|
flow
Kokkos cut-cell IBM incompressible Navier-Stokes solver + pnm pore extraction
|
Functions | |
| inlet_profile (H, S, nz, U_in) | |
| reattach (u_bottom) | |
| bfs (S, Re, dt, mode, Lr=12, U_in=1.0, nz=4, max_steps=6000, tol=2e-5, vcyc=4, vel_iter=60, outer=2) | |
| main () | |
Velocity-MG benchmark on the backward-facing step (high-resolution 2D; BFS is a channel, not a cube). Same two questions as the cavity bench: does the velocity-MG (a) speed up the march to steady at high resolution, and (b) stay free of any dt restriction (no immersed solid -> geometry-exact coarse operator)? Re_S = U_in*S/nu; the step is the inlet condition (set_domain_bc_profile), +x outflow, -y/+y no-slip walls.
| bench_velocity_mg_bfs.inlet_profile | ( | H, | |
| S, | |||
| nz, | |||
| U_in | |||
| ) |
Definition at line 19 of file bench_velocity_mg_bfs.py.
Referenced by bfs().
| bench_velocity_mg_bfs.reattach | ( | u_bottom | ) |
Definition at line 28 of file bench_velocity_mg_bfs.py.
Referenced by bfs().
| bench_velocity_mg_bfs.bfs | ( | S, | |
| Re, | |||
| dt, | |||
| mode, | |||
Lr = 12, |
|||
U_in = 1.0, |
|||
nz = 4, |
|||
max_steps = 6000, |
|||
tol = 2e-5, |
|||
vcyc = 4, |
|||
vel_iter = 60, |
|||
outer = 2 |
|||
| ) |
Definition at line 38 of file bench_velocity_mg_bfs.py.
References inlet_profile(), and reattach().
Referenced by main().
| bench_velocity_mg_bfs.main | ( | ) |
Definition at line 67 of file bench_velocity_mg_bfs.py.
Referenced by main().