gridpack::math::NonlinearSolverInterface< T, I > Class Template Reference

Interface to a solve system of nonlinear equations in parallel. More...

#include <nonlinear_solver_interface.hpp>

Inheritance diagram for gridpack::math::NonlinearSolverInterface< T, I >:

[legend]

List of all members.

Public Types
typedef VectorT< T, I >	VectorType
typedef MatrixT< T, I >	MatrixType
Public Member Functions
	NonlinearSolverInterface ()
	Default constructor.
virtual	~NonlinearSolverInterface (void)
	Destructor.
double	tolerance (void) const
	Get the solution tolerance.
void	tolerance (const double &tol)
	Set the solver tolerance.
int	maximumIterations (void) const
	Get the maximum iterations.
void	maximumIterations (const int &n)
	Set the maximum solution iterations.
void	solve (VectorType &x)
	Solve w/ the specified initial estimated, put result in same vector.

---

Detailed Description

template<typename T, typename I = int>
class gridpack::math::NonlinearSolverInterface< T, I >

Interface to a solve system of nonlinear equations in parallel.

This serves as a base for classes that solve a system of nonlinear equations. While not strictly abstract, it has no function if instantiated on its own.

It encapuslates the nonlinear system solver of some underlying implementation. The Pimpl idiom is used for implementation, so user code is completely independent of the underlying library. This class simply provides an interface to a specific implementation. Subclasses are required to call p_set_impl() at construction to set the implementation.

---

Member Typedef Documentation

template<typename T, typename I = int>

typedef MatrixT<T, I> gridpack::math::NonlinearSolverInterface< T, I >::MatrixType

Reimplemented in gridpack::math::NewtonRaphsonSolverT< T, I >, gridpack::math::NonlinearSolverT< T, I >, and gridpack::math::NonlinearSolverImplementation< T, I >.

template<typename T, typename I = int>

typedef VectorT<T, I> gridpack::math::NonlinearSolverInterface< T, I >::VectorType

Reimplemented in gridpack::math::NewtonRaphsonSolverT< T, I >, gridpack::math::NewtonRaphsonSolverImplementation< T, I >, gridpack::math::NonlinearSolverT< T, I >, and gridpack::math::NonlinearSolverImplementation< T, I >.

---

Constructor & Destructor Documentation

template<typename T, typename I = int>

gridpack::math::NonlinearSolverInterface< T, I >::NonlinearSolverInterface

(

)

Default constructor.

template<typename T, typename I = int>

virtual gridpack::math::NonlinearSolverInterface< T, I >::~NonlinearSolverInterface

(

void

)

[virtual]

Destructor.

---

Member Function Documentation

template<typename T, typename I = int>

void gridpack::math::NonlinearSolverInterface< T, I >::maximumIterations

(

const int &

n

)

Set the maximum solution iterations.

Parameters:

n

new maximum number of iterations

template<typename T, typename I = int>

int gridpack::math::NonlinearSolverInterface< T, I >::maximumIterations

(

void

)

const

Get the maximum iterations.

Returns:

current maximum number of solution iterations

template<typename T, typename I = int>

void gridpack::math::NonlinearSolverInterface< T, I >::solve

(

VectorType &

x

)

Solve w/ the specified initial estimated, put result in same vector.

This solves the system of nonlinear equations using the contents of x as an initial solution estimate. The final result is placed back in x upon completion.

Parameters:

x

solution Vector

template<typename T, typename I = int>

void gridpack::math::NonlinearSolverInterface< T, I >::tolerance

(

const double &

tol

)

Set the solver tolerance.

Parameters:

tol

new solution tolerance

template<typename T, typename I = int>

double gridpack::math::NonlinearSolverInterface< T, I >::tolerance

(

void

)

const

Get the solution tolerance.

Returns:

current solution tolerance

---

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

• nonlinear_solver_interface.hpp