Skip to content
GitLab
Projects Groups Snippets
  • /
  • Help
    • Help
    • Support
    • Community forum
    • Submit feedback
    • Contribute to GitLab
  • Sign in / Register
  • openfoam openfoam
  • Project information
    • Project information
    • Activity
    • Labels
    • Members
  • Repository
    • Repository
    • Files
    • Commits
    • Branches
    • Tags
    • Contributors
    • Graph
    • Compare
  • Issues 418
    • Issues 418
    • List
    • Boards
    • Service Desk
    • Milestones
  • Merge requests 7
    • Merge requests 7
  • Deployments
    • Deployments
    • Releases
  • Wiki
    • Wiki
  • Activity
  • Graph
  • Create a new issue
  • Commits
  • Issue Boards
Collapse sidebar
  • Development
  • openfoamopenfoam
  • Merge requests
  • !454

ENH: turbulentDFSEMInlet: various improvements

  • Review changes

  • Download
  • Email patches
  • Plain diff
Merged Kutalmış Berçin requested to merge feature-normalisation-dfsem into develop May 19, 2021
  • Overview 2
  • Commits 3
  • Changes 87

Summary

  • ENH: Input entries of R, L and U are now PatchFunction1 type.
  • ENH: Adds two scalar normalisation factors for input R, L and U: Uref and Lref (default=1).
  • ENH: Adds a scalar factor to enable users to tune the C1 normalisation coefficient: scale (default=1).
  • BUG: MappedFile: separates scopes of entries.
  • TUT: Replaces chan395DFSEM and PCF tutorials with planeChannel and oneCellThickPlaneChannel tutorials.
    • Corrects the integral-length scale input files.
    • Updates the Extended Code Guide's channel flow case, and allows users to reproduce the results.

Resolved bugs (If applicable)

#2098 (closed) #2097 (closed) #2090 #1004 #1744 #2089 (closed)

Details of new models (If applicable)

A set of results obtained from the planeChannel tutorial (and its plot script available to users) with a different set of settings can be seen below:

Ruv vs y

Ruv_vs_y

u vs y

u_vs_y

Ruu vs y

Ruu_vs_y

x vs Cf

x_vs_cf

Rvv vs y

Rvv_vs_y

Rww vs y

Rww_vs_y

Risks

User input

Users (my apologies) will have to change the input syntax for R, L and U entries. For example, for reading the input files in constant/boundaryData/inlet/0/, users were using the following minimal syntax:

inlet
{
    type            turbulentDFSEMInlet;
    delta           1;
    mapMethod       nearestCell;
    value           $internalField;
}

For the same effect, one of the syntax examples could be as follows:

inlet
{
    type            turbulentDFSEMInlet;
    delta           1;

    U               mappedFile;
    R               mappedFile;
    L               mappedFile;
    mapMethod       nearest;

    value           $internalField;
}

or as follows:

inlet
{
    type            turbulentDFSEMInlet;
    delta           1;
        
    U
    {
        type        mappedFile;
        mapMethod   nearest;
    }
    R
    {
        type        mappedFile;
        mapMethod   nearest;
    }
    L
    {
        type        mappedFile;
        mapMethod   nearest;
    }

    value           $internalField;
}

Users can, however, use any of the PatchFunction1 input syntax exemplified in this link and this link.

Regressions
  • The default value of nCellPerEddy entry was changed from 5 to 1, which is the default value corresponding to the original journal paper.
  • The default value of nCellPerEddy entry was kept the same as 5.
Remaining issues
  • Arguably, the DFSEM method has various theoretical issues that need to be addressed (see #2090). We have been trying to reach the originators of the method during the last couple of years; and we will continue our efforts, if possible, to cooperate with them to resolve some of these issues.
Edited Jun 05, 2021 by Kutalmış Berçin
Assignee
Assign to
Reviewers
Request review from
Time tracking
Source branch: feature-normalisation-dfsem