[mtgoncalves]

I want to modify boundary conditions during a simulation from inside a custom solver (derived from pimpleFoam) without creating my own custom boundary condition type.

So far I have been successful for fixedValue boundary conditions. I managed this by using, for example

type fixedValue;
value uniform (0 0 0);

in the 0/U file. Then, during the simulation I modify the boundary values from inside my custom solver using

int index = 0;

forAll( U.boundaryFieldRef()[boundary_label(patchName)], i )
{
    for ( int j = 0; j < 3; j++ )
    {
        U.boundaryFieldRef()[boundary_label(patchName)][i][j] = velocityBC[index];
        index++;
    }
}

Thus, for a fixedValue b.c. I managed to modify the specification of "value". How can I do the same for the specification of "gradient" for a fixedGradient b.c.?

[RGS]

Originally Posted by mtgoncalves

I want to modify boundary conditions during a simulation from inside a custom solver (derived from pimpleFoam) without creating my own custom boundary condition type.

So far I have been successful for fixedValue boundary conditions. I managed this by using, for example

type fixedValue;
value uniform (0 0 0);

in the 0/U file. Then, during the simulation I modify the boundary values from inside my custom solver using

int index = 0;

forAll( U.boundaryFieldRef()[boundary_label(patchName)], i )
{
    for ( int j = 0; j < 3; j++ )
    {
        U.boundaryFieldRef()[boundary_label(patchName)][i][j] = velocityBC[index];
        index++;
    }
}

Thus, for a fixedValue b.c. I managed to modify the specification of "value". How can I do the same for the specification of "gradient" for a fixedGradient b.c.?

Hi,

Did you manage to figure out how to do this? I am trying to do something similar.


Thanks!