Chombo + EB  3.0
Public Types | Public Member Functions | Private Member Functions | Private Attributes | List of all members
CoefficientInterpolator< LevelData_, SolutionLevelData_ > Class Template Reference

#include <CoefficientInterpolator.H>

Inheritance diagram for CoefficientInterpolator< LevelData_, SolutionLevelData_ >:
Inheritance graph
[legend]

Public Types

typedef LevelData_ LevelDataType
 
typedef SolutionLevelData_ SolutionLevelDataType
 

Public Member Functions

 CoefficientInterpolator (int a_numComps)
 
virtual ~CoefficientInterpolator ()
 Destructor. More...
 
int numComps () const
 Returns the number of components in the interpolated coefficient. More...
 
virtual void interpolate (LevelDataType &a_result, Real a_time)
 
virtual void interpolate (LevelDataType &a_result, const SolutionLevelDataType &a_solution, Real a_time)
 
virtual bool dependsUponSolution () const
 
virtual void interpolatePrime (LevelDataType &a_prime, const SolutionLevelDataType &a_solution, Real a_time)
 
virtual void solve (SolutionLevelDataType &a_phi, const SolutionLevelDataType &a_f, Real a_time, const SolutionLevelDataType &a_phi0, Real a_tolerance)
 
void NewtonRaphson (SolutionLevelDataType &a_phi, const SolutionLevelDataType &a_f, Real a_time, const SolutionLevelDataType &a_phi0, Real a_tolerance)
 

Private Member Functions

 CoefficientInterpolator ()
 
 CoefficientInterpolator (const CoefficientInterpolator &)
 
CoefficientInterpolator & operator= (const CoefficientInterpolator &)
 

Private Attributes

int m_numComps
 
bool m_inCall
 

Detailed Description

template<typename LevelData_, typename SolutionLevelData_ = LevelData_>
class CoefficientInterpolator< LevelData_, SolutionLevelData_ >

This abstract base class provides an interface to time-dependent spatial coefficient data for various partial differential equations. One obtains constant access to coefficient data by specifying the desired time centering. Subclasses of this base class define how that coefficient data is computed.

Template Parameters
LevelDataTypeThe LevelData type used to store coefficient data.

Member Typedef Documentation

◆ LevelDataType

template<typename LevelData_, typename SolutionLevelData_ = LevelData_>
typedef LevelData_ CoefficientInterpolator< LevelData_, SolutionLevelData_ >::LevelDataType

◆ SolutionLevelDataType

template<typename LevelData_, typename SolutionLevelData_ = LevelData_>
typedef SolutionLevelData_ CoefficientInterpolator< LevelData_, SolutionLevelData_ >::SolutionLevelDataType

Constructor & Destructor Documentation

◆ CoefficientInterpolator() [1/3]

template<typename LevelData_ , typename SolutionLevelData_ >
CoefficientInterpolator< LevelData_, SolutionLevelData_ >::CoefficientInterpolator ( int  a_numComps)
explicit

Base class constructor. Called by every subclass.

Parameters
a_numCompsThe number of components in the interpolated coefficient.

◆ ~CoefficientInterpolator()

template<typename LevelData_ , typename SolutionLevelData_ >
CoefficientInterpolator< LevelData_, SolutionLevelData_ >::~CoefficientInterpolator ( )
virtual

Destructor.

Referenced by CoefficientInterpolator< LevelData< FluxBox >, LevelData< FArrayBox > >::CoefficientInterpolator().

◆ CoefficientInterpolator() [2/3]

template<typename LevelData_, typename SolutionLevelData_ = LevelData_>
CoefficientInterpolator< LevelData_, SolutionLevelData_ >::CoefficientInterpolator ( )
private

◆ CoefficientInterpolator() [3/3]

template<typename LevelData_, typename SolutionLevelData_ = LevelData_>
CoefficientInterpolator< LevelData_, SolutionLevelData_ >::CoefficientInterpolator ( const CoefficientInterpolator< LevelData_, SolutionLevelData_ > &  )
private

Member Function Documentation

◆ numComps()

template<typename LevelData_ , typename SolutionLevelData_ >
int CoefficientInterpolator< LevelData_, SolutionLevelData_ >::numComps ( ) const

Returns the number of components in the interpolated coefficient.

Referenced by CoefficientInterpolator< LevelData< FluxBox >, LevelData< FArrayBox > >::~CoefficientInterpolator().

◆ interpolate() [1/2]

template<typename LevelData_ , typename SolutionLevelData_ >
void CoefficientInterpolator< LevelData_, SolutionLevelData_ >::interpolate ( LevelDataType &  a_result,
Real  a_time 
)
virtual

Interpolate the coefficient at the given time, placing the result in the given LevelData object. This method must be overridden by subclasses.

Parameters
a_resultThe LevelData object that will store the result.
a_timeThe time at which the coefficient is to be evaluated.

Referenced by CoefficientInterpolator< LevelData< FluxBox >, LevelData< FArrayBox > >::dependsUponSolution(), CoefficientInterpolator< LevelData< FluxBox >, LevelData< FArrayBox > >::interpolate(), and CoefficientInterpolator< LevelData< FluxBox >, LevelData< FArrayBox > >::NewtonRaphson().

◆ interpolate() [2/2]

template<typename LevelData_ , typename SolutionLevelData_ >
void CoefficientInterpolator< LevelData_, SolutionLevelData_ >::interpolate ( LevelDataType &  a_result,
const SolutionLevelDataType &  a_solution,
Real  a_time 
)
virtual

Interpolate the coefficient at the given time, placing the result in the given LevelData object. This method assumes that the coefficient depends upon the solution to the partial differential equation in question, so the solution is passed into it as an argument.

Parameters
a_resultThe LevelData object that will store the result.
a_solutionThe solution to the partial differential equation.
a_timeThe time at which the coefficient is to be evaluated.

◆ dependsUponSolution()

template<typename LevelData_ , typename SolutionLevelData_ >
bool CoefficientInterpolator< LevelData_, SolutionLevelData_ >::dependsUponSolution ( ) const
virtual

Returns true if the coefficient depends on the solution to the partial differential equation (rendering it nonlinear), false otherwise. By default, the coefficient is assumed not to depend upon the solution.

Referenced by CoefficientInterpolator< LevelData< FluxBox >, LevelData< FArrayBox > >::interpolate(), CoefficientInterpolator< LevelData< FluxBox >, LevelData< FArrayBox > >::interpolatePrime(), and CoefficientInterpolator< LevelData< FluxBox >, LevelData< FArrayBox > >::numComps().

◆ interpolatePrime()

template<typename LevelData_ , typename SolutionLevelData_ >
void CoefficientInterpolator< LevelData_, SolutionLevelData_ >::interpolatePrime ( LevelDataType &  a_prime,
const SolutionLevelDataType &  a_solution,
Real  a_time 
)
virtual

Computes the derivative of the coefficient with respect to the solution at the desired time. By default, this sets a_deriv to 0.

Parameters
a_primeThe coefficient derivative data will be stored here.
a_solutionThe solution to the partial differential equation.
a_timeThe time at which to compute the coefficient data.

Referenced by CoefficientInterpolator< LevelData< FluxBox >, LevelData< FArrayBox > >::interpolate(), and CoefficientInterpolator< LevelData< FluxBox >, LevelData< FArrayBox > >::NewtonRaphson().

◆ solve()

template<typename LevelData_ , typename SolutionLevelData_ >
void CoefficientInterpolator< LevelData_, SolutionLevelData_ >::solve ( SolutionLevelDataType &  a_phi,
const SolutionLevelDataType &  a_f,
Real  a_time,
const SolutionLevelDataType &  a_phi0,
Real  a_tolerance 
)
virtual

This virtual void method performs the iterative nonlinear solve $A(\phi) \phi - f(\vec{x}) = 0$ for $\phi$.

Parameters
a_phiThe solution to the equation, $\phi$, will be stored here.
a_fThe term $f(\vec{x})$ in the equation.
a_timeThe time at which the equation is solved.
a_phi0The initial estimate for $\phi$.
a_toleranceThe threshold for the error in the equation that dictates when iteration should cease.

Referenced by CoefficientInterpolator< LevelData< FluxBox >, LevelData< FArrayBox > >::interpolatePrime().

◆ NewtonRaphson()

template<typename LevelData_ , typename SolutionLevelData_ >
void CoefficientInterpolator< LevelData_, SolutionLevelData_ >::NewtonRaphson ( SolutionLevelDataType &  a_phi,
const SolutionLevelDataType &  a_f,
Real  a_time,
const SolutionLevelDataType &  a_phi0,
Real  a_tolerance 
)

This helper method solves the nonlinear equation $A(\phi) \phi - f(\vec{x}) = 0$ for $\phi$ using Newton-Raphson iteration. It can be used by subclasses to implement the solve() virtual method.

Parameters
a_phiThe solution to the equation, $\phi$, will be stored here.
a_fThe term $f(\vec{x})$ in the equation.
a_timeThe time at which the equation is solved.
a_phi0The initial estimate for $\phi$.
a_toleranceThe threshold for the error in the equation that dictates when iteration should cease.

Referenced by CoefficientInterpolator< LevelData< FluxBox >, LevelData< FArrayBox > >::solve().

◆ operator=()

template<typename LevelData_, typename SolutionLevelData_ = LevelData_>
CoefficientInterpolator& CoefficientInterpolator< LevelData_, SolutionLevelData_ >::operator= ( const CoefficientInterpolator< LevelData_, SolutionLevelData_ > &  )
private

Member Data Documentation

◆ m_numComps

template<typename LevelData_, typename SolutionLevelData_ = LevelData_>
int CoefficientInterpolator< LevelData_, SolutionLevelData_ >::m_numComps
private

Referenced by CoefficientInterpolator< LevelData< FluxBox >, LevelData< FArrayBox > >::numComps().

◆ m_inCall

template<typename LevelData_, typename SolutionLevelData_ = LevelData_>
bool CoefficientInterpolator< LevelData_, SolutionLevelData_ >::m_inCall
private

Referenced by CoefficientInterpolator< LevelData< FluxBox >, LevelData< FArrayBox > >::interpolate().

The documentation for this class was generated from the following file:
Generated by   doxygen 1.8.13