Wrapper class for 'beuler' integrator We derive the class from the class Integrator. More...

#include <BEulerInt.h>

Inheritance diagram for BEulerInt:

Collaboration diagram for BEulerInt:

Public Member Functions
	BEulerInt ()
	The default constructor doesn't take an argument.

virtual	~BEulerInt ()
	Destructor.

virtual void	setTolerances (double reltol, size_t n, double *abstol)
	Set or reset the number of equations.

virtual void	setTolerances (double reltol, double abstol)
	Set error tolerances.

virtual void	setProblemType (int probtype)
	Set the problem type.

virtual void	initializeRJE (double t0, ResidJacEval &func)

virtual void	reinitializeRJE (double t0, ResidJacEval &func)

virtual double	integrateRJE (double tout, double tinit=0.0)

virtual doublereal	step (double tout)
	Integrate the system of equations.

virtual void	setSolnWeights ()

virtual double &	solution (size_t k)
	The current value of the solution of equation k.

double *	solution ()
	The current value of the solution of the system of equations.

int	nEquations () const
	The number of equations.

virtual int	nEvals () const
	The number of function evaluations.

virtual void	setMethodBEMT (BEulerMethodType t)

virtual void	setIterator (IterType t)
	Set the linear iterator.

virtual void	setMaxStep (double hmax)

virtual void	setMaxNumTimeSteps (int)

virtual void	setNumInitialConstantDeltaTSteps (int)

void	print_solnDelta_norm_contrib (const double const soln0, const char const s0, const double const soln1, const char const s1, const char const title, const double const y0, const double *const y1, double damp, int num_entries)

virtual void	setPrintSolnOptions (int printSolnStepInterval, int printSolnNumberToTout, int printSolnFirstSteps=0, bool dumpJacobians=false)

void	setNonLinOptions (int min_newt_its=0, bool matrixConditioning=false, bool colScaling=false, bool rowScaling=true)

virtual void	setPrintFlag (int print_flag)

virtual void	setColumnScales ()

virtual double	soln_error_norm (const double *const, bool printLargest=false)
	calculate the solution error norm

virtual void	setInitialTimeStep (double delta_t)

void	beuler_jac (GeneralMatrix &, double const, double, double, double const, double *const, int)

virtual void	setSensitivityTolerances (doublereal reltol, doublereal abstol)
	Set the sensitivity error tolerances.

virtual void	initialize (doublereal t0, FuncEval &func)
	Initialize the integrator for a new problem.

virtual void	reinitialize (doublereal t0, FuncEval &func)

virtual void	integrate (doublereal tout)
	Integrate the system of equations.

virtual void	setMaxOrder (int n)
	Set the maximum integration order that will be used.

virtual void	setMethod (MethodType t)
	Set the solution method.

virtual void	setMaxStepSize (double hmax)
	Set the maximum step size.

virtual void	setMinStepSize (double hmin)
	Set the minimum step size.

virtual void	setMaxSteps (int nmax)

virtual void	setBandwidth (int N_Upper, int N_Lower)

virtual int	nSensParams ()

virtual double	sensitivity (size_t k, size_t p)

Protected Member Functions
void	internalMalloc ()
	Internal routine that sets up the fixed length storage based on the size of the problem to solve.

void	calc_y_pred (int)
	Internal function to calculate the predicted solution at a time step.

void	calc_ydot (int, double , double )
	Internal function to calculate the time derivative at the new step.

double	time_error_norm ()
	Internal function to calculate the time step truncation error for a predictor corrector time step.

double	time_step_control (int m_order, double time_error_factor)
	Internal function to calculate the time step for the next step based on the time-truncation error on the current time step.

int	solve_nonlinear_problem (double const y_comm, double const ydot_comm, double CJ, double time_curr, GeneralMatrix &jac, int &num_newt_its, int &num_linear_solves, int &num_backtracks, int loglevel)
	Solve a nonlinear system.

void	doNewtonSolve (double, double , double , double *, GeneralMatrix &, int)
	Compute the undamped Newton step.

double	boundStep (const double const y, const double const step0, int loglevel)
	Bound the Newton step while relaxing the solution.

int	dampStep (double, const double , const double , const double , double , double , double , double &, GeneralMatrix &, int &, bool, int &)

void	computeResidWts (GeneralMatrix &jac)

double	filterNewStep (double, double , double )

double	getPrintTime (double time_current)

Protected Attributes
IterType	m_iter
	IterType is used to specify how the nonlinear equations are to be relaxed at each time step.

BEulerMethodType	m_method
	MethodType is used to specify how the time step is to be chosen.

int	m_jacFormMethod
	m_jacFormMethod determines how a matrix is formed.

bool	m_rowScaling
	m_rowScaling is a boolean.

bool	m_colScaling
	m_colScaling is a boolean.

bool	m_matrixConditioning
	m_matrixConditioning is a boolean.

int	m_itol
	If m_itol =1 then each component has an individual value of atol.

double	m_reltol
	Relative time truncation error tolerances.

double	m_abstols
	Absolute time truncation error tolerances, when uniform for all variables.

double *	m_abstol
	Vector of absolute time truncation error tolerance when not uniform for all variables.

double *	m_ewt
	Error Weights.

double	m_hmax
	Maximum step size.

int	m_maxord
	Maximum integration order.

int	m_order
	Current integration order.

int	m_time_step_num
	Time step number.

int	m_time_step_attempts

int	m_max_time_step_attempts
	Max time steps allowed.

int	m_numInitialConstantDeltaTSteps
	Number of initial time steps to take where the time truncation error tolerances are not checked.

int	m_failure_counter
	Failure Counter -> keeps track of the number of consequetive failures.

int	m_min_newt_its
	Minimum Number of Newton Iterations per nonlinear step default = 0.

int	m_printSolnStepInterval
	Step Interval at which to print out the solution default = 1; If set to zero, there is no printout.

int	m_printSolnNumberToTout
	Number of evenly spaced printouts of the solution If zero, there is no printout from this option default 1 If set to zero there is no printout.

int	m_printSolnFirstSteps
	Number of initial steps that the solution is printed out.

bool	m_dumpJacobians
	Dump Jacobians to disk default false.

int	m_neq
	Number of equations in the ode integrator.

double *	m_y_n

double *	m_y_nm1

double *	m_y_pred_n

double *	m_ydot_n

double *	m_ydot_nm1

double	m_t0
	Initial time at the start of the integration.

double	m_time_final
	Final time.

double	time_n

double	time_nm1

double	time_nm2

double	delta_t_n

double	delta_t_nm1

double	delta_t_nm2

double	delta_t_np1

double	delta_t_max
	Maximum permissible time step.

double *	m_resid

double *	m_residWts

double *	m_wksp

ResidJacEval *	m_func

double *	m_rowScales

double *	m_colScales

GeneralMatrix *	tdjac_ptr
	Pointer to the jacobian representing the time dependent problem.

int	m_print_flag
	Determines the level of printing for each time step.

int	m_nfe
	Number of function evaluations.

int	m_nJacEval
	Number of Jacobian Evaluations and factorization steps (they are the same)

int	m_numTotalNewtIts
	Number of total newton iterations.

int	m_numTotalLinearSolves
	Total number of linear iterations.

int	m_numTotalConvFails
	Total number of convergence failures.

int	m_numTotalTruncFails
	Total Number of time truncation error failures.

int	num_failures

Detailed Description

Wrapper class for 'beuler' integrator We derive the class from the class Integrator.

Definition at line 61 of file BEulerInt.h.

Constructor & Destructor Documentation

BEulerInt ( )

The default constructor doesn't take an argument.

Definition at line 48 of file BEulerInt.cpp.

~BEulerInt ( )

virtual

Destructor.

Definition at line 110 of file BEulerInt.cpp.

References BEulerInt::m_abstol, BEulerInt::m_ewt, mdp::mdp_safe_free(), and BEulerInt::tdjac_ptr.

Member Function Documentation

void setTolerances	(	double	reltol,
		size_t	n,
		double *	abstol
	)

virtual

Set or reset the number of equations.

Set error tolerances.

Parameters

reltol	scalar relative tolerance
n	Number of equations
abstol	array of N absolute tolerance values

Reimplemented from Integrator.

Definition at line 127 of file BEulerInt.cpp.

References BEulerInt::m_abstol, BEulerInt::m_itol, BEulerInt::m_neq, and BEulerInt::m_reltol.

void setTolerances	(	double	reltol,
		double	abstol
	)

virtual

Set error tolerances.

Parameters

reltol	scalar relative tolerance
abstol	scalar absolute tolerance

Reimplemented from Integrator.

Definition at line 143 of file BEulerInt.cpp.

References BEulerInt::m_abstols, BEulerInt::m_itol, and BEulerInt::m_reltol.

void setProblemType ( int probtype )

virtual

Set the problem type.

Parameters

probtype Type of the problem

Reimplemented from Integrator.

Definition at line 150 of file BEulerInt.cpp.

References BEulerInt::m_jacFormMethod.

double step ( double tout )

virtual

Integrate the system of equations.

Parameters

tout	integrate to this time. Note that this is the absolute time value, not a time interval.

Reimplemented from Integrator.

Definition at line 1231 of file BEulerInt.cpp.

References BEulerInt::calc_y_pred(), BEulerInt::calc_ydot(), BEulerInt::delta_t_max, BEulerInt::m_failure_counter, BEulerInt::m_method, BEulerInt::m_numInitialConstantDeltaTSteps, BEulerInt::m_numTotalConvFails, BEulerInt::m_numTotalTruncFails, BEulerInt::m_order, BEulerInt::m_print_flag, BEulerInt::m_time_step_num, ckr::max(), mdp::mdp_copy_dbl_1(), ckr::min(), BEulerInt::solve_nonlinear_problem(), BEulerInt::tdjac_ptr, BEulerInt::time_error_norm(), and BEulerInt::time_step_control().

virtual double& solution ( size_t k )

inlinevirtual

The current value of the solution of equation k.

Reimplemented from Integrator.

Definition at line 78 of file BEulerInt.h.

double* solution ( )

inlinevirtual

The current value of the solution of the system of equations.

Reimplemented from Integrator.

Definition at line 81 of file BEulerInt.h.

int nEquations ( ) const

inlinevirtual

The number of equations.

Reimplemented from Integrator.

Definition at line 84 of file BEulerInt.h.

References BEulerInt::m_neq.

int nEvals ( ) const

virtual

The number of function evaluations.

Reimplemented from Integrator.

Definition at line 349 of file BEulerInt.cpp.

References BEulerInt::m_nfe.

void setIterator ( IterType t )

virtual

Set the linear iterator.

Reimplemented from Integrator.

Definition at line 211 of file BEulerInt.cpp.

References BEulerInt::m_iter.

double soln_error_norm	(	const double * const	delta_y,
		bool	printLargest = `false`
	)

virtual

calculate the solution error norm

L2 Norm of a delta in the solution.

The second argument has a default of false. However, if true, then a table of the largest values is printed out to standard output.

Definition at line 1648 of file BEulerInt.cpp.

References Cantera::error(), BEulerInt::m_ewt, and BEulerInt::m_neq.

void internalMalloc ( )

protected

Internal routine that sets up the fixed length storage based on the size of the problem to solve.

Definition at line 361 of file BEulerInt.cpp.

References BEulerInt::m_colScaling, BEulerInt::m_ewt, BEulerInt::m_neq, BEulerInt::m_rowScaling, and BEulerInt::tdjac_ptr.

void calc_y_pred ( int order )

protected

Internal function to calculate the predicted solution at a time step.

Definition at line 817 of file BEulerInt.cpp.

References ResidJacEval::filterSolnPrediction(), and BEulerInt::m_neq.

Referenced by BEulerInt::step().

void calc_ydot	(	int	order,
		double *	y_curr,
		double *	ydot_curr
	)

protected

Internal function to calculate the time derivative at the new step.

Definition at line 880 of file BEulerInt.cpp.

References BEulerInt::m_neq.

Referenced by BEulerInt::solve_nonlinear_problem(), and BEulerInt::step().

double time_error_norm ( )

protected

Internal function to calculate the time step truncation error for a predictor corrector time step.

Definition at line 925 of file BEulerInt.cpp.

References Cantera::error(), BEulerInt::m_ewt, BEulerInt::m_neq, and BEulerInt::m_print_flag.

Referenced by BEulerInt::step().

double time_step_control	(	int	m_order,
		double	time_error_factor
	)

protected

Internal function to calculate the time step for the next step based on the time-truncation error on the current time step.

Definition at line 1021 of file BEulerInt.cpp.

References BEulerInt::m_print_flag, ckr::max(), and ckr::min().

Referenced by BEulerInt::step().

int solve_nonlinear_problem	(	double *const	y_comm,
		double *const	ydot_comm,
		double	CJ,
		double	time_curr,
		GeneralMatrix &	jac,
		int &	num_newt_its,
		int &	num_linear_solves,
		int &	num_backtracks,
		int	loglevel
	)

protected

Solve a nonlinear system.

Find the solution to F(X, xprime) = 0 by damped Newton iteration. On entry, y_comm[] contains an initial estimate of the solution and ydot_comm[] contains an estimate of the derivative. On successful return, y_comm[] contains the converged solution and ydot_comm[] contains the derivative

Parameters

y_comm[]	Contains the input solution. On output y_comm[] contains the converged solution
ydot_comm	Contains the input derivative solution. On output y_comm[] contains the converged derivative solution
CJ	Inverse of the time step
time_curr	Current value of the time
jac	Jacobian
num_newt_its	number of newton iterations
num_linear_solves	number of linear solves
num_backtracks	number of backtracs
loglevel	Log level

Definition at line 2165 of file BEulerInt.cpp.

References BEulerInt::calc_ydot(), BEulerInt::doNewtonSolve(), BEulerInt::m_min_newt_its, BEulerInt::m_numTotalLinearSolves, BEulerInt::m_numTotalNewtIts, BEulerInt::m_order, and mdp::mdp_copy_dbl_1().

Referenced by BEulerInt::step().

void doNewtonSolve	(	double	time_curr,
		double *	y_curr,
		double *	ydot_curr,
		double *	delta_y,
		GeneralMatrix &	jac,
		int	loglevel
	)

protected

Compute the undamped Newton step.

The residual function is evaluated at x, but the Jacobian is not recomputed.

Definition at line 1711 of file BEulerInt.cpp.

References Cantera::Base_ResidEval, GeneralMatrix::begin(), ResidJacEval::evalResidNJ(), GeneralMatrix::factored(), BEulerInt::m_colScaling, BEulerInt::m_matrixConditioning, BEulerInt::m_neq, BEulerInt::m_nfe, BEulerInt::m_numTotalLinearSolves, BEulerInt::m_rowScaling, BEulerInt::m_time_step_num, ResidJacEval::matrixConditioning(), ResidJacEval::nEquations(), and GeneralMatrix::solve().

Referenced by BEulerInt::solve_nonlinear_problem().

double boundStep	(	const double *const	y,
		const double *const	step0,
		int	loglevel
	)

protected

Bound the Newton step while relaxing the solution.

Return the factor by which the undamped Newton step 'step0' must be multiplied in order to keep all solution components in all domains between their specified lower and upper bounds. Other bounds may be applied here as well.

Currently the bounds are hard coded into this routine:

Minimum value for all variables: - 0.01 * m_ewt[i] Maximum value = none.

Thus, this means that all solution components are expected to be numerical greater than zero in the limit of time step truncation errors going to zero.

Delta bounds: The idea behind these is that the Jacobian couldn't possibly be representative if the variable is changed by a lot. (true for nonlinear systems, false for linear systems) Maximum increase in variable in any one newton iteration: factor of 2 Maximum decrease in variable in any one newton iteration: factor of 5

Parameters

y	Current value of the solution
step0	Current raw step change in y[]
loglevel	Log level. This routine produces output if loglevel is greater than one

Returns: Returns the damping coefficient

Definition at line 1910 of file BEulerInt.cpp.

References BEulerInt::m_ewt, BEulerInt::m_neq, ckr::max(), and ckr::min().

virtual void setSensitivityTolerances	(	doublereal	reltol,
		doublereal	abstol
	)

inlinevirtualinherited

Set the sensitivity error tolerances.

Parameters

reltol	scalar relative tolerance
abstol	scalar absolute tolerance

Definition at line 94 of file Integrator.h.

virtual void initialize	(	doublereal	t0,
		FuncEval &	func
	)

inlinevirtualinherited

Initialize the integrator for a new problem.

Call after all options have been set.

Parameters

t0	initial time
func	RHS evaluator object for system of equations.

Reimplemented in CVodeInt.

Definition at line 111 of file Integrator.h.

Referenced by ImplicitSurfChem::initialize(), and ImplicitSurfChem::integrate().

virtual void integrate ( doublereal tout )

inlinevirtualinherited

Integrate the system of equations.

Parameters

tout	Integrate to this time. Note that this is the absolute time value, not a time interval.

Reimplemented in CVodeInt.

Definition at line 124 of file Integrator.h.

Referenced by ImplicitSurfChem::integrate(), and ImplicitSurfChem::integrate0().

virtual void setMaxOrder ( int n )

inlinevirtualinherited

Set the maximum integration order that will be used.

Reimplemented in CVodeInt.

Definition at line 163 of file Integrator.h.

virtual void setMethod ( MethodType t )

inlinevirtualinherited

Set the solution method.

Reimplemented in CVodeInt.

Definition at line 168 of file Integrator.h.

Referenced by ImplicitSurfChem::ImplicitSurfChem().

virtual void setMaxStepSize ( double hmax )

inlinevirtualinherited

Set the maximum step size.

Reimplemented in CVodeInt.

Definition at line 178 of file Integrator.h.

Referenced by ImplicitSurfChem::integrate().

virtual void setMinStepSize ( double hmin )

inlinevirtualinherited

Set the minimum step size.

Reimplemented in CVodeInt.

Definition at line 183 of file Integrator.h.

Member Data Documentation

IterType m_iter

protected

IterType is used to specify how the nonlinear equations are to be relaxed at each time step.

Definition at line 254 of file BEulerInt.h.

Referenced by BEulerInt::setIterator().

BEulerMethodType m_method

protected

MethodType is used to specify how the time step is to be chosen.

Currently, there are two choices, one is a fixed step method while the other is based on a predictor-corrector algorithm and a time-step truncation error tolerance.

Definition at line 261 of file BEulerInt.h.

Referenced by BEulerInt::step().

int m_jacFormMethod

protected

m_jacFormMethod determines how a matrix is formed.

Definition at line 265 of file BEulerInt.h.

Referenced by BEulerInt::setProblemType().

bool m_rowScaling

protected

m_rowScaling is a boolean.

If true then row sum scaling of the Jacobian matrix is carried out when solving the linear systems.

Definition at line 271 of file BEulerInt.h.

Referenced by BEulerInt::doNewtonSolve(), and BEulerInt::internalMalloc().

bool m_colScaling

protected

m_colScaling is a boolean.

If true, then column scaling is performed on each solution of the linear system.

Definition at line 276 of file BEulerInt.h.

Referenced by BEulerInt::doNewtonSolve(), and BEulerInt::internalMalloc().

bool m_matrixConditioning

protected

m_matrixConditioning is a boolean.

If true, then the Jacobian and every rhs is multiplied by the inverse of a matrix that is suppose to reduce the condition number of the matrix. This is done before row scaling.

Definition at line 283 of file BEulerInt.h.

Referenced by BEulerInt::doNewtonSolve().

int m_itol

protected

If m_itol =1 then each component has an individual value of atol.

If m_itol = 0, the all atols are equal.

Definition at line 288 of file BEulerInt.h.

Referenced by BEulerInt::setTolerances().

double m_reltol

protected

Relative time truncation error tolerances.

Definition at line 292 of file BEulerInt.h.

Referenced by BEulerInt::setTolerances().

double m_abstols

protected

Absolute time truncation error tolerances, when uniform for all variables.

Definition at line 297 of file BEulerInt.h.

Referenced by BEulerInt::setTolerances().

double* m_abstol

protected

Vector of absolute time truncation error tolerance when not uniform for all variables.

Definition at line 302 of file BEulerInt.h.

Referenced by BEulerInt::setTolerances(), and BEulerInt::~BEulerInt().

double* m_ewt

protected

Error Weights.

This is a surprisingly important quantity.

Definition at line 306 of file BEulerInt.h.

Referenced by BEulerInt::boundStep(), BEulerInt::internalMalloc(), BEulerInt::soln_error_norm(), BEulerInt::time_error_norm(), and BEulerInt::~BEulerInt().

double m_hmax

protected

Maximum step size.

Definition at line 309 of file BEulerInt.h.

int m_maxord

protected

Maximum integration order.

Definition at line 313 of file BEulerInt.h.

int m_order

protected

Current integration order.

Definition at line 317 of file BEulerInt.h.

Referenced by BEulerInt::solve_nonlinear_problem(), and BEulerInt::step().

int m_time_step_num

protected

Time step number.

Definition at line 321 of file BEulerInt.h.

Referenced by BEulerInt::doNewtonSolve(), and BEulerInt::step().

int m_max_time_step_attempts

protected

Max time steps allowed.

Definition at line 326 of file BEulerInt.h.

int m_numInitialConstantDeltaTSteps

protected

Number of initial time steps to take where the time truncation error tolerances are not checked.

Instead the delta T is uniform

Definition at line 332 of file BEulerInt.h.

Referenced by BEulerInt::step().

int m_failure_counter

protected

Failure Counter -> keeps track of the number of consequetive failures.

Definition at line 337 of file BEulerInt.h.

Referenced by BEulerInt::step().

int m_min_newt_its

protected

Minimum Number of Newton Iterations per nonlinear step default = 0.

Definition at line 342 of file BEulerInt.h.

Referenced by BEulerInt::solve_nonlinear_problem().

int m_printSolnStepInterval

protected

Step Interval at which to print out the solution default = 1; If set to zero, there is no printout.

Definition at line 351 of file BEulerInt.h.

int m_printSolnNumberToTout

protected

Number of evenly spaced printouts of the solution If zero, there is no printout from this option default 1 If set to zero there is no printout.

Definition at line 358 of file BEulerInt.h.

int m_printSolnFirstSteps

protected

Number of initial steps that the solution is printed out.

default = 0

Definition at line 365 of file BEulerInt.h.

bool m_dumpJacobians

protected

Dump Jacobians to disk default false.

Definition at line 371 of file BEulerInt.h.

int m_neq

protected

Number of equations in the ode integrator.

Definition at line 379 of file BEulerInt.h.

Referenced by BEulerInt::boundStep(), BEulerInt::calc_y_pred(), BEulerInt::calc_ydot(), BEulerInt::doNewtonSolve(), BEulerInt::internalMalloc(), BEulerInt::nEquations(), BEulerInt::setTolerances(), BEulerInt::soln_error_norm(), and BEulerInt::time_error_norm().

double m_t0

protected

Initial time at the start of the integration.

Definition at line 391 of file BEulerInt.h.

double m_time_final

protected

Final time.

Definition at line 395 of file BEulerInt.h.

double delta_t_max

protected

Maximum permissible time step.

Definition at line 409 of file BEulerInt.h.

Referenced by BEulerInt::step().

GeneralMatrix* tdjac_ptr

protected

Pointer to the jacobian representing the time dependent problem.

Definition at line 423 of file BEulerInt.h.

Referenced by BEulerInt::internalMalloc(), BEulerInt::step(), and BEulerInt::~BEulerInt().

int m_print_flag

protected

Determines the level of printing for each time step.

0 -> absolutely nothing is printed for a single time step. 1 -> One line summary per time step 2 -> short description, points of interest 3 -> Lots printed per time step (default)

Definition at line 433 of file BEulerInt.h.

Referenced by BEulerInt::step(), BEulerInt::time_error_norm(), and BEulerInt::time_step_control().

int m_nfe

protected

Number of function evaluations.

Definition at line 440 of file BEulerInt.h.

Referenced by BEulerInt::doNewtonSolve(), and BEulerInt::nEvals().

int m_nJacEval

protected

Number of Jacobian Evaluations and factorization steps (they are the same)

Definition at line 445 of file BEulerInt.h.

int m_numTotalNewtIts