Simulation API

For more control over the simulation, use the solver classes directly. These accept physical SI units and expose intermediate state for custom post-processing.

The high-level functions (tortuosity_lbm, permeability_lbm) use these classes internally.

TransientDiffusion

Solves the transient diffusion equation using LBM (D3Q7 BGK).

from poromics.simulation import TransientDiffusion

solver = TransientDiffusion(im, axis=1, D=1e-9, voxel_size=1e-6)
solver.run(n_steps=100_000, tol=1e-2)

c = solver.concentration    # Steady-state concentration field
J = solver.flux(axis=1)     # Mean flux at midplane
print(f"Converged: {solver.converged}, dt: {solver.dt:.2e} s")

Property / Method	Description
concentration	Concentration field, shape (nx, ny, nz)
flux(axis)	Mean diffusive flux at the domain midplane
converged	Whether convergence was reached
n_iterations	Total number of time steps taken
dt	Physical time step in seconds
voxel_size	Voxel edge length in metres
step()	Advance by one time step
run(n_steps, tol)	Run to steady state

TransientFlow

Solves incompressible Stokes flow using LBM (D3Q19 MRT).

from poromics.simulation import TransientFlow

solver = TransientFlow(im, axis=1, nu=1e-6, voxel_size=1e-6)
solver.run(n_steps=100_000, tol=1e-3)

v = solver.velocity   # Velocity field in m/s, shape (nx, ny, nz, 3)
P = solver.pressure   # Pressure field in Pa (gauge)
print(f"Converged: {solver.converged}, dt: {solver.dt:.2e} s")

Property / Method	Description
velocity	Velocity field in m/s, shape (nx, ny, nz, 3)
pressure	Gauge pressure field in Pa, shape (nx, ny, nz)
converged	Whether convergence was reached
n_iterations	Total number of time steps taken
dt	Physical time step in seconds
voxel_size	Voxel edge length in metres
step()	Advance by one time step
run(n_steps, tol)	Run to steady state