Re: [Paddlewise] Design

On Aug 17, 2004, at 7:36 AM, John Winters wrote:

> PeterO wrote;
>
>> Interesting analysis, though I hope and suspect that like computer
> generated
>> music, computer designed kayaks will only ever be a small niche 
>> market.
> Are
>> cars designed this way?
>
> Some notes about using computers in design work.
>
> Computers do not design much of anything. I asked mine to design a 
> kayak and
> left it on for  week but nothing happened. Turns out you have to do it
> yourself. The computer does help with the drawing and calculations 
> though.
> It helps if you actually know something about design first since you 
> can
> design bad boat faster on a computer just as easily (or maybe more 
> easily)
> than designing a good boat..

The Godzilla portion of the Michlet analysis program is as close to a 
computer designing a hull as I know of at this point. Of course this 
only "designs" the part of the boat below the waterline and leaves the 
above water shape to the human designer.

For those unfamiliar with Godzilla, it works with the drag prediction 
program Michlet. It creates a random family of hull shapes which are 
constrained to things like length, width, displacement, etc by the 
user. It then lets Michlet analyze all the hulls. It selects the one 
that best fits a suitability factor. This factor would typically be the 
lowest drag at a given speed or range of speeds. From this choice it 
randomly distorts the hull shape by a small amount to create another 
family of hull shapes. Which it then analyzes to find the best one. It 
repeats this process until the user gets tired of waiting. In this way 
it can attempt to evolve an "optimum" hull shape within the desired 
constraints.

While it is still up to the user to determine the constraints (thus 
acting as a designer) it is the computer that determines the ultimate 
shape. I have used this system to define the below-the-water hull shape 
for one design as a science experiment and I am happy with the results. 
There was still a lot of work to create a above-water-shape but the 
wetted hull shape seems good.

With ever increasing capabilities to analyze varying aspects of 
performance, it is at least conceivable that this type of system could 
provide more and more sophisticated design creation possibilities. For 
example, rough water drag components could be added to find an 
"optimum" design for going at a desired speed into a given sea state. 
Performance characteristics such as maneuverability could at least 
theoretically be included. It would still be up to the user/designer to 
choose the constraints and desired characteristics to optimize. But 
then maybe you could leave the computer for a week and actually see 
something kayak-like result when you come back. Of course it might take 
a month to create the appropriate set of constraints.

Nick Schade

Guillemot Kayaks
824 Thompson St
Glastonbury, CT 06033
USA
Ph/Fx: (860) 659-8847
http://www.guillemot-kayaks.com/
***************************************************************************
PaddleWise Paddling Mailing List - Any opinions or suggestions expressed
here are solely those of the writer(s). You must assume the entire
responsibility for reliance upon them. All postings copyright the author.
Submissions:     PaddleWise_at_PaddleWise.net
Subscriptions:   PaddleWise-request_at_PaddleWise.net
Website:         http://www.paddlewise.net/
***************************************************************************