Drainage Direction Script

I made a python script that estimates the direction of drainage paths on a surface.

It does not deal with puddling or accumulation, and if you have any suggestions on how to make it faster or to improve it in any way, let me know -- this is a first draft.

you can see the python code in this github gist:

https://gist.github.com/1054969

Attachments:

drainScript.ghx, 218 KB

Replies to This Discussion

Permalink Reply by kleerkoat on July 1, 2011 at 12:36pm

cool, thank you!

Permalink Reply by Nay Soe on October 11, 2011 at 3:59am

is this possible in rh 4? or does the python script only works in rh5?

Permalink Reply by Benjamin Golder on October 11, 2011 at 7:21am

@Nay: This script only works in Rhino 5, though it would be possible to rewrite it in a language other than python in order to run it in Rhino 4.

Permalink Reply by Nay Soe on October 12, 2011 at 1:27am

thanks for the reply ben, i downloaded rh5, and it works very pretty, and fast. But do you think this could be applied to a mesh face instead of a surface?

Permalink Reply by Benjamin Golder on October 12, 2011 at 4:10am

The same basic logic could be applied to a mesh. The current logic of the looping algorithm is like this:

find the closest point on the surface
if the new point passes the test (the point is not higher than or too close to the previous point) then ...
get a tangent plane at that point
figure out the "downmost" direction in that plane
move the stepsize distance in that direction

Step 1 would have to be edited to get the closest point on a mesh (instead of a surface).

Step 3 should be different. Instead of falling on a mesh face, the closest point may fall on a mesh edge, and you would have to decide how you want to calculate the new drainage direction at that point (perhaps along the edge in the downmost direction?). Or in that case you could choose which mesh face to use for calculating a new tangent plane.

Perhaps instead of finding points in space and then closest points on the surface geometry, a mesh version should traverse the edges of the mesh?

Permalink Reply by taz on October 24, 2011 at 4:24pm

Ben,

I've been thinking about drainage with a mesh...

Worse comes to worse, Kangaroo can be used to make it rain and trace a flowline..

Permalink Reply by Nay Soe on October 30, 2011 at 2:48am

i've been just converting it from mesh to srf, but that's a great idea using kangaroo. thanks

Permalink Reply by Rémy Maurcot on December 8, 2011 at 6:01am

Hi all,
The post speak a surface, but it's intersting for à mesh ...
À mesh is more complicate to assign drainage système ??

Regards

Permalink Reply by Rémy Maurcot on December 9, 2011 at 5:05am

Hi Benjamin,

I test your procedure, but I can not generate a point on a mesh with GHPython.

And for convert:

newFrame = self.mesh.SurfaceAt(self.u, self.v)

For a mesh, GHPython product an error.

Thansk for your helping

Permalink Reply by Benjamin Golder on December 13, 2011 at 2:59am

for people who are interested in how this could be applied to a mesh, it occurred to me that there is a much simpler algorithm that could be used to simulate drainage on a mesh: traveling the path of least resistance along edges from a particular node in the mesh.

Basically, you start at a particular point in the mesh, and then determine which edge connected to that point is most steeply sloped downwards, and then follow that edge to the next point until reaching a point that has no downward-sloping edges.

I'll post an example in a few days, when I 'm not so busy.

Permalink Reply by Rémy Maurcot on December 13, 2011 at 4:03am

Thank you in advance, I see the procedure for calculating these things.
The thing is more complex if the mesh is made of himself in the mesh.
We should be able to "merge" between the cracks it.