tardis.simulation.convergence module

class tardis.simulation.convergence.ConvergenceSolver(strategy)

Bases: object

Convergence solver. Sets convergence strategy and assigns a method to the converge property.

Parameters:

strategystring

Convergence strategy for the physical property

Raises:

NotImplementedError

Custom convergence type specified

ValueError

Unknown convergence type specified

adaptive_damped(value, estimated_value)

Adaptive damped convergence solver

Dynamically updates the damping factor by locally searching around the current value and selecting the step that minimizes the residual

Parameters:

valuenp.float64

The current value of the physical property

estimated_valuenp.float64

The estimated value of the physical property

Returns:

np.float64

Updated value after applying adaptive damping

Notes

The damping factor is updated in place and constrained in the interval [lambda_min, lambda_max]

damped_converge(value, estimated_value)

Damped convergence solver

Parameters:

valuenp.float64

The current value of the physical property

estimated_valuenp.float64

The estimated value of the physical property

Returns:

np.float64

The converged value

get_convergence_status(value, estimated_value, no_of_cells)

Get the status of convergence for the physical property

Parameters:

valuenp.float64, Quantity

The current value of the physical property

estimated_valuenp.float64, Quantity

The estimated value of the physical property

no_of_cellsnp.int64

The number of cells to measure convergence over

Returns:

bool

True if convergence is reached