3D plots for measure of turbulence resolution inside FDS

[Muthu-V]

Inside FDS there is an output quantity for 2D/3D slice called resolved kinetic energy (filtered by grid?) and subgrid kinetic energy (modelled). If that 2D/3D slice data of both these variables are time averaged, can it be substituted in the measure of turbulence resolution (MTR) equation as given below to check the grid resolution for turbulence in 3D field?

Thanks

[rmcdermo]

Sorry for the slow reply. Yes, this idea should work. I would double check things again DEVC.

I also aways caution about using MTR as a final metric. It is neither necessary nor sufficient to prove a simulation is adequate.

[Muthu-V]

No worries. Thank you for your comments, Randy. I understand several other factors to be considered as well for simulation adequacy. But this gives an idea about how much TKE is resolved by the grid in the area of interest (asper the classical definition of LES 80 % TKE in the inertial subrange) and how much is modelled. When there are experimental measurements to compare with simulation results, for example checking RMS fluctuations in the wake regions behind obstacles in atmospheric boundary layer, it gives some insights about the grid resolution applied to that flow and to address the deviations.

[Vojta-Salek]

@Muthu-V, have you tried comparing the RESOLVED KINETIC ENERGY defined in the Guide as:

to the TKE you mentioned above:

calculated using devices?

Thank you!

[rmcdermo]

@Vojta-Salek For a flow with non-zero ubar, vbar, wbar (almost any real flow), these are of course different quantities. The theory really only applies to the homogeneous isotropic turbulence case, which never truly exists, which is why I don't get too hung up on this metric. I don't think it would be very easy to get TKE from slices. You'd have to use pyfdstools or similar to basically do what the devices would do.

[Vojta-Salek]

@rmcdermo Thank you for the clarification!

[Muthu-V]

Thank you @rmcdermo