Class for evolving plasma equations using a split step Godunov method. More...

#include <CD_CdrPlasmaGodunovStepper.H>

Inheritance diagram for Physics::CdrPlasma::CdrPlasmaGodunovStepper:

[legend]

Collaboration diagram for Physics::CdrPlasma::CdrPlasmaGodunovStepper:

[legend]

Classes
class	CdrStorage
	Utility class for memory handling in CdrPlasmaStepper. More...

class	FieldStorage
	Utility class for memory handling in CdrPlasmaStepper. More...

class	RtStorage
	Utility class for memory handling in CdrPlasmaStepper. More...

class	SigmaStorage
	Utility class for memory handling in CdrPlasmaStepper. More...

Public Member Functions
	CdrPlasmaGodunovStepper ()=delete
	Disallowed constructor – use strong construction.

	CdrPlasmaGodunovStepper (RefCountedPtr< CdrPlasmaPhysics > &a_physics)
	Full constructor.

virtual	~CdrPlasmaGodunovStepper ()
	Destructor.

void	parseOptions () override
	Parse startup options.

void	parseRuntimeOptions () override
	Parse run-time adjustable parameters.

Real	advance (const Real a_dt) override
	Implementation of the advance method.

void	preRegrid (const int a_lbase, const int a_oldFinestLevel) override
	Pre-regrid method. Used for storing important variables for the solvers.

void	regrid (const int a_lmin, const int a_oldFinestLevel, const int a_newFinestLevel) override
	Regrid method.

void	postRegrid () override
	Perform post-regrid operations.

Real	computeDt () override
	Compute the time step – this will be different for the different supported algorithms.

void	postCheckpointSetup () override
	Run post-checkpoint setup operations.

Public Member Functions inherited from Physics::CdrPlasma::CdrPlasmaStepper
	CdrPlasmaStepper ()
	Empty constructor.

	CdrPlasmaStepper (RefCountedPtr< CdrPlasmaPhysics > &a_physics)
	Full constructor.

	~CdrPlasmaStepper ()
	Destructor.

virtual void	setupSolvers () override
	Instantiate solvers.

virtual void	allocate () override
	Allocate data for the time stepper and solvers.

virtual void	initialData () override
	Fill all solvers with initial data.

virtual void	postInitialize () override
	Post-initialize operations to be performed after filling solvers with initialData.

virtual void	registerRealms () override
	Register realms to be used for the simulation.

virtual void	registerOperators () override
	Register operators to be used for the simulation.

virtual void	prePlot () override
	Compute some thing that go into plot files.

virtual int	getNumberOfPlotVariables () const override
	Get the number of plot variables for this time stepper.

virtual Vector< std::string >	getPlotVariableNames () const override
	Get plot variable names.

virtual void	writePlotData (LevelData< EBCellFAB > &a_output, int &a_icomp, const std::string a_outputRealm, const int a_level) const override
	Write plot data to output holder.

virtual void	synchronizeSolverTimes (const int a_step, const Real a_time, const Real a_dt) override
	Synchronize solver times.

virtual void	printStepReport () override
	Print a step report.

virtual void	preRegridInternals (const int a_lmin, const int a_oldFinestLevel)
	Do a preRegrid operation for internal storage (if needed).

virtual void	computeJ (EBAMRCellData &a_J) const
	Compute the current density.

virtual void	computeElectricField (MFAMRCellData &a_electricFieldCell, const MFAMRCellData &a_potential) const
	Compute the cell-centered electric field on both phases.

virtual void	computeElectricField (EBAMRCellData &a_electricFieldCell, const phase::which_phase a_phase) const
	Compute the cell-centered electric field on a specific phase using whatever is available in m_fieldSolver.

virtual void	computeElectricField (EBAMRCellData &a_E, const phase::which_phase a_phase, const MFAMRCellData &a_potential) const
	Compute the cell-centered electric field on a specific phase.

virtual void	computeElectricField (EBAMRFluxData &a_electricFieldFace, const phase::which_phase a_phase, const EBAMRCellData &a_electricFieldCell) const
	Compute the face-centered electric field. This will only do the normal component.

virtual void	computeElectricField (EBAMRIVData &a_electricFieldEB, const phase::which_phase a_phase, const EBAMRCellData &a_electricFieldCell) const
	Compute the EB-centered electric field.

virtual void	computeMaxElectricField (Real &a_maxElectricField, const phase::which_phase a_phase)
	Compute the maximum of the electric field.

virtual Real	getTime () const
	Return time.

virtual bool	stationaryRTE ()
	Check if RTE solvers are stationary.

virtual void	deallocate ()
	Deallocation function. This will deallocate internal storage in the subclasses as well as the solvers.

virtual void	computeSpaceChargeDensity ()
	Compute the centroid-centered space charge density by using the data inside the CDR solvers.

virtual void	computeSpaceChargeDensity (MFAMRCellData &a_rho, const Vector< EBAMRCellData * > &a_cdrDensities)
	Compute the centroid-centered space charge density.

virtual void	computeCellConductivity (EBAMRCellData &a_cellConductivity) const
	Compute the cell-centered conductivity.

virtual void	computeFaceConductivity (EBAMRFluxData &a_conductivityFace, EBAMRIVData &a_conductivityEB, const EBAMRCellData &a_conductivityCell) const
	Compute the face-centered conductivity from a cell-centered conductivity.

virtual void	setupSemiImplicitPoisson (const Real a_dt)
	Set up a semi-implicit Poisson solver.

virtual void	setupSemiImplicitPoisson (const EBAMRFluxData &a_conductivityFace, const EBAMRIVData &a_conductivityEB, const Real a_factor)
	Set up a semi-implicit Poisson solver.

virtual bool	solvePoisson ()
	Solve the electrostatic Poisson equation.

virtual bool	solvePoisson (MFAMRCellData &a_potential, MFAMRCellData &a_rho, const Vector< EBAMRCellData * > a_cdrDensities, const EBAMRIVData &a_sigma)
	General electrostatic Poisson solver routine.

virtual void	advanceReactionNetwork (const Real a_time, const Real a_dt)
	Advance the reaction network. This will compute the electric field (on the appropriate phase) and call the other version.

virtual void	advanceReactionNetwork (Vector< EBAMRCellData * > &a_cdrSources, Vector< EBAMRCellData * > &a_rteSources, const Vector< EBAMRCellData * > &a_cdrDensities, const Vector< EBAMRCellData * > &a_rteDensities, const EBAMRCellData &a_E, const Real &a_time, const Real &a_dt)
	Compute reaction network source terms.

virtual void	advanceReactionNetwork (Vector< EBAMRCellData * > &a_cdrSources, Vector< EBAMRCellData * > &a_rteSources, const Vector< EBAMRCellData * > &a_cdrDensities, const Vector< EBAMRCellData * > &a_cdrGradients, const Vector< EBAMRCellData * > &a_rteDensities, const EBAMRCellData &a_E, const Real &a_time, const Real &a_dt)
	Compute reaction network source terms. This is the AMR version that does all levels.

virtual void	advanceReactionNetwork (Vector< LevelData< EBCellFAB > * > &a_cdrSources, Vector< LevelData< EBCellFAB > * > &a_rteSources, const Vector< LevelData< EBCellFAB > * > &a_cdrDensities, const Vector< LevelData< EBCellFAB > * > &a_cdrGradients, const Vector< LevelData< EBCellFAB > * > &a_rteDensities, const LevelData< EBCellFAB > &a_E, const Real &a_time, const Real &a_dt, const int a_level)
	Compute reaction network source terms. This is the level version.

virtual void	advanceReactionNetworkRegularCells (Vector< FArrayBox * > &a_cdrSources, Vector< FArrayBox * > &a_rteSources, const Vector< FArrayBox * > &a_cdrDensities, const Vector< FArrayBox * > &a_cdrGradients, const Vector< FArrayBox * > &a_rteDensities, const FArrayBox &a_E, const Real &a_time, const Real &a_dt, const Real &a_dx, const Box &a_cellBox)
	Compute reaction network sources. This is the regular version which does not do the cut-cells.

virtual void	advanceReactionNetworkIrreg (Vector< EBCellFAB * > &a_cdrSources, Vector< EBCellFAB * > &a_rteSources, const Vector< EBCellFAB * > &a_cdrDensities, const Vector< EBCellFAB * > &a_cdrGradients, const Vector< EBCellFAB * > &a_cdrVelocities, const Vector< EBCellFAB * > &a_rteDensities, const EBCellFAB &a_E, const Real &a_time, const Real &a_dt, const Real &a_dx, const Box &a_cellBox, const int a_lvl, const DataIndex &a_dit)
	Compute reaction network sources. This is the irregular version which only does the cut-cells.

virtual void	advanceReactionNetworkIrregInterp (Vector< EBCellFAB * > &a_cdrSources, Vector< EBCellFAB * > &a_rteSources, const Vector< EBCellFAB * > &a_cdrDensities, const Vector< EBCellFAB * > &a_cdrGradients, const Vector< EBCellFAB * > &a_cdrVelocities, const Vector< EBCellFAB * > &a_rteDensities, const EBCellFAB &a_E, const Real &a_time, const Real &a_dt, const Real &a_dx, const Box &a_box, const int a_lvl, const DataIndex &a_dit)
	Compute reaction network sources. This version interpolates everything to the centroid and runs with that data.

virtual void	advanceReactionNetworkIrregUpwind (Vector< EBCellFAB * > &a_cdrSources, Vector< EBCellFAB * > &a_rteSources, const Vector< EBCellFAB * > &a_cdrDensities, const Vector< EBCellFAB * > &a_cdrGradients, const Vector< EBCellFAB * > &a_cdrVelocities, const Vector< EBCellFAB * > &a_rteDensities, const EBCellFAB &a_E, const Real &a_time, const Real &a_dt, const Real &a_dx, const Box &a_box, const int a_lvl, const DataIndex &a_dit)
	Compute reaction network sources.

virtual void	computeCdrDiffusion ()
	Compute diffusion things using whatever is available in the solvers.

virtual void	computeCdrDiffusion (const EBAMRCellData &a_electricFieldCell, const EBAMRIVData &a_electricFieldEB)
	Compute diffusion things using whatever is available in the solvers. This version uses the input electric fields.

virtual void	computeCdrDiffusionCell (Vector< EBAMRCellData > &a_cdrDcoCell, const Vector< EBAMRCellData * > &a_cdrDensities, const EBAMRCellData &a_electricFieldCell, const Real &a_time)
	Compute CDR diffusion coefficients on cell centers. This will call the level version.

virtual void	computeCdrDiffusionCell (Vector< LevelData< EBCellFAB > * > &a_cdrDcoCell, const Vector< LevelData< EBCellFAB > * > &a_cdrDensities, const LevelData< EBCellFAB > &a_electricFieldCell, const int a_lvl, const Real &a_time)
	Compute diffusion coefficients. This is the level version and it will iterate through grid patches.

virtual void	computeCdrDiffusionCellRegular (Vector< FArrayBox * > &a_cdrDcoCell, const Vector< FArrayBox * > &a_cdrDensities, const FArrayBox &a_electricFieldCell, const Box a_cellBox, const Real a_dx, const Real a_time)
	Compute diffusion coefficients on cell centers. This does all the regular cells.

virtual void	computeCdrDiffusionCellIrregular (Vector< EBCellFAB * > &a_cdrDcoCell, const Vector< EBCellFAB * > &a_cdrDensities, const EBCellFAB &a_electricFieldCell, const Real a_dx, const Real &a_time, const int a_lvl, const DataIndex &a_dit)
	Compute diffusion coefficients on cell centers. This version does the irregular cells.

virtual void	computeCdrDiffusionFace (Vector< EBAMRFluxData * > &a_cdrDcoFace, const Vector< EBAMRCellData * > &a_cdrDensities, const EBAMRCellData &a_electricFieldCell, const Real &a_time)
	Compute diffusion coefficients on face centers.

virtual void	computeCdrDiffusionEb (Vector< EBAMRIVData * > &a_cdrDcoEB, const Vector< EBAMRIVData * > &a_cdrDensitiesEB, const EBAMRIVData &a_electricFieldEB, const Real &a_time)
	Compute diffusion coefficients on the EB. This is the AMR version – it will call the level version.

virtual void	computeCdrDiffusionEb (Vector< LevelData< BaseIVFAB< Real > > * > &a_cdrDcoEB, const Vector< LevelData< BaseIVFAB< Real > > * > &a_cdrDensitiesEB, const LevelData< BaseIVFAB< Real > > &a_electricFieldEB, const Real &a_time, const int a_lvl)
	Compute diffusion coefficients on the EB. This is a level version but it will go through all patches.

virtual void	computeCdrDriftVelocities ()
	Compute the CDR drift velocities.

virtual void	computeCdrDriftVelocities (Vector< EBAMRCellData * > &a_velocities, const Vector< EBAMRCellData * > &a_cdrDensities, const EBAMRCellData &a_E, const Real &a_time)
	Compute the cell-centered CDR velocities. This will call the level version.

virtual void	computeCdrDriftVelocities (Vector< LevelData< EBCellFAB > * > &a_cdrVelocities, const Vector< LevelData< EBCellFAB > * > &a_cdrDensities, const LevelData< EBCellFAB > &a_electricField, const int a_lvl, const Real &a_time)
	Compute the cell-centered CDR velocities.

virtual void	computeCdrDriftVelocitiesRegular (Vector< FArrayBox * > &a_cdrVelocities, const Vector< FArrayBox * > &a_cdrDensities, const FArrayBox &a_electricField, const Box &a_cellBox, const Real &a_time, const Real &a_dx)
	Compute the cell-centered CDR velocities.

virtual void	computeCdrDriftVelocitiesIrregular (Vector< EBCellFAB * > &a_velocities, const Vector< EBCellFAB * > &a_cdrDensities, const EBCellFAB &a_electricField, const Real &a_time, const Real &a_dx, const int a_lvl, const DataIndex &a_dit)
	Compute the cell-centered CDR velocities.

virtual void	computeCdrFluxes (Vector< EBAMRIVData * > &a_cdrFluxesEB, const Vector< EBAMRIVData * > &a_extrapCdrFluxes, const Vector< EBAMRIVData * > &a_extrapCdrDensities, const Vector< EBAMRIVData * > &a_extrapCdrVelocities, const Vector< EBAMRIVData * > &a_extrapCdrGradients, const Vector< EBAMRIVData * > &a_extrapRteFluxes, const EBAMRIVData &a_electricField, const Real &a_time)
	Compute CDR fluxes on the EB. This is the AMR version – it will call the level version.

virtual void	computeCdrFluxes (Vector< LevelData< BaseIVFAB< Real > > * > &a_cdrFluxesEB, const Vector< LevelData< BaseIVFAB< Real > > * > &a_extrapCdrFluxes, const Vector< LevelData< BaseIVFAB< Real > > * > &a_extrapCdrDensities, const Vector< LevelData< BaseIVFAB< Real > > * > &a_extrapCdrVelocities, const Vector< LevelData< BaseIVFAB< Real > > * > &a_extrapCdrGradients, const Vector< LevelData< BaseIVFAB< Real > > * > &a_extrapRteFluxes, const LevelData< BaseIVFAB< Real > > &a_electricField, const Real &a_time, const int a_lvl)
	Compute CDR fluxes on the EB. This is the level version which will fill EB fluxes on both electrode and dielectric cells.

virtual void	computeCdrDomainFluxes (Vector< EBAMRIFData * > &a_cdrFluxes, const Vector< EBAMRIFData * > &a_extrapCdrFluxes, const Vector< EBAMRIFData * > &a_extrapCdrDensities, const Vector< EBAMRIFData * > &a_extrapCdrVelocities, const Vector< EBAMRIFData * > &a_extrapCdrGradients, const Vector< EBAMRIFData * > &a_extrapRteFluxes, const EBAMRIFData &a_field, const Real &a_time)
	Compute CDR fluxes on domain faces for usage in boundary conditions. This is the AMR version.

virtual void	computeCdrDomainFluxes (Vector< LevelData< DomainFluxIFFAB > * > a_cdrFluxes, const Vector< LevelData< DomainFluxIFFAB > * > &a_extrapCdrFluxes, const Vector< LevelData< DomainFluxIFFAB > * > &a_extrapCdrDensities, const Vector< LevelData< DomainFluxIFFAB > * > &a_extrapCdrVelocities, const Vector< LevelData< DomainFluxIFFAB > * > &a_extrapCdrGradients, const Vector< LevelData< DomainFluxIFFAB > * > &a_extrapRteFluxes, const LevelData< DomainFluxIFFAB > &a_electricField, const Real &a_time, const int a_lvl)
	Compute CDR fluxes on domain faces for usage in boundary conditions. This is the level version.

virtual void	computeExtrapolatedFluxes (Vector< EBAMRIVData * > &a_extrapCdrFluxesEB, const Vector< EBAMRCellData * > a_cdrDensities, const Vector< EBAMRCellData * > a_cdrVelocities, const phase::which_phase a_phase)
	Extrapolate cell-centered fluxes to the EB centroid.

virtual void	computeExtrapolatedDomainFluxes (Vector< EBAMRIFData * > &a_cdrDomainFluxes, const Vector< EBAMRCellData * > a_cdrDensities, const Vector< EBAMRCellData * > a_cdrVelocities, const phase::which_phase a_phase)
	Extrapolate drift fluxes to domain faces.

virtual void	computeExtrapolatedVelocities (Vector< EBAMRIVData * > &a_cdrVelocitiesEB, const Vector< EBAMRCellData * > a_cdrVelocitiesCell, const phase::which_phase a_phase)
	Extrapolated cell-centered velocities to the EB.

virtual void	deallocateSolverInternals ()
	Deallocate internal solver storages.

virtual void	extrapolateToEb (Vector< EBAMRIVData * > &a_ebData, const phase::which_phase a_phase, const Vector< EBAMRCellData * > &a_cellData)
	Extrapolated cell-centered data to the EB. This is the AMR version for all species.

virtual void	extrapolateToEb (EBAMRIVData &a_ebData, const phase::which_phase a_phase, const EBAMRCellData &a_cellData)
	Extrapolated cell-centered data to the EB. This is the AMR version.

virtual void	extrapolateToEb (LevelData< BaseIVFAB< Real > > &a_ebData, const phase::which_phase a_phase, const LevelData< EBCellFAB > &a_cellData, const int a_lvl)
	Extrapolated cell-centered data to the EB. This is the level version.

virtual void	extrapolateToDomainFaces (Vector< EBAMRIFData * > &a_domainData, const phase::which_phase a_phase, const Vector< EBAMRCellData * > &a_cellData)
	Extrapolate cell-centered data to domain faces. This is the AMR version that does all species.

virtual void	extrapolateToDomainFaces (EBAMRIFData &a_domainData, const phase::which_phase a_phase, const EBAMRCellData &a_cellData)
	Extrapolate cell-centered data to domain faces. This is the AMR version.

virtual void	extrapolateToDomainFaces (LevelData< DomainFluxIFFAB > &a_domainData, const phase::which_phase a_phase, const LevelData< EBCellFAB > &a_cellData, const int a_lvl)
	Extrapolate cell-centered data to domain faces. This is the AMR version.

virtual void	extrapolateVectorToDomainFaces (Vector< EBAMRIFData * > &a_domainData, const phase::which_phase a_phase, const Vector< EBAMRCellData * > &a_cellData)
	Extrapolate vector data to domain faces. This is the AMR version that does all species.

virtual void	extrapolateVectorToDomainFaces (EBAMRIFData &a_domainData, const phase::which_phase a_phase, const EBAMRCellData &a_cellData)
	Extrapolate vector data to domain faces. This is the AMR version.

virtual void	extrapolateVelocitiesToDomainFaces (Vector< EBAMRIFData * > &a_domainVelocities, const phase::which_phase a_phase, const Vector< EBAMRCellData * > &a_cellVelocities)
	Extrapolate velocities data to domain faces. This is the AMR version that does all species.

virtual void	getCdrMax (Real &a_cdrMax, std::string &a_solverName) const
	Get maximum density in the CDR species. This will fetch both the value and the solver name.

virtual void	initialSigma ()
	Initialize the surface charge.

virtual void	projectFlux (EBAMRIVData &a_projectedFlux, const EBAMRIVData &a_flux)
	Project a flux onto the EB normal. This is the AMR version.

virtual void	projectFlux (LevelData< BaseIVFAB< Real > > &a_projectedFlux, const LevelData< BaseIVFAB< Real > > &a_flux, const int a_lvl)
	Project a flux onto the EB normal. This is the level version.

virtual void	projectDomain (EBAMRIFData &a_projectedFlux, const EBAMRIFData &a_flux)
	Project flux onto domain edges/faces.

virtual void	regridSolvers (const int a_lmin, const int a_oldFinestLevel, const int a_newFinestLevel)
	Regrid all of our solvers.

virtual void	resetDielectricCells (EBAMRIVData &a_data) const
	Set data defined over dielectric cells to zero.

virtual void	sanityCheck () const
	Do a sanity check. Only works when DEBUG=TRUE.

virtual void	setupCdr ()
	Setup the CDR solvers.

virtual void	setupPoisson ()
	Setup the field solver.

virtual void	setupRadiativeTransfer ()
	Setup the RTE solvers.

virtual void	setupSigma ()
	Set up the surface charge solver.

virtual void	setSolverVerbosity ()
	Set solver verbosity.

virtual void	solverDump ()
	All solvers writes plot files. These are not the same as the plot files written by Driver.

virtual void	setVoltage (std::function< Real(const Real a_time)> a_voltage)
	Set voltage curve.

virtual void	solveRadiativeTransfer (const Real a_dt)
	Solve the radiative transfer problem.

virtual void	setCdrSolvers (RefCountedPtr< CdrLayout< CdrSolver > > &a_cdr)
	Set the CDR layout.

virtual void	setFieldSolver (RefCountedPtr< FieldSolver > &a_fieldSolver)
	Set the field solver.

virtual void	setRadiativeTransferSolvers (RefCountedPtr< RtLayout< RtSolver > > &a_rte)
	Set the RTE solvers.

virtual Real	computeElectrodeCurrent ()
	Compute the ohmic current on electrodes.

virtual Real	computeDielectricCurrent ()
	Compute the ohmic current on dielectrics.

virtual Real	computeDomainCurrent ()
	Compute the ohmic current through the domain faces.

virtual Real	computeOhmicInductionCurrent ()
	Compute induced current in external circuit due to ohmic conduction.

virtual Real	computeRelaxationTime ()
	Compute the relaxation time as dt = eps0/conductivity.

virtual Real	getDt ()
	Get dt.

Public Member Functions inherited from TimeStepper
	TimeStepper ()
	Default constructor (does nothing)

virtual	~TimeStepper ()
	Default destructor (does nothing)

void	setAmr (const RefCountedPtr< AmrMesh > &a_amr)
	Set AmrMesh.

void	setComputationalGeometry (const RefCountedPtr< ComputationalGeometry > &a_computationalGeometry)
	Set the computational geometry.

virtual void	postPlot ()
	An option for calling special functions prior to plotting data. Called by Driver in the IMMEDIATELY after writing the plot file.

virtual Vector< long int >	getCheckpointLoads (const std::string a_realm, const int a_level) const
	Get computational loads to be checkpointed.

virtual bool	needToRegrid ()
	Fuction which can have Driver do regrids at arbitrary points in the simulation.

virtual bool	loadBalanceThisRealm (const std::string a_realm) const
	Load balancing query for a specified realm. If this returns true for a_realm, load balancing routines will be called during regrids.

virtual void	loadBalanceBoxes (Vector< Vector< int > > &a_procs, Vector< Vector< Box > > &a_boxes, const std::string a_realm, const Vector< DisjointBoxLayout > &a_grids, const int a_lmin, const int a_finestLevel)
	Load balance grid boxes for a specified realm.

Protected Types
enum class	FieldCoupling { Explicit , SemiImplicit }
	Enum for representing the coupling between the electric field and charge transport.

enum class	AdvectionSolver { Euler , RK2 , MUSCL }
	Enum for switching between various advection solvers.

enum class	DiffusionAlgorithm { Explicit , Implicit , Automatic }
	Enum for distinguishing various ways we handle diffusion.

Protected Types inherited from Physics::CdrPlasma::CdrPlasmaStepper
enum class	SourceTermComputation { Interpolated , Upwind }
	Enum for switching between source term computations. 'Interpolated' interpolates the cell-centered data to cell centroids. InterpolatedStable is the same as Interpolated but with a stability fix near inflow EBs. CellAverage assumes centroid-centered data, and Upwind uses the upwind formulation in Villa et. al.

enum class	TimeCode { Advection , Diffusion , AdvectionDiffusion , Source , RelaxationTime , Restricted , Hardcap , Error , Physics }
	An enum for encapsulating how time steps were restricted.

Protected Member Functions
RefCountedPtr< CdrStorage > &	getCdrStorage (const CdrIterator< CdrSolver > &a_solverIt)
	Function for getting the transient storage assocaited with a particular CDR solver.

RefCountedPtr< RtStorage > &	getRtStorage (const RtIterator< RtSolver > &a_solverIt)
	Function for getting the transient storage assocaited with a particular RTE solver.

void	regridInternals (const int a_lmin, const int a_oldFinestLevel, const int a_newFinestLevel) override
	For regridding internal storage. Does nothing.

void	allocateInternals () override
	Allocate internal storage.

void	allocateScratch ()
	Allocate internal storage.

void	deallocateInternals () override
	Deallocate transient memory.

void	deallocateScratch ()
	Allocate internal storage.

void	computeElectricFieldIntoScratch ()
	Compute electric field into scratch storage.

void	computeCdrGradients ()
	Compute gradients of the CDR densities into scratch storage.

void	extrapolateWithSourceTerm (const Real a_dt)
	Extrapolate cell-centered states with the source term.

void	extrapolateCdrToEB ()
	Extrapolate the cell-centered states to the EB.

void	computeCdrFluxesEB ()
	Compute the CDR boundary condition fluxes on the EBs.

void	extrapolateCdrToDomain ()
	Extrapolate the CDR cell-centered densities to domain edges/face.

void	computeCdrDomainFluxes ()
	Compute the CDR boundary condition fluxes on the domain faces.

void	computeSigmaFlux ()
	Compute the surface flux for the surface charge solver.

void	advanceTransport (const Real a_dt)
	Advance the transport problem.

void	advanceTransportExplicitField (const Real a_dt)
	Advance the transport problem using the explicit or explicit-implicit Euler rule.

void	advanceTransportSemiImplicit (const Real a_dt)
	Advance the transport problem using a semi-implicit formulation for the electric field.

void	advanceCdrReactions (const Real a_dt)
	Advance the reactive problem over a time step dt.

void	advanceRadiativeTransfer (const Real a_dt)
	Advance the radiative transfer problem.

void	floorMass (EBAMRCellData &a_data, const std::string a_message, const RefCountedPtr< CdrSolver > &a_solver) const
	Print how much mass was injected into the system.

void	postStep ()
	Perform post-step operations.

void	computeCdrDriftVelocities (const Real a_time)
	Compute CDR drift velocities.

void	computeCdrDiffusionCoefficients (const Real a_time)
	Compute CDR diffusion coefficients.

void	computeSourceTerms (const Real a_dt)
	Compute source terms for the CDR and RTE equations.

void	parseAdvection ()
	Parse centering for the advective integrator.

void	parseDiffusion ()
	Parse how we handle diffusion.

void	parseField ()
	Parse the transport algorithm.

void	parseFloor ()
	Parse whether or not we should floor the mass.

void	parseDebug ()
	Parse debug mode.

void	parseProfile ()
	Parse profiling mode.

void	parseFHD ()
	Parse inclusion of random diffusion flux.

void	parseRegridSlopes ()
	Parse whether or not to use slopes when regridding.

void	parseFiltering ()
	Parse filtering settings.

bool	solveSemiImplicitPoisson ()
	Solve the semi-implicit Poisson equation.

Protected Member Functions inherited from Physics::CdrPlasma::CdrPlasmaStepper
virtual void	parseVerbosity ()
	Parse class verbosity.

virtual void	parseSolverVerbosity ()
	Parse solver verbosities.

virtual void	parseCFL ()
	Parse the CFL number.

virtual void	parseRelaxationTime ()
	Parse the relaxation time restriction.

virtual void	parseMinDt ()
	Parse the minimum allowed time step.

virtual void	parseMaxDt ()
	Parse the maximum allowed time step.

virtual void	parseFastPoisson ()
	Parse the "fast poisson" method, i.e. how often we solve the Poisson equation.

virtual void	parseFastRadiativeTransfer ()
	Parse the "fast rte" method, i.e. how often we solve the Poisson equation.

virtual void	parseSourceComputation ()
	Parse how we compute the source terms.

virtual void	writeJ (LevelData< EBCellFAB > &a_output, int &a_comp, const std::string a_outputRealm, const int a_level) const
	Compute and put the current density in output data holder.

virtual void	computePhysicsPlotVars (EBAMRCellData &a_physicsPlotVars) const noexcept
	Calculate the user-supplied plot variables.

virtual void	writeData (LevelData< EBCellFAB > &a_output, int &a_comp, const EBAMRCellData &a_data, const std::string a_outputRealm, const int a_level, const bool a_interpToCentroids, const bool a_interpGhost) const noexcept
	Write data to output. Convenience function.

Protected Attributes
FieldCoupling	m_fieldCoupling
	For figuring out which transport algorithm we use.

DiffusionAlgorithm	m_diffusionAlgorithm
	For figuring out how we handle diffusion.

AdvectionSolver	m_advectionSolver
	Advection solver.

std::unique_ptr< Timer >	m_timer
	Timer for run-time profiling.

Vector< RefCountedPtr< CdrStorage > >	m_cdrScratch
	Scratch storage for the CDR solvers.

Vector< RefCountedPtr< RtStorage > >	m_rteScratch
	Scratch storage for the RTE solvers.

RefCountedPtr< FieldStorage >	m_fieldScratch
	Scratch storage for the field solver.

RefCountedPtr< SigmaStorage >	m_sigmaScratch
	Scratch storage for the surface charge solver.

EBAMRCellData	m_semiImplicitRho
	Use for holding the "space charge" when doing semi-implicit solves.

EBAMRCellData	m_conductivityFactorCell
	Used for storing conductivity*dt/eps0 on the cell center.

EBAMRFluxData	m_conductivityFactorFace
	Used for storing conductivity*dt/eps0 on the face centers.

EBAMRIVData	m_conductivityFactorEB
	Used for storing conductivity*dt/eps0 on the EB centers.

EBAMRCellData	m_scratchSemiImplicitRho
	Scratch storage for storing m_semiImplicitRho on the old grids during regrid operations.

EBAMRCellData	m_scratchConductivity
	Scratch storage for storing m_conductivityFactorCell on the old grids during regrid operations.

bool	m_extrapAdvect
	A special flag for letting the CDR solvers know that we want to compute advective derivatives at half time steps.

bool	m_debug
	Enable for debugging this class.

bool	m_profile
	Enable performance profiling.

bool	m_floor
	True if we floor the CDR densities. Due to reactions/EBs, they are not generally non-negative.

bool	m_fhd
	If true, add a stochastic diffusion flux.

bool	m_regridSlopes
	Regrid slopes or not.

bool	m_filterCompensate
	Use compensation step in filter or not.

Real	m_implicitDiffusionThreshold
	Diffusion threshold factor.

int	m_diffusionOrder
	Integration order for diffusion equations.

int	m_numFilterRho
	Number of filterings for space charge.

std::vector< bool >	m_useImplicitDiffusion
	If true, we are using implicit diffusion. Otherwise it's explicit diffusion.

Protected Attributes inherited from Physics::CdrPlasma::CdrPlasmaStepper
std::string	m_realm
	Realm where the time stepper is registered.

std::string	m_className
	Time stepper class name.

phase::which_phase	m_phase
	Plasma phase.

SourceTermComputation	m_whichSourceTermComputation
	Which source term computation.

RefCountedPtr< MultiFluidIndexSpace >	m_multifluidIndexSpace
	Index space.

RefCountedPtr< CdrPlasmaPhysics >	m_physics
	Plasma kinetics.

RefCountedPtr< CdrLayout< CdrSolver > >	m_cdr
	CDR solvers.

RefCountedPtr< RtLayout< RtSolver > >	m_rte
	Radiative transfer solvers.

RefCountedPtr< FieldSolver >	m_fieldSolver
	Poisson solver.

RefCountedPtr< SurfaceODESolver< 1 > >	m_sigma
	Surface charge solver.

EBAMRCellData	m_currentDensity
	Storage for the current.

EBAMRCellData	m_physicsPlotVars
	Storage for physics-observables.

std::function< Real(const Real a_time)>	m_voltage
	Set the potential.

Real	m_minDt
	Minimum allowed time step.

Real	m_maxDt
	Maximum allowed time step.

Real	m_cfl
	CFL number.

Real	m_relaxTime
	Fudge factor for relaxation time.

Real	m_time
	TIme.

Real	m_dt
	Previous time step size.

Real	m_dtCFL
	Computed CFL time step.

TimeCode	m_timeCode
	Time code for step restriction.

int	m_solverVerbosity
	Verbosity for solvers.

int	m_fastRTE
	Fast RTE solver.

int	m_fastPoisson
	Fast RTE solver.

int	m_upwindFactor
	Upwind factor.

Protected Attributes inherited from TimeStepper
int	m_verbosity
	Class verbosity.

int	m_timeStep
	Time step.

Real	m_time
	TIme.

Real	m_dt
	Previous time step size.

RefCountedPtr< AmrMesh >	m_amr
	AmrMesh.

RefCountedPtr< ComputationalGeometry >	m_computationalGeometry
	Computational geometry.

Detailed Description

Class for evolving plasma equations using a split step Godunov method.

This class can use semi-implicit formulations for the transport.

Constructor & Destructor Documentation

◆ CdrPlasmaGodunovStepper()

CdrPlasmaGodunovStepper::CdrPlasmaGodunovStepper ( RefCountedPtr< CdrPlasmaPhysics > & a_physics )

Full constructor.

Parameters

[in] a_physics Reference to the plasma physics implementation.

Member Function Documentation

◆ advance()

Real CdrPlasmaGodunovStepper::advance ( const Real a_dt )

overridevirtual

Implementation of the advance method.

This will switch between the various available implementations.

Parameters

[in] a_dt Time step to try

Returns: Returns a_dt (no adjustable time step for this class)

Implements Physics::CdrPlasma::CdrPlasmaStepper.

◆ advanceCdrReactions()

void CdrPlasmaGodunovStepper::advanceCdrReactions ( const Real a_dt )

protected

Advance the reactive problem over a time step dt.

Parameters

[in] a_dt Time step

Note: Only does the CDR solvers.

◆ advanceRadiativeTransfer()

void CdrPlasmaGodunovStepper::advanceRadiativeTransfer ( const Real a_dt )

protected

Advance the radiative transfer problem.

Parameters

[in] a_dt Time step

Note: This should be called after advanceReactions because that routine sets up the source terms

◆ advanceTransport()

void CdrPlasmaGodunovStepper::advanceTransport ( const Real a_dt )

protected

Advance the transport problem.

Parameters

[in] a_dt Time step.

Note: This will switch between transport algorithms and call the other methods.

◆ advanceTransportExplicitField()

void CdrPlasmaGodunovStepper::advanceTransportExplicitField ( const Real a_dt )

protected

Advance the transport problem using the explicit or explicit-implicit Euler rule.

Parameters

[in] a_dt Time step.

◆ advanceTransportSemiImplicit()

void CdrPlasmaGodunovStepper::advanceTransportSemiImplicit ( const Real a_dt )

protected

Advance the transport problem using a semi-implicit formulation for the electric field.

Under the hood, this is still an Euler solve.

Parameters

[in] a_dt Time step.

◆ allocateInternals()

void CdrPlasmaGodunovStepper::allocateInternals ( )

overrideprotectedvirtual

Allocate internal storage.

This will set up extra memory needed in the semi-implicit coupling.

Implements Physics::CdrPlasma::CdrPlasmaStepper.

◆ allocateScratch()

void CdrPlasmaGodunovStepper::allocateScratch ( )

protected

Allocate internal storage.

Set up some scratch storage needed in the time advancement algorithms.

◆ computeCdrDiffusionCoefficients()

void CdrPlasmaGodunovStepper::computeCdrDiffusionCoefficients ( const Real a_time )

protected

Compute CDR diffusion coefficients.

This is just like the parent method, except that we use the electric field stored in the scratch storage.

Parameters

[in] a_time Time step

◆ computeCdrDomainFluxes()

void CdrPlasmaGodunovStepper::computeCdrDomainFluxes ( )

protected

Compute the CDR boundary condition fluxes on the domain faces.

Prior to calling this routine, one must extrapolated the CDR states to the domain faces (calling extrapolateCdrToDomain)

◆ computeCdrDriftVelocities()

void CdrPlasmaGodunovStepper::computeCdrDriftVelocities ( const Real a_time )

protected

Compute CDR drift velocities.

This is just like the parent method, except that we use the electric field stored in the scratch storage.

Parameters

[in] a_time Time step

◆ computeCdrFluxesEB()

void CdrPlasmaGodunovStepper::computeCdrFluxesEB ( )

protected

Compute the CDR boundary condition fluxes on the EBs.

Prior to calling this routine, one must extrapolated the CDR states to the EB (calling extrapolateCdrToEB)

◆ computeDt()

Real CdrPlasmaGodunovStepper::computeDt ( )

overridevirtual

Compute the time step – this will be different for the different supported algorithms.

Implements Physics::CdrPlasma::CdrPlasmaStepper.

◆ computeElectricFieldIntoScratch()

void CdrPlasmaGodunovStepper::computeElectricFieldIntoScratch ( )

protected

Compute electric field into scratch storage.

This computes the electric field into the scratch storages in m_fieldScratch

◆ computeSourceTerms()

void CdrPlasmaGodunovStepper::computeSourceTerms ( const Real a_dt )

protected

Compute source terms for the CDR and RTE equations.

Parameters

[in] a_dt Time step

◆ deallocateInternals()

void CdrPlasmaGodunovStepper::deallocateInternals ( )

overrideprotectedvirtual

Deallocate transient memory.

Implements Physics::CdrPlasma::CdrPlasmaStepper.

◆ deallocateScratch()

void CdrPlasmaGodunovStepper::deallocateScratch ( )

protected

Allocate internal storage.

Set up some scratch storage needed in the algorithms

◆ extrapolateWithSourceTerm()

void CdrPlasmaGodunovStepper::extrapolateWithSourceTerm ( const Real a_dt )

protected

Extrapolate cell-centered states with the source term.

This will compute cell-centered states (in the scratch storage) as n = n(t) + 0.5*a_dt * S(t)

Parameters

[in] a_dt Time step size.

◆ floorMass()

void CdrPlasmaGodunovStepper::floorMass	(	EBAMRCellData &	a_data,
		const std::string	a_message,
		const RefCountedPtr< CdrSolver > &	a_solver
	)		const

protected

Print how much mass was injected into the system.

Occasionally, mass injection can be necessary due to negative densities arising the reactive problem, or due to redistribution near the embedded boundaries (which does not preserve non-negativeness).

Parameters

[in]	a_message	Base message
[in]	a_solver	Which CDR solver to compute the mass for.

◆ getCdrStorage()

RefCountedPtr< CdrStorage > & CdrPlasmaGodunovStepper::getCdrStorage ( const CdrIterator< CdrSolver > & a_solverIt )

protected

Function for getting the transient storage assocaited with a particular CDR solver.

Parameters

[in] a_solverIt Solver iterator

Returns: This returns the entry in m_cdrScratch corresponding to the solver.

◆ getRtStorage()

RefCountedPtr< RtStorage > & CdrPlasmaGodunovStepper::getRtStorage ( const RtIterator< RtSolver > & a_solverIt )

protected

Function for getting the transient storage assocaited with a particular RTE solver.

Parameters

[in] a_solverIt Solver iterator

Returns: This returns the entry in m_rteScratch corresponding to the solver.

◆ parseOptions()

void CdrPlasmaGodunovStepper::parseOptions ( )

overridevirtual

Parse startup options.

Implements Physics::CdrPlasma::CdrPlasmaStepper.

◆ parseRuntimeOptions()

void CdrPlasmaGodunovStepper::parseRuntimeOptions ( )

overridevirtual

Parse run-time adjustable parameters.

Implements Physics::CdrPlasma::CdrPlasmaStepper.

◆ postCheckpointSetup()

void CdrPlasmaGodunovStepper::postCheckpointSetup ( )

overridevirtual

Run post-checkpoint setup operations.

The override is only relevant for the semi-implicit scheme because the field needs to be computed from a different equation. Fortunately, we've stored the necessary quantities (conductivity and space charge) to HDF5 so we can just set up a standard semi-implicit Poisson solve here.

Reimplemented from Physics::CdrPlasma::CdrPlasmaStepper.

◆ postRegrid()

void CdrPlasmaGodunovStepper::postRegrid ( )

overridevirtual

Perform post-regrid operations.

This releases transient memory needed during the regrid.

Reimplemented from Physics::CdrPlasma::CdrPlasmaStepper.

◆ preRegrid()

void CdrPlasmaGodunovStepper::preRegrid	(	const int	a_lbase,
		const int	a_oldFinestLevel
	)

overridevirtual

Pre-regrid method. Used for storing important variables for the solvers.

This is overriden because the semi-implicit scheme requires us to store some extra variables on the old grids.

Parameters

[in]	a_lmin	The coarsest level that changes
[in]	a_oldFinestLevel	The finest level before the regrid.

Reimplemented from Physics::CdrPlasma::CdrPlasmaStepper.

◆ regrid()

void CdrPlasmaGodunovStepper::regrid	(	const int	a_lmin,
		const int	a_oldFinestLevel,
		const int	a_newFinestLevel
	)

overridevirtual

Regrid method.

This overrides the parent class regrid method when using a semi-implicit scheme.

Parameters

[in]	a_lmin	The coarsest level that changed.
[in]	a_oldFinestLevel	The finest level before the regrid.
[in]	a_newFinestLevel	The finest level after the regrid.

Reimplemented from Physics::CdrPlasma::CdrPlasmaStepper.

◆ regridInternals()

void CdrPlasmaGodunovStepper::regridInternals	(	const int	a_lmin,
		const int	a_oldFinestLevel,
		const int	a_newFinestLevel
	)

overrideprotectedvirtual

For regridding internal storage. Does nothing.

Parameters

[in]	a_lmin	The coarsest level that changed.
[in]	a_oldFinestLevel	The finest level before the regrid.
[in]	a_newFinestLevel	The finest level after the regrid.

Implements Physics::CdrPlasma::CdrPlasmaStepper.

◆ solveSemiImplicitPoisson()

bool CdrPlasmaGodunovStepper::solveSemiImplicitPoisson ( )

protected

Solve the semi-implicit Poisson equation.

Note: This is just like the parent method for solving it, except that we use m_semiImplicitRho for the space charge density. Note that the user should first have set up the semi-implicit Poisson equation by computing the conductivities and calling CdrPlasmasetupSemiImplicitPoisson.

Member Data Documentation

◆ m_semiImplicitRho

EBAMRCellData Physics::CdrPlasma::CdrPlasmaGodunovStepper::m_semiImplicitRho

protected

Use for holding the "space charge" when doing semi-implicit solves.

This is different from the actual space charge because it also includes a diffusive contribution

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

Physics/CdrPlasma/Timesteppers/CdrPlasmaGodunovStepper/CD_CdrPlasmaGodunovStepper.H
Physics/CdrPlasma/Timesteppers/CdrPlasmaGodunovStepper/CD_CdrPlasmaGodunovStepper.cpp

Classes

Public Member Functions

Protected Types

Protected Member Functions

Protected Attributes

Detailed Description

Constructor & Destructor Documentation

◆ CdrPlasmaGodunovStepper()

Member Function Documentation

◆ advance()

◆ advanceCdrReactions()

◆ advanceRadiativeTransfer()

◆ advanceTransport()

◆ advanceTransportExplicitField()

◆ advanceTransportSemiImplicit()

◆ allocateInternals()

◆ allocateScratch()

◆ computeCdrDiffusionCoefficients()

◆ computeCdrDomainFluxes()

◆ computeCdrDriftVelocities()

◆ computeCdrFluxesEB()

◆ computeDt()

◆ computeElectricFieldIntoScratch()

◆ computeSourceTerms()

◆ deallocateInternals()

◆ deallocateScratch()

◆ extrapolateWithSourceTerm()

◆ floorMass()

◆ getCdrStorage()

◆ getRtStorage()

◆ parseOptions()

◆ parseRuntimeOptions()

◆ postCheckpointSetup()

◆ postRegrid()

◆ preRegrid()

◆ regrid()

◆ regridInternals()

◆ solveSemiImplicitPoisson()

Member Data Documentation

◆ m_semiImplicitRho