Lewis 



CONCLUSIONS 



A method of computing the response of a ship to irregular waves is avail- 

 able which is non-dimensional and convenient for graphical presentation. Use 

 of this log- slope form of plotting shows that for most ship responses the wave 

 components at the peak of the wave spectrum may be less significant than the 

 shorter wave components. It also shows that the cut-off point, or maximum 

 wave length present, is of considerable importance. 



Samples of the application of the procedure lead to certain general conclu- 

 sions: 



Non-dimensional wave bending moment coefficients in very severe seas 

 show a distinct downward trend as ship size increases. 



Wave-induced bending moments in severe storm seas are affected relatively 

 little by increase of speed, but relative motion between bow and wave is appre- 

 ciably affected. High values of length/draft ratio show a distinct advantage in 

 this respect which leads to less shipping of water forward with a given 

 freeboard/length ratio. But possible danger of slamming from greater forefoot 

 emergence should be considered. 



Change of heading has a significant effect on wave bending moments, but 

 much more so in long-crested than in short- crested seas. Relative bow motion 

 is greatly reduced by a change from head to beam seas. 



FUTURE POSSIBILITIES 



It is of interest to consider some of the further possibilities in the applica- 

 tion of results of seakeeping research. One of the obvious steps being under- 

 taken at M.I.T. [13] and elsewhere is to make use of calculated response ampli- 

 tude operators instead of model test values. This requires perhaps some further 

 refinement in ship theory along the line of work by Grim [14] and Gerritsma[l5]. 

 It also requires that the theory be extended to oblique seas in order that short- 

 crestedness can be properly taken into account. Preliminary investigation of 

 this important problem indicates that it may not be too difficult [16]. Lalangas 

 [17] has shown that pitching and heaving motions in oblique seas can be predicted 

 quite well simply by allowing for the effect of heading on effective wave length, 

 frequency of encounter, and ship-wave interaction effects such as "Smith effect." 

 In due course it will be possible to evaluate the seagoing performance of any 

 number of alternative designs entirely by electronic computer. 



Meanwhile, systematic model tests at all headings to regular waves can 

 provide the needed inputs (response amplitude operators) into our calculations. 

 For ships of very unusual characteristics, such as semi-submerged types for 

 supercritical operation [18], model tests are the only reliable basis for the cal- 

 culations. The excellent work of Vossers [3] should be extended to cover a 

 wider range of ship characteristics and speeds. From the viewpoint of naval 

 ship design, the need for systematic model tests is particularly great, for very 

 little complete information is now available. For example, many reports on 



206 



