[lr103476]

Hi,

I am just reading an article about code verification and validation. To what extent is OpenFOAM verified before it was released?

Which cases are used to check the accuracy of the schemes employed in OpenFOAM. I mean, are there any grid convergence studies performed such that the mathematical order of the schemes is checked.

I performed simulations of the flow around a cylinder at Re=150. Validation with experiments is Ok, but I did not succeed in getting a grid converged solution. Although the flow looks pretty much ok, the mean drag is fluctuating on grids with increasing sizes.

Thats why I am interested which cases the developers used for code verification before OpenFOAM was released.

Regards, Frank

[hjasak]

The question of verification has been raised a number of times before and I am happy to say that OpenFOAM is probably one of the most validated codes to date. Vadidation cases are distributed through a number of PhD Theses where the work has been done with the code. This ranges from simple numerics validation to complete model runs and comparison with experimental data across the applicationspectrum. Some of the work may be out of date because the models or impelemntation has improved, which leaves more space for the current users :-)

For example, numerics validation rates of error reduction, adaptive mesh refinement etc (see Jasak and Juretic PhDs). For LES, we've got Eugene de Villiers on line (waiting for that Thesis!), two-phase flows by Rusche and Brennan, moving mesh FVM and free surface work by Tukovic, for stress analysis and fluid-structure interaction see the work in prof. Ivankovic's group at UC Dublin, for Diesel combustion see the papers by Niklas Nordin and the Chalmers group, recent paper by G. Tabor and H Weller on combustion etc.

Some of this may benefit from re-organisation, but the info is still there - if you wish to have a go at organising it, most of the stuff is already available on-line.

Enjoy,

Hrv