basis yachts and their evaluation and the use of design tables or formulae. The result derived by the ANN can be seen instantly. If the calculated output parameters do not appeal changes can be made easily by adjusting the input parameters. To make these adjustments more target-orientated the functional relations between the different parameters are plotted by the network.


To give justice to different design approaches – starting with design parameters or starting with performance targets - the network can be run in both directions. The design approach starting with the performance targets is in the following referred to as In-Out whereas the


approach, approach starting with the design parameters is referred to as Out-In approach.


4. CHOICE OF INPUT AND OUTPUT PARAMETERS


The values for the set of input variables that are applied to the network are specified by the customer. Although these variables define specific attributes of the vessel, they are selected as the input parameters because they each capture elements of the desired performance of the vessel, as discussed below


 Displacement


 Dellenbaugh angle  Aspect ratio of sails  Sail area


[t] [°] [-]


[m²]


The displacement combines factors such as size, number of crew members, range, spaciousness of the interior and cost.


The Dellenbaugh angle is a widely used parameter to judge on the stability characteristics of a yacht in an early design stage. The angle can be calculated with a simple formula [1]:


DA


GM  





279 S A HA


(1)


The Aspect ratio of the sails and sail area together can be seen as parameters to classify performance of the yacht.


the sailing


The aspect ratio of the sails is defined with an empirical formula[1].


AR 


(1.1 ) S


MH A


2 (2)


The output parameters derived by the network are the design parameters, required in the very first steps of the design process.


 Length over all  Beam over all  Draft


 Mast height


 Required vertical centre of gravity


©2007: Royal Institution of Naval Architects B-35


[m] [m] [m] [m]


Figure 2: The evaluation


Variation of Length (Maxsurf) of the


BL OA 0.31 OA 


The same is feasible for the length overall to draft ratios of the yachts, although they have different keel forms. However, because the draft


is one of the design


parameters that is likely to be restricted, the draft is chosen as one of the parameters that are varied for each length in the


training data


Tc L  with 1


set. The factor varies


between the lowest, the mean and the highest value of the length to draft ratio derived from the basis yachts.


iOA c 0.14 , 2c  0.17 and 3c  0.2 .


An example for the variation of the draft for constant length over all can be seen in Figure 3.


from waterline


[m] 5. GENERATING TRAINING DATA


In former applications of ANNs to design problems, existing data bases of a specific ship type have been used as training data sets. However, no data base for sailing yachts is available that can be used for the problem at hand. Therefore a way has to be found to create a sufficient amount of training data that represent a certain range of yachts.


The training data is chosen to contain yachts of a length between 10 to 15 meters, in steps of one meter, covering a wide range of series produced yachts. Based on information gained from basis yachts – cruisers and performance cruisers produced


by


companies - a method is found to vary the main dimensions systematically


to generate


four different a


number of training data sets. This method is chosen to shorten the time consuming search for a huge number of basis yachts.


For the standardisation , the length over all is chosen to be the characteristic parameter. An example of the variation of the length over all with constant length to draft ratio can be seen in Figure 2.


sufficient


basis yachts justifies the


expression of the beam as a linear function of the length over all. The arithmetic mean of the values of all yachts is taken factor.


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48  |  Page 49  |  Page 50  |  Page 51  |  Page 52  |  Page 53  |  Page 54