RANS Vs LES

kamal tewari · December 17, 2016, 13:20

My Question is , will LES give the same field after averaging over time(any flow field Quantity) as given by a RANS model?
I am trying to simulate jet flows and I am getting matched profiles with experiments in case when I used RANS Model(K-omega) but when I applied averaging over time in LES then i was not able to get same results.

flotus1 · December 17, 2016, 13:52

A LES -when performed correctly- is supposed to yield the same statistics as a DNS or an experiment. Not necessarily the same statistics that a RANS method would produce because of the inherent flaws of this approach.

FMDenaro · December 17, 2016, 13:54

The two fields are not exactly the same... for RANS you compute directly for <v> while for LES, after you average the solution you get <v_bar>.
If the difference is disregardable or not would depend on many factors.

kamal tewari · December 17, 2016, 14:31

Thank u Sir , for your Quick Reply

But I am little confused as I want this difference to be as small as possible. I am trying to model the jet noise for which I need to do LES to resolve audible range of frequency emitted from the jet. When I applied my boundary conditions initially For RANS model ( Free stream Exterior, Nozzle as Free Stream inlet) I got the flow profile quiet close to experimental data, on the other hand the averaging over time done in LES gave different results. According to you what parameters should I change(experiment) in an standad LES model as it is very time consuming.

FMDenaro · December 17, 2016, 15:31

The difference exists always by a theoretical point of view and depends on the size of the filter in LES. It can or not be highlighted by the statistics.

However, I first suggest to check carefully that your LES solution is good!
Therefore: check the grid resolution, use the dynamic SGS model, use a high order discretization (space and time), be careful to compute the time-averaging only after the numerical transient (due to the numerical initial condition) is disregarded, perform the averaging on a sufficient number of samples.

It would be useful if you post the results in terms of plot of RANS vs. LES

December 17, 2016, 13:20	RANS Vs LES	#1
kamal tewari Member Polard Join Date: Oct 2015 Posts: 31 Rep Power: 11	My Question is , will LES give the same field after averaging over time(any flow field Quantity) as given by a RANS model? I am trying to simulate jet flows and I am getting matched profiles with experiments in case when I used RANS Model(K-omega) but when I applied averaging over time in LES then i was not able to get same results.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Model Switch, RANS to LES?	jennap223	STAR-CCM+	0	July 7, 2015 12:30
Transition RANS to LES during simulation	LorenzN	OpenFOAM Running, Solving & CFD	4	November 17, 2011 11:47
Laminar Flamlet RANS to LES	LB	FLUENT	0	September 5, 2007 06:35
Mapping RANS data onto an LES	christian	OpenFOAM Running, Solving & CFD	0	April 13, 2007 06:31
RANS results for LES	Li Yang	Main CFD Forum	5	June 17, 2002 06:34

December 17, 2016, 13:52		#2
flotus1 Super Moderator Alex Join Date: Jun 2012 Location: Germany Posts: 3,427 Rep Power: 49	A LES -when performed correctly- is supposed to yield the same statistics as a DNS or an experiment. Not necessarily the same statistics that a RANS method would produce because of the inherent flaws of this approach.

December 17, 2016, 13:54		#3
FMDenaro Senior Member Filippo Maria Denaro Join Date: Jul 2010 Posts: 6,897 Rep Power: 73	The two fields are not exactly the same... for RANS you compute directly for <v> while for LES, after you average the solution you get <v_bar>. If the difference is disregardable or not would depend on many factors.

December 17, 2016, 14:31		#4
kamal tewari Member Polard Join Date: Oct 2015 Posts: 31 Rep Power: 11	Thank u Sir , for your Quick Reply But I am little confused as I want this difference to be as small as possible. I am trying to model the jet noise for which I need to do LES to resolve audible range of frequency emitted from the jet. When I applied my boundary conditions initially For RANS model ( Free stream Exterior, Nozzle as Free Stream inlet) I got the flow profile quiet close to experimental data, on the other hand the averaging over time done in LES gave different results. According to you what parameters should I change(experiment) in an standad LES model as it is very time consuming.

December 17, 2016, 15:31		#5
FMDenaro Senior Member Filippo Maria Denaro Join Date: Jul 2010 Posts: 6,897 Rep Power: 73	The difference exists always by a theoretical point of view and depends on the size of the filter in LES. It can or not be highlighted by the statistics. However, I first suggest to check carefully that your LES solution is good! Therefore: check the grid resolution, use the dynamic SGS model, use a high order discretization (space and time), be careful to compute the time-averaging only after the numerical transient (due to the numerical initial condition) is disregarded, perform the averaging on a sufficient number of samples. It would be useful if you post the results in terms of plot of RANS vs. LES