The Specialist Committee on Prediction of Extreme Ship ... · capsizing with sub-harmonic rolling in case (a) and capsizing with harmonic rolling in case (d) were well predicted

23rd International Towing Tank Conference

Proceedings of the 23rd ITTC – Volume II 619

1. INTRODUCTION

1.1. Membership, meetings and organisation

Membership: The Committee appointed by the 22nd ITTC consisted of the following members:

Professor D. Vassalos (Chairman) Universities of Glasgow and Strath-clyde, UK

Dr. M. Renilson (Secretary) Australian Maritime College, Australia, and QinetiQ, Haslar, UK

Mr. A Damsgaard Danish Maritime Institute, Denmark

Professor H.Q. Gao China Ship Scientific Research Centre,

Mr. D. Molyneux Institute for Marine Dynamics, Canada

Professor A. Papanikolaou National Technical University of Ath-ens, Greece

Professor N. Umeda Osaka University, Japan

In addition, the following corresponding members contributed greatly to the work of the committee:

Dr. J.O. De Kat MARIN, The Netherlands

Professor A. Francescutto University of Trieste, Italy

Professor J. Matusiak Helsinki University of Technology, Finland

Meetings: Seven Committee meetings were held as follows:

Shanghai, China, September 1999 Launceston, Australia, February 2000 Osaka, Japan, October 2000 Glasgow, Scotland, UK, May 2001 Trieste, Italy, September 2001 Heraklion, Greece, October 2001 Glasgow, Scotland, UK, February 2002

(Editorial meeting)

Organisation: The following working groups were established and chairmen ap-pointed:

Benchmark Testing for Intact Ship Sta-bility (Umeda)

Benchmark Testing for Damaged Ship Stability (Papanikolaou)

Guidelines for Experimental Testing of Intact Ship Stability (de Kat)

Guidelines for Experimental Testing of Damage Ship Stability (Damsgaard)

Questionnaire (Molyneux) Symbols and Terminology (Frances-

cutto)

The Specialist Committee on Prediction of Extreme Ship Motions and Capsizing

Final Report and Recommendations to the 23rd ITTC

620 The Specialist Committee on Prediction of Exteme Ship Motions and Capsizing 23rd International

Towing Tank Conference

Liaisons: The following Committees and organisations have been contacted: Loads and Responses; Manoeuvring; Waves; IMO (Re-vision of 1966 ICLL, Intact Stability, Harmonisation Group); WEGEMT; CRN; SNAME Technical Panel; EU Thematic Net-work − SAFER EURORO; SRA of Japan − Panel RR71; COREDES.

1.2. Tasks from the 22nd ITTC

Coordinate a comparative study of mathematical models for the prediction of intact and damage stability in waves. The mathematical models will be com-pared to the results of benchmark tests for two test ships, Ships A and B, as specified in Section 7.2 of the report of the Stability Committee of the 22nd ITTC.

Present the guidelines for experimental testing of intact and damage stability, as given in Appendix A of the report of the Stability Committee of the 22nd ITTC, in the format defined in the ITTC Qual-ity Manual.

Symbols and terminology should agree with those used in the 1999 version of the ITTC S&T List; if necessary, new symbols should be proposed.

1.3. Contents of the 23rd ITTC Report

The following chapters detail the tasks undertaken by the Committee: Chapter 2: Benchmark Testing for Intact

Ship Stability Chapter 3: Benchmark Testing for Dam-

age Ship Stability Chapter 4: Guidelines for Model Testing

of Intact and Damage Stability Chapter 5: Questionnaire Chapter 6: Symbols and Terminology Chapter 7: Conclusions and Recommen-

dations Chapter 8: References and Nomenclature

2. BENCHMARK TESTING FOR INTACT SHIP STABILITY

2.1. Introduction

This chapter describes results of the ITTC benchmark testing of intact stability. For these tests, a container ship and a fishing vessel were selected and their hull forms, captive test data and results of capsizing model experi-ments were provided in advance. On this ba-sis, eight research organisations submitted numerical results. Comparisons between nu-merical and experimental results revealed that some numerical models are able to predict extreme motions qualitatively, including cap-sizing due to parametric resonance and due to broaching. Moreover, the importance of sev-eral factors necessary for capsize prediction is noted by mutual comparisons of the numerical studies.

List of Participating Organisations

Ship A-1:

Flensburger Schiffbau Gesellschaft1 (Ms. Heike Cramer)

Helsinki University of Technology (Prof. Jerzy Matusiak)

Maritime Research Institute Netherlands (Dr. Jan O. de Kat)

Osaka University (Prof. Naoya Umeda)

Technical University of Malaysia (Dr. Adi Maimun)

Universities of Glasgow and Strath-clyde, The Ship Stability Research Cen-tre (SSRC) (Prof. Dracos Vassalos)

University of Tokyo (Prof. Masataka Fu-jino).

1 The computer program at FSG was originally

developed at Universitat Hamburg.



Ship A-2:

Helsinki University of Technology (Prof. Jerzy Matusiak)

Memorial University of Newfoundland (Prof. Don Bass)

Osaka University (Prof. Naoya Umeda)

Universities of Glasgow and Strath-clyde, The Ship Stability Research Cen-tre (Prof. Dracos Vassalos)

This order is not related to the code used in this report.

2.2. Background

The trend towards adopting performance-based criteria in favour of rules-based criteria aiming at safety improvement at sea continues unabated at the International Maritime Or-ganisation (IMO), the rule making body of the United Nations. To facilitate this process, model experiments and numerical simulations tools need to be developed and validated. However, a standard numerical prediction technique for capsizing has not yet been es-tablished. Therefore, the 22nd ITTC (ITTC, 1999) organised a specialist committee for this purpose and planned benchmark testing of numerical predictions with selected data from free running model experiments. This chapter summarises the results of these benchmark tests and highlights the importance of a number of factors to the numerical pre-diction of ship capsizing.

2.3. Framework of ITTC Benchmark Testing

In the intact benchmark testing pro-gramme, two sets of free running model ex-periments were utilised. The first set was car-ried out with a 1/60 scaled model of a 15000 gross tonnes container ship (Ship A-1) at the seakeeping and manoeuvring basin of the Ship Research Institute by Hamamoto et al.

(1996). Here the ship model capsized mainly due to parametric resonance in the lower speed region. The second set was carried out with a 1/15 scaled model of a 135 gross ton-nes purse seiner (Ship A-2) at the seakeeping and manoeuvring basin of the National Re-search Institute of Fisheries Engineering (NRIFE) by Umeda et al. (1999). In these tests, the model capsized mainly due to broaching in the higher speed region. The principal particulars and body plans of these ships are shown in Table 2.1 and Figures 2.1 and 2.2. In the experiments each ship model was self-propelled and free from any re-straints, steered on a specified course by using an auto pilot in regular following and quarter-ing waves. The angular velocities and angles were measured using an optical gyroscope, and were recorded on an onboard computer. The reference system used in this report is shown in Figure 2.3.

Table 2.1 Principal particulars of the test ships.

Items Ship A-1 Ship A-2

LPP (length) 150.0 m 34.5 m B (breadth) 27.2 m 7.60 m D (depth) 13.5 m 3.07 m Tf (draught at FP) 8.5 m 2.50 m T (mean draught) 8.5 m 2.65 m Ta (draught at AP) 8.5 m 2.80 m Cb (block coefficient) 0.667 0.597 kyy/LPP (pitch radius of gyration)

0.244 0.302

xCG (longitudinal position of centre of gravity from mid-ships)

1.01 m aft

1.31 m aft

GM (metacentric height) 0.15 m 1.00 m TE (natural roll period) 43.3 s 7.4 s AR (rudder area) 28.11 m2 3.49 m2 DP (propeller diameter) 5.04 m 2.60 m TE (time constant of steering gear)

1.24 s 0.63 s

KR (proportional gain) 1.2 1.0 KR TD (differential gain) 53.0 s 0.0 s



Among several hundreds of model runs, four runs were selected for each ship for the purpose of ITTC benchmark tests as described in Tables 2.2 and 2.3. Here the nominal Froude number, Fr, and the auto pilot course from the wave direction, χc, are control pa-rameters and the wave height, H, and wave length, λ, are the wave parameters. The initial values of ship motion were specified based on measured data except for the sway velocity, which was assumed to be zero because of measurements limitation.

For ships A-1 and A-2, the captive model experiments, e.g. resistance test, self- propul-sion test, propeller open test, circular motion tests (CMT), roll decay test and so on, were carried out mainly in NRIFE’s seakeeping and manoeuvring basin using an X-Y towing car-riage. These data together with hull offset data and the above mentioned initial values were provided to the participating organisations prior to undertaking any numerical simula-tions.

2.4. Results

The ITTC benchmark test programme for intact stability commenced in March 2000 with numerical results submitted by March 2001. Numerical prediction methods used by the participating organisations are outlined in Umeda (2001) with numerical results shown in Figures 2.4 to 2.6 together with the experi-mental results. In agreement with the partici-pating organisations the results have been pre-sented anonymously throughout this bench-mark programme.

Figure 2.1 Body plan of Ship A-1.

Figure 2.2 Body plan of Ship A-2.

X

Y

Z

G

ROLL

PITCH

YAWRUDDER

Figure 2.3 Reference system.

The numerical predictions are firstly re-quired to qualitatively agree with the corre-sponding model experiments. Thus, the quali-tative nature of the results obtained from ex-periments and numerical calculations are overviewed in Tables 2.4 and 2.5. This in-cludes capsize, non-capsize, harmonic roll, sub-harmonic roll, surf-riding and broaching. Here as a judging criterion of broaching the proposal of Umeda (1999) is used. That is, broaching is a phenomenon in which both the yaw angle and yaw angular velocity increase despite the application of maximum opposite rudder angle. Lack of qualitative agreement between numerical and experimental results identified with shading.

Table 2.2 Calculated conditions for Ship A-1.

H/λ λ/LPP Fr χc

degrees (a) 1/25 1.5 0.2 0 (b) 1/25 1.5 0.2 45 (c) 1/25 1.5 0.3 30 (d) 1/25 1.5 0.4 30



Table 2.3 Calculated conditions for Ship A-2.

H/λ λ/LPP Fr χc

degrees (a) 1/10 1.637 0.30 -30 (b) 1/10 1.637 0.43 -10 (c) 1/8.7 1.127 0.30 -30 (d) 1/8.7 1.127 0.43 -30

2.5. Discussion (Tables 2.6-2.7 & Figures 2.4-2.6)

Ship A-1

For Ship A-1 all the participating organi-sations used 6 degrees of freedom (DOF) models. However, only Organisation-A sub-mitted results that qualitatively agree with the experiments.

Organisation-A calculated radiation and diffraction forces using a strip theory and dealt with manoeuvring forces by the MMG model, utilizing a body coordinate system. It evaluated the Froude-Krylov forces, including roll restoring moment in waves, by integrating incident wave pressure up to the instantaneous water surface. With this numerical model, capsizing with sub-harmonic rolling in case (a) and capsizing with harmonic rolling in case (d) were well predicted.

Organisation-G also shows similar agree-ment but numerical results in case (a) predict capsizing with harmonic rolling, which was not observed in the corresponding experiment. The method used here is almost the same as that of Organisation-A except for radiation and diffraction modelling.

Organisation-E has problems in the pre-diction of the heading angle. In some cases the ship course changes to bow sea and then a completely different situation occurs. This model is different from the above two organi-sations in a number of ways. The radiation forces were calculated using a a strip theory with hydrodynamic memory effects. The ma-noeuvring forces, roll damping moments, re-sistance and propulsion forces were estimated

using available databases instead of the cap-tive test data provided.

Table 2.4 Overview of qualitative results for Ship A-12.

Experiment A B C

(a) cap (s) cap (s) cap (h) no roll

(b) (s) (s) (s) N/A

(c) (h) (h) N/A N/A

(d) cap (h) cap (h) N/A N/A

D E F G

(a) (s) cap (s) cap (h) cap (h)

(b) (s) (h) (h) (s)

(c) (h) cap cap (h)

(d) (h) cap cap cap (h)

The method used by Organisation-B is based on a conventional seakeeping approach. That is, heave, pitch, sway and yaw are as-sumed to be linear around the averaged course. This organisation reported that this method is not able to deal with ship runs at Froude number greater than 0.3.

Organisation-D proposed a method to avoid this limitation of the seakeeping model by a two-stage approach. Here the motions are assumed to be the sum of linear parts with hydrodynamic memory effects and nonlinear contributions. This means that linear motion was calculated around the instantaneous head-ing angle instead of the auto pilot course. However, agreement between predictions us-ing this calculation and experiments is not satisfactory. This may be partly because the initial values used in order to take memory effects into account are different to those specified.

Organisation-F is a unique example where diffraction forces were ignored, but the nu- 2 Here (h) and (s) mean harmonic and sub-

harmonic roll motions, respectively and cap indicates capsizing.



merical results do not agree well with the ex-perimental results, particularly the case (b) concerning calculated pitch motion ampli-tudes.

Following successful application to seakeeping predictions, Organisation-C at-tempted to apply CFD to the present problem. Here the Euler equation was solved by a finite difference method with fully nonlinear free surface and body surface conditions. How-ever, it can provide a solution only for case (a) without lateral motions. If the specified initial values for lateral motions are input, even for case (a) the calculation process failed. In addi-tion, it cannot deal with cases (b), (c) and (d), in which the desired heading angles are not zero. This fact demonstrates that the CFD ap-proach is not yet appropriate for practical use in capsize prediction.

Ship A-2

For Ship A-2, only Organisation-A ob-tained qualitative agreement with experi-ments. Here a 4 DOF model was used by as-suming that heave and pitch motions trace their static equilibria, which are calculated as the limit of solution sets of a strip theory at zero encounter frequency. The manoeuvring forces were estimated using the MMG model and the wave-induced forces, including hy-drodynamic lift due to wave fluid velocity, were calculated using Ohkusu’s slender body theory. The wave effects on roll restoring moment and manoeuvring forces were ig-nored as higher order terms. As a result, this organisation succeeded in predicting capsiz-ing due to broaching associated with surf-riding as well as periodic motions.

Organisation-C used a method that is al-most the same as that of Organisation-A but the nonlinear terms in the manoeuvring mod-els, deriving from the Froude-Krylov and ra-diation forces were added. As a result, for case (b) it predicted capsizing without surf-riding and with a smaller rudder angle com-pared to the results from the experiment and those predicted by Organisation-A.

Table 2.5 Overview of qualitative results for Ship A-23.

exp. A B C D

(a) non-cap

non-cap

non-cap

non-cap

non-cap

(b) surf

broach cap

surf broach

cap cap cap

surf non-cap

(c) non-cap

non-cap

non-cap

non-cap

non-cap

(d) cap cap cap cap cap

Organisation-B applies a 6 DOF model in which radiation and diffraction were calcu-lated with a 3D Green function for zero for-ward velocity. Here the change of roll restor-ing moment due to waves was taken into ac-count but the hydrodynamic lift due to wave fluid velocity was ignored. The hydrodynamic memory effect was included in this calcula-tion, although the initial values were not ex-actly as specified. While the predictions of mean yaw angle for cases (a), (c) and (d) are better than those from the other organisations, the predicted rudder angle for case (b) is smaller than the corresponding experimental results.

Organisation-D also takes memory effects into account but with a strip theory. Like or-ganisation-B, the initial conditions are differ-ent from those specified. This organisation predicts stable surf-riding in the case (b). This may be a result of some shift of the stable equilibrium point towards a wave crest or in-accuracy of hydrodynamic lift due to wave.

As a whole, the four participating organi-sations predicted the results relatively well compared to experiments for Ship A-2, the obvious exception being broaching.

3 Here surf and broach mean surf-riding and

broaching, respectively.



2.6. Factors Affecting Prediction Accuracy

As mentioned above, the mathematical models for capsizing prediction involve a number of factors without clear guidance in place on which of these should be taken into account in which case. Mutual comparisons among the organisations do not easily clarify the importance of each particular factor be-cause more than two factors are often differ-ent between the organisations. Therefore, this report reviews comparative studies of numeri-cal simulations with and without each particu-lar factor for Ships A-1, A-2 or indeed other ships.

6 DOF vs. 4 DOF or 1 DOF

Although all organisations submitted re-sults with 6 DOF models for Ship A-1, many theoretical studies with 1 DOF models can be found for capsizing due to parametric rolling. Munif (2000) estimated the capsizing bounda-ries for Ship A-1 with a 1 DOF model, a 4 DOF model ignoring heave and pitch motions (4 DOF A model), a 4 DOF model with static equilibria of heave and pitch motions (4 DOF B model) and a 6 DOF model. Here the first three models were obtained by simplifying the 6 DOF model. As a result, the following con-clusions were drawn:

The 1 DOF model overestimates capsiz-ing danger.

The difference between the 4 DOF A model and the 6 DOF model can be sig-nificant.

The results from the 4 DOF B model are in reasonable agreement with those from the 6 DOF model and the experiment. The small difference between the 4 DOF B model and the 6 DOF model derives from the fact that the natural frequency of heave and pitch motions is far from the encounter frequency with the ship running in following and quartering seas (Matsuda et al., 1997). This conclusion suggests also that coupling effects of heave and pitch on the extreme roll mo-tion are not very important.

Hydrodynamic memory effect

It is well known that the linear transient motions of a ship with frequency-dependent hydrodynamic forces can be calculated using the convolution integral for hydrodynamic memory effect. However, it is not so clear for capsizing prediction whether the hydrody-namic memory effect should be taken into ac-count or not. This is because an extreme mo-tion leading to capsizing is nonlinear and the hydrodynamic forces acting on a ship running in following and quartering seas do not sig-nificantly depend on the encounter frequency.

Hamamoto & Saito (1992) carried out a comparative study for a container ship in fol-lowing seas with and without the memory ef-fect in heave and pitch motions. They con-cluded that no significant difference exists if the added mass and damping coefficients are calculated for the natural frequency of heave and pitch motions. Matusiak (2001) investi-gated this problem and concluded that mem-ory effects can improve agreement with ex-periments for Ship A-1. Here it is noteworthy that exact calculation with memory effects should be carried out from the start of the waves. Thus the present benchmark testing, which does not specify the initial conditions of fluid motions, is not appropriate for this purpose.

Manoeuvring coefficients

In following and quartering waves, predic-tion of manoeuvring coefficients is important because hydrodynamic lift is dominant. The first question here is whether the effect of nonlinear terms of manoeuvring forces on capsizing prediction is important or not. For Ship A-2, Umeda et al. (2000) produced time domain simulations with and without these non-linear terms and concluded that the effect of non-linear terms is negligibly small. This is because the sway velocity and yaw angular velocity non-dimensionalised with the higher forward velocity are not large even during the process of broaching.



The next problem is wave effect on the linear manoeuvring coefficients. This problem has been discussed for many years but its ef-fect on capsizing prediction has not yet been fully investigated. Hashimoto & Umeda (2001) tackled this problem with Ship A-2. Their main conclusion is that the effect of waves on the derivatives of manoeuvring forces can be important with respect to sway velocity but it is not significant with respect to yaw angular velocity.

Nonlinearity in yaw

In seakeeping theory, ship motions, such as yaw, are often linearised around the inertial system moving with the averaged speed and course of a ship. In the field of manoeuvring on the other hand, ship motions are described with a body fixed coordinate system. Hama-moto & Kim (1993) introduced a horizontal body coordinate system, which is body-fixed but not allowed to roll. Cramer (2001) re-ported the limitation of linearisation of yaw motion with an inertial coordinate system.

Radiation and diffraction

In following and quartering seas, the en-counter frequency of a ship with forward speed is generally low and hence the wave-making effect is not so significant. In this re-spect, it is difficult to predict pitch and heave motions near zero encounter frequency be-cause of divergence of the 2D added mass. Matsuda et al. (1997) solved the problem by calculating the limit of the solution set of strip theory for the zero encounter frequency and confirmed that the new method explains the experimental results. In case of a 3D theory at very low encounter frequency, it is essential to use the Green function with both forward speed effect and frequency effect taken into account. This is because the wave related to frequency, the k2 wave, disappears at the zero encounter frequency and only the wave related to forward speed, the k1 wave, remains.

Hydrodynamic lift due to wave fluid ve-locity

Very small effect of wave-making does not mean small effect of incident waves as a ship behaves like a lifting surface with a time-varying angle of attack due to wave fluid ve-locity and ship forward velocity. Within the assumption of small wave steepness, this hy-drodynamic lift can be calculated as an end term of slender body theory or strip theory, which represents trailing vortices as a line doublet shed from the aft end (Umeda, 1988). For a 3D theory, it is necessary to include free vortex layers shed from the hull surface. Comparison between calculations with and without the hydrodynamic lift due to wave fluid velocity for Ship A-2 can be found in Umeda (2000). The results indicate that pre-diction of broaching is largely affected by this term.

Roll damping moment

Roll damping moment consists of wave-making, eddy-making, lift and friction com-ponents, the main non-linearity deriving from the eddy-making component. However, as the experimental work of Umeda (2000) showed, roll damping can be regarded as linear when the Froude number is greater than 0.2. Since eddies are shed away at high speed, the eddy making component disappears. In addition, the wave-making component is not significant because of the low encounter frequency and the friction component is generally small. Therefore, roll damping relating to this benchmark testing scheme consists of mainly the lift component, which is linear and de-pends on forward velocity. A comparison of predictions of broaching boundary using em-pirical methods of calculating the lift compo-nent was presented by Ikeda et al. (1988) for the Ship A-2, indicating that the predicted re-sults depend on the selection of empirical methods. Because of this, roll decay tests with forward velocity were carried out for the Ship A-2 to obtain reliable results for use in the benchmark study.



Wave effect on roll restoring moment

While all organisations took into account the wave effect on roll restoring moment to simulate parametric resonance, some organi-sations ignored this effect for broaching pre-diction. This is because the wave effect esti-mated on the basis of hydrostatics can be too large in case of high forward velocity. Results from captive model experiments (Umeda & Yamakoshi, 1986) for a small trawler indicate that the measured metacentric height in waves is smaller than the calculated value, derived by using hydrostatics. It is, therefore, neces-sary to develop theoretical or empirical meth-ods for a more accurate prediction of the wave effect on roll restoring, particularly for higher forward speeds.

Roll-yaw coupling

When a ship runs in calm water with a constant heel angle, sway force, yaw moment and roll moment act on the hull in addition to conventional manoeuvring forces and mo-ments. In this benchmarking scheme, these data from the captive tests for Ship A-2 at NRIFE were provided in advance. However, such data are not always available and em-pirical or theoretical methods have not yet been established. Renilson & Manwarring (2000) reported a comparison in predictions of broaching boundary with and without the roll-yaw coupling for a trawler and the re-sults indicate that the prediction without the roll-yaw coupling can underestimate the danger of broaching.

Resistance and propulsion

It is well accepted that predictions of hull resistance and propulsive performance have been the most crucial issue at ITTC. This is also the case in the prediction of capsizing of intact ships. This is because surf-riding, which can trigger off broaching and then capsizing, depends on hull resistance and propeller thrust in addition to wave-induced surge force. In the benchmarking study, data from calm-water

model tests were provided in advance. How-ever, since such data are not always available, an accurate prediction method is still desirable.

Wave irregularity and short-crestedness

The applicability of numerical models to realistic seaways, that is, short-crested irregu-lar waves, should be examined in the future. Although capsizing model experiments for Ships A-1 and A-2 were carried out in both long-crested and short-crested irregular waves (Umeda et al., 1995), the benchmark testing programme deals with only the case of regular waves. The experimental results indicate that capsizing danger is least in short-crested ir-regular waves, followed by long-crested and finally regular waves. However, numerical simulations in time domain for capsizing in short-crested irregular waves are very limited. Only recently Sera & Umeda (2001) executed numerical calculation in both short-crested and long-crested irregular waves with a 1 DOF model, and confirmed the qualitative conclusion, from the experiments, that wave short-crestedness reduces capsizing danger.

2.7. Concluding Remarks

As a result of benchmark testing of intact stability, it was found that numerical models can qualitatively predict capsizing due to pa-rametric resonance and due to broaching in the limited cases tested here. In the case of parametric resonance, a 6 DOF model, hydro-statics for the wave effect on roll restoring moment, strip theory for wave radiation and diffraction and experimental data of manoeu-vring forces, hull resistance, propulsion force and roll damping are used. In the case of broaching, a 4 DOF model, slender body the-ory for the added mass and hydrodynamic lift due to wave fluid velocity and experimental data of manoeuvring forces, hull resistance, propulsion force and roll damping are used. However, minimum requirements for accurate modelling of intact ship capsize have not yet been fully established.



Table 2.6 Comparison of numerical prediction methods for Ship A-1.

Organisations DOF Radiation Manoeuvring

Damping Roll Restoring

A 6

strip theory exp.

(non-linear) hydrostatics in

waves

B 6 (linear in sway, heave, pitch, yaw)

strip theory ignored hydrostatics in

waves

C 6 (coupled with fluid motion)

CFD (non-linear)

ignored CFD

(non-linear)

D 6 (two stage model)

strip theory exp. (linear) hydrostatics in

waves

E 6 strip theory empirical hydrostatics in

wave

F 6 empirical empirical hydrostatics in

waves

G 6 strip theory exp. (non-linear) hydrostatics in

waves

Organisations Roll Damping Froude-Krylov Diffraction Hydrodynamic

Lift due to Wave

Hull Resistance

A experimental+

empirical forward speed

non-linear strip theory ignored experimental

B empirical linear strip theory ignored experimental

C no viscous effect non-linear

(cfd) CFD

(non-linear) ignored no viscous effect

D empirical non-linear strip theory ignored experimental

E empirical non-linear strip theory cross-flow model empirical

F empirical non-linear ignored ignored empirical

G experimental+

empirical forward speed

non-linear slender body theory at ω=0

end term experimental

Organisations Propeller

Thrust Rudder Force Incident Wave

Hydrodynamic Memory Effect

Specified Initial Conditions

A experimental experimental linear ignored yes

B adjusted ignored linear ignored yes

C adjusted ignored linear included no

D experimental empirical linear included no

E empirical empirical linear included no

F empirical empirical linear ignored yes

G experimental experimental linear ignored yes



Table 2.7 Comparison of numerical prediction methods for Ship A-2.

Organisations DOF Radiation Manoeuvring

Damping Roll

Damping Roll

Restoring Froude- Krylov

Diffraction

A 4 static heave & pitch

slender body

theory at ω=0

experimental (linear)

exp.+ empirical forward

speed effect

hydrostatics in calm water

linear slender

body theory at ω=0

B 6

3D theory （Green

function at Fr=0）

ignored empirical +

tuned hydrostatics

in waves non- linear

3D theory（Green

function at Fr=0）

C 4 static heave & pitch

strip theoryexperimental (non-linear)

exp.+ empirical forward

speed effect

hydrostatics in calm water

non- linear

slender body theory

at ω=0

D 6 two stage model

strip theoryexperimental

(linear) empirical

hydrostatics in waves

non- linear

strip theory

Organisations

Hydro- dynamic

Lift due to Waves

Hull Resistance

Propeller Thrust

Rudder Force

Incident Wave

Hydro- dynamic Memory

Effect

Specified Initial Con-

dition

A end term exp. exp. exp. linear ignored yes

B ignored empirical tuned empirical non-linear included no

C ignored exp. exp. exp. linear ignored yes

D ignored exp. exp. empirical linear included no

To improve quantitative prediction accu-racy, it is essential that the contribution of several factors should be further investigated through comparative studies of capsizing pre-dictions with and without these factors being accounted for. These should include wave ef-fect on manoeuvring coefficients and roll re-storing moment, roll damping prediction, ra-diation and diffraction and resistance at higher speeds with use of captive model experi-ments. It is noteworthy that stability predic-tion should be based on accurate predictions of seakeeping, manoeuvring, resistance and propulsion, subjects which have been the fo-cus of considerable effort by the relevant ITTC technical committees over many years. This alone offers ample justification of the problems encountered in attempting to accu-

rately predict ship capsize. For wider valida-tion studies, it is desirable to execute bench-mark tests concerning the prediction of cap-sizing boundary curves as shown by Munif (2000) for Ship A-1 and by Umeda & Hashi-moto (2002) for Ship A-2. This is because capsizing boundaries can be complicated as a result of system nonlinearities. Furthermore, practical application to ship design, operation and regulation necessitates the extension of the current predictive capability to more real-istic seaways.

This, in turn, warrants the undertaking of a new benchmark testing of numerical codes with relevant experimental data from the most advanced model basins having multi-component wave makers.



0 10 20 30

-10

-5

0

5

10

t(s)

P itch(degrees)

0 10 20 30

-90-60-30

0306090

t(s)

Roll(degrees)

0 10 20 30

-45-30-15

0153045

t(s)

Yaw(degrees)

0 10 20 30

-15-10

-505

1015

t(s)

Rudder(degrees)

0 10 20 30

-10

-5

0

5

10

t(s)

P itch(degrees)

Calculation

0 10 20 30

-90-60-30

0306090

t(s)

Experiment

Roll(degrees)

0 10 20 30

-45-30-15

0153045

t(s)

Yaw(degrees)

0 10 20 30

-15-10

-505

1015

t(s)

Rudder(degrees)

0 20 40 60

-10

-5

0

5

10

t(s)

P itch(degrees)

0 20 40 60

-90-60-30

0306090

t(s)

Roll(degrees)

0 20 40 60

-60-40-20

0204060

t(s)

Yaw(degrees)

0 20 40 60

-15-10

-505

1015

t(s)

Rudder(degrees)

0 20 40 60

-10

-5

0

5

10

t(s)

P itch(degrees)

Calculation

0 20 40 60

-90-60-30

0306090

t(s)

Experiment

Roll(degrees)

0 20 40 60

-60-40-20

0204060

t(s)

Yaw(degrees)

0 20 40 60

-15-10

-505

1015

t(s)

Rudder(degrees)

(a)

(b)

0 10 20 30 40

-10

-5

0

5

10

t(s)

P itch(degrees)

0 10 20 30 40

-90-60-30

0306090

t(s)

Roll(degrees)

0 10 20 30 40

-60-40-20

0204060

t(s)

Yaw(degrees)

0 10 20 30 40

-15-10

-505

1015

t(s)

Rudder(degrees)

0 10 20 30 40

-10

-5

0

5

10

t(s)

P itch(degrees)

Calculation

0 10 20 30 40

-90-60-30

0306090

t(s)

Experiment

Roll(degrees)

0 10 20 30 40

-60-40-20

0204060

t(s)

Yaw(degrees)

0 10 20 30 40

-15-10

-505

1015

t(s)

Rudder(degrees)

Organisation-A Organisation-E

0 5 10 15 20

-10

-5

0

5

10

t(s)

P itch(degrees)

0 5 10 15 20

-90-60-30

0306090

t(s)

Roll(degrees)

0 5 10 15 20

-60-40-20

0204060

t(s)

Yaw(degrees)

0 5 10 15 20

-15-10

-505

1015

t(s)

Rudder(degrees)

0 5 10 15 20

-10

-5

0

5

10

t(s)

P itch(degrees)

Calculation

0 5 10 15 20

-90-60-30

0306090

t(s)

Experiment

Roll(degrees)

0 5 10 15 20

-60-40-20

0204060

t(s)

Yaw(degrees)

0 5 10 15 20

-15-10

-505

1015

t(s)

Rudder(degrees)

Organisation-G

(c)

(d)

Figure 2.4 Experimental results and numerical results for Ship A-1 from three organisations.



0 10 20 30

-10

-5

0

5

10

t(s)

P itch(degrees)

0 10 20 30

-90-60-30

0306090

t(s)

Roll(degrees)

0 10 20 30

-45-30-15

0153045

t(s)

Yaw(degrees)

0 10 20 30

-15-10

-505

1015

t(s)

Rudder(degrees)

0 10 20 30

-10

-5

0

5

10

t(s)

P itch(degrees)

Calculation

0 10 20 30

-90-60-30

0306090

t(s)

Experiment

Roll(degrees)

0 10 20 30

-45-30-15

0153045

t(s)

Yaw(degrees)

0 10 20 30

-15-10

-505

1015

t(s)

Rudder(degrees)

0 20 40 60

-10

-5

0

5

10

t(s)

P itch(degrees)

0 20 40 60

-90-60-30

0306090

t(s)

Roll(degrees)

0 20 40 60

-90-60-30

0306090

t(s)

Yaw(degrees)

0 20 40 60

-15-10

-505

1015

t(s)

Rudder(degrees)

0 20 40 60

-10

-5

0

5

10

t(s)

P itch(degrees)

Calculation

0 20 40 60

-90-60-30

0306090

t(s)

Experiment

Roll(degrees)

0 20 40 60

-90-60-30

0306090

t(s)

Yaw(degrees)

0 20 40 60

-15-10

-505

1015

t(s)

Rudder(degrees)

(a)

(b)

0 10 20 30 40

-10

-5

0

5

10

t(s)

P itch(degrees)

0 10 20 30 40

-90-60-30

0306090

t(s)

Roll(degrees)

0 10 20 30 40

-60-40-20

0204060

t(s)

Yaw(degrees)

0 10 20 30 40

-15-10

-505

1015

t(s)

Rudder(degrees)

0 10 20 30 40

-10

-5

0

5

10

t(s)

P itch(degrees)

Calculation

0 10 20 30 40

-90-60-30

0306090

t(s)

Experiment

Roll(degrees)

0 10 20 30 40

-60-40-20

0204060

t(s)

Yaw(degrees)

0 10 20 30 40

-15-10

-505

1015

t(s)

Rudder(degrees)

Organisation-B Organisation-C

0 5 10 15 20

-10

-5

0

5

10

t(s)

P itch(degrees)

0 5 10 15 20

-90-60-30

0306090

t(s)

Roll(degrees)

0 5 10 15 20

-60-40-20

0204060

t(s)

Yaw(degrees)

0 5 10 15 20

-15-10

-505

1015

t(s)

Rudder(degrees)

0 5 10 15 20

-10

-5

0

5

10

t(s)

P itch(degrees)

Calculation

0 5 10 15 20

-90-60-30

0306090

t(s)

Experiment

Roll(degrees)

0 5 10 15 20

-60-40-20

0204060

t(s)

Yaw(degrees)

0 5 10 15 20

-15-10

-505

1015

t(s)

Rudder(degrees)

Organisation-F Organisation-D

(c)

(d)

Figure 2.5 Experimental results and numerical results for Ship A-1 from four organisations.



0 5 10 15 20

-20

-10

0

10

20

t(s)

P itch(degrees)

0 5 10 15 20

-90-60-30

0306090

t(s)

Roll(degrees)

0 5 10 15 20

-60-40-20

0204060

t(s)

Yaw(degrees)

0 5 10 15 20

-45-30-15

0153045

t(s)

Rudder(degrees)

0 5 10 15 20

-20

-10

0

10

20

t(s)

P itch(degrees)

Calculation

0 5 10 15 20

-90-60-30

0306090

t(s)

Experiment

Roll(degrees)

0 5 10 15 20

-60-40-20

0204060

t(s)

Yaw(degrees)

0 5 10 15 20

-45-30-15

0153045

t(s)

Rudder(degrees)

0 5 10 15 20

-20

-10

0

10

20

t(s)

P itch(degrees)

0 5 10 15 20

-90-60-30

0306090

t(s)

Roll(degrees)

0 5 10 15 20

-60-40-20

0204060

t(s)

Yaw(degrees)

0 5 10 15 20

-45-30-15

0153045

t(s)

Rudder(degrees)

0 5 10 15 20

-20

-10

0

10

20

t(s)

P itch(degrees)

Calculation

0 5 10 15 20

-90-60-30

0306090

t(s)

Experiment

Roll(degrees)

0 5 10 15 20

-60-40-20

0204060

t(s)

Yaw(degrees)

0 5 10 15 20

-45-30-15

0153045

t(s)

Rudder(degrees)

(a)

(b)

0 5 10 15 20

-20

-10

0

10

20

t(s)

P itch(degrees)

0 5 10 15 20

-90-60-30

0306090

t(s)

Roll(degrees)

0 5 10 15 20

-90-60-30

0306090

t(s)

Yaw(degrees)

0 5 10 15 20

-45-30-15

0153045

t(s)

Rudder(degrees)

0 5 10 15 20

-20

-10

0

10

20

t(s)

P itch(degrees)

Calculation

0 5 10 15 20

-90-60-30

0306090

t(s)

Experiment

Roll(degrees)

0 5 10 15 20

-90-60-30

0306090

t(s)

Yaw(degrees)

0 5 10 15 20

-45-30-15

0153045

t(s)

Rudder(degrees)

Organisation-A Organisation-B

0 5 10 15 20

-20

-10

0

10

20

t(s)

P itch(degrees)

0 5 10 15 20

-90-60-30

0306090

t(s)

Roll(degrees)

0 5 10 15 20

-60-40-20

0204060

t(s)

Yaw(degrees)

0 5 10 15 20

-45-30-15

0153045

t(s)

Rudder(degrees)

0 5 10 15 20

-20

-10

0

10

20

t(s)

P itch(degrees)

Calculation

0 5 10 15 20

-90-60-30

0306090

t(s)

Experiment

Roll(degrees)

0 5 10 15 20

-60-40-20

0204060

t(s)

Yaw(degrees)

0 5 10 15 20

-45-30-15

0153045

t(s)

Rudder(degrees)

Organisation-C Organisation-D

(c)

(d)

Figure 2.6 Experimental results and numerical results for Ship A-2 from four organisations.



3. BENCHMARK TESTING FOR DAMAGE SHIP STABILITY

3.1. Introduction

This chapter highlights the results of the benchmark testing on damage stability for Ship B-2 of the 22nd ITTC (ITTC, 1999). The benchmark test programme commenced in March 2000, inviting ITTC member organisa-tions and other qualified research institutions to express their interest in participating at the launched study. Based on this call, five or-ganizations submitted numerical results.

The selected ship for this investigation is a Ro-Ro Passenger vessel, which has been model tested in the Denny Tank (the test tank of the Ship Stability Research Centre) accord-ing to the Model Test Method of IMO SOLAS ’95 Resolution 14 (Vassalos and Jasionowski, 2000). The condition addressed in the study concerns the midship damage as described in SOLAS (1997). Wave conditions were set in the benchmark guidelines (Umeda & Papanikolaou, 2000) as well as the range of tests in regular and irregular seas, as shown in Table 3.1. In particular it was requested to in-vestigate the intact ship performance for a

number of incoming regular beam waves and the damage ship capsizing boundaries in ir-regular beam seas for varying KG values.

Lack of uniformity in the interpretation of the initial guidelines and specifications (Pa-panikolaou & Spanos, 2001) led to specifica-tions being revised (Vassalos, Umeda and Pa-panikolaou, 2001) and resubmission of results (Papanikolaou, 2001; Jasionowski, 2001).

Details of the experimental results and ship characteristics are reported by Jasionowski & Vassalos (2001).

Despite the fact that finally only a small number of organisations were able to partici-pate in the benchmark study and that only one ship could be benchmark tested during, it is felt that the final outcome enabled the draw-ing of some important conclusions. Even if some numerical results cannot be considered satisfactory, compared to model experiment results, the study clearly identified advances and gaps in the present state of knowledge on the prediction of extreme ship motions and capsizing of damaged ships leading to rec-ommendations on further studies concerning specifically identified problem areas.

Table 3.1 Overview of benchmark test results.

Participants × results available; − results not available

P1 P2 P3 P4 P5

GZ curves (intact/damaged) × × × × ×

Simulated free roll decay curves (intact) × × × × ×

Simulated free roll decay curves (damaged) − × × × ×

Simulated frequency roll response curves (intact/damaged RAOs), constant wave height

× × × × ×

Simulated roll response curves (intact), constant wave slope (1:25)

× × − × ×

Simulated survivability boundaries (0% capsize, 100% capsize for KG=12.892 m)

× × × × ×

Wave, roll, heave, water on deck time series for KG=12.892 m and Hs=4.0 m, 4.25 m and 4.5 m (5 runs)

× × × × ×



List of Participants. The following five or-ganisations participated in the ITTC ship B-2 damage stability benchmark study:

Flensburger Schiffbau Gesellschaft, FSG (Ms. Heike Cramer)

Maritime Research Institute Netherlands (Dr. Jan O. de Kat)

National Technical University of Athens (Prof. Apostolos Papanikolaou)

Osaka University (Prof. Naoya Umeda) Universities of Glasgow and Strath-

clyde, SSRC (Prof. Dracos Vassalos).

3.2. Software employed

All the software employed is non-linear time domain codes. Six degrees of freedom (DOF) models are used by all participants, except for Participant 3 where a three DOF model is used (sway, heave and roll).

Table 3.2 Potential theory approaches to ship motions employed by participants.

Participant Approach

P1 Strip theory, 6 DOF

P2 3D source panel theory, 6 DOF

P3 Newly modified strip theory, 3 DOF

P4 3D source panel theory, 6 DOF

P5 Strip theory, 6 DOF

Table 3.3 Modelling of damping forces.


P1 Non-linear roll damping according to P. Blume

P2 Equivalent linear roll damping estimated from the available intact ship roll decay measurements

P3 Non-linear roll damping according to Ikeda

P4 Adaptive-linear roll damping according to Ikeda

P5 Non-linear (equivalent linear & quadratic) roll damping based on a modified Ikeda’s approach

Radiation and wave diffraction forces are calculated by a number of approaches, all in the framework of potential flow theory, as in-dicated in Table 3.2.

Additional information for the calculation of damping forces is shown in Table 3.3.

The floodwater is generally considered as independent variable masses moving inside the flooded compartments and interacting with the ship. Modelling of floodwater motion can be generally described by the modelling of the floodwater free surface condition, as shown in Table 3.4.

Table 3.4 Floodwater free-surface model-ling.


P1 Plane and free movable (when period away from natural); Glimm’s equations (when period close to natural)

P2 Plane and free movable

P3 Plane and horizontal

P4 Plane and free movable

P5 Plane and horizontal

Water ingress/egress through the damage opening is commonly based on hydraulic models following application of Bernoulli’s dynamic pressure head equation. All partici-pants used empirically determined coefficients to take into account the actual water in-gress/egress flow through the specified dam-age opening.

3.3. The Test Ship

The general particulars of the test ship in full and model scale are given in Table 3.5. The model scale is 1:40.



Table 3.5 Main ship particulars

Full Scale

Model Scale

LOA 179.00 m 4,475.0 mm

LBP 170.00 m 4,250 mm

B 27.80 m 695.0 mm

T 6.25 m 156.3 mm

DCARDECK 9.00 m 225.0 mm

Displacement (even keel)

17,300 tonnes

270.3 kg

Intact KG (above BL)

12.89 m 322.0 mm

Intact Design GM

2.63 m 65.8 mm

The ship was studied in intact and damage condition. Some details of the model charac-teristics pertaining to these conditions are given next.

Metacentric height (intact): GM = 65.76 mm. This is determined by inclining ex-periment.

Roll radius of gyration, ixx/B= 0.235 (ixx= 163 mm). The roll radius of gyra-tion ixx was estimated by analysis of the free roll decay measurements for the in-tact condition and given to the partici-pants. This radius refers to the inertia of ship structural mass (derived from the relevant decay measurements by ac-counting for the hydrodynamic added inertia) and is a characteristic constant of the model, for the specific loading condition.

Intact natural roll period: Tni = 2.056 seconds. This period was determined by analysis of records of free rolling tests in intact condition.

Damaged natural roll period: Tnd = 2.300 seconds. This period was determined by

analysis of records of free rolling tests in damaged condition.

Pitch radius of gyration: iyy/LPP = 0.217 (iyy = 872 mm). Radius iyy was estimated by analysis of free pitching experiments in air (Vassalos and Jasionowski, 2000).

Yaw Radius of Gyration: izz/LPP = 0.238 (iyy = 960 mm). Radius izz was assumed to be 10% greater than pitch radius of gyration. This increase can be justified by the fact that for models in a damaged condition the mass of superstructures is normally absent and hence the lateral mass distribution of the model is ex-pected to be greater than the vertical one.

Figure 3.1 depicts the damaged case con-sidered in the benchmark study with the GZ curves in intact and damaged cases given in Figures 3.2 and 3.3, respectively.

3.4. GZ curves

The accuracy of hydrostatic calculations of the numerical codes used in the benchmark depends on the discretisation of the ship’s ge-ometry. This explains the differences ob-served in Figures 3.2 and 3.3, which should be borne in mind when analysing the predicted ship responses by the participating organisa-tions, as any inaccuracy in the geometry and ship hydrostatics will affect the estimated stiffness (restoring ability) of the ship and hence her natural frequencies. In the intact ship case the observed differences are minor. However, in the damaged case, the GZ curve by Participant 3 shows higher initial stiffness for the flooded ship, whereas the range of sta-bility computed by Participants 2 and 5 is no-ticeably lower. Only Participants 1 and 4 properly capture the hydrostatic properties of the benchmark ship over the entire stability range.



Figure 3.1 Midship Damage Case (Vassalos & Jasionowski, 2000).

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1.1

0 10 20 30 40 50 60

Heel angle [deg]

Rig

htin

g le

ver

[m]

Participant 1

Participant 2

Participant 3

Participant 4

Participant 5

Figure 3.2 Computed GZ curves for the in-tact ship.

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4

0 5 10 15 20

Heel angle [deg]

Rig

htin

g le

ver

[m]

Participant 1Participant 2Participant 3Participant 4Participant 5

Figure 3.3 Computed GZ curves for the damaged ship.



3.5. Free Rolling Simulations

The results of free rolling simulations are shown in Figure 3.4. Simulation of free roll decay in the intact condition presents no diffi-culty, as generally good agreement with the experiments was achieved by all participants. However, similar attempts to simulate the free roll response in the damaged condition were less successful. Results presented by all par-ticipants show a distinctive overestimation of the natural roll frequency. In this respect, three possible sources for this discrepancy may be cited:

(a) Lack of understanding of the complete hydrodynamics of the damaged ship

(b) Inaccurate representation of the floodwa-ter dynamics and its coupling with ship motion

(c) Possible inconsistencies in the available experimental data (clarification of ex-perimental conditions and way of analysis of data).

As a first step towards improving this situation, it would seem necessary to under-take a new benchmark study in the future for at least another ship case and to perform addi-tional experimental verifications of free roll tests in damaged conditions.

PRR1, KG=12.892m, Free roll decay, Intact condition

-8

-6

-4

-2

0

2

4

6

8

0 20 40 60 80 100 120

Time [s]

Rol

l [de

g]

EXPERIMENT

Participant 1

Not available


-8

-6

-4

-2

0

2

4

6

8

0 20 40 60 80 100 120

Time [s]

Rol

l [de

g]

EXPERIMENT

Participant 2

PRR1, KG=12.892m, Free roll decay, Damage condition

-10

-8

-6

-4

-2

0

2

4

6

0 20 40 60 80 100 120

Time [s]

Rol

l [de

g]

EXPERIMENT

Participant 2




-8

-6

-4

-2

0

2

4

6

8

0 20 40 60 80 100 120

Time [s]

Rol

l [de

g]

EXPERIMENT

Participant 3


-8

-6

-4

-2

0

2

4

6

0 20 40 60 80 100 120

Time [s]

Rol

l [de

g]

EXPERIMENT

Participant 3


-8

-6

-4

-2

0

2

4

6

8

0 20 40 60 80 100 120

Time [s]

Rol

l [de

g]

EXPERIMENT

Participant 4


-8

-6

-4

-2

0

2

4

6

0 20 40 60 80 100 120

Time [s]

Rol

l [de

g]

EXPERIMENT

Participant 4


-8

-6

-4

-2

0

2

4

6

8

0 20 40 60 80 100 120

Time [s]

Rol

l [de

g]

EXPERIMENT

Participant 5


-10

-8

-6

-4

-2

0

2

4

6

0 20 40 60 80 100 120

Time [s]

Rol

l [de

g]

EXPERIMENT

Participant 5

Figure 3.4 Free roll decay for the intact and damaged ship (measured and simulated).



3.6. Ship Performance in Regular Waves

The results of the benchmark study for the intact and damaged ship roll response in regu-lar beam waves, with

constant wave height (Hw = 1.2 m, 2.0 m and 2.4 m) and

constant wave slope, namely constant wave height to wavelength ratio equal to 1/25,

are presented in this section. The purpose of this particular study was to gain insight into the modelling of roll motion by the partici-pants in normal operational and extreme con-ditions (significant wave height up to 20 m).

Roll Response Amplitude Operators (RAOs) for Constant Wave Height Excitation. Based on the results presented in Figure 3.5, the following can be concluded4:

Simulation of intact ship roll RAO for constant wave height monochromatic wave excitation: All participants have generally accomplished the simulation of the basic intact ship roll RAO suc-cessfully. Some differences in the predicted peak values of ship response, occurring at the natural roll frequency (more pronounced in the simulation by Participant 1), are due to the differences in the modelling of roll damping, which is based on semi-empirical coefficients and approaches that need further improvement.

Simulation of damaged ship roll RAO for constant wave height monochro-matic wave excitation: The simulation of the basic damaged ship roll RAO could not be predicted satisfactorily by the participants. The reasons given ear-lier concerning free rolling simulation apply equally here and become even

4 Note that experiments were performed in

wave heights equal to 1.2 m and 2.4 m, whereas numerical simulations were under-taken in the range between 1.2 and 2.4 m.

more apparent in the simulation of the damaged ship roll RAO.

In particular, none of the participants ob-tained the natural frequency of the damaged ship close to the value derived experimentally. In fact, the predicted natural frequency of the damaged ship is quite inconsistent among the participants: Participants 2, 4 and 5 predict a slight decrease of this frequency in relation to the natural frequency of the intact ship, a trend shown also in the damaged model experiments, but not to the extent measured in these experiments. Participant 1 is even pre-dicting an increase of the intact natural fre-quency, whereas the predicted natural fre-quency by Participant 3 remains practically unchanged.

The above results suggest that the pre-dicted hydrodynamic added moment of inertia by all participants deviates significantly from the experimental value.

Regarding the predicted peak values of the damaged ship roll response, practically all participants predict higher roll amplitudes in-dicating a clear underestimate of roll damp-ing.

The experimentally measured damaged ship roll response indicates the existence of a second resonance frequency at approximately twice the main roll resonance frequency. This phenomenon is predicted by Participants 4 and 5 but at much higher frequencies and for higher secondary resonance peak values.

The above conclusions call for additional research in this area and a reassessment of the damaged ship roll RAO results in the future, when more experimental and numerical benchmark results become available.



PRR1, KG=12.892m, Roll RAO, Intact condition

0

2

4

6

8

10

12

14

16

18

0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2

ω [rad/s]

Rol

l [de

g/m

]

Experiment, (Hw=1.2m)


Participant 1 (Hw=2.0m)

PRR1, KG=12.892m, Roll RAO, Damage condition

0

2

4

6

8

10

12

14

16

18

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2

ω [rad/s]

Rol

l [de

g/m

] g



Participnt 1, (Hw=2.0m)


0

2

4

6

8

10

12

14

16

18

0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2

ω [rad/s]

Rol

l [de

g/m

]



Participant 2, (Hw=2.4m)


0

2

4

6

8

10

12

14

16

18

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2

ω [rad/s]

Rol

l [de

g/m

] g





0

2

4

6

8

10

12

14

16

18

0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2

ω [rad/s]

Rol

l [de

g/m

]





0

3

6

9

12

15

18

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2

ω [rad/s]

Rol

l [de

g/m

] g







0

2

4

6

8

10

12

14

16

18

0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2

ω [rad/s]

Rol

l [de

g/m

]





0

2

4

6

8

10

12

14

16

18

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2

ω [rad/s]

Rol

l [de

g/m

] g





0

2

4

6

8

10

12

14

16

18

0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2

ω [rad/s]

Rol

l [de

g/m

]





0

2

4

6

8

10

12

14

16

18

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2

ω [rad/s]

Rol

l [de

g/m

] g




Figure 3.5 Free roll decay for the intact and damaged ship (measured and simulated).

Roll Response Amplitude Operators (RAOs) for Constant Wave Slope Excitation. This study was restricted to only the intact ship

case and no experimental data were available to cross check the numerical predictions. The results are presented in Figures 3.6 and 3.7.

PRR1, KG=12.892m, Roll RAO, Constant wave slope 1:25, Intact condition

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

0 0.5 1 1.5 2 2.5 3λ/L [-]

Rol

l Am

pl /

kA [-

]

Participant 1

Participant 2

Participant 4

Participant 5

Figure 3.6 Roll response in regular beam waves with constant wave slope over λ/L.

PRR1, KG=12.892m, Roll RAO, Constant wave slope 1:25, Intact condition

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

0 0.5 1 1.5 2 2.5 3λ/L [-]

Rol

l Am

pl /

kA [-

]

Participant 1

Participant 2

Participant 4

Participant 5

Figure 3.7 Roll response in regular beam waves with constant wave slope over ω.



Based on the derived results, the following can be stated:

Overall, the results produced by the par-ticipating organisations differ significantly. Results for large wavelength to ship length ratios (λ/L) appear satisfactory but this is not the case for small λ/L ratios. Only results by Participants 2 and 4 appear to agree closely throughout the frequency range considered in the study.

Predictions in the resonance region deviate substantially indicating clearly the differences of the semi-empirical damping models used by the various participants for the extreme motion amplitude predictions.

In conclusion, it appears necessary that a more comprehensive study should be carried out in the future to investigate the relationship between the damping models used by the benchmark study participants. Unfortunately, experimental measurements were not avail-able to enable a more thorough evaluation of the employed numerical procedures for the intact, large amplitude and large slope motion studies. The only apparent result from this comparison is that numerical modelling of

highly nonlinear ship motion problems is not yet satisfactory.

3.7. Ship Performance in Irregular Waves

The prediction of the damaged ship per-formance is assessed on the basis of analysis of the numerically simulated time series for the exciting wave, the ship motion response and the amount of floodwater, in comparison with the corresponding time series of model experiments, simulation of the damaged ship survival boundaries and finally identification of critical wave heights. Detailed results of this study (experimental and numerically simulated time series) are given by Jasionowski & Vassalos (2001). In this sec-tion only sample results are presented (Figures 3.8 to 3.19) and discussed.

Two representative runs per participant, one for survival and one for capsizal cases, are presented, all corresponding to a signifi-cant wave height excitation of 4.0 m. Wave elevation, heave and roll motions as well as the amount of water accumulated on the car deck are shown.

Figure 3.8 Experimental measurements – damaged ship model (survival case).



Figure 3.9 Experimental measurements – damaged ship model (capsizal case).

Figure 3.10 Numerical simulations by Participant 1 – damaged ship model (survival case).

Figure 3.11 Numerical simulations by Participant 1 – damaged ship model (capsizal case).



Figure 3.12 simulations by Participant 2 – damaged ship model (survival case).














A visual comparison between the numeri-cally predicted and experimentally measured time series shows a rather unsatisfactory level of agreement. Indeed, none of the numerical time series matches qualitatively the experi-mental values and only some of the numerical results agree qualitatively (Participant 1 and to some extent Participants 4 and 5). The roll response predicted by Participants 2 and 3 displays again noticeably higher amplitudes, possibly due to inaccurate roll damping mod-els.

A Fourier spectral analysis of the calcu-lated time series records enables a better un-derstanding of the differences between the numerical simulations and the response char-acteristics derived by physical model tests (Figure 3.20) as outlined next:

Participant 2 failed to reproduce exactly the exciting wave spectrum characteris-tics, and hence used a wave spectrum with its peak slightly shifted towards

lower frequencies, closer to natural roll frequency of the ship.

The predicted roll response spectra by Participants 1, 2, 4 and 5 indicate under-estimation of roll damping, with reso-nant roll amplitudes significantly higher than the experimental values.

Participants 2 and 3 predict considerably higher roll spectral densities, partly as a consequence of the predicted higher peak roll values and partly due to the shift of the peak frequency of the excit-ing wave spectrum.

A noticeable peak appears in the roll motion spectrum by Participant 5 at a frequency of about 1.0 rad/s unlike the experimental results.

The spectrum of heave response derived by Participant 3 shows a peculiar second lower peak around 0.6 rad/s.



Figure 3.20 Spectral analysis of experimental and numerical ship responses.



3.8. Survivability Boundaries

The reported survivability boundaries show consistently high accuracy by most the participants. Both, the “lower” and “upper” boundaries distinguishing between sea states leading to ship survival, marginal survivabil-ity or capsize were predicted with a spread of approximately ±0.5 m in Hs. Only Participant 3 fails to predict the capsizal boundary (no capsize predictions for the specified condi-tions). It must be mentioned here, that the simulation time is of importance in establish-ing the boundary consistently, as the longer the duration of the simulation, the lower the boundary tends to be. This, however, was clearly defined as a benchmark test constraint and some variation between the participants is noticeable. The demonstrated accuracy in pre-dicting the critical sea states seems quite satis-factory from the point of view of application of such information to practical survivability assessment procedures. An overview of rele-vant results is shown In Table 3.6 and de-picted in Figure 3.21. In compiling this table the ship survival boundary, for the particular damage case and sea state tested, is identified on the basis of zero capsize events occurring for five consecutive numerical simulations using different irregular wave realisations in each case. On the other hand a capsizal boundary is identified on the basis of all five runs leading to a capsize event.

Table 3.6 Experimentally observed and numerically simulated capsize events.

Experiment P1 P2 P3 P4 P5

Hs=3.50m - - - - - 0

Hs=3.75m - - 0 - 0 1

Hs=4.00m 0 0 2 0 1 3

Hs=4.25m 3 0 3 0 2 4

Hs=4.50m 5 2 5 1 3 5

Hs=4.75m - 5 - 1 3 -

Hs=5.00m - - - 2 5 -

SURVIVE/CAPSIZE BOUNDARIES

0

1

2

3

4

5

6

Hs

[m]

Exp

P1

P2

P3P4

P5

Figure 3.21 Comparison between predicted and measured survival/capsizal boundaries.

3.9. Concluding remarks

Considering the relatively low number of the benchmark study participants and the com-plicate nature of the benchmark study problem being addressed it is felt that the main objec-tives of the present study have been met, though a similar study should be repeated in the future with the aim to alleviate the effects of some of the identified weaknesses. It has been ascertained that at the present state of knowledge, model experiments remain indis-pensable for assessing the survivability of damaged ships in waves, though theoretical-numerical prediction methods can greatly con-tribute to the assessment of the survivability of damaged ships in waves.

4. GUIDELINES FOR MODEL TESTING OF INTACT AND DAMAGE STABILTY

The purpose of these guidelines is to pro-vide ITTC member organisations, intending to undertake intact and damage stability model tests in waves, with a sound basis for carrying out these tests. The derived guidelines are based on those presented to the 22nd ITTC, upgraded to reflect member organisation ex-perience. The full recommended guidelines for model test on intact stability are included in the ITTC Quality Manual as Procedure 7.5-02-07-02.5 and on damage stability as Proce-dure 7.5-02-07-02.6.



5. QUESTIONNAIRE

5.1. Questionnaire on Model Experiments

The questionnaire was formatted with the intention of validating the ITTC guidelines for Intact and Damaged model experiments, pre-sented at the 1999 ITTC.

The following organisations replied to the questionnaire:

Institute for Marine Dynamics, National Research Council of Canada

MARIN Naval Ship Research and Development

Centre Osaka University, Department of Naval

Architecture & Ocean Engineering QinetiQ The Universities of Glasgow and Strath-

clyde, SSRC

A summary of the questionnaire is given in Table 5.1. The full questionnaire, including details of how to fill it out and submit it, can be found on the web site of the committee: www.ssrc.na-me.ac.uk/ittc/scexcap/

Table 5.1 Summary of questionnaire on model test procedures.

1. Past experience 2. Model design, construction and outfit

Construction material Uppermost limit of accurate modelling Factors considered in scaling flooded com-partments, deck, etc. Scaling flow through pipes and openings Modelling of flooded spaces Model scale Instrumentation and equipment Model Power supply systems Data recording Ballasting Control of models

3. Experiments Initial conditions Start of data acquisition Distance from wave maker Wave types used Model speed measurement

4. Data analysis

Four of the organisations described ex-periments on intact and damaged ship models that fitted within the operating practice de-scribed by the 1999 ITTC Guidelines. Two of these organisations had experience with ex-treme motions, but not capsize.

For experiments in oblique waves with forward speed, three organisations used bat-tery based power systems onboard the model for intact extreme motions or capsize experi-ments and data was handled in one of the fol-lowing ways:

Collected and stored on the model Transferred to storage unit by telemetry Transferred to storage unit by cable.

For capsize/extreme motions, all three or-ganisations used a free running model, pow-ered by a model propeller, with autopilot to keep the model on course, once the experi-ment had started.

For the damaged case, all four organisa-tions supplied power to the model and trans-ferred the data to storage via an umbilical ca-ble. The models were allowed to drift under the action of waves. Scaling of openings through which water would flow was based on geometry. No allowance was made for scale effects. Some notable exceptions to the guidelines were:

Damaged stability model experiments at one organisation were carried out with length and scale requirements outside the guidelines.

For one organisation, damaged stability experiments were on simplified hull forms and permeability of under deck spaces was 100%.

Two organisations conducted damage stability tests using a much more sophis-ticated procedure than that recom-mended for the typical Ro-Ro ferry case, and this was to study damage stability of warships. For these tests, the experiment starts with an intact, self- propelled model. The preparation and procedures for this part of the experiment fit within those described in the ITTC guidelines



for intact model experiments. At some point in the experiment, the damage is simulated and the model floods. At this point the requirement is for the model to behave as a damaged ship and flooding is monitored with water depth sensors and video cameras. The internal struc-ture of the model is much more complex than a Ro-Ro model, due to the more complex internal structure of the ship. Also, the requirement to have the model self-propelled at the start of the experi-ments brings in extra challenges for model construction. Model scale for these tests was 1:24 giving model lengths between 5 and 6 metres. The preferred approach is to use a com-pletely free running model, with teleme-try to transfer data from the model to a shore station.

For intact stability tests most organisations use models ranging in length from 3 to 5 m, although there are cases of using models un-der 2 m.

In conclusion, the analysis of the small sample of questionnaires suggests that there are no major flaws in the ITTC guidelines for model tests as proposed in 1999. The commit-tee was lucky in that much of the work done by researchers in this area is well documented in the literature and the performers have accepted the academic tradition of exchanging informa-tion freely on techniques and processes.

Based on the results of the questionnaire, each organization has slightly different proce-dures, but all appear to be well founded on good experiment practice. The challenge for the future is to ensure that the guidelines ap-ply to a wider range of ships than Ro-Ro fer-ries, and should also consider ships where portions of the ship are flooded in normal op-erations (cruise ships, landing platform docks, ships with moon pools, etc).

5.2. CFD Survey

The following organisations replied to the questionnaire:

MARIN Memorial University of Newfoundland,

Department of Naval Architecture and Ocean Engineering

National Technical University of Ath-ens, Department of Naval Architecture and Marine Engineering

W. S. Atkins Consultants Ltd. Universities of Glasgow and Strath-

clyde, Department of Naval Architecture and Marine Engineering.

The survey is summarised in Table 5.2.

Table 5.2 Summary of survey on the use of CFD.

1. What was the original motivation for you in-corporating CFD into ship motion predictionsoftware?

2. Give a brief description of the ship motionprediction code used

3. Briefly describe the CFD code used 4. Give a brief description of how you inte-

grated the CFD code with the ship motionscode

5. Flow types modelled by CFD component 6. Linking Issues between ship motions code

and CFD code 7. Post Processing 8. State of Development of Codes 9. Validation 10. Special Limitations 11. Future Developments 12. References

The survey focused on combining codes to predict ship motions in waves with CFD, to model the combined motion of ship and fluid trapped on the deck or within a ship’s hull. It should also be recognised that CFD has been used for other areas relevant to extreme mo-tions including capsizing, such as modelling steep breaking waves, and also for seakeep-ing. Using CFD presents a means of overcom-ing the classical assumptions of linear theo-ries, combining forward speed with motion in



waves and non-viscous flow. Beck & Reed (2001) give a good overview of the develop-ment of numerical seakeeping predictions from their earliest beginnings to the latest un-steady RANS approaches. They also point out the huge computational requirements for us-ing CFD based seakeeping predictions, and the level of uncertainty in the results. As such, CFD methods for full ship motion predictions are a long way from practical application in engineering situations.

CFD can also be applied to the reduced problem of predicting viscous roll damping (Salui et al., 2000). This has always been a severe limitation for potential flow methods, which cannot realistically predict roll motion. Most work with CFD in this area has focused on forced rolling in calm water, to obtain pre-dictions of the added mass and damping coef-ficients directly from the numerical procedure, hence eliminating the empirical assumptions.

Three of the replies are related to the clas-sical assumptions of CFD methods, and one was a simplified approach. For the three ‘clas-sical’ CFD approaches, the following obser-vations were made.

Three methods use a 6DOF time domain code for predicting ship motions, linked to a CFD code to predict the motion of water on the deck of the ship. Two ship motion codes have some linear and non-linear elements in the hydrodynamic coefficients and restoring moments. The third method uses a linear response time domain method as the basis for its motion code.

All three CFD codes used variations of the ‘Volume of Fluid’ technique, where the grid is fixed (body fitted) and the free surface is tracked either by grid elements that are partially full or by special free surface cells. In one case studying a flooded Ro-Ro, the method included fluid entering and leaving the deck area, based on the relative motion given by the ‘seakeeping’ code and matching the in-ternal fluid height with the external height at the damage opening. In another case studying a fishing vessel with anti-

roll tanks, the volume of water was fixed. Two methods kept the CFD and ship mo-

tions codes separate, but the methodology for linking them together varied. In one case, the exchange between the codes was by a data file. Another file controls which program runs, but each code runs completely separately on different com-puters. In the other case, the motion code was considered as an external subroutine by the CFD code. This was possible be-cause the code was designed for analys-ing sloshing in spacecraft, and so it was possible to include externally computed body motions as input to the problem. The advantage of this approach is that the CFD code needs no modification.

For one organisation, the CFD code was integrated into the ship motion code as a subroutine. Using linear motions to com-pute the ship motions is a simpler (and faster) process. However, it is limited to ‘operational’ rather than extreme mo-tions. Synchronising time steps between the two commercial CFD codes and the ship motion codes was an issue in both cases. In the Ro-Ro case, using files to transfer the data makes the coupling ex-plicit, but the limitation of very small time steps within the CFD code makes this solution acceptable for the applica-tion considered, but this may not be so for other cases. In the fishing boat case, even though the motions code is treated as a subroutine by the CFD code, the two can function independently and the time step in each one can be varied internally to obtain solutions. The CFD code uses a non-inertial reference frame. The re-quired ‘gravity’ components are interpo-lated at whatever time step the CFD code needs to obtain a solution. The limitation of the non-inertial reference frame is reached when the fluid reaches the top of the boundary. This limitation is less criti-cal for an enclosed tank than for water trapped on the deck, when at this point the water would spill out, which is not modelled by the CFD code. For one or-



ganisation, the CFD code adjusts the time steps internally to reach a solution, but since the two codes are combined the in-ternal clock is consistent.

All methods have been partially validated against experimental data (Bass & Cumming, 2000; van Daalen et al., 2000; Woodburn et al., 2001).

The two replies, also studying Ro-Ro fer-ries, took a different approach, which was out-side the classical definition of CFD. In one case a 6DOF non-linear time domain motions code has been expanded to 9DOF by considering the fluid as a lump mass. Hydrodynamic coeffi-cients related to ship-wave interaction are cal-culated externally with a frequency domain panel method. Calculated quantities are then transferred to the time domain by the Impulse Response Function technique. Flooding and draining of compartments can be modelled. The method has been validated against experi-ment data (Papanikolaou et al., 2000).

In the other case, the underlying equations of the ship are derived from conservation of linear and angular momentum applied to rigid bodies, extended to include the internal fluid mass in six degrees of freedom. The Froude-Krylov and restoring forces and moments are integrated up to the instantaneous wave eleva-tion, the radiation and diffraction forces and moments are derived from linear potential flow theory and expressed in time domain based on convolution and spectral techniques, respectively. The hull asymmetry due to ship flooding is taken into account by a “database” approach, whereby the hydrodynamic coeffi-cients are predicted beforehand, and then in-terpolated during the simulation. The correc-tion for viscous effects on roll and yaw modes of motion is applied based on empirical meth-ods. The second order drift and current effects are also catered for, based on parametric for-mulations. Fluid sloshing has been modelled by a free mass point moving due to the acceleration field and restrained geometrically by predetermined potential surfaces of centre of buoyancy for given amount of floodwater.

This model is derived from simple rigid body motion consideration. Finally, an artificial co-efficient is introduced to represent damping of floodwater motion. An ad hoc value of 0.15 is adopted for this coefficient derived for simple box-shaped compartment from comparisons with experimental data. With the geometric information about the tank stored in a data-base, the model is complete. The method has been validated against experiment data (Jasionowski & Vassalos, 2001).

The major advantage of the lumped mass approach is that it is computationally more efficient than the traditional CFD methods, and provided that there is not a significant amount of sloshing of the fluid, the prediction of ship motions, including the dynamics of floodwater is of acceptable accuracy. Flood-ing and draining issues are handled with em-pirical methods.

Using CFD for predicting water motion on the deck and its effect on ship motions was the focus of the survey. Other approaches that have been used in the past are to use either lumped mass (as one survey respondent did) or to use potential flow but for non-linear waves (Huang & Hsiung, 1996). In this case, Euler’s equations were used for non-linear shallow water flow on the deck, with a flux split based on the characteristic directions of motion. The method was employed for two and three-dimensional decks. Whilst this is not CFD in the classical sense, it is a valid approach for tackling the problem, provided that the volume of water on the deck is con-stant or changing very slowly with time.

An area that is similar to the extreme mo-tions and capsize problem is sloshing of fluid within tanks on a ship. Related to this, Cariou & Cassela (1999) give a comparison of nu-merical results for 11 different CFD codes. This paper summarizes the time and space do-mains, viscosity, compressibility free surface and wall condition for each of the codes, and compares results for specified cases. The first was a simple two-dimensional problem, and compared the surface elevation of the fluid in



an oscillating rectangular tank for a range of amplitudes and periods. The second case was a 3D problem, consisting of an LNG cargo tank, excited for a combination of pitch and roll, again for a range of amplitudes and periods.

An alternative to the grid based CFD ap-proaches is Smoothed Particle Hydrodynam-ics (SPH). In this method the fluid is idealised as elemental particles, and the tracks of these particles are followed during the computation. The method was originally developed for as-trophysics problems, but has since been ap-plied to free surface problems. The method can give a good prediction of breaking waves, but according to Beck & Reed (2001), de-tailed comparisons of pressures and fluid ve-locities for simplified problems are not avail-able. The method has been used to obtain some interesting results for the sloshing prob-lem and post-breaking behaviour of waves. Naito & Sueyoshi (2001) presented results for an SPH prediction of water moving on the deck of a flooded Ro-Ro ferry.

6. SYMBOLS AND TERMINOLOGY

A number of international organisations have been involved in ship stability for many years and have developed their own symbols and terminology for the field.

As these organisations have been using their symbols and terminology for a number of years, it is felt that it is now not practical to develop a single set of symbols and terminol-ogy to be used by all working in this field. However, the need has been identified, fol-lowing a comprehensive review by Frances-cutto (2002), for taking immediate action on some items that are directly linked to stability and recommendations to this effect are put forward. Moreover, a table comparing the symbols used by ITTC; ISO 12217; HSC 2000; and IMO has also been prepared and submitted to the Symbols & Terminology Group (Francescutto, 2002). This report is also posted in the web site of the committee.

7. CONCLUSIONS AND RECOMMENDATIONS

7.1. General Technical Conclusions

As the maritime industry progressively moves towards performance based criteria to address safety issues, there is wide scope and a major opportunity for member organisations to benefit from these developments.

However, the severe limitations identified in the existing numerical models for predict-ing ship capsizing and extreme motions needs to be addressed as summarised below:

7.1.1 Prediction of Intact Ship Capsizing and Extreme Motions

At this stage there is a limited number of numerical models for predicting intact ship capsizing and extreme motions with a range of different levels of sophistication and pa-rameters. Only a few of these models consis-tently agree qualitatively with all the extreme motions and modes of capsize identified in free running model experiments. None of the models does so quantitatively.

More work is required to improve the agreement between physical and numerical model tests results.

7.1.2 Prediction of Damaged Ship Capsizing

At this stage there is a limited number of numerical models for predicting damaged ship capsizing. Unlike the case for intact ships, most models can consistently predict the cap-size boundaries obtained from model experi-ments in realistic sea conditions, albeit for the specific damage scenario and mode of capsize considered in the benchmark tests.

There are, however, fundamental differ-ences in the way these models handle flood-water/ship dynamics, none of the models giv-ing good agreement, either qualitatively or quantitatively, with results from physical model experiments.



Before confidence can be gained in these models, wider application to different damage scenarios and ship types is required.

7.1.3 Guidelines and Procedures

Those member organisations involved with experimental testing of intact and dam-age stability generally follow the guidelines recommended by the 22nd ITTC as appropri-ate, justifying the adoption of procedures based on these guidelines as refined by the work undertaken by this committee.

7.1.4 Symbols and Terminology

The rationalisation of symbols and termi-nology for stability has not been made possi-ble, as international organisations directly dealing with the subject are too ingrained in the use of symbols different to those adopted by the ITTC. Notwithstanding this, it is rec-ommended that steps are taken to communi-cate widely the use and encourage the adop-tion of the Greek symbol “φ” to denote heel/list/roll angle.

7.2. Recommendations to the Conference

Adopt the Procedure “Seakeeping – Model Tests on Intact Stability” 7.5-02-07-02.5.

Adopt the Procedure “Seakeeping – Model Tests on Damage Stability” 7.5-02-07-02.6.

8. REFERENCES AND NOMENCLATURE

8.1. References

Bass, D., and Cumming, D., 2000, “Numeri-cal and Experimental Investigation of Wa-ter Trapped on Deck on a Small Fishing Boat”, STAB 2000.

Bass, D., and Cumming, D., 2000, “Numeri-cal and Experimental Investigation of Wa-ter Trapped on Deck on a Small Fishing

Boat”, STAB 2000.

Beck, R.F., and Reed, A.M., 2001, “Modern Computational Methods for Ships in a Seaway”, SNAME Annual Meeting, 2001.

Cariou, A., and Cassela, G., 1999, “Liquid Sloshing in Ship Tanks: A Comparative Study of Numerical Simulations”, Marine Structures Vol. 12, No. 3, pp. 183-198.

Cramer, H., 2001, “Effect of Non-Linearity in Yaw Motion on Capsizing Prediction” Proceedings of the 5th International Work-shop on Stability and Operational Safety of Ships, Trieste, Italy.

van Daalen, E.F.G., Kleefsman, K.M.T., Ger-rits, J., Luth, H.R. and Veldman, A.E.P., 2000, “Anti-roll Tank Simulations with a Volume of fluid Based Navier-Stokes Solver”, 23rd Symposium on Naval Hy-drodynamics, Val de Reuil, France.

Francescutto, A., 2002, “A Critical Review of the Symbols and Terminology Relevant to Ship Stability”, Department Naval Archi-tecture, Ocean and Environmental Engi-neering, University of Trieste, Italy.

Hamamoto, M., and Enomoto, T., 1996, “Model Experiment of Ship Capsize in Astern Seas − 2nd Report”, J. Society of Naval Architects of Japan, Vol. 179, pp. 77-87.

Hamamoto, M., and Kim, Y. S., 1993, “A New Coordinate System and the Equations Describing Manoeuvring Motions of a Ship in Waves” J. Society of Naval Archi-tects of Japan, Vol. 173, pp. 209-220, (in Japanese).

Hamamoto, M., and Saito, K., 1992, “Time Domain Analysis of Ship Motions in Fol-lowing Waves” Proceeding of the 11th Australian Fluid Mechanics Conference, Hobart, Vol. 1, pp. 355-358.



Hashimoto, H., and Umeda, N., 2001, “Impor-tance of Wave Effects on Manoeuvring Coefficients for Capsizing Prediction”, Proceedings of the 5th International Work-shop on Stability and Operational Safety of Ships, Trieste, Italy.

Ikeda, Y., Umeda, N., and Tanaka, N., 1988, “Effect of Forward Speed on Roll Damp-ing of a High-Speed Craft” Journal of Kansai Society of Naval Architects, Japan, Vol. 208, pp. 27-34 (in Japanese).

ITTC, 1999, “Specialist Committee on Ship Stability. Final Report and Recommenda-tions to the 22nd ITTC “Proceeding of the 22nd ITTC, Seoul, Korea and Shanghai, China, Vol. 2, pp. 399-431.

Jasionowski, A., 2001, “Detailed Analysis of the Revised Damage Benchmark Results”. University of Strathclyde − The Ship Sta-bility Research Centre, October 2001.

Jasionowski, A., and Vassalos, D., 2001, “De-tailed Analysis of the Final Revised Dam-age Benchmark Results”, University of Strathclyde − The Ship Stability Research Centre, December 2001.

Matsuda, A., Umeda, N., and Suzuki, S., 1997, “Vertical Motions of a Ship Running in Following and Quartering Seas”, J. Kansai Society of Naval Architects, 227 : 47-55, (in Japanese).

Matusiak, J., 2001, “Importance of Memory Effect for Capsizing Prediction” Proceed-ings of the 5th International Workshop on Stability and Operational Safety of Ships, Trieste, Italy.

Munif, A., 2000, “Numerical Modelling on Extreme Motions and Capsizing of an In-tact Ship in Following and Quartering Seas”, Doctor Thesis, Osaka University, Japan.

Naito, S., and Sueyoshi, M., 2001, “A Nu-

merical Analysis of Violent Free Surface Flow on an Flooded Car Deck Using Par-ticle Method”, STAB2001 5th Interna-tional Workshop on Stability and Opera-tional Safety of Ships, Trieste,Italy,.

Papanikolaou, A., Zaraphonitis, G., Spanos, D., Boulougouris, E., and Eliopoulou, E., 2000, “Investigation into the Capsizing of Damaged RO-RO Passenger Ships in Waves”, STAB2000.

Papanikolaou, A., and Spanos, D., 2001, “Benchmark Study on the Capsizing of a Damaged Ro-Ro Passenger Ship in Waves” Draft Final Report, May 2001.

Papanikolaou, A., 2001, “Benchmark Study on the Capsizing of a Damaged Ro-Ro Passenger Ship in Waves”, Revised Final Report, October 2001.

Renilson, M.R., and Manwarring, T., 2000, “An Investigation into Roll/Yaw Coupling and its Effect on Vessel Motions in Fol-lowing and Quartering Seas” STAB2000, Vol. A, pp. 452-459.

Salui K.B., Sarkar, T., and Vassalos, D., 2000, “An Improved Method for Deter-mining Hydrodynamic coefficients in Roll Motion using CFD Techniques” Ship Technology Research, Vol. 47, No. 4.

Sera, W., and Umeda, N., 2001, “Effect of Short-Crestedness of Waves on Capsize of a Container Ship in Quartering Seas” J Ja-pan Institute of Navigation, Vol. 104, pp. 141-146, (in Japanese).

SOLAS, 1997 “Consolidated edition 1997- Annex 5: Resolutions of the 1995 SOLAS Conference. Model test method” IMO , Resolutions of the Conference of Contract-ing Governments to the International Convention for the Safety of Life at Sea, 1974, adopted on 29 November 1995.

Umeda, N., and Yamakoshi, Y., 1986, “Ex-



perimental Study on Pure Loss of Stability in Regular and Irregular Following Seas”, STAB86.

Umeda, N., 1988, “Application of Slender Body Theory to Lateral Ship Motions with Both Free Surface and Free Vortex Layers”, Bulletin of National Research Institute of Fisheries Engineering, No. 9, 185-203.

Umeda, N., Hamamoto, M., and Takaishi, Y., 1995, “Model Experiments of Ship Cap-size in Astern Seas” J. Society of Naval Architects of Japan, Vol. 177, pp. 207-217.

Umeda, N., Matsuda A., and Takagi M., 1999, “Model Experiment on Anti-Broaching Steering System”, J. Society of Naval Ar-chitects of Japan, Vol. 185, pp. 41-48.

Umeda, N., 2000, “Effects of Some Seakeep-ing/Manoeuvring Aspects on Broaching in Quartering Seas”, Contemporary Ideas on Ship Stability, Elsevier Science Publica-tions (Amsterdam), pp. 423-433.

Umeda, N., Munif A. and Hashimoto H., 2000, “Numerical Prediction of Extreme Motions and Capsizing for Intact Ships in Follow-ing/Quartering Seas”, Proceeding of the 4th Osaka Colloquium on Seakeeping Per-formance of Ships, Osaka, Japan, 368-373.

Umeda, N., and Papanikolaou, A., 2000, “Re-vised Guidelines for ITTC committee on the prediction of Extreme Ship Motions and Capsizing Benchmark Tests”.

Umeda, N., “Final Report on the Benchmark Tests for Intact Ships”, ITTC SCEXCAP Progress Report, December 2001.

Umeda, N., and Hashimoto, H., 2002, “Quali-tative Aspects of Nonlinear Ship Motions in Following and Quartering Seas with High Forward Velocity” Journal of Marine Science and Technology, in press.

Umeda, N., and Hashimoto, H., 2002, “Quali-

tative Aspects of Nonlinear Ship Motions in Following and Quartering Seas with High Forward Velocity” Journal of Marine Science and Technology, in press.

Woodburn, P., Galagher, P., and Letizia, L., 2001, “Fundamentals of Damaged Ship Survivability”, RINA, Spring Meeting, 2002.

Vassalos, D., and Jasionowski, A., 2000, Stockholm Agreement Water on Deck Model Experiments for Passenger/Ro-Ro Vessel, Final Report, PSBG-RE-004-AY. University of Strathclyde − The Ship Sta-bility Research Centre, February 2000.

Vassalos, D., Umeda, N., and Papanikolaou, A., 2001, “2nd Revised Guidelines for ITTC committee on the Prediction of Ex-treme Ship Motions and Capsizing Benchmark Tests”, June 2001.

8.2. Nomenclature COREDES Committee on Research and

Development in European Shipbuilding

CRN Co-operative Research, Navy

JSNAJ Journal of the Society of Na-val Architects of Japan

RINA Royal Institution of Naval Architects

SNAME Society of Naval Architects and Marine Engineers

STAB Int. Conf. on Stability of Ships and Ocean Vehicles

WEGEMT European Association of Universities in Marine Tech-nology and Related Sciences

9. ACKNOWLEDGEMENT

The committee would like to thank wholeheartedly all the organisations and indi-viduals who contributed to this work over the last three years.

Documents

The Specialist Committee on Prediction of Extreme Ship ... · capsizing with sub-harmonic rolling in case (a) and capsizing with harmonic rolling in case (d) were well predicted