Ask for help! In the les turbulent model, how to deal with the wall boundary? And if the y+ is needed , how to calculate it?

In the boundary layer, you must use DNS resolution, in other words, y+ < 1. (see any standard book on turbulence for y+ calculation)

Wall models are an alternative, and you might want to try them.

agg, can talk about the calculation of y+ in les ? The y+ fomulation need k, but how to get it in les??

y+ is just a non-dimensional distance based on friction velocity

y+ = y*utau/viscosity utau = (wall shear stress/density)^0.5

You can calculate wall shear stress from LES using velocity gradient at the wall and hence y+

agg, you mean the wall shear stress can get by using velocity gradient at the wall, so must use DNS resolution, in other words, y+ < 1??? But when the y+>30, the wall shear stress is diff to get?

When y+ is not resolved, you can use wall models.

An example would be to use:

1) u+ = y+ in the viscous sublayer (if y+ <= 10) 2) u+ = 2.5 ln(y+) + 5.5 (if 10 < y+ <= 300)

For (2), an iterative procesure is required, since wall stress is an unknown in both LHS and RHS.

Once u+ is known, you can back calculate u at that point.

Best information can be found in papers. You should be able to find papers with LES using wall models.

Visit here http://www.lstm.uni-erlangen.de/~breuer/ and see this

CFD - Code LESOCC (Large Eddy Simulation on Curvilinear Coordinates)

Numerical Method

Discretization

Subgrid Scale Model

=== Wall Boundary Conditions ===

Good Luck!

Halim

i am reading, thanks halim!

hi....halim

the Code LESOCC is shared?

where can get some code for learning?

(1) Ask to Dr. Breuer (E-mail: breuer_at_lstm.uni-erlangen.de )

(2) Sorry, I do not know. I know that if you buy this book, you can have a source code for LES. "Fluid Flow Phenomena : A Numerical Toolkit" by Paolo Orlandi, Kluwer Academic Publishers, 2000.

Halim