StokesFlowSolver< DIM > Class Template Reference

#include <StokesFlowSolver.hpp>

Inherits AbstractContinuumMechanicsSolver< DIM >.

Collaboration diagram for StokesFlowSolver< DIM >:

[legend]

List of all members.

Public Member Functions
	StokesFlowSolver (AbstractTetrahedralMesh< DIM, DIM > &rQuadMesh, StokesFlowProblemDefinition< DIM > &rProblemDefinition, std::string outputDirectory)
virtual	~StokesFlowSolver ()
void	Solve ()
void	SetKspAbsoluteTolerance (double kspAbsoluteTolerance)
std::vector< c_vector< double, DIM > > &	rGetSpatialSolution ()
std::vector< c_vector< double, DIM > > &	rGetVelocities ()
Private Member Functions
void	AssembleSystem ()
Private Attributes
StokesFlowProblemDefinition < DIM > &	mrProblemDefinition
StokesFlowAssembler< DIM > *	mpStokesFlowAssembler
StokesFlowPreconditionerAssembler < DIM > *	mpStokesFlowPreconditionerAssembler
ContinuumMechanicsNeumannBcsAssembler < DIM > *	mpNeumannBcsAssembler
double	mKspAbsoluteTol
Friends
class	TestStokesFlowSolver

---

Detailed Description

template<unsigned DIM>
class StokesFlowSolver< DIM >

Finite element solver for Stokes flow problems

Definition at line 55 of file StokesFlowSolver.hpp.

---

Constructor & Destructor Documentation

template<unsigned DIM>

StokesFlowSolver< DIM >::StokesFlowSolver	(	AbstractTetrahedralMesh< DIM, DIM > &	rQuadMesh,
		StokesFlowProblemDefinition< DIM > &	rProblemDefinition,
		std::string	outputDirectory
	)			[inline]

Constructor.

Parameters:

	rQuadMesh	Quadratic mesh
	rProblemDefinition	Problem definition
	outputDirectory	the output directory to use

Definition at line 141 of file StokesFlowSolver.hpp.

References StokesFlowSolver< DIM >::mpNeumannBcsAssembler, StokesFlowSolver< DIM >::mpStokesFlowAssembler, StokesFlowSolver< DIM >::mpStokesFlowPreconditionerAssembler, StokesFlowSolver< DIM >::mrProblemDefinition, and AbstractContinuumMechanicsSolver< DIM >::mrQuadMesh.

template<unsigned DIM>

StokesFlowSolver< DIM >::~StokesFlowSolver

(

)

[inline, virtual]

Destructor.

Definition at line 157 of file StokesFlowSolver.hpp.

References StokesFlowSolver< DIM >::mpNeumannBcsAssembler, StokesFlowSolver< DIM >::mpStokesFlowAssembler, and StokesFlowSolver< DIM >::mpStokesFlowPreconditionerAssembler.

---

Member Function Documentation

template<unsigned DIM>

void StokesFlowSolver< DIM >::AssembleSystem

(

)

[inline, private]

Assemble the linear system and preconditioner matrix.

Definition at line 287 of file StokesFlowSolver.hpp.

References AbstractContinuumMechanicsSolver< DIM >::AddIdentityBlockForDummyPressureVariables(), AbstractContinuumMechanicsSolver< DIM >::ApplyDirichletBoundaryConditions(), PetscMatTools::Finalise(), PetscVecTools::Finalise(), AbstractContinuumMechanicsSolver< DIM >::mLinearSystemRhsVector, StokesFlowSolver< DIM >::mpNeumannBcsAssembler, AbstractContinuumMechanicsSolver< DIM >::mPreconditionMatrix, StokesFlowSolver< DIM >::mpStokesFlowAssembler, StokesFlowSolver< DIM >::mpStokesFlowPreconditionerAssembler, AbstractContinuumMechanicsSolver< DIM >::mSystemLhsMatrix, and PetscMatTools::SwitchWriteMode().

Referenced by StokesFlowSolver< DIM >::Solve().

template<unsigned DIM>

std::vector< c_vector< double, DIM > > & StokesFlowSolver< DIM >::rGetSpatialSolution

(

)

[inline, virtual]

Returns:

the flow. Note: return_value[i](j) = u_j for node i.

Implements AbstractContinuumMechanicsSolver< DIM >.

Definition at line 327 of file StokesFlowSolver.hpp.

References AbstractMesh< ELEMENT_DIM, SPACE_DIM >::GetNumNodes(), AbstractContinuumMechanicsSolver< DIM >::mCurrentSolution, AbstractContinuumMechanicsSolver< DIM >::mrQuadMesh, and AbstractContinuumMechanicsSolver< DIM >::mSpatialSolution.

Referenced by StokesFlowSolver< DIM >::rGetVelocities().

template<unsigned DIM>

std::vector< c_vector< double, DIM > > & StokesFlowSolver< DIM >::rGetVelocities

(

)

[inline]

Returns:

the flow. Note: return_value[i](j) = u_j for node i. Just calls rGetSpatialSolution().

Definition at line 342 of file StokesFlowSolver.hpp.

References StokesFlowSolver< DIM >::rGetSpatialSolution().

template<unsigned DIM>

void StokesFlowSolver< DIM >::SetKspAbsoluteTolerance

(

double

kspAbsoluteTolerance

)

[inline]

Set the absolute tolerance to be used when solving the linear system. If this is not called a relative tolerance is used.

Parameters:

kspAbsoluteTolerance

the tolerance

Definition at line 320 of file StokesFlowSolver.hpp.

References StokesFlowSolver< DIM >::mKspAbsoluteTol.

template<unsigned DIM>

void StokesFlowSolver< DIM >::Solve

(

)

[inline]

Solve the system.

Definition at line 165 of file StokesFlowSolver.hpp.

References StokesFlowSolver< DIM >::AssembleSystem(), GenericEventHandler< 7, MechanicsEventHandler >::BeginEvent(), PetscTools::Destroy(), GenericEventHandler< 7, MechanicsEventHandler >::EndEvent(), PetscTools::GetMyRank(), AbstractContinuumMechanicsSolver< DIM >::mCurrentSolution, StokesFlowSolver< DIM >::mKspAbsoluteTol, AbstractContinuumMechanicsSolver< DIM >::mLinearSystemRhsVector, AbstractContinuumMechanicsSolver< DIM >::mNumDofs, AbstractContinuumMechanicsSolver< DIM >::mPreconditionMatrix, StokesFlowSolver< DIM >::mrProblemDefinition, AbstractContinuumMechanicsSolver< DIM >::mSystemLhsMatrix, AbstractContinuumMechanicsSolver< DIM >::mVerbose, PETSC_DESTROY_PARAM, Timer::PrintAndReset(), AbstractContinuumMechanicsSolver< DIM >::RemovePressureDummyValuesThroughLinearInterpolation(), Timer::Reset(), AbstractContinuumMechanicsSolver< DIM >::WriteCurrentPressureSolution(), AbstractContinuumMechanicsSolver< DIM >::WriteCurrentSpatialSolution(), and PetscVecTools::Zero().

---

Member Data Documentation

template<unsigned DIM>

double StokesFlowSolver< DIM >::mKspAbsoluteTol [private]

Absolute tolerance for linear systems. Can be set by calling SetKspAbsoluteTolerances(), but default to -1, in which case a relative tolerance is used.

Definition at line 77 of file StokesFlowSolver.hpp.

Referenced by StokesFlowSolver< DIM >::SetKspAbsoluteTolerance(), and StokesFlowSolver< DIM >::Solve().

template<unsigned DIM>

ContinuumMechanicsNeumannBcsAssembler<DIM>* StokesFlowSolver< DIM >::mpNeumannBcsAssembler [private]

Assembler for adding the surface integral arising from natural Neumman boundary conditions to the RHS vector

Definition at line 70 of file StokesFlowSolver.hpp.

Referenced by StokesFlowSolver< DIM >::AssembleSystem(), StokesFlowSolver< DIM >::StokesFlowSolver(), and StokesFlowSolver< DIM >::~StokesFlowSolver().

template<unsigned DIM>

StokesFlowAssembler<DIM>* StokesFlowSolver< DIM >::mpStokesFlowAssembler [private]

Assembler for computing volume integral part of matrix and RHS vector

Definition at line 58 of file StokesFlowSolver.hpp.

Referenced by StokesFlowSolver< DIM >::AssembleSystem(), StokesFlowSolver< DIM >::StokesFlowSolver(), and StokesFlowSolver< DIM >::~StokesFlowSolver().

template<unsigned DIM>

StokesFlowPreconditionerAssembler<DIM>* StokesFlowSolver< DIM >::mpStokesFlowPreconditionerAssembler [private]

Assembler for computing volume integral part of preconditioner matrix (which is the same as the system matrix except has a mass matrix in the pressure-pressure block

Definition at line 64 of file StokesFlowSolver.hpp.

Referenced by StokesFlowSolver< DIM >::AssembleSystem(), StokesFlowSolver< DIM >::StokesFlowSolver(), and StokesFlowSolver< DIM >::~StokesFlowSolver().

template<unsigned DIM>

StokesFlowProblemDefinition<DIM>& StokesFlowSolver< DIM >::mrProblemDefinition [private]

Object containing all the information about the problem to solve

Reimplemented from AbstractContinuumMechanicsSolver< DIM >.

Definition at line 55 of file StokesFlowSolver.hpp.

Referenced by StokesFlowSolver< DIM >::Solve(), and StokesFlowSolver< DIM >::StokesFlowSolver().

---

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

• continuum_mechanics/src/solver/StokesFlowSolver.hpp