTECH International of Sweden.77 0.80 10.73 1. 2002).5 0. Salas and A.07 2.90 8.49 0.30 7.

93 170. The boundary layer theory is used to compute the ﬂuid characteristics in zone 2.29 22.10 147. 315 SAOS 2007 Vol.54 21.69 230.11 30.97 307.92 11. development of wave proﬁles and sequential matters consisting of trim and sinkage characteristics.45 0. iii. Zone 3: Navier–Stokes method.57 0.41 19.43 0.80 22.20 166.00 2. changes in velocities and pressure ﬁeld Copyright C The laminar ﬂow starts from the stagnation point.13 2.60 10.69 6.11 15.48 0.40 151. 2004).20 L/V1/3 6.91 113. According to Larsson (1993).22 2. around objects such as propellers.00 151.27 22.20 155.71 289.50 11.90 91. Zone 1: Potential ﬂow method.27 20.60 8.43 7. it breaks down and becomes turbulent. the development of SHIPFLOW is based on three major methods each applied in its most efﬁcient zone of ﬂuid condition: Figure 4 Body plans of model 1 to 7 (Sahoo. Some of these problems remain a challenge to researchers in order to produce more sophisticated CFD program to handle the complex phenomenon of ﬂuid and object interactions.37 30.50 B/T 1.67 0. and extending up to the aft end.00 8. Navier–Stokes theory is applied in this zone to calculate the energy and hence the corresponding resistance incurred.57 3.57 231.90 S(m2 ) 130.99 44. regions used by SHIPFLOW to maximize computational efﬁciency. 2 No.88 2.94 0.08 246.92 0. Major areas in which SHIPFLOW has been found to be highly applicable include calculation of ship hull resistance both viscous and wave-related.69 6. and when they reach the transition point where the viscous force is insufﬁciently strong to bond the streamlines.00 50.89 12.00 50.70 21. diverge gradually as it moves downstream.33 4.00 4.62 0.93 2.00 50.99 21. The remaining region of the free surface is fully turbulent and will have wakes.40 131.00 3.50 152.89 181.49 0.46 0.15 CB 0.33 1.71 42. Potential ﬂow method is used to analyse the ﬂuid-ﬂow in the outermost area of the free surface designated as Zone 1 in Figure 5. Table 15 Geometric parameters of vessels [Subramanian and Joy (2004)] Parameter Form 1 Form 2 Form 3 Form 4 Form 5 Form 6 Form 7 Form 8 Form 9 L/B 9.32 44.49 0.33 7. Browne and Salas.00 50.52 0.53 i.33 3.) i E (deg) β(deg) WSA (m2 ) LCB (m) LCF (m) 1 2 3 4 5 6 7 50.29 1. It is speciﬁed as zone 3 and extending far aft from the transition point.37 246.00 50. Zone 2: Boundary layer method.62 13.47 0.27 22.37 8.69 8.60 0.20 195.10 173.76 1.30 22.03 4.10 256.43 CP 0.33 3.15 11.54 21.06 2.77 5.22 1. In this zone the ﬂuid-ﬂow is treated as continuous streamlines starting from fore end of the ship.33 2.69 12.18 7. which is usually about amidships.54 21. The nature of ﬂuid ﬂow change as the ﬂuid moves along the hull in this region.54 Figure 5 Zonal distribution for ﬂuid ﬂow computation in SHIPFLOW.67 1.10 125.00 50.00 5. ii.00 169.92 8.27 22.33 3.92 8. 4 2007 Taylor & Francis .Practical evaluation of resistance of high-speed catamaran hull forms—Part I Table 14 Hydrostatics of systematic series [Sahoo.25 2.65 0.32 24.15 11. The region of free surface that describes the thin boundary layers along the ship hull is deﬁned as Zone 2.60 2.67 0.94 23.79 1. Browne and Salas (2004)] Downloaded By: [Australia Maritime College] At: 04:52 2 April 2008 Model Length (m) Beam (m) Draught (m) Displacement ( ) (Max.34 22.

The work by Armstrong (2000) is limited to the applicable range of low Froude numbers that 316 Copyright C 2007 Taylor & Francis . these results cannot be considered as accurate until they have been scaled according to some model testing or full-scale data. at lower speeds it is not necessary to extend the free and transom surfaces further aft of the body group. The change in grid density was applied to this analysis to account for changes in Froude numbers. for NPL models and regression analysis for comparison purposes. Conversely. However. The NPL series tested by Molland et al. Due to the ﬂexibility of SHIPFLOW to be applied to many different applications. As in the previous section. it is necessary to deﬁne the number of stations and points required to produce a grid. when the speed is increased SHIPFLOW becomes increasingly unstable in its ability to model spray and wave breaking phenomena. the number of stations aft of the body must be the same for the free surface as it is for the transom group so that the panels are aligned. therefore. Sahoo.5 the correlation of results is extremely good. This group is. respectively. Therefore. and therefore. however. Therefore. however. It is interesting to note the variables that have been included in the different equations for each Froude number for a catamaran and demi-hull. The form factor due to viscous resistance interference factor is another aspect of catamaran resistance that could be further analysed. Schwetz The free group represents the free surface. a closer correlation would hopefully exist when the input is more accurate. Therefore. The value especially for dead-rise amidships and half angle of entrance have been kept constant at 7◦ and 30◦ . all of the co-ordinates are non-dimensionalised by the length between perpendiculars L P P . the grid must be systematically altered as the speed is increased to take into account the larger wavelength. 2 No. It is interesting to note the variables that have been included in the different equations for each Froude number for a catamaran and demi-hull. It is important to note that SHIPFLOW undergoes its analysis by non-dimensionalising the vessel down to a model of unit length. Therefore. Therefore. necessary to input the maximum number of iterations that are to be used. When considering the validity of results there are two key aspects. therefore. After the limits of the free surface have been introduced.P. quite long and only as wide as the vessel. The main variable that becomes apparent as having the most inﬂuence on resistance is the slenderness ratio (L/∇ 1/3 ). At low Froude numbers the transom wave has a small wavelength and a large wave height. The module XPAN is the solver that iteratively converges on the value of coefﬁcient of wave resistance. If SHIPFLOW is used correctly very precise results may be obtained. then the linear solver is used as the default. Therefore. As mentioned XMESH module enables the user of SHIPFLOW to construct a grid of panels to illustrate the scenario to be tested. it can produce varying results. as previously mentioned. however. caution must be taken when comparing results at different speeds. In addition to this. On the other hand. if the grid is not well set-up the result does not have much validity. at high Froude numbers the transom wave has a large wavelength SAOS 2007 Vol. it is necessary to once again deﬁne the number of stations along the length of the free surface and the number of points across each station to create the grid. Salas and A. which is used to model the waves created. Half angle of entrance. This was achieved by increasing the free and transom surfaces further aft until two wavelengths are included as a guideline. The free surface can be considered as the water surrounding the body group. and a considerable distance abeam of the vessel. using this software is a balance of stability and precision and to produce valid results an extensive amount of time is required to analyse the different scenarios. The other problem with using this work is the insufﬁcient information available in their paper on the hydrostatics of each model. The main variable that becomes apparent as having the most inﬂuence on resistance is the wet volume slenderness ratio (L/∇ 1/3 ). but it will be necessary to include smaller panels in the grid to account for the signiﬁcant changes in wave height.0 considered as the upper limit. which will not match model testing or fullscale data. It is. it is necessary to create a free surface that extends forward of the bow. RESULTS Only three vessels from the series tested by Molland et al. which can then be altered for accuracy. the type of solver that will be used must be speciﬁed. To overcome this problem. The other important feature of XPAN is whether the model is enabled to freely sink and trim. M. when constructing the grid in SHIPFLOW the aim is to achieve precise results. If the grid is not altered it can be expected that as the Froude number is increased the results can be considered as becoming increasingly precise. The breadth to draft ratio becomes signiﬁcant only at the higher speeds. it is more unstable particularly at high speeds and the solution may not converge. the investigation has been restricted to this speed. If reference is not made to the type of solver. The transom group represents a part of the free surface. One of the major limitations of SHIPFLOW is its inability to model spray and wave-breaking phenomena at high speeds with a Froude number of 1. dead-rise and separation ratio are also signiﬁcant throughout the speed range. The non-linear solver will generally produce a more accurate result than the linear solver. 4 Downloaded By: [Australia Maritime College] At: 04:52 2 April 2008 with small wave amplitude. For consistency. which extends directly aft of the transom. It is observed that above a Froude number of 0. the precision and the accuracy. K. (1994) coincided with the series produced for this analysis. well aft of the model (approximately two wave lengths). Therefore. (1994) is based on a mono-hull series that has been put into a catamaran conﬁguration. if a constant grid is applied to all of the models at the full range of speeds the degree of precision varies. The program will produce an accurate result of coefﬁcient of wave resistance based on the grid supplied.

so as to differentiate the single hard chine hulls and the round bilge hulls. the following empirical formulae may be used: Wetted Surface Area: ∇ (14) S = 1.000 10.300 11.00 0. Rintala and Karppinen method.276 0. an equation for form factor of catamarans could be produced.145 0.397 9. and if they are not respected the results can lose their numerical reliability.844 × 10−3 6.678 × 10−3 1. (1994). g (m/s2 ) υ SW (m2 /s) ρ SW (t/m3 ) WSA demi (m2 ) WSA cat (m2 ) CA Transom wedge L/∇ 1/3 B/T LCB/WL LCB/LCF s /L L/B 1. 1992. if a similar analysis was undertaken with carefully monitored SHIPFLOW and model testing results.025 0. i E (◦ ) Dead rise angle β M (◦ ) LCB (m) LCF (m) CB Gravitational constant. 4 .92 0. 2004).97 0.95 0.0 ◦ are split into two. s (m) Half angle of entrance.65) T (15) Figure 6 Comparison of CW against C R values of Zips (1995).Practical evaluation of resistance of high-speed catamaran hull forms—Part I Table 16 Parameters of a chine hull catamaran model Downloaded By: [Australia Maritime College] At: 04:52 2 April 2008 vessel (tonne) demi (tonne) ∇demi (m3 ) L(m) T(m) B(m) Separation. Copyright C 2007 Taylor & Francis 317 SAOS 2007 Vol. Therefore. In order to calculate the total resistance. 2 No.7LT + m 2 as per Mumford T S= 1.553 4. The results obtained from the different methods are very close and reliable. For different values of the catamaran parameters. 1995 and Subramanian and Joy method. Results The variation of grid density must be very precise in order to obtain accurate results over the whole speed range. Hanhirova. except the ones for which data were not available (Millward’s method.600 0. time must be spent before analysis to set-up a grid for each Froude number so that the results are consistently precise. Final remarks can be used.7 B 2 ∇ + m B C B − 0. if actual data is unavailable.1881 × 10−6 1.497 2. Indeed they are different for each method. Therefore. Figure 7 represents a comparative analysis between the round bilge catamaran resistance undertaken by the authors and the NPL-hull form catamaron series investigated by Molland et al. All the methods explained here have been integrated into the program. The results for a model catamaran with chine hull (particulars shown in Table 16) have been shown in Figure 6.073 0.81 1.48 8 30 0.00 × 10−4 10 8.2(C B − 0. We have attempted to demonstrate here an example of the results obtained so as to summarise and compare the different methods. In order to perform suitable analysis on round bilge catamaran hull forms the constraints as shown in Table 10 should be strictly adhered to. the program allows us to plot the graphs of the predicted wave resistance coefﬁcients versus Froude number and of the power predicted versus velocity. So these methods can easily be used for a predimensioning.369 × 10−2 6. The only limit of these methods is the range of the catamaran parameters. This seems to be the least researched aspect of determining catamaran resistance.

Figure 8 C B against L/B as used by various authors. It can be seen from the ﬁgure that the range of applicability have been signiﬁcantly enhanced by inclusion of higher values of C B . with experimental results for the three NPL models whose geometrical parameters closely match the constraints of the regression analysis. 318 Copyright C In view of the analysis and validation process undertaken in this research work the following conclusions can be drawn: SAOS 2007 Vol. K. Schwetz Downloaded By: [Australia Maritime College] At: 04:52 2 April 2008 Figure 7 Comparison of round bilge catamaran hull forms against results of catamaran hull forms of Molland et al. r A regression analysis has been performed based on CFD results and has been compared with experimental results of NPL series hull forms as conducted by Molland et al. (1994) have a constant C B value of 0. Finally the Figure 8 has been illustrated to show the range of C B values against L/B ratios used by various authors to arrive at the regression equations. 2 No. (1994). r It appears that the regression equation is robust enough as it compares favourably. Salas and A.P. (1994).5 < Fn < 1. especially at 0. Sahoo. M. It may be noted that the catamaran hull forms investigated by Insel and Molland (1992) and that of Molland et al. CONCLUSIONS AND DISCUSSION r A systematic series of round bilge catamaran hull forms have been designed and wave resistance coefﬁcients for demi-hull as well as catamaran hull form conﬁgurations have been determined using CFD (SHIPFLOW) for a range of Froude numbers.397. 4 2007 Taylor & Francis .0.

1994. pp 355–368. Germany. of Marine Technology. Subramanian VA. France).Practical evaluation of resistance of high-speed catamaran hull forms—Part I r It is expected that in Part II. Drewry Shipping Consultants. Kantimahanthi K. Wellicome JF. Trans. of Royal Institute of Naval Architects. Couser PR. UK. pp. December 2003. In Proceedings of 2nd International Euro Conference on High Performance Marine Vehicles (HIPER’01). VWS Hard Chine. who have done some of the regressions (catamaran form factor and NPL series) and developed the spread sheets which calculate the wave resistance coefﬁcient and the power prediction for several methods (NPL Series. 134:1–20. 2002. Zips JM. Insel M. An investigation into resistance components of high-speed displacement catamarans. The University of New South Wales. Hamburg. Australia. Resistance experiments on a systematic series of high speed displacement catamaran forms: Variation of length-displacement ratio and breadth-draft ratio. They also want to extend sincere gratitude to Nick Browne of BMT Defence Services. Vol 1. Drewry Shipping Consultants. Australia and David Pontaut from the ENSIETA Engineering School (Brest. Single Chine. Chile for their support.and multihull vessels. Ship Science Report No. Luebeck. SHIPFLOW User’s Manual. Wave resistance of semi-displacement high speed catamarans through CFD and regression analysis. 1992. 2001. University of Southampton. Numerical resistance prediction based on results of the vws hard chine catamaran hull series ‘89. Preliminary resistance prediction method for fast mono. 1995. Norway. 2001 and Round Bilge. UK. In Proceeding of High-Speed Vessels for Transport and Defence. Sahoo PK. 38(1):5–11. ACKNOWLEDGMENTS The authors would like to express their sincere gratitude to the Australian Maritime College. Sahoo PK. which implies that the range could be further enlarged and more rigorous validation is required against experimental results. Australia and the University Austral. REFERENCES Armstrong NA. France. Nantes.Royal Society of Arts. 2000. Rintala S. Ph. Copyright C 2007 Taylor & Francis 319 SAOS 2007 Vol. Joy P. Wave resistance prediction of hard-chine catamarans through regression analysis. 1– 227. In Proceedings of 3rd International Euro Conference on High Performance Marine Vehicles (HIPER’02). 67–74. Pham XP. Sydney. FAST FERRIES: Shaping the Ferry Market for the 21st Century. Without their valuable contributions this paper would not have seen the light of the day. London. Trans. Molland AF. encouragement throughout the course of this research work. Bergen. 1995. It is imperative to note that a limited number of models have been created in this instance. A method for rapid hull form development and resistance estimation of catamarans. On the Viscous Resistance and Form Factor of High-Speed Catamaran-Ferry Hull Forms. Schwetz A.71. L. 4 . 1993. London. 1997. pp. Germany. 2002). Edition 1.D Thesis. Molland AF. Flowtech International. 2004. the paper will attempt Downloaded By: [Australia Maritime College] At: 04:52 2 April 2008 to validate the theoretical results against experimental results from a random selection of catamaran hull forms. pp. Resistance and ﬂow predictions using SHIPFLOW code. Hanhirova K. UK. In Proceedings of 4th International Conference on Fast Sea Transportation (FAST ‘95). 2 No. Spring. Larsson. Karppinen T. 19th WEGEMNT school. 382–394.

196250 0. 1995) Fn∇/2 1.817200 0.000000 0.000000 0.429128 0.866917 0.000000 0.000000 0.096328 0.0 Fn > 1.635988 −2.877430 0.000000 −0.371713 −0.000000 −0.706875 0.241426 0.000000 1.276875 0.000000 Copyright C 3.629531 −2.000000 1.5 0.148359 0.194250 0.000000 0.5 < Fn < 1.243674 −3.246017 0.000000 0.059200 0.15 (L/B)−1.076125 −0.67 (B/T)0.000000 0.103767 1 r Wetted surface area coefﬁcient is given by: SW(V=0) = C S ∗ ∇ 2/3 XS ∗ 10 Table 17 Matrix of regression coefficients C R (Zips.555179 −2.5 (L/∇ 1/3 )−1.656719 0.000000 0.554213 1.651877 + 1.007700 0.090188 −0.000000 0.190125 0.72 − f(L/∇ 1/3 )g (B/T)−0.571875 0.5 3 × 106 < Rn < 5×106 6.422339 −0.029385 0.450245 0.000000 0.391296 −0.209794 −2.211125 0.1 −2.25Fn2 + 4.00 5.266895 −0.000000 0.5 Validity 0.332042 1.P.416250 0.000000 0.832031 0.690865 −2. 4 320 2007 Taylor & Francis .000000 0.6 < Fn < 1.000000 0.000000 −0.000000 0.342187 0.151312 −0.000000 0.251026 0.000000 0.328047 0.00 1.000000 0.052496 ∗ X12 ∗ X2 (β M − 27◦ )/11 and X3 = δ W /12 X1 0.767250 0.225938 0.000000 −0.000000 0.165938 0.453125 0.000000 0.50 5.000000 −0.137585 2.000000 0.900769 −0.036289 −0.332250 −0.5 < L/∇ 1/3 < 9.000000 0.683505 −0.76 − 1.65 + 350(Fn )−0.409500 0.256967 0. 2 No.25 4.000000 0.000000 0.000000 0.189350 0.348312 −0.083789 SAOS 2007 Vol.000000 0.000000 C R = −0.000000 0.000000 0.5 1.148359 0.272668 0.000000 0.704463 0.000000 0.47Fn − 1.000000 0.000000 0.349687 0. K.152163 0.61 0.000000 1.000000 −0.694413 ∗ X1 (∇/2)1/3 +0.588758 0.55)/3.000000 0.884844 −0.000000 0.139 (L/∇ 1/3 )0.5 < B/T < 2.000000 0.00983 C S = −0.000000 0.75 5.447655 0.000000 −0.000000 0.447266 0.0 (1+k)model 1.000000 −1. Schwetz APPENDIX I Downloaded By: [Australia Maritime College] At: 04:52 2 April 2008 As per Armstrong (2000) the form factors are given by Parameter 1+k Equation 0.09f APPENDIX II Detailed regression analysis had been performed on this series (Zips 1995)) and in order to carry out resistance prediction some of the parameters have been reproduced below: r Variables deﬁned as X1 = (LW L /BXD H − 10.135563 0. Salas and A.000000 0.6 (B/T)−0.000000 0.000000 0.708625 −0.5 <B/T < 2.45 − 0.000000 0.000000 0. X2 = r Length–displacement ratio of demi-hull is given by: LW L = 7.000000 −1.257507 3 7.000000 0.50 6.000000 0.000000 0.496875 0.000000 −0.1 (1+k)ship f f g 1.151489 X22 0.647421 0.000000 2.000000 0.0 109 < Rn < 2×109 Fn < 1.701719 0.149062 −0.288437 0.586730 0.61 0.000000 0.008195 X1X22 −0.322734 −0.282139 ∗ X12 − 0. Sahoo.50 7. M.009085 XS = X12 2 X1 X2 0.000000 0.000000 0.000000 0.000000 0.041762 X13 X2 1.000000 0.120516 r Residual drag-weight ratio is given by: ε R (Fn∇/2 ) = r Vector of regression parameter XR is given by: (XRT ∗C R) 100 2.428999 0.000000 1.500808 −0.484800 0.000000 0.089625 −0.000000 −0.000000 0.627470 −2.000000 0.0 6 < L/∇ 1/3 < 8 12 < L/B < 15 1.090188 −0.

Practical evaluation of resistance of high-speed catamaran hull forms—Part I Downloaded By: [Australia Maritime College] At: 04:52 2 April 2008 X9 = X31/4 X0 = 1 X10 = X1 ∗ X2 X1 = f (LW L /BXD H ) X11 = X1 ∗ X6 X2 = f (β M ) X12 = X4 ∗ X6 X3 = f (δW ) X13 = X4 ∗ X2 X4 = X12 XR = XR = X14 = X1 ∗ X3 X5 = X13 X15 = X4 ∗ X7 X6 = X22 X16 = X1 ∗ X9 X7 = X31/2 X17 = x1 ∗ X73 1/3 X8 = X3 X18 = X2 ∗ X8 The wave resistance coefﬁcient correction formula for high L/B mono-hulls is: C W = C R − C W = C 0 + C 1 Fn sin λFne + C 2 cos λFne + C 4 sin λFne 2 A AP + C 3 sin λFne AX TAP + C 5 Fn3 TX C2 P Fn TAP TX 2 2 CP 2 r Procedure for calculation is as follows: + C 6 (C ∇ )2 cos λFne + C 7 + C8 A AP 2 AX • Given the following parameters LW L /B XD H transformed to X1. ∇. Copyright C w B 2 2007 Taylor & Francis SAOS 2007 Vol. W(V=0) and ε R (Fn∇/2 ) ∇ 2/3 • Rn = LW L (∇/2)1/3 Fn + C 9 sin λFne TAP TX 2 × Fn∇/2 × ( ∇ g )1/2 2 ν + C 10 cos λFne + C 11 TAP 1 TX Fn2 A AP AX + C 12 [log (90 − i E )]3 + C 13 cos λFne C ∇ • CF = 0. 2 No. C A. ρ and ν S LW • Determine (∇/2)L1/3 . 4 . β M transformed to X2 and δW transformed to X3.075 (log Rn − 2)2 SW(V=0) ∇ 2/3 The wave resistance coefﬁcient correction formula for catamaran is: × ∇ × g × 22/3 • • RF = 2 (C F + C A) ρ Fn∇/2 × 4 C W = C R − C W = C 0 + C 1 (C ∇ )2 1 Fn 2 RT = RF + (ε R × ∇ × ρ × g ) + C 2 C P [log (90 − i E )] + C 3 cos λFne + C 4 C 2 log (90 − i E ) + C 5 C 3 P P +C 6 TAP TX 2 BAP BX APPENDIX III The wave resistance coefﬁcient correction formula for low L/B mono-hulls is: C W = C R − C W = C 0 + C 1 cos λFne + C 2 Fne C ∇ exp + C4C∇ C P + a Fn2 BAP BX sin λFne + C 7 BAP BX 2 2 sin λFne cos λFne 2 + C 8 C 2 + C 9 C ∇ + C 10 sin λFne P 2 + C 3 C ∇ cos λFne + C 11 cos λFne sin λFne exp − + C 12 exp − w B 2 w B 2 C 5 Fne 2 cos λFne log (90 − i E ) a Fn2 C ∇ + C 13 cos λFne sin λFne 1 Fn w B TAP TX + C 6 cos λFne C ∇ + C 7 Fne C X exp + C 14 cos λFne 2 BX + C 8 (C ∇ )2 + C 9 [log (90 − i E )]3 TX + C 10 Fne cos λFne + C 11 Fne cos λFne C P + C 12 exp a Fn2 2 2 + C 15 C ∇ sin λFne + C 16 sin λFne 2 + C 17 C ∇ exp a Fn2 2 + C 18 C ∇ exp − TAP TAP + C 13 (C X)2 TX TX + C 19 cos λFne exp − + C 20 exp − 321 w TAP B TX where λ = a 1 C P + a 2 C ∇ .

2 No.P. Schwetz APPENDIX IV Downloaded By: [Australia Maritime College] At: 04:52 2 April 2008 Figure A1 Body plans of demi-hull models as per Pham. Kantimahanthi and Sahoo (2001). Salas and A. SAOS 2007 Vol. K. Sahoo. M. 4 322 Copyright C 2007 Taylor & Francis .

2509 0.87 0.1907 −0. C Wdemi = C 1 L/∇ 1/3 C2 C (LCB/LCF)C3 (B/T)C4 C B 5 (18) The generalized wave resistance coefﬁcient for a catamaran can be predicted from Equation (19).92 0.2 to 9.543 C4 0.042 0.98 0.0933 −0.1637 1.509 × 103 2.38 1.042 0.97 to 1.5821 −1.116 Table 21 Range of parameters for Equation 18 Geometric Parameters Range of Application Copyright C L/∇ 1/3 6.637 −7.2795 −1.012 0.4272 −2.0 1.938 −8. using the constants C1 to C4 from Table 19.649 C7 −3.1 1.000 −0.064 −8. C Wc at = C 1 L/∇ 1/3 C2 C (s /L)C3 (LC B/LC F)C4 i E 5 (16) The wave resistance coefﬁcient for a catamaran can be predicted from Equation (17).6142 −0.65 C2 −1.3948 −3.073 −0.7 0. using the constants C1 to C7 Table 20 Coefficients for Equation 17 Fn 0.47 to 2.9 1.261 × 106 1.235 −0.2 CB Range of 9.5526 0.8 0.8 0.5 0.6728 −3.020 0.4089 −1.302 3.440 × 1010 1.4 C1 2.215 −2.98 0.00 × 102 3.591 −0.5 0.46 to 0.95 0.2913 −0.5593 −1.113 −6.68 0.6 0.4 from Table 20.4 C1 1.9 1 1.6934 −1.92 to 1.023 0.392 −1.108 −0. 2002) are reproduced below.98 0.1339 −0.83 × 103 2.1 1.9787 C3 −2.16 1.000 0.285 −0.2 B/T 1.6 0.388 −9.99 0.995 −9.194 × 104 5.303 × 104 5.98 0.99 0.180 −0.590 −8.180 −8.99 0.9655 −1. C Wc at = C 1 L/∇ 1/3 C2 C (s /L)C3 (LC B/LC F)C4 i E 5 C (B/T)C6 C B 7 (17) APPENDIX V Some of the details of calculating the demi-hull and catamaran resistance (Schwetz and Sahoo.035 −2.191 −1.571 C3 −0.5754 1.8 0.6 0.5211 −1.466 −0.4 Table 19 Coefficients for Equation 16 Fn 0.98 0.207 −8.3 CB 0.4 to 1.98 0.0981 −0.96 0.97 0.99 0.35234 C5 1.99 0.99 × 104 1.99 0.97 0.982 −1.5075 −2.015 0.39 × 106 5.03 × 106 5.230 3.004 0. C Wdemi = C 1 L/∇ 1/3 C2 C (LC B/LC F)C3 C B 4 The generalized wave resistance coefﬁcient for a demihull can be predicted from Equation (18).41 0.5536 −1.3903 R2 0.97 Fn Downloaded By: [Australia Maritime College] At: 04:52 2 April 2008 Geometric Parameters L/∇ 1/3 LCB/LCF 0.5547 C4 −2.052 R2 0.102 0.3 1. 2 No.264 −2. using the constants C1 to C5 from Table 22.97 0.5148 −1.775 −0.003 C4 −0.9 1.6947 −2.87 × 102 1.3178 −0.002 −12.942 1.9213 −3.519 −5.0263 0.50 × 107 C2 −5.018 0.99 0.1 1.46 to 0.4 to 1.96 0.92 0.3 1.4110 0. whose validity range is shown in Table 21.473 −3.2168 −1.992 2.3 to 9.99 0.195 −0. whose validity range is shown in Table 18.715 −6.309 −1.81 × 103 4.4304 −1.029 1.54 × 102 3.422 −0.106 −0.00 1.5 0.1202 −0.026 −0.932 −10.551 −11. using the constants C1 to C7 from Table 23.072 0.182 −0.0 1.8731 0.583 1.69 C3 −0.4 SAOS 2007 Vol.027 0.151 × 107 8.6 LCB/LCF 0.000 0.78 0.Practical evaluation of resistance of high-speed catamaran hull forms—Part I Table 18 Range of parameters for Equation 16 Table 22 Coefficients for Equation 18 Fn C1 0.050 0.488 −0. The wave resistance coefﬁcient for a demi-hull can be predicted from Equation (16). 4 2007 Taylor & Francis 323 .99 0.7080 0.96 0.66 0.3228 0.2 1.61 × 102 6.016 0.508 × 104 1.046 −0.488 × 105 1.1409 −3.93 1.757 −1.98 0.30 0.4599 −2.6111 −1.4 0.5278 −1.155 −8.7 0.2 1.2 1.090 −18.000 −0.403 −0.96 0.648 −0.68 Fn 0.292 −0.6 Application 0.4337 1.7 0.438 × 103 8.348 C5 C ·C B 6 (B/T)C7 (19) C6 2.399 −1.1154 −0.305 −0.1421 −2.99 0.97 0.058 −5.303 2.690 −4.3 1.398 −2.327 −2.965 × 1012 C2 −12.99 0.164 0.012 2.840 −9.818 1.97 × 105 1.99 0.005 −16.000 R2 0.8650 −0.

999 0.216 −1.081 −0.055 0.138 −0.476 −3.999 0.554 −1.658 C5 −0.396 −0.300 C6 0.8 0.368 −0.317 −1.471 0.7 0.613 0.99 0.324 2.614 R2 0.1 1.9 1.873 −0.179 1.000 0.048 −0.025 0.129 −3.984 0.171 −0.600 0.259 −1.000 0.000 0.9 1.032 C4 −1.110 0.5 0. Schwetz Table 23 Coefficients for Equation 19 Downloaded By: [Australia Maritime College] At: 04:52 2 April 2008 Fn 0.365 −1.997 0.523 −4.995 0.379 −0.001 1.6 0.4 C1 1.356 0.221 3. Salas and A.339 −1.069 −0.702 0.5 0.051 1.422 −1.221 −3.000 0.262 −3.963 −0.615 −4.999 0.436 0.072 C6 1.128 0.092 −0.282 0.000 0.000 0.035 −0.087 −0.175 −3.99 0.455 1.006 0.000 0.337 −0.124 −3.000 0.266 0.000 −0.688 −0.000 −0.625 −3.P.632 −2.652 −2.585 3. Sahoo.000 −0.246 0.082 0.809 1.000 −0.756 −1.338 C7 −1.878 1.460 −1.290 −1.998 Table 25 Regression coefficients and R2 for catamaran configuration for Equation 21 Fn 0.3 0.199 0.000 −0.049 0.239 −0.091 −0. 4 324 Copyright C 2007 Taylor & Francis .989 0.8 0.162 0.000 −0.775 0.962 −0.617 −3.000 0. K.000 C8 0.974 1.965 C5 −0.600 −0.272 0.173 0.000 0.180 2.0 C1 2.322 −3.815 0.999 0.046 C6 −0.640 −1.2 0.3 1.031 1.645 −1.594 2.519 2.988 0.117 C4 0.515 0.2 0.8 0.7 0.278 −0.130 0.000 0.817 −2.99 0.086 C2 −0.264 C7 −0.314 −0.282 −3.122 0.178 −0.114 −4.668 −2.063 −0.000 0.194 0.265 −1.997 0.203 R2 0.052 −0.410 −0.079 −3.439 1.075 R2 0.002 0.995 SAOS 2007 Vol.037 −0.603 −0.070 0.7 0.181 −0.201 −0.99 0. M.337 C2 −2.3 0.158 −2.000 C3 0.4 0.6 0.068 −0.220 −0.9 1.6 0.064 0.209 0.466 −1.99 0.5 0.043 0.105 −0.000 0.666 −3.089 −0.390 0. 2 No.111 −0.998 0.99 0.472 −1.000 C5 −4.869 −3.000 −0.999 0.228 −0.377 −0.000 0.996 1.0 C1 3.195 −0.395 −1.132 −0.99 0.99 APPENDIX VI C Wd emi = e C1 L B C2 (C B )C3 L ∇ 1/3 C4 (i E )C5 (β)C6 (20) C Wc at = e C1 (β)C7 L B s L C2 B T C3 (C B )C4 L ∇ 1/3 C5 (i E )C6 (21) C8 Table 24 Regression coefficients and R2 for demi-hull configuration for Equation 20 Fn 0.069 −0.0 1.99 0.056 0.000 0.999 0.000 0.445 −3.985 1.000 0.820 −3.964 0.999 0.402 −3.000 0.061 0.218 −0.000 0.971 0.501 1.967 0.4 0.221 C2 0.000 −0.000 0.99 0.000 0.571 0.056 C3 −0.734 −2.000 C3 0.090 −0.000 0.039 0.2 1.305 −0.222 0.159 0.052 −0.095 −0.000 0.945 −0.396 C4 −3.064 −0.

Are you sure?

This action might not be possible to undo. Are you sure you want to continue?

We've moved you to where you read on your other device.

Get the full title to continue

Get the full title to continue reading from where you left off, or restart the preview.

scribd