View Issue Details
|ID||Project||Category||View Status||Date Submitted||Last Update|
|0003292||OpenFOAM||Bug||public||2019-06-12 22:15||2019-06-14 15:00|
|Fixed in Version|
|Summary||0003292: Incorrect filtering and expression in LES dynamicKEqnModel|
|Description||In the referred paper "Kim, W and Menon, S. (1995). A new dynamic one-equation subgrid-scale model for large eddy simulation. In 33rd Aerospace Sciences Meeting and Exhibit, Reno, NV, 1995.", the expression for L_ij, M_ij and Ck (or Cv as in the paper) are given in equations 14 to 16. The corresponding expressions in the code (dynamicKEqn.C) involve an additional filter, |
e.g. for M_ij, lines 50-53 in the code are:
const volSymmTensorField MM
, but we think the simpleFilter_ used in the front is not required as per the expression in the paper (Eq. 15). Also, in the formulation found in the code, the two terms on the RHS of Eqn 15 are different from the above expression.
We think this usage of the additional filter can have an additional damping effect on the test scale motions.
|Tags||No tags attached.|
Do you have any cases or results which demonstrate that the additional filter reduces the accuracy of the model? My understanding is that it was added to improve stability and range of application but if you can provide results for a range of cases which show that it is not needed and/or not appropriate I would be happy to remove it.
> Also, in the formulation found in the code, the two terms on the RHS of Eqn 15 are different from the above expression.
What other changes do you want to make to the model? Do you have a complete and tested alternative implementation?
||I'll upload comparison cases within a couple of days. I guessed that the use of the additional filter must be for enhancing stability, but couldn't get why the RHS of Eqn 15 was changed significantly.|