Abstract: Training on ship handling is largely recommended by IMO as one of the most effective method for improving the safety at sea. The goal of the above training is to gain theoretical and practical knowledge on ship handling in a wide number of different situations met in practice at sea. Pilot organisations and shipping companies for many reasons consider manned model ship handling training as very effective means of improving ship operator skill. The objection most often raised by opponents of manned model methodology is that "feeling for a ship" could be strongly affected because of time scale effect. It is obvious that in manned model methodology all manoeuvres are realised faster than in reality and for same trainees difficulties resulting from different "feeling for a ship" during first few hours of training have been observed. To clarify this problem special research program has been carried out. The aim of the programme was to realise identical manoeuvres on manned models and on real time desktop computer simulator especially programmed (own ship's parameters and environment effects the same as close as possible) and then to compare different elements: trajectory realised, ship's speed, rudder and main engine settings. The same group of operators made on manned models and on real time simulator manoeuvres. In general no significant differences were found, however some observations concerning differences in operator behaviour can be made.

　

 As is well known, two methods of training in ship handling are available. The first one uses real ships for training, and is called "on board" or "on job" training. In opinion of many ship handlers there is nothing more effective than shipboard training. Such a method of training includes mainly bridge demonstrations, discussions with analysis of situations met at sea and action taken. Special training sessions at sea for operators are rather rare: there is no enough time and money for realisation of such sessions. Many companies consider also such a type of training too dangerous because of increasing level of risk. This is one of reasons of development of the second method of training based on simulator training facilities. Generally it is a computer simulator, but some training centres use large self-propelled manned models equipped with more or less sophisticated simulators of ship systems.

 The manned model technique needs a man-made basin or small lake equipped with mock-ups of piers, canals and other restricted waterways, mooring buoys and other training areas especially designed to present difficult problems and conditions of ship handling. The example of such training areas used at Ilawa Training Centre is shown in figure 1 and photo no 2 and no 3 . A system of control of water level and a protection from strong winds (wooded shore) in case of natural lake is strongly recommended.

　

　

　

 Manned models although they are small, are ships by nature and physical laws govern their behaviour when manoeuvring. In other words manned models represent realistically all hydrodynamic phenomena. For example the model generates low and high pressure areas along the hull in the same way as the ship does. It is very important for restricted water manoeuvres and ship-to-ship interaction where bow cushion and suction areas influence much ship behaviour. Also stopping, slowing down, and all other manoeuvres where hull-propeller-rudder interaction plays a significant role, can be on models in realistic way reproduced.

　

　

 Training on manned models assures the psychological aspect of training by better feeling of effects of grounding, ramming and collisions, also environmental effects such as wind and current are much better visible from the wheelhouse of training models. When performing specific manoeuvring exercise something goes wrong the trainee can immediately see that the result is wrong and understand why it is so. Further explanations by staff may help to understand the physical phenomena and to perform the exercise correctly next time.

　

 Training model's kinematics is based on well-known Froude's identity widely used in free running model manoeuvring tests. Because of not satisfying of Reynold's identity, so-called scale effect occurs. It means that the behaviour of the ship and the representing model may be different. However large dimensions of models using for training can reduce this disadvantage to the minimum not important for training purposes.

 The main disadvantage resulting from principle of model testing technique is that all manoeuvres on models are realised in so called "model time":

　

　

 For example at Ilawa Training Centre the scale is 24, in this connection everything happens 4.9 times faster than in the reality. It results from this that "feeling for a ship" - very important factor in ship handling based on correct timing - can he affected by the above time scale.

 Trainees raise problem of ,,feeling for a ship" very rare and according to their opinion it concerns mainly angular velocity . This may change feeling for ship's dimensions (and also her inertia). Trainees say also that it lasts rather very shortly.

　

 To clarify this problem special research program has been carried out during last year. The goal of the programme was to find answer to the following questions: if the time scale effect involves changes in operator behaviour and how these changes affect realised on manned models manoeuvres, for examples by differences in trajectory?

 The aim of the programme was to follow identical path on manned models and on real time desktop computer simulator especially programmed (own ship's parameters and environment effects the same as close as possible) and then to compare different elements: trajectory realised, ship speed, rudder and main engine settings. The same group of operators has made manoeuvres in case of manned models and of real time simulator.

 For realisation of the research programme one of training areas commonly used during ship handling courses was selected. This area consists of 4 straight waterways intersected at the angle 60°, as was shown in figure 4. The total length of ship's trajectory travelled during experiment was about 6 N.M., three sharp turns either to port or to starboard was included in the programme. All axes of waterways were marked by typical leading marks.

 The task of all participants of the experiment was the same. After unberthing from one of Ilawa Port piers and after passing the reporting line the operator followed the leading mark in so-called Deep Water Route, than in way point between buoys marked as C1 and C2 sharp turn to port side was realised. After travelling 1Nm onto new leading marks the next turn to starboard side took place. Than the ship has followed the new course until the buoy Tl was reached. After making a turn in the close proximity of this buoy (120°of change of heading) the ship approached the next way point situated between buoys D1 and D2. The new change of heading by 60°to starboard side was made there and after travelling next 2 Nm and passing through reporting line the experiment was completed.

　

　

 Participants of experiment were operators -experienced masters - often cooperating with the Centre and that's why having good knowledge of experimental areas. All operators realised all manoeuvres by themselves without giving orders to helmsmen. The experiment started after crossing the reporting line. For assuring the same initial condition for manned model and computer simulator it was proposed to begin experiment from the position as close as possible to the axe of Deep Water Route and from initial velocity corresponding to dead slow ahead setting of main engine. No one other suggestion concerning realisation of manoeuvres was made.

 Water depth was sufficient to consider the training area used for experiment as deep water, during the whole experiment no influence of wind and current was observed.

 In case of computer simulation typical MMG model was adapted. Special attention was paid to equal settings of main engine for model and ship, i,e. in both cases given setting corresponded to the same velocity. Also transversal force due to the propeller rotation was adjusted taking into account results of model tests. The landscape visualisation was assured by means of one 17" high-resolution monitor. Only buoys and leading marks marking waterways used in experiment were visualised.

 The ship used in experiment was 150 000 DWT LCC tanker with rather good turning ability and small degree of unstability. Her main parameters as well parameters of corresponding model are shown in table no 1. The ship and the model were equipped with one fixed pitch right handed propeller having the same diameter and the same pitch. It was assumed that the ship and the model have Diesel type main engine.

　

　

 During the experiment on manned model were measured: trajectory (DGPS), rudder angle deflection, propeller revolution, heading (Gyro Anschutz). Of course in case of desktop simulator all the above parameters were also available.