I'm doing fundamental investigation into the flat plate turbulent boundary layer using LES in OpenFoam. which parameter would lead us to better near-wall results in terms of accuracy and cost?
1- Wall-layer models (cost effective)
2- Isotropic and homogeneous assumption in the near-wall region while the nature of the flow is inhomogeneous and anisotropic
3- Switch into hybrid RANS/LES or other models
It should be noted that grid independent study has been done and y+ is fine enough
Try to read that paper, I think it will help you so much.Pleas, copy and past
https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0ahUKEwia3KuagZ3TAhXHtRQKHZADB-8QFggeMAA&url=http%3A%2F%2Fwww.iaeng.org%2Fpublication%2FIMECS2009%2FIMECS2009_pp2165-2170.pdf&usg=AFQjCNFs4sr9Iv4q5-UvpGAq5VdDbFTxMQ&sig2=iwwXVu2U47uIfu7mjD10RQ
Wall function is not related to LES. As far as I know, it is is a concept in RANS and especially in k-e model.
inhomogenous and anisotropic assumptions are not assumtions. they are reality of flow.
in the case of turbulence model, you dont need to use turbulence model in LES. you should use SGS models for scales smaller than your cut-off wave numbers.
Y+ should not be a large number. the y+ should be less than 5 to capture laminar region near the wall in turbulent flow.
Dear Ali,
Please have a look to this article.
"Some insights for the prediction of near-wall
turbulence "by:
Farid Karimpour and Subhas K. Venayagamoorthy
Hope this helps.
Please refers paper AIAAJournal, Vol. 24, No. 7, July 1986, pp. 1207-1209.
Dear Foad Vashahi,
I really appreciate your response. Yeah, you are right. But I am trying to make my simulation cost effective so I have applied a wall layer model. In this case, it can be treated as a wall-function.
Dear Hamid Reza Karbasian,
I really appreciate your clear explanations. I am using wall layer models to make my simulation cost effective (missed in the question). I have already modified my question.
Regarding the inhomogeneous and anisotropic, I meant we assume isotropic in the near wall while the reality is anisotropic.
It depends strongly on the goal of the study...if you are interested in the physics of the spatial evolving boundary layer along the flat plate, then you have to resolve the BL using a fine grid such that you set 3-4 nodes for y+
- Badges
- Science topic