CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   Turbulence Model L.E.S (https://www.cfd-online.com/Forums/fluent/65615-turbulence-model-l-e-s.html)

hamoudi June 21, 2009 06:12

Turbulence Model L.E.S
 
Good Moning Everybody,

Please if every one can answer me, i am doing a simulation in 2D
Can i use L.E.S as a turbulence Model in 2D simulation
because L.E.S turbulence model is very robust

Thank you in advance

ak6g08 June 21, 2009 17:03

Hi,

Turbulence is inherently a three-dimensional phenomenon, i.e. the creation of large scale motion and then dissipation into small eddie scales can only be resolved in 3 dimensions. Modelling such as RANS, averages the navier stokes equations...i.e. it does not solve the NS equations directly, it only provides an average picture of what is going on and so is acceptable to use in 2D. LES however is a hybrid between RANS and DNS (direct numerical simulation), the latter of which solves the navier stokes equations directly. What LES does is, it averages the small scales, and directly resolves the large scales, to increase computation efficiency however should NEVER be used for 2D. I honestly dont understand why FLUENT even offer this model in two dimensions...read any turbulence book out there and you will agree with me. Moral of the story, if modelling turbulence in 2D, only use a fully averaged model i.e. RANS otherwise you will be calculating rubbish, if in 3D use whatever your PC can take...but DNS is not applicable for 'industrial' simulations, because the grid size scales with the Reynolds number so u would need millions and millions of cells to accurately capture something.

hope that helps.

AK


All times are GMT -4. The time now is 23:20.