2. SYNOPSIS
feelpp_toolbox_fsi [--fsi.filename <path to .json specification file>] [--config-file <path to .cfg file] [--case <directory to model case>] [--case.dim 3] [--case.discretization P2P1] [--fsi.ksp-monitor=<0,1>] [--fsi.ksp-view=<0,1>] [--fsi.ksp-type=<see below>] [--fsi.pc-view=<0,1>] [--fsi.pc-type=<see below>] [--fsi.snes-monitor=<0,1>] [--fsi.ksp-type=<see below>]
3. DESCRIPTION
feelpp_toolbox_fsi is an application which can run fluid structure interaction model defined by a set of files : a geometry or mesh, a config file(.cfg
) and a model file(.json
)
The execution steps are as follows:
-
load the model
-
load the mesh
-
setup model (materials, initial and boundary condition)
-
execute the steady or unsteady simulation
-
post-process the results, viewable in paraview or ensight
3.1. Options
feelpp_toolbox_fsi requires some options.
Name | Description | Default value | Possible Values |
---|---|---|---|
case.dim |
dimension of problem |
3 |
2,3 |
case.discretization |
discretization of problem |
P2P1 |
P2P1 |
fsi.filename |
path to the .json specification case file |
no default value |
required from user |
fsi.ksp-monitor |
monitor the Krylov Subspace iterations |
0 |
0,1 |
fsi.ksp-view |
view the Krylov Subspace structure |
0 |
0,1 |
fsi.ksp-type |
change the Krylov Subspace method |
|
|
fsi.pc-view |
view the Krylov Subspace structure |
0 |
0,1 |
fsi.pc-type |
change the Krylov Subspace strategy |
|
|
fsi.snes-monitor |
if applicable, monitor the non-linear solver iterations |
0 |
0,1 |
fsi.snes-type |
if applicable, change the non-linear solver iterations |
|
- NOTE
-
0 as option value means that the option is not enabled, 1 means it is enabled.
4. EXAMPLES
We are now using the feelpp/feelpp-toolboxes:latest
docker images as described in docs.feelpp.org/user/0.110.0-rc.5/install/containers/#_docker to demonstrate feelpp_toolbox_fsi usage.
feelpp/feelpp-toolboxes:latest
docker run --rm -ti -v $HOME/feel:/feel ghcr.io/feelpp/feelpp-toolboxes:v0.110.0-rc.5
docker run (1)
--rm (2)
-ti (3)
-v $HOME/feel:/feel (4)
feelpp/feelpp-toolboxes (5)
1 | run docker executable |
2 | remove container from docker once execution is done |
3 | run in interactive mode and enter the docker container |
4 | mount $HOME/feel on host to /feel/ in docker: the results will be exported there |
5 | the docker container to execute, by default it uses the latest tag |
mpirun --bind-to core -np 10 feelpp_toolbox_fsi --case "github:{path:toolboxes/fsi/cases/TurekHron}"
mpirun --bind-to core (1) -np 10 (2) (3) --case "github:{path:toolboxes/fsi/cases/TurekHron}" (4)
1 | execute in parallel using MPI with processes bound to cores |
2 | execute in parallel usisng 10 processes |
3 | the name of the toolbox application |
4 | case directory containing the model files (geometry, config and json). This case file is available remotely on github. |
Other test cases can be found
-
locally in /usr/share/feelpp/data/testcases/toolboxes/fsi.
5. RESOURCES
- Feel++ Docs
- Feel++ Cases for feelpp_toolbox_fsi
- Feel++ Toolbox Docs for feelpp_toolbox_fsi
6. SEE ALSO
- Feel++ Mesh Partitioner
-
Mesh partitioner for Feel++ Toolboxes docs.feelpp.org/user/latest/using/mesh_partitioner/
- Feel++ Remote Tool
-
Access remote data(model cases, meshes) on Github and Girder in Feel++ applications. docs.feelpp.org/user/latest/using/remotedata/