[GeomechanicsApplication] Make quasi newton available for dsettlement#14547
Draft
aronnoordam wants to merge 35 commits into
Draft
[GeomechanicsApplication] Make quasi newton available for dsettlement#14547aronnoordam wants to merge 35 commits into
aronnoordam wants to merge 35 commits into
Conversation
Temporarily disabled the CROW case tests with Mohr-Coulomb materials. It now runs the model with linear elastic materials only.
The goal is to have a single set of expected results for this case, regardless of the adopted solution method.
This extra subdirectory is meant to distinguish between variants.
… as D-Sheet Piling These results are independent of the variant that is being run.
Solutions obtained with Broyden and L-BFGS are now compared to the same expected results. To make this work, we had to relax the applied relative tolerance. In addition, we had the supply an appropriate absolute tolerance, which is calculated using the maximum absolute result value and the relative tolerance.
Also reverted most of the convergence settings back to what they were when we were still using linear iteration, except for stage 6 (i.e. the second excavation), since otherwise Broyden's method would not converge. Also relaxed the relative tolerance used for checking the obtained results from 1% to 2%.
In order to obtain a properly converged solution, we had to increase the maximum number of iterations in stage 6 (i.e. the second excavation). Now, for all three solution methods (i.e. linear iteration, Broyden and L-BGFS), we obtain the same solution within a tolerance of 2%.
The extracted function returns the maximum absolute value from a list of nodal values.
Also reordered a few statements
It's back at using Newton-Raphson (or, more precisely, linear iteration). When testing both quasi-Newton methods, the test script replaces/adds the appropriate settings.
It now also (optionally) receives a function object that modifies the project parameters.
An optional analysis runner can be supplied, which is useful for the test case with save-and-load. If not provided, it will use the orchestrator-based approach to run a multi-stage analysis.
Also moved the test case with linear iteration a bit up.
Also updated the expected results of the model with Mohr-Coulomb materials.
…' into geo/quasi_newton_raphson # Conflicts: # applications/GeoMechanicsApplication/tests/crow_validation/common/staged_construction.json
aronnoordam
changed the title
aronnoordam
changed the title
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Make quasi newton available for dsetlement