Str^m-Tejsen and Chislett 



a semitheoretical technique which utilizes the experimental results of captive- 

 model tests in conjunction with the equations of motion expanded to include sig- 

 nificant nonlinear terms. The captive model tests may be experiments using 

 either a rotating arm or a planar -motion mechanism, and the predictions of 

 manoeuvring characteristics are obtained from a solution of the equations of 

 motion by means of either a digital or an analog computer. 



A semitheoretical technique of this kind has been adopted at the Hydro- and 

 Aerodynamics Laboratory (HyA) for investigating steering and manoeuvring 

 qualities of ships and is dealt with in this paper. The HyA- method utilizes a 

 planar-motion mechanism system for experimental measurement of the hydro- 

 dynamic coefficients in the equations of motion, and uses the HyA-GIER digital 

 computer for the numerical solution of the equations. Before launching into 

 technical details, however, it is relevant to consider briefly the reasons which 

 make such an approach desirable. 



Free-running model tests constitute the simplest and most direct means 

 of assessing the behaviour of a given ship design during the execution of a par- 

 ticular manoeuvre. The tests are made by simulating full-scale conditions as 

 closely as possible. The disadvantages of free-running model tests are firstly 

 that they can only be conducted with difficulty in a conventional long narrow 

 towing tank and preferably require a large manoeuvring basin. Secondly, 

 whereas free -running tests provide information enabling engineering decisions 

 to be made as to the suitability or otherwise of a given design, the reasons for 

 the observed performance are not apparent from the test results. The tests are 

 unsatisfactory in that little insight is gained into the hydrodynamic phenomena 

 involved, and a rational basis for improvements in design is difficult to estab- 

 lish. Finally, even when large facilities are available for making free -running 

 model tests, certain scaling problems are difficult to avoid. It is, for instance, 

 difficult to obtain the correct conditions of propeller loading corresponding to 

 the ship propulsion point. When making captive-model tests this is very simply 

 achieved by applying a towing force via the rigid connection to the carriage. 



The semitheoretical technique employing captive-model tests in combina- 

 tion with a mathematical model is a more analytical and potentially more power- 

 ful approach which overcomes these disadvantages. If a planar -motion mecha- 

 nism is employed for the execution of the captive model tests, the disadvantages 

 of the free-running tests mentioned above are eliminated and scale effects in 

 general are reduced because of the large models which can conveniently be used. 

 If a rotating arm is used for the captive model experiments, the disadvantage 

 that a special manoeuvring basin is necessary is not eliminated. Furthermore, 

 because of the small size of most rotating arm facilities, scale effects are 

 likely to be troublesome. 



The equations of motion, which constitute the mathematical model of the 

 situation under investigation, must closely represent the real physical occur- 

 rences and be capable of yielding results with an accuracy at least as good as 

 those obtainable from free -running model tests. The ease with which this can 

 be done depends largely on the degree of complexity of the situation considered. 

 Course stability characteristics involving motions only deviating infinite simally 

 from zero are thus more amenable to simulation by these means than are for 



318 



