hydromodpy.solver.boussinesq.runtimes.local#

In-process nonlinear runtime for the Boussinesq backend.

This backend is intentionally simple and transparent:

  • the residual is assembled in pure NumPy,

  • the Jacobian is approximated by dense finite differences,

  • the nonlinear solve uses a damped Newton loop.

It is therefore not the fastest option, but it is the easiest one to inspect when validating the physics or debugging convergence problems.

Functions

solve_backward_euler_step(mesh, *, ...[, ...])

Compatibility wrapper around solve_transient_step().

solve_steady_problem(inputs)

Solve one steady nonlinear balance from the normalized runtime contract.

solve_steady_state(mesh, *, head_initial_guess_m)

Compatibility wrapper around solve_steady_problem().

solve_transient_step(inputs)

Solve one implicit transient step from the normalized runtime contract.