ippl::EvalFunctor< T, Dim, numElementDOFs > Struct Template Reference

Representation of the lhs of the problem we are trying to solve. More...

#include <FEMMaxwellDiffusionSolver.h>

Collaboration diagram for ippl::EvalFunctor< T, Dim, numElementDOFs >:

Public Member Functions
	EvalFunctor (Vector< T, Dim > DPhiInvT, T absDetDPhi)
	Constructor.
KOKKOS_FUNCTION auto	operator() (size_t i, size_t j, const QuadratureData< Vector< T, Dim >, Vector< T, Dim >, numElementDOFs > &qd) const
	Returns the evaluation of (curl(b_i)*curl(b_j) + b_i*b_j)*absDetDPhi.
	EvalFunctor (Vector< Tlhs, Dim > DPhiInvT, Tlhs absDetDPhi)
KOKKOS_FUNCTION auto	operator() (const size_t &i, const size_t &j, const QuadratureData< Tlhs, Vector< Tlhs, Dim >, numElemDOFs > &qd) const

Public Attributes
const Vector< T, Dim >	DPhiInvT
	The inverse transpose Jacobian.
const T	absDetDPhi
	The determinant of the Jacobian.
const Vector< Tlhs, Dim >	DPhiInvT
const Tlhs	absDetDPhi

Detailed Description

template<typename T, unsigned Dim, unsigned numElementDOFs>
struct ippl::EvalFunctor< T, Dim, numElementDOFs >

Representation of the lhs of the problem we are trying to solve.

In our case this corresponds to the variational formulation of the curl(curl(E)) + E and is curl(b_i)*curl(b_j) + b_i*b_j.

Template Parameters

T	The type we are working with.
Dim	the dimension of the space.
numElementDOFs	the number of DOFs per element that we have.

Member Function Documentation

operator()()

template<typename T , unsigned Dim, unsigned numElementDOFs>

KOKKOS_FUNCTION auto ippl::EvalFunctor< T, Dim, numElementDOFs >::operator() ( size_t  i, size_t  j, const QuadratureData< Vector< T, Dim >, Vector< T, Dim >, numElementDOFs > &  qd  ) const

inline

Returns the evaluation of (curl(b_i)*curl(b_j) + b_i*b_j)*absDetDPhi.

This function takes as input the basis function values and their curl for the different DOFs and returns the evaluation of the inner part of the integral of the variational formuation, which corresponds to (curl(b_i)*curl(b_j) + b_i*b_j), but note that we additionally also multiply this with absDetDPhi, which is required by the quadrature rule. In theroy this could also be done outside of this.

Parameters

i	The first DOF index.
j	The second DOF index.
qd	Per-quadrature basis values and curls.

Returns

(curl(b_i)*curl(b_j) + b_i*b_j)*absDetDPhi

Member Data Documentation

absDetDPhi

template<typename T , unsigned Dim, unsigned numElementDOFs>

const T ippl::EvalFunctor< T, Dim, numElementDOFs >::absDetDPhi

The determinant of the Jacobian.

As we have a unirectangular grid it is the same for all the differnt Elements and we therefore have to store it only once.

DPhiInvT

template<typename T , unsigned Dim, unsigned numElementDOFs>

const Vector<T, Dim> ippl::EvalFunctor< T, Dim, numElementDOFs >::DPhiInvT

The inverse transpose Jacobian.

As we have a unirectangular grid it is the same for all the differnt Elements and we therefore have to store it only once.

---

The documentation for this struct was generated from the following files:

• ippl-src/src/MaxwellSolvers/FEMMaxwellDiffusionSolver.h
• ippl-src/src/PoissonSolvers/EvalFunctor.h