Hi everyone,
I was trying to use another mapping method instead of neares-neighbor mapping.
Therefore I changed my config from:
<mapping:nearest-neighbor direction="write"
from="Fluid-Mesh"
to="Solid-Mesh"
constraint="conservative"/>
<mapping:nearest-neighbor direction="read"
from="Solid-Mesh"
to="Fluid-Mesh"
constraint="consistent"/>
to
<mapping:rbf-thin-plate-splines direction="write" from="Fluid-Mesh" to="Solid-Mesh" constraint="conservative" z-dead="true"/>
<mapping:rbf-thin-plate-splines direction="read" from="Solid-Mesh" to="Fluid-Mesh" constraint="consistent" z-dead="true"/>
My mesh is something like this:
With nearest neighbor-mapping, everything is working fine, but switching to rbf, there is happening a huge displacement and therefore a crash of the simulation. Does anybody know, what could go wrong in this case?
Hi,
is it possible to add a picture of the nearest neighbor displacement and RBF displacement. Have you only tried thin plate splines? I would suggest also a compact RBF polynomial with a support radius of 3 or 4 times the mesh width. You can simply use
rbf-compact-polynomial-c0 support-radius="value"
Alternatively, is using z-dead=true actually required? Does it help to remove this?
1 Like
Hi,
sorry for the late reply.
Here you can see a picture of the fluid domain:
1st Timestep:
2nd Timestep (scaling is an issue - not wanted):
My solid domain is just a vertical beam which can bend along its vertical axis.
As it seems, a beam with 11 nodes on the y - axis can not be mapped with precice. Therefore, I just extruded the beam in x - direction 10 times.
I have now managed to get a deformation with rbf mapping but, it seems to map the data in a wrong way (see scaling issue).
I think my approach is wrong, but I do not know how to set up a mesh of a beam correctly.
Yes, since it is a 2D case, I have to mark the z-axis as dead.
