[zkzkzk010537]

Dear ALL,

I have a question regarding the time averaged mixture fraction for non-premixed combustion.

Before i start running my transient case, the mixture fraction option has been chosen under sampling options. However, after running for certain physical time, i have noticed there is no time averaged mixture fraction but only instantaneous mean mixture fraction available, under the option of unsteady statistics (contour plot).

Has anyone ever met the same problem, i remember i have seen such questions in the forum before, but i can't find it.

Regards,
Kai.

[CeesH]

Hi Kai,

You can go for 2 approaches; under `run calculation' there is the option 'sample time statistics' that allows to set time averaged variables. I never use it, but I guess it would be available for mixture fractions, too.

Another option is to export your data every N timesteps via calculation activities, and post-process in your favorite program. For example it's quite straightforward to average a large number of files in MATLAB.

Does that answer your question?
Cees

[zkzkzk010537]

Hi cees,

Thanks for your help, but that's exactly my problem. Under sample statistics, although i have chosen the averaged mixture fraction, it is not really saved after calculation. The export of data does not have the time averaged mixture fraction either. (all other variables are time averaged properly, except for mixture fraction.)

Kai.

[CeesH]

That is odd indeed; can you make monitor points or so for it or does it not show up under that tab either?

I guess the post-processing option would still be a possibility, but of course, a more tedious and time-taking one.

[zkzkzk010537]

The option is not available anywhere, not even in post-processing....weird one.

[CeesH]

Is it possible for you to use MATLAB or so for post-processing? Then it is certainly possible.

[Andrea1984]

According to my experience when collecting statistics, for some reason the time-averaged quantities for some variables are not available under Unsteady Statistics until you save your .cas and .dat file. Try to save your results, close Fluent and open you results again and the mean-f variable should be there.

[zkzkzk010537]

Well, i did have saved my cas and data file as i run all my cases on cluster, and save the file every few time steps. None of those saved data include the mean variable i want.

But anyway, thanks for your help.
Kai.

[zkzkzk010537]

I think the mixture fraction data is not even saved for post-processing..so no point to use a different software to post-process.

Kai.

[Andrea1984]

Hi Kai,

which Fluent version are you using? I am with Fluent 15.0 and I am having a strange issue when collecting statistics for the mixture fraction as well.
Although f is supposed to be bounded between 0 and 1 (and the instantaneous values are indeed in the correct range), the mean mixture fraction goes well above 1, so I am guessing perhaps there is a bug in the averaging of this particular variable? I am trying to define a custom field function equal to f and see if the statistics for this function will look sensible, I'll keep you posted

Andrea

[zkzkzk010537]

I am using 14.5, and i have never seen mean mixture fraction go above 1, though i can't find the averaged mixture fraction.

[Andrea1984]

Hi Kai,

It looks like I have solved my issues collecting statistic for the mixture fraction by defining a cff equal to the instantaneous mixture fraction value, and sampling this cff instead of the variable solved for in Fluent.
I am attaching a few screenshots to show you what I meant (I have been collecting statistics only for a short time, this is why the time-averaged values look a bit wiggly).

You can see that the instantaneous value is bounded between 0. As I mentioned in my previous post, the time-averaged value for Fluents variable is bounded between 0 and 9 (for some reason I am not able to explain) and the cff time-averaged value is bounded (correctly) between 0 and 1.
Hope this can help to solve your issue as well
Andrea