Detailed Description

template<class T = real_t, int MatOrder = RowMajor>
class gismo::gsINSSolverUnsteady< T, MatOrder >

The unsteady incompressible Navier-Stokes solver.

Template Parameters

T	coefficient type
MatOrder	sparse matrix storage order (ColMajor/RowMajor)

Inheritance diagram for gsINSSolverUnsteady< T, MatOrder >:

Collaboration diagram for gsINSSolverUnsteady< T, MatOrder >:

Public Member Functions
virtual void	applySolver (gsMatrix< T > &solution)
	Solve the linear system.

int	checkGeoJacobian (int npts=-1, T dist=-1, T tol=-1)
	Check values of jacobian near boundaries of all patches.

gsField< T >	constructSolution (int unk) const
	Construct solution field for the unknown unk for the current solution vector.

gsINSAssemblerUnsteady< T, MatOrder > *	getAssembler () const
	Returns a pointer to the assembler.

virtual const T	getAssemblyTime () const
	Returns the total time spent on matrix assembly.

T	getAvgPicardIterations () const
	Returns the average number of Picard iterations per time step.

virtual const T	getInitAssemblyTime () const
	Returns the time of the initial matrix assembly.

unsigned	getIterationNumber () const
	Returns the current iteration number.

gsFlowLinSystSolver< T, MatOrder > *	getLinSolver () const
	Returns a pointer to the linear system solver.

virtual std::string	getName ()
	Retrurns the name of the class as a string.

virtual gsFlowSolverParams< T >	getParams ()
	Retrurns the solver parameters.

const gsMatrix< T > &	getSolution () const
	Returns the current solution vector.

virtual const T	getSolverSetupTime () const
	Returns the total time spent on linear solver setup.

virtual const T	getSolveTime () const
	Returns the total time spent on solving of the linear systems.

	gsINSSolverUnsteady (gsFlowSolverParams< T > &params)
	Constructor.

virtual void	initialize ()
	Initialize the solver.

virtual void	initIteration ()
	Prepare for the solution process.

virtual void	initIteration (const gsSparseMatrix< T, MatOrder > &mat)
	Prepare for the solution process.

virtual void	markDofsAsEliminatedZeros (const std::vector< gsMatrix< index_t > > &boundaryDofs, const int unk)
	Eliminate given DOFs as homogeneous Dirichlet boundary.

virtual void	nextIteration ()
	Perform next iteration step.

virtual void	nextIteration (const unsigned numberOfIterations)
	Perform several iteration steps.

int	numDofs () const
	Returns the total number of DOFs (the matrix size).

virtual T	residualRelNorm () const
	Compute and return the relative residual norm for the current solution.

virtual T	residualRelNorm (const gsMatrix< T > &solution) const
	Compute and return the relative residual norm for the given solution.

virtual void	setSolution (const gsMatrix< T > &solVector)
	Set a given solution vector as current solution.

T	solutionChangeRelNorm () const
	Compute and return the relative norm of the solution change.

T	solutionChangeRelNorm (gsMatrix< T > solOld, gsMatrix< T > solNew) const
	Compute and return the relative norm of the solution change given the two successive solutions.

void	solve (const int maxIterations, const T epsilon=1e-3, const int minIterations=1)
	Solve the incompressible Navier-Stokes problem.

virtual void	solveGeneralizedStokes (const int maxIterations, const T epsilon, const int minIterations=1)
	Solve the generalized Stokes problem.

virtual void	solveStokes ()
	Compute the Stokes problem and save the solution into m_solution.

virtual void	updateAssembler (bool updateSol=true)
	Update the assembler with current solution.

virtual void	updateAssembler (const gsMatrix< T > &sol, bool updateSol=true)
	Update the assembler with a given solution.

virtual void	writeSolChangeRelNorm (gsMatrix< T > solOld, gsMatrix< T > solNew)
	Compute and display the relative norm of the solution change given the two successive solutions.

Protected Member Functions
virtual void	createOutputFile ()
	Create output file.

virtual void	initMembers ()
	Initialize all members.

real_t	stopwatchStart ()
	Start measuring time (decides whether to use gsStopwatch or MPI_Wtime)

real_t	stopwatchStop ()
	Stop measuring time (decides whether to use gsStopwatch or MPI_Wtime)

virtual void	updateSizes ()
	Update sizes of members (when DOF numbers change, e.g. after markDofsAsEliminatedZeros()).

Member Function Documentation

◆ applySolver()

template<class T , int MatOrder>

virtual void applySolver ( gsMatrix< T > & solution )

inlinevirtualinherited

Solve the linear system.

Parameters

[out] solution a reference to the vector, where the computed solution will be stored

◆ checkGeoJacobian()

template<class T , int MatOrder>

int checkGeoJacobian	(	int	npts = `-1`,
		T	dist = `-1`,
		T	tol = `-1`
	)

inherited

Check values of jacobian near boundaries of all patches.

Parameters

npts[in]	number of evaluation points along a patch side in each direction
dist[in]	distance from patch boundaries (in the parametric space)
tol[in]	critical (positive) value of jacobian to throw warning (too close to zero)

Returns: error code (0 = all jacobians OK, 1 = jacobian close to zero in some points, -1 = jacobian negative in some points)

◆ constructSolution()

template<class T , int MatOrder>

gsField< T > constructSolution ( int unk ) const

inlineinherited

Construct solution field for the unknown unk for the current solution vector.

Parameters

unk	the considered unknown (0 - velocity, 1 - pressure)

Returns

◆ initIteration()

template<class T , int MatOrder>

virtual void initIteration ( const gsSparseMatrix< T, MatOrder > & mat )

inlinevirtualinherited

Prepare for the solution process.

Parameters

[in] the linear system matrix

◆ markDofsAsEliminatedZeros()

template<class T , int MatOrder>

void markDofsAsEliminatedZeros	(	const std::vector< gsMatrix< index_t > > &	boundaryDofs,
		const int	unk
	)

virtualinherited

Eliminate given DOFs as homogeneous Dirichlet boundary.

Parameters

[in]	boundaryDofs	indices of the given boundary DOFs
[in]	unk	the considered unknown

◆ nextIteration()

template<class T , int MatOrder>

void nextIteration ( const unsigned numberOfIterations )

virtualinherited

Perform several iteration steps.

Parameters

[in] numberOfIterations the number of iterations to be performed

◆ setSolution()

template<class T , int MatOrder>

virtual void setSolution ( const gsMatrix< T > & solVector )

inlinevirtualinherited

Set a given solution vector as current solution.

Parameters

[in] solVector the given solution

◆ solutionChangeRelNorm()

template<class T , int MatOrder>

T solutionChangeRelNorm	(	gsMatrix< T >	solOld,
		gsMatrix< T >	solNew
	)		const

inherited

Compute and return the relative norm of the solution change given the two successive solutions.

Parameters

[in]	solOld	the old solution
[in]	solNew	the new solution

◆ solve()

template<class T , int MatOrder>

void solve	(	const int	maxIterations,
		const T	epsilon = `1e-3`,
		const int	minIterations = `1`
	)

inherited

Solve the incompressible Navier-Stokes problem.

Parameters

[in]	maxIterations	the maximum number of linearization method iterations (in the steady case) or time steps (in the unsteady case)
[in]	epsilon	the stopping tolerance
[in]	minIterations	the minimum number of iterations/time steps

◆ updateAssembler() [1/2]

template<class T , int MatOrder>

virtual void updateAssembler ( bool updateSol = true )

inlinevirtualinherited

Update the assembler with current solution.

Parameters

[in] updateSol save current solution into m_solution in the assembler

◆ updateAssembler() [2/2]

template<class T , int MatOrder>

void updateAssembler	(	const gsMatrix< T > &	sol,
		bool	updateSol = `true`
	)

virtualinherited

Update the assembler with a given solution.

Parameters

[in] updateSol save the given solution into m_solution in the assembler

◆ writeSolChangeRelNorm()

template<class T , int MatOrder>

void writeSolChangeRelNorm	(	gsMatrix< T >	solOld,
		gsMatrix< T >	solNew
	)

virtualinherited

Compute and display the relative norm of the solution change given the two successive solutions.

Parameters

[in]	solOld	the old solution
[in]	solNew	the new solution

Detailed Description

Public Member Functions

Protected Member Functions

Member Function Documentation

◆ applySolver()

◆ checkGeoJacobian()

◆ constructSolution()

◆ initIteration()

◆ markDofsAsEliminatedZeros()

◆ nextIteration()

◆ setSolution()

◆ solutionChangeRelNorm()

◆ solve()

◆ updateAssembler() [1/2]

◆ updateAssembler() [2/2]

◆ writeSolChangeRelNorm()