[mali28]

I am using Fluent to do simulation of air flow in a geometry. I want to compare the turbulence level obtained from RANS models with LES model. In RANS I can get turbulence intensity, kinetic energy etc.. but in LES which parameters do I use?
Fluent has resolved UV, VW, UW Reynolds stresses for LES but I do not understand what this is.

[LuckyTran]

Originally Posted by mali28

Fluent has resolved UV, VW, UW Reynolds stresses for LES but I do not understand what this is.

If you do not understand what resolved UV, VW, UW Reynolds stresses are then you do not understand how LES works and I would encourage you to slow down, stop what you are doing, learn the basics, and only then proceed.

Turbulence intensity is just a normalized turbulent kinetic energy. If you know what turbulent kinetic energy is, it is straight forward to calculate the resolved part of the turbulent kinetic energy from the filtered velocities (the RMSE X-velocity, Y-velocity, and Z-velocity). Alternatively you can also back-calculate it from the resolved part of the Reynolds stresses but in my opinion it is already easy enough to do.

The tricky part is getting the contribution of the turbulent kinetic energy of the subgridscales. It depends slightly on what subgrid scale model you are using. There's a very simple algebraic formula that involves a filter length and subgridscale turbulent visosity (it's basically the subgridscale model itself), but it only makes sense if you know how LES works in the first place. For Smagorinsky, the subgrid kinetic energy is the turbulent viscosity divided by the sgs filter length and then squared.

First you must understand what turbulent intensity is, what turbulent kinetic energy is, what LES does, and what is the output of LES. It's my personal opinion, but I refuse to teach people how to do anything if they don't understand what it is or how it works because they often go on to draw naive conclusions. And then they teach it to others.