[ssa_cfd]

Hi all,

I have a velocity time series data (U vs time) from an wind tunnel experiment for a number of points along the height of the wind tunnel.

I need to calculate the integral length scale and reynolds stress tensor from this data for my simulation.

How to obtain these values.? Is there any matlab script available. ?


Thanks.
ssa.

[LuckyTran]

U is the full 3D vector or just one component? You can't calculate the Reynolds stress with only one velocity component. Reynolds stresses are straight forward to calculate. You determine u'(t)=u(t)-umean, v'=v(t)-vmean, w'=w(t)-wmean and then take all their products. Hopefully you don't need anyone else's code to do this...

For estimating length scale, there's several methods mentioned in doi: 10.1115/1.1391277 needing only the u' at a single point.

[ssa_cfd]

I have three velocity components.

So If I take a time series, say may be 100 values over a period of 5 seconds, I should calculate the u mean from those values and then get u'(t).

so I will have 100 u'(t) values and then I should average those values to get u'.

Am I correct.?

[LuckyTran]

Originally Posted by ssa_cfd

I have three velocity components.

So If I take a time series, say may be 100 values over a period of 5 seconds, I should calculate the u mean from those values and then get u'(t).

so I will have 100 u'(t) values and then I should average those values to get u'.

Am I correct.?

You will need a lot more than 100 values! But okay, supposing you have a lot of them...


If you average 100 u'(t) values you will get 0.


The Reynolds stresses are calculated by taking pairwise products u'(t)*u'(t) and then averaging the 100 product-pairs. There are 6 stresses from u'u', u'v', u'w', v'v', v'w', w'w'. This is all done at 1 point in the wind tunnel. You can repeat this across all the points in the wind tunnel if you like. The Reynolds stresses are a tensor field over the domain, just like the velocity is a vector field over the entire domain.


Similarly, the integral length scale is not one number for the entire wind tunnel but a local flow property.