[GeoMechanicsApplication] setup a test for the triaxial experiment#14354
[GeoMechanicsApplication] setup a test for the triaxial experiment#14354Sabine-van-Dijk wants to merge 18 commits intomasterfrom
Conversation
…d the right result with the element_integration_point_values_at_time function.
WPK4FEM
left a comment
There was a problem hiding this comment.
Hi Sabine,
Thank you for a good start of this test. You will find a lot of comments which are largely a direct result of how it was done for the element test software setup, hopefully they just help to make it even better. Please have a look at the remarks and also the following:
- Maybe a natural place for this new test is withing tests/test_element_lab
Regards, Wijtze Pieter
| from applications.GeoMechanicsApplication.tests import test_helper | ||
| from kratos.python_scripts import KratosUnittest | ||
|
|
||
| class KratosGeoMechanicsTriaxialTests(KratosUnittest.TestCase): |
There was a problem hiding this comment.
To have this test running automatically in one of the test suites, its class name needs to be added in one of the lists of test cases in test_GeoMechanicsApplication.py
There was a problem hiding this comment.
Yes, it should be added to small_test_cases list, I think.
| """ | ||
| The output for the test was generated for the Mohr Coulomb model using the following material properties: | ||
| "YOUNG_MODULUS": 20000.0, | ||
| "POISSON_RATIO": 0.25, | ||
| "COHESION": 2.0, | ||
| "FRICTION_ANGLE": 25.0, | ||
| "DILATANCY_ANGLE": 2.0 | ||
| """ |
There was a problem hiding this comment.
This kind of description I would expect in a REAMDE.md file in this directory with a schematic picture of the test and maybe graphs of the results in time. Please have a look at e.g. the README.md file in tests/building_pit for a very elaborate example.
indigocoral
left a comment
There was a problem hiding this comment.
Thank you, Sabine, for adding this Triaxial test. It's a very good start indeed. The main change is for the set up of the test which was done by the ElementTest GUI. As Wijtze Pieter has also mentioned, we would need to add a README.md document for this test.
| "problem_name": "triaxial", | ||
| "parallel_type": "OpenMP", | ||
| "echo_level": 1, | ||
| "start_time": 0.0, |
There was a problem hiding this comment.
Changing this to -0.1 so the cell confining pressure can happen beforehand between time -0.1 till 0.0. And then the prescribed displacement can happen within 100 steps between time 0.0 and time 1.0.
| "start_time": 0.0, | |
| "start_time": -0.1, |
| from applications.GeoMechanicsApplication.tests import test_helper | ||
| from kratos.python_scripts import KratosUnittest | ||
|
|
||
| class KratosGeoMechanicsTriaxialTests(KratosUnittest.TestCase): |
There was a problem hiding this comment.
Yes, it should be added to small_test_cases list, I think.
WPK4FEM
left a comment
There was a problem hiding this comment.
Hi Sabine,
Its coming closer. Please the test to the testsuite to let it run automagically, change the bottom condition in the picture to rollers and put some well chosen text + the nice picture in a README.md file and then its good to go for me.
Thank you, Wijtze Pieter
There was a problem hiding this comment.
The fully fixed condition at the bottom of the picture does not match the roller conditions ( only Y fixed ) in the ProjectParameters. As the json file decides what the computation is, please adapt the picture.
| }, | ||
| "time_stepping": { | ||
| "time_step": 8.640000000000002e+03, | ||
| "time_step": 8.64e+03, |
There was a problem hiding this comment.
This is a change ( and the other dsettlement project parameters file too ) unrelated to the purpose of the pull request. Still, it is an improvement.
|
Because I added the test to the already existing test_element_lab it was already in the testsuite. |
There was a problem hiding this comment.
Hi Sabine,
Firstly, apologies from my side about the schematic's base condition. Thanks @WPK4FEM for your comment on it.
Thanks for moving the test to the test_element_lab.py and the corresponding folder. And also for applying the requested changes.
I just have a question about adding the material file to the exception list. I thought we would always format the new files and this exception list is for the old files? Perhaps I have missed the decision here....
Although I saw in your commits that the DYNAMIC_VISCOSITY had an unexpected change...
…m the gid output that we do not assert in the test)
|
I changed the code a bit to do some testing yesterday and thought it was nice to leave it this way with the assertAlmostEqual instead of the assertEqual for future tests |
WPK4FEM
left a comment
There was a problem hiding this comment.
Hi Sabine,
Lets get this on master. In a later revision, you may want to alter the README sligthly: the top nodes that are moved go from y=1.0 to y=0.8, its the displacement u_y that goes from 0.0 to -0.2
That you form creating a complete set with picture and documentation.
Wijtze Pieter
📝 Description
Set up a test for the triaxial experiment.
Key changes
The Project Properties file needed some changes to be able to run the element_integration_point_values_at_time function.
🆕 Changelog