Coflowing-rectangular Jet, Adaptive Mesh Refinement (AMR) | Canonical Flows 3 (OpenFOAM)
In this video a rectangular, coflowing jet was studied. The effects of the initial mesh resolution and adaptive mesh refinement where studied. In the table shown you can see the cases that where investigated, 4 without and 3 with AMR.
OpenFOAM case parameters
mesh: blockMesh
solver: pimpleFoam
turbulence model: KOmegaSST
Adaptive Mesh Refinement (AMR) is the process of dynamically adapting the accuracy of a solution in certain areas of interest specified via some criteria.
The idea of AMR is to get the simulation accuracy of a fine mesh while keeping the computational costs as low as possible.
For this simulation I injected dye (or a passive scalar) through one of the inlets, this way I use the interface of the two streams as my refinement criteria. This method work particularly well for this simulation, but it may be necessary to use another criteria, like turbulent viscosity, in other cases.
The cases and video presented are for educational purposes and should serve as a launching point in conducting and setting up your investigation.
Case Files: https: github.com/Interfluo/OpenFOAM...
Consider supporting me on Patreon: / interfluo
#openfoam #cfd #aerospace # engineering #fluids #paraview #fluiddynamics #computationalfluiddynamics
Пікірлер: 5
Timeline: 0:00 - Introduction 1:23 - Cases and Set-up 2:09 - blockMesh 3:55 - Boundaries 4:29 - Results 5:00 - A bit on AMR 5:18 - Scalar transport animations 6:20 - Velocity Profiles 7:50 - Closing Remarks
Really good content. Thank you for sharing. I get that computational cost is reduced when using AMR, but what is the results and time difference from using AMR with something like the kEpsilon or KOmegaSST model versus doing a full LES simulation. I know RANS is much better computationally, but what about quantitative values?
@interfluo6420
2 жыл бұрын
Thanks for the comment, it is kind of a complicated topic that you bring up. I am by no means an expert in LES, but as far as I am familiar the decisive parameter for the mesh resolution is the turbulent kinetic energy compared to RANS where the mesh is resolving gradients. Because of this the answer is highly dependent on your specific case, and I can't really give quantitative values without running a case (which is definitely somewhere in the pipeline).
Hi, could you kindly explain how to make the contour? it is very professional...
When I run the code, I get the error below, How can I fix it? --> FOAM FATAL IO ERROR: Essential value for keyword 'field' not set In function entry: scalarTransport In file: /Desktop/OpenFoam/OpenFOAM-Cases-Interfluo-main/Canonical-Flows/jet/case1/system/controlDict at line 56 The function entry should be: scalarTransport(field = )