CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   Turbulence Intensity Vs CFD Simulation (

Apple L S Chan December 14, 1998 04:24

Turbulence Intensity Vs CFD Simulation
Turbulence intensity of a point inside space can be obtained by using measured velocities of that point varied with time. However, in CFD simulation, only steady state results of velocity, temperature, pressure, turbulence kinetic energy and dissipation rate of turbulence kinetic energy are calculated. Could anyone please tell me if there is any direct relationship/equation linking turbulence intensity and CFD simulated results, i.e. can I use the CFD simulated results to calculate or estimate the turbulence intensity at any point of the indoor space.

Thank you!

Gary Dantinne December 14, 1998 04:40

Re: Turbulence Intensity Vs CFD Simulation

At first, for me turbulence intensity and turbulence kinetic energy are the same so it would be fine if you define what you mean by turbulence intensity.

Second, I am not sure that measuring turbulence intensity at a specified location at a specified time will give useful information. You need mean values in order to compare efficiently different approaches to study turbulence. Therefore it seems to me that only "mean (in time or in space or ...) turbulence intensity" (rms of the velocity fluctuations) deserves to be measured and compared to theory or numerical computations. And this quantity (the rms of the velocity fluctuations) can in principle be found using classical CFD tools.

Hope it helps,


Fabien Coppens December 14, 1998 04:48

Re: Turbulence Intensity Vs CFD Simulation
If I may, I'd like to add this : with CFD you do not only get steady states !! You can run temporal simulations (time-marching) of unsteady phenomena, with either RANS or DNS/LES approaches. It is then possible to carry out local spatial averaging of flow variables at each time step, thus obtaining an "average" and a "fluctuating" velocity field, which allow you to define a form of turbulent kinetic energy. It is also theoretically possible to carry out spatial simulations (marching in space instead of in time) and you can then carry out your time-averaging process just like in the "classical" definition of stochastic averaging in the "turbulent" Navier-Stokes equations. Hope I was clear enough !


Erwin Platvoet December 15, 1998 18:28

Re: Turbulence Intensity Vs CFD Simulation
For fully developed pipe flows the turbulent intensity at the core can be estimated by I = 0.16 (Re)^-1/8

All times are GMT -4. The time now is 04:50.