[eric-meng]

Hi everyone, I just finished a lot of cases for my project, while I found there were two cases I forgot to uncomment the fieldAverage section in my controlDict so that it did not wrote down the average information in run-time for velocity and pressure fields.

my question is, is it possible using any post processing utilities to obtain the mean and prime2Mean values for these two fields?

I tried postProcess -func fieldAverage -time xxx but got prompt "Cannot find functionObject file fieldAverage", while fieldAverage is indeed a function object listed in the mannual

There seems be an utility called valueAverage for postProcessing while it can not return prime2Mean, I am not sure

Does any one have any solution?

BTW I hope this can be done only using the data in processors, and no need to reconstruct

[mAlletto]

This utility is averaging in the fly. So I doubt it is possible. Did you store the fields at the time steps.

[eric-meng]

Originally Posted by mAlletto

This utility is averaging in the fly. So I doubt it is possible. Did you store the fields at the time steps.

Hi Michael

Thank you for your reply

I actually stored the results in processors

and I can reconstruct the result calculate the average one by one

but this is a bit huge project for me, and may involve some programming job,

I have been searching for a couple of days while it seems no good solution

so I have to use the reconstruct way

[mAlletto]

You can try to make a small utility which loops over the time folders and calculated the statistics. But maybe the best would be to just restart the simulations in order that the results are comparable

[ssa_cfd]

If you have your time steps, you can reconstruct and then run temporal statistics filter in paraview which gives mean. But the results are not accurate as fieldAverage function object. It depends on how many time steps you have.

If its difficult to reconstruct, go to 'paraFoam -builtin' where u can see the data without reconstructing and then try temporal statistics filter.

[eric-meng]

Originally Posted by mAlletto

You can try to make a small utility which loops over the time folders and calculated the statistics. But maybe the best would be to just restart the simulations in order that the results are comparable

yes it can be but I am not very familiar with openfoam programming for the moment, may be try it in the future

[eric-meng]

Originally Posted by ssa_cfd

If you have your time steps, you can reconstruct and then run temporal statistics filter in paraview which gives mean. But the results are not accurate as fieldAverage function object. It depends on how many time steps you have.

If its difficult to reconstruct, go to 'paraFoam -builtin' where u can see the data without reconstructing and then try temporal statistics filter.

This is also another concern about it, my dt is 0.0025 thats as so small one

can the fluid field are also in transition to turbulence, so I think I'd better just run another enough time steps to keep my result in safe

Thank you for your reply and suggestion