Repository navigation
RBE3: judge the rank of the retained REFC system with a relative tolerance - #340
Open
Valladarex wants to merge 1 commit into
Open
Valladarex wants to merge 1 commit into
Valladarex wants to merge 1 commit into
Conversation
…rance RBE3_PROC (RIGID_ELEMENT_PROCESSING) checked the reduced system of the retained REFC components with DGELSY at RCOND = EPS1 (machine precision). A system that is singular in exact arithmetic comes out with sigma_min/sigma_max of 1e-16 to 1e-12 from round-off, so identical elements were accepted or rejected (ERROR 1957) depending on their orientation. Example: two weighting grids with 123, REFC 123, the reference grid off the line through them: rotation about that line is not determined by the weighting grids, so the reference grid's motion is not either. Of 24 orientations of that element, current dev rejects 10 and solves 14 with an arbitrary reference-grid motion. The final rank check now uses RANK_RCOND = 1.0E-10; the elimination of the discarded components keeps EPS1. Such an element is now rejected (ERROR 1957) in every orientation. Well-posed RBE3s (the reference grid on the line with REFC 123, or three non-collinear weighting grids with REFC 123456) are accepted in every orientation, as before. Test (MYSTRAN_Validation): rbe3_rank_singular_orient2.bdf, that element in an orientation current dev accepts: the run must stop and write no displacements.
Collaborator
|
Looks like this just loosens the tolerance. Wouldn't that move the threshold for orientation dependent failure somewhere else? Or is the point of it that the tolerance was too tight before and allowing bad RBE3s to solve? |
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.
Fixes #339.
One commit, in
Source/LK1/L1D/RIGID_ELEMENT_PROCESSING.f03.Behaviour change, stated plainly: some models that ran on
devnow stop.devsuch an RBE3 was rejected or accepted depending on its orientation, and when accepted it gave an arbitrary reference-grid motion.Test: MYSTRAN_Validation MystranSolver/MYSTRAN_Validation#46,
rbe3_rank_singular_orient2.bdf: the undetermined element in an orientationdevaccepts.devit solves. With this PR it stops with ERROR 1957.Also checked on this commit alone:
dev: 10 of 24);dev;Validation suite: with both PRs, 1 failure of 3,256 checks, the same one current
devhas without them ("mitc4 thermal strain free rectangle cylinder"). Every existing RBE3 case passes, and the new check passes.