CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   Turbulent Fluctuations (http://www.cfd-online.com/Forums/main/2855-turbulent-fluctuations.html)

Thomi December 2, 2000 11:45

Turbulent Fluctuations
 
Hi there,

Have you got any suggestions for calculating the turbulent fluctuations/intensities for a time series? Is it the standard deviation of the mean? I appreciate any comments and suggestions. Many thanx,

Thomi

Hallo December 6, 2000 15:52

Re: Turbulent Fluctuations
 
If your time series is steady then the fluctuating velocity (u') is just the instantaneous velocity (u) minus the mean velocity (U). Where the mean is calculated from a given time interval. the turbulent intensity (TI) is usually defined as some estimation of the small scale (or turbulent) velocity scales, non-dimensionalized with the large scale velocities. You can pick different velocity scales depending on the problem you have. Usually, the turbulent velocity scale is the root mean square (urms) where urms = sqrt(1/N * SUM(u'u')) where N is the number of discrete u' measurements and the large scale velocities is typically taken to be U.

Note this can be found in any book on Turbulence... (Lumley, Hinze, ...)

Thomi December 6, 2000 17:39

Re: Turbulent Fluctuations
 
Hi Hallo,

Thanks a lot. that helps for a start. so there is no standard deviation involved?

Hallo December 7, 2000 10:34

Re: Turbulent Fluctuations
 
If you use the standard definitions ie. u=U+u' and the u'rms defined in the previous thread then the turbulent intensity is just the average variations from the mean velocity (or the standard deviation) normalized with the mean velocity.

(ie. u'rms is the standard deviation)


All times are GMT -4. The time now is 13:54.