Transactions on the Built Environment vol 5, © 1994 WIT Press, www.witpress.

com, ISSN 1743-3509

Load prediction due to green water on deck

mounted equipment forfloatingbodies
B. Hamoudi & K.S. Varyani
Hydrodynamics Laboratory of Naval Architecture
and Ocean Engineering, Glasgow University,
Acre Road, Glasgow, G20 OTL, UK

ABSTRACT

This paper presents a new method to predict the load in the longitudinal direction on a vertical
superstructure or deck mounted equipment due to green water shipped on deck of floating bodies
(ship, platforms) with or without forward speed during the operation time. The study consists of
theoretical and experimental analysis along with some assumptions to simplify the problem which
is highly non-linear. Theoretically, the load in the longitudinal direction is assumed to be a function
of the wave velocity, the distance of the vertical wall (superstructure) from where the water is
shipped and forward speed. Experimentally, the applied load in the longitudinal direction is
measured using plates mounted on load cells on the deck, where green water is expected to be
shipped. A catch tank technique is used to collect the mass of water. The results of this study can
be used as a tool for the development and design stage of advanced marine vehicles.

1. INTRODUCTION

In rough weather and in head seas in particular a marine vehicle is subjected to deck wetness
phenomenon. This phenomenon can be very severe and may cause serious damage on the deck
level. Moreover, the crew's task on the wet deck may become all the more difficult. The
performance and effectiveness of the operation of the vehicle will be governed primarily by the
wetness of the deck, The severity of deck wetness event can be decreased by slowing down the
speed of the marine vehicle or changing the heading angle until the storm is over.

The deck wetness phenomenon has been investigated by many researchers. Conclusions in
reference [5], [6] and [7] state that the main parameters which influence this event is the freeboard
(the above water-line structure) and to certain extent the forward speed [8].
 Transactions on the Built Environment vol 5, © 1994 WIT Press, www.witpress.com, ISSN 1743-3509

146 Marine, Offshore and Ice Technology

In this paper analytical and experimental investigations have been performed for the high speed
container ship S175 in regular waves.

2. IMPACT FORCE IN THE LONGITUDINAL DIRECTION

2.1. Definition
The impact force in the longitudinal direction is the force due to the mass of water hitting the
vertically positioned equipment. The mass of water is randomly distributed after the occurrence of
deck wetness.

2.2. Formulation of the impact force

As shown in Figure 1 as the relative vertical motion exceeds the freeboard, deck wetness occurs
and an undefined mass of water, which has high velocity comes on deck and hits the
superstructure or any equipment mounted on deck such as gun, etc. This force is variable in nature
and is a function of the velocity of the undefined mass of water, forward speed of the marine
vehicle, quantity of the mass of water and the magnitude of the relative vertical motion when
exceeding the freeboard level.

Because the vessel is navigating in head seas with certain forward speed if any, the mass of
water is moving in the opposite direction with a certain velocity which is the wave velocity plus the
forward speed of the vehicle.

It is a very complex task to predict the volume of the mass shipped on deck, but some
assumption have to be made in order to linearise the complexity of the problem. The mass of the
water is estimated from the water shipped on deck and it can be calculated from the volume of water
as follows:

m = pV (1)

where:
p : water density
V : volume of water.

The volume of water is given in [1] by the summation of parallel sectional elements as follows:
L
V = jA(x)dx (2)
o

where:

A(x) = B(x)Z(x) (3)

 Transactions on the Built Environment vol 5, © 1994 WIT Press, www.witpress.com, ISSN 1743-3509

Marine, Offshore and Ice Technology 147

B(x) : deck transverse beam at x

Z(x) : longitudinal wave shape function.

The phenomenon is very complex and highly disturbed; for this reason the volume calculated
in equation (2) consists of mixture of water and air, a dispersion factor is incorporated in the above
equation which can be written as follows:

L
V = JkA(x)dx (4)
0

where:
0<k<1.0

The dispersion coefficient k depends on how homogenous the mass is, which is mixture of
water and air, it can be chosen by intuition and/or can be calculated from experiments as is the case
in this investigation. Moreover, this coefficient can be predicted approximately by comparing the
calculated and the measured forces.

If the water is crossing certain distance called the wetted length L', which can be identified by
specifying at which station the impact force is going to be exerted, and from which reference point
in the fore end the water is coming. The crossing time can be approximated by dividing the crossed
length by the velocity of the mass of the water. The approximate expression for the impact force in
the longitudinal direction is given in [11] as follows:

(5)
.L/

Finally the force in the longitudinal direction can be written as follows:

(6)

The calculated force in the longitudinal direction in equation (6) is predicted under some
assumptions which are as follows:

- the mass of water does not vary along the crossed length,
- the mass collected in the catch tank did hit the vertical wall without any water coming from
other directions,
- the mass of water has got a constant energytillit hits the obstacle,
- the crossed length reference starts from the very end of the overhang and
- the mass of water moves in the longitudinal direction only.
 Transactions on the Built Environment vol 5, © 1994 WIT Press, www.witpress.com, ISSN 1743-3509

148 Marine, Offshore and Ice Technology

Equation (6) can be interpreted as the impact force whixh is inversely proportional to the
crossed length. In another words, if the vertical wall (superstructure, appendages, deck mounted
equipment, guns...etc.) is at a distance from where the water is getting shipped, the impact force
applied decreases. Such remark is in line with the results presented in Figure 7 in reference [1].
Figure 2 presents the impact force per unit mass as a function of the crossed length.

The mass of water, while moving, is varying in quantity and shape before hitting the vertical
wall. This mass of water, which will be shipped on deck after hitting the wall depends on the
shape of the forebody of different types of ship (freeboard, flare angle...etc). Furthermore, the
knowledge of quantifying the mass of water is still at its initial stage. For example, Mizoguchi [2]
found that the water heights on the triangular or sharp deck form with narrower deck area, are
higher than that of the semi-circle or blunt deck form with wider deck area. And the speed of the
water particle of the triangle deck form are faster than that of the semi-circle deck form. However,
Lloyd [3] found in his experimental investigation that the effect of deck shapes, bluntness or
sharpness, are less conclusive. He noticed that both blunt and sharp stems appear beneficial at 17.5
knots for his narrow beam LEANDERfrigate,but sharpness has much less effect at 20 knots.

2.3. Experimental investigation on the impact force

Because of difficulties in the analytical approach of deck wetness field, predictions of wetness are
often made directly by testing a model in regular or irregular waves and observing the phenomenon
of shipping of water. A video tape records and catch tanks are often used to witness and clarify the
event.

The model of the container ship was run in regular waves for two draughts, the ballast and the
loaded draught. The event of deck wetness did not occur for ballast conditions except for some
sprays. However, for loaded conditions, the deck wetness occurred for most of the cases. One
can conclude that freeboard plays an important role in reducing the phenomenon of deck wetness.

In order to quantify the wetness impact force in the longitudinal direction on a vertical wall, in
the present investigation, nine square plates were mounted in array over the deck at station 91/2, as
shown in Figure 3. Each plate was mounted on a load cell (Censotec 'D' series, range 0.5 Ibs).
As the top array of load cells failed before the set of experiments all the runs were conducted with
the middle and the bottom load cells only. A catch tank was placed inside (at the centre of gravity)
the model of container ship S175 to collect the incoming water on deck. Using this technique and
equation (6) water impingement on superstructure and deck mounted equipement can be predicted.
Details of model and test are given in reference [4]. The mass of water was weighed after each run.
Table 1 presents the magnitude of the impact force converted to an impact pressure (impact
load/plate area, model scale) in the longitudinal direction (six plates mounted on load cells).

The measured impact pressure generally increases with increase in forward speed as is
presented in Table 1. The results are consistent except for some frequencies. These inconsistent
results may be due to non-linearities and some error from the calibration procedure during the
experiments.
 Transactions on the Built Environment vol 5, © 1994 WIT Press, www.witpress.com, ISSN 1743-3509

Marine, Offshore and Ice Technology 149

The dispersion factor proposed by Oliver [1] has been calculated in this investigation by
dividing the measured impact pressure from experiment and the calculated pressure from equation
(6). As shown in Table 2 this factor lies between 0 and 1.0. It is equal to 0 when there is no force
applied on the vertical wall and it is always lower than 1.0 due to non-homogeneousity of the mass
of water. This factor can be calculated by the following equation:

k=Measured Load/Calculated Load (7)

The dispersion factor which is presented in Table 2 may not be accurate at this stage due to the
assumptions given above. However, it can be used asfirstapproximation in the design stage.

3. DISCUSSION

Our knowledge is not adequate enough regarding the above water-line structural design, because
the investigation in the environmental loading on any marine structure occurs below the water-line.
Moreover, the event of green seas does not occur with the same regularity as is the case of loads
applied constantly on the under water-line hull. In this investigation, a simple model was used to
linearise a very complex phenomenon. In real life, the event is complicated [9] however, the
present theory is a good starting point despite the heavy assumptions made. There are other
parameters which have to be included in future, such as wind effect, variation of water mass
quantity, speed during its crossing period till it hits the barrier (vertical wall) and the direction of
water when it is disturbed by the presence of the hull.

The experimental tests were carried out in the towing tank, where the model was free only in
heave and pitch as presented in Figure 4 and 5 in [10] and compared with the results presented in
the 15th ITTC, there are discrepancies in the comparison, and this disagreement is due to the
restriction from other motions, so the deck wetness event which occurred during the test was
influenced, therefore the results given here are partially consistent in their nature. One can say that
the full scale test is the best way and real trial to investigate the green seas event.

4. CONCLUSIONS and RECOMMENDATIONS

This paper presents a new method to predict the load in the longitudinal direction on a vertical
superstructure or deck mounted equipment due to green water shipped on deck offloatingbodies
(ship, platforms) with or without forward speed during the operation time. The same
phenomenon was investigated experimentally and one may conclude the following points:

- The freeboard plays an important role in reducing the deck wetness event.
- The force in the longitudinal direction presented has only been recently formulated and it is
inversely proportional to the distance at which the vertical wall is positioned from where the
water was shipped.
 Transactions on the Built Environment vol 5, © 1994 WIT Press, www.witpress.com, ISSN 1743-3509

150 Marine, Offshore and Ice Technology

- The measured load is always lower than the calculated load and the dispersion factor always
lies between 0.0 and 1.0. This is because not all the collected mass of shipped water, which
by its mean the predicted load was calculated, hits the vertical wall on the deck of the floating
body. In another words, a partial quantity of water came from other directions, such as sides
and from above the walL.etc..
- The dispersion factor tabulated here is only for the case of the container ship and for certain
conditions. Similar experiments need to be performed for other type of vessels. However
these results serve as a good outline for initial stage design purposes.
- It is recommended that the mass of water, shipped on deck, needs more investigations in its
geometric, kinematic and dynamic properties. However, to satisfy these properties, more
theoretical work using experimental investigations needs to be carried out.

ACKNOWLEDGEMENTS

The author sincerely acknowledges Dr. H.S. Chan and Dr. A. Incecik for the rich discussion.
Financial supports from the Government of Algeria and The British Council are gratefully
acknowledged.

REFERENCES

1. Oliver, J.C. "Analytical Methods to Predict Transient Loads on Floating Platforms", IOS
conference, 1981.
2. Mizoguchi, S. 'Analysis of Shipping water with Experiments and the Numerical
Calculations', J.S.N.A. Japan, Vol 163, June 1988.
3. Lloyd, A.R.J.M., Hammond, M.A. 'The Effect of Bow Shape on Ship Motions and Deck
Wetness', Report AMTE(H) R82012, Unclassified, July 1982.
4. Hamoudi, B 'Effect of Speed and Draught on the Bottom Slamming and Deck WEtness',
Dept. Report, NAOE-92-18, University of Glasgow, Glasgow, Apr. 1992.
5. Lloyd, A.R J.M., Salsich, J.O., Zseleczky, J.J. 'The Effect of Bow Shape on Deck Wetness
in Head Seas', RINA, 1985.
6. Lloyd, A.RJ.M. 'Deck Wetness Experiments', 20th ATTC, Vol.2, New Jersey, Aug.
1983.
7. Bales, N.K. 'Minimum Freeboard Requirements for Dry Deck Foredecks: A Design
Procedure' SNAME/STAR Symp., Texas, Apr. 1979.
8. Fukuda, J.,Shinhai, A. 'Speed Loss of a Container Ship on the Different Routes in the North
Pacific Ocean in Winter', ISSHES-83, El Pardo, Spain, Sep. 1983.
9. Hamoudi, B., Varyani, K.S, Incecik, A., Chang, H.S. 'Private Discussion', HLMRP
Meeting, Glasgow University, 26th March 1993.
10. Hamoudi, B., Varyani.K.S. 'On the Prediction of Impact Pressure due to Bottom Slamming
on Container Ship', CADMO'92 Proc., El-Pardo, Spain, Oct. 1992.
11. Hamoudi, B 'Dynamic Response of Hull Due to Bottom Slamming and Deck Wetness:
Review and Extension', Dept. Report, NAOE-93-18, University of Glasgow, May 1993.
Transactions on the Built Environment vol 5, © 1994 WIT Press, www.witpress.com, ISSN 1743-3509

Marine, Offshore and Ice Technology 151

U+C

C: Wave velocity
U: Forward speed
m: Water mass
F: Force
L: Crossed length
Figure 1 Force in the longitudinal direction.

a O Load (m/sec^2). Fn=0.0

175- A
I 4- Load (m/sec^2X Fr=0. 1
150 - * A. Load (rn/sec^2). Fr=0. 15
.1
125- * Load ( m/sec*2) . Fr=*0.20
I -K A Load(m/sec^2\Fr»).278
1 100 -
75 - ° S.
'§> *
5 50 -
25- Q £ f
q o g
1 , 1 S i
0.0 0.2 0.4 0.5 0.3 1.0
Vertical Wall (m)
Rgure 2 Force in the longtudinal direction per unit mass for different
Froude Numbers and frequencies.

Figure 3 Load measurement device on the deck of container ship model S175.
Transactions on the Built Environment vol 5, © 1994 WIT Press, www.witpress.com, ISSN 1743-3509

152 Marine, Offshore and Ice Technology

' 3 O O O 3 O CM ' CM0 >~m fi '\a sO=3-v5553

0 0rM 0ro w,
r- CM >n CM oo
- — CM
d d d d d d d o d d do d d d d d d d d d d d dd
,1 * -1
5 "3 -j v-t\OO\ — r^oocMoo ^ooi§"*"^^~3 OCM — OOOJ^ — OJ 35SSSocM^""o
E s 3 00000000 00000000 0-0- -.000
S"-i
i in KN/m*2
on of Ihc dr

•»»• O CM
oo oe _;-. r^> m
r- CN«o*%•«/~sCMrsiUsJ5— wiO '— r-; CM »r> CM rr ' ' w->in op
P^ un Ov
\C g^ m C\
oc
TT — rn
»o -«r x *"
2 ^ •
§
r*- *~« vo oo Vi CM o* O3§S2SS2g teg^sssss r~O«^.oo«QCMONO
NCgTr-Oocu~oo
1 §1
-
£ | = 1 5

J:vc—,rr r^
r» r*- ; Cs \C TT \O O <n OO
oddddddd TTwi—TTo— OOr*-—"OJOCMCM
jdddddd
— —0000000
« — — rsjoorN^ — Or-r- Cr-w. — OO
oo — — c«J rrr«^fNo5o
c-d o d d o o o d d d d d d d d d O — — rsmooo
doddoddo ddddoddd
v> mO vO ov -vr-»r^>"O—O QOO
»o r*^ O O — OC — K~. O
5:00000000 )i dO dO dO d— oO oO dO O O O O — 'OOO t>dd — — odd
00
—o• a\ r» o oo oo u-> ooo\ — -q-OOOr- oooem — x?r>-
om cw~,—OoCM 5 oo OOf**. M f N m O O O Or^OfN
O — COr^r»"^O^ nrou-, C M—O OOQO
00000000 d d d d d d d d d d d d d d d d d d d d d d o d
- 'mcMQO 3\TrO\tCrn(NOu") TTCM«n
oo— — \corM
82538552 0 f!T«ncN — —' m O—O—O —O
O O O O O O O O cncNOOOOO
dddddod ddddddod dddddddd
ooooooo
d d d d d d d d d d d d d d d d —d —drdMdr d rOOO d d d d d d d d
dd
\)c-i o m —•^CMp—oopOoo—"
CMmoorroo
>i 8 5 p p p 5 p > p 5 5 p p p S > CMpCMp•?•
p p p p — — r n p—pO—•pp
d d d d d d d > d d d d d d d dddddddd dddddddd