NumLib::EquationSystem Class Reference

Detailed Description

Collection of basic methods every equation system must provide.

Definition at line 30 of file EquationSystem.h.

#include <EquationSystem.h>

Inheritance diagram for NumLib::EquationSystem:

Public Member Functions
virtual bool	isLinear () const =0
virtual bool	requiresNormalization () const =0
virtual void	preIteration (const unsigned iter, GlobalVector const &x)
virtual IterationResult	postIteration (GlobalVector const &x)
virtual MathLib::MatrixSpecifications	getMatrixSpecifications (const int process_id) const =0
virtual	~EquationSystem ()=default

Constructor & Destructor Documentation

~EquationSystem()

virtual NumLib::EquationSystem::~EquationSystem ( )

virtualdefault

Member Function Documentation

getMatrixSpecifications()

virtual MathLib::MatrixSpecifications NumLib::EquationSystem::getMatrixSpecifications ( const int process_id ) const

pure virtual

isLinear()

virtual bool NumLib::EquationSystem::isLinear ( ) const

pure virtual

Check whether this is actually a linear equation system.

Remarks

Depending on its parameters an in general nonlinear equation system can be linear in special cases. With this method it is possible to detect that at runtime and thus save some computations.

postIteration()

virtual IterationResult NumLib::EquationSystem::postIteration ( GlobalVector const & x )

inlinevirtual

Post-processes an iteration in the solution process of this equation.

Parameters

x	the current approximate solution of the equation.

Returns

A status flag indicating id the current iteration succeeded.

Definition at line 68 of file EquationSystem.h.

    {
        (void)x;  // by default do nothing
        return IterationResult::SUCCESS;
    }

References NumLib::SUCCESS.

preIteration()

virtual void NumLib::EquationSystem::preIteration ( const unsigned iter, GlobalVector const & x )

inlinevirtual

Prepares a new iteration in the solution process of this equation.

Parameters

iter	the current iteration number, starting from 1.
x	the current approximate solution of the equation.

Definition at line 56 of file EquationSystem.h.

    {
        (void)iter;
        (void)x;  // by default do nothing
    }

requiresNormalization()

virtual bool NumLib::EquationSystem::requiresNormalization ( ) const

pure virtual

Check whether normalization of A and rhs is required.

Remarks

In some processes, a normalization operation is required, to calculate A^T * A, and overwrite A; also calculate A^T * rhs and overwrite rhs. This parameter reflect whether such operation is required.

---

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

• NumLib/ODESolver/EquationSystem.h