anuhar
April 7, 2011, 4:35pm
#1
Hi everybody

Does anyone know the possible reasons why a simulation doesn’t converge?
Here are the printed results of a porous medium simulation:

…
Iteration 30000
Iteration 31000
Iteration 32000
simulation diverged.
End of simulation at iteration 32298
Permeability:

Computing the permeability.
Average velocity = nan
Lattice viscosity nu = 0.0042735
Grad P = 2.61233e-05
Permeability = nan

thank you for your advice

Anuhar

Anton
September 7, 2015, 3:24pm
#2
I have the same problem, but it finishes at

End of simulation at iteration 499
Permeability:

Computing the permeability.
Average velocity = nan
Lattice viscosity nu = 927500
Grad P = 2.04082e+12
Permeability = nan

500 is number of iteration at which it make picture of pressure distribution.

Did you find solution of this problem? Any one have ideas?

Thanks.

Anton
September 24, 2015, 8:13am
#3
Hello all.

I find out, that process can be checked by computeAverageDensity(), so I understand that I have some really low omega and it make huge instability of the simulation. My density increased from 0.6 to 10^308 between 60 and 80 iteration. So next step it becomes NaN.

Hope it will helps somebody in the future.