1

1. GENERAL

1.1 Membership and meetings

The following members of the 21st ITTCleft the Manoeuvring Committee : Prof Kijima,Dr Renilson, Dr Aage, and Prof Wu. The mem-bers of the present Committee wish to expresstheir appreciation for their effort during theirterm. The Committee appointed by the 21st

ITTC consisted of the following members :

Dr. Rod Barr (Secretary)Hydronautics Res. Inc.

Dr. Giovanni CapurroCETENA

Dr. Stéphane Cordier (Chairman)Bassin d’essais des carènes

Dr. Masayoshi HiranoAkishima Laboratory

Dr. J. Buus PedersonDanish Maritime Institute

Prof. Key Pyo RheeSeoul National University

Prof. Marc VantorreFlanders Hydraulic, University of Ghent

Dr. Ing. Zou ZaojianWuhan Transportation University

The Committee meetings were as follows :January 20th and 21st 1997

CETENA, Genoa, ItalySeptember 8th and 9th 1997

DMI, Lyngby Denmark

April 16th and 17th 1998Akishima Lab., Yokohama, Japan

September 16th, 17th, and 18th 1998Bassin d’essais des carènes, France

January 25th, 26th, and 27th 1999Flanders Hydraulics, Antwerpen, Belgium

1.3 Tasks assigned by the Advisory Council

The advisory council defined the followingtasks to be performed by the Committee:

1. Review the state-of-the-art, comment on thepotential impact of new developments on theITTC, and identify the need for research anddevelopment into manoeuvrability. Monitorand follow the development of new experi-mental techniques and extrapolation meth-ods.

2. Review the ITTC recommended procedures,benchmark data, and test cases for validationand uncertainty analyses and update as re-quired. Pass the information to the QualitySystems Group for publication in 1999.

3. Identify the requirements for new proce-dures, benchmark data, validation, uncer-tainty analyses and stimulate the necessaryresearch for their preparation.

4. Prepare an up-to-date bibliography of rele-vant technical papers and reports.

5. Strongly promote comparative model testsand force predictions including experimen-tal, semi-empirical, computational methods,and comparisons with the results of sea trialsfor modern ship types in deep water. Specificinterest is in the full-load condition, waterjetpropulsion, and the effect of aft-body varia-

The Manoeuvring Committee

Final Report andRecommendations to the 22nd ITTC

2

tions.6. Develop a reliable method of predicting ma-

noeuvring in shallow and restricted water,including squat.

7. Continue to promote research into manoeu-vrability standards, including the IMO in-terim standards, in order to provide advice toorganisations who set standards, such as theIMO, and pilot organisations.The committee has attempted to perform

these tasks with the exception of the 6th taskconcerning the development of a « reliablemethod for predicting manoeuvring in shallowwater.. » which was felt to go beyond the com-monly accepted ITTC scope of work.

2. SPECIAL GROUPS

2.1 RR74

In relation to the IMO Interim Standards forShip Manoeuvrability (IMO ResolutionA.751(18)), the panel of RR74 ManoeuvrabilityWG was established by Japan ShipbuildingResearch Association in 1995. The primary taskof this panel is to develop a database of fullscale manoeuvring trials mainly for newly-builtships with modern hull forms, and to review theStandards on the basis of the database devel-oped. Trial results of more than 200 ships havebeen collected and manoeuvrability analyses arebeing carried out with respect to adequacy ofthe criteria of the Standards. Some results ob-tained through RR74 activity are described byHaraguchi et al (1998). Besides the primarytask, basic studies for ship manoeuvring pre-diction are also being made by focusing resear-ch targets to the IMO Manoeuvrability Stan-dard.

2.2 SNAME panel H10

SNAME Panel H10 (Ship Controllability)has been active since 1996 in a number ofSNAME sponsored research projects and co-operative projects with U. S. pilots. A study ofthe prediction of slow speed manoeuvring wasinitiated in1997. A survey of current methodsused to address the unique problems ofvery lowspeed manoeuvring including criteria for meas-

uring accuracy was made. In October 1998 aworkshop was held at the U. S. Merchant Mari-ne Academy. The workshop, which was attend-ed by more than 20 designers, hydrodynami-cists, pilots and ship operators, explored allfacets of this problem and refined the plan ofaction for the project, which should be com-pleted by the end of 1999. A project to obtainfull scale manoeuvring data in the HoustonShip Canal was initiated because of the chan-nel’s pending widening and deepening and theability this offered to conduct trials in a beforeand after situation. This is a co-operative pro-ject with other groups concerned with the safetyand operation of the waterway. The panel willhelp develop a data acquisition plan usingDGPS, with special attention to vertical mo-tions (squat), which can provide the data mostuseful in advancing understanding of ship be-haviour in highly restricted waters. Work onanalysis of a large body of ship trials data hascontinued. It is intended to add these data to theexisting SNAME/Coast Guard Ship Manoeu-vring Data Base. Track data for several hundredship have been analysed, but suitable ship char-acteristic data, which are required by the database, have been found for few of these ships.Co-operative projects with pilots have includedan on-ship evaluation of various portable DGPSNavigation Units, preparation of recommenda-tions for a more detailed pilot card and prepa-ration of a Hydrodynamics Handbook for useby pilots. The first two of these projects havebeen completed while the last is ongoing. Thepanel supported a workshop on best practicesfor master-pilot communications. This work-shop which was sponsored by the AmericanPilots Association, the Maritime Administrationand the U. S. Coast Guard resulted in a bestpractices document that is now being used byindustry as a standard. Proceedings of the panelsponsored workshop on squat and the final re-port on modular manoeuvring models will bepublished in 1999

2.3 MARIN Co-operative Research, Ships

The 3 years program (1994-95-96) to devel-op a manoeuvring prediction code for singlescrew vessels was completed and the MPP codeis now in use among the various CRS memberorganisations.

3

At the 1996 Annual General Meeting(AGM), the Working Group received a newtask for the extension of the MPP code to twin-screw vessels. A one year project (1997) wasaccepted as a pilot study. The aim of the studywas to investigate the possibilities and limits ofan extension to twin-screw vessels of the ma-noeuvring prediction program (MPP) developedfor single-screw ships. A modified version ofthe MPP (MPP 97-1), covering both single andtwin screw ships (with one or two rudders) wasdeveloped, but still based on hydrodynamiccoefficients of single screw hulls.

A new validation process started in July1997 and the Working Group discussed the re-sults at meeting in October 1997. More than 30ships were used, and the comparison betweencalculations and measurements (model or full-scale) were not always satisfactory: especiallyturning circle parameters showed a large scatterand some systematic deviations. It was thusdecided by the Group that it would be desirableto modify the MPP97-1 program in order toachieve an accuracy similar to that of single-screw ships.

The proposal for a two years project (1998-99) was accepted by the AGM (1997) with themain objective to develop a reliable predictiontool for manoeuvring performances of twin-screw vessels.

3. HYDRODYNAMIC FORCES

A great deal of effort has been devoted todeveloping theoretical, semi-empirical as wellas experimental methods for estimating thehydrodynamic forces acting on manoeuvringships and for predicting ship manoeuvrabilitymore accurately at the initial design stage. Inparticular, the effect of hull form andhull/rudder/propeller interaction have receivedmore attention.

At the same time, more and more effortshave been devoted to applying advanced nu-merical techniques to calculate the hydrody-namic forces, and great progress has beenachieved in this respect. Although thesemethods are still not wholly reliable, their useas a design tool will become widely acceptedand used in the near future.

3.1 Hull Forces in Deep Water

Numerical Methods: RANS Solvers. Re-markable achievements have been made duringthe last few years in the prediction of the hy-drodynamic forces acting on a manoeuvringship by using 3D viscous flow methods.

After several years of code developmentseveral calculations have been performed bydifferent groups of researchers in order to testand to assess the validity of codes solving theRANS equations. Test cases include constantdrift and steady turning motions of a Series 60hull, the ESSO OSAKA, and the SR221 tankers:Alessandrini & Delhommeau (1998), CuraHochbaum (1998), Tahara et al (1998), Berth etal (1998), Ohmori et al (1996), Makino & Ko-dama (1997), Sowdon (1996). Generally, theresults presented show good agreement withavailable experimental data. These methods areparticularly interesting since they can be usedfor a wide range of different geometries. How-ever, a point which needs improvement is thedependence of the results to the grid size andtopology.

Sato et al (1998) proposed a numericalsimulation method for solving the manoeuvringmotion of blunt ships by coupling the equationsof motion with the Reynolds Averaged NavierStokes (RANS) equations or RANSE.

Figure 3.1 : Free-surface calculated around aSeries 60 at 10° drift angle (Alessandrini and

Delhommeau, 1998)

4

Numerical Methods: Potential flow solvers.The use of potential flow based numericalmethods has continued with several recent de-velopments. One of the principal difficulties inmodelling manoeuvring forces using potentialflow methods is to include the lift generatingvortices in the computation: their position alongthe hull, as well as their intensity, constant orvarying. Although this problem can usually besolved for lifting surface type bodies such askeels, rudders, etc., these simulations on ships,and particularly on full form ships, requiresome prior knowledge of the vortex field. As aconsequence, and given the complexity of thewake in steady turning motion, calculations aresometimes limited to steady oblique towing.Ando et al (1997) and Nakatake et al (1998)developed a surface panel method and present-ed numerical results for three VLCC modelswith different after-body shapes were comparedwith experiments.

Figure 3.2 : Comparison of forces and momentsin oblique towing (Nakatake et al, 1998)

By solving a 3D lifting potential flow prob-lem, Landrini & Campana (1996), and Zou(1996), calculated the hydrodynamic forces on asurface piercing plate in steady drift and turningmotion, in which the free surface conditionswere linearised with respect to the doublemodel flow and the wake was modelled by asystem of trailing vortices shedding from thetrailing edge and keel edge.

Combining a 3D Rankine Panel methodwith the unsteady linear lifting surface theory,Zou & Soeding (1995) simulated the forcedsway and yaw oscillating motion of a surface-piercing plate and calculated the linear hydro-dynamic derivatives.

Pinziy et al (1995) developed a panelmethod using the wave resistance Green’sfunction for solving the Neumann-Kelvinproblem for a surface-piercing body moving atforward speed with lifting effects. Results werepresented for a simple shape wing at yaw anglesand for a Wigley hull in symmetric motion.

Kijima et al (1995a) and Furukawa & Ki-jima (1996) proposed a prediction method forthe cross flow drag acting on a ship hull basedon the vortex shedding model with damping ofthe free vortices. It was shown that the longitu-dinal distribution of cross flow drag along theship length can be predicted with good accu-racy.

Kijima et al (1995b, 1996a, and 1996b)propose a prediction method to estimate thelateral force and yaw moment acting on a shiphull in oblique and turning motion. They im-proved the method by assuming the separationline position based on the results of captivemodel tests and investigated the influence ofafterbody shape based on SR221 ships A, B, C.

Several publications present developmentsusing slender body theory: Kose et al (1996,1997), and Xiong & Kose (1996), Wellicome etal (1995), Liu et al (1996), Clarke & Horn(1997b). Yumuro (1997) calculated manoeu-vring hydrodynamic forces on a ship hull withheel angle by using Bollay’s lifting surface the-ory. Beukelman (1997) proposed a theoreticalmethod to determine the lift forces on hull andrudder as well as the manoeuvring derivativesmaking use of the rate of change of fluid mo-

5

mentum. Based on slender body theory Tanaka(1996, 1998) proposed a practical method forpredicting hydrodynamic forces acting on a shipmoving with large drift angles. The flow wasmodelled by two dimensional cross flow at eachcross section, whereas both bound vortices andfree vortices are distributed to represent the twodimensional separated flow. Kim & Rhee(1996) applied parameter identification tech-niques to estimation of the manoeuvring coeffi-cients of a slender body. Karasuno & Maekawa(1996, 1997) presented a component typemathematical model of hydrodynamic forces insteering motion. They estimated the ideal flowforce, viscous and induced drag by applying asimplified vortex theory.

Experimental Methods. Ishida & Fujiwara(1995) conducted a large amplitude forced swaymotion test to investigate the effect of hullforms on the non-linear sway force and mo-ment. Takano et al (1995) conducted obliquetowing test to investigate the effect of bow andstern shape on ship hydrodynamic forces.Nonaka et al (1996) measured stern flow fieldsand hydrodynamic forces acting on three VLCCmodels in oblique towing motion. Kijima et al(1997) carried out measurements of hydrody-namic lateral force and yaw moment acting on aship hull to clarify the effect of roll motion onthe hydrodynamic forces. Nakatake et al (1995)measured wake distributions of three ship mod-els in oblique towing. Sadakane (1996) meas-ured the lateral drag coefficient on modelsmoving laterally from rest. Longo & Stern(1997) performed extensive force, flow fieldand wave measurements in the vicinity of aseries 60 model in oblique tow for CFD valida-tion purposes.

Semi-Empirical Methods. Clarke & Horn(1997a) developed new empirical expressionsfor the hydrodynamic velocity derivatives. Al-ternative predictor variables were suggested.Sutulo & Kim (1997) developed a regressionmodel for estimation of hydrodynamic forcesacting on the hull of a submersible in arbitrarythree dimensional manoeuvring motion. Bian-cardi (1997) proposed a method for calculatingthe hydrodynamic coefficients of surface shipsby applying adjusted formula obtained previ-ously for submerged bodies. The calculatedsway force and yaw moment were comparedwith the model scale measurements.

Kodan et al (1996) performed model tests ofrecent ships to obtain their hydrodynamic coef-ficients, and developed a new prediction proce-dure of ship manoeuvrability based on thesedata.

3.2 Hull Forces in Restricted Water

Theoretical/Numerical Methods. By using afinite volume method Ohmori (1998) calculatedthe viscous flow around a ESSO OSAKA tankermodel in steady drift motion and turning mo-tion in shallow water. The calculated hydrody-namic forces were compared with experimentaldata, and qualitatively good agreement wasobtained. Assuming that a ship hull can be re-placed by a rectangular flat wing, Yumuro(1995) proposed a simplified method for cal-culating the manoeuvring hydrodynamic forcesand the shedding angles of the trailing vorticesin shallow water.

Yasukawa (1997, 1998) applied unsteadyslender body theory to predict the hydrody-namic coefficients of the ship hull and rudder.The hydrodynamic memory effect of wake vor-tices generated by a slender ship advancingwith sinusoidal steering in shallow water wasinvestigated theoretically. Nakao et al (1995)calculated the hydrodynamic forces acting on amanoeuvring ship in confined water using slen-der body theory. Xiong & Wu (1996) applied a3D Rankine source method to calculate forcesand wave patterns of ship hulls moving in re-stricted water. The effect of free surface andcanal bank effects on the forces was empha-sised. Yumuro (1996) proposed a simplifiedmethod for calculating forces on a ship on anoff-centreline course in a narrow water channel.

Experimental and Semi-Empirical Methods.Vantorre & Eloot (1996) described unsteadyhydrodynamic phenomena which were ob-served during systematic captive model testseries carried out in shallow water. Ishibashi etal (1996) conducted captive model tests cover-ing a wide range of yaw and sway motion inshallow water to identify characteristics of hull,propeller and rudder as well as interactionforces. Clarke (1997) pointed out an error foundin a simple shallow water correction of hydro-dynamic derivatives published previously andsuggested new equations for hulls with rectan-

6

gular sections. Gronarz (1995) proposed a for-mulation of an exponential equation built up ofa constant term and a term describing the de-pendency of the manoeuvring hydrodynamiccoefficients from the water depth.

Laforce et al (1996) presented experimentalresults of systematic captive model tests onthree ship models of different lengths in openshallow water and in restricted water includinga cross section of the canal and a scale model ofthe geometry of the bends. The influence of theship’s length, water depth and canal banks onhydrodynamic forces were discussed.

3.3 Manoeuvring devices

Conventional rudders: Chau (1998) com-puted the turbulence flow around ship ruddersin uniform inflow by solving the RANSE, usingthe standard k-C turbulence model with wallfunction.

Gong et al (1995) conducted a series ofmodel tests to investigate the effect of rudderarea on the manoeuvrability of a ship with largeB/T. Rudder open water characteristics weredetermined by open water tests, and HPMMtests were carried out for the ship with ruddersof different areas. Ding et al (1997) conductedship model tests for measuring rudder lateralforce in both still water and following seas. Odaet al (1996) measured the open water normalforce and chordwise force of the mariner rud-der.

Jordan (1989) investigated the loads on therudder and especially the rudder stock duringemergency manoeuvres.

Non-conventional rudders: Zhu & Tang(1995) and Zhu & Wang (1996) conducted ex-periments on low aspect ratio circulation-controlled rudders in a circulating water chan-nel which demonstrated lift augmentation capa-bility.

Tachi & Endo (1996) conducted rudderopen water tests with a Schilling rudder.Hamamoto & Enomoto (1997) investigatedanalytically and experimentally the forces on aVec Twin Rudder system as well as the interac-tion between the two rudders. They proposed a

model of Vec Twin Rudder performance for theMMG model.

Other manoeuvring devices. Hirayama et al(1996a, 1996b) and Hirayama & Niihara (1996)performed model experiments and numericalsimulations to investigate the effectiveness ofan active vertical fin on improvement of thetransverse stability and manoeuvrability of highspeed displacement mono-hull ships.

Knowles et al (1996) performed a numericalstudy of the unsteady hydrodynamics of anUAV thruster, using a vortex-lattice, lifting-surface model modified to handle unsteady op-erating conditions during dynamic positioningand manoeuvring.

Endo et al (1997) conducted model tests andproposed hydrodynamic mathematical modelsfor side-thrusters.

Jukola & Castleman (1995) tested tractorand stern drive tugs and concluded that vesselsequipped with Z-drives are capable of othermeans of producing arresting and steeringforces with reduced risk of placing the escortingtug in a potentially dangerous situation.

The increasing use of podded propulsion isnotable and is causing large changes in the ma-noeuvrability of these ships, essentially cruiseships. However, little published data is avail-able on the manoeuvring characteristics ofpods.

3.4 Hull/Propeller/Rudder Interaction

Theoretical/Numerical Methods. A numberof models of the hull/propeller/rudder forcesand interactions have been developed, some ofwhich include viscous effects.

Suzuki et al (1996) computed the flow fieldaround a rudder behind a propeller by a viscousflow code and compared the results with meanflow measurements. Hinatsu et al (1995) cal-culated the viscous flow field around a tankerand its rudder. The propeller effect was consid-ered using equivalent body force distribution.

Kulczyk & Tabaczek (1995) applied an ad-vanced computational method to calculate

7

forces and moments on a rudder located in apropeller slipstream. Molland & Turnock(1996, 1998) developed a theoretical method topredict the hull/propeller/rudder interactionforces. Wang et al (1994) studied the propulsiveperformance of the propeller/rudder system. Avortex lattice method was applied to describethe performance of the rudder behind the pro-peller.

Yasukawa et al (1996) proposed a method tocalculate hydrodynamic forces on a ship mov-ing with constant rudder angle. Tamashima et al(1995) developed a theoretical method for pre-dicting the performance of a propeller/rudderbehind a ship. The propeller and the rudderwere modelled separately, and the mutual inter-action was taken into account by an iterativeprocedure.

Li & Dyne (1995) presented a linear methodto calculate the steady forces on the propel-ler/rudder combination working in a uniformflow.

Experimental and Semi-Empirical Methods.In order to develop a rational model for shipmanoeuvring, a series of rotating arm and lineartowing test were conducted to isolate forces onthe hull, propeller and rudder, and to study theinteractions between them (Lewandowski &Klosinski, 1992, Klosinski & Lewandowski,1993). A systematic set of model tests withfourteen different ship models and various rud-der-propeller configurations was performed bySedat & Fuller (1995). The testing was made tocreate data which could be used for a modularsimulation model

Nakatake et al (1996a, 1996b, 1997) con-ducted tests and calculations using the SQCMmethod to clarify the hull/propeller/rudder in-teraction mechanisms and found good agree-ment between their numerical techniques andthe experimental data.

Molland & Turnock (1995) conducted ex-perimental investigations to study the influenceof changes in the relative position of the rudderand propeller and concluded that significantchanges in both manoeuvring and propulsiveperformance could occur when the relativeposition of the rudder and propeller was altered.Based on wind tunnel tests, Molland & Turnock

(1998) also demonstrated the importance of theflow straightening influence of upstream hullform on the performance of the propeller/ruddercombination which can be used for the devel-opment of prediction methods which wouldinclude the effects of the drift angle and theupstream hull geometry.

3.5 External influences

Proximity effects: Based on slender bodytheory, Kijima (1997) predicted hydrodynamicinteraction forces between two ships and be-tween a ship and a pier. Manoeuvring motion inthe proximity of the pier was studied. Usingpotential flow around a slender body with arigid free surface, Varyani et al (1997) calculat-ed the hydrodynamic interactive forces betweenthree ships in a restricted waterway. The effectsof water depth and separation distance betweenships were investigated.

Figure 3.3 : Effect of rudder lateral separationon rudder lift slope (dcl/da) and propeller thrustcoefficient (dKt) (Molland & Turnock, 1995)

8

Wind. Blendermann (1995, 1996) proposedan empirical method to estimate the windloading of ships in uniform and non-uniformflow based on wind-tunnel tests. Blendermann(1997) also investigated the effect of beamwind on overtaking manoeuvres using wind-tunnel test data and numerical simulation. Shi-gehiro et al (1997) conducted model tests in awind tunnel and in a circular tank, and studiedthe course stability of towed fish preserves inthe presence of wind. Fukuchi et al (1997) pro-posed a method to improve the course-keepingability of a small vessel scudding under strongwind by spreading canvas around the flyingbridge. Model experiments were carried out in awind/water facility to measure wind loads.Simulations were conducted using the experi-mental data. Yamano & Saito (1997) developeda practical estimation method for wind forces.

Wave. Feng et al (1996) predicted the non-linear motion responses of a submerged slenderbody running near the free-surface due to waveexciting forces.

3.6 Non-Conventional Ships

Zhang & Andrews (1998) investigated themanoeuvrability of a trimaran ship. The hydro-dynamic forces due to the presence of the sidehulls have been analysed using a combinationof theoretical and empirical methods. The inter-action effects between the hulls were neglected.

Brizzolara et al (1998) presented a semi-empirical method for the study of dynamiccourse stability of waterjet propelled mono-hulls. PMM tests were carried out to measurethe hydrodynamic lateral force and yaw mo-ment for small drift angles and yaw rates withand without fins, whereas the influence of thewaterjet steering forces and waterjet inlets weretheoretically evaluated. Lewandowski (1997)developed a method to evaluate the coupledroll/yaw/sway dynamic stability of planingcrafts and presented expressions for the linearstability derivatives.

Sahin et al (1997) used a low-order singu-larity panel method to predict the hydrodynamiccharacteristics of underwater vehicles. Chiu etal (1997) investigated the lateral stability of anAUV using captive model test and empirical

data. Caccia et al (1997) performed tests toidentify hydrodynamic derivatives on a ROV.

Hiroshima et al (1997) and Kataoka et al(1997) developed and applied a 6 DOF perfor-mance prediction simulation method for yachts,in which the forces and moments were derivedfrom the CFD computation. Suzuki & Yoshi-hara (1995) computed the hydrodynamic forcesacting on sailing yachts by means of a surfacepanel method. Tahara (1995, 1996) developed anumerical method for calculating boundarylayer and wake flows around a sailing yachtwith yaw angle. The RANS equations weresolved with the Baldwin-Lomax turbulencemodel.

Figure 3.4 : Calculated flow around a IACCyacht in oblique flow (Tahara, 1996)

4. SIMULATION OF DYNAMICS

The work performed in the area of ship ma-noeuvring simulation covers the developmentand applications of mathematical models usedin manoeuvring simulations. This activity hasbeen stimulated by the need to meet IMO ma-noeuvring standards at design stage and to pre-dict manoeuvres in restricted waters (harbours,waterways....). Specific issues related to theapplication of the IMO standards are in section6.4

9

4.1 Modelling of Ship Dynamics.

Mathematical models of ship dynamics havebeen improved through use of more refinedtechniques, comparisons between manoeuvringmodels, and parametric studies. It is now pos-sible to incorporate very sophisticated simula-tion models in desktop simulators.

Mathematical Model Structure. It is now ac-cepted practice to divide mathematical ma-noeuvring models into « whole ship models »and « modular models ». In « whole ship mod-els », equations of motion are composed ofterms representing the total hydrodynamicforces acting on the hull/propeller/rudder com-bination, and the hydrodynamic force coeffi-cients required in these equations of motion aredetermined from tests of a model or from theo-retical predictions for a ship in which the pro-peller and rudder are installed and the propelleris operating at the appropriate loading condi-tion(s). In modular models, forces acting on thehull, propeller and rudder and the forces due tointeraction of these components or « modules »of the ship are each represented by differentterms in the equations, and forces of force coef-ficients are measured or predicted separately forthe hull, propeller and rudder. In modular mod-els, interactions between hull, propeller andrudder are sometimes measured in model tests,but are more typically determined from empiri-cal relationships incorporating parameters thatdepend on the geometry and position of therudder and propeller relative to the hull.« modular ship models » should not be con-fused with « modular computer programs orcodes » in which various physical and compu-tational functions are incorporated in separatesoftware modules or subroutines to facilitatemodification and de-bugging of the software.

Finally, the increase in computer power nowenables the resolution of the equations of mo-tions with forces calculated at each time step byan unsteady CFD code (Sato et al, 1998). Anexample of this approach is given by McDonaldand Whitfield (1997) for a change of depth ma-noeuvre of a self-propelled submarine (figure4.1). This approach has the advantage of beingtruly unsteady as opposed to the usual quasisteady models.

Figure 4.1 : Change of depth manoeuvre

McDonald and Whitfield (1997)

Lee et al (1997) compared the MMGmathematical manoeuvring model with a typi-cal whole ship model such as the Abkowitzmathematical model (Strom-Tejsen and Chis-lett, 1966). Results of simulation with a productcarrier using PMM model test data are given.The purpose of introducing a modular simula-tion model compared with a “whole ship”model is the ability to split the mathematicalmodel into relevant physical phenomenon, typi-cally a separation of hull forces and rudder-propeller-hull interaction forces, but also othermathematical models can be included such asthe behaviour of the engine, either a slow ormedium speed diesel engine or a turbine. Fur-thermore, research can be made with individualmodules.

Several authors presents work concentratedon models for the hull-rudder-propeller interac-tion. Chislett (1996) describes how hull forcesrelated to yaw rate, and rudder-propeller forcescan be non-dimensionalized in all four quad-rants. Molland et al (1996) describe an en-hanced rudder propeller model. Based onsimulation studies it is indicated that the en-hanced rudder propeller model should lead toimprovements of rudder propeller interactioneffects in a manoeuvring simulator. Lee et al(1996) and Kobayashi et al (1994) studiedmathematical models for a twin screw and twinrudder ships.

Perdon (1998) suggests a model for controlforces due to a hydrojet based on experimentaldata.

10

Alternative ways of defining the equationsof motions have been investigated by Bailey etal (1995, 1997, 1998a). Their work combinesthe knowledge of impulse response functionsfrom linear sea-keeping theory with traditionalmanoeuvring equations of motion, creating aunified general theory of ship motions. An in-teresting finding has been made that the tradi-tional manoeuvring derivative Yr was foundexperimentally to be approximated by

Y B UAr = − −26 22

in contradiction to the traditional theoryfound in numerous textbooks that

Y B UAr = − −26 11

where B26 is the frequency dependent sway-yaw damping and A11, A22 the frequency de-pendent surge and sway added mass coeffi-cients.

Pawlowski (1996) proposed a link betweenformal hydrodynamic models and CFD hydro-dynamic models. A general mathematical modelfor ship manoeuvring simulation is proposedand a discussion of various mathematical ma-noeuvring simulation model approaches isgiven.

Degrees-of-Freedom (DOF). The need insome cases to include more than 3 DOF in theequations of motion has been accepted; in par-ticular roll motion for surface ships. The influ-ence of GM and thereby the roll motion on shipmanoeuvrability was investigated among othersby Kijima et al (1997) and Kijima & Furukawa(1998). Measurements of hydrodynamic forcesas a function of speed and GM were made andnumerical simulations including roll motionwere performed. Figure 4.2 shows the signifi-cant effect of varying these parameters on thesimulated 1st overshoot angle in 10-10 and 20-20 zig-zag manoeuvres. The effect of includingroll in the equations of motion can be seen toincrease with ship speed.

A similar conclusion was obtained for amodern over-panamax container carrier byOltmann (1996) who showed that yaw instabil-ity increases with increasing approach speed.These results confirm earlier conclusions aboutthe effect of roll motion on the manoeuvrability

of ships with low transverse stability (low GM).

Sutulo & Kim (1997) present a unrestrictedmathematical model of submersible dynamicsbased on regression of parameters for the 6DOF forces and moments.

Prediction of Dynamics. Simulation of stan-dard manoeuvres has become more relevantsince the IMO Res. 751 was adopted.

A method for predicting ship manoeuvringwhich paid special attention to stern shape wasdeveloped by Kang & Kim (1995) using slenderbody theory, low-aspect-ratio theory, and cross-flow theory. For the cases considered the simu-lation results show good agreement with meas-ured manoeuvres. Hooft & Quadvlieg (1996)also use cross-flow drag and slender body theo-ry for prediction of forces used in a simulationmodel. Comparisons between simulated ma-noeuvres and full scale measurements showacceptable agreement.

When sea trials are performed in a “semi-loaded” condition means of extrapolating thedata to full load condition are required. Kijimaet al (1995) developed such a method whereturning ability is predicted satisfactorily but theprediction of overshoot angles in zig-zag ma-noeuvres is less satisfactory.

Figure 4.2.a: 10-10 zig-zag manoeuvre

11

Figure 4.2.b: 20-20 zig-zag manoeuvre

Figure 4.2: Influence of approach speed andGM on first overshoot angles (Kijima & Furu-

kawa, 1998)

Kristensen (1998) describes the design andservice experience with the manoeuvrability ofthree designs of double ended ferries with lowL/B and high B/T ratios. The manoeuvrabilitywas significantly improved through model testsresulting in slight changes in hull form or theaddition of skegs or bulbs/fins

Yoshimura et al (1997) developed a methodfor predicting the effect of a flapped rudder. Themethod is validated against free running modeltests. It is concluded that for course keepingsmaller flap angles are better for stable shipsand larger flap angles better for unstable ships.

Sohn et al (1996) made model tests withvarying rudder sizes for a full-formed ship. Oneconclusion is that the course stability does notnecessarily improve with increasing ruddersize. Another paper by Gong et al (1995) inves-tigates the influence of rudder area on the ma-noeuvrability of a ship with large beam-to-draught ratio by use of PMM tests and simula-tions. It is concluded that the rudder size im-proves the manoeuvrability of the tested shipexcept for straight course keeping.

The manoeuvring of a twin screw tankerwith rudder failure modes was investigated byTrägärd (1998). It is concluded that the tankercomplies with the IMO criteria even with onlyone rudder working. The behaviour of the en-gine during the last part of a turning circle ma-

noeuvre or during the stopping manoeuvre hasinfluence on the result of the manoeuvre. Also,for harbour manoeuvring a realistic modellingof the engine is essential.

Benvenuto et al (1996) suggests a mathe-matical model for simulating the behaviour of adiesel engine. The emergency stopping ma-noeuvre was investigated by Schmidli (1996)and he suggests that changes in the engine sys-tem could improve the stopping distance forsome ship types.

Simulation of submersible manoeuvres hashad some attention in this period. Li (1997)calculated a manoeuvring index and frequencycharacteristics for a submersible.

Sensitivity Analysis. Sensitivity analysisprovides a basis for determining the importanceand the required accuracy of the individualterms in the mathematical model. This is donethrough analysis of simulations performed withsystematic variations of each term. Such studieshave been stimulated by the need for more ac-curate manoeuvring predictions at the designstage.

Vassalos et al (1995) performed a sensitivityanalysis based on a naval ship manoeuvringmodel. He identified Yv, Yr, Nv, Nr as the mostimportant coefficients for the steady turningmotion.

Ishiguro et al (1996) studied the sensitivityof simulated IMO manoeuvres to values of thecoefficients in the MMG model using threedifferent ship types. The coefficients Yr, Yβ, Nr,Nβ as well as the flow straightening factor γ andthe wake fraction ratio ε were identified as themost important as shown in figure 4.3.

The knowledge from the SR221 researchproject with three tankers with the same mainparticulars but different stern frame sectionswere used to suggest a correction as function ofaft body shape (U to V shaped) to the proposedmathematical model. Significant improvementsin prediction were obtained, especially for di-rectionally unstable ships (Ishiguro et al, 1996,Kose et al, 1996).

The sensitivity of selected manoeuvringoutput variables (advance, transfer, overshoot

12

angles etc.) to various hull form parameters(stern shape, LCB, L/B) was investigated byKang et al (1995). Simulations were madebased on measured hull derivatives determinedfrom systematic PMM and free running tests on19 slow speed, full form hulls with stern bulbsand with horn type rudders. The stern shapeparameter,

σ awa

pa

CC

=−

11

where Cwa is the water plane area coefficientof the aft body and Cpa is the aft body prismaticcoefficient, was found to have a large influenceon the overshoot angles in the 10-10 zig-zag

manoeuvre but limited influence on the tacticaldiameter. However, the σa parameter was notincluded in the suggested regression equations.

Lee & Shin (1998) also used 19 PMM testresults to suggest another set of regressionequations to estimate the hydrodynamic hullcoefficients as well as the ε and γ parametersfor the MMG model. The regression includedthe parameters L, B, T, CB and a stern bulb areaparameter. The sensitivities of the 1st and 2nd

overshoot angles in the 10-10 zig-zag manoeu-vre to various parameters in the MMG modelare shown in Figure 4.4.

Figure 4.3: Sensitivity of MMG model parameters on 1st overshoots in the 10-10 zig-zag ma-noeuvre for three different types of ships (Ishiguro et al 1996).

Figure 4.4: Sensitivity of MMG parameters on 1st overshoots in the 10-10 zig-zag manoeuvre forthree different ship models (Lee & Shen, 1998)

13

The sensitivity study shows that ε and Nβ have aparticularly large influence on the overshoot angles inthe 10-10 zig-zag manoeuvre. Simulation resultsbased on estimated hydrodynamic parameters showgood agreement with simulated results based onmeasured hydrodynamic parameters.

Generally, the sensitivity of simulated manoeu-vres to each mathematical parameter will depend onthe following items:• the mathematical model itself.• the hydrodynamic of the ship• the manoeuvre that is investigated, i.e. turning

circle, zig-zag or other manoeuvres.

The work performed in the period covers pri-marily the items 2 and 3. Most of the sensitivitystudies use the MMG mathematical model. Thesesensitivity studies identify the parameters which aremost important, and where to put the emphasis ofresearch. Improvements in regression equations havebeen suggested by many authors to estimate the mostimportant parameters.

Modelling of Fast Ship Dynamics. Due to in-creased interest in fast transportation at sea, the ma-noeuvrability of fast ships is particularly importantand research in this area is increasing. However, re-search efforts concentrate on individual projects andnot on general tools which is due to the wide diver-sity of fast ship designs and the limited experience inthe area.

Hirayama et al (1996a) (1996b) investigated theeffect of anti-rolling active vertical fins on the ma-noeuvrability for displacement-type super high speedship. Kobayashi et al (1995) studied the manoeu-vrability of a high speed boat. The manoeuvrabilityof an air cushion vehicle was investigated by Huanget al (1996). Plante et al (1998) investigated the ma-neuverability of a planing craft. It is found that addedmasses depend on the forward speed of the planingcraft.

Enhancement Features. Simulation of the stan-dard manoeuvres via the Internet is described by Ha-segawa & Sasaki (1997). The model is based on thewell known MMG model and the users can via theInternet simulate standard manoeuvres on their owncomputer by downloading the Java source code. Shipsimulators of today include many features for model-ling realistic harbour manoeuvres.

The use of desktop simulators for harbour designstudies has become widely used. Kose et al (1995a),Galor (1997), Yang (1996) all present mathematicalmodels suited for harbour design studies.

4.2 Modelling of External Influences

Tugs and Towlines. An important aspect in har-bour manoeuvring is the use of tug assistance as thenumber or size of tugs can be the influencing factorfor a successful harbour manoeuvre. Also, oceantowing of large ships has received attention.

Harbour manoeuvres including tugs has been ad-dressed by Laible & Gray (1997) and Rooij (1996)who discusses the simulation of tugs. The use ofvarying modelling details of tugs in ship simulators isdescribed in Jakobsen et al (1996).

The combined motion of a tug towing a largetanker were investigated by Jiang (1997) and Jiang etal (1998). The modelling included the non-linear re-storing forces of the elastic towline. Numericalsimulations shows that the dynamic behaviour of thesystem is qualitatively different from results obtainedwith simpler models. It is also shown that it is ofgreat importance to include the tug dynamics in themathematical modelling of a tug-tanker tow system.Also Milgram (1995) investigated tow line tension inopen ocean towing

Position and length of tow wire for a tug towing abarge in shallow water was studied by Kijima & Fu-rukawa (1995) by use of simulations. Varyani (1997)describes a method for simulating of a tow includingthree ships.

Shigehiro et al (1996 & 1997) investigated the in-fluence of the course stability on a high speed towedfish preserve. Kreibel & Zieleski (1990) investigatedthe effect of different viscous damping models forsingle point mooring simulations.

Restricted Water Influences. The influence of re-stricted water covers both shallow water manoeu-vring and manoeuvres in channels or in the vicinityof banks.

Different formulations of lateral force and yawingmoment were compared with model experiment re-sults for shallow water manoeuvring for all drift an-gles by Vantorre & Eloot (1996). It was found that atabular formulation of the lateral force and the yaw-

14

ing moment was needed to cover the whole range ofdrift angles. Propeller action seems to have signifi-cant influence on lateral force and yawing moment,especially at large drift angles.

A study of the influence of the length of bulk car-riers in a canal was undertaken by Laforce & Van-torre (1996), Laforce et al (1996). Captive modeltests with three different ship sizes were undertaken,multi harmonic tests were performed, and fast timesimulations were made to evaluate the influence ofship length for the risk in the canal. Unsteady phe-nomena in restricted water were observed duringmodel testing.

Shallow water harbour manoeuvring was also in-vestigated by Ishibashi et al (1996). Kobayshi (1995)developed a method for simulating shallow watermanoeuvres based on deep water hydrodynamic de-rivatives which produces results with enough accura-cy for practical purposes.

The determination of hydrodynamic forces actingon a ship during berthing was investigated by Chen etal (1996) using a RANS code in 2D suited for un-steady time domain simulations.

Environmental Influences. There has been littlepublished research on the influence of wind, currentand waves.

Spyrou (1995) investigated the yaw stability ofships in steady wind. It is concluded that dynamicinstability can occur in following or head windswhereas for other relative wind directions the shipwill be course stable. Larjo (1994) defined windlimits on large cruise ships. Wind tunnel experimentswere performed to determine the wind loads for usein the simulation model.

The phenomena leading to broaching of ships infollowing waves were investigated by Spyrou (1996& 1997). Both steady state and transient analysis iscarried out. Yang and Fang (1998) addressed shipmanoeuvring in a non-uniform current. More gener-ally, the influence of shear current on the ship hullseems to have had little attention in the past.

5. SCALE EFFECTS AND VALIDATION

Publications on the topics of this chapter havebeen very scarce during the period of 22nd ITTC.The state of the art is reviewed and summarised

based mainly on committee reports of previous ITTCproceedings.

5.1 Scale Effects

In spite of a great deal of efforts to reveal scale ef-fects, the ship-model correlation is still one of the keyissues related to model testing techniques in shipmanoeuvrability.

Model tests are generally classified into two cate-gories. One is free-running model tests, where thesame tests as in full scale are typically performed.Manoeuvring characteristics of a full scale ship canbe predicted directly from model tests or throughsimulations using coefficients obtained by systemidentification techniques. The other is captive modeltests where hydrodynamic forces in manoeuvringmotion are measured. Full scale predictions are madewith the use of a mathematical model in which re-sults of model tests are used as input data.

The most important scale effect in manoeuvringtests is caused by the inability to achieve ReynoldsNumber similarity. As a result, scale effects in thearea of ship manoeuvrability are essentially caused bythe lack of similarity for velocity field in stern regionespecially over the rudder.

Other possible factors affecting the fidelity ofmodel tests (accuracy of model geometry, surfacetension, cavitation number, roughness, engine con-troller...) are generally less important or can be ac-counted for if deemed important.

Prediction from Free-running Model Tests. Thereare two principal aspects of Reynolds Number relatedscale effects in free-running model tests: decreasedvelocity field in stern region due to thicker boundarylayer and increased flow velocity over the rudder dueto higher propeller loading at the model self-propulsion point. As a result of these scale effects,rudder effectiveness of a model may generally beover-estimated compared with that of a real ship.Accordingly, free running models tend to be morestable (or less unstable) with respect to course-keeping ability. This effect is typically less significantfor fine ships because of their inherent stable coursekeeping ability.

One typical example which illustrates the above-mentioned effect is given in figure 5.1, where steadyturning performances obtained by free-running model

15

tests for 100 KDWT crude oil carrier are presentedtogether with full scale trial results (Okamoto et al,1972). It is seen in Fig. 5.1 that small-sized models(both L = 2 & 6 m) show stable course keeping abil-ity while large-sized model (L = 14.5 m) has unstablecharacteristics similar to that at full scale.

On the other hand, an opposite result has been re-ported for a ULCC with the use of three models of L= 4, 10 & 30 m (Sato et al, 1973). In this free-runningmodel tests, the larger-sized model was less unstablethan the smaller model (figure 5.2). However, fullscale trial results indicated that the ship was moreunstable than the smallest model.

Figure 5.1 : Spiral test results for 100 KDWTtanker model

Figure 5.2 : Scale effects on K’ and T’ indices(ULCC model)

Attempts to apply additional towing force to aself-propelled model have been made in order tocompensate the over-estimated rudder effectivenessmentioned above. Oltmann et al (1986) used thistechnique on a 20/20 zig-zag manoeuvre with amodel of the ESSO OSAKA. It was found that a pro-peller loading between model and ship self-propulsion points gives the best agreement with fullscale results.

The contradictory results on scale effects on slow,full form ships raise questions about our under-standing of scale effects even for large models. Forfiner hull forms such inconsistencies have not beenreported and free-running model tests continue to bewidely used to predict the manoeuvring behaviour ofships.

Prediction from Captive Model Tests. Hydrody-namic forces measured in captive model tests areused as input data to a mathematical model of shipmanoeuvring motion, where scale effects can be ap-plied to each hydrodynamic coefficient. In this re-spect, full scale predictions from captive model testsare widely understood to be more scientifically basedfor most manoeuvres than free-running model tests.

While scale effects on the ahead resistance aretaken into account in the conventional manner, thelateral forces and yaw moments obtained by captivemodel tests are generally not corrected. Results ob-tained in a co-operative effort with geosim models ofthe ESSO OSAKA (L = 2.5 - 7.256 m) shown in figure5.3 (17th ITTC proceedings) indicate that no signifi-cant scale effects exist on the linear coefficients. Alt-hough scale effects are expected in non-linear coeffi-cients which are mainly dependent on cross flowdrag, non-linear coefficients derived from the samemodel tests do not reveal clear scale effects. This maybe due to the fact that non-linear coefficients dependon the range of drift angles and yaw rates used inthese sets of experiments as well as the regressiontechniques used to identify the coefficients.

Given the importance of scale effects onhull/propeller/rudder interactions, scale effect correc-tions should be applied. These can readily be intro-duced in modular models (e.g. MMG) through scal-ing of interaction parameters such as propeller andrudder wake factors and the flow straightening coef-ficients. Scale effects on the first two parameters arevery significant, while the flow straightening coeffi-cient may not be affected (Yumuro & Yamamoto,1992).

16

Figure 5.3 : Comparison of hydrodynamic linear de-rivatives for different model sizes (ESSO OSAKA)

5.2 Validation

Validation of the predicted manoeuvres is per-formed by comparison with full scale trial results.Hence, the quality of the validation will largely de-pend on the reliability and quality of full scale trials.It should be noted that adequate validation requiresmore than a “good agreement” on limited manoeu-vres for limited ship type.

However, both prediction methods and full scaletrial results are subject to different types of errors anduncertainty. It is therefore essential that in both casesthe uncertainty in the results should be known in or-der to validate the method.

The method needed for validation depends on themethod used to predict the manoeuvres ; typicallyfree-running model tests and simulation models(Berlekom, 1992). In the first case, validation israther simple and can be performed in a straight for-ward manner. In the case of simulation models, vali-

dation of the prediction method requires the valida-tion of the different stages (we identified 5 stages)which comprise the construction of a ship manoeuvresimulation software.

Free-running model tests. The validation of free-running model tests is performed by comparison ofmodel scale manoeuvres with full scale test data for agiven set of conditions and ship state parameters(rpm, rudder angles, etc..).. The validity of test resultsdepends on the accuracy of the measured model ma-noeuvres and on scale effects which are difficult toaccount for in these tests. The accuracy of the modelmanoeuvres can be assessed through an analysis ofthe experimental errors (uncertainty analysis) whicharise in the course of following a specific test proce-dure for these experiments.

Validation of Simulation Models. The validity ofa simulation model depends on a number of factorsas outlined in figure 5.4

The validation of force data as determined bycaptive model tests, and computational methods reli-es on the availability of reference benchmark data.Chapter 8 discusses and attempts to establish abenchmark data base for this purpose using the ESSOOSAKA hull form. There is a need for benchmark datafor other hull types. The CFD community has beenusing the Series 60 although little manoeuvring datais known for this form.

The accuracy of measured forces depend on theuncertainty in the measurement and on the use of anappropriate test procedures. The document written bythis Committee and incorporated in the ITTC QMpresents such a test procedure and a method for as-sessing the uncertainty of forces measured and, tosome extent, the uncertainty in the coefficients de-rived from PMM tests. However, adherence to thetest procedure alone does not guarantee the accuracyof the results, since the parameters of the test (e.g.model size, amplitude and frequency of oscilla-tions...) are crucial. For this purpose recommendedguidelines are included in Chapter 7.

Uncertainties arise in the identification of coeffi-cients from force data measurements such as obliquetowing or PMM tests. Hence, an uncertainty analysisof these derived quantities should be made.

17

Validation procedure for predictingtrials manoeuvres using simulations.

1 - Forces

Source : Force measurement (steady or harmonic), Numerical

computations Validation requirement : Benchmark data

2 - Coefficients Source : Empirical formulas Force data regression System identification from free-running tests Validation requirement : Documentation of method used Benchmark data

3 - Math Model Structure Source : Adapted to the type of manoeuvre (DOF....) Validation requirement : To be determined

4 - Simulation Software Validation requirement : Test cases (debug)

5 - Simulated Manoeuvres Type of manoeuvres:, Std. manoeuvres (i.e. IMO), Controlled ship operations (i.e. harbour) Validation requirement : Trials data Mariner’s input

Figure 5.4: Validation procedure for predicting trialsmanoeuvres.

Empirical formulae are widely used at the designstage to provide estimates of the coefficients basedon principal characteristics of the ship. The mostcommonly used parameters in the regression equa-tions for describing the ship are the L, B, T, CB, LCB,and σa which to some extent describes the shape ofthe aft body. It is the opinion of the committee thatmore parameters describing the ship are needed toincrease the accuracy of the most important parame-ters. This, however, requires a large data base of reli-

able model test and/or computational data. The re-sults shown in Chapter 8 imply that the compilationof such a data base would require prior verification ofthe sources of data (tests or calculations).

Two applications of simulations can be identified:• simulations performed in the course of the design

of a ship to verify the performance for standardmanoeuvres.

• simulations of controlled ship operations (e.g.simulators)The validation of a mathematical model should

reflect the simulation requirements of the applicationconsidered: ship geometry, manoeuvring devices,expected ship motions (roll), restricted waterway,etc..

Research in ship manoeuvrability has shown thatwith 20 to 30 coefficients and parameters in amathematical model the standard manoeuvres of aship may be predicted with sufficient accuracy.

Each model in a ship simulator has a specificpurpose, sometimes a very simple model working inthe first quadrant is perfectly suitable, at other timesdetailed modelling of all four quadrants at slow speedand large drift angles is necessary. Environmentalinteraction from wind current, waves, shallow water,banks, ship-ship interaction etc. can be of utmostimportance. It is therefore obvious that the validationof a simulation model is closely connected with thepurpose for which each model is going to be used.

In the case of controlled ship operations, quanti-tative validation is not practical. Therefore the ac-ceptance of such models is based on the qualitativeassessments of professional mariners. The committeerecommends that a set of standard manoeuvres cov-ering the parameters (and subsequent ranges) whichwill be used in a specific simulation should be per-formed and documented before a simulation isstarted.

Full Scale Benchmark Data. There are very fewfull scale trials carried out at a scientific level withwhich validation can successfully be made. The ESSOOSAKA trial is one of such scarce and valuable fullscale trials, where extensive manoeuvring tests hadbeen carried out both in deep and shallow water con-ditions. Taking the opportunity of ESSO OSAKA trial,extensive model tests have been made at many placesfrom both aspects of hydrodynamic forces and ma-noeuvring motions. A benchmark study is made forESSO OSAKA by the Manoeuvring Committee as de-

18

scribed in Chapter 8, where efforts are made toprovide a means to validate hydrodynamic forces.Another set of extensive full scale trials data obtainedon a Mariner hull was used as a basis for validationin the 14th ITTC proceedings (Eda, 1975).

Full scale trials are always carried out for anewly-built ship before delivery according to thecontract and a large number of full scale trial resultshave been collected and accumulated in ship yards.Unfortunately few of them have been reported andfewer have been used for validation purposes and inthe past the quality of these results has sometimesbeen poor due to lack of accurate tracking, environ-mental conditions, motivation...

A practical way to validate predictions is throughcomparisons with trials data for a large number ofship types and manoeuvres. Hirano (1981) has com-pared full scale trial results for seven merchant ships,covering a wide range of ship types from 10 KDWTtraditional cargo boat to 400 KDWT ULCC withblock coefficients of CB = 0.52 - 0.83, and for threetypes of manoeuvring motions covering a wide rangeof rudder angle. A suitable level of validation wasconfirmed through these comparisons.

A methodology to evaluate validity of the predic-tion method on a scientific basis was discussed in20th ITTC Proceedings. In order to assess the degreeof “agreement” in quantity in a sophisticated manner,a concept of error bands had been proposed, Dand(1992).

Accuracy of Full Scale Measurements. Since1994, the NAVSTAR GPS (Global Positioning Sys-tem), consisting of a constellation of 24 satellites, isin operation giving world-wide coverage 24 hours aday. The absolute positional accuracy of GPS in theautonomous mode can be as low as 100m. In order toimprove accuracy, DGPS (Differential Global Posi-tioning System) has been developed, which is nowused widely as a means to get accurate (1 - 5 m) andcontinuous position information in a reliable andcost-effective manner. Applications of DGPS formonitoring full scale trials in general and standardmanoeuvres in particular are described by Cortellini& Lauro (1995), Stenson (1995), Yum et al (1996)and Youn et al (1997).

In addition, an advanced technique of KGPS(Kinematic Global Positioning System) has recentlybeen developed to further improve the measurementsaccuracy. Full scale measurements with the use of

KGPS and RTKGPS (Real Time KGPS) have beenmade by Hirata et al (1997) for standard manoeuvresof a small-sized cargo ship (L = 70 m). The meas-urement accuracy obtained was 2 cm in the horizon-tal direction and 5cm in the vertical direction. Similarfull scale measurements by both KGPS and RTKGPShave been made by Takase et al (1997) and Suzuki etal (1998).

In general, the use of DGPS is recommended inorder to keep a position accuracy better than 10 m asrequired by Norske Standard NS2780 (1985), or 3%of the turning diameter as suggested by the 20th IT-TC Manoeuvring Committee.

6. SHIP OPERATION AND SAFETY

Because safety during navigation depends on in-teraction between ships, the environment, and theoperators, studies have been conducted to promotemarine safety through improvements in shiphandlingsimulators and application of modern control algo-rithms. Also, inherent ship manoeuvrability has be-come an important design factor since the adoptionof “Interim Standard for Ship Manoeuvrability” bythe IMO.

6.1 Ship-handling Simulators

Ship-handling simulators have been widely usedin the design of ports and fairways, and for trainingand demonstrating competence of many maritimeskills and objectives, and to evaluate the performanceof newly developed controller.

The primary emphasis and publications in simu-lator development are concerned with the fidelity ofthe simulator environment, including the display.Improvements in the underlying manoeuvring modelsare reported in section 4.

Some of the recent publications in the use of shipsimulators to assess ship safety and operations arediscussed in this section.

Gong, et al (1996) developed a simulation systemfor the assessment of harbour capabilities from theview point of safety of ship navigation in congestedharbour area. Pourzanjani (1996) examined the ef-fectiveness of electronic chart system on collisionavoidance behaviour in coastal water navigation.Endo, et al (1996) used a ship handling simulator to

19

evaluate the control performance of the joy stickcontroller for berthing based on decoupling controltheory. Takahashi, et al (1996) verified the effective-ness of the navigation support systems, leading lightsand the Ramark Beacon, when entering port of theTechno Super Liner, and the results of simulator ex-periments were compared with the records of theexperimental ship HISHO. Kose, et al (1995) des-igned a prototype Integrated Navigation System(INS)taking into consideration the human decision makingprocess, and Ishioka, et al (1996) used real time shiphandling simulator to evaluate the supporting systemin INS for collision avoidance.

6.2 Control

Autopilot. The first application of control theoryto the ship is an autopilot for course keeping. Nowa-days, autopilots have been used widely in ship steer-ing, and several kinds of control theories are adoptedto enhance the robustness or to increase the effec-tiveness of the controller.

Jiang (1997) studied the influence of tow-hooklocation, towline length, and control parameters of aPID autopilot on the non-linear dynamic behaviour ofthe tow when a tug-tanker tow is operating in calmwater by use of locally linearised stability analysis,time-domain simulation and Poincare map. Nakataniet al (1996) proposed a simple and safe automaticPID gain tuning methods using relay control for typi-cal marine PID controller, and they applied thismethod to an autopilot system, a yaw control systemthrough bow thruster and a diesel engine governorsystem.

Neural networks was applied to the autopilot fortanker conning by Logan (1995). He envisaged theintelligent autopilot that will learn ship manoeuvringdynamics through experience, allowing the neuro-controller to apply rudder and engine speed control inthe same manner as humans. Sutton, et al (1996) usedartificial neural networks in the design of fuzzyautopilots and later Sutton & Marsden (1997) used agenetic algorithm to optimise a fuzzy rule basedautopilot, and showed that such approaches can pro-duce effective designs.

Adaptive controllers which can compensate thedisturbance such as wind and wave during manoeu-vring motion were designed. Zhang et al (1996 &1997) designed a robust autopilot system for courseregulation or directional stability control in assumed

wave disturbances and in a random sea by applying acontrol strategy of variable structure control. Wang etal (1996) applied the fuzzy control method for theadaptive control of heading and position under uni-form wind. An adaptive autopilot for submarine viagain scheduling was designed by Dumlu et al (1995)based on the stochastic controller and observer tech-niques which possesses robustness against possiblechanges in the external environment. Ogawara et al(1995) showed that the Learning Feed-Forward Con-trol System has good controllability to compensatefor the wind disturbance.

A robust digital servo control method incorporat-ing the concept of the annihilator polynomial wasapplied to auto-pilot control system in course changeto the specified direction by Han et al (1995) andconfirmed the effectiveness of the proposed controlmethod by model tests. and Zuev et al (1996) appliedimpulsive course-keeping autopilot to unstable ship.

Collision Avoidance. For safe and effective navi-gational assistance, collision avoidance systems havebeen developed in two ways. One is to reason thedegree of collision risk, and the other is to determinethe collision avoidance manoeuvre. Imazu (1996)assumed that the collision avoidance was consisted ofan information processing ability and the avoidingaction ability, and developed decision model for col-lision avoidance action considering the actions on asecond and subsequent stages based on a forecast ofthe encounter condition. An algorithm is proposedfor the real-time detection of encounter situationcompatible with the real behaviour of ship’s officersby Zec (1996). Hilgert et al (1996) created a commonrisk level from the actions requested by relevantsteering and sailing rules of the International Regula-tion for Preventing Collisions at Sea (COLREGS).The moment at which a collision avoidance manoeu-vre should be executed in a dangerous two-ship en-counter in order to obtain a certain passing distancewas calculated by Kwik (1996) based on the ships’equations of motion in conjunction with the kine-matics of ship encounters. Lisowski et al (1996) de-termined ship’s optimum safe trajectory in a collisionsituation of passing many moving targets by using amultistage decision-making process.

6.3 IMO Standards

Since IMO has adopted Resolution A.751 (18)“Interim Standards for Ship Manoeuvrability”, sever-al studies have compared the performance of existing

20

ship populations to the Standards. Based on thesecomparisons, certain changes in the criteria havebeen proposed.

In order to review the IMO standards, extensiveefforts to compile full scale manoeuvring databaseswere made by Capurro & Sodomaco (1996), andKang et al (1996). Raestad (1996) found that the cri-teria generally are based upon sound principles andthat most ships behaving “normally” will be in com-pliance with the standards. Kijima, et al (1997)pointed out through use of numerical simulation theimportance of GM, which is not considered in theinterim standards.

Oltmann (1998) presented an overview of the ac-tivities concerning the definition of various proposedmanoeuvring standards during the last fifty years,with special attention to the IMO. A constant limitingvalue of 17° for the 1st overshoot angle in a 10° zig-zag regardless of speed is proposed.

Following the adoption of the IMO interim stan-dards, a special effort was made by the Ship ResearchInstitute of Japan to collect full scale trial data fornewly-built ships (Haraguchi et al, 1998). The da-tabase consists of trial results of 226 ships in total,most of which have been built during the last decade.The database covers a wide range of ship types, inwhich about two third of the ships are full hull formships such as oil tankers and bulk carriers with mod-ern hull forms (pram stern with semi-balanced rud-der). Full scale tests have been carried out for 73ships in service condition, of which 23 dry cargo ves-sels and bulk carriers. According to the IMO Ma-noeuvrability Standards, data for three types of ma-noeuvring motions, namely turning motion with35deg. rudder, 10/10 and 20/20 zig-zag manoeuvreand full astern stopping motion, have been collectedand stored.

As shown in figures 6.1 and 6.2, there exist aconsiderable number of ships which do not complywith the criteria for the second overshoot angle in10/10 zig-zag manoeuvre and the first overshoot an-gle in 20/20 zig-zag manoeuvre. Moreover it ispointed out that about a half of the ships which donot comply with the above-mentioned criteria docomply with the criteria for the first overshoot anglein 10/10 zig-zag manoeuvre.

Based on these results, Japan has formally sub-mitted a proposal for revision of the interim stan-dards (IMO MSC 70/20/6 dated 28/7/98). This pro-

posal indicates that the criteria for the 2nd overshootangles in a 10/10 and the 1st overshoot angle in a20/20 zig-zag manoeuvres are not appropriate be-cause "...the present criteria may regard ships withgood manoeuvring performance as having poor ma-noeuvring performance."

Furthermore, this document indicates that "...thecriteria on stopping ability should be improved basedon results of research work considering physicalparameters e.g. displacement, horsepower of a shipand so on...".

Norrbin (1998) discussed the procedures and ex-periences of the crash-stop test against a review oftrial and scale model results, track reach estimates,and presented a guideline formula for the track reachnot to be exceeded.

6.4 Squat.

Definition of squat. Tuck defines squat as fol-lows: "Squat is not a change of draft (...). It is anoverall lowering of the ship together with the waterin the neighbourhood of the ship. Hence it is almostunseen in the open sea, where it is nevertheless pres-ent. However, squat is mainly of concern in restrictedwater (...)". For this reason, papers handling the sink-age due to forward speed in deep, unrestricted waterare not discussed in this report.

Need for reliable squat data. The organisation ofmeetings on squat (SNAME Workshop, WashingtonDC, 1995; Nautical Institute Seminar, Hull, 1995)and the attention paid to this subject by internationalmaritime associations (PIANC/IAPH, 1997) reveal arenewed interest in this topic. A need is recognisedfor more reliable information about a ship's sinkage,which is an essential element in determining an ap-propriate under-keel clearance for safe transit throughchannels with restricted depth Overestimation ofsquat may lead to excessive dredging expenses ornon-optimal use of navigation areas while underesti-mation of squat can lead to unsafe situations.

Squat predictions. Reviews of practical, empiricalmethods allowing an estimation of squat based on alimited number of parameters characterising shipgeometry, waterway configuration and ship speed arepublished by PIANC/IAPH (1997), Dand (1999),Vantorre (1999a), Millward (1996). Substantial de-viations can be observed between the results of suchformulae.

21

A new simple prediction method based on nu-merical calculations using slender body theory andmodel test data was published by Kijima & Higashi(1999). Ankudinov et al (1996) and Ankudinov &Jakobsen (1999) developed a semi-empirical engi-neering and simulation tool to estimate squat forconventional ship types with a minimum of inputvariables and computational efforts.

Figure 6.3 : Comparison of squat predictions (PI-ANC/IAPH, 1997)

Full scale measurements. Use of GPS techniquesfor real-time measurements of a ship's squat leads tomore reliable results with a standard deviation of lessthan 0.1 m. Queensland Transport (1996) reportssquat effects on a large passenger ship transiting anentrance channel using DGPS and RTKGPS posi-tioning software. Comparison with empirical squatformulae shows a good correlation with the formulaof Millward (1996), while the values extracted fromthe ship's squat estimation table were considerably inexcess. A GPS survey of deep-draft vessels in theChesapeake Bay channels, including squat measure-ments, was described by Hewlett (1999). The use ofGPS for compensating hydrographic surveys for thesquat of the survey vessel was discussed by Huff(1999).

Special conditions. Effects of channel confine-ment and asymmetry on squat are described byNorrbin (1999). Vantorre (1999b) discusses the influ-ence of fluid mud covering the bottom of a naviga-tion area on a ship's squat.

Special craft. Bertram & Grollius (1994) present a3D potential flow panel method for computing resis-tance, sinkage and trim of SWATH ships in shallowwater. Results are good, except near critical depthFroude numbers. The wavemaking resistance andsquat of a fast catamaran moving uniformly in astraight rectangular shallow water channel was theo-retically investigated by Jian et al (1995) using thetechnique of matched asymptotic expansions.

Figure 6.4. Measured and calculated sinkage and trimof twin-hull moving in shallow water channel with

H/T = 2.27 (Jiang et al, 1995)

7. MODEL TEST TECHNIQUES

Considering experimental research with shipmodels in the area of manoeuvrability, a distinction ismade between free-running and captive model tech-niques. Progress in both techniques has been re-viewed and a comprehensive survey of current prac-tice in captive model tests was organised and ana-lysed with respect to the choice of experimentalparameters for captive model tests. The presentCommittee has also formulated an adapted version ofprocedures of captive model testing on behalf of theITTC Quality Manual.

7.1 Review of free-running model tests

Test techniques. The requirements to be met byan outdoor test location for conducting standard ma-noeuvres with radio-controlled large models are dis-

22

cussed by Rossignol (1995). A detailed description ofthe NSWCCD facilities (Lake Needwood) is given;outdoors and indoor test results are compared.

An application of a GPS technique to free-running model tests has been attempted by Ueno et al(1997), and compared with the existing UPS (Ultra-sonic Positioning System). Sufficient accuracy andreliability for position measurement was obtained.

Analysis techniques. The use of neural networksfor the identification of a 4 DOF model based onfree-running tests (zigzag manoeuvres, harmonicrudder angle variations, steady turn) is discussed byCaux & Jean (1996). They concluded that a classicallinear neural network model generally provides goodpredictions, although results are less reliable if ruddervariations are applied in the resonance frequencyrange for roll.

7.2 Review of captive model tests

Test techniques. Several efforts were made tooptimise captive manoeuvring test programs by usingnovel techniques.

A very compact apparatus for circular motiontests (CMT), to be used autonomously or combinedwith a towing carriage, is described by Karasuno et al(1996). The application of this apparatus to severaltypes of tests with the intent to reduce test duration isdiscussed.

Reduction of the time required for a captive testprogram may be achieved by combining several testparameters in one run; e.g. combined oscillatory testscan replace separate yaw and drift tests (Rhee et al,1998). Results of alternative tests in shallow water(h/T<1.2) are compared with stationary tests by Eloot& Vantorre (1998). Although in some cases goodagreement was obtained, discrepancies may occurdue to non-stationary effects and incomplete flowdevelopment around hull or rudder.

A technique using a rotating arm (RA) facility fordetermining equilibrium conditions (speed, drift an-gle, rudder angle) as an alternative to free runningtests was described by Perdon (1998).

Agdrup et al (1998) investigated the applicabilityof a wind tunnel PMM to ship manoeuvrability as-sessment; except for Yr’, a qualitative agreementwith tank results was obtained.

Test program and analysis. The literature reflectsa tendency towards an optimised standard captive testprogram and analysis technique developed for a par-ticular simulation model. The PMM test programpresented by Blok et al (1998) contains 97 runs,comprising both bare and appended hull tests, whileSutulo & Kim (1998) claim that a specially optimisedexperimental program of only 20 combinedsway/yaw tests is sufficient for use with their mathe-matical manoeuvring model.

Existing guidelines for selecting suitable pa-rameters (amplitude, frequency) for harmonic captivemanoeuvring tests are reviewed by Vantorre & Eloot(1997). A relation between non-stationary phenomenaand interference with the model's swept path duringPMM tests is discussed, and a alternative non-stationary sway test is proposed. A method for opti-mising PMM test conditions, yielding most reliableresults, is suggested by Rhee et al (1998), based onsensitivity analysis. Several estimators applied todetermine manoeuvring coefficients from PMM testsare compared; the non-recursive least square estima-tor appears to be preferable to the recursive and ge-netic algorithm estimators.

7.3 Current practice in captive model tests

Captive model test techniques have been used forthe last 30 years. During this period, each institutionhas developed its own methods, mainly based onsemi-empirical considerations. The ITTC identified aneed for guidelines in order to ensure the quality oftest results. The 21st ITTC Manoeuvring Committeeformulated a "Recommended standard PMM testprocedure" which has been extended in three ways:

• to cover rotating arm tests (RA),• to provide quantitative guidelines,• to suggest an analysis procedure for the uncer-

tainty.

The quantitative data are based on two sources:literature on captive testing published during the lastdecades, and the results of a questionnaire distributedamong all ITTC member organisations in 1997. Apositive response was received from 37 institutions,providing a solid base for an overview of actualpractice.

Questionnaire. Taking account of an increasingneed for guidelines and even standard test proce-dures, the Committee considered a thorough insightin present methodologies for selecting the experi-mental parameters for captive model tests - being the

23

result of years of experience of many institutions - asa requirement. For this reason, a questionnaire wascirculated among 110 ITTC Member Organisations inorder to obtain an overview of the actual practice ofcaptive model testing. A positive answer was re-ceived from 37 institutions, covering a total of 61facilities. This report summarises the response to thequestionnaire. A more detailed overview will be pub-lished.

The questionnaire consisted of three parts:1. Experimental facilities: main specifications and

physical limitations.2. Experimental program: actual practice.3. Data acquisition and processing.

Test types. Taking account of the mechanism in-volved and the motion imposed to the ship model, adistinction can be made between different types oftests:(a) Stationary straight line tests in a towing tank

(a1) straight towing;(a2) straight towing with rudder deflection;(a3) oblique towing;(a4) oblique towing with rudder deflection;

(b) Harmonic tests, requiring a towing tank equippedwith a PMM:(b1) pure sway;(b2) pure yaw;(b3) pure yaw with rudder deflection;(b4) pure yaw with drift;

(c) Stationary circular tests, by means of a rotatingarm or a x-y-carriage:(c1) pure yaw;(c2) yaw with drift;(c3) yaw with rudder deflection;(c4) yaw with drift and rudder deflection.

Tests a1, a3, b1, b2, b4, c1, c2 are carried out fordetermining hull forces; a2, a4, b3, c3, c4 yield rud-der induced forces, and are therefore non-applicablein case the model is not fitted with rudder and pro-peller (bare hull testing).

The questionnaire covered the following numbersof facilities for each category:(a) Stationary straight line tests ..... 53 facilities(b) Harmonic tests.......................... 34 facilities(c) Stationary circular tests ............ 14 facilities

Experimental facilities. Figure 7.1 presents dif-ferential and cumulative distributions of the data onship model length L. When an average value wasgiven, the limiting values of model length were as-

sumed to be 33 % lower and higher than the meanvalue. The following conclusions can be drawn fortest types (a) and (b):• comparable lengths are used for (a) and (b);• the median value for L appears to be 4.5 m;• the distribution reaches a peak at a L ≈ 3 m;• 95% of all tests are carried out with L> 2 m.On the average, circular tests (c) are performed withsmaller models. The median length is only 3 m, thepeak in the distribution is reached at 2.2 m, and the95% limit is 1.5 m.

Figure 7.2 shows that most tests of types (a) or (b)are carried out in a tank with a length of 35 times theship model length, which is also approximately themedian value. Most circular tests (c) are carried outin a tank the largest dimension of which is about 20times the model length.

According to figure 7.3, a median value for modellength to tank width ratio (L/W) is 0.47 for stationarystraight-line tests (a), and somewhat smaller (0.42)for harmonic tests (b). This difference can be ex-plained by the fact that PMM mechanisms aremounted in tanks with a width, which is larger thanthe average towing tank. As most circular tests areexecuted in circular or wide tanks, the median valueof L/W for test type (c) is much smaller (0.09).

Experimental program: test type (a). Table 7.1gives an overview of the number of ship speeds, pro-peller loadings, drift angles and rudder angles appliedduring captive model tests. A larger number of modelspeeds is used for resistance-propulsion tests (a1), asthe self-propulsion point has to be determined by thiskind of tests. For other types of tests (a2, a3, a4), themedian value appears to be 1 or 2.

The majority of the tests are carried out at the(model or ship) self-propulsion point. Straight towingtests without rudder action (a1) and rudder force tests(a2) are often carried out at other propeller loadingsas well.

The number of drift angles applied in tests a3-a4is on the average smaller for oblique towing testswith rudder action. The highest frequency is observedat 12 angles for type (a3), and 5 angles for type (a4).A similar distribution is obtained for the number ofrudder angles at which tests a2/a4 are carried out.The way drift and rudder angles are selected is dis-played in figures 7.4(a3) and 7.5(a2), respectively.

Besides parameters related to the ship model

24

kinematics and control, the questionnaire requestedfor details concerning some parameters related toexperimental and analysis techniques: waiting timebetween runs, acceleration phase, settling phase,steady phase, deceleration phase. An overview isgiven in Table 7.2.

Experimental program: test type (b). An overviewof the number of parameters determining the shipmodel kinematics and control is shown in Table 7.1.Most harmonic tests are carried out at only onespeed-rpm combination.

The number of sway or yaw velocity amplitudesapplied during tests of types (b1) and (b2), respec-tively, varies between 1 and 20, 4 being a medianvalue. There is only a slight difference between thedistributions for (b1) and (b2), which is remarkable,as in general sway tests are only performed for de-termining the sway acceleration derivatives, whileyaw tests also provide data on yaw rate dependentforces and moments. Median ranges for nondimen-sional sway and yaw velocities are [0.1 ; 0.35] and[0.16 ; 0.58], respectively.

The number of amplitudes applied in a harmonicsway (b1) and yaw (b2) test program may vary be-tween 1 and 10, 3 being a median value. The mediannumber of frequencies selected for such types of testsis 2.

The ratio of lateral amplitude yA to tank width Wis in some cases restricted by the technical limitationof the driving mechanism, but even if the lateral mo-tion extends over the full width, yA/W is selected tobe not larger than a certain value in order to avoidwall effects. As shown in figure 7.6, the sway ampli-tude typically takes less than 10% of the tank width.

An important issue concerns the selection of thePMM frequency ω, which can be expressed non-dimensionally in various ways:

′ =ω ω1

Lu

(7.1)

′ = = ′ω ω ω2 1Lg

Fn (7.2)

′ = = ′ω ω ω3 12u

gFn (7.3)

Figure 7.7 and table 7.3 present an overview ofactual practice in selecting ω'1, ω'2, ω'3. For this pur-

pose, a Froude number range between 0.05 and 0.3was assumed if no indication could be found on thistopic in the completed questionnaires. Table 7.3 alsomentions recommended values according to empiri-cal rules of thumb formulated by several authors (seeITTC Quality Manual: Manoeuvring - Captive ModelTest Procedure, discussed in paragraph 7.4).

Interaction of yawing with drift and rudder actionis typically verified at four drift angles and three rud-der deviations, but are only seldom combined. Notendency can be observed concerning the selection ofthe rudder angle range (see figure 7.5(b3)); drift an-gles for combination with yawing are selected in therange |β| < 30 deg, [0 deg;16 deg] being a medianrange (see figure 7.4(b4)).

Common practice concerning execution parame-ters, such as the number of cycles considered foranalysis is given in Table 7.4, which also gives anindication about the number of cycles skipped in or-der to obtain a steady state. Waiting times betweentests of types (a) or (b) are comparable.

Experimental program: test type (c). An overviewof the number of ship speeds, propeller loadings, driftangles and rudder angles applied during rotating armor circular motion tests is given in Table 7.1.

Non-dimensional yaw rates r’ vary from 0.07 to1; a median range appears to be [0.2 ; 0.75]. Thenumber of yaw rates varies between 2 and 16, 4 beinga median value.

The maximum drift angle applied during tests oftype (c2) varies between 10 and 20 deg. About 50%of the respondents apply an asymmetric range.

Data acquisition and processing. The memberorganisations were requested to answer, which of alist of data were always, sometimes or never meas-ured during captive manoeuvring tests. The replies,reflected in figure 7.8, can be summarised as follows:• longitudinal and lateral hull force components

and yawing moment are (of course) always meas-ured;

• a majority always measures the position and/orspeed components of the driving mechanism, aswell as parameters characterising the control ofthe ship model steering and propulsion equipment(rudder angle, propeller rpm), and thrust and tor-que acting on the propeller(s);

• following data are always or sometimes measuredby a majority of the respondents: rolling moment,

25

forces and moments on rudder(s) and, to somelower extend, sinkage and trim.

Sampling rates vary between 4 and 250 Hz, 20 Hzbeing a median value.

7.4 Captive model test procedures and analysistechniques

Captive model test techniques have been used forthe last 30 years. During this period, each institutionhas developed its own methods, mainly based onsemi-empirical considerations.

The ITTC identified a need for guidelines in orderto ensure the quality of test results. The 21st ITTCManoeuvring Committee (1996) formulated a "Re-commended standard PMM test procedure". On be-half of the ITTC Quality Group, the present Com-mittee proposed an updated version of this procedure,entitled “Manoeuvring - Captive Model Test Proce-dure”, which can be considered as an extension of thelatter in three ways.

In the first place, the considered techniques arenot restricted to PMM testing, but other captivemethods are also discussed. Procedures for rotatingarm tests, however, are still in development.

Secondly, an attempt is made to provide quantita-tive data, unlike the former procedure, which inten-tionally was given a qualitative character. The quan-titative data are based on two sources: literature oncaptive testing published during the last decades, andthe results of the questionnaire discussed in section7.3.

Finally, basic ideas for an uncertainty analysis areformulated. It is clear that in comparison with othertests, such as resistance tests, such an analysis is farmore complex in the case of captive manoeuvringtests, for several reasons:

• The number of possible causes of uncertainty isvery large, and substantially depends on the con-cept and the characteristics of the experimentalfacility.

• Several data have to be measured simultaneouslyduring captive manoeuvring tests.

• According to the kinematics imposed to the shipmodel, a large number of test types can be distin-guished.

• A rather important number of parameters has to

be selected for determining a captive model test.• Several techniques may be applied to analyse the

measured data.• The structure of the mathematical manoeuvring

model is also of importance for assessing the un-certainty of test results.

For these reasons, it is not possible to formulate auniversal uncertainty analysis procedure that can beapplied for any captive model test. Instead, someexamples are given, indicating the importance of theselection of test parameters.

For further details reference is made to the ITTCQuality Manual.

Figure 7.1. Differential and cumulative distributionof the length of ship models used for several types ofcaptive model tests.

Figure 7.2. Differential and cumulative distributionof the ratio of ship model length to tank length forseveral types of captive model tests.

26

Figure 7.3 : Differential and cumulative distribution of the ratioof ship model length to tank width for several types of captivemodel tests.

Figure 7.6. Harmonic sway tests (b1): distribution ofsway amplitude to tank width ratio.

Table 7.1. Test types (a), (b), (c):number of test parameters.

# forward speeds #propeller loadingscum. distr. (%) Max cum. distr. (%) max0 50 80 100 Freq 0 50 80 100 freq

a1 1 3 9 15 1 1 2 5 20 1a2 1 2 4 6 1 1 1 5 10 1a3 1 2 3 9 1 1 1 5 10 1a4 1 1 3 5 1 1 1 8 10 1B 1 1 3 10 1 1 1 1 10 1C 1 1 2 4 1 1 1 3 8 1

# drift angles # rudder anglescum. distr. (%) Max cum. distr. (%) max0 50 80 100 Freq 0 50 80 100 freq

a2 - - - - - 2 10 15 17 9a3 3 11 15 23 12 - - - - -a4 3 8 14 20 5 2 8 14 20 10b1 - - - - - 1 1 1 10 1b3 - - - - - 2 3 4 6 3b4 2 4 6 10 4 1 1 4 10 1c2 3 7 12 24 6 - - - - -c3 - - - - - 2 6 17 24 6

Table 7.2. Stationary straight line tests (a):experimental parameters

Cumul. distr. (%) max0 50 80 100 freq.

Acceleration (L) 0.07 1.7 5.5 33.3 0.8Settling (L) 0.1 2.2 5.5 13.3 1.5Steady (L) 0.3 8.7 17.2 80.0 3.5Deceleration (L) 0.07 1.7 5.3 20.0 0.7Waiting time(min)

15 15 20 20 15

27

Table 7.3. Harmonic tests (b):frequency selection

max.Freq.

P=50% P=80% empiri-cal

ω1' 0.5 - 1.5 5.0 14 1 - 4ω2' 0.1 - 0.2 0.5 0.9 0.15-0.2ω3' 0.02- 0.04 0.08 0.22 << 0.25

Table 7.4. Harmonic tests (b)execution parameters.

P = 50% P = 80% max.freq.

Transient 1 cycle 3 cycles 1 cycleSteady 2 cycles 4 cycles 2 cyclesWaiting time 15 min 25 min 15 min

Figure 7.8. Data-acquisition: data measured duringcaptive manoeuvring tests.

28

Figure 7.4. Distributions of limits of drift angle range applied for tests (a3) and (b4).

Figure 7.5. Distributions of limits of rudder angle range applied for tests (a2) and (b3) (legends: seefigure 7.4).

Figure 7.7. Distributions of non-dimensional PMM frequencies for tests (b).

29

8. ESSO OSAKA Benchmark Study

One of the tasks assigned to the committeeand which was considered essential by thiscommittee was to develop a quantitative basisfor assessing the ability and limitations of cur-rent captive model test techniques and numeri-cal computation methods for predicting impor-tant hydrodynamic forces acting on a ship (barehull or appended hull) under typical manoeu-vring conditions (drift angle, yaw rate or rudderangle). Of particular interest are the variationsexisting in measured or predicted force, theinfluence of experimental factors such as modelscale ratio, and the capabilities of numericalmethods.

The ESSO OSAKA was selected as a bench-mark ship for assessing current state of the artin prediction of manoeuvring forces and mo-tions for the following reasons:

• Data are available from a very extensiveand very carefully conducted set of shiptrials in deep and finite (shallow) waterdepths;

• Many model tests at various model scaleratios, and many manoeuvring simulationstudies, have been conducted for the ESSOOSAKA;

• The ESSO OSAKA was extensively studiedin Japan and reported by JAMP (1985).

• An extensive comparison of available re-sults for the ESSO OSAKA was presentedby Barr (1993).

Other ships considered, but not selected be-cause of their more limited database, includedthe Mariner which was previously consideredin detail by the 15th ITTC Manoeuvring Com-mittee, and a ship from the MARAD series ofmore modern, shallow draft, full form hulls,described by Roseman (1987). It is recognisedthat there is an essential need for a similar setof data comparable to the ESSO OSAKA set butfor other hull forms.

8.1 ITTC Member Survey

To determine the expected level of supportand co-operation in using the ESSO OSAKA as abenchmark ship, a survey was prepared and

sent to all ITTC member organisations. Thissurvey is summarised in Table 8.1.

Table 8.1: Questionnaire on ESSO OSAKA Ma-noeuvring Studies

1. Have you previously conducted ESSOOSAKA model tests? Can raw data or otherdata be provided?

2. Have you done manoeuvring simulations?Can results or simulation model be pro-vided?

3. Have you made numerical computations forthe ESSO OSAKA ? Can results be pro-vided, and are methods described in theopen literature?

4. Would you be willing to conduct newmodel tests using an existing, new or bor-rowed model? Would you be willing tobuild a new ESSO OSAKA model?

5. Do you currently have a model of the ES-SO OSAKA ? What is scale ratio, wouldyou loan to others?

6. Would you conduct numerical computa-tions?

7. Indicate where responses apply to deepwater, shallow water or deep and shallowwater.49 organisations responded to this survey.

More than 30 organisations indicated a willing-ness to carry out numerical calculations of hy-drodynamic forces, while very few indicated awillingness to undertake new model tests. Table2 summarises responses to the survey. Not allrespondents answered all questions.

Table 8.2 Summary Responses

Question/Response Yes No

Conducted previous tests 19 30

ESSO OSAKA model now avail-able

11 37

Willing to build new model 10 29

Willing to conduct new tests 28 20

Deep water only 20

Deep and shallow water 8

30

Conducted previous simulations 19 30

Conducted numerical predictions 39 10

Willing to make new numericalmanoeuvring predictions

22 23

As prediction of forces was of greatest in-terest to the Committee, all organisationsagreeing to calculate forces were requested todo so for the following set of typical operatingconditions:

WaterDepth

Drift Angleβ=(°)

Turning Rater’ (rad/s)

Deep 20 0Deep 0 0.5Deep 0 0.75Deep 8 0.5

H/T=1.2 8 0H/T=1.2 0 0.5 H/T=1.2 8 0.5

Electronic data files defining the hull andrudder lines and propeller characteristics wereprepared from the drawings and sent to eachorganisation.

The present investigation was restricted tohull damping. Propeller and rudder forces,which are of equal or greater interest, were notconsidered due to the lack of consistency in, orabsence, of data.

Forces and moments are non-dimensionalized as follows:

Y’ = Y/ (½ ρ U2L2)

N’ = N/ (½ρU2L3)

where:Y sway forceN yaw momentρ= water mass densityU ship forward speed in m/sL ship length (LPP) in m.

Forces are plotted as a function of drift an-gle, β=(in degrees), and non-dimensional yawrate, r’:

β = sin-1 (v/U)

r’ = r.L/U

where:v sway velocity in m/sr yaw rate in rad/s

8.2 ESSO OSAKA Hull Force Data Sources

Sources of available ESSO OSAKA forcedata and coefficients are presented in Table 8.3.Included are only those sources used in thepresent comparisons. Data sources which couldnot be easily used due to the data format (plotsonly..) were not included.

Table 8.3 Sources of ESSO OSAKA Force Data

References SourceData

Format

JAMP(1985) Model test CoefficientsMiller (1980) Model test CoefficientsAnkudinov (1979) Empirical CoefficientsAbkowitz (1981),(1984)

Systems ID(SI)

Coefficients

Gronarz (1988) Model test CoefficientsDand (1983) Model test CoefficientsEda (1983) Model test CoefficientsOgawa (1977) K.U. Model test CoefficientsShiraka (1997) Model test CoefficientsBogdonov (1987) Model test CoefficientsBigot (1997) Tests/CFD Force tablesLech (1998) Model test CoefficientsCopenchov (1998) Model test CoefficientsVaryani (1995) Empirical CoefficientsOhmori (1998) CFD CoefficientsZou (1998) Theory CoefficientsHirano (1985) Model test Coefficients

Data presented by Bigot (1997) is note-worthy for the extensive CFD and model testresults.

8.3 Deep Water Force Data

Non-dimensional force data obtained fromthe sources listed in Table 8-3 were compared,and statistical variation of these forces deter-mined. Figures 8.1 through 8.4 present com-parisons of sway and yaw forces as a functionof drift angle, for r' = 0 and 0.75, for both barehull and hull with rudder. Comparisons for oth-

31

er yaw rates are similar. The present compari-sons are more comprehensive than those ofBarr (1993), as they include many new theore-tical and experimental results obtained in re-sponse to the Committee's request to memberorganisations. Tables 8.4 and 8.5 present stan-dard deviations of forces divided by the meanvalue. The largest values typically result fromsmall mean values.

When considering the results in figures 8.1to 8.4 and the tables 8.4 and 8.5, it is evidentthat extremely large variations exist in meas-ured hydrodynamic forces which are funda-mental in the prediction of manoeuvres (Chap-ter 4).

Figure 8.1: Comparison of bare hull swayforces as a function of drift angle (r'=0)

Figure 8.2: Comparison of bare hull yawmoments as a function of drift angle (r'=0.75)

Figure 8.3: Comparison of hull and ruddersway forces as a function of drift angle (r'=0)

Figure 8.4: Comparison of hull and rudder yawmoments as a function of drift angle (r'=0.75)

Table 8.4: Statistics of forces and moments forbare hull

Sway Force Yaw MomentDrift r’ Mean Std./ Mean Std./Angle .103 Mean .103 Mean

0 0.30 1.61 0.36 -1.24 0.1540.50 2.83 0.35 -2.20 0.130.75 4.74 0.33 -3.54 0.13

8 0.30 6.05 0.20 -0.04 2.770.50 8.06 0.20 -1.21 0.180.75 11.3 0.19 -2.99 0.13

32

Table 8.5: Statistics of forces and moments forhull with rudder and propeller

Sway Force Yaw MomentDrift r’ Mean Std./ Mean Std./Angle .103 Mean .103 Mean

0 0.30 2.14 0.54 -1.17 0.240.50 3.64 0.52 -2.02 0.250.75 5.58 0.52 -3.16 0.28

8 0.30 5.85 0.13 -0.07 4.060.50 7.87 0.20 -1.06 0.460.75 10.83 0.24 -2.46 0.37

Figures 8.5 through 8.8 present standarddeviations of forces and moments. On thesegraphs the mean value of the force and momentcorresponding to a typical for a large rudderangle turn (r’=1, β = 8°) is plotted to provide aabsolute reference of forces.

Figure 8.5: Std. deviation of sway forces ona bare hull as a function of yaw rate

Figure 8.6: Std. deviation of yaw moments on abare hull as a function of yaw rate

Figure 8.7: Std. deviation of sway forces onhull and rudder as a function of yaw rate

Figure 8.8: Std. deviation of yaw moments on hull and rudder as a function of yaw rate

These figures show that the standard devia-tion is relatively constant for r'<0.2, but thatabove this value, it increases monotonically.The standard deviation is much less sensitive todrift angle. It should be noticed that the com-puted standard deviations are of the same orderof magnitude as the forces or moments actingon the ship when manoeuvring (value at r'=1.0,β=8°).

Scale and speed effects: One issue on whichit was hoped to shed some new light was thatof scale effects. Previous studies, Barr (1993),have found evidence, but no clear confirmation,of scale effects. Figure 8.9 shows a very largeamount of scatter and indicates no correlationof sway force at 12° drift angle with modellength (correlation factor of 0.37). There wasalso no correlation of yaw moment at 12° driftangle with model length. Identical conclusionswere drawn from sway forces or yaw momentsfor r’ = 0.75 with or without drift. The effect ofReynolds number could not be investigated astest speeds were often not published.

Figure 8.10, from Bigot (1997) shows ro-

33

tating arm and oblique towing test data atvarying speeds which show no significant ef-fect of speed in the speed range tested. There islittle difference in yawing moments at signifi-cant yaw rates or in sway forces for any driftangles or yaw rates. However, large differencesdo exist between hull yaw moments due to puredrift at a ship speed of 5 knots. These resultsindicate that important scale effects may existat these low model Reynolds number for yawmoment due to pure drift.

Figure 8.9: Sway force as a function of modellength (β=12°)

Figure 8.10: Yawing moment as a function ofdrift angle for different scaled ship speeds

(r' = 0 and 0.75)

Force Data from Systems Identification.Abkowitz (1981 and 1984) investigated systemidentification (SI) procedures for estimatingforces and moments from ESSO OSAKA trials indeep and shallow water. In the 1984 report hediscusses the use of different numbers of statevariables for the identification of deep waterforce coefficients and concludes that little is

gained by using four measured parameters (u,v, ψ and r) rather than three parameters (u, v,ψ) plus a value of r calculated from ψ. Table8.6, which compares identified values of linearcoefficients for various input parameters, indi-cates large differences in yaw moment coeffi-cients.

Figure 8.11 compares yaw moments calcu-lated using these three sets of parameters for arange of yaw rates for two drift angles (0° and12°). Also shown on the figure are the extremevalues of all model test data for appended mod-els. These results indicate that the differencesin yaw moments obtained from trials data usingdifferent SI methods are as large as the scatterof test data from all model tests. It is thereforeconcluded that values of the identified coeffi-cients have to be validated by simulating ma-noeuvres other than those used in the identifi-cation. Furthermore, the choice of suitableidentification parameters has to be validated forother manoeuvres and other ships.

Table 8.6: Linear coefficients for differentidentification parameters, Abkowitz (1984)

Parameters 4 3 3+Coefficient u, v, r, ψ u, v, ψ u, v, ψ ψ, �

Yv’ -0.0261 -0.0255 -0.0257 Nv’ -0.0141 -0.0061 -0.0145 Yr’ 0.0037 0.0034 0.0036 Nr’ -0.0048 -0.0025 -0.0066

Figure 8.11: Yaw moments obtained through SImethod using different sets of parameters

8.4 Shallow Water Force Data

Force data and coefficients for operation inwater of finite depth (water depth-draft ratios

34

of 2 to 1.2) were available from many sources.This comparison was limited only to lineardamping coefficients. This is appropriate forthe slower responses experienced in shallowwater.

Figures 8.12 and 8.13 compare the variationof two linear damping coefficients with waterdepth (ratio of coefficient at finite water depthto that in deep water) Table 8.7 presents statis-tics (mean values and standard deviations orSTD.) of each coefficient and the ratio Nv’/Yv’(measure of static yaw or weathervane stabil-ity). There are increasingly large variations inthe force ratios derived from this data as waterdepth is reduced.

Table 8.7: Statistics of shallow water forcecoefficients from 16 sources

H/T=1.50Coefficient Mean Std. Std./MeanYv’/Yv’∞ 2.270 1.652 0.73Nv’/Nv’∞ 2.233 0.896 0.40Yr’/Yr’∞ 0.914 0.874 0.96Nr’/Nr’∞ 1.324 0.456 0.34Nv’/Yv’ 1.239 0.776 0.63

H/T=1.20Coefficient Mean Std. Std./MeanYv’/Yv’∞ 5.238 2.818 0.54Nv’/Nv’∞ 3.798 2.203 0.58Yr’/Yr’∞ 1.847 0.587 0.32Nr’/Nr’∞ 1.745 0.693 0.40Nv’/Yv’ 0.769 0.344 0.45

Figure 8.12: Effect of water depth on swaydamping coefficient Yv'

Figure 8.13: Effect of water depth on yawdamping coefficient Nr'

9. TECHNICAL CONCLUSIONS

1. CFD has been shown to be useful in pre-dicting forces associated with manoeuvring andits use is becoming more widespread.

2. Work should be pursued on CFD ap-proaches to reduce the required amount of ex-periments

3. It has been demonstrated that when rollangle exceeds a specific value, this motion andits coupling with other modes of motion has tobe modelled in the simulation.

4. Sensitivity analysis performed withmathematical models for conventional shiptypes has proven useful in identifying the mostimportant terms in simulation models and de-termining their necessary level of accuracy.

5. Systematic investigations of the influenceof local hull geometry on forces are needed togenerate more accurate simulation methods.

6. Work needs to continue to develop moreaccurate simulation methods.

7. Some scale effects can be taken into ac-count in simulations using modular models byapplying corrections to hull resistance, and ef-fective wakes at propeller and rudder. Otherscale effects on hull forces may exist but arenot currently taken into account.

8. Validation of the hydrodynamic coeffi-cients used in simulation models requiresbenchmark data at model scale.

9. Validation of simulations, and of free-running model test results, require benchmarkdata at full scale.

10. Research is required to quantify scale ef-

35

fects on hydrodynamic forces acting on a ma-noeuvring vessel, particularly the lateral forcesand yawing moments.

11. Clarification is required on the scale ef-fects present in free-running models.

12. There is evidence that manoeuvres pre-dicted using free-running model tests on fullform ships can be subject to significant scaleeffects which are not yet well understood.

13. Research is needed to improve the accu-racy of practical prediction methods, includingnumerical methods, for squat.

14. Research has to be conducted to assessthe relevance of existing IMO criteria, particu-larly overshoot angles, to practical ship ma-noeuvring performance. The development ofcriteria which best quantify ship manoeuvringperformance should be pursued.

15. Comparison of trials data for a largenumber of newer ships with the IMO standardsindicated that a large majority of these shipsmet all IMO performance criteria. The criteriamost frequently not met are the overshoot an-gles in zig-zag manoeuvres and the reach in acrash stop.

16. Differential Global Positioning Systemmeasurements of squat provide a new source ofdata for validation of squat predictions.

17. New methods have been proposed andapplied to reduce the number of tests requiredto obtain a complete set of hydrodynamic coef-ficients. The general applicability of thesemethods has to be demonstrated.

18. Current practice for captive model tests,and in particular PMM tests, shows a widespread in the relative size of model and facility.

19. The choice of parameters used in PMMtests, particularly the frequency of oscillation,often does not follow published guidelines.

20. Based on proposed guidelines for captivemodel tests and described uncertainty analysis,the number of sources of uncertainty in captivemodel tests is significantly greater than typi-cally considered (i.e. in resistance tests).

21. The accuracy of test results, in particularfor PMM tests, depends to a great extent on thephysical mechanism, the choice of test pa-rameters, and the method of the analysis.

22. Based on the comprehensive collection ofexisting and new results, surprisingly largedifferences exist in published hull forces and

moments for the ESSO OSAKA for both deepwater and shallow water.

23. The available data for the ESSO OSAKA inits current form is not suitable for use asbenchmark .

10. RECOMMENDATIONS TO THECONFERENCE

1. Adopt the procedure on captive modeltests (ITTC Procedure 4.9-03-04-03)

2. Standardised precision limits should beprovided with both predictions and full scaleresults. “Good agreement” should be definedbased on these precision limits.

3. A systematic validation procedure, suchas the one outlined in figure 5.4 for simulationmodels, should be applied and presented withthe results of simulations.

11. RECOMMENDATIONS FOR FU-TURE WORK

1. Model test procedures should be devel-oped for free-running model tests.

2. The committee strongly recommends thatthe Esso Osaka benchmark data effort be con-tinued in the following five areas:• Reduce the scatter in existing data either

by eliminating suspect data sets, or bystimulating new, benchmark quality ex-periments.

• Compare propeller and rudder forces andpropeller-hull-rudder interactions.

• Carry out a systematic series of simula-tions using one reference mathematicalmodel (e.g. MMG with fixed propeller andrudder forces and interactions) using avail-able sets of hull damping coefficients(linear and non-linear).

• Compare the results of these systematicsimulations with available track data andparticularly the full scale trials data.

• Promote the disclosure of benchmark datathrough the organisation of a workshop

3. Work should be pursued to define thecritical roll angle values above which 4 DOFare needed in the simulation.

4. Work should be pursued on the modellingof a ship manoeuvring in a non-uniform orshear current.

5. Trials data for all ship types at fully load-

36

ed condition should be collected to supportevaluation of the IMO standards.

6. The lack of benchmark data for all ma-noeuvring problems needs to be addressed byconducting suitable (free-running and captive)benchmark quality tests at model and full scale.

REFERENCES

Abkowitz, M., 1981, “System Identification inShip Maneuvering Trials : Analysis of theESSO OSAKA Trials Data, MIT Dept. ofOcean Engineering Report 81-9.

Agdrup, K., Jensen, A.G., Aage, C., 1998,"Wind Tunnel Assessment of Ship Manoeu-vrability using PMM Technique", MAN’98,Val de Reuil, France, pp. 1-9F.

Alessandrini, B., Delhommeau, G., 1998, “Vis-cous Free Surface Flow Past a Ship in Driftand Rotating Motion”, 22nd Symposium onNaval Hydrodynamics, Washington, D.C.,USA.

Ando, J., Yamamoto,T., Maita, S., Nakatake,K., 1997, “An Estimation of HydrodynamicForces Acting on a Ship in Oblique Towingby a Simple Surface Panel Method(SQCM)”, (in Japanese), TWCNA, No. 94,pp. 13-20.

Ankudinov, V., and Miller, E., 1979, “PredictedManeuvering Characteristics of the TankerEsso Osaka in Deep and Shallow Water,”Hydronautics Report 778-1.

Ankudinov, V., Daggett, L., et al, 1996, “SquatPredictions for Manoeuvring Applications”,MARSIM'96, Copenhagen, Denmark,Vol.1, pp. 467-495.

Ankudinov, V.K., Jakobsen, B.K., 1999, "SquatPredictions at an Early Stage of Design",Workshop on Ship Squat in Restricted Wa-ters, SNAME Technical and Research Bul-letin.

Bailey, P. A., Price, W.G., Temarel, P., 1995,“The Dynamic Stability and Manoeuvringof a Ship in a Sea-way”, PRADS’95, Seoul,Korea, pp. 1.479-1.493.

Bailey, P. A., Price, W.G., Temarel, P., 1997, “AUnified Mathematical Model Describingthe Manoeuvring of a Ship Travelling in aSeaway” Trans. RINA, W6.

Bailey, P. A., Hudson, D.A., Price, W.G.,Temarel, P., 1998a, « Theoretical and Ex-perimental Techniques for PredictingSeakeeping and Manoeuvring Ship Char-acteristics », International Conference onShip Motions & Manoeuvrability, London,England, RINA.

Bailey, P. A., Hudson, D.A., Price, W.G.,Temarel, P., 1998b, «The Measurement andPredicition of Fluid Action Experienced bya Manoeuvering vessel », MAN’98, Val deReuil, France, pp. 117-123.

Barr, R. A., 1993, “A Review and Comparisonof Ship Maneuvering Simulation Methods,”Transactions of the SNAME, Vol. 101. pp.609-671.

Benvenuto, G., Carrera, G. et al, 1996, “A Dy-namic Diesel Engine Model for Ship Ma-noeuvrability Prediction Studies”,MARIND’96, P.A. Bogdanov, Procs. Vol.III, pp. 393-404.

Berlekom, W., B., 1992, “Experimental Ma-noeuvring Facilities”, Hydrodynamics:Computations, Model Tests and Realty,Proceedings of MARIN Workshops,Wageningen, The Netherlands, pp. 533-537.

Bertram, V., Grollius, W., 1994, “Shallow watereffects for SWATH ships.”, Ship Technolo-gy Res, Vol. 41 n° 2, pp. 55-57

Beukelman, W., 1997, “Fluid Momentum inShip Hydrodynamics”, Ship en Werf deZee, Vol. 7, No.3, pp. 30-36

Bigot, E., 1997, « Etude Expérimentale et nu-mérique des efforts hydrodynamiques sur letanker Esso Osaka », Study E1009 Reportno.1 and 2, Bassin d’essais des carènes

Berth, F., Bigot, E., Laurens, J-M., 1998,« Numerical Simulation on the EssoOsaka », MAN’98, Val de Reuil, France,pp. 109-116.

37

Biancardi, C.G., 1997, “An Alterative Method-ology for Calculating Ship ManoeuvringCoefficients at the Design Stage”, Interna-tional Shipbuilding Progress, Vol. 44, No.440, pp. 273-297.

Blendermann, W., 1995, “Estimation of WindLoads on Ships in Wind with a Strong Gra-dient”, Copenhagen, Denmark, Vol. 1, PtA,pp. 271-277.

Blendermann, W., 1996, “Wind Loads of Ships– Collected Data from Wind Tunnel Tests inUniform Flow”, Institut fuer Schiffbau derUniversitaet Hamburg, Report No. 574.

Blendermann, W., 1997, “Overtaking Manoeu-vres of Ships in Beam Wind”, Ship Tech-nology Research, Vol. 44, pp. 193-197.

Blok, J.J., Quadvlieg, F.H.H.A., Toxopeus,S.L., 1998, "Computerised PMM approachfor ESSO OSAKA", MAN’98, Val de Reuil,France, pp. 101-107.

Bogdonov, P., et al, 1987, “Esso Osaka TankerManoeuvrability Investigation in Deep andShallow Water,” Intl. Shipbuilding ProgressVol. 34, No. 340.

Brizzolara, S., Grossi, L., Chislett, S., 1998,“Course Keeping Aspects in the Design ofFast Deep V Monohulls”, InternationalConference on Ship Motions & Manoeu-vrability, London, England, RINA, PaperNo.6

Caccia, M., Indiveri,G., Tiano,A., Veruggio, G.,1997, “Experimental Comparison of Iden-tification Methods for an Open FrameROV”, MCMC’97, Brijuni, Croatia, pp. 16.

Capurro, G., Sodomaco, P., 1996, “Applicationsof the Manoeuvring Prediction ProgramSIMSUP to Meet the New IMO Stran-dard », MARSIM’96, Copenhagen, Den-mark, pp. 563-576.

Caux, C., Jean, P., 1996, "Neural NetworksApplied on Identification of Ship Motions",MARSIM'96, Copenhagen, Denmark, pp.577-590.

Chau, S. W., 1998, “Computation of RudderForces and Moment in Uniform Flow”,Ship Technology Research, Vol. 45.

Chen, M., Chen, H.C., Davis,D.A., 1996,“Numerical Simulation of Transient FlowsInduced by a Berthing Ship”, ISOPE 96,Los Angeles, USA, Vol. IV, pp. 533 - 542.

Chislett, M.S., 1996, “A Generalized MathModel for Manoeuring”, MARSIM 96, Co-penhagen, Denmark, pp. 593-606.

Chiu, F. C., Guo, J., Huang, C. C., Wang, J. P.,1997, “On the Linear Hydrodynamic Forcesand the Manoeuvrability of an UnmannedUntethered Submersible with StreamlinedBody (2nd Report: Lateral Motion)”,JSNAJ, Vol. 182, pp. 151-159.

Clarke, D., 1997, “The Shallow Water Effecton Linear Derivatives” Proceedings ofMCMC’97, Brijuni, Croatia, pp. 87-92.

Clarke, D., Horn, J.R., 1997a, “Estimation ofHydrodynamic Derivatives”, 11th ShipControl System Symposium, Southampton,UK, Vol. 2, pp. 275-290.

Clarke, D., Horn, J.P., 1997b, “The Effects ofTrailing vortices and Stern Design on ShipManoeuvring”, IMDC’97, Newcastle uponTyne, UK. Vol.1, pp. 595-608.

Copchenov, V., 1998, Communication to theITTC Manoeuvring Committee, KrylovShipbuilding Research Institute.

Cortellini, L., Lauro, G., 1995, “Assessment ofShip Performance Using A DGPS System”,International Seminar on Differential GPSApplications to Ship Sea Trials, Genova, It-aly.

Cura Hochbaum, A., 1998, “Computation ofthe Turbulent Flow Around a Ship Model inSteady Turn and in Steady Oblique Mo-tion”, 22nd Symposium on Naval Hydrody-namics, Washington, D.C., USA

Dand, I., W., 1992, “Measurement Errors andManoeuvring Prediction”, Workshop onPrediction of Ship Manoeuvrability, Fu-kuoka, TWSNA, pp. 1-16.

38

Dand, I.W., 1999, "Prediction of squat at sub-,trans- and supercritical speeds for dis-placement ships in shallow water", Work-shop on Ship Squat in Restricted Waters.SNAME Technical and Research Bulletin.

Dand, I., and Hood, D., 1983, “ManoeuvringExperiments Using Two Geosims of theEsso Osaka,” NMI Report R 163.

Ding, Y., Min, M., 1997, “Experimental Studyon Rudder Effect Decreasing of a Ship inFollowing Seas’, (in Chinese), Shipbuildingof China, No. 138, pp. 23-27.

Dumlu, D., Istefanopulos, Y., 1995,“Design ofan Adaptive Controller for Submersibles viaMulti-modal Gain Scheduling.”, OceanEng., Vol. 22, No. 6, pp. 953-975.

Eda, H., 1975, “The Mariner Model Co-operative Test Program – Correlations andApplications,” Appendix V, Report of theManoeuvrability Committee, FourteenthITTC, Vol. 2, pp. 414-427.

Eda, H., 1983, “Studies on Vessel ManeuveringCharacteristics - Shallow Water Effects,”Davidson Laboratory Report SIT-DL-79-9-2012.

Eloot, K., Vantorre, M., 1998, "AlternativeCaptive Manoeuvring Tests: Possibilitiesand Limitations", MAN’98, Val de Reuil,France, pp. 89-98.

Endo, M., Kobayachi, H., Murayama, Y., 1996,“Ability of Berthing Assisted by Joy Stickcontroller.”, MARSIM '96,. Copenhagen,Denmark, Vol. 1, pp. 349-355.

Endo, M., Yokoi, K., Tachi, K., Kobayashi, H.,Takeda, M., 1997, “Estimated Manoeu-vrability of T.S. Wakashio – maru – II –Side Thruster Model Tests –“ (in Japanese)JJIN, Vol. 96, pp. 185-193.

Feng, X., Miao, Q., Jiang, Q., 1996, “Predic-tion of Non-linear Motion Responses of aSubmerged Slender body Running near FreeSurface due to Wave Exciting Forces”, (inChinese), Shipbuilding of China, No. 134,pp. 29-35.

Fukuchi, A, Fukutani, T., Hori, A., Tanaka, S.,1997, “Effects of a Canvas on the CourseKeeping Ability of a Small Vessel Scuddingunder Strong Wind”, (in Japanese), JJIN,Vol. 97, pp. 183-190.

Furukawa, Y., Kijima, K., 1996, “Study on Pre-diction Method for Hydrodynamic ForcesActing on a Ship Hull in Sway Motion(Continued)”, (in Japanese), JSNAJ, Vol.180, pp. 235-240.

Galor, W., 1997, “The Navigation Tests ofDesigning the Entrance Into the Port ofLeba”, 2nd International Conference onMarine Technology, ODRA’ 97, Szczecin,Poland.

Gong, I.., Kim, Y., Lee, C.-, Kim, J., Lee, H.,1996, “Development of harbor capabilityassessment simulation system by the appli-cation of fuzzy algorithm”, MARSIM '96;Copenhagen, Denmark, pp. 9-13.

Gong, I.Y., Kang, C.G., Kim, Y.G., You, Y.B.,Huh, D., Kim, Y.H., 1995, “The Influenceof rudder Area on the Manoeuvrability of aShip with Large Beam to Draft Ratio”,PRADS’95, pp. 1.700-1.711.

Gronarz, A., 1995, “A Numerical Simulation ofthe Ship’s Motion in Manoeuvres with Spe-cial Consideration of the Dependency of theWater Depth”, (in German), Ph.D. Thesis,Gerhard Mercator Universität, Duisburg.

Gronarz, A., 1988, “Manoeuvring Behavior ofa Full Ship in Water of Limited Depth,”VBD (Duisburg, Germany) Report 1225.

Hamamoto, M., Enomoto, T., 1997, “Manoeu-vring Performance of a Ship with Vec TwinRudder System”, JSNAJ, Vol. 181, pp. 197-204.

Hasegawa, K., Sasaki, Y., 1997, “Java-basedsimulation tool for ship manoeuvring”,MCMC 97, Brijuni,Croatia.

Hewlett, C., 1999, "Methods for MeasuringSquat in the Field", Workshop on ShipSquat in Restricted Waters. SNAME Tech-nical and Research Bulletin.

39

Hilgert, H., Baldauf, M., 1996, “A commonrisk model for the assessment of encountersituations on board ships”, Maritime Colli-sion and Prevention, The Hill Taylor Dick-inson Seminar on Maritime Collision,Vol.2, p. 136.

Hinatsu, M., Hino, T., et al 1995, “NumericalSimulation of Flow around a Ship Hull withRudder in Self Propulsion Condition”, (inJapanese), TWSNA, No. 90, pp. 1-10

Hirano, M., 1981, “A Practical CalculationMethod of Ship Maneuvering Motion atInitial Design Stage”, Naval Architectureand Ocean Engineering (published by theSociety of Naval Architects of Japan), Vol.19, pp. 68-80.

Hirano, M., et al, 1985, “An ExperimentalStudy on Maneuvering HydrodynamicForces in Shallow Water,” Transactions ofWJSNA, Vol. 60.

Hirata, N., Kose, K., Takase, S., Nakamura, Y.,Terada, K., Ito, R., 1997, “Applications of aKinematic GPS to Ship Maneuvering Tri-als” (in Japanese), TWSNA, No. 94, pp.111-119.

Hirayama, T., Niihara, Y., 1996, “Improvementof Transverse Stability and Manoeuvrabilityof Displacement Type High Speed Mono-Hull Ship by an Active Vertical Fin”, HighSpeed Marine Craft, No. 20, pp. 1-21.

Hirayama, T., Niihara, Y., Saito, Y., 1996a “In-troduction of Anti-Rolling Active VerticalFin and its Application to Manoeuvrabilityfor Displacement-Type Super High SpeedShip (2nd report)” (in Japanese), JSNAJ,Vol. 180, pp. 251-262.

Hirayama, T., Niihara, Y., Nishimura, K.,1996b, “Introduction of Anti-Rolling ActiveVertical Fin and its Application to Manoeu-vrability for Displacement Type Super HighSpeed Ship”, (in Japanese), JSNAJ, Vol.178, pp. 329 - 339.

Hiroshima, F., Miyata, H., Akimoto, H., 1997,“Design Method for Sailing Boats by CFDPerformance Simulation”, (in Japanese),JSNAJ, Vol.181, pp.15-23

Hooft, J.P. & Quadvlieg, F.H.H.A., 1996.“Non-Linear Hydrodynamic Hull ForcesDerived from Segmented Model Tests”,MARSIM ’96, Copenhagen, Denmark, Vol.1. pp. 399-409.

Huff, L.C., 1999, "On-the-fly GPS for VerticalControl of Hydrographic Surveys", Work-shop on Ship Squat in Restricted Waters.SNAME Technical and Research Bulletin.

Imazu, H., 1995, “The effect of ship's speed oncollision avoidance”, FAST '95, LubeckTravemunde, Germany. Procs, Ed by C.F.L.Kruppa. Publ by SchiffbautechnischeGesellschaft, Germany. Vol. 1, pp. 393-405.

Ishibashi, A., Kobayshi, H., Ugajin, T., 1996,“A Study on Ship Manoeuvrability Charac-teristics in Shallow Water On the HarborManoeuvring at Low Speed Range”, (inJapanese), JJIN, Vol.95, pp. 371-380.

Ishida, S., Fujiwara, T., 1995, “Measurement ofHydrodynamic Sway Force Acting onTwodimensional Models of Small Ship”, (inJapanese), JSNAJ, Vol.178, pp. 147-159.

Ishiguro, T., Tanaka, S., Yoshimura, Y., 1996,“A Study On the Accuracy of the RecentPrediction Technique of Ship’s Manoeu-vrability at Early Design Stage”, MAR-SIM’96, Copenhagen Denmark, pp. 547-561.

Ishioka. Y., Nakamura, S., et al, 1996, “A Su-tudy on a Supporting of Decision-makingfor Collision Avoidance in INS.”, MARSIM'96, Copenhagen, Denmark, Vol. 1, pp. 49-58.

Jakobsen, B.K., Miller, E.R., Wulder, J. H.,Hensen, H., 1996, “The Simulation of TugOperations in a Multiple Simulator Envi-ronment”, MARSIM’96, Copenhagen,Denmark, pp. 29-36.

JAMP report, 1985, “Prediction of Manoeu-vrability of a Ship,” Bulletin No. 668 ofJSNAME.

Jian, T., Sharma, S.D., Chen, X-N., 1995, “Onthe Wavemaking Resistance and Squat of

40

Catamaran Moving at High Speed in aShallow Water Channel.”, FAST '95, Lu-beck-Travemunde, Germany. Procs. Ed byC.F.L. pp. 1313-1325.

Jiang, T., 1997, “Autonomous Oscillations andBifurcations of a Tug-Tanker Tow.”, ShipTechnology Res, Vol. 44, No.1, pp. 4-13.

Jiang, T., Henn, R., Sharma, S.,D., 1998, “Dy-namic Behaviour of a Tow System under anAutopilot on the Tug”, MAN’98, Val deReuil, France, pp. 183-193.

Jordan, K., 1989, Rudder Loads at EmergencyManoeuvres (Evaluation Procedure and Re-sults)”, Versuchsanstalt fuer Wasserbau undSchiffbau, Final Report, Part 1.

Jukola, H., Castleman, G.E., 1995, “Z-DriveEscort Tug Operating Modes”, MarineTechnology, Vol. 32, No. 3, pp. 193-196.

Kang, C-G., Kim, J-H., 1996, “Assessment ofShip Manoeuvrability Based on IMO Re-solution No.A.751(18)”, MARSIM’96, Co-penhagen, Denmark, pp. 293-302.

Kang, C.G., Kim, Y.G., 1995, “A PredictionMethod of Manoeuvrability Including theEffects of Side Profile at the Initial DesignStage”, PRADS, pp. 1.629-1.637.

Kang, C.,G, Gong, I.,Y., Kim, S., Y., 1998,“The Manoeuvrability of Full Form Shipswith Low Speed”, MAN’98, Val de Reuil,France, pp. 165-171.

Karasuno, K., Maekawa, K., 1996, “A Compo-nenttype Mathematical Model of Hydrody-namic Forces in Steering Motion Derivedby Simplified Vortex Model”, (in Japanese),JSNAJ, Vol. 179, pp. 139-151.

Karasuno, K., Maekawa, K., 1997, “A Compo-nenttype Mathematical Model of Hydrody-namic Forces in Steering Motion Derivedby Simplified Vortex Model (2) – Applied3D Cross Flow Model ”, (in Japanese),JSNAJ, Vol. 182, pp. 141-150.

Karasuno, K., Yasunaga, T., Maekawa, K., Ta-kano, K., Januma, S., 1996, "Measurementand Analyzed Results of Hydrodynamic

Forces Acting on a Ship in Oblique andTurning Motion by Means of a Very Com-pact Apparatus for Circular Motion Test, 1stReport"

Kataoka, H., Miyata,H., Kanai, A., 1997,“Design Method for Sailing Boats by CFDPerformance Simulation (2nd Report)”, (inJapanese), JSNAJ, Vol. 182, pp. 113-20.

Kijima, K., Furukawa, Y., 1995, “On a Studyfor Towing Method in Narrow Water”(inJapanese), TWSNA, Vol. 89, pp. 167-172.

Kijima, K., Nakira, Y. et al, 1995, “On a studyfor influence of loading condition on a pre-diction of ship manoeuvrability” (in Japa-nese), TWSNA, No. 89, pp. 155-166.

Kijima, K., Furukawa, Y., Tsukamoto, K.,1995a, “Study on Prediction Method forHydrodynamic Force Acting on a Ship Hullin Swaying Motion”, (in Japanese), JSNAJ,Vol. 178, pp. 273-280.

Kijima, K., Yukawa, K., Maekawa, K., 1995b,“An Estimation of Hydrodynamic ForcesActing on a Ship Hull in Oblique Motion”,(in Japanese), TWSNA, No. 90, pp. 115-125.

Kijima, K., Yukawa, K., Maekawa, K., 1996a,“An Estimation of Hydrodynamic ForcesActing on a Ship Hull in Oblique Motion(continued)”, (in Japanese), TWSNA, No.91, pp. 95-106.

Kijima, K., Yukawa, K., Maekawa, K., 1996b,“An Estimation of Hydrodynamic ForcesActing on a Ship Hull in Turning Motion”,(in Japanese), TWSNA, No. 92, pp. 105-116.

Kijima, K., 1997, “Prediction Method for ShipsManoeuvring Motion in the Proximity of aPier”, Ship Technology Research, Vol.44,No.1, pp. 22-31.

Kijima, K., Furukawa, Y., Eshiwa, M., Yama-moto, K., 1997, “Influence of Roll Motionon Ship Manoeuvrability”, (in Japanese),TWSNA, No. 93, pp. 35-46.

41

Kijima, K., Furukawa, Y., 1998, “Effect on rollmotion on manoeuvrability of ship”,MAN’98, Val de Reuil, France, pp. 157-163

Kim, C.K., Rhee, K.P., 1996, “Estimation ofManoeuvring Coefficients of a SubmergedBody Using Parameter Identification Tech-niques”, Journal of the Hydrospace Tech-nology, Vol. 2, No. 2, pp. 24-35.

Klosinski, W.E., Lewandowski, E.M., 1993,“Systematic Series Model Tests”, TechnicalReport SITDL9392689, Davidson Labora-tory, Stevens Institute of technology, USA.

Knowles, J., Grosenbaugh, M., Keenan, D.,1996, “Numerical Simulation of an AUVThruster during Manoeuvring”, ISOPE’96,Los Angles, USA. Vol. 2, pp. 247-250.

Kobayashi, E., 1995, “The development ofpractical simulation system to evaluate shipmanoeuvrability in shallow water”, PRADS95, pp. 1712-1723

Kobayashi, H., Arai,, Y. et al, 1995 “A study onthe manoeuvrability of a high-speed boat”(in Japanese), JKSNA, No. 223, pp. 81-89.

Kobayashi, H., Ishibashi, A., et al, 1994, “Astudy on a mathematical model for the ma-noeuvring motions of a twin propeller andtwin rudder ship” (in Japanese), JJIN, Vol.91, pp. 263-270.

Kodan, N., Yamasaki, K., et al, 1996, Predic-tion Procedure of Ship Manoeuvrability inConsideration with Hull Forms”, NKKTechnical Review, No. 75, pp. 37-40.

Kose, K., Kobayashi, H., et al, 1995b, “A studyon a supproting procedure for collisionavoidance in INS.”, JSNAJ, Vol. 177, pp.409-418.

Kose, K., Mislag, W.A., Hashizume, Y., No-gami, S., Ishioka, Y., 1995a, “A PracticalApproach To The Mathematical Model ofShip Maneuvering In Harbours”, Ship Ma-neuvering Symposium - Ilawa’ 95, Poland.

Kose, K., Mislag, W.A., Xiong, X.M., 1996,“Systematic Approach for Ship Manoeu-

vrability Prediction”, MARSIM’96, Copen-hagen, Denmark, pp. 535-545.

Kose, K., Xiong, X.M., Mislag, W.A., 1997, “ANew Prediction Approach for Ship Ma-noeuvring Hydrodynamics”, TWSNA, No.93, pp. 47-61.

Kreibel, D., Zieleski, J., 1990, “Evaluation ofViscous Damping Models for Single-PointMooring Simulations » US Naval AcademyReport, July 1990.

Kristensen, H. O., 1998, “The Manoeuvrabilityof Double-Ended Ferries Design, Consid-erations, Construction and Service Experi-ence” International Conference on ShipMotions and Manoeuvrability, London.

Kulczyk, J., Tabaczek, T., 1995, “Calculationof Influence of Propeller Slipstream onMarine Rudder Performance”, Ship Ma-noeuvring Symposium, Ilawa, Poland, Vol.1, pp. 164-168 .

Kwik, K.H., 1996, “Mathematical calculationof collision avoidance.”, Maritime Collisionand Prevention, The Hill Taylor DickinsonSeminar on Maritime Collision; Dalian,China. Procs. Publ by Chiavari Publishing,UK, ISBN 09520592 4x. Vol. 2, pp. 99-115.

Laforce, E., Claeyssens, P., Vantorre, M., 1996,“Influence of the Ship’s Length on the Ma-noeuvrability in a Canal”, Proceedings of11th International Harbor Congress, Ant-werpen, Belgium, pp .523-534.

Laforce, E., Vantorre, M., 1996, “Experimentaldetermination and modelling of restrictedwater effects on bulk carriers”, MARSIM96, Copenhagen, Denmark, pp. 451-464.

Laible, D.H., Gray, D.L., 1997, “Escort tugcomputer simulation - standards and ad-vances”, Tug & Salvage Convention, Seat-tle, Washington, USA 1996, Thomas ReedPublications, pp. 201-208.

Landrini, M., Campana, E.F., 1996, “SteadyWaves and Forces About a Yawing FlatPlate”, JSR, Vol. 40, No. 3, pp. 179-192.

42

Lech, R., 1998, Communication to the ITTCManoeuvring Committee - “Model Test ofEsso Osaka,” Centrum Techniki Okretowek,Gdansk, Poland.

Lee, H.Y., Shin, S.S., 1998, ”The prediction ofship’s manoeuvring performance in initialdesign stage”, Practical Design of Ships andMobile Units, The Netherlands.

Lee, S.K., Kim, Y.S., Lee, S.J., 1996, “Resear-ch on the Prediction of Maneuvring Motionfor a Twin-Screw Twin-Rudder Ship” (inJapanese), JSNAK, Vol. 33, No. 4, pp. 60-65.

Lewandowski, E.M., 1997, “Transverse Lee,H.Y., Yum, D.J., Shin, S.S., Lee, T.I., 1997,“Prediction of Manoeuvrability using PMMModel Tests - Comparative Study ofMathematical Models” (in Korean),JSNAK, Vol. 34, No. 2, pp. 48-55.

Dynamic Stability of Planing Craft”, MarineTechnology, Vol. 34, No. 2, pp. 109-118.

Lewandowski, E.M., Klosinski, W.E., 1992,“Rotating Arm and Linear Towing TankTests on a Mariner Model”, Technical Re-port SITDL9292679, Davidson Laboratory,Stevens Institute of Technology, USA.

Li, D.Q., Dyne G., 1995, “Study of PropellerRudder Interaction based on a LinearMethod”, International Shipbuilding Prog-ress, Vol. 42, No. 431, pp. 235-257.

Li, T., 1997, “Calculation of manoeuvring in-dex and frequency characteristics for sub-mersible” (in Chinese) Shipbuilding ofChina, No. 136, pp. 15-24.

Lisowski, J., Mohamed-Seghir, M., 1996,“Ship's optimum safe trajectory in a colli-son situation of passing many moving tar-gets in fuzzy environment.”, Polish Mari-time Res, Vol. 3, No. 4, pp. 29-32.

Liu, Z.Y., Zhang, X.D., Wu, X.H, 1996, “Cal-culation of Manoeuvring HydrodynamicForce including the Effect of Viscosity”,MARSIM’96, Copenhagen, Denmark, pp.419-424.

Logan, K.P., 1995, “Neural network-basedautopilot for tanker conning.”, Sea Tech-nology, Vol. 36, No. 4, pp. 51-54.

Longo, J, and Stern, F. , 1997, « Yaw effects onModel-Scale Ship Flow », Twenty-FirstSymposium on Naval Hydrodynamics,Trondheim, Norway, pp. 312-327.

Makino, M., Kodama, Y., 1997, “Computationof Flows around Full Hull Forms inOblique Towing or Steady Turning UsingNICE Code”, (in Japanese), JSNAJ, Vol.181, pp. 67-73.

McDonald, H. and Whitfield, D. 1997, « Self-propelled Manoeuvring Underwater Vehi-cles », 21st Symposium on Naval Hydrody-namics, Trodheim, Norway, pp. 478-489.

Milgram, J., 1995, “Extreme Tensions in OpenOcean Towing”, Journal of Ship Research,SNAME, Vol. 39, No. 4, pp. 328-346.

Miller, E., and Ankudinov, V., 1980, “ModelTest and Simulation Study Based on theFull Scale “Esso Osaka”ManeuveringTrial,“ Hydronautics Report 8007-1.

Millward, A., 1996, “A review of the predictionof squat in shallow water.”, JJIN, Vol. 49,No. 1, pp. 77-89.

Molland, A.F., Turnock, S.R., 1995, “SomeEffects of Rudder-Propeller-Hull Arrange-ments on Manoeuvring and Propulsion”,PRADS’95, Seoul, Korea, pp. 1.333-1.345.

Molland, A.F., Turnock, S.R., Wilson, P.A.,1996, “Performance of an enhanced rudderforce prediction model in a ship manoeu-vring simulator”, MARSIM 96, Copenha-gen, Denmark, pp. 425-434.

Molland, A.F., Turnock, S.R., 1996, “A Com-pact Computation Method for PredictingForces on a Rudder in a Propeller Slip-stream”, RINA, Spring Meeting, London,Paper No. 2.

Molland, A.F., Turnock, S.R., 1998, “The Ef-fect of the hull on the Manoeuvring Perfor-mance of Rudders”, International Confer-

43

ence on Ship Motion & Manoeuvrability,London, RINA, Paper No. 8.

Nakao, I., Iseki, T., Ohtsu, K., 1995, “The Ef-fect of Heading Angle on Ship Manoeu-vring in Confined Water”, (in Japanese),JJIN, Vol. 93, pp. 171-178.

Nakatake, K., Yoshitake, A., Shimomichi,Y.,Yamamoto, T., 1995, ”Measurement ofWake Distributions Behind Ships with YawAngle”, (in Japanese), TWSNA, No. 90, pp11-20.

Nakatake, K., Ando,J., Ohmori, T., Kataoka,K., 1996a, “A Method to Calculate Interac-tion Forces during Rudder Angle Tests”, (inJapanese), TWSNA, No. 91, pp. 71-80.

Nakatake, K., Oda, K., Yoshitake,A., Fujita, K.,Nakajima, A., 1996b, “Experimental Studyon Interaction Forces during Rudder AngleTest”, (in Japanese), TWSNA, No. 92,pp.95-103.

Nakatake, K., Ando, J., Yoshitake, A., Hashi-moto, K., 1997, “Hull Propeller Rudder In-teractions during Rudder Angle Test”, In-ternational Conference, NAV & HSMV’97,SorrentoHotel Cocumella, pp. 361-368.

Nakatake, K., Ando, J., Maita, S., Yamamoto,T., 1998, “Prediction of Forces Acting onShip with Different Afterbody Shapes inOblique Towing”, MAN’98, Val de Reuil,France, pp. 173-180.

Nakatani, T., Ohtsu, K., et al, 1996, “A studyon automatic turning of ship's PID regula-tors.”, JSNAJ, Vol. 179, pp. 231-238.

Nonaka, K, Nimura, T., Haraguchi, Ueno, M.,1996, “Measurements of Stern Flow Fieldof a Ship in Oblique Towing Motion”,MARSIM’96, Copenhagen, Denmark, pp.389-398.

Norrbin, N. H., 1998, “ On the stopping ofships, including the proposal for a newstandards for the crash-stop-astern capabili-ties”, Proc. International Conference onShip Motions & Manoeuvrability, London,pp.

Norrbin, N.H., 1999, "The effects of flow con-finement and asymmetry on a ship in afairway passage", Workshop on Ship Squatin Restricted Waters. SNAME Technicaland Research Bulletin.

Norrbin, N.H., 1996, “Further notes on the dy-namic stability parameter and the predictionof manoeuvring characteristics”, MAR-SIM’96, Copenhagen, Denmark, A.A.Balkema, Rotterdam, Netherlands, pp. 527-531.

Oda, K., Yamashita, S., Ichinose, T., Nakatake,K., 1996, “On Normal Force and Drag ofMariner Rudder”, (in Japanese), TWSNA,No. 91, pp. 15-23.

Ogawa, A., et al, 1977, “Report on MMG, I,Mathematical Models of ManoeuvringMotions,” Bulletin No. 575, SNAJ.

Ogawara, Y., Iwamoto, S. Yoshimura, G., 1995,“Basic study on the compensation controlsystem of the wind disturbance influence tothe ship manoeuvring motion with thelearning feed-forward control”, JSNAJ, Vol.178, pp. 321-329.

Ohmori, T., 1998, “A Study on HydrodynamicCharacteristics of a Manoeuvring Ship inShallow Water by a FiniteVolume Method”,MAN’98, Val de Reuil, France, pp. 15-20.

Ohmori.T., Fujino, M., Tatsumi, K., Kawa-mura, T., Miyata, H., 1996, “A Study onFlow Field Around Full Ship Forms in Ma-noeuvring Motion (3rd Report: Flow FieldAround Ship’s Hull in Steady Turning Con-dition)”, (in Japanese), JSNAJ, Vol. 179,pp. 125-138.

Okamoto, H., Tamai, H., Oniki, H., 1972,“Correlation Studies on Manoeuvrability ofFull Ships” (in Japanese), JSNAJ, Vol. 131,pp. 189-202.

Oltmann, P. , 1998, « Reflections on the As-sessement of the Manoeuvring Behavior ofShips », International Conference on ShipMotions & Manoeuvrability, London, Eng-land, RINA, Paper No. 10.

44

Oltmann, P. , 1996, “On the influence of speedon the manoeuvring behaviour of a contain-er carrier”, MARSIM’96, Copenhagen,Denmark, pp. 527-531.

Oltmann, P., Wolff, K., Muller, E., Baumgarten,B., 1986, “On the Correlation betweenModel and Full-Scale Ship for Manoeu-vring Tests in Deep and Shallow Waters”(in German), Jahrbuch STG, Vol. 80, pp.217-245.

Pawlowski, J.S., 1996, “Hydrodynamic model-ling for ship manoeuvring simulation”,MARSIM’96, Copenhagen, Denmark, A.A.Balkema, Rotterdam, Netherlands, pp. 625-634.

Plante, M., Toxopeus,S. , Blok, J., Keuning, A.1998, « Hydrodynamic manoeuvering as-pects of planing craft », MAN’98, Val deReuil, France, pp. 147-155.

Perdon, P., 1998, "Rotating arm manoeuvringtest and simulation for waterjet propelledvessels", MAN’98,Val de Reuil, France,pp.41-46.

PIANC/IAPH, 1997, "Approach Channels - AGuide for Design", PTC II-30:Final reportof joint Working Group PIANC-IAPH II, incooperation with IMPA and IALA, Supple-ment to PIANC Bulletin No. 95.

Pinziy,B., Ba, M., Guilbaud, M., 1995, “Hy-drodynamic Force Computations on Yawedand Surfacepiercing Bodies”, ISOPE’95,The Hague, The Netherlands, Vol. 3, pp.308-313.

Pourzanjani, M., “A Study of collision avoid-ance vehaviour, using a PC based simula-tor.”, Maritime Collision and Prevention,The Hill Taylor Dickinson Seminar onMaritime Collision, Dalian, China. Procs.Publ by Chiavari Publishing, UK, pp .50-60.

Queensland Transport, 1996, “QueenslandTransport Transport Technology Kivision -Document No R-007 Squat measurementsin the Port of Cairns”, BOOK ; QueenslandTransport Place: Queensland, AustraliaFormat: REPORT

Raestad, A., E., 1996, “A Classification Soci-ety’s Experience with IMO ResolutionNo.A.751(18)”, MARSIM’96, Copenhagen,Denmark, pp. 309-318.

Rhee, K.-P., Lee, S.Y., Sung, Y.J., 1998, "Esti-mation of manoeuvring coefficients fromPMM test by genetic algorithm", MAN’98,Val de Reuil, France, pp.77-87.

Roseman, D., 1987, The MARAD SystematicSeries of Full-Form Ship Models, TheSNAME, New York.

Rossignol, G., 1995, "Outdoor radio-controlledmaneuvering experiments", 24th ATTC,Session 9, pp. 323-330.

Rooij, h. van, 1996, “OPA ’90 - response orprevention ? The European perspective”,Tug & Salvage Convention, Seattle, Wash-ington, USA, Thomas Reed Publications,pp. 209-212.

Sadakane, H., 1996, “A Study on Lateral DragCoefficient for Ship Moving Laterally fromRest”, (in Japanese), JJIN, Vol. 95, pp 193-200.

Sahin, I., Crane, J.W., Watson, K.P., 1997,“Application of a Panel Method to Hydro-dynamics of Underwater Vehicles”, OceanEngineering, Vol. 24, No. 6, pp. 501-512.

Sato, S., Nakamura, I., Takagi, M., Jono, T.,Yamada, K., Hikino, M., Takai, T., Yao, H.,1973, “On a Study of Ship-Controllabilityof a Wide-Beam Tanker” (in Japanese),JSNAJ, Vol. 134, pp. 205-222.

Sato, T., Izumi, K., Miyata, H., 1998, “Numeri-cal Simulation of Maneuvering Motion”,22nd Symposium on Naval Hydrodynamics,Washington, D.C., USA.

Shiraka, A., 1997, Communication to ITTCManoeuvring Committee, “Collection ofthe Manoeuvring Model Test Data on theEsso Osaka, NKK Corporation.

Schmidli, C., 1996, “Reducing the stoppingdistance of deep sea vessels”, Institute fürSchiffbau der Universität Hamburg, ReportNo. 562.

45

Sedat, R., Fuller, N., 1995, “U.S. Coast GuardModular Model Manoeuvring Tests”, 24thATTC, Session 9, pp. 309-315.

Shigehiro, R., Ueda, K., Arii, T., 1996, “CourseStability of the High-Speed-Towed FishPreserve” (in Japanese), JKSNA, No. 227,pp. 79-87.

Shigehiro, R., Ueda, K., Arii, T., Nakayama,H.,1997, “Course Stability of the High-Speed-Towed Fish Preserve with Wind Ef-fect” (in Japanese), JKSNA, No. 228, pp167-174.

Sohn, K.H., Lee, G.W., Kim, H.S., Kim, Y.S.,Ha, M.K., 1996, “A study on the Effect ofRudder Area with Reference to Changes inSpan Distance on Course Stability of aShip” (in Korean), Transactions of theSNAK, Vol. 33, No

Sowdon, A., 1996, “CFD Computations of Hy-drodynamic Forces Acting on Full HullForms in Oblique Tow”, Ship Research In-stitute Japan Papers, Vol. 33, No. 1, pp. 1-34.

Spyrou, K., 1995, “Yaw stability of ships insteady wind”, Ship Technology Research,Vol. 42, No. 1, pp. 21-30.

Spyrou, K., 1996, “Dynamic Instability inQuartering Seas: The behavior of a ShipDuring Broaching”, Journal of Ship Re-search, SNAME, Vol. 40, No. 1, pp. 46-59.

Spyrou, K., 1997, “Dynamic Instability inQuartering Seas- part III: non-linear effectson periodic motion”, Journal of Ship Re-search, SNAME, Vol. 41, No. 3, pp. 210-225.

Stenson, R., J., 1995, “Use of the Global Posi-tioning System for the Conduct of US NavyPerformance and Special Trials”, Interna-tional Seminar on Differential GPS Appli-cations to Ship Sea Trials, Genova, Italy.

Sutton, R., Marsden, G.D., 1997, “A fuzzyautopilot optimized using a genetic algo-rithm.”, JJIN, Vol. 50, No. 1, pp. 120-132.

Sutton, R., Taylor, S.D.H. Roberts, G.N., 1996,“Neuro-fuzzy techniques applied to a shipautopilot design.”, JJIN, Vol. 49, No. 3, pp.410-431.

Sutulo, S., Kim, S.Y., 1998, "Systematic ap-proach to PMM/rotating arm experimentplanning, parameter estimation, and uncer-tainty analysis", MAN’98, Val de Reuil,France, pp. 57-67.

Sutulo, S., Kim, Y.-G., 1997, “Structure ofMathematical Model for Description of Hy-drodynamic Forces and Moments ActingUpon the Hull of a Submersible in Arbitrary3D Manoeuvring Motion”, Proceeding ofthe Annual Spring Meeting, SNAK, Chin-hae 1997.

Suzuki, H., Ichihawa, M., et al, 1996, “Com-putation of a Flow Field around a Rudderbehind a Propeller”, (in Japanese), JKSNA,No. 225, pp. 15-24.

Suzuki, K., Yoshihara, A., 1995, “Computationof Hydrodynamic Forces Acting on SailingYacht by Means of Surface Panel Method”,(in Japanese), JSNAJ, Vol. 178, pp. 113-124.

Suzuki, O., Nakamura, T., Namie, H., Yasuda,A., 1998, “Real Time Anchor Watch byGPS” (in Japanese), JJIN, Vol. 98, pp. 25-32.

Tachi, K., Endo, M., 1996, “Estimated Ma-noeuvrability of T.S. Wakashiomaru WindTunnel Test and Rudder Open Test”, (inJapanese), JJIN, Vol. 94, pp. 17-26.

Tahara, Y., 1995, “Computation ofBoundarylayer and Wake Flows aroundIACC Sailing Yacht For a Canoe BodyCase ”, JKSNA, No. 224, pp. 1-11

Tahara, Y., Longo, J., Stern, F., Himeno, Y.,1998, “Comparison of CFD and EFD forthe Series 60 Cb=0.6 in Steady Yaw Mo-tion”, 22nd Symposium on Naval Hydrody-namics, Washington, D.C., USA

Tahara,Y., 1996, “A Multi-Domain Method forCalculating Boundary layer and Wake

46

Flows around IACC Sailing Yacht”,JKSNA, No. 226, pp. 63-76.

Takahashi, H., Hagiwara, H., et al, 1996,“Study on support systems of entering aport for techno super liner.”, JJIN, Vol. 95,pp. 71-79.

Takano, K., Thamby, A., Karasuno, K.,Maekawa, V., Jyanuma, S., 1995, “Funda-mental Study by Experiment on Ship Hy-drodynamic Forces in Oblique Motion Af-fected by Bow and Stem Shape”, (in Japa-nese), JKSNA, No. 224, pp. 59-67.

Takase, S., Kose, K., Hirata, N., Furumoto, M.,Minh, L-M., Yamada, T., 1997, “Accuracyand Application for Measurement of ShipMotion by Kinematic GPS” (in Japanese),JJIN, Vol. 97, pp. 15-22.

Tamashima, M., Yang, J.M., Matsui, S., Yama-zaki, R., 1995, “The Performance of Pro-pellerRudder System with Rudder Angle”,Proceedings of PRADS’95, Seoul, Korea,Vol. 1, pp. 1.321-1.332

Tanaka, S., 1996, “On Hydrodynamic ForcesActing on a Ship at Large Drift Angles”, (inJapanese), TWSNA, No. 91, pp. 81-94.

Tanaka, S., 1998, “On Hydrodynamic ForcesActing on a Ship Moving with Large DriftAngles, MAN’98, Val de Reuil, France, pp.31-38.

Trägärdh, P., 1998, “Manoeuvring of a twinscrew tanker including engine and rudderfailure modes”, Symposium on forcesacting on a manoeuvring vessel, Val deReuil, France.

Ueno, M., Nimura, T., Fujiwara, T., Nonaka,K., 1997, “Evaluation of RTK-OTF Posi-tioning System for Free Running Manoeu-vrability Test of a Model Ship” (in Japa-nese), JKSNA, No. 228, pp. 113-121.

Vantorre, M. ,Eloot, K., 1996, “Hydrodynamicphenomena affecting manoeuvres at lowspeed in shallow navigation areas”, 11thInternational Harbour Congres, Antwerpen,Belgium, pp 535-546.

Vantorre, M., 1999a, "A review of practicalmethods for prediction of squat", Workshopon Ship Squat in Restricted Waters.SNAME Technical and Research Bulletin.

Vantorre, M., 1999b, "Influence of fluid mudlayers on squat effects", Workshop on ShipSquat in Restricted Waters. SNAME Tech-nical and Research Bulletin.

Vantorre, M., Eloot, K., 1997, "Requirementsfor standard harmonic captive manoeuvringtests", MCMC'97, Brijuni, Croatia, pp. 93-98.

Varyani, K. S., Incecik,A., Hannah, S., 1995,“Manoeuvring Simulations of BritishBombadier, Mariner and Frigate Forms,”Report NAOE-95-19, and “Shallow WaterManoeuvring Simulation of ESSO OSAKAand BRITISH BOMBADIER,” ReportNAOE-95-3, University of Glasgow.

Varyani, K.S., Hamoudi, B., McGregor, R.,1997, “Interactive Forces between ThreeShips in a Restricted Waterway”, Proceed-ings of MCMC’97, Brijuni, Croatia, pp.127-133.

Vassalos, D., Park, S.H., Lee, B.S., 1995, “De-veloping a manoeuvring design capabilityfor naval vessels”, PRADS‘95, pp. 1616-1628.

Wang, D., Zhu, S., Jia, X., 1996, “Design andresearch on an intelligent collision avoid-ance system.”, Maritime Collision and Pre-vention, The Hill Taylor Dickinson Seminaron Maritime Collision; Dalian, China.Procs. Publ by Chiavari Publishing, Vol. 2,pp.182-191.

Wang, G. Jiang, S., Yang, C., 1994, “On Pro-pusive Performance of Propeller and Rud-der System”, Selected Papers of the ChineseSociety of Naval Architects and Marine En-gineers, Vol. 9, pp. 13-31.

Wellicome, J.F., Wilson, P.A., Cheng, X., 1995,“Prediction of Sway Force and Yaw Mo-ment on Slender Ships”, International Ship-building Progress, Vol. 42, No. 431, pp.259-276.

47

Xiong, X.M., Kose, K., 1996, “A Study onNumerical Methods for the ManoeuvringHydrodynamics Based on the Slender BodyTheory”, TWSNA, No. 91, pp. 61-70.

Xiong, X.M., Wu, X.H., 1996, “A Study onManoeuvring Hydrodynamic Forces Actingon 3D ShipHulls with Free Surface Effectin Restricted Water”, International Ship-building Progress, Vol. 43, No. 433, pp. 48-69.

Yamano, T, Saito, Y, 1997, “An EstimationMethod of Wind Forces Acting on Ships”(in Japanese), JKSNA, No. 228, pp. 91-100.

Yang, Y., 1996, “Study on ship manoeuvringmathematical model in ship handlingsimulator”, MARSIM’96, Copenhagen,Denmark, pp. 607-614.

Yang, Y., Fang, X., 1998, “Simulation model ofship manoeuvring motion in uneven currentand it’s application”, Shipbuilding ofChina, No. 1 (Serial No. 140),(In Chinese).

Yasukawa, H., Yoshimura, Y., Nakatake, K.,1996, “Hydrodynamic Forces on a ShipMoving with Constant Rudder Angle”,MARSIM’96, Copenhagen, Denmark, pp.435-447.

Yasukawa, H. 1997, “Manoeuvring Responseof a Slender Ship with Sinusoidal Steeringin Shallow Water”, (in Japanese), TWSNA,No. 94, pp. 89-99.

Yasukawa, H. 1998, “Computation of EffectiveRudder Forces of a Ship in Shallow Water”,MAN’98, Val de Reuil, France, pp. 125-133.

Yoshimura, Y., Sasaki, H., Takekawa,, M., 1997“Prediction of Ship Manoeuvrability with aFlapped Rudder” (in Japanese), JSNAJ,Vol. 181, pp. 191-196.

Youn, Y-P., Kim, Y-S., 1997, “The Measure-ment and Analysis of Manoeuvrability atSea Trial” (in Korean), Annual SpringMeeting, SNAK, pp. 257-260.

Yum, D-J., Le, T-I., Lee, H-Y., Lee, H-W.,1996, “New Manoeuvring Sea Trials system

Using DGPS”, MARSIM’96, Copenhagen,Denmark, pp. 333-340.

Yumuro, A., Yamamoto, K., 1992, “An Ex-perimental Study on Scale Effect on Direc-tion of Flow Behind Ship Hulls MovingObliquely” (in Japanese), JKSNA, No. 217,pp. 137-143.

Yumuro, A., 1995, “Manoeuvring Hydrody-namic Forces Calculated Using CirculationRetention Characteristics”, (in Japanese),JKSNA, No. 224, pp. 69-76.

Yumuro, A., 1996, “An Approximate Calcula-tion Method for Manoeuvring Hydrody-namic Forces on Ship Hulls in a NarrowWater Channel”, (in Japanese), JKSNA, No.225, pp. 133-139.

Yumuro, A., 1997, “An Approximatie Calcula-tion Method for Manoeuvring Hydrody-namic Forces on a Ship Hull with Heel An-gle”, (in Japanese), JKSNA, No. 228, pp.137-142.

Zec, D., 1996, “An algorithm for a real-timedetection of encounter situations”, JJIN,Vol. 49, No. 1, p. 121.

Zhang, J.W., Andrews, D., 1998, “Manoeu-vrability Performance of a Trimaran Ship”,International Conference on High SpeedCraft Motions & Manoeuvrability, London,RINA, Paper No. 9.

Zhang, R., Sun, Z., Xu, H., 1996, “On the ap-plication of a novel control scheme to shipsteering”, Intl Shipbuilding Progress, Vol.43, No. 434, pp.167-185.

Zhang, R., Sun, Z., Xu, H., 1997, “Robustautopilot design for ship steering in waves”,Intl Shipbuilding Progress, Vol. 44, No.437, pp. 59-75.

Zhu, S.L., Tang, Z.G., 1995, “An ExperimentalStudy of Circulation Controlled Rudderwith Grooves at After-body”, (in Chinese),Ship Engineering, No. 1, pp. 8-11.

Zhu, S.L., Wang,X.F., 1996, “ExperimentalInvestigation of a High-Lift Rudder – Cir-culation Control Rudder”, International

48

Shipbuilding Progress, Vol. 43, No. 436, pp.347-357.

Zou, Z.J., 1996, “Hydrodynamic Forces on aSurfacepiercing Plate in Steady Manoeu-vring Motion”, Proceedings of 21st Sympo-sium on Naval Hydrodynamics, Trond-heimn Norway, pp. 986-995.

Zou, Z.J., Soeding, H., 1995, “A 3D NumericalMethod for Hydrodynamic Forces on aPlate in Manoeuvring Motion”, Ship Ma-neuvering Symposium, Ilawa, Poland, pp.21-38.

Zou, Zao-Jian, 1998, Communication to theITTC Manoeuvring Committee - “NewCalculation Results of ESSO OSAKA,”Wuhan University

Zuev, V.A., Lukomsky, J.A., Miroshnikov,A.N., 1996, “Impulsive course-keepingautopilot for unstable ships.”, MARIND'96,Varna, Bulgaria.

Abbreviation of Journal, Transaction and Con-ference

• JJIN: Journal of Japan Institution of Navi-gation

• JKSNA: Journal of Kansai Society of NavalArchitects

• JSNAJ: Journal of the Society of Naval Ar-chitects of Japan

• TWSNA: Transaction of the West-JapanSociety of Naval Architects

• JSR : Journal of Ship Research edited bySNAME

• MARSIM’96: Marine Simulation and ShipManoeuvrability Symposium, Ed. S. Chis-lett, A.A. Balkema, Rotterdam, Netherlands.

• MAN’98 : Symposium and Workshop onForces Acting on a Manoeuvring Vessel. Ed.S. Cordier, Bassin d’essais des carènes, Valde Reuil, France.

• RINA : Royal Institute of Naval Architects

�

��������������

������������������ �� �������������������

��������������������������

����������� ��������������

������������������� ����������������������

���� ���� ����� �������� � �� � �� �����!���

�� �����"��������������!��������!�����������

��� � ��� ���� ��������� ��� � �������� !�����

������������� ������!� !��� �!���������� � ��

�������� ���� ����������!� ��� ���� ��!��!� ���

����!!��������� ����������������!��!��������!��

��� ������

������#��� ������� ������� ��!�!�� ��$%!� &�!������

'()�� *������� +��� �� !� ���� +��

��������������*� ������ � � �� ����������

���� � ������� � �� � �������� �,-�� �� ���� ����

�����������!�!�!��� ���������� ���� ����������!�

)./� �����!�� ���� ���� � �!� ����� !� ���

!������������� � �� �� �!�� ��� ����� ��!� ����

0����������������!���������!��!���#�����!���!

���� ������� ��!�� ��!��!� ������� �� � ������ ���

����������� ��� ������������� ������!� ���� �����

��0���������

�����1�� ���� !�������� ���� 2��32��� ��!�!� �!� 4���

���������� � 5����� �!� ��� ����� 0����� ����������

���� � ���� 2��32��� ��!�� ����� ������ ���

������������� ��� �����!� ���� ����� �����

�!��� �����!����������!����!��!���5�������!���!

�/�6�/��2��32���!�����������!��!������ �

!���� �� ��� ��������� � �� �� ���� ��� ���

��������!������� ��� �������������������!����

7� � ����!����� �����!� ���� ��� 4���� ���������

1��� ���������!� ���� ��!�� ������ � � �� ���� ��

������������ ��������!����� ����-�������

�������!� ������ ���� ����� ���� ����!� ����� ��$%!

��������� �!� 4����������� ���� �� � ����� ��� �

�� ������ ��4������!� ��� �� ����� ����� ��

����� ���� !��� ������ ��!��!� ���� ���� !����� ��

���� ������ ���� ���� ������������ �� � �- ��������

!� ��� ������� ������� ���������������� � ���

2��32���������� �� �� ��� ��� � �� ��� �55�

�������

��������������� ������ ��������������

������������ ����������

���������������� ��������������

�������������������������

���������������� �������� ������� ��! ����"����#��

-1500 -1200 -900 -600 -300 0 300 600 900 1200 1500Meter

-300

0

300

600

900

1200

Met

er

Raw Data

Corrected Data

7

������������������ �� �������������������

��������������������������

��������������+8'��9��! �

������������������� ��� ���� ���������

�������� ���� ������ ����� ���� :���� "���� �

�������� �� �� � ����!!��� �� � ��� �

�������������!�� !�����������!�;

�����������!����������� �!���������������

�� ���� ��� �0�� !��� ����0!� ��� ���� ���!�

�������������- �������������!�����!!�

������������ ����

7� ������������������� ��$�!��� �� � �� ���!!

!��������!�������������������!�� !������

�����!�����������!��

<� �� � ��� ��� � ���� ����������� ��� ���

�������!����+�� �&�!��������!���������=� ��

������!������>�"�����!����������������������

��������������!��� �� ��� ����� ���������

!���������

.� ����� � ������� ��� �!� ����!!���� ��� ���� ����

���!� ��������������

������������������ �� �������������������

��������������������������

���9�����?�:����?�!��@�����!����A�������������B

���������� �� ��� �

����� �� ������������������������ �����������������

�����5��� ���������� �������� ����0!� ���

�� �� � ���� ���� ���� ������ �!�!!���� ����

��������������������� !���������������

5������������!� ��� �����������������������

������� ��� �!� �� ������ ����� �� 4���������

��������� ��0�� ���� ����!����� �����!� ��� ���� �/3�/

���3�����!� �������������!��������!������0����

����� ���� �/3�/� ���3������������ �!� �����!�

��������������� �� ������ ��� ���������������

��$��������!��5������!����!�����������!���

� ��� �����!� ���� ������������ ���� 0��� �����!�

�������!��������!��!��������������

�����5��� �������� ��!� ���� !��� ���� ������ ��

!�������� ������ � !������� ��� ���!� ������� '!

�������!���������������������������� ������

���� �/� ��� �)� 6!� ������ �����! �� !� ��� �

������� ��� ���� ���!� ������ � ���� ��� � �����!

�������0�!���������� ��������

�����@����� !��� �� ���� � �!� ����� � � ���

�������� ���� ����� !���!�� ��� ��� � �

!�������� � ��� ��� ���� !�� �� @!�� ���

!��������� ��� ��� �� ���� ���� �-���� � �����!

���� ���� ���� ������������� ��� �����!� ��

������� !��� !���� ��� ������������ �������� �����!

�� � !�������!�� 5���� !������ ���� ��4���

��������������������������� ������ ���������� �����

�� �������������

��� ������!���" �� ��� #$�

��� ���������%���#��� �

&������� ���!���"'�� #(� '��

&����)����!*��"������+

�,-

�� ���

�� �� ����� ������!���".(�#���#'�� /������$�� �(�(���0�' �/�#

#��� ����� ������!���".'(�/���#��/ '$�+���##�( (�'���#��# '��#

.�- ����� �������1� ���� ����������2��� ��� ��� ������)� ������

<

�������!����������� ��� ������������������

��!���

�����+����� ���� ���������� �����!� ���� �����

���0� ����� ���� ���������� �������� ���� ���

7<� � �55�� ���!� ������ �!� ���������� !�������

�����������-��������������������������0������

����� �� ������������������������ ����������

�����5��� �������� ����0!� ���� ��������� ���

��!��������� �!�!!����

�����&���� ���� ���� ���!�� �!�!!����� �������

������!� ��� ���� ���!�� 0�� ���� �������

AC� ������ ���� !���������D� ��� ���!� �!�!!���B

���� ���� �!����� � ��� ��� �� ���� ����!�� ��� ���

������ ��� �55�� ���������� �������� �� ����

�����!� ���� ���� ���!�� 0�� ���� �������� ���

� �� ��� ���� ���� ��� � ������� ���� ��!������ ���

�!�!!���� ��� !����� ������!� ��� ���� ���!�

0�� ���� �������� ��� ���� ����� �������� ��� ��!�

�� �!�����

�����"������! ��������������!��0�� ��������������

��$������������ ������� +��� �� �� ��� ����

������������������������!������������������ ���!�

���� ��� ���� ���!�� ����!����� ������ ��� �/E6�/E

���3���� �������� A���������� � �!� C�2�DB�� ���

!���� � ����!����� ������ ��� 7/E67/E� 2�� '���

���������!������������ ��!��� �������������!�

0�� ������������������������� ��!���� �������

���!�� �� �-� ��� ���� ���!�� ����!����� ������ ��

�/E6�/E�2�

�����&���� ���� ���� ���� � ������ �!� �!����� � ��

��� ���� 7�� ���� ����� ��� &&(.� �!� ������!� � ��

=� ��� +�� ��� ���� &�!������ '!!��������� ��

�� ��� ��� ������� �� � �������� ���� ��$

����������� ������� +��� �� �� '� �����!�� ��

���������� ���� !����� �����!� ��!� ����

����� � � ��� +�� � &�!������ ��!������ ��� =� ��

������� &&(.� ���������!�� 1�!� � ��� ���

�-���!���� !� �� ����� !�� ��� ���� �����!��

=� ���!��F�����������!�!����� ��� �� �!��

��� �!�!!� �����!� ��������� ��� ���� !��� ��

A���!���� ��� ���� � �4���� ��� ��������� ���� ���

���!��0�� �����������B�����+�������$�������

����GGG��"���� ������� �������0�������������

�������� ����+��� �� ������ !����� ����,���� ��$

������-�������

�����#��������������������������������H!�����

��� ���� ����!!���� ��� ���� ����� ���� �����

��������� ��� ���� ��������� ���� ���� ���!�� 0�� ���

�����������������������0�

����� �� ������������������������ �� ����!�"���

�����5��� �������� ����0!� 9����� ?�:��� ���� ��!

���������!�����!�� �� ��������

��$� +��� �� !� ���� +�� � ����������

9���������� ���� ���� !�� ��� !� ��� �� ���� ���

�!���� ����!� ������ ���� �� ���� !��

���������� ������������� �!����!�����

�����5��� � ���� �� �� � ��!� ��� ���������

�����!� �� �!� � ��� ������������ �� �����

F�� � ��� =� ��� ���� ���� �� ���� !� � ��� ���

�!���� ������!� ��� !�� ��� !�� $��� ���!��� ���!

��!� ���� ��!� ��� !� ������� ��� �� ������� ���

�� �� ������ �����������!� !���� � ������

���������������- ������!�

�����1�� ���� �� ���� !� � � ������� ������ ���

���� �� �� ������ �����������!� ���� ���� ������

���� ���������� ��� !����� !�� �!� ������ !��� ��

����� ��� ������!� ��� ���� ���������

������������������!�� !�!����!����0��!�

�����'!� ��!� ���� � ��� ���� �� ����� ?���� ��� ��� �

�������������� ��!������ ��� � ��� ?����� ��!�

��� � ��� �- ������!� ����� ���� !����� !�� �

��������!��������� ����� ����� ������������ ��

���� ���� �� �� �� � ���� ���� �� �� ��!����

������������ !����� ��� ���� �� ���� ����!� �����

��������������������!����������!� ��� �����/3�/

���3������������

�����+��� ���� ����!���� ��� ���� !����� !�� �

�������� ��� ���� �����!!���� �4������ !��!� ��

��� ����!!���� �!� ��!� ������ � ��� 9����

?�:���