As far as I know in order to solve the Navier Stokes partial differential equations with Computational Fluid Dynamics - Large Eddy Simulation you need to construct a computational domain and to provide initial and boundary conditions at the planes of the computational domain.
This issue is still object of research in the LES community and there is no a unique approach. I can suggest to read in the book Sagaut and the cited papers.
This way is quite simple but you need a very long inlet to let the flow correlate physically...but in some case where the separation is driven by the geometry (e.g. the backward facing step) that can work quite well