Turbulent kinetic energy in k-epsilon model
 

Hello,
I am a newbie here and I got a trouble when calculating turbulent kinetic energy in standard k-epsilon model which I hope anyone out there please help me :)

My problem is like this: I picked 1 point in my cubit domain, run Fluent and exported turbulent kinetic energy at (for example) time step 100, that was fine

But I have to do the manual calculation, then I used the equation:
turbulent kinetic energy: k=1.5*(u_avg*I)^2
turbulence intensity: I=u'/u_avg
where: u_avg, u' are mean flow velocity and velocity fluctuation

I did study the theory and user guides but I still don't know how to calculate u_avg and u', spatial average or time average? then I did 3 cases:
- u_avg = u of that point which I exported from Fluent after running
- u_avg = spatial average velocity (mean velocity of total cubic domain at time step 100)
- u_avg = time average velocity (mean velocity of that single point over 100 time steps)

actually, I used these values to multiply with turbulence intensity I that i export from Fluent to get k (as equation above) because I am not sure how to calculate u', but unfortunately they were all wrong (compare to the value of k I exported)

I know that maybe this is a stupid question but I really confused, so can you guys plz explain for me? It would be very appreciated.

Thank you in advance.

Regards,
Thao Nguyen