|
Items in cart [0] |
Questions? Call 888-624-8373 |
| Copyright © 2009. National Academy of Sciences. All rights reserved. Terms of Use and Privacy Statement |
Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 957
A Design Tool for High Speed Ferries Washes
D. Aelbrecht ~ J.-C. Dern 2~3 y. Doutreleau4
('ELF - Laboratoire National d'Hydraulique et Environnement, Consultant,
Ocda3 ide BOO FIRST, 4DGA-Bassin d'Essais des Carenes, France)
ABSTRACT
The high waves generated by fast ferries may have
den imentcl effects v ham cpproahing She co t
The aim of this st dy is to detemmme She
characteristics of ship wash or g oups of high- peed
ship waves m coastal Ed shallow water regions The
pmpose of She medhod developed herem is to provide
c tool for authorities as regards speed limits Ed
routes for fast ferries cpprocchmg the coasts
The waves generated by high-sped ships are
represented by th ir free wave spectrum This
sp ctrum is determined digitally using c wave
resi tance prod cm POTFLO) based on the Ne marm
Kelvin model
It is ass med that She he waves propagate m She
same maimer es c short Rested see, defmed by c
di ecticrurl energy spectrum This energy pectrum is
expressed explicitly in terms of She he wave
sp ctrum
Spatial wave propagation is then modeled over
time using c thi d-generation spechal wave model
named TOMAWAC, based on c Quite element
tech iq~x Evolution over time is depicted et certain
critical positions along the shore Results me given in
terms of sig fficmt wave heights Ed me m wave
periods for deferent ship ro tes Ed peeds
INTRODUCTION
In recent years, m merous papers have been
published on fat ship wash finish MA 1997,
Hem k Kofoed et al 1996, Kirkegsard et al 1998,
St mbo et al 1999) The temm "ship wash" (or "wake
wash") refers to She arrival on the coast of waves
mectedbyfc t ships havellmg otlihore
If he ship is tnavelling et high speed, he
cmplit de of th waves arrivmg on the coast may be
high
These high waves me potentially d mgerous in that
Hey provoke unacceptable motions Ed reduced
stability for smell vessels Ed ships m coastal waters,
phs unacceptable wave agitation Ed risks for
swimmers Ed other users along the con t Ed on
beaches
This paper desk es c simulation medhod
developed es c tool for maritime safety authorities Ed
shipyards involved in She desig Ed building of high-
peed ships
Recent simulation results me peeve ted for She
case of c standard high-speed fer y refened to es
'N t" irLt~i e rapide type), cpprocchmg She Po t of
Nice from Corsica
The characteri tics of She N t me es follows:
- lend h et waterline l pp = 87 metner
displaced vol me V = 1200 ma
- lend h / widdh ratio equal to 5 7
- lend h / d aft ratio equal to 36
- max peed up = 37 nomads i e c Froude
n mber F = uO /^ = 0 65 corre pending
to t had odynamiccllyfastship
- the Froude m mber based on She di placed
vol me FV = uO /~ is equal to 1 865,
hence categori ing 6 is ship as c fast ship
according to IMO regmhtions (safety code
chapter 10 of SCOT AS) According to His rule,
adopted by Bme m Veritas in She framework of its
regmhtions concerning high- peed ships, c ship is
consideredtobefastffFv 21181
SHIP WAVE FIELD AND AMPLITUDE
SPECTRUM
Ships havellmg et c constmt speed in calm waters
generate c complex wave ystem ('efened to es "ship
wave field"l mo ing et She same speed as the ship
itself Ed therefore appearing es immobile to
observers on board She ship
Using c fixed pomt, the wash m be modeled in
She form of superimposed plane waves N wmm
1977):
+~ f[~"YcosO+kYsinOO to]
(.')= I l ~ (k,O)e kGkdO (1)
withy c Cartesim coordinate >! rem QXYZ whereby:
- Z = 0 signifies mdistmbed water,
- Z is di ected upwards,
OCR for page 958
- th point Q is fi cd wi6h ~egards to th see
bottom,
- QX is di~ection tdo Of the ship's sped md
th refore situAed on 6he umdi tu bed water,
- mgle O is the mgle of the di ection of
propsgation of th phne wave wi6h ~espect
to QX,
- QY is 6he diection of 6he umdistubed
water perpff~dicohr to QX, so that the co-
ordincte system is di ect
In formoh (1), the followmg rektionships cppe
between pulsation m, wave number k, speed c, water
dep6h h md ship speed uo:
t=~=
c = uO cos O
g thkh
kuo cos0 = 2)
uO cosO
(3)
+
/3 +I d (t s) ~ it~ 0063 + tY 6t 3) tdtdS (9)
7 o
with:
A(k,O)= H(k,O)K (k,O)secO (10)
27 uO
K(k,O)= i757 +Pf
k ko sec2O
(11)
whereby Pf g k, O) is 6he p s udo -funct ion cs soc icte d
with 6he n m-mteg ctle fu tion g k,O) md 5~ is 6he
Di~ac mecsure di tobuted owx cu ve ( 7 ) havmg 6he
followmg polar equstion:
k = K(0) = k o sec2O
k = K(0) thkh (4) m the pkme ~, ky)
wi6h:
K(0) = ko sec2O
ko = U2
One fcrmuac (2) gives 6he wave frequ ncies
mecsu cd by m immobile observer watching th ship
go by
Using c point rekted to 6he ship md pmallel to Q
XYZ, 6he wave feld is no longer time-dependent md
formoh (1) becomes:
5(7)= | 7 A(k,O)s [k cosO+k ysino] kd dO C)
(6)
Free wave amplitude spectrum
7he ship wave field usmg c point ~ehted to 6he
ship is determined by formok (7) whereby A(k,O)
is cclcokted using 6he Kochin fu tion, ~esultmg in:
H(k o)=4xl ~k +zk CO50+zkysirO ~(~ y z ) dS
for c smgle density hyer a(x',y',z')
7he impotance of 6he Kochm fu tion remits
fi om 6he followmg formok which em~bles cclcoktion
of the ship wave feld Eggers et o/ 1967):
i (8)
(12)
Formohs (9), (10) md (11) cm be simplffied by
noti g that if
K(k,O) = 2i1~5K. + Ke (k,O)
(I 3)
the conh ibution of 6he second term conesponds to
loccl waves which me Imked to the ship, i e which do
not radiste energy outwards fr m the ship md in
particular towards the coc t (see Eggers et o/ 1967 for
c demonshation usmg mteg ction m 6he complex
pkme; c second demon trati m is based on
Sokhotsky's formoke using m qu mt m physics)
7he fi st term only c m th refore be rebined for
calcokting ship wash; formoh (9) 6he~efme becomes:
5(XY)= I A
with:
A(7) ~ z[~(O)xoosO+~(O)yst °] dO (13)
A(0)= 0 H(K(O),O)sec30= g3 (3) (14)
gi o gi o Cos O
H(0) = H K(0), O)
(I 5)
Prheipal of rdeudadon of the free wave amplitude
spectrum
- 7he dish~bution of singukrities [eg a(x,y,z)]
equivalent to 6he hull is determined initislly
OCR for page 959
N merically, with 6he zid of z calcuhtion code
based on the Ne mar~Kelvm model: dem ity a
(x,y,z) results di~ectly from 6his calcuhtion
Exp rimentally, by meas xing the ship wave
fleld usmg z towmg tmk (longit dinal cuts) md
pinpomtmg 6he sing krity di tobution which is
most rep~esenbtive of the measmed field
7he Kochm f mction H k,0) is thff~ calcohted
using formoh (8)
7he flee wave zmplit de spectrum is 6hen
calcuhtedusmgformoh(l4)
7he ship wave fleld is 6hen deduced 5 (x,y) usmg
formoh (I 3)
7he ship waves are sig iflcmt m 6he followmg
sector
x<0 et Y <2 3/2 =tg(19°28') (16)
19°28'
the Kelvin mgle
(I 7)
- formoh (13) is rektive to m O y point ~ehted
to the ship; O bemg the bow, O bei g m 6he
mmetrical plane md di~ected to the fme md
Oy being in the t ms xse phne di ected to
po tside
- 6he scope of 6he wave field is x = 0 to x = -oo
Par'de~dar ease of ~dtra tdgh-speed s tdp s
Ult~a high-speed ships sail zt higher speeds 6 m
6he st mdard high-speed ships xrently in service
Ult~a high- peed ships c m briefly be descobed zs
ships havellmg zt z higher speed 6 m thA
cone ponding to the kst b mp on the wave ~esistmce
curve Associated wi6h these high speds, the ship
wave far-field is composed mainly of diverge t
waves, pro idmg 6he hull is st~eamlmed (Ogilvie
1 977)
As regards wave resistmce, hansverse waves
~emain importmt zs d mon tr ted by 6he followmg
formoh valid to i dinite depths Newm m 1977):
+
t:
R W= P u | A(0) scos 0 dO
(I 8)
whereby the factor cos3 0 implies that the portion
of wave ~esistance is more shongly-zssociated with
h msverse waves (0 = 0) th m with divergent waves (0
x )
7he zmplitude pech m of ulhz high- peed ships
is given for 6he flrst zpproximation order by z very
simple formoh (demonstmted in Kost kov 1968):
A(0) = [V.K (0) ezcK(O)
whereby zc <0 is 6he hull centx immersion, W is
6he hull vol me md K(0)= kO sec20
If the formok is written zs follows:
A ( 7 ) = e F ~ 7 2 0 )
6he zmplit de spech m tends t wards :mro when 6he
Froude n mber based on hull center immersion
F = u0 /~ rises indefmitely ~ 6his limit case,
6he wave sy tem is reduced to z smgle wave
enveloping 6he ship, whose amplit de, which is
mdependent of speed, d ops ve y mpid y zs w head
zwayfr m 6he ship:
~ (~, y) = c3 when R = ,1~ ~ + 21 )
7his wave is z 1ocal dist xbance; zll flee waves hz ing
disappeared
Formoh (19) gives the zmplit de of energy-
~adi tmg flee waves meating coastsl wash 7his
formoh is zpplicdole to high Froude n mbers, before
~eachmg the degerexated case ~epresented by formuh
21)
An ult of urt ship m y therfore creote less wash
6hon o high speedship
We 6herefore obtsin 6he chssic result ii l u tr ted by
(Shxmz 1968), m which the decrease in 6he
zmplitud spectrum with z Froude n mber
betweff~ 0 5 md 0 7 was demon tmted for z
particohr case
OCR for page 960
Numerical computation of the wave amplitude
spectrum
We fir t find She singularity dish~bution [for
mstance c son e distribution o(x,y,z) ] equivalent to
She hull using c Neum um-Kelvm code Rrard 1972)
Sol ing th Neumarm liel in >! em of equations
is c rather dffhcut task for which c c mpletely
satisfato y cnsw r m't be given for She moment:
how ver, some m meri 31 code have been
implemented that c m give i retesting results
POTFLO implements She medhod of son e
dishibution of singularity, which allows to solve She
Neumum-Kelvin problem by solving m integ al
equation over She hull
As soon es the son ce di Ablution O is k own, w
c m compute:
eThe wave amplitude A(q), th ought She
computation of Kochm function H (q) (formula (8)
Ed (15)) This is She direct method
eThe fiee surface elevation 11 clo g longitudinal
cuts Then, using c kind of Fou ier crudysis ( c code
called ACVCL) of it, w m obtain the wave
amplitude spechum This is th indi ect method
This second method is useful to check She
computations done by the first method
Resudts
Neum arm-Ke lvin the cry assume s Ulna t She ship is
slender Ed Nat the peed is not too high As She case
of c high peed ship is perhaps out of the scope of this
Theory, w have h led to have the most co fident resell
es w c m To achieve this, w did c sensitivity study
of She resuts with re pect to She refinement of She
mesh of the hull The case celled N It contsim 320
elements, whereas Nrt2 has 477 Ed Nrt3 580
Longitudinal cut
Fig I shows I longitudinal Ant computed by
POTFLO for the 3 meshes et c hansverse distance of
50 meters for c speed of 35 k ots The origin of She
axis is located et the bow of the ship
We see c fairly good convergence of These curves
with the refinement of She hull mesh: The two h t
meshes (mt2 et mt3) give closer remits f m She fi t
one (Ott) How ver, the wave elevation et 35 k ots
em to be c bit e ague u d: d is is undoubtedly du
to the limitation of Neumum-Kelvm theory et high
speed
Wove omplitudespect?um
The wave amplitude spectrum A(q) is expressed
fi om Kochin f motion H(q) using formals (14)
Figure 2 shows She module of the Kochin
f motion with respect to q et 35 k ots using the two
method of computation ( direct use of POTFLO or
mdnect u e of POTFLO f ough ACVL)
We find c rel.ttne good cg cement between She
results of the two method How we c maximum
cppe s around 75 deg ees: w c m be doubtfu Shoed
She real Existence of it, Ed it cone pondg according
to the dispersion equsti m (12) to high fiequ ncies
Also, these waves should not propagate why far
Geneeally p mug, big males (clove 70°) are
problematic: indeed, the Kochin f motion sh mid
Dish ohm q lends to 90' But, computations are
hardier, es this area cone ponds to fiequ y going to
i fmity The di ect consequ nce is that the valet s of
She Kochm fumcti m are not why accu ate m this men,
Ed the divisi m by con q to obtain A(q) makes it
worse it is necessary to humcate the wave amplitude
p an m m She vicinity of 90°
Conclusion
Although some limitations of the code, She wave
amplitude speck computed are Representative So
Hey c m be u ed es Inputs of c wave propagation
code
The spectra have been computed at 3 speeds, 15,
25 Ed 35 k ots with th mesh Nrt3 For th sake ok
clarity, thei 3mplitud s has been multiplied by costs
We note f(3~ig(3)=cos 3A(3) Fig 3 shows reel part
f Ed im33~3ry pan g at 35 k ots
COASTAL WAVE PROPAGATION
310DELLING
O ce She ship wave field has been computed, one
has to detemmme the wave characteristics m She
nearshme zone
It is Hen necessary to describe the wave field in c
fixed geog 3phi 31 f 3me h fi is fixed f 3me Q'X'Y'
Z', the Z' axis is di ected ve ticclly upward, X'Y'
defining the hori motel pi m coincident with the still
water level The ship route is oriented with m male d
mch that it is possible to express quantities fiom She
fi 3me O ye to the f 3me Q'X'Y'Z' f ou h following
formula:
a= X'cos,d+Y'sin,d up t)
y= X'sm,d +Y'cos,d
22)
In She following text, She prime (') symbol will be
omitted in ord r to simplify She no umni
Then, She wave field descry ed by equation (13) is
expressed in She QXYZ fi 3me by:
OCR for page 961
.
(xYt)=~(~)~ i[~(~)xc~(~+~)+~(~)~(~+~) ~q
-
(23)
whe~e m is 6he pulsation given by ~ehti m (2) with
6hkh = 1, wh n 6he ship evolves m d ep water To
eah propagation mgle 0, w c m cssociate c
fi e qu y f m th fi e d f mme such 6~t:
f 2~Uo cos(o + 3)
(24)
Couphug betweea amplitude speetrum and
varirmee (or energy) speetrum
The 6hird-generction spechal wave model
TOMAWAC is used here for modellmg 6he wave
propc,3tion towards 6he shore it computes 6he
diecticrud wave varia e (or energy) spechum As
for cll pect~al models, it cssumes 6~t the phcses
~ehted to eah wave component (i e eah fiequ ncy)
me distributed m c r mdom way A lot of details on
6his model m be foumd m 6he foil wmg references
(Aekrecht et ol 1998, Benoit 1995, Benoit et ol
997)
When comparmg 6he expressions for 6he see level
varictiom th ough the cmplitude spectrum cpproah
m one hmd, md 6 ou h 6he wave varimce spectrum
cpproah m the odher h md, we c m propose c rehtion
between th cmplitude pectrum A(0) md 6he
di ectiorul variance pechum F(f,0):
F(f O) 60 A2(0)
(25)
where A0 md Af are the di ectiomd md
fiequ ncicl discretizations used by 6he model
~espectively This is c pu ely m merical rektion,
which hcs no real sig fficance m termes of
math matical equivalence The valu s of F rekted to
A(0 ) depff~ds on the discretizations used in the wave
propagation model
Pratically pecki g, for eah mgle 0, one
computes c fiequ ncy f following equ~ti m 24), md
one computes valu s of F(f,0) following equ~tion
25), once Af md A0 are fxed
The msbbilities that h~ve beff~ foumd by 6he
Bcssin d'Esscis des Carenes m 6he m mericcl remits
~ehted to 6he cmplitude pectrum for 0 mgles g ecter
6 m 70°, suggest to cdopt c tnsncatme of the wave
cmplitude pectrum md to forget the wave amplitude
sp dchumi fcrmationforsuchhigh mgles
Highdighthg simudations
Wave propagation
nerfommed on c maritimr
simohtions have been
domcin concerned by 6he
ferries routes coming fiom Corsicc (towm of Bcstic
md Ajacio) md ~eahmg the harbou of Nice along
6he French tivie~a Figme 4) Bcthymeby md map
SHOM n° 5176 of 6he F'ench Nx y A minimum
water depfh of Sm is imposed in fhe model: fhe
TOMAWAC model used here is no more valid for
~egiom of sh~llower water depfhs
We investigated few routes md ship speeds along
c mpubtiorud d mcm is comcident wifh the ship
route We report here only f w of the m merous
simohtiom which have been perfommed
Si~lohon 2: Rout A B Horbour Speed of 35
Imot on A B. 15 l~ot on B Horbour
The computaticrud domcm is illushated on
Figme 5 Point B is located I mmtic mile soubhward
fiom the East hfht of the Nice harbou Pomt A is
roughly located I mmtic mile clong the di~ection
313°N from B. md conesponds to the miticl position
of the ship in ou simulations We cssume fnat fhe
ship wave feld that h~s been genemted before its
position in A hcs no influ nce on the cocsbl mec of
mte~est
Si~lohon 4: Rout A'B' Horboux Speed of 35
Imot on A'B' 15 Imot on B' Horbour
In addition to simulation 4 ctove, it hcs been
decided to look at fhe case of c ship cpproahing very
near the coc tline md keepmg its cruise speed of 35
k ots on fhis A'-B'-Harbou route This lecd to:
Si~lohon 4bu: RouteA'B'Horbour Speed of
35 Imot on A' B' 35 Imot on B' Horbour
The c mpubtiorud domain for these two
simohtions is given on Figme 6 Point B'is located I
mtic mile m 6he 159°N cape fiom 6he east iight of
6he Nice harbou Point A' is roughly located et 0 6
m~u ic mile (1000 m) elong the 309°N di ection from
pomt B', md es before conesponds to the initi~l
position of 6he ship Agam, w essume that 6he ship
wave field that hes been generated befme its position
m A'hes no influ nce on the coastal mee of mte~est
Si~lohon 6: Rout A" B Horbom: Speed of 35
Imot on A" B, 15 Imot on B Horbour
The compubtiorud domem for this simohtion is
given on Figme 7 Pomt B is 6he same es for
simohtion 2 (see texte before) Pomt A" is roughly
located et I mmtic mile elong the 332° d6rection from
pomt B, md es before cone ponds to 6he initi~l
position of the ship in 6his simulation Again, w
OCR for page 962
ass me that the ship wave field that hr. been
generated before its position m A" hr. no influence on
She coastal area of interest
Simohtions are performed in c m teddy mode, in
order to reproduce She time-varymg development of
She wave field generated by the high-speed ferry At
each time tep, one determines the ship position md
assigns the wave vari mce spectrum deduced from She
ship speed et each node along the bo mdary of She
domain, that coincides wish the ship position This is c
realistic procedure which enables to reproduce She
wave energy q mtity that is h m fenedbythefer y to
She cc mic domain
With She help of fmite element tech iq~x, mesh
size for all simulations varies fiom 80 m offshore to
less thm 15 m near She shore Time step is fixed et
I s in addition to the usual processes that effect She
wave propagation geomehic refiation md shoaling
m~mh 1, the model also taco mts for She dissipative
process th o gh depth-induced wave breaking
The wave variance sp ctrum is discretied using
25 he.p~ e. fj dime ted wish re pect to c
ge met i serie if = lo 05 (I I'd ~ I, ad 'J dmYnons
leg larly pa d with AO = 15
For i formation, one simulation, which roughly
cone ponds to 6 minutes duration in reel time (time
necessary for She ship to reach She harbour from She
Initial positi m A or A' or A'), requites 10 hours of
computational time on c HP B132 work tation This
Restively high CPU time is due to c ve y mall time
step ad c line discretization in fiequency md
di ection of the wave vari mce (or energy) sp cm m
Results for the 4 simulations presented here are
given on figures 8 to I I
Note: Results of nesrshore wave propagation
mod Ihng are given m terms of c spechal sig if cat
wave height computed et each node of She
c mputatiom~l domain The spectral rigmfic mt wave
height is c charateri tic q mtity of She wave energy
pectrum For t mdard see tates, this value is
equivalent to the tati tic sig i li mt wave height, i e
She arithmetic m m of the 1/3 pper percentile of She
wave heights serie of c wave train
The time varymg evol tion of -. sig ffic mt wave
height is particularly useful to determine She
maxim m value reached during She simulation et all
locations, i e H= ,,, which corresponds to the peak of
wave energy observed et th se pomts
For all simohtiom, except simulation 4bis which
cone ponds to a exceptiom~l case, values of H=L.~. do
not exceed 0 20 m near She coast Cone pending
maxim m mare heights Hmax could be 0 30 m
Two ocher i formations c m also be d duced from
She simulations:
- First i fommation deals with She time-hg between
th time so h n the feny retch s the htit de of one
con till location ad th one when maxim m
waves re tch the same con till location This time-
hg could re tch tbord 2 minutes
- Second is She mddem ss of She ship-wash
development when arrivmg et the coast, which
could be me tnued by c time a,,,= necessary for he
For each simulation, w provide, for 4 "strategic"
pomts along the shore Ptl, Ply, Pt3, sty, She time-
evolution of She sigmifc mt wave height Hs computed
by She model fr m She di ectiorul variance spectr m
Water depths et these 4 locations are:
Pomt Ptl: water depth :6 m
Pomt Pt2: water depth: 9 m
Pomt Pt3: water depth: 10 m
Pomt Pt4: water depth :20 m
This i 1 3nn an on particularly emibles to d term me,
for each simnlition, (1) the maxim m sigmifomt
wave height Ha,,, reached at these 4 positions NB:
Hs ,, is different fiom She maxim m wave height
Hmax of a wave Lain), md 2) She time T.....
cone pondmg to 6 is maxim m value
value He,,,, at one location This time a,,,,, depends
on She ship route md speed, md on the coastal
location of mterest For simnl trion 4bis, value of
That point 2 is about 20 s This clearly illushates
th rapidity of the ship wash phenomena at She
coast
OCR for page 963
FUTURE WORK AND CONCLUSIONS
Ships cpproahing ports et Restively high speeds
generate waves which propagate m~tmclly towards She
coast Ed die down on She shore m the fomm of so-
celled "wash"
The initial aim of this tudy was to assess whether
She height of th se wash waves is sigmifcant or not
The study was condo ted using available tools,
which w w re able to u e m c c mplementary
maimer thinks to a original methodology d awn up
for the st dy The main cdvmbge of this
methodology is that it emibles c non- tationary
calcoktion of wave propagation over time Th ship's
hajecto y with ~ rd. to the coast is therefore
mimpo tat The feasibility Ed validity of She
proposed methodology have been d monshated m c
st mdard high-spee d ship
The st dy erect led the pinpomtmg of coastal melts
most exposed to She wash phen menon As regards
She example of She Port of Nice, wash wave height in
She most exposed cocsbl ares is I O cm to 50 cm on
average (in water depths of 5 meters), et speeds
compli mt with the legal limit kid d wn on June 3,
1998 (d me 23/98) At maxim m valet ., the wash
waves could reach c h ight of 65 cm
If no peed limit is imposed, avemge wash wave
The development of this method for computing
ship wash was requ sted by She "Bu eau des
E q 3tes Techmiqu s et Admmi t Hi es ares
Accid nts en Mer BEA-M R)" ihhrnime
Techmiccl & Adminishative Board of h qui y)
Ed baked by the French Minist y of T mspo I,
Department of R se ch Ed Scientific &
Techmiccl if Ens)
The authors would like to thick the '7 stit t
Fr mgais de Navigation" French Shipping
Institute) Ed the experts fiom BEAM R who
ernt led the authors to d cw up 6 is practi al tool
thanks to Heir advice, expertise ad in-depth
k owl age of She site
REFERENCES
Aelbreeht D., Benoit M., fiascos F., Goasguen
G. 1998 Prediction of 11 bore Ed Nearshore
Stomm Waves Using c Third Generation Specnal
Wave Model Proc of ISOPE .98 conference, Vol.
111, 71 76, MontreaL Conodo, M y 1998
Benmt M 1995 Log~ciei Tomawa de
fneoriqu de h ersion 1 0 R ppo t EDF-DER
tnntt~r is /0 cm on avenge with c maxmm m BE-42/95/047/A
6beorencal valet of 90 m
The shipping authorities Ed shipyards now base
a ess to c rage of calcoktion codes emiblmg She
prior q mtitative evalusti m of high-speed ship wash,
using con till zone relief Ed badhymet y maps for She
are Is crossed by the ship Operators will ah ~ elk e, in
She futme, be Cole to eq ip themselves with new high-
sp cd car ferries satisfying the safety standards laid
doss by shipping authorities
The aim of cunent research is to optimize
calcoktion ules, in parncular es regards hull
discretication (m mber of son es), badhymet y, She
amplitude Ed energy spectrum ( mgle Ed f~equ ncsl
Ed She le gth of She Kelvin sector to be taken i to
tcco mt Of her parameters should also be considered,
su h es She effect of She ship's smkage Ed trim
In She longer to m, st dies could be mdertaken on
She i lie nce of She ship's non-stationary character on
wash Ed in parti ular She effect of speed red s non
Ed heeding charge du ing port cpproah This t pe
of non-sbtionary situsti m is w 11 represented by She
TOMAWAK software but not by She POTFLO p~e-
processor
Acknowledgments:
Benoit M., fiascos F., Beeq F. 1997
D velopment of c third generation shot w water
wave model with mshuctmed pttul meshing
Proc of ICCE 96 . or. em. e, Orb ado, USA,
September 1996
Brand, R. 1972 The representation of c given
ship fiom by singularity di tributiom when She
boumduy condition on She Ire surface is
Imeari:osd Joumtl of Shin Research, Vol 16,
number I, March
Danish hlantime Authority, Janary 1997
R port on She Lmp tot of High Speed Ferries on the
E corral E s no merit
Eggers, K.W.H, Sharma, S D, Ward, L.W.
1967 An assessment of some experimental
method for determining the wave mckmg
characteristics of c ship form, Tr msactions Srume
75,112-144
Heurik Kofoed Hansen, October 1996
Tech t al h vestigation of Wake Wash fi om Fe t
Ferries DH DaishMaritimeAubhorities
Kirkegaard J., Hajhrd N., Holmegaard
krsitensen H.O., 1998 Fast Feny Operation in
OCR for page 964
Danish Waters—XXIX session de l' association ~ ~ <
internationale des congres de navigation, La Haye. ~ ~ ;
Kostyukov, A.R - Theory of Ship Waves and
Wave resistance - ECI7 Towa City Iawa, 1968. ~ 0~2 ~ ~ ~~ ~ i- $~
~ :E : ~ ': ~~ i' ' .: .
Newman, J.N. 1977 - Marine Hydrodynamics - ;~ 13~1~ - ~6~ ' ~ ~ ~~ ~
MIT Press, Massachusetts. ~ ~ . ~ ~ .~ ~ ~ ~~ ~ ~ ~ ~~ ..~:
Ogilvie, T.F. 1977 - Singular perturbation ~ ~1 i- :- ~ ~ ~ ~~ ~ i
problems in ship hydrodynamics Adv. Appl. ~ ~ ~6
Rech. 17, Academic Press. ,~ ~~ ;
Sharma, S.D., 1969 - Some Results Concerning ~ ~436~ '6, ~<
the Wavemaking of a Thin Ship Journal of Ship ~~ ~~
Research 13 72-81. ~~ ~ 10 20 SiD 40 ~ 50 ~ 70 -A
7 7 to
Stumbo, S., Fox, K., Dvorak, F. Elliot, L. The
Prediction Measurement and Analysis of Wake Figure3: Wave amplitude spectrum at35 knots
Wash from Marine Vessels - Marine Technology,
vol 36, no 4, Winter 1999, pp 248-260.
Em____ ~
100 200 300 400 500 x(m)
Figc`'e 1: tongi/ud'r~a/ ~ As ',
1 6000 r r i ~ :
14000~in Function 35 knots: potflo-acvcl ~ I ~ Ash i~
~2000 ~ :. ~ 1 I== =~—
10000 acvf~cl ~ ' I
8000 \\
6000 \,
4000
2000
O 1 1 1 1 1 1 1 it.,
0 10 20 30 40 50 60 70 80 90
theta(deg)
Figure2: Kochin function
OCR for page 965
| SIMI]LATIONI: 25 knots onA-B; 15 knots on B-Harh~our | N~cel~rhour \
late Mann
SIMULATION 2: 35 knots on A- B; 15 knots on B- Harh~our |
SIMULATION ~ his: Simulation ~ in steady state
B==
Nice
C - e
Cape 00 |
anon -
~nnn -
SIMULATION 5: 25 knots on All- B; 15 knots on B1l- Harh~our
C TACT IT . ~ T T Nan ~ · ~ q knntc a" ~ I I _ T] · 1 q knntc a" TO I I _ T-T=rLn~lr
750 1000 1250 1500 1750 2000 X (m)
Figure 5 - Configurations for simulations 11 2 and 2 his on the A- B- Harh~our route
(normal route for ferries coming from Bastia - Corsica)
SIMULATIONS: 25 knots onA1-Bl; 15 knots on B1-Harh~our
SIMULATION4: 35 knots onA1-Bl; 15 knots on B1-Harh~our
SIMULATION Allis: 37 knots on A1 - B1- Harh~our
Y (m)
fir
too -
~ooo -
2500 -
~'
Care 309°
Aim ~~ Q; ~~
A,
2000- 1 1 1 1 1 1 X
750 1000 1250 1500 1750 2000 (m)
Figure ~ - Configurations for simulations 31 4 and This on the A1- B1- Harh~our route
(Exceptional ease of ferries coming from Bastia very near the euastline)
0 20 -
0 10 -
0 05 -
noo-
,^
(m)
d000
Cape 0~ 1 3500
2500
2000
750 1000 1250 1500 1750 2000 X (m)
Figure ~ - Configurations for simulations 5 and ~ on the All- B- Harh~our route
1 normal route for ferries coming from Ajaceio - Corsica)
4500
4400
4300
4200
4100
4000
3900
SSOO
3700
3600
0 25 - Hs (m) ~ ~ ]
' . .
000- ~ _~f .... ..
0 125 250 375 500 0 125 250 375 500
- Pll
~ 1
pi ~ ~ Nice
~ ~ Cape
I.
Pl3 ~
Pl4
_-
800 900 1000 1100 1200 1300 1400
Figure 8
SIMULATION 2
Time evolution
of the significant wee height Hs
at 4 coastal locations
n A -
0 125 250 375 500 0 1 I
Time (s)
.-
. . .[:
. ... . ... ... ....
.
250 375 500
Time (s)
OCR for page 966
4500
4400
4300
4 2 0 0
4100
4000
3900
SSOO
S700
3600
025 r
02-~
015 -~
01
0.0s -
O -,
02s~
0
015
0.1 -
0.05 -
o -
o
4sOO
4400
4300
4200
4100
4000
3900
SSOO
3700
~nn
0.70 -
0.60 -
0 50 -
0 40 -
0.30 -
0 20 -
0 1 0 -
0.00 -
r~n
Pll ~
~ 1
- \ Pt ~ C~ Niee
~
\~ Pl3 ~
~ Pl4
_ ~\\ (fer7y Fo~e)
800 900 1000 1100 1200 1300 1400
F~ure ~
SIMULATION 4
4sOO
4400
4300
4200
4100
Time evoIution 4000
of the significant w~e haght Hs 39°°
~ 4 coastal locations ssoo
~oo
acoo
Pll ~
~ 1
Pt ~ ~ Niee
~ ~ Cape
~.
Pl3 ~
Pl4
800 900 1000 1100 1200 1300 1400
Figure 11
SIhIUL ATION
Time evolution
of the significant w~e h~ght Hs
at 4 coast31 locations
0 125 250 375 500 0 125 250 375 500
Hs Im)
nQ~ .
Hs (m) [ - - ~ u 2~ - Hs (m) L -
.............................. . . 020- ......................
\ ;;{ ' ~ i DDb
125 250 375 500 0 125 250 375 500 0 100 200 300 400 SC
Time (s) Time (s) Time (s)
Pll ~
- \ Pt ~ C~ N'ee
Cape
Pl3 ~
Pl4
_ \~ eny ro~e)
. .
O 70 - Hs (m) [: , ,~
0.60 - .
O . s O . ~ , .. ... . .
040- (\
010~~ / ~ _~ ...
0.00
0 1 25 250 375
Time (s)
100 200 300 400 500 0 100 200 300 400 500
I
1 ~ ~ ~ 1 1
1 . . . 1Pl41
4 : ~ : 1 1
1 ' ' ' ' 1
~ : : : : 1
~ 1 . ~ 1 ~
. ~
I : : f : : I
1 ~ eJ ~ ~ I
I . f . . I
—- ' ' ' 1
i ~ i
.~D 0 100 200 300 400 500
Time (s)
Figlre 10
SIhIULATION 4 bis
Time evolution
of the significant w~e heght Hs
~ 4 coasta1 locations
~oo
'E~
... . ... . ... ... ... I
~::::::::
500 0 125 250 375 500
Time (s)
OCR for page 967
DISCUSSION
L Doctors
The University of New South Wales, Au m clip
Cou id the mthors kindly co firm that the first
part of their calculation employs c deep-water
qucsi-stecdy-state calculation for She e timation
of the wave spectrum Presumably the second
part of She cclcubtion is tmly unsteady es He
procedure follows the waves es Hey enter the
shallow water Ed refract Would Here be c
large en or for Hose cases where She vessel is, in
fact, accelemtmg or deccelemting?
I would like to th mk the mfhors for m
interesting Ed p tcticsl paper on m import mt
topic
AUTHOR'S REPLY
Referring to Dr Doctor's question, w confirm
that m the first part of She calculation w use c
deep water quasi- tecdy-state c Mu I In on for the
estimation of the wave spectrum That is, w
neglect the phase during which the vessel pass
from c const mt speed to c smeller one es it
arrives near She harbour in fact the effect of m
accelemti g motion onthe wave mckmg of c
vessel is c ve y complicated problem for which
solutions exist only in some cases We make for
instance reference to the work by D octors L J.
Ed Sharmc S. D on the wave resistance of m
cir~ushion vehicle m steady Ed accelerated
moti m (J. of Ship Res 16,1972) For She general
case of c ship, to our knowledge, there is no
validated prog cmme computing the wa ve
resist mce m accelerated moti m Our application
to high speed ferries washes cone ponds to She
decelerated motion for which no validated
solution seems to exist L t us add Nat She
as mmption of deep water for the calcu he on of
the ship wave spectrum is correct for the case of
the approach of the harbour of Nice Indeed, She
water depth is very large along She con t except
very near the harbour ent mce We co firm also
that our calculation is tmly unsteady in the
see ond part when the wa ve s pr opagate mto
shallow water
DISCUSSION
D Hend ix
Naval Surface W art ne C nter, Carderock
Division, USA
This pap r presents c very mtere ting application
of potential flow to c current design problem
Since, es She hors observe, "Solving the
Neumarm-Kelvin system of equations is c rather
difficult task " I would I ke to suggest Nat the
mthors consider using the slender ship theory to
detemmme She wave amplitude function As we
repo ted earlier this week, (in Practical CFD
application to design of Wave Cancellation
Multihull Ship) the use of slender ship theo y
allows the computation of m my mme conditions
thm other competing methods while conectly
capturing trends due to geometry The use of
this wave amplitude f notion should allow
design of high speed ferries to consider the effect
of wake wash
AUTHOR'S REPLY
Refen ing to Dr Hend i '. comment, w
consider favourably its suggestion to use mother
ship .. a ve theory to determine the wa ve
spechum One c mdidate is the slender ship
theory which ht. been successfully used for She
design of wave cancellation multihull ship es
mentioned by Dr Hend i We have been se y
much impressed by the ~ em its obtained by him
Ed his co mthors in fact, one of our project
was to use the high speed slender theory of Tuck
E O. presented in 1 9SS et the The d Inremtrional
Workshop on Water Wa r e. Ed Floctmg Bodies
(c strip ah or. for wave resist UK e) But it seems
that this theo y is valid only for ve y large
Froude mmmbers Prot at h She slender theo y
used in Dr Hend ix 's pap r is more efficient
Representative terms from entire chapter:
wave amplitude
