Create a perpendicular-flap tutorial case for Dune #239
Create a perpendicular-flap tutorial case for Dune #239davidscn merged 5 commits intoprecice:developfrom
Conversation
davidscn
left a comment
davidscn
left a comment
There was a problem hiding this comment.
This is a very nice contribution fulfilling a lot of aspects already, thank you very much. However, in a first step we need to sort the files a bit.
| @@ -0,0 +1,22 @@ | |||
| ## Setup | |||
There was a problem hiding this comment.
You have essentially two places to provide all relevant information: tutorial specific information belong into the tutorial README and the adapter specific information belong onto an adapter webpage (to be created https://github.com/precice/precice.github.io/edit/master/pages/docs/adapters/dune) . I can help with this if you like. However, the participant directory doesn't contain a README.
There was a problem hiding this comment.
I would appreciate some help ;)
| @@ -0,0 +1,29 @@ | |||
| cmake_minimum_required(VERSION 3.13) | |||
There was a problem hiding this comment.
This directory (dune-perpendicular-flap) would go to the adapter repository. I understand that the applicability is limited from a physical perspective to this tutorial, but that's how we handle separate the tutorials (configuration, mesh, etc.) from the actual solver. There is only one exception, namely one openfoam tutorial we have.
|
I reordered the files as described by @davidscn. This means the directory Is the file structure for the tutorial ok as it is right now? |
|
Yes this looks perfectly fine. The documentation you deleted here needs to be added to the adapter docs then. |
perpendicular-flap/README.md
Outdated
|
|
||
| * deal.II. For compatibility reasons, the solver reads `Force` data from other `Fluid` participants. Reading and applying `Force` data is currently only implemented in the linear elastic solid model (`Model = linear` in the parameter file). Thus, the default simulation setup is only compatible with the linear elastic solid model. If you want to apply the non-linear solid model, please use `Stress` data for your read data, which is supported by the OpenFOAM adapter as well (example given in Turek-Hron-FSI). The `./run.sh` script takes the compiled executable `elasticity` as input argument (`run.sh -e=/path/to/elasticity`) and is required in case the executable is not discoverable at runtime (e.g. has been added to the system `PATH`). For more information, have a look at the [deal.II adapter documentation](https://precice.org/adapter-dealii-overview.html). | ||
|
|
||
| * Dune. The structural model is limited to linear elasticity. The solver reads `Force` data and writes `Displacement` data. The `./run.sh` script needs the compiled executable `solid-dune` in the same directory, which can be obtained by building the dune-adapter. |
There was a problem hiding this comment.
The executable generated from your module is called dune-perpendicular-flap and not solid-dune.
There was a problem hiding this comment.
You're absolutely right! Thanks, I will change that.
3 participants
Creating a new solid-participant for Dune in the perpendicular-flap tutorial case.
More information can be found in the README!
The test-case shows the right behavior, but I wasn't able to reproduce the exact results
of the other solid participants. I assume that there is some kind of little mistake, maybe
in the dune-adapter or dune-elastodynamics module (needed as dependency).
This should provide a nice test setup to be able to run a first simulation with the dune-adapter (precice/dune-adapter#1).
Of course this also depends on the dune-core-modules!