#include <calmls.h>

Inheritance diagram for oofem::CylindricalALM:

Collaboration diagram for oofem::CylindricalALM:

Public Member Functions
	CylindricalALM (Domain d, EngngModel m)
virtual	~CylindricalALM ()
ConvergedReason	solve (SparseMtrx &K, FloatArray &R, FloatArray R0, FloatArray &X, FloatArray &dX, FloatArray &F, const FloatArray &internalForcesEBENorm, double &ReachedLambda, referenceLoadInputModeType rlm, int &nite, TimeStep ) override
double	giveCurrentStepLength () override
void	setStepLength (double s) override
void	initializeFrom (InputRecord &ir) override
bool	referenceLoad () const override
void	saveContext (DataStream &stream, ContextMode mode) override
void	restoreContext (DataStream &stream, ContextMode mode) override
void	setDomain (Domain *d) override
void	reinitialize () override
const char *	giveClassName () const override
const char *	giveInputRecordName () const
SparseLinearSystemNM *	giveLinearSolver () override
Public Member Functions inherited from oofem::SparseNonLinearSystemNM
	SparseNonLinearSystemNM (Domain d, EngngModel m)
	Constructor.
virtual	~SparseNonLinearSystemNM ()
	Destructor.
virtual void	printState (FILE *outputStream)
void	initializeFrom (InputRecord &ir) override
virtual void	convertPertMap ()
virtual void	applyPerturbation (FloatArray *displacement)
std::string	errorInfo (const char *func) const
	Error printing helper.
Public Member Functions inherited from oofem::NumericalMethod
	NumericalMethod (Domain d, EngngModel m)
virtual	~NumericalMethod ()
	Destructor.
EngngModel *	giveEngngModel ()

Protected Types
enum	calm_ControlType { calm_hpc_off = 0 , calm_hpc_on , calml_hpc }
	CALM mode type; determines the calm step length control. More...
enum	calm_NR_ModeType { calm_modifiedNRM , calm_fullNRM , calm_accelNRM }
	Controlling mode of Newton-Raphson Method. More...
enum	RootSelectionType { RST_Cos =0 , RST_Dot =1 }
typedef std ::set< DofIDItem >	__DofIDSet

Protected Member Functions
void	convertHPCMap ()
int	computeDeltaLambda (double &deltaLambda, const FloatArray &dX, const FloatArray &deltaXt, const FloatArray &deltaX_, const FloatArray &R, double RR, double eta, double deltaL, double DeltaLambda0, int neq)
void	search (int istep, FloatArray &prod, FloatArray &eta, double amp, double maxeta, double mineta, int &status)
bool	checkConvergence (const FloatArray &R, const FloatArray *R0, const FloatArray &F, const FloatArray &X, const FloatArray &ddX, double Lambda, double RR0, double RR, double drProduct, const FloatArray &internalForcesEBENorm, int nite, bool &errorOutOfRange)
	Evaluates the convergence criteria.
void	do_lineSearch (FloatArray &X, const FloatArray &XInitial, const FloatArray &deltaX_, const FloatArray &deltaXt, const FloatArray &dXm1, FloatArray &dX, FloatArray &ddX, const FloatArray &R, const FloatArray R0, const FloatArray &F, double &DeltaLambda, double &DeltaLambdam1, double &deltaLambda, double &Lambda, double &ReachedLambda, double RR, double &drProduct, TimeStep tStep)
	Perform line search optimization of step length.

Protected Attributes
int	nsmax
int	maxRestarts
double	Psi
double	deltaL
double	minStepLength
double	maxStepLength
int	solved
int	numberOfRequiredIterations
calm_NR_ModeType	calm_NR_Mode
calm_NR_ModeType	calm_NR_OldMode
int	calm_NR_ModeTick
int	calm_MANRMSteps
int	minIterations
	Minimum hard number of iteration.s.
int	calm_hpc_init
	Variables for HyperPlaneControl.
calm_ControlType	calm_Control
FloatArray	calm_HPCWeights
IntArray	calm_HPCIndirectDofMask
	Array containing equation numbers of dofs under indirect control.
IntArray	calm_HPCDmanDofSrcArray
	Input array containing dofmanagers and corresponding dof numbers under indirect control.
FloatArray	calm_HPCDmanWeightSrcArray
	Input array of dofman weights (for hpcmode 2).
std ::unique_ptr< SparseLinearSystemNM >	linSolver
	Linear system solver.
LinSystSolverType	solverType
	linear system solver ID.
int	lsFlag
	Line search flag.
double	ls_tolerance
	Line search tolerance.
double	amplifFactor
	Line search amplification factor.
double	maxEta
	Line search parameters (limits).
double	minEta
int	nccdg
	Number of convergence criteria dof groups.
std ::vector< __DofIDSet >	ccDofGroups
	Convergence criteria dof groups.
FloatArray	rtolf
	Relative unbalanced force tolerance for each group.
FloatArray	rtold
	Relative iterative displacement change tolerance for each group.
ParallelContext *	parallel_context
	Parallel context for computing norms, dot products and such.
RootSelectionType	rootselectiontype
	Root selection type.
FloatArray	old_dX
	previous increment of dX, needed by root selection type 1
Protected Attributes inherited from oofem::SparseNonLinearSystemNM
double	deltaL =0.0
	Load level.
double	randPertAmplitude =0.0
	Amplitude of a random perturbation applied on the solution before the iteration process.
int	randSeed =0
bool	pert_init_needed
IntArray	igp_PertDmanDofSrcArray
FloatArray	igp_PertWeightArray
IntArray	igp_Map
FloatArray	igp_Weight
Protected Attributes inherited from oofem::NumericalMethod
Domain *	domain
	Pointer to domain.
EngngModel *	engngModel
	Pointer to engineering model.

Additional Inherited Members
Public Types inherited from oofem::SparseNonLinearSystemNM
enum	referenceLoadInputModeType { rlm_total =0 , rlm_incremental =1 }

Detailed Description

Implementation of sparse nonlinear solver with indirect control. It uses Cylindrical Arc-Length Method algorithm.

DESCRIPTION : Perform solution of non-linear problem in the form Kt * deltaR = g where g is defined as g = g(Lambda,X) = Lambda* R - F(X). During iteration process, we want g became zero vector.

This method uses Modified Newton Raphson iteration scheme

If we solve non-linear static we can interpret symbols as follows:

Kt - tangential stiffness dX - increment of displacements g - vector of unbalanced forces (at the end should be zero one) Lambda - Load level Rt - Quasi Total Load vector Rt = R0 + Lambda*R X - total displacement vector F(X) - Nodal representation of (real) internal forces. Psi - control parameter (=0. means displacement control) Bergan_k0 - variable used to compute bergan's parameters of current stiffness. calm_NR_Mode - variable controlling the mode of NRM (ModifiedNR, Full NRM (stiffness update after each iteration), Modified Accelerated NRM (we perform iteration with stiffness matrix updated only after calm_MANRMSteps) calm_NR_OldMode - variable containing the old mode of NRM, which will be restored after calm_NR_ModeTick iterations. calm_NR_ModeTick - see calm_NR_OldMode. calm_MANRMSteps - if calm_NR_Mode == calm_accelNRM, it specifies, that new updated stiffness matrix is assembled after calm_MANRMSteps. calm_Control - variable indicating the ALM control. calm_HPCIndirectDofMask - Mask, telling which dofs are used for HPC. calm_HPCWeights - dofs weights in constrain.

Tasks:

solving problem: solveYourselfAt.
returning results (increment of displacement, reached level of loading and so on)

Variable description :

K(N,N) = Stiffness matrix (assumed positive definite)
dX(N) = Iterative increment of displacement
Rt = "quasi" total load vector
R0 = Fixed (Initial) load vector
deltaL = Step length
deltaLambda = Increment of load level
Psi = Determines loading or displacement control
dX = Current total increment
F = Nodal representation of (real) internal forces
DeltaLambda = Current total increment of load level
Lambda = Total load level
RTOL= Convergence tolerance
NSMAX = Maximum number of sweeps allowed

OUTPUT : (after call solveYourselfAt)

dX = Reached displacement increment
Lambda = Total load level
nite = Number of iterations required to fulfill balance
status = NM_status with flags set to reached state

Definition at line 146 of file calmls.h.

Member Typedef Documentation

◆ __DofIDSet

typedef std :: set< DofIDItem > oofem::CylindricalALM::__DofIDSet

protected

Definition at line 163 of file calmls.h.

Member Enumeration Documentation

◆ calm_ControlType

enum oofem::CylindricalALM::calm_ControlType

protected

CALM mode type; determines the calm step length control.

Enumerator
calm_hpc_off	Full ALM with quadratic constrain and all dofs.
calm_hpc_on	Full ALM with quadratic constrain, taking into account only selected dofs.
calml_hpc	Linearized ALM (only displacements), taking into account only selected dofs with given weight.

Definition at line 150 of file calmls.h.

◆ calm_NR_ModeType

enum oofem::CylindricalALM::calm_NR_ModeType

protected

Controlling mode of Newton-Raphson Method.

Enumerator
calm_modifiedNRM	Keeps the old tangent.
calm_fullNRM	Updates the tangent every iteration.
calm_accelNRM	Updates the tangent after a few steps.

Definition at line 157 of file calmls.h.

◆ RootSelectionType

enum oofem::CylindricalALM::RootSelectionType

protected

Type characterizing how the roots of the quadratic equation for the arc-length parameter are selected 0 – cosinus based criterion 1 – dot product based criterion

Enumerator
RST_Cos
RST_Dot

Definition at line 219 of file calmls.h.

Constructor & Destructor Documentation

◆ CylindricalALM()

oofem::CylindricalALM::CylindricalALM	(	Domain *	d,
		EngngModel *	m )

References CylindricalALM(), and solve().

Referenced by CylindricalALM().

◆ ~CylindricalALM()

oofem::CylindricalALM::~CylindricalALM ( )

virtual

Definition at line 103 of file calmls.C.

Member Function Documentation

◆ checkConvergence()

bool oofem::CylindricalALM::checkConvergence	(	const FloatArray &	R,
		const FloatArray *	R0,
		const FloatArray &	F,
		const FloatArray &	X,
		const FloatArray &	ddX,
		double	Lambda,
		double	RR0,
		double	RR,
		double	drProduct,
		const FloatArray &	internalForcesEBENorm,
		int	nite,
		bool &	errorOutOfRange )

protected

Evaluates the convergence criteria.

Definition at line 484 of file calmls.C.

References oofem::FloatArray::add(), oofem::FloatArray::at(), CALM_MAX_REL_ERROR_BOUND, calm_SMALL_ERROR_NUM, ccDofGroups, oofem::NumericalMethod::domain, oofem::Element_local, nccdg, OOFEM_LOG_INFO, parallel_context, rtold, rtolf, oofem::FloatArray::subtract(), oofem::FloatArray::sum(), oofem::FloatArray::times(), and oofem::FloatArray::zero().

Referenced by solve().

◆ computeDeltaLambda()

int oofem::CylindricalALM::computeDeltaLambda	(	double &	deltaLambda,
		const FloatArray &	dX,
		const FloatArray &	deltaXt,
		const FloatArray &	deltaX_,
		const FloatArray &	R,
		double	RR,
		double	eta,
		double	deltaL,
		double	DeltaLambda0,
		int	neq )

protected

Returns: If 0 then ok, 1 then failure (restart).

Definition at line 995 of file calmls.C.

References oofem::FloatArray::at(), calm_Control, calm_hpc_off, calm_hpc_on, calm_HPCIndirectDofMask, calm_HPCWeights, calm_SMALL_NUM, calml_hpc, deltaL, OOFEM_ERROR, OOFEM_WARNING, parallel_context, Psi, rootselectiontype, RST_Cos, oofem::Vec2(), and oofem::Vec4().

Referenced by do_lineSearch(), and solve().

◆ convertHPCMap()

void oofem::CylindricalALM::convertHPCMap ( )

protected

Definition at line 900 of file calmls.C.

References oofem::FloatArray::at(), oofem::IntArray::at(), calm_Control, calm_HPCDmanDofSrcArray, calm_HPCDmanWeightSrcArray, calm_HPCIndirectDofMask, calm_HPCWeights, calml_hpc, oofem::NumericalMethod::domain, OOFEM_WARNING, parallel_context, oofem::FloatArray::resize(), and oofem::IntArray::resize().

Referenced by solve().

◆ do_lineSearch()

void oofem::CylindricalALM::do_lineSearch	(	FloatArray &	X,
		const FloatArray &	XInitial,
		const FloatArray &	deltaX_,
		const FloatArray &	deltaXt,
		const FloatArray &	dXm1,
		FloatArray &	dX,
		FloatArray &	ddX,
		const FloatArray &	R,
		const FloatArray *	R0,
		const FloatArray &	F,
		double &	DeltaLambda,
		double &	DeltaLambdam1,
		double &	deltaLambda,
		double &	Lambda,
		double &	ReachedLambda,
		double	RR,
		double &	drProduct,
		TimeStep *	tStep )

protected

Perform line search optimization of step length.

Definition at line 1235 of file calmls.C.

References oofem::FloatArray::add(), amplifFactor, oofem::FloatArray::at(), oofem::FloatArray::clear(), computeDeltaLambda(), deltaL, oofem::NumericalMethod::domain, oofem::NumericalMethod::engngModel, oofem::FloatArray::giveSize(), oofem::TimeStep::incrementStateCounter(), oofem::InternalRhs, ls_tolerance, maxEta, minEta, OOFEM_LOG_INFO, parallel_context, and search().

Referenced by solve().

◆ giveClassName()

const char * oofem::CylindricalALM::giveClassName ( ) const

inlineoverridevirtual

Reimplemented from oofem::SparseNonLinearSystemNM.

Definition at line 256 of file calmls.h.

◆ giveCurrentStepLength()

double oofem::CylindricalALM::giveCurrentStepLength ( )

inlineoverridevirtual

Returns step length.

See also: solve For more details on the step length s.

Returns: Current step length

Reimplemented from oofem::SparseNonLinearSystemNM.

Definition at line 238 of file calmls.h.

References deltaL.

◆ giveInputRecordName()

const char * oofem::CylindricalALM::giveInputRecordName ( ) const

inline

Definition at line 257 of file calmls.h.

References _IFT_CylindricalALM_Name.

◆ giveLinearSolver()

SparseLinearSystemNM * oofem::CylindricalALM::giveLinearSolver ( )

overridevirtual

Constructs (if necessary) and returns a linear solver. Public method because some problems require it for sensitivity analysis, etc. even for nonlinear problems (e.g. tangent relations in multiscale simulations).

Reimplemented from oofem::SparseNonLinearSystemNM.

Definition at line 977 of file calmls.C.

References oofem::classFactory, oofem::NumericalMethod::domain, oofem::NumericalMethod::engngModel, linSolver, OOFEM_ERROR, and solverType.

Referenced by initializeFrom(), and solve().

◆ initializeFrom()

void oofem::CylindricalALM::initializeFrom ( InputRecord & ir )

overridevirtual

Reimplemented from oofem::NumericalMethod.

Definition at line 681 of file calmls.C.

◆ referenceLoad()

bool oofem::CylindricalALM::referenceLoad ( ) const

inlineoverridevirtual

Returns true if reference loads are used (i.e. arc length methods).

Reimplemented from oofem::SparseNonLinearSystemNM.

Definition at line 241 of file calmls.h.

◆ reinitialize()

void oofem::CylindricalALM::reinitialize ( )

inlineoverridevirtual

Reinitializes the receiver. This is used, when topology of problem has changed (for example after adaptive refinement or load transfer in parallel applications). This is necessary for numerical methods, that cache some data between solution steps and that may depend on domain or problem topology. The caching of data by receiver is intended only for speeding up the calculation, but numerical method must be always able to generate this data again. This method clears receiver cached data dependent on topology, when it changes.

Reimplemented from oofem::NumericalMethod.

Definition at line 250 of file calmls.h.

References calm_hpc_init, and linSolver.

◆ restoreContext()

void oofem::CylindricalALM::restoreContext	(	DataStream &	stream,
		ContextMode	mode )

overridevirtual

Reimplemented from oofem::NumericalMethod.

Definition at line 968 of file calmls.C.

References oofem::CIO_IOERR, deltaL, oofem::DataStream::read(), and THROW_CIOERR.

◆ saveContext()

void oofem::CylindricalALM::saveContext	(	DataStream &	stream,
		ContextMode	mode )

overridevirtual

Reimplemented from oofem::NumericalMethod.

Definition at line 959 of file calmls.C.

References oofem::CIO_IOERR, deltaL, THROW_CIOERR, and oofem::DataStream::write().

◆ search()

void oofem::CylindricalALM::search	(	int	istep,
		FloatArray &	prod,
		FloatArray &	eta,
		double	amp,
		double	maxeta,
		double	mineta,
		int &	status )

protected

Definition at line 1149 of file calmls.C.

References oofem::FloatArray::at(), and oofem::max().

Referenced by do_lineSearch().

◆ setDomain()

void oofem::CylindricalALM::setDomain ( Domain * d )

inlineoverridevirtual

Reimplemented from oofem::NumericalMethod.

Definition at line 244 of file calmls.h.

References oofem::NumericalMethod::domain, and linSolver.

◆ setStepLength()

void oofem::CylindricalALM::setStepLength ( double s )

inlineoverridevirtual

Sets the step length.

See also: solve For more details on the step length s.

Parameters

s	New step length.

Reimplemented from oofem::SparseNonLinearSystemNM.

Definition at line 239 of file calmls.h.

References deltaL.

◆ solve()

ConvergedReason oofem::CylindricalALM::solve	(	SparseMtrx &	K,
		FloatArray &	R,
		FloatArray *	R0,
		FloatArray &	X,
		FloatArray &	dX,
		FloatArray &	F,
		const FloatArray &	internalForcesEBENorm,
		double &	s,
		referenceLoadInputModeType	rlm,
		int &	nite,
		TimeStep *	tStep )

overridevirtual

Solves the given sparse linear system of equations \( s R + R_0 - F(X) = 0 \). Total load vector is not passed, it is defined as \( s R + R_0 \), where s is scale factor.

See also: EngngModel::updateComponent Used to update the stiffness matrix and load vector.

Parameters

K	Coefficient matrix ( \(\displaystyle K = \frac{\partial F}{\partial X} \); stiffness matrix).
R	Reference incremental RHS (incremental load).
R0	Initial RHS (initial load).
X	Total solution (total displacement).
dX	Increment of solution (incremental displacements).
F	InternalRhs (real internal forces).
internalForcesEBENorm	Norm of internal nodal forces (evaluated on element by element basis) (split into each DOF id).
s	RHS scale factor (load level).
rlm	Reference load mode.
nite	Number of iterations needed.
tStep	Time step to solve for.

Returns: ConvergedReason value.

Implements oofem::SparseNonLinearSystemNM.

Definition at line 108 of file calmls.C.

References oofem::FloatArray::add(), oofem::FloatArray::at(), calm_accelNRM, calm_Control, CALM_DEFAULT_NRM_TICKS, calm_fullNRM, calm_hpc_init, calm_hpc_off, calm_hpc_on, calm_HPCIndirectDofMask, calm_HPCWeights, calm_MANRMSteps, calm_NR_Mode, calm_NR_ModeTick, calm_NR_OldMode, CALM_RESET_STEP_REDUCE, calml_hpc, checkConvergence(), oofem::FloatArray::clear(), computeDeltaLambda(), convertHPCMap(), oofem::CR_CONVERGED, oofem::CR_DIVERGED_ITS, oofem::CR_UNKNOWN, deltaL, do_lineSearch(), oofem::NumericalMethod::domain, oofem::NumericalMethod::engngModel, giveLinearSolver(), oofem::TimeStep::giveNumber(), oofem::FloatArray::giveSize(), oofem::TimeStep::incrementStateCounter(), oofem::TimeStep::incrementSubStepNumber(), oofem::InternalRhs, linSolver, lsFlag, maxRestarts, maxStepLength, minIterations, minStepLength, nccdg, oofem::NonLinearLhs, nsmax, numberOfRequiredIterations, old_dX, OOFEM_ERROR, OOFEM_LOG_INFO, OOFEM_WARNING, parallel_context, Psi, oofem::FloatArray::resize(), rootselectiontype, RST_Cos, oofem::sgn(), solved, oofem::FloatArray::subtract(), oofem::FloatArray::times(), oofem::Vec2(), and oofem::FloatArray::zero().

Referenced by CylindricalALM().

Member Data Documentation

◆ amplifFactor

double oofem::CylindricalALM::amplifFactor

protected

Line search amplification factor.

Definition at line 198 of file calmls.h.

Referenced by do_lineSearch(), and initializeFrom().

◆ calm_Control

calm_ControlType oofem::CylindricalALM::calm_Control

protected

Definition at line 179 of file calmls.h.

Referenced by computeDeltaLambda(), convertHPCMap(), initializeFrom(), and solve().

◆ calm_hpc_init

int oofem::CylindricalALM::calm_hpc_init

protected

Variables for HyperPlaneControl.

Definition at line 178 of file calmls.h.

Referenced by initializeFrom(), reinitialize(), and solve().

◆ calm_HPCDmanDofSrcArray

IntArray oofem::CylindricalALM::calm_HPCDmanDofSrcArray

protected

Input array containing dofmanagers and corresponding dof numbers under indirect control.

Definition at line 184 of file calmls.h.

Referenced by convertHPCMap(), and initializeFrom().

◆ calm_HPCDmanWeightSrcArray

FloatArray oofem::CylindricalALM::calm_HPCDmanWeightSrcArray

protected

Input array of dofman weights (for hpcmode 2).

Definition at line 186 of file calmls.h.

Referenced by convertHPCMap(), and initializeFrom().

◆ calm_HPCIndirectDofMask

IntArray oofem::CylindricalALM::calm_HPCIndirectDofMask

protected

Array containing equation numbers of dofs under indirect control.

Definition at line 182 of file calmls.h.

Referenced by computeDeltaLambda(), convertHPCMap(), and solve().

◆ calm_HPCWeights

FloatArray oofem::CylindricalALM::calm_HPCWeights

protected

Definition at line 180 of file calmls.h.

Referenced by computeDeltaLambda(), convertHPCMap(), initializeFrom(), and solve().

◆ calm_MANRMSteps

int oofem::CylindricalALM::calm_MANRMSteps

protected

Definition at line 172 of file calmls.h.

Referenced by initializeFrom(), and solve().

◆ calm_NR_Mode

calm_NR_ModeType oofem::CylindricalALM::calm_NR_Mode

protected

Definition at line 170 of file calmls.h.

Referenced by initializeFrom(), and solve().

◆ calm_NR_ModeTick

int oofem::CylindricalALM::calm_NR_ModeTick

protected

Definition at line 171 of file calmls.h.

Referenced by solve().

◆ calm_NR_OldMode

calm_NR_ModeType oofem::CylindricalALM::calm_NR_OldMode

protected

Definition at line 170 of file calmls.h.

Referenced by initializeFrom(), and solve().

◆ ccDofGroups

std :: vector< __DofIDSet > oofem::CylindricalALM::ccDofGroups

protected

Convergence criteria dof groups.

Definition at line 206 of file calmls.h.

Referenced by checkConvergence(), and initializeFrom().

◆ deltaL

double oofem::CylindricalALM::deltaL

protected

Definition at line 168 of file calmls.h.

Referenced by computeDeltaLambda(), do_lineSearch(), giveCurrentStepLength(), initializeFrom(), restoreContext(), saveContext(), setStepLength(), and solve().

◆ linSolver

std :: unique_ptr< SparseLinearSystemNM > oofem::CylindricalALM::linSolver

protected

Linear system solver.

Definition at line 189 of file calmls.h.

Referenced by giveLinearSolver(), reinitialize(), setDomain(), and solve().

◆ ls_tolerance

double oofem::CylindricalALM::ls_tolerance

protected

Line search tolerance.

Definition at line 196 of file calmls.h.

Referenced by do_lineSearch(), and initializeFrom().

◆ lsFlag

int oofem::CylindricalALM::lsFlag

protected

Line search flag.

Definition at line 194 of file calmls.h.

Referenced by initializeFrom(), and solve().

◆ maxEta

double oofem::CylindricalALM::maxEta

protected

Line search parameters (limits).

Definition at line 200 of file calmls.h.

Referenced by do_lineSearch(), and initializeFrom().

◆ maxRestarts

int oofem::CylindricalALM::maxRestarts

protected

Definition at line 166 of file calmls.h.

Referenced by initializeFrom(), and solve().

◆ maxStepLength

double oofem::CylindricalALM::maxStepLength

protected

Definition at line 168 of file calmls.h.

Referenced by initializeFrom(), and solve().

◆ minEta

double oofem::CylindricalALM::minEta

protected

Definition at line 200 of file calmls.h.

Referenced by do_lineSearch().

◆ minIterations

int oofem::CylindricalALM::minIterations

protected

Minimum hard number of iteration.s.

Definition at line 175 of file calmls.h.

Referenced by initializeFrom(), and solve().

◆ minStepLength

double oofem::CylindricalALM::minStepLength

protected

Definition at line 168 of file calmls.h.

Referenced by initializeFrom(), and solve().

◆ nccdg

int oofem::CylindricalALM::nccdg

protected

Number of convergence criteria dof groups.

Definition at line 204 of file calmls.h.

Referenced by checkConvergence(), initializeFrom(), and solve().

◆ nsmax

int oofem::CylindricalALM::nsmax

protected

Definition at line 165 of file calmls.h.

Referenced by initializeFrom(), and solve().

◆ numberOfRequiredIterations

int oofem::CylindricalALM::numberOfRequiredIterations

protected

Definition at line 169 of file calmls.h.

Referenced by initializeFrom(), and solve().

◆ old_dX

FloatArray oofem::CylindricalALM::old_dX

protected

previous increment of dX, needed by root selection type 1

Definition at line 227 of file calmls.h.

Referenced by solve().

◆ parallel_context

ParallelContext* oofem::CylindricalALM::parallel_context

protected

Parallel context for computing norms, dot products and such.

Definition at line 212 of file calmls.h.

Referenced by checkConvergence(), computeDeltaLambda(), convertHPCMap(), do_lineSearch(), and solve().

◆ Psi

double oofem::CylindricalALM::Psi

protected

Definition at line 167 of file calmls.h.

Referenced by computeDeltaLambda(), initializeFrom(), and solve().

◆ rootselectiontype

RootSelectionType oofem::CylindricalALM::rootselectiontype

protected

Root selection type.

Definition at line 225 of file calmls.h.

Referenced by computeDeltaLambda(), initializeFrom(), and solve().

◆ rtold

FloatArray oofem::CylindricalALM::rtold

protected

Relative iterative displacement change tolerance for each group.

Definition at line 210 of file calmls.h.

Referenced by checkConvergence(), and initializeFrom().

◆ rtolf

FloatArray oofem::CylindricalALM::rtolf

protected

Relative unbalanced force tolerance for each group.

Definition at line 208 of file calmls.h.

Referenced by checkConvergence(), and initializeFrom().

◆ solved

int oofem::CylindricalALM::solved

protected

Definition at line 169 of file calmls.h.

Referenced by solve().

◆ solverType

LinSystSolverType oofem::CylindricalALM::solverType

protected

linear system solver ID.

Definition at line 191 of file calmls.h.

Referenced by giveLinearSolver(), and initializeFrom().

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

Public Member Functions

Protected Types

Protected Member Functions

Protected Attributes

Additional Inherited Members

Detailed Description

Member Typedef Documentation

◆ __DofIDSet

Member Enumeration Documentation

◆ calm_ControlType

◆ calm_NR_ModeType

◆ RootSelectionType

Constructor & Destructor Documentation

◆ CylindricalALM()

◆ ~CylindricalALM()

Member Function Documentation

◆ checkConvergence()

◆ computeDeltaLambda()

◆ convertHPCMap()

◆ do_lineSearch()

◆ giveClassName()

◆ giveCurrentStepLength()

◆ giveInputRecordName()

◆ giveLinearSolver()

◆ initializeFrom()

◆ referenceLoad()

◆ reinitialize()

◆ restoreContext()

◆ saveContext()

◆ search()

◆ setDomain()

◆ setStepLength()

◆ solve()

Member Data Documentation

◆ amplifFactor

◆ calm_Control

◆ calm_hpc_init

◆ calm_HPCDmanDofSrcArray

◆ calm_HPCDmanWeightSrcArray

◆ calm_HPCIndirectDofMask

◆ calm_HPCWeights

◆ calm_MANRMSteps

◆ calm_NR_Mode

◆ calm_NR_ModeTick

◆ calm_NR_OldMode

◆ ccDofGroups

◆ deltaL

◆ linSolver

◆ ls_tolerance

◆ lsFlag

◆ maxEta

◆ maxRestarts

◆ maxStepLength

◆ minEta

◆ minIterations

◆ minStepLength

◆ nccdg

◆ nsmax

◆ numberOfRequiredIterations

◆ old_dX

◆ parallel_context

◆ Psi

◆ rootselectiontype

◆ rtold

◆ rtolf

◆ solved

◆ solverType