CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   Time-averaged result (https://www.cfd-online.com/Forums/fluent/97539-time-averaged-result.html)

lgas222 February 20, 2012 02:53

Time-averaged result in transient FLUENT simulation
 
Hi,

I'm working on FLUENT with transient mode. And I hope to get the time-averaged contours of the results.

I have tried with paraview. It seems like paraview can only read data with a time step; however, a set of time series transient data can not be imported to the paraview.

Does anyone know how to post process it? Thanks a lot.

doronzo February 20, 2012 08:20

Re
 
hi

not totally sure, but if you are using for example RANS, then the results are already time-averaged, i.e. they are averaged on your time step, whatever time level you are looking at.

regards

lgas222 February 20, 2012 08:38

Quote:

Originally Posted by doronzo (Post 345318)
hi

not totally sure, but if you are using for example RANS, then the results are already time-averaged, i.e. they are averaged on your time step, whatever time level you are looking at.

regards

:)
Thanks.

I use k-epsilon viscous model. Do you know how can I get the time-averaged contour between, eg. 1-100s, with paraview or tecplot? As I knew from my colleague, we can use matlab to deal with it. But I haven't tried that.

doronzo February 20, 2012 11:39

Re
 
hi

what i know is that if you are simulating the RANS, so not LES/DNS, then your solution is already integrated over the time step you are using.

aside from the numerical aspects your time step can lead, it should be longer than the typical duration of fluctuating components, but it should also allow for the possibility of variations in mean quantities to describe unsteady conditions in the mean fluid motion, thus the time step cannot be too long.

this means it depends on the problem you are solving, but however i don't know how to average in ''other'' times outside Fluent; i know it is possible with Tecplot in a relatively easy way, but honestly i don't know how

regards


All times are GMT -4. The time now is 09:34.