Provides a convergence criterion for the linear solvers which looks at the weighted maximum of the difference between two iterations. More...

#include <fixpointcriterion.hh>

Inheritance diagram for Opm::Linear::FixPointCriterion< Vector, CollectiveCommunication >:

Public Member Functions
	FixPointCriterion (const CollectiveCommunication &comm)

	FixPointCriterion (const CollectiveCommunication &comm, const Vector &weightVec, Scalar reduction)

void	setWeight (const Vector &weightVec)
	Sets the relative weight of a primary variable.

Scalar	weight (int outerIdx, int innerIdx) const
	Return the relative weight of a primary variable.

void	setTolerance (Scalar tol)
	Set the maximum allowed weighted maximum difference between two iterations.

Scalar	tolerance () const
	Return the maximum allowed weighted difference between two iterations for the solution considered to be converged.

void	setInitial (const Vector &curSol, const Vector &) override
	Set the initial solution of the linear system of equations.

void	update (const Vector &curSol, const Vector &, const Vector &) override
	Update the internal members of the convergence criterion with the current solution.

bool	converged () const override
	Returns true if and only if the convergence criterion is met.

Scalar	accuracy () const
	Returns the accuracy of the solution at the last update.

Public Member Functions inherited from Opm::Linear::ConvergenceCriterion< Vector >
virtual	~ConvergenceCriterion ()
	Destructor.

virtual bool	failed () const
	Returns true if the convergence criterion cannot be met anymore because the solver has broken down.

virtual void	printInitial (std::ostream &=std::cout) const
	Prints the initial information about the convergence behaviour.

virtual void	print (Scalar, std::ostream &=std::cout) const
	Prints the information about the convergence behaviour for the current iteration.

Detailed Description

template<class Vector, class CollectiveCommunication>
class Opm::Linear::FixPointCriterion< Vector, CollectiveCommunication >

Provides a convergence criterion for the linear solvers which looks at the weighted maximum of the difference between two iterations.

For the FixPointCriterion, the error of the solution is defined as

$e^k = \max_i\{ \left| w_i \delta^k_i \right| \}\;,$

where $\delta = x^k - x^{k + 1}$ is the difference between two consequtive iterative solution vectors $x^k$ and $x^{k + 1}$ and $w_i$ is the weight of the $i$ -th degree of freedom.

This criterion requires that the block type of the vector is a Dune::FieldVector

Member Function Documentation

◆ accuracy()

template<class Vector , class CollectiveCommunication >

Scalar Opm::Linear::FixPointCriterion< Vector, CollectiveCommunication >::accuracy ( ) const

inlinevirtual

Returns the accuracy of the solution at the last update.

A value of zero means that the solution was exact.

Implements Opm::Linear::ConvergenceCriterion< Vector >.

◆ converged()

template<class Vector , class CollectiveCommunication >

bool Opm::Linear::FixPointCriterion< Vector, CollectiveCommunication >::converged ( ) const

inlineoverridevirtual

Returns true if and only if the convergence criterion is met.

Implements Opm::Linear::ConvergenceCriterion< Vector >.

◆ setInitial()

template<class Vector , class CollectiveCommunication >

void Opm::Linear::FixPointCriterion< Vector, CollectiveCommunication >::setInitial	(	const Vector &	curSol,
		const Vector &
	)

inlineoverridevirtual

Set the initial solution of the linear system of equations.

This version of the method does NOT take the two-norm of the residual as argument. If the two-norm of the defect is available for the linear solver, the version of the update() method with it should be called.

Parameters

curSol	The current iterative solution of the linear system of equations
curResid	The residual vector of the current iterative solution of the linear system of equations

Implements Opm::Linear::ConvergenceCriterion< Vector >.

◆ setTolerance()

template<class Vector , class CollectiveCommunication >

void Opm::Linear::FixPointCriterion< Vector, CollectiveCommunication >::setTolerance ( Scalar tol )

inline

Set the maximum allowed weighted maximum difference between two iterations.

Set the maximum allowed maximum difference between two iterationsfor the solution considered to be converged.

◆ setWeight()

template<class Vector , class CollectiveCommunication >

void Opm::Linear::FixPointCriterion< Vector, CollectiveCommunication >::setWeight ( const Vector & weightVec )

inline

Sets the relative weight of a primary variable.

For the FixPointCriterion, the error of the solution is defined as

$e^k = \max_i\{ \left| w_i \delta^k_i \right| \}\;,$

where $\delta = x^k - x^{k + 1}$ is the difference between two consequtive iterative solution vectors $x^k$ and $x^{k + 1}$ and $w_i$ is the weight of the $i$ -th degree of freedom.

This method is specific to the FixPointCriterion.

Parameters

weightVec A Dune::BlockVector<Dune::FieldVector<Scalar, n> > with the relative weights of the degrees of freedom

◆ update()

template<class Vector , class CollectiveCommunication >

void Opm::Linear::FixPointCriterion< Vector, CollectiveCommunication >::update	(	const Vector &	curSol,
		const Vector &	,
		const Vector &
	)

inlineoverridevirtual

Update the internal members of the convergence criterion with the current solution.

This version of the method does NOT take the two-norm of the residual as argument. If the two-norm of the defect is available for the linear solver, the version of the update() method with it should be called.

Parameters

curSol	The current iterative solution of the linear system of equations
changeIndicator	A vector where all non-zero values indicate that the solution has changed since the last iteration.
curResid	The residual vector of the current iterative solution of the linear system of equations

Implements Opm::Linear::ConvergenceCriterion< Vector >.

◆ weight()

template<class Vector , class CollectiveCommunication >

Scalar Opm::Linear::FixPointCriterion< Vector, CollectiveCommunication >::weight	(	int	outerIdx,
		int	innerIdx
	)		const

inline

Return the relative weight of a primary variable.

For the FixPointCriterion, the error of the solution is defined as

$e^k = \max_i\{ \left| w_i \delta^k_i \right| \}\;,$

where $\delta = x^k - x^{k + 1}$ is the difference between two consequtive iterative solution vectors $x^k$ and $x^{k + 1}$ and $w_i$ is the weight of the $i$ -th degree of freedom.

This method is specific to the FixPointCriterion.

Parameters

outerIdx	The index of the outer vector (i.e. Dune::BlockVector)
innerIdx	The index of the inner vector (i.e. Dune::FieldVector)

The documentation for this class was generated from the following file:

opm/simulators/linalg/fixpointcriterion.hh

Public Member Functions

Detailed Description

Member Function Documentation

◆ accuracy()

◆ converged()

◆ setInitial()

◆ setTolerance()

◆ setWeight()

◆ update()

◆ weight()