Skip to content
Snippets Groups Projects

Feature iso advector AMR

Merged Johan Roenby requested to merge feature-isoAdvector-AMR into develop-pre-release
All threads resolved!
  1. interIsoFoam rewritten based on the new interFoam including DyM functionality.
  2. isoAdvection, isoCutFace and isoCutCell classes modified to work with AMR (dynamicRefineFvMesh).
  3. Refactoring of isoAdvection, isoCutFace and isoCutCell (not related to AMR functionality).
  4. Included damBreakWithObstacle test case for interIsoFoam with dynamicRefineFvMesh.
  5. Added discInConstantFlowCyclicBC case for interIsoFoam to confirm proper behaviour with cyclic BC's (not AMR related).

I have tested the changes by compiling, running all interIsoFoam tutorials and verified visually that everything looks OK.

Known issue: isoCutFace sometimes gives a warning related to an edge being cut multiple times by the isoFace (There are around 20 such warnings in the damBreakWithObstacle log). It does not seem to have any effect on the results. I have not so far been able to locate the cause of these warnings.

Best regards Johan

Edited by Andrew Heather

Merge request reports

Loading
Loading

Activity

Filter activity
  • Approvals
  • Assignees & reviewers
  • Comments (from bots)
  • Comments (from users)
  • Commits & branches
  • Edits
  • Labels
  • Lock status
  • Mentions
  • Merge request status
  • Tracking
  • Andrew Heather
  • Andrew Heather
  • Johan Roenby added 1 commit

    added 1 commit

    • fbed78f6 - Updated tutorial headers. Added copyright note to isoAdvector src. Removed…

    Compare with previous version

  • I now updated the headers and cleaned the code. See my new commit.

  • I tried to write out alpha1.oldTime() and alpha1.prevIter() in the damBreakWithObstacle case and confirmed that these fields "live" on the mesh of the current time step so that adding alpha1, phi.oldTime() and alpha1.prevIter() makes sense.

    I am currently rewriting alphaEqn.H to use prevIter in case of nOuterCorrectors > 2 instead of defining alpha10, U0 and phi0 in createFields.H. This will avoid the doubling of the number of fields in cases where nOuterCorrectors = 1.

  • Johan Roenby added 1 commit

    added 1 commit

    • ddc5e043 - Reimplemented treatment of alpha1, phi and U in case of nOuterCorrectors > 1…

    Compare with previous version

  • Andrew Heather added 4 commits

    added 4 commits

    • 1da89da4 - INT: Initial update of isoAdvector and interIsoFoam to work with AMR.
    • eb3a069c - TUT: Modified tutorials and added cyclic test case.
    • e3a22868 - ENH: multiple updates to interIsoFoam related code
    • 63239d15 - ENH: interIsoFoam updates

    Compare with previous version

  • Andrew Heather unmarked as a Work In Progress

    unmarked as a Work In Progress

  • Andrew Heather changed title from WIP: Feature iso advector amr to Feature iso advector AMR

    changed title from WIP: Feature iso advector amr to Feature iso advector AMR

  • Andrew Heather changed the description

    changed the description

  • Andrew Heather changed milestone to %v1806

    changed milestone to %v1806

  • Andrew Heather added 59 commits

    added 59 commits

    • 63239d15...55679efd - 55 commits from branch develop-pre-release
    • fb32b109 - INT: Initial update of isoAdvector and interIsoFoam to work with AMR.
    • 26e3f7e2 - TUT: Modified tutorials and added cyclic test case.
    • ac8411e6 - ENH: multiple updates to interIsoFoam related code
    • 0840aecf - ENH: interIsoFoam updates

    Compare with previous version

  • Andrew Heather resolved all discussions

    resolved all discussions

  • Andrew Heather mentioned in commit 17f26c6c

    mentioned in commit 17f26c6c

  • Please register or sign in to reply
    Loading