Add chimera app

[jginternational]

Not ready yet
Missing JSON files

[jginternational]

Chimera_Case.gid.zip

[adityaghantasala]

@jginternational The zip file , is it the same case as Ruben posted on the other thread ?
And is it going to be a example case or something similar ?

[adityaghantasala]

Also I checked the interface, looks good. Thank you very much for your efforts. One minor thing I noticed is that the case is always 2D. There should also be an option to have 3D too.

[jginternational]

Hi @adityaghantasala
The zip is the same case Ruben posted (maybe the condition values are not the same) and it can be the given example.

I'll activate the 3D also.

[rubenzorrilla]

Can we add this cross section example as predefined example in the GUI?

[adityaghantasala]

yes we can ... !

[rubenzorrilla]

yes we can ... !

I'll try to create it.

[adityaghantasala]

@rubenzorrilla ... How can I help here to continue with finishing the problem type. From the last commits, there is minor change in the JSON format. I will attach the JSON here by the end of the day. Niw JSON is similar to the Mesh motion solver.

[rubenzorrilla]

@rubenzorrilla ... How can I help here to continue with finishing the problem type. From the last commits, there is minor change in the JSON format. I will attach the JSON here by the end of the day. Niw JSON is similar to the Mesh motion solver.

Hi! I think that in the last iteration we had some problems with the writing. Could you please attach an example compatible with the latest version of your branch (including the json you mentioned above).

Thanks!!

[adityaghantasala]

Hi guys
Here are the examples which can also go in to problem types. Inside you will also find the json files in their newest format. Do let me know if they are easy to realize with the problemtype.

examples_for_chimera_problem_type.tar.gz

[adityaghantasala]

      "chimera_settings":{
         "chimera_echo_level":2,
         "reformulate_chimera_every_step":true,
         "chimera_parts":[
            [
               {
                  "model_part_name":"FluidModelPart.Parts_background_surface",
                  "overlap_distance":0.7,
                  "model_import_settings":{
                     "input_type":"mdpa",
                     "input_filename":"sep_meshes/chimera_background"
                  }
               }
            ],
            [
               {
                  "model_part_name":"FluidModelPart.Parts_patch_surface",
                  "overlap_distance":0.7,
                  "model_import_settings":{
                     "input_type":"mdpa",
                     "input_filename":"sep_meshes/chimera_patch_for_embedded"
                  }
               }
            ]
         ],
         "internal_parts_for_chimera":[
         ]
      }

[jginternational]

Hi @adityaghantasala
in the json:

"chimera_parts":[
        [
           {
              "model_part_name":"FluidModelPart.Parts_background_surface",
              "overlap_distance":0.7,
              "model_import_settings":{
                 "input_type":"mdpa",
                 "input_filename":"sep_meshes/chimera_background"
              }
           }
        ],
        [
           {
              "model_part_name":"FluidModelPart.Parts_patch_surface",
              "overlap_distance":0.7,
              "model_import_settings":{
                 "input_type":"mdpa",
                 "input_filename":"sep_meshes/chimera_patch_for_embedded"
              }
           }
        ]
     ]

Do you need this nested list structure? It's a list of lists of patches. Wouldn't it be better something like:

"chimera_parts":[
{
    "model_part_name":"FluidModelPart.Parts_background_surface",
    "overlap_distance":0.7,
    "model_import_settings":{
        "input_type":"mdpa",
        "input_filename":"sep_meshes/chimera_background"
    }
} ,
{
    "model_part_name":"FluidModelPart.Parts_patch_surface",
    "overlap_distance":0.7,
    "model_import_settings":{
        "input_type":"mdpa",
        "input_filename":"sep_meshes/chimera_patch_for_embedded"
    }
}
]

In addition, I can place the submodelpart name of the internal boundaries in each patch

[jginternational]

Ok, in this branch, you can test this example
test_chimera.gid.zip

Notice that we have 2 patches created:
-The first one has 2 internal boundaries and the second one 1

3 mdpa files are generated:

• 1 for the fluid
• 2 for the patches (One for each patch)

Chimera settings are:

"chimera_settings"            : {
            "chimera_echo_level"             : 1,
            "reformulate_chimera_every_step" : true,
            "chimera_parts"                  : [[{
                "model_part_name"            : "FluidModelPart.Patch_2",
                "overlap_distance"           : 0.7,
                "model_import_settings"      : {
                    "input_type"     : "mdpa",
                    "input_filename" : "Patch_2"
                },
                "internal_parts_for_chimera" : ["FluidModelPart.ChimeraInternalBoundary2D_Cross_2"]
            }],[{
                "model_part_name"            : "FluidModelPart.Patch_1",
                "overlap_distance"           : 0.8,
                "model_import_settings"      : {
                    "input_type"     : "mdpa",
                    "input_filename" : "Patch_1"
                },
                "internal_parts_for_chimera" : ["FluidModelPart.ChimeraInternalBoundary2D_Cross_1_A","FluidModelPart.ChimeraInternalBoundary2D_Cross_1_B"]
            }]]
        }

[adityaghantasala]

Awesome @jginternational thank you very much. I will test this tomorrow as I need to change some parts of the code to take care of the internal boundary parts.

[adityaghantasala]

@jginternational There are theree small issues :

1. The MainKratos.py is somehow outdated.
2. It the problemtype is not outputting the FluidMaterials.json
3. The Background is not being written in the "chimera_parts" of the "chimera_settings".

[jginternational]

@adityaghantasala

@jginternational There are theree small issues :

1. The MainKratos.py is somehow outdated.

Done.

2. It the problemtype is not outputting the FluidMaterials.json

Done

3. The Background is not being written in the "chimera_parts" of the "chimera_settings".

Would it be possible to assume that the background part is the one in volume_model_part_name? As far as I understand the background model part is always unique. Besides I find it easier from the user perspective as it would require to only specify the patches.

[adityaghantasala]

@jginternational good suggestion, I will see how I can do that and will get back to you.

[jginternational]

@adityaghantasala @KratosMultiphysics/chimera

The current Gui in this branch generates this example, that does not run. I don't know how to follow

chimera.gid.zip

Error message:

+ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    + CategoryInfo          : NotSpecified: (Traceback (most recent call last)::String) [], RemoteException
    + FullyQualifiedErrorId : NativeCommandError
 
  File "C:\Users\garat\Desktop\chimera.gid\MainKratos.py", line 32, in <module>
    simulation.Run()
  File "E:\Deploy_Kratos\KratosMultiphysics\analysis_stage.py", line 47, in Run
    self.Initialize()
  File "E:\Deploy_Kratos\KratosMultiphysics\analysis_stage.py", line 93, in Initialize
    self._GetSolver().Initialize()
  File "E:\Deploy_Kratos\KratosMultiphysics\ChimeraApplication\navier_stokes_solver_vmsmonolithic_chimera.py", line 
47, in Initialize
    super(NavierStokesSolverMonolithicChimera,self).Initialize()
  File "E:\Deploy_Kratos\KratosMultiphysics\FluidDynamicsApplication\navier_stokes_solver_vmsmonolithic.py", line 324, 
in Initialize
    solution_strategy.Initialize()
RuntimeError: Error: The following errors occured in a parallel region!
Thread #7 caught exception: Error: In initialization of Element QSVMS #10340: No CONSTITUTIVE_LAW defined for property 
0.


in E:/PROYECTOS/KRATOS/Kratos/applications/FluidDynamicsApplication/custom_elements/fluid_element.cpp:88:FluidElement<c
lass QSVMSData<2,3,0> >::Initialize
   E:/PROYECTOS/KRATOS/Kratos/applications/FluidDynamicsApplication/custom_elements/fluid_element.cpp:100:FluidElement<
class QSVMSData<2,3,0> >::Initialize
Thread #6 caught exception: Error: In initialization of Element QSVMS #10181: No CONSTITUTIVE_LAW defined for property 
0.


in E:/PROYECTOS/KRATOS/Kratos/applications/FluidDynamicsApplication/custom_elements/fluid_element.cpp:88:FluidElement<c
lass QSVMSData<2,3,0> >::Initialize
   E:/PROYECTOS/KRATOS/Kratos/applications/FluidDynamicsApplication/custom_elements/fluid_element.cpp:100:FluidElement<
class QSVMSData<2,3,0> >::Initialize

[philbucher]

@adityaghantasala is not working at the chair any more, not sure who could solve this problem :/
@rubenzorrilla this seems like a regular fluid problem, do you have an idea?

[rubenzorrilla]

@jginternational I think that the developments in the ChimeraApplication are kind of discontinued. Speaking frankly, IMO it isn't worth to put effort on its GUI.