[GeoMechanicsApplication] Add reset displacement to interfaces #12871
Conversation
| if (stresses_on_integration_points.empty()) { | ||
| rElement.CalculateOnIntegrationPoints( | ||
| CAUCHY_STRESS_VECTOR, stresses_on_integration_points, mrModelPart.GetProcessInfo()); | ||
| } |
There was a problem hiding this comment.
I now made the change here that the CAUCHY_STRESS_VECTOR can also be used. Let me know if you think this is a good idea, or if we should stick with PK2_STRESS_VECTOR only
| if (!rCurrentProcessInfo[IS_RESTARTED] || | ||
| mConstitutiveLaws.size() != shape_function_values_at_integration_points.size()) { |
There was a problem hiding this comment.
The ! is hard to comprehend for the human mind. I guess the else branch is for restarted models with correct constitutive law size. If so, I suggest to use:
if ( rCurrentProcessInfo[IS_RESTARTED] && mConstitutiveLaws.size() == shape_function_values_at_integration_points.size()) {}
else {}
and in this way switch the contents of the if and else branch.
There was a problem hiding this comment.
You're right, made the switch
| @@ -33,6 +33,10 @@ void ResetDisplacementProcess::ExecuteInitialize() | |||
| std::vector<Vector> stresses_on_integration_points; | |||
| rElement.CalculateOnIntegrationPoints(PK2_STRESS_VECTOR, stresses_on_integration_points, | |||
| mrModelPart.GetProcessInfo()); | |||
| if (stresses_on_integration_points.empty()) { | |||
There was a problem hiding this comment.
This check apparently means that the PK2_STRESS_VECTOR was not computed or computable for this element. Is to let the element tell what is natural stress and strain measure are and then compute that one?
There was a problem hiding this comment.
Ideally yes, I don't really know if there is such a mechanism?
| "model_part_name": "PorousDomain.LoadedSide", | ||
| "variable_name": "TANGENTIAL_CONTACT_STRESS", | ||
| "active": [false,true], | ||
| "value": [0.0,667.0], |
There was a problem hiding this comment.
For the normal contact stress the value array is filled with zeros, values come from the table. Here a value is given , but it will be taken from the table anyway. The syntax itself is already confusing, may leave the unused value at zero.
There was a problem hiding this comment.
Yes, done (and in the test I based this one on)
| def test_multi_stage_3_plus_3_line_interface_element_with_neumann_conditions_and_reset_displacements(self): | ||
| file_path = test_helper.get_file_path(os.path.join('line_interface_elements', 'Neumann_multi_stage_reset_displacements')) | ||
|
|
||
| initial_cwd = os.getcwd() |
There was a problem hiding this comment.
Initial current working directory? Behaviour reads like I remember where I am now and will return here when finished wandering. initial_directory is more clear to me as variable name.
There was a problem hiding this comment.
No objections there!
There was a problem hiding this comment.
@rfaasse
Thank you for fixing this issue. It looks OK to me. I just have very minor remarks.
| "strategy_type": "newton_raphson", | ||
| "convergence_criterion": "residual_criterion", | ||
| "displacement_relative_tolerance": 1.0E-4, | ||
| "displacement_absolute_tolerance": 1.0E-4, |
There was a problem hiding this comment.
Usually we set "displacement_absolute_tolerance": 1.0E-9,
Did you set it intentially to 1e-4?
There was a problem hiding this comment.
Put them both to 1e-9, it doesn't really matter since we use residual_criterion anyway, but it's better for consistency, nice catch!
| "variable_name": "TANGENTIAL_CONTACT_STRESS", | ||
| "active": [false,true], | ||
| "value": [0.0,667.0], | ||
| "table": [0,1], |
There was a problem hiding this comment.
The value comes from the table. Value is not relevant here. It can be set to zero
There was a problem hiding this comment.
Agreed, changed (also in the test on which I based this one)
| "active": [false,true], | ||
| "value": [0.0,667.0], | ||
| "table": [0,1], | ||
| "fluid_pressure_type": "Uniform" |
There was a problem hiding this comment.
Preferly to have "fluid_pressure_type" before "value". It makes it easier to follow. However, it is matter of preference.
| "strategy_type": "newton_raphson", | ||
| "convergence_criterion": "residual_criterion", | ||
| "displacement_relative_tolerance": 1.0E-4, | ||
| "displacement_absolute_tolerance": 1.0E-4, |
There was a problem hiding this comment.
1.0e-9? see prevoues comment
There was a problem hiding this comment.
Put them both to 1e-9, it doesn't really matter since we use residual_criterion anyway, but it's better for consistency, nice catch!
📝 Description
Added a test case for reset-displacement and made the necessary changes to the line interface element