intersecting volume accumulation

hi all,

I'm trying to figure out a way of to addition intersecting volumes, intead of performing a substraction of the intersected parts, just sum up the volumes, one on top of the other.

the diagrams show:

-sections for two volumes

-the intersected part

-how the intersected parts sum up (pile up) to original volume to form the new volume.

the red lines are the same height in both diagrams, just that one they all start at the same height.

maybe the logic could be done by creating a grid of points that generate a grid of vertical lines like the one in the diagram above and then have a cloud of points in space that create a mesh that would be similar to the resuting volume?

would appreciate anyone who has any ideas on how to make an approach to this problem.

I've started to put together a definition that's still super far from what I want.

cheers

Attachments:

accumulation.gh, 12 KB

Replies to This Discussion

Permalink Reply by Andrew Heumann on February 10, 2016 at 10:37am

I think your approach with the vertical lines is a smart one. I've created a definition that does what you're after on the basis of that idea. It's a bit "brute force" - you can increase the sliders to increase the accuracy. In the case of polylines I think you'd be better off placing your "section" curves at any discontinuity in the graph for better precision.

Attachments:

StackedGraphs.gh, 9 KB

Permalink Reply by Rabbit on February 10, 2016 at 12:41pm

Hi Andrew!

Thanks a lot!

Wow, it's quite accurate with not so many lines, I was trying to use your solution for a 3D volume, and couldn't make the boxes work...

Here's what I've accomplished with points so far, imagine that this is the simplest of cases, in more complex ones there should be up to 5 intersecting volumes...

so if it gets quite slow I would have no problem on scaling down the resolution...

do you think boxes would still work with 3D, I just found a component named grid structure...

Attachments:

StackedGraphs_3D.gh, 10 KB

Permalink Reply by Andrew Heumann on February 10, 2016 at 1:09pm

Ah, I didn't see you were after 3d volumes. I think something like the attached will be the most effective/efficient way to do it - provided you are dealing primarily with volumes with planar faces. Try this. If you mean to add more freeform volumes together you will have to do some kind of point-based sampling approach.

Attachments: