[StructuralMechanicsApplication] Add initial state and stress calculation to truss element and constitutive law #12525

rfaasse · 2024-07-10T11:36:02Z

📝 Description
This PR has two additions to the structural mechanics components, to allow for the GeoMechanicsApplication to reset displacements using structural elements, instead of using the geo structural elements. The following additions are being made:

The TrussConstitutiveLaw now adds the initial stress and initial strain to the calculation of the PK2 stress vector if the ConstitutiveLaw::InitialState is set in CalculateMaterialResponsePK2. This is very similar to how it is done in the ElasticIsotropic3D::CalculateMaterialResponsePK2.
The TrussLinearElement3D2N is now able to calculate the PK2_STRESS_VECTOR on the integration points.

Unit tests are added for the changed functions. When no ConstitutiveLaw::InitialState is set for the truss constitutive law, the observed behavior is the same as before.

…el and added unit tests

…pe of the strain and stress vectors (since they are refs in the parameters)

avdg81

This looks like a clean and straightforward set of changes to me. And many thanks for adding tests that demonstrate the intended behavior (i.e. it reads like documentation). I only have a few minor suggestions, nothing blocking.

applications/StructuralMechanicsApplication/custom_elements/truss_element_linear_3D2N.cpp

applications/StructuralMechanicsApplication/custom_elements/truss_element_linear_3D2N.hpp

applications/StructuralMechanicsApplication/tests/cpp_tests/test_truss.cpp

applications/StructuralMechanicsApplication/tests/cpp_tests/test_truss_constitutive_law.cpp

…t semi-colons after macros)

philbucher · 2024-07-11T11:59:35Z

LGTm

AlejandroCornejo · 2024-07-11T15:24:25Z

applications/StructuralMechanicsApplication/custom_constitutive/truss_constitutive_law.cpp

@@ -141,6 +141,7 @@ void TrussConstitutiveLaw::CalculateMaterialResponsePK2(Parameters& rValues)
    Vector& stress_vector = rValues.GetStressVector();
    if (stress_vector.size() != 1) stress_vector.resize(1, false);
    stress_vector[0] = this->CalculateStressElastic(rValues);
+    AddInitialStressVectorContribution(stress_vector);


no initial strains? :)

Good point, we do not need it for our application at the moment, but added it now for completeness (as well as to the unit test)

AlejandroCornejo · 2024-07-11T15:25:06Z

applications/StructuralMechanicsApplication/custom_elements/truss_element_linear_3D2N.cpp

@@ -189,20 +189,20 @@ void TrussElementLinear3D2N::WriteTransformationCoordinates(
    BoundedVector<double, TrussElement3D2N::msLocalSize>
    & rReferenceCoordinates)
 {
-    KRATOS_TRY;
+    KRATOS_TRY
    rReferenceCoordinates = ZeroVector(msLocalSize);


Indeed, I didn't really look into this function, just removed the redundant semi-colon, but I would actually argue, since rReferenceCoordinates is a BoundedVector of size 6 and we set all values, we don't need to initialize it with a ZeroVector again (on top of that, the only place where we use this function the input is already initialized as a ZeroVector).

I removed it for now, let me know if you agree, or if you think I should bring it back with noalias.

AlejandroCornejo · 2024-07-11T15:25:30Z

applications/StructuralMechanicsApplication/custom_elements/truss_element_linear_3D2N.cpp

@@ -189,20 +189,20 @@ void TrussElementLinear3D2N::WriteTransformationCoordinates(
    BoundedVector<double, TrussElement3D2N::msLocalSize>
    & rReferenceCoordinates)
 {
-    KRATOS_TRY;
+    KRATOS_TRY
    rReferenceCoordinates = ZeroVector(msLocalSize);
    rReferenceCoordinates[0] = GetGeometry()[0].X0();
    rReferenceCoordinates[1] = GetGeometry()[0].Y0();
    rReferenceCoordinates[2] = GetGeometry()[0].Z0();


maybe create a const auto& r_geom = GetGeometry() ?

actually better call only once the .InitialPosition();

Done, thanks for the suggestion!

AlejandroCornejo · 2024-07-11T15:26:42Z

applications/StructuralMechanicsApplication/custom_elements/truss_element_linear_3D2N.cpp

@@ -189,20 +189,20 @@ void TrussElementLinear3D2N::WriteTransformationCoordinates(
    BoundedVector<double, TrussElement3D2N::msLocalSize>
    & rReferenceCoordinates)
 {
-    KRATOS_TRY;
+    KRATOS_TRY
    rReferenceCoordinates = ZeroVector(msLocalSize);
    rReferenceCoordinates[0] = GetGeometry()[0].X0();
    rReferenceCoordinates[1] = GetGeometry()[0].Y0();
    rReferenceCoordinates[2] = GetGeometry()[0].Z0();
    rReferenceCoordinates[3] = GetGeometry()[1].X0();
    rReferenceCoordinates[4] = GetGeometry()[1].Y0();


you can use const auto& r_initial_coords = GetGeometry()[0].InitialPosition();

rfaasse

Thank you for the quick review, I processed the comments, so feel free to have another look!

rfaasse · 2024-07-12T06:51:46Z

applications/StructuralMechanicsApplication/custom_constitutive/truss_constitutive_law.cpp

@@ -141,6 +141,7 @@ void TrussConstitutiveLaw::CalculateMaterialResponsePK2(Parameters& rValues)
    Vector& stress_vector = rValues.GetStressVector();
    if (stress_vector.size() != 1) stress_vector.resize(1, false);
    stress_vector[0] = this->CalculateStressElastic(rValues);
+    AddInitialStressVectorContribution(stress_vector);


Good point, we do not need it for our application at the moment, but added it now for completeness (as well as to the unit test)

rfaasse · 2024-07-12T06:56:53Z

applications/StructuralMechanicsApplication/custom_elements/truss_element_linear_3D2N.cpp

@@ -189,20 +189,20 @@ void TrussElementLinear3D2N::WriteTransformationCoordinates(
    BoundedVector<double, TrussElement3D2N::msLocalSize>
    & rReferenceCoordinates)
 {
-    KRATOS_TRY;
+    KRATOS_TRY
    rReferenceCoordinates = ZeroVector(msLocalSize);


Indeed, I didn't really look into this function, just removed the redundant semi-colon, but I would actually argue, since rReferenceCoordinates is a BoundedVector of size 6 and we set all values, we don't need to initialize it with a ZeroVector again (on top of that, the only place where we use this function the input is already initialized as a ZeroVector).

I removed it for now, let me know if you agree, or if you think I should bring it back with noalias.

rfaasse · 2024-07-12T06:57:47Z

applications/StructuralMechanicsApplication/custom_elements/truss_element_linear_3D2N.cpp

@@ -189,20 +189,20 @@ void TrussElementLinear3D2N::WriteTransformationCoordinates(
    BoundedVector<double, TrussElement3D2N::msLocalSize>
    & rReferenceCoordinates)
 {
-    KRATOS_TRY;
+    KRATOS_TRY
    rReferenceCoordinates = ZeroVector(msLocalSize);
    rReferenceCoordinates[0] = GetGeometry()[0].X0();
    rReferenceCoordinates[1] = GetGeometry()[0].Y0();
    rReferenceCoordinates[2] = GetGeometry()[0].Z0();


Done, thanks for the suggestion!

rfaasse · 2024-07-12T06:57:53Z

applications/StructuralMechanicsApplication/custom_elements/truss_element_linear_3D2N.cpp

@@ -189,20 +189,20 @@ void TrussElementLinear3D2N::WriteTransformationCoordinates(
    BoundedVector<double, TrussElement3D2N::msLocalSize>
    & rReferenceCoordinates)
 {
-    KRATOS_TRY;
+    KRATOS_TRY
    rReferenceCoordinates = ZeroVector(msLocalSize);
    rReferenceCoordinates[0] = GetGeometry()[0].X0();
    rReferenceCoordinates[1] = GetGeometry()[0].Y0();
    rReferenceCoordinates[2] = GetGeometry()[0].Z0();
    rReferenceCoordinates[3] = GetGeometry()[1].X0();
    rReferenceCoordinates[4] = GetGeometry()[1].Y0();


rfaasse · 2024-07-16T07:11:25Z

@AlejandroCornejo Thank you for your review, I processed the comments, could you re-review the updated state of the branch?

avdg81 · 2024-07-25T07:08:12Z

@AlejandroCornejo @KratosMultiphysics/structural-mechanics Just a friendly reminder that we would like to request you to re-review the changes made by @rfaasse in response to earlier code reviews. Thank you.

…al-state-and-stress-calculation-to-trusses

rfaasse · 2024-09-03T14:33:06Z

@AlejandroCornejo @KratosMultiphysics/structural-mechanics Would it be possible to re-review the contents of this PR, since the review comments were processed on July 16th? Thank you in advance!

AlejandroCornejo

I'm very sorry for the delay! I did not see the notification before...

… takes into account initial state (#12902) The Timoshenko beam constitutive law now takes into account any initial state when calculating the Cauchy material response. The changes are in line with previous changes for the truss element (see #12525). Two new C++ unit tests cover the changed behavior.

rfaasse added 5 commits July 10, 2024 11:47

Added the addition of the initial state to the truss constitutive mod…

57d346b

…el and added unit tests

Reverted some refactoring steps because they give issues with the sco…

8788bf0

…pe of the strain and stress vectors (since they are refs in the parameters)

Added a test for the calculation of the PK2 stress

56024fe

Refactored the additions to the truss element

67dc960

Added pre-stress, to make it more comparable to the non-linear variant

1ab70da

rfaasse linked an issue Jul 10, 2024 that may be closed by this pull request

[GeoMechanicsApplication] Adopt truss element 'reset displacement' approach with structural elements #12496

Closed

rfaasse added 2 commits July 10, 2024 14:18

Cleaning up and making the prestress vector 1 length again

e4953a3

Changed license for the new test file

a9d80a0

avdg81 reviewed Jul 11, 2024

View reviewed changes

rfaasse added 2 commits July 11, 2024 11:01

Processed review comments and removed some empty statements (redundan…

a47d09f

…t semi-colons after macros)

Add [[nodiscard]] to a few stub functions in the truss test

2a93bc5

rfaasse marked this pull request as ready for review July 11, 2024 11:07

rfaasse requested a review from a team as a code owner July 11, 2024 11:07

rfaasse changed the title ~~[DRAFT] Add initial state and stress calculation to truss element and constitutive law~~ [StructuralMechanicsApplication] Add initial state and stress calculation to truss element and constitutive law Jul 11, 2024

philbucher requested a review from AlejandroCornejo July 11, 2024 11:59

AlejandroCornejo reviewed Jul 11, 2024

View reviewed changes

Processing review comments

7e28fe4

rfaasse commented Jul 12, 2024

View reviewed changes

rfaasse requested a review from AlejandroCornejo July 12, 2024 07:25

rfaasse self-assigned this Jul 16, 2024

Merge remote-tracking branch 'origin/master' into geo/12496-add-initi…

efd8e60

…al-state-and-stress-calculation-to-trusses

AlejandroCornejo approved these changes Sep 3, 2024

View reviewed changes

rfaasse merged commit d21311e into master Sep 4, 2024
9 of 11 checks passed

rfaasse deleted the geo/12496-add-initial-state-and-stress-calculation-to-trusses branch September 4, 2024 08:36

avdg81 mentioned this pull request Dec 4, 2024

[StructuralMechanicsApplication] The Timoshenko beam constitutive law takes into account initial state #12902

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[StructuralMechanicsApplication] Add initial state and stress calculation to truss element and constitutive law #12525

[StructuralMechanicsApplication] Add initial state and stress calculation to truss element and constitutive law #12525

rfaasse commented Jul 10, 2024 •

edited

Loading

avdg81 left a comment

philbucher commented Jul 11, 2024

AlejandroCornejo Jul 11, 2024

rfaasse Jul 12, 2024

AlejandroCornejo Jul 11, 2024

rfaasse Jul 12, 2024

AlejandroCornejo Jul 11, 2024

AlejandroCornejo Jul 11, 2024

rfaasse Jul 12, 2024

AlejandroCornejo Jul 11, 2024

rfaasse Jul 12, 2024

rfaasse left a comment

rfaasse Jul 12, 2024

rfaasse Jul 12, 2024

rfaasse Jul 12, 2024

rfaasse Jul 12, 2024

rfaasse commented Jul 16, 2024

avdg81 commented Jul 25, 2024

rfaasse commented Sep 3, 2024

AlejandroCornejo left a comment

[StructuralMechanicsApplication] Add initial state and stress calculation to truss element and constitutive law #12525

[StructuralMechanicsApplication] Add initial state and stress calculation to truss element and constitutive law #12525

Conversation

rfaasse commented Jul 10, 2024 • edited Loading

avdg81 left a comment

Choose a reason for hiding this comment

philbucher commented Jul 11, 2024

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

rfaasse left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

rfaasse commented Jul 16, 2024

avdg81 commented Jul 25, 2024

rfaasse commented Sep 3, 2024

AlejandroCornejo left a comment

Choose a reason for hiding this comment

rfaasse commented Jul 10, 2024 •

edited

Loading