[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: complex boundaries
From: |
Jean-Noël Grad |
Subject: |
Re: complex boundaries |
Date: |
Fri, 18 Mar 2022 02:03:52 +0100 |
User-agent: |
Roundcube Webmail/1.3.17 |
Hi,
At the moment, 4.2.0 only supports basic shapes and metashapes (unions
of basic shapes [1]). But in 4.3.0, it will be possible to provide a 3D
bitmask numpy array to construct arbitrary geometries, as well as a 3D
float numpy array to define the slip-velocity on individual nodes. In
the attached pypresso script, a random landscape is generated with
numpy.random.uniform and a smoothing kernel (lb_input_geometry.jpg), and
it is forwarded to the walberla LB fluid class to define the boundary
nodes, which are then visualized in the OpenGL visualizer
(lb_espresso_visu.png). However the particles won't bounce off the LB
boundary. To address that issue, one could for example derive normal
vectors from the boundary surface to construct a 3D force field, which
can then be used to repel particles via the field coupling feature [2].
We are still working on the 4.3.0 version of ESPResSo, though. You can
try the development version by pulling the walberla branch locally [3];
the attached script should work on commit 9e6ddbb9c18f839450 (the
`viridis` visualization option won't work).
Best,
JN
Links:
[1]
https://espressomd.github.io/doc/espressomd.html#espressomd.shapes.Union
[2]
https://espressomd.github.io/doc/espressomd.html#espressomd.constraints.ForceField
[3] https://github.com/espressomd/espresso/pull/2701
On 3/17/22 22:50, zzzaaalll@virgilio.it wrote:
Hello everyone,
I'm writing to you to ask if the last version of Espresso provides for
the possibility of inserting boundaries of arbitrary shape? Actually a
few years ago I obtained this result in the publication reported in the
link below by inserting a mesh kept fixed during the simulation (the
green mesh in fig. 4). However, the drawback of this method is that the
resulting boundaries didn't behave correctly with the Lattice-Boltzmann
fluid (the flow lines crossed the boundaries).
Thank you in advance for your support.
Best regards
G
link:
https://www.dropbox.com/s/ly3m8837pobcm51/lazzini_etal_2021.pdf?dl=0
mwe.py
Description: Text document
lb_input_geometry.jpg
Description: JPEG image
lb_espresso_visu.png
Description: PNG image