pyopensn.solver.SCDSAAcceleration

class pyopensn.solver.SCDSAAcceleration

Construct an SCDSA accelerator for the power iteration k-eigenvalue solver.

Wrapper of opensn::SCDSAAcceleration.

Requires a single groupset and, with the AAH sweep, l_max_its greater than one. sdm='pwlc' is not supported in 1D spherical geometry. The owning solver checks these requirements in Initialize(). With l_max_its greater than one, each power iteration performs one sweep after the groupset solve to form the scalar-flux change, i.e. one sweep more than l_max_its.

__init__(self: pyopensn.solver.SCDSAAcceleration, **kwargs) → None

SCDSA acceleration for the power iteration k-eigenvalue solver.

Parameters:

  • problem (pyopensn.solver.DiscreteOrdinatesProblem) – Existing DiscreteOrdinatesProblem instance.

  • l_abs_tol (float, default=1.0e-10) – Absolute residual tolerance.

  • max_iters (int, default=100) – Maximum allowable iterations.

  • verbose (bool, default=False) – If true, enables verbose output.

  • petsc_options (str, default="ssss") – Additional PETSc options.

  • pi_max_its (int, default=50) – Maximum allowable iterations for inner power iterations.

  • pi_k_tol (float, default=1.0e-10) – k-eigenvalue tolerance for the inner power iterations.

  • sdm (str, default="pwld") –

    Spatial discretization method to use for the diffusion solver. Valid choices are:

    • ’pwld’ : Piecewise Linear Discontinuous

    • ’pwlc’ : Piecewise Linear Continuous