#include <AbstractFunctionalCalculator.hpp>

Inheritance diagram for AbstractFunctionalCalculator< ELEMENT_DIM, SPACE_DIM, PROBLEM_DIM >:

Collaboration diagram for AbstractFunctionalCalculator< ELEMENT_DIM, SPACE_DIM, PROBLEM_DIM >:

Public Member Functions
virtual	~AbstractFunctionalCalculator ()
double	Calculate (AbstractTetrahedralMesh< ELEMENT_DIM, SPACE_DIM > &rMesh, Vec solution)
double	CalculateOnElement (Element< ELEMENT_DIM, SPACE_DIM > &rElement)
Private Member Functions
virtual double	GetIntegrand (ChastePoint< SPACE_DIM > &rX, c_vector< double, PROBLEM_DIM > &rU, c_matrix< double, PROBLEM_DIM, SPACE_DIM > &rGradU)=0
virtual bool	ShouldSkipThisElement (Element< ELEMENT_DIM, SPACE_DIM > &rElement)
Private Attributes
ReplicatableVector	mSolutionReplicated

Detailed Description

template<unsigned ELEMENT_DIM, unsigned SPACE_DIM, unsigned PROBLEM_DIM>
class AbstractFunctionalCalculator< ELEMENT_DIM, SPACE_DIM, PROBLEM_DIM >

This is an abstract class for computing user-defined integral-based functionals of a solution on the mesh. The user needs to define GetIntegrand() in the concrete class, and this class can then be used to compute the integral of f(x,u,grad_u) over the mesh, where x is position, u is the solution at x (possibly with multiple components, for which PROBLEM_DIM>1), grad_u the gradient of u and f the integrand as defined in the concrete class.

Note linear basis functions and 2 quad points per dimension are currently hardcoded.

Definition at line 58 of file AbstractFunctionalCalculator.hpp.

Constructor & Destructor Documentation

template<unsigned ELEMENT_DIM, unsigned SPACE_DIM, unsigned PROBLEM_DIM>

virtual AbstractFunctionalCalculator< ELEMENT_DIM, SPACE_DIM, PROBLEM_DIM >::~AbstractFunctionalCalculator ( ) [inline, virtual]

Destructor.

Definition at line 89 of file AbstractFunctionalCalculator.hpp.

Member Function Documentation

template<unsigned ELEMENT_DIM, unsigned SPACE_DIM, unsigned PROBLEM_DIM>

double AbstractFunctionalCalculator< ELEMENT_DIM, SPACE_DIM, PROBLEM_DIM >::Calculate	(	AbstractTetrahedralMesh< ELEMENT_DIM, SPACE_DIM > &	rMesh,
		Vec	solution
	)

Calculate the integral over the given mesh, using the given solution vector on the mesh.

Note that, in parallel, this method uses a collective reduction step and should therefore always be called collectively.

Parameters:

rMesh	The mesh
solution	The solution vector

Definition at line 181 of file AbstractFunctionalCalculator.hpp.

References AbstractTetrahedralMesh< ELEMENT_DIM, SPACE_DIM >::CalculateDesignatedOwnershipOfElement(), EXCEPTION, AbstractTetrahedralMesh< ELEMENT_DIM, SPACE_DIM >::GetElementIteratorBegin(), AbstractTetrahedralMesh< ELEMENT_DIM, SPACE_DIM >::GetElementIteratorEnd(), AbstractMesh< ELEMENT_DIM, SPACE_DIM >::GetNumNodes(), and PetscTools::ReplicateException().

template<unsigned ELEMENT_DIM, unsigned SPACE_DIM, unsigned PROBLEM_DIM>

double AbstractFunctionalCalculator< ELEMENT_DIM, SPACE_DIM, PROBLEM_DIM >::CalculateOnElement ( Element< ELEMENT_DIM, SPACE_DIM > & rElement )

Compute the contribution to the integral from one element.

Parameters:

rElement The element

NOTE: This assumes that the Jacobian is constant on an element, ie no curvilinear bases were used for position

Definition at line 118 of file AbstractFunctionalCalculator.hpp.

References AbstractTetrahedralElement< ELEMENT_DIM, SPACE_DIM >::CalculateInverseJacobian(), LinearBasisFunction< ELEMENT_DIM >::ComputeBasisFunctions(), LinearBasisFunction< ELEMENT_DIM >::ComputeTransformedBasisFunctionDerivatives(), AbstractElement< ELEMENT_DIM, SPACE_DIM >::GetNode(), AbstractElement< ELEMENT_DIM, SPACE_DIM >::GetNodeGlobalIndex(), AbstractElement< ELEMENT_DIM, SPACE_DIM >::GetNumNodes(), GaussianQuadratureRule< ELEMENT_DIM >::GetNumQuadPoints(), GaussianQuadratureRule< ELEMENT_DIM >::GetWeight(), ChastePoint< DIM >::rGetLocation(), and GaussianQuadratureRule< ELEMENT_DIM >::rGetQuadPoint().

template<unsigned ELEMENT_DIM, unsigned SPACE_DIM, unsigned PROBLEM_DIM>

virtual double AbstractFunctionalCalculator< ELEMENT_DIM, SPACE_DIM, PROBLEM_DIM >::GetIntegrand	(	ChastePoint< SPACE_DIM > &	rX,
		c_vector< double, PROBLEM_DIM > &	rU,
		c_matrix< double, PROBLEM_DIM, SPACE_DIM > &	rGradU
	)		`[private, pure virtual]`

Get the integrand. Must be defined by the user.

Parameters:

rX	The point in space
rU	The unknown as a vector, u(i) = u_i
rGradU	The gradient of the unknown as a matrix, rGradU(i,j) = d(u_i)/d(X_j)

Implemented in PseudoEcgCalculator< ELEMENT_DIM, SPACE_DIM, PROBLEM_DIM >.

template<unsigned ELEMENT_DIM, unsigned SPACE_DIM, unsigned PROBLEM_DIM>

bool AbstractFunctionalCalculator< ELEMENT_DIM, SPACE_DIM, PROBLEM_DIM >::ShouldSkipThisElement ( Element< ELEMENT_DIM, SPACE_DIM > & rElement ) [private, virtual]