LES always 3D?
 

Hi. I'm trying to understand why LES is not valid for 2D solutions. Can someone explain with details?

Re: LES always 3D?
 

For starters, LES requires the modeling be done in 3-D because it is NOT isotropic in nature as in many RANS (for which 2-D simulations are permitted) models. And since turbulence is inherently 3-D, 3-D simulations in LES must be performed for acceptable results.

Re: LES always 3D?
 

You are trying to simulate turbulence in fluid flows, with the dominant mode of energy transfer between the scales being vortex stretching. This is a 3-D phenomenon and thus you have to model it in 3-D.

Incidentally, in 3-D turbulence the energy moves from large scales to small. In 2-D turbulence - you can consider Earth's atmosphere as an example - the enery transfer is in the opposite direction because vortex stretching does not occur. This is why you get weather systems in the atmosphere: large-scale vortices which last for a long time.

Hrv