You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I recently came across PhiFlow and I've been really enjoying the accessible API design and ease of use. Thank you so much!
I've been experimenting with the Karman Vortex Street demo and wanted to simulate the movement of a solid object within the flow.
My attempt to do so by advecting a PointCloud in the velocity field was successful, however, when I try to do it with a CenteredGrid instantiated with Sphere for the values, the field slowly detoriates, which is to be expected. I'm definitely missing something here and would greatly appreciate receiving any guidance about how I could possibly advect a solid geometry.
Here are some of my attempts attached below.
The text was updated successfully, but these errors were encountered:
Hi, I appreciate the kind words!
As for your problem, this is indeed an expected numerical artefact that can't be fully eliminated. The amount of numerical diffusion depends on the method you use and of course the grid resolution.
Semi-Lagrangian advection generally incurs the largest diffusion.
MacCormack is still relatively fast but leads to less diffusion.
Higher-order advection might be more accurate. You can try setting order=4 or order=6, see here
I recently came across PhiFlow and I've been really enjoying the accessible API design and ease of use. Thank you so much!
I've been experimenting with the
Karman Vortex Street
demo and wanted to simulate the movement of a solid object within the flow.My attempt to do so by advecting a
PointCloud
in the velocity field was successful, however, when I try to do it with aCenteredGrid
instantiated withSphere
for the values, the field slowly detoriates, which is to be expected. I'm definitely missing something here and would greatly appreciate receiving any guidance about how I could possibly advect a solid geometry.Here are some of my attempts attached below.
The text was updated successfully, but these errors were encountered: