Top-level class for the electromagnetic finite-difference solver. More...

#include <FiniteDifferenceSolver.H>

Public Member Functions
	FiniteDifferenceSolver (ElectromagneticSolverAlgo fdtd_algo, std::array< amrex::Real, 3 > cell_size, ablastr::utils::enums::GridType grid_type)
	Initialize the finite-difference Maxwell solver (for a given refinement level)

void	EvolveB (ablastr::fields::MultiFabRegister &fields, int lev, PatchType patch_type, std::array< std::unique_ptr< amrex::iMultiFab >, 3 > &flag_info_cell, std::array< std::unique_ptr< amrex::LayoutData< FaceInfoBox > >, 3 > &borrowing, amrex::Real dt)
	Update the B field, over one timestep.

void	EvolveE (ablastr::fields::MultiFabRegister &fields, int lev, PatchType patch_type, ablastr::fields::VectorField const &Efield, std::array< std::unique_ptr< amrex::iMultiFab >, 3 > const &eb_update_E, amrex::Real dt)
	Update the E field, over one timestep.

void	EvolveF (amrex::MultiFab Ffield, ablastr::fields::VectorField const &Efield, amrex::MultiFab rhofield, int rho_comp, amrex::Real dt)
	Update the F field, over one timestep.

void	EvolveG (amrex::MultiFab *Gfield, ablastr::fields::VectorField const &Bfield, amrex::Real dt)

void	EvolveECTRho (ablastr::fields::VectorField const &Efield, ablastr::fields::VectorField const &edge_lengths, ablastr::fields::VectorField const &face_areas, ablastr::fields::VectorField const &ECTRhofield, int lev)
	Update the B field, over one timestep.

void	ApplySilverMuellerBoundary (ablastr::fields::VectorField &Efield, ablastr::fields::VectorField &Bfield, amrex::Box domain_box, amrex::Real dt, amrex::Array< FieldBoundaryType, 3 > field_boundary_lo, amrex::Array< FieldBoundaryType, 3 > field_boundary_hi)
	Update the B field at the boundary, using the Silver-Mueller condition.

void	ComputeDivE (ablastr::fields::VectorField const &Efield, amrex::MultiFab &divE)
	Update the F field, over one timestep.

void	MacroscopicEvolveE (MacroscopicSolverAlgo macroscopic_solver_algo, ablastr::fields::VectorField const &Efield, ablastr::fields::VectorField const &Bfield, ablastr::fields::VectorField const &Jfield, std::array< std::unique_ptr< amrex::iMultiFab >, 3 > const &eb_update_E, amrex::Real dt, std::unique_ptr< MacroscopicProperties > const &macroscopic_properties)
	Macroscopic E-update for non-vacuum medium using the user-selected finite-difference algorithm and macroscopic sigma-method defined in WarpXAlgorithmSelection.H.

void	EvolveBPML (ablastr::fields::MultiFabRegister &fields, PatchType patch_type, int level, amrex::Real dt, bool dive_cleaning)
	Update the B field, over one timestep.

void	EvolveEPML (ablastr::fields::MultiFabRegister &fields, PatchType patch_type, int level, MultiSigmaBox const &sigba, amrex::Real dt, bool pml_has_particles)
	Update the E field, over one timestep.

void	EvolveFPML (amrex::MultiFab *Ffield, ablastr::fields::VectorField Efield, amrex::Real dt)
	Update the E field, over one timestep.

void	HybridPICSolveE (ablastr::fields::VectorField const &Efield, ablastr::fields::VectorField &Jfield, ablastr::fields::VectorField const &Jifield, ablastr::fields::VectorField const &Bfield, amrex::MultiFab const &rhofield, amrex::MultiFab const &Pefield, std::array< std::unique_ptr< amrex::iMultiFab >, 3 > const &eb_update_E, int lev, HybridPICModel const *hybrid_model, bool solve_for_Faraday)
	E-update in the hybrid PIC algorithm as described in Winske et al. (2003) Eq. 10. https://link.springer.com/chapter/10.1007/3-540-36530-3_8.

void	CalculateCurrentAmpere (ablastr::fields::VectorField &Jfield, ablastr::fields::VectorField const &Bfield, std::array< std::unique_ptr< amrex::iMultiFab >, 3 > const &eb_update_E, int lev)
	Calculation of total current using Ampere's law (without displacement current): J = (curl x B) / mu0.

void	ComputeCurlA (ablastr::fields::VectorField &Bfield, ablastr::fields::VectorField const &Afield, std::array< std::unique_ptr< amrex::iMultiFab >, 3 > const &eb_update_B, int lev)
	Calculation of B field from the vector potential A B = (curl x A) / mu0.

template<typename T_Algo>
void	EvolveBCylindrical (ablastr::fields::VectorField const &Bfield, ablastr::fields::VectorField const &Efield, int lev, amrex::Real dt)

template<typename T_Algo>
void	EvolveECylindrical (ablastr::fields::VectorField const &Efield, ablastr::fields::VectorField const &Bfield, ablastr::fields::VectorField const &Jfield, std::array< std::unique_ptr< amrex::iMultiFab >, 3 > const &eb_update_E, amrex::MultiFab const *Ffield, int lev, amrex::Real dt)

template<typename T_Algo>
void	EvolveFCylindrical (amrex::MultiFab Ffield, ablastr::fields::VectorField const &Efield, amrex::MultiFab rhofield, int rho_comp, amrex::Real dt)

template<typename T_Algo>
void	ComputeDivECylindrical (ablastr::fields::VectorField const &Efield, amrex::MultiFab &divE)

template<typename T_Algo>
void	HybridPICSolveECylindrical (ablastr::fields::VectorField const &Efield, ablastr::fields::VectorField const &Jfield, ablastr::fields::VectorField const &Jifield, ablastr::fields::VectorField const &Bfield, amrex::MultiFab const &rhofield, amrex::MultiFab const &Pefield, std::array< std::unique_ptr< amrex::iMultiFab >, 3 > const &eb_update_E, int lev, HybridPICModel const *hybrid_model, bool solve_for_Faraday)

template<typename T_Algo>
void	CalculateCurrentAmpereCylindrical (ablastr::fields::VectorField &Jfield, ablastr::fields::VectorField const &Bfield, std::array< std::unique_ptr< amrex::iMultiFab >, 3 > const &eb_update_E, int lev)
	Calculate total current from Ampere's law without displacement current i.e. J = 1/mu_0 curl x B.

template<typename T_Algo>
void	ComputeCurlACylindrical (ablastr::fields::VectorField &Bfield, ablastr::fields::VectorField const &Afield, std::array< std::unique_ptr< amrex::iMultiFab >, 3 > const &eb_update_B, int lev)
	Calculate B from the curl of A i.e. B = curl(A) output field on B field mesh staggering.

Private Attributes
ElectromagneticSolverAlgo	m_fdtd_algo

ablastr::utils::enums::GridType	m_grid_type

amrex::Real	m_dr

amrex::Real	m_rmin

int	m_nmodes

amrex::Vector< amrex::Real >	m_h_stencil_coefs_r

amrex::Vector< amrex::Real >	m_h_stencil_coefs_z

amrex::Gpu::DeviceVector< amrex::Real >	m_stencil_coefs_r

amrex::Gpu::DeviceVector< amrex::Real >	m_stencil_coefs_z

Detailed Description

Top-level class for the electromagnetic finite-difference solver.

Stores the coefficients of the finite-difference stencils, and has member functions to update fields over one time step.

Constructor & Destructor Documentation

◆ FiniteDifferenceSolver()

FiniteDifferenceSolver::FiniteDifferenceSolver	(	ElectromagneticSolverAlgo	fdtd_algo,
		std::array< amrex::Real, 3 >	cell_size,
		ablastr::utils::enums::GridType	grid_type )

Initialize the finite-difference Maxwell solver (for a given refinement level)

This function initializes the stencil coefficients for the chosen finite-difference algorithm

Parameters

fdtd_algo	Identifies the chosen algorithm, as defined in WarpXAlgorithmSelection.H
cell_size	Cell size along each dimension, for the chosen refinement level
grid_type	Whether the solver is applied to a collocated or staggered grid

Member Function Documentation

◆ ApplySilverMuellerBoundary()

void FiniteDifferenceSolver::ApplySilverMuellerBoundary	(	ablastr::fields::VectorField &	Efield,
		ablastr::fields::VectorField &	Bfield,
		amrex::Box	domain_box,
		amrex::Real	dt,
		amrex::Array< FieldBoundaryType, 3 >	field_boundary_lo,
		amrex::Array< FieldBoundaryType, 3 >	field_boundary_hi )

Update the B field at the boundary, using the Silver-Mueller condition.

◆ CalculateCurrentAmpere()

void FiniteDifferenceSolver::CalculateCurrentAmpere	(	ablastr::fields::VectorField &	Jfield,
		ablastr::fields::VectorField const &	Bfield,
		std::array< std::unique_ptr< amrex::iMultiFab >, 3 > const &	eb_update_E,
		int	lev )

Calculation of total current using Ampere's law (without displacement current): J = (curl x B) / mu0.

Parameters

[out]	Jfield	vector of current MultiFabs at a given level
[in]	Bfield	vector of magnetic field MultiFabs at a given level
[in]	eb_update_E	indicate in which cell E should be updated (related to embedded boundaries)
[in]	lev	level number for the calculation

◆ CalculateCurrentAmpereCylindrical()

template<typename T_Algo>

void FiniteDifferenceSolver::CalculateCurrentAmpereCylindrical	(	ablastr::fields::VectorField &	Jfield,
		ablastr::fields::VectorField const &	Bfield,
		std::array< std::unique_ptr< amrex::iMultiFab >, 3 > const &	eb_update_E,
		int	lev )

Calculate total current from Ampere's law without displacement current i.e. J = 1/mu_0 curl x B.

Parameters

[out]	Jfield	vector of total current MultiFabs at a given level
[in]	Bfield	vector of magnetic field MultiFabs at a given level
[in]	eb_update_E	specifies where the plasma current should be calculated.
[in]	lev	refinement level

◆ ComputeCurlA()

void FiniteDifferenceSolver::ComputeCurlA	(	ablastr::fields::VectorField &	Bfield,
		ablastr::fields::VectorField const &	Afield,
		std::array< std::unique_ptr< amrex::iMultiFab >, 3 > const &	eb_update_B,
		int	lev )

Calculation of B field from the vector potential A B = (curl x A) / mu0.

Parameters

[out]	Bfield	vector of current MultiFabs at a given level
[in]	Afield	vector of magnetic field MultiFabs at a given level
[in]	eb_update_B	array indicating where the field should be updated with respect to the position of the embedded boundary
[in]	lev	level number for the calculation

◆ ComputeCurlACylindrical()

template<typename T_Algo>

void FiniteDifferenceSolver::ComputeCurlACylindrical	(	ablastr::fields::VectorField &	Bfield,
		ablastr::fields::VectorField const &	Afield,
		std::array< std::unique_ptr< amrex::iMultiFab >, 3 > const &	eb_update_B,
		int	lev )

Calculate B from the curl of A i.e. B = curl(A) output field on B field mesh staggering.

Parameters

[out]	Bfield	output of curl operation
[in]	Afield	input staggered field, should be on E/J/A mesh staggering
[in]	eb_update_B	specifies where the plasma current should be calculated.
[in]	lev	refinement level

◆ ComputeDivE()

void FiniteDifferenceSolver::ComputeDivE	(	ablastr::fields::VectorField const &	Efield,
		amrex::MultiFab &	divE )

Update the F field, over one timestep.

◆ ComputeDivECylindrical()

template<typename T_Algo>

void FiniteDifferenceSolver::ComputeDivECylindrical	(	ablastr::fields::VectorField const &	Efield,
		amrex::MultiFab &	divE )

◆ EvolveB()

void FiniteDifferenceSolver::EvolveB	(	ablastr::fields::MultiFabRegister &	fields,
		int	lev,
		PatchType	patch_type,
		std::array< std::unique_ptr< amrex::iMultiFab >, 3 > &	flag_info_cell,
		std::array< std::unique_ptr< amrex::LayoutData< FaceInfoBox > >, 3 > &	borrowing,
		amrex::Real	dt )

Update the B field, over one timestep.

◆ EvolveBCylindrical()

template<typename T_Algo>

void FiniteDifferenceSolver::EvolveBCylindrical	(	ablastr::fields::VectorField const &	Bfield,
		ablastr::fields::VectorField const &	Efield,
		int	lev,
		amrex::Real	dt )

◆ EvolveBPML()

void FiniteDifferenceSolver::EvolveBPML	(	ablastr::fields::MultiFabRegister &	fields,
		PatchType	patch_type,
		int	level,
		amrex::Real	dt,
		bool	dive_cleaning )

Update the B field, over one timestep.

◆ EvolveE()

void FiniteDifferenceSolver::EvolveE	(	ablastr::fields::MultiFabRegister &	fields,
		int	lev,
		PatchType	patch_type,
		ablastr::fields::VectorField const &	Efield,
		std::array< std::unique_ptr< amrex::iMultiFab >, 3 > const &	eb_update_E,
		amrex::Real	dt )

Update the E field, over one timestep.

◆ EvolveECTRho()

void FiniteDifferenceSolver::EvolveECTRho	(	ablastr::fields::VectorField const &	Efield,
		ablastr::fields::VectorField const &	edge_lengths,
		ablastr::fields::VectorField const &	face_areas,
		ablastr::fields::VectorField const &	ECTRhofield,
		int	lev )

Update the B field, over one timestep.

◆ EvolveECylindrical()

template<typename T_Algo>

void FiniteDifferenceSolver::EvolveECylindrical	(	ablastr::fields::VectorField const &	Efield,
		ablastr::fields::VectorField const &	Bfield,
		ablastr::fields::VectorField const &	Jfield,
		std::array< std::unique_ptr< amrex::iMultiFab >, 3 > const &	eb_update_E,
		amrex::MultiFab const *	Ffield,
		int	lev,
		amrex::Real	dt )

◆ EvolveEPML()

void FiniteDifferenceSolver::EvolveEPML	(	ablastr::fields::MultiFabRegister &	fields,
		PatchType	patch_type,
		int	level,
		MultiSigmaBox const &	sigba,
		amrex::Real	dt,
		bool	pml_has_particles )

Update the E field, over one timestep.

◆ EvolveF()

void FiniteDifferenceSolver::EvolveF	(	amrex::MultiFab *	Ffield,
		ablastr::fields::VectorField const &	Efield,
		amrex::MultiFab *	rhofield,
		int	rho_comp,
		amrex::Real	dt )

Update the F field, over one timestep.

◆ EvolveFCylindrical()

template<typename T_Algo>

void FiniteDifferenceSolver::EvolveFCylindrical	(	amrex::MultiFab *	Ffield,
		ablastr::fields::VectorField const &	Efield,
		amrex::MultiFab *	rhofield,
		int	rho_comp,
		amrex::Real	dt )

◆ EvolveFPML()

void FiniteDifferenceSolver::EvolveFPML	(	amrex::MultiFab *	Ffield,
		ablastr::fields::VectorField	Efield,
		amrex::Real	dt )

Update the E field, over one timestep.

◆ EvolveG()

void FiniteDifferenceSolver::EvolveG	(	amrex::MultiFab *	Gfield,
		ablastr::fields::VectorField const &	Bfield,
		amrex::Real	dt )

◆ HybridPICSolveE()

void FiniteDifferenceSolver::HybridPICSolveE	(	ablastr::fields::VectorField const &	Efield,
		ablastr::fields::VectorField &	Jfield,
		ablastr::fields::VectorField const &	Jifield,
		ablastr::fields::VectorField const &	Bfield,
		amrex::MultiFab const &	rhofield,
		amrex::MultiFab const &	Pefield,
		std::array< std::unique_ptr< amrex::iMultiFab >, 3 > const &	eb_update_E,
		int	lev,
		HybridPICModel const *	hybrid_model,
		bool	solve_for_Faraday )

E-update in the hybrid PIC algorithm as described in Winske et al. (2003) Eq. 10. https://link.springer.com/chapter/10.1007/3-540-36530-3_8.

Parameters

[out]	Efield	vector of electric field MultiFabs updated at a given level
[in]	Jfield	vector of total plasma current MultiFabs at a given level
[in]	Jifield	vector of ion current density MultiFabs at a given level
[in]	Bfield	vector of magnetic field MultiFabs at a given level
[in]	rhofield	scalar ion charge density Multifab at a given level
[in]	Pefield	scalar electron pressure MultiFab at a given level
[in]	eb_update_E	indicate in which cell E should be updated (related to embedded boundaries)
[in]	lev	level number for the calculation
[in]	hybrid_model	instance of the hybrid-PIC model
[in]	solve_for_Faraday	boolean flag for whether the E-field is solved to be used in Faraday's equation

◆ HybridPICSolveECylindrical()

template<typename T_Algo>

void FiniteDifferenceSolver::HybridPICSolveECylindrical	(	ablastr::fields::VectorField const &	Efield,
		ablastr::fields::VectorField const &	Jfield,
		ablastr::fields::VectorField const &	Jifield,
		ablastr::fields::VectorField const &	Bfield,
		amrex::MultiFab const &	rhofield,
		amrex::MultiFab const &	Pefield,
		std::array< std::unique_ptr< amrex::iMultiFab >, 3 > const &	eb_update_E,
		int	lev,
		HybridPICModel const *	hybrid_model,
		bool	solve_for_Faraday )

◆ MacroscopicEvolveE()

void FiniteDifferenceSolver::MacroscopicEvolveE	(	MacroscopicSolverAlgo	macroscopic_solver_algo,
		ablastr::fields::VectorField const &	Efield,
		ablastr::fields::VectorField const &	Bfield,
		ablastr::fields::VectorField const &	Jfield,
		std::array< std::unique_ptr< amrex::iMultiFab >, 3 > const &	eb_update_E,
		amrex::Real	dt,
		std::unique_ptr< MacroscopicProperties > const &	macroscopic_properties )

Macroscopic E-update for non-vacuum medium using the user-selected finite-difference algorithm and macroscopic sigma-method defined in WarpXAlgorithmSelection.H.

Parameters

[in]	macroscopic_solver_algo	macroscopic Maxwell solver algorithm (BackwardEuler or Lax-Wendroff)
[out]	Efield	vector of electric field MultiFabs updated at a given level
[in]	Bfield	vector of magnetic field MultiFabs at a given level
[in]	Jfield	vector of current density MultiFabs at a given level
[in]	eb_update_E	indicate in which cell E should be updated (related to embedded boundaries)
[in]	dt	timestep of the simulation
[in]	macroscopic_properties	contains user-defined properties of the medium.

Member Data Documentation

◆ m_dr

amrex::Real FiniteDifferenceSolver::m_dr

private

◆ m_fdtd_algo

ElectromagneticSolverAlgo FiniteDifferenceSolver::m_fdtd_algo

private

◆ m_grid_type

ablastr::utils::enums::GridType FiniteDifferenceSolver::m_grid_type

private

◆ m_h_stencil_coefs_r

amrex::Vector<amrex::Real> FiniteDifferenceSolver::m_h_stencil_coefs_r

private

◆ m_h_stencil_coefs_z

amrex::Vector<amrex::Real> FiniteDifferenceSolver::m_h_stencil_coefs_z

private

◆ m_nmodes

int FiniteDifferenceSolver::m_nmodes

private

◆ m_rmin

amrex::Real FiniteDifferenceSolver::m_rmin

private

◆ m_stencil_coefs_r

amrex::Gpu::DeviceVector<amrex::Real> FiniteDifferenceSolver::m_stencil_coefs_r

private

◆ m_stencil_coefs_z

amrex::Gpu::DeviceVector<amrex::Real> FiniteDifferenceSolver::m_stencil_coefs_z

private

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

/home/docs/checkouts/readthedocs.org/user_builds/warpx/checkouts/6102/Source/FieldSolver/FiniteDifferenceSolver/FiniteDifferenceSolver.H
/home/docs/checkouts/readthedocs.org/user_builds/warpx/checkouts/6102/Source/FieldSolver/FiniteDifferenceSolver/ApplySilverMuellerBoundary.cpp
/home/docs/checkouts/readthedocs.org/user_builds/warpx/checkouts/6102/Source/FieldSolver/FiniteDifferenceSolver/ComputeCurlA.cpp
/home/docs/checkouts/readthedocs.org/user_builds/warpx/checkouts/6102/Source/FieldSolver/FiniteDifferenceSolver/ComputeDivE.cpp
/home/docs/checkouts/readthedocs.org/user_builds/warpx/checkouts/6102/Source/FieldSolver/FiniteDifferenceSolver/EvolveB.cpp
/home/docs/checkouts/readthedocs.org/user_builds/warpx/checkouts/6102/Source/FieldSolver/FiniteDifferenceSolver/EvolveBPML.cpp
/home/docs/checkouts/readthedocs.org/user_builds/warpx/checkouts/6102/Source/FieldSolver/FiniteDifferenceSolver/EvolveE.cpp
/home/docs/checkouts/readthedocs.org/user_builds/warpx/checkouts/6102/Source/FieldSolver/FiniteDifferenceSolver/EvolveECTRho.cpp
/home/docs/checkouts/readthedocs.org/user_builds/warpx/checkouts/6102/Source/FieldSolver/FiniteDifferenceSolver/EvolveEPML.cpp
/home/docs/checkouts/readthedocs.org/user_builds/warpx/checkouts/6102/Source/FieldSolver/FiniteDifferenceSolver/EvolveF.cpp
/home/docs/checkouts/readthedocs.org/user_builds/warpx/checkouts/6102/Source/FieldSolver/FiniteDifferenceSolver/EvolveFPML.cpp
/home/docs/checkouts/readthedocs.org/user_builds/warpx/checkouts/6102/Source/FieldSolver/FiniteDifferenceSolver/EvolveG.cpp
/home/docs/checkouts/readthedocs.org/user_builds/warpx/checkouts/6102/Source/FieldSolver/FiniteDifferenceSolver/FiniteDifferenceSolver.cpp
/home/docs/checkouts/readthedocs.org/user_builds/warpx/checkouts/6102/Source/FieldSolver/FiniteDifferenceSolver/HybridPICSolveE.cpp
/home/docs/checkouts/readthedocs.org/user_builds/warpx/checkouts/6102/Source/FieldSolver/FiniteDifferenceSolver/MacroscopicEvolveE.cpp

Public Member Functions

Private Attributes

Detailed Description

Constructor & Destructor Documentation

◆ FiniteDifferenceSolver()

Member Function Documentation

◆ ApplySilverMuellerBoundary()

◆ CalculateCurrentAmpere()

◆ CalculateCurrentAmpereCylindrical()

◆ ComputeCurlA()

◆ ComputeCurlACylindrical()

◆ ComputeDivE()

◆ ComputeDivECylindrical()

◆ EvolveB()

◆ EvolveBCylindrical()

◆ EvolveBPML()

◆ EvolveE()

◆ EvolveECTRho()

◆ EvolveECylindrical()

◆ EvolveEPML()

◆ EvolveF()

◆ EvolveFCylindrical()

◆ EvolveFPML()

◆ EvolveG()

◆ HybridPICSolveE()

◆ HybridPICSolveECylindrical()

◆ MacroscopicEvolveE()

Member Data Documentation

◆ m_dr

◆ m_fdtd_algo

◆ m_grid_type

◆ m_h_stencil_coefs_r

◆ m_h_stencil_coefs_z

◆ m_nmodes

◆ m_rmin

◆ m_stencil_coefs_r

◆ m_stencil_coefs_z