We are conducting a study on katabatic winds. We are trying to simulate the generation of this kind of winds using CFD software (mainly ANSYS-Fluent) but we have not been able to obtain any reasonable results so far.
Applying very simple boundary conditions to emulate a free atmosphere setting (regular temperature and pressure gradient with no initial wind speed), yields very unreasonable results (even with a very simple geometry like the one shown in the attached images) such as incredibly high wind speed in random directions. This is obviously very far from the "natural convection" flow that one would expect.
Does anyone have any experience with CFD studies of wind flow in a free atmosphere setting? Is Ansys Fluent just not suitable enough for this kind of studies? What other alternatives can you think of to conduct a study like this?
Thank you very much
Thanks. I know that forum since a long time ago. It has been very useful many times before. We have tried posting also the same question there, but so far we have not succeeded in having a reply that solves our problems.
Physically based equations describing transient diffusion induced flows (analogues of katabatic winds in a stratified fluid at rest), their analytical or numerical solutions for a number of geometries and data of laboratory observations are presented in papers of A.V.Kistovich, V.G.Bydulov, I.V.Zagumennyi, N.F. Dimitrieva written together with Yuli D Chashechkin (Available on R/g). Please read, think and adopt.
Sincerely,
Yuli D. Chashechkin
I think that the height of domain is too short (judged by your figure), as the top wall of the domain will affect the velocity field. You may try to increase its height or apply a symmetry wall condition to the top wall.
- Badges
- Science topic