Hi developers and users,
Two questions about BC in the ex 8.
1. position of fill_dirichlet_bc() in Newmark and Newton nonlinear solver
I am using newton nonlinear solver and fill_dirichlet_bc() function in a
Newmark system. Base on the the ex8, fill_dirichlet_bc() is inserted in
the time loop . If there is a newton nonlinear in the time loop, should
I plug the fill_dirichlet_bc() in the newton nonlinear loop also? or
just put it in the time loop but outside of the nonlinear loop.
************************ the fill_dirichlet_bc() part
************************
t_system.update_rhs();
if (time_step == 0)
{
equation_systems.parameters.set<bool>("Newmark set BC for
Matrix") = true;
fill_dirichlet_bc(equation_systems, "Wave");
equation_systems.parameters.set<bool>("Newmark set BC for
Matrix") = false;
}
else
fill_dirichlet_bc(equation_systems, "Wave");
********************************************************************
2. the neumann BC at the end of assemble_wave() subroutine.
Beside the fill_dirichlet_bc(), anther boundary condition is added at
the end of assemble_wave() subroutine, which looks like a neumann BC,
please correct me if it is wrong. So does it mean that the neumann BC
should be set in the assemble() subroutine and the dirichlet BC should
be set in the main() part for the Newmark method? Or even the third BC
could be set as a combination of two BCs.
Thanks!
Liang
------------------------------------------------------------------------------
_______________________________________________
Libmesh-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/libmesh-users
