LES smagorinsky and SST turbulence

drsidd10 · January 9, 2015, 16:22

What differences do we get when simulating a tidal turbine blade geometry using SST (steady state) and LES-Smagorinsky (transient)?

The current problem has been simulated for 2.5m/s inlet velocity, what difference would I get if I were to simulate the same problem for 1.5m/s or even 3m/s inlet velocity for both the analyses?

Chris_321 · January 10, 2015, 04:59

If you simulate with RANS(+ SST Turbulence model), the set of equations your solving is (because of the high instationary behavior of turbulence) time average . This is done by saperate the velocity in an average and a fluctuating part. u = u_ave + u_fluc

To close the set of equations some modeling approaches are required( Turbulence models like SST).

Thats why you will get a time average solution of the velocity field in your simulation when simulating with RANS.

The LES simulates is not time average, it solves the big turbulance scales directly, while modeling the small ones.

https://www.youtube.com/watch?v=eP_eytKCCFY

Before starting to simulate with LES keep in mind that you will need a finer mesh and a small timestep, based on the turbulence timescale. A rule of thumb for les is that it takes the solver around 100 times longer than with rans. What works best for your simulation depends on what you are interessting in.

January 9, 2015, 16:22	LES smagorinsky and SST turbulence	#1
drsidd10 New Member Siddharth Kulkarni Join Date: Oct 2010 Location: Birmingham, UK Posts: 6 Rep Power: 15	What differences do we get when simulating a tidal turbine blade geometry using SST (steady state) and LES-Smagorinsky (transient)? The current problem has been simulated for 2.5m/s inlet velocity, what difference would I get if I were to simulate the same problem for 1.5m/s or even 3m/s inlet velocity for both the analyses?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
k-omega SST SAS High Wavenumber Damping Limiter	PonchO	OpenFOAM Programming & Development	5	March 21, 2016 12:01
LES required y+	zoozoozoo	Main CFD Forum	6	June 12, 2010 07:25
Tutorials for incompressible flow and different turbulence models	Sandu69	OpenFOAM	2	April 20, 2010 12:17

January 10, 2015, 04:59		#2
Chris_321 Member Christian Join Date: Sep 2013 Location: Germany Posts: 88 Rep Power: 12	If you simulate with RANS(+ SST Turbulence model), the set of equations your solving is (because of the high instationary behavior of turbulence) time average . This is done by saperate the velocity in an average and a fluctuating part. u = u_ave + u_fluc To close the set of equations some modeling approaches are required( Turbulence models like SST). Thats why you will get a time average solution of the velocity field in your simulation when simulating with RANS. The LES simulates is not time average, it solves the big turbulance scales directly, while modeling the small ones. https://www.youtube.com/watch?v=eP_eytKCCFY Before starting to simulate with LES keep in mind that you will need a finer mesh and a small timestep, based on the turbulence timescale. A rule of thumb for les is that it takes the solver around 100 times longer than with rans. What works best for your simulation depends on what you are interessting in.