Need assistance optimising lines from voronoi cells

Hello guys,

I have created a voronoi grid through populating geo on a lofted surface, and then brep | brep the surface and voronoi cells to get a voronoi grid. I created another of the same voronoi grid by moving it by a certain amplitude.

At the moment i can loft these two grids to get the sort of effect i am after, but i need to make them planar instead of a lofted surface. I have kangaroo script after that to create planarised surfaces, but the simulation would not work, and I am suspecting it is due to the complicated line segments from the voronoi grids. In other words, i need to optimise the voronoi grids in order to make it work, but I am not sure how to accomplish that, or should i try making my grids in another fashion rather than using a brep|brep component.

Hopefully I can receive some suggestions to try out!

Cheers!

Attachments:

Cellular Tess tb.3dm, 69 KB
Cellular Tess tb.gh, 13 KB

Replies to This Discussion

Permalink Reply by Will Pearson on April 26, 2015 at 2:13am

I'm making an assumption here that you want to build this!

The trouble is you have three conflicting constraints: (a) the original "tunnel" geometry, (b) planar elements, and (c) nice connection details. Assuming the tunnel geometry is fixed and that each lofted surface is modelled independently of it's neighbours (i.e. not a brep), you could force planarity by moving the fourth corner point into the plane of the other three, but you'll find you need to design a connection detail as the structure will no longer node out. Totally feasible of course, but every connection detail will likely be unique...

Permalink Reply by Andy L. on April 26, 2015 at 2:32am

Hey will,

I am developing this technique so that i could later use it on a different surface, i.e. not this 'tunnel' geometry. therefore, yes essentially i do want to build this, well maybe just a model of it. My idea was to develope the technique to create a voronoi grid over any surface based on a random populate geo points. Then somehow isolate them, so that like you say, each cell is independent and would not be sharing the same edge surface. (however, i haven't gotten up to that part yet). I am a bit lost with what you meant by 'force planarity by moving the fourth corner into the plane of the other three'? which corners were you referring to?
Actually the technique i am trying to develope is through reverse engineering the Cellular Tessellation project http://monograph.io/knapp/cellular-tessellation i only hoped to be able to create something of the similar sort.
Perhaps there is another approach i should take in terms of creating a voronoi cells, instead of trying to grid them onto a surface?

Cheers

RSS