voronoi/delaunay diagram on non-planar surface

Hi!

i want to do a voronoi/delaunay diagram on a wavy surface (not a projected diagram, but work in the 2d surface as it were planar). is this possible?

thanks for the interest!

Replies to This Discussion

Permalink Reply by Sedef Çatalkaya on December 28, 2011 at 5:54am

thank you for your reply I actually tried this but it didnt work I dont know the reason

Permalink Reply by Manuel on December 28, 2011 at 6:15am

probably a paths/branchs problem.. maybe you could post an image of the definition?

if you are using the brep intersect component, try flatten on your polysurface and graft on your 3d voronoi cells (save the file before).

Permalink Reply by Austin Samson on October 17, 2011 at 3:25pm

Is there a way to force the cells, that lie on the curved surface, to be planar?

Permalink Reply by Andrei Raducanu on October 18, 2011 at 1:53am

don't think it's geometrically possible if the cells have more than 3 edges and the overall shape is not uniform.

if you want to close the cells with planar sheets, i think your best bet is to make some tall profiles that can accommodate the different angles between the sheets (or use some pneumatic foil cells)

Permalink Reply by Austin Samson on October 18, 2011 at 3:48pm

So after reading this I decided to try and make what I described manually and realized a few things. Like you said, it is impossible using only cells that are more than 3 sides. However, it is possible if triangles are placed in strategic areas, for example: where the curvature changes directions. I then got to thinking that it is almost like trying to merge two halves of a geodesic dome thus making a almost an s-shape. The triangles enable for the two halves to come together with a relatively smooth slope. I've attached images of both cases.

In short, what I am looking for is a more randomized way of constructing a complex surface by using cells with a combination of sides, while minimizing the amount of 3-sided cells.

Attachments: