Implements the mixed Generalized-Lagrange Multiplier version of divergence cleaning in: Dedner et al., JCP 175, 645-673 (2002). More...

#include <mixed_glm.h>

Inheritance diagram for wxm::apps::mhd::divergence_cleaning::DednerMixedGLM:

Detailed Description

Implements the mixed Generalized-Lagrange Multiplier version of divergence cleaning in: Dedner et al., JCP 175, 645-673 (2002).

This verison implements a combination of hyperbolic divergence cleaning and parabolic divergence cleaning where the parabolic part enters as a source term.

Specifically, this implements the form in Eqs. 1a, 1b, 4, 1d, and 5, with the constraint given by Eq. 17, except that the ideal part is taken care of separately in the ideal mhd app (wxm::apps::imhd::WmApplication_IMHD_Flux).

So this only adds the \(\nabla \psi\) part of Eqn. 4 and the full Eqn. 5 using Eq. 17.

The numerical flux is derived and given in the line of text after Eq. 42.

Note that in the limit \(c_{p}\rightarrow \infty\), the source term goes to 0 and the parabolic part of the divergence cleaning disappears, so this becomes purely hyperbolic divergence cleaning. This app also has a flag for the source term which can be set to false to turn off. Conversely, \(c_{p}\rightarrow 0\) leads to increased source term (and thus parabolic divergence cleaning) contribution.

For Warpy class see warpy.apps.mhd.divergence_cleaning.divergence_cleaning.mixed_glm

Public Member Functions
	DednerMixedGLM ()

	~DednerMixedGLM () override

void	setup (const WxCryptSet &wxc) override

const std::vector< int > &	getInputVariableIndexes (int flag) const override

const std::vector< int > &	getOutputVariableIndexes (int flag) const override

real	internal_flux (const real q, const real aux, const solverVariables_t *pSV, std::vector< std::vector< real > > &internalFlux) const override

real	numerical_flux (const real q_l, const real q_r, const real aux_l, const real aux_r, const solverVariables_t pFV, real numericalFlux) const override

real	source (const real q, const real aux, const elementGeometry_t pEG, real source) const override

Public Member Functions inherited from WmApplication
	WmApplication ()

virtual	~WmApplication ()=default

virtual void	setup (const WxCryptSet &wxc)

virtual const std::vector< int > &	getAuxiliaryVariableIndexes (int flag=WMAPPLICATIONFLAG_NONE) const

virtual const std::vector< int > &	getCrossVariableIndexes (int flag=WMAPPLICATIONFLAG_NONE) const

const std::vector< std::string > &	getBoundaryNames () const

bool	isOnBoundary (const std::string &boundaryName) const

bool	has (int flag) const

virtual real	numerical_flux (const real q_l, const real q_r, const real aux_l, const real aux_r, const solverVariables_t pFV, real numericalFlux) const

virtual real	internal_flux (const real q, const real aux, const solverVariables_t *pSV, std::vector< std::vector< real > > &internalFlux) const

virtual real	source (const real q, const real aux, const elementGeometry_t pEG, real source) const

virtual void	bc_q (const real q_in, const real aux_in, const real aux_out, const solverVariables_t pFV, real *q_out) const
	Boundary Condition Application which sets the boundary condition on ghost nodes.

virtual real	bcNumericalFlux (const real q_l, const real q_r, const real aux_l, const real aux_r, const solverVariables_t pFV, real numericalFlux) const

virtual void	conserved_to_primitive (const real q, const real aux, real *w) const

virtual void	primitive_to_conserved (const real w, const real aux, real *q) const

virtual void	evaluate_function (const real q, const real aux, const solverVariables_t pSV, real result) const

virtual void	bc_q_kinetic (const real q_in, const real aux_in, const solverVariables_t pFV, real q_out) const

std::shared_ptr< std::string >	app_name ()

virtual const std::vector< int > &	getInputVariableIndexes (int flag=0) const

virtual const std::vector< int > &	getOutputVariableIndexes (int flag=0) const

Protected Attributes
real	_gas_gamma

std::vector< int >	_input_variable_idcs

std::vector< int >	_output_variable_idcs

real	_ch_factor
	Multiplier on max mhd wave speed (fast magnetosonic speed, cf), to calculate ch = cf * _ch_factor \(\in (0, \infty)\).

real	_cr
	Factor for calculating cp = sqrt(cr * ch) \(\in (0, \infty)\), which constributes to the strength of the source term adding the parabolic component of the divergence cleaning.

real	_min_density_floor

real	_min_pressure_floor

bool	_use_psi_source

Protected Attributes inherited from WmApplication
std::string	_app_name

WxLogStream	_debStrm

std::vector< int >	_allowedFlags

std::vector< std::string >	_onBoundaries

Additional Inherited Members
Public Types inherited from WmApplication
typedef Application_Variable	WmApplication_Variable

typedef std::map< std::string, std::string >	replaceMap_t

Static Public Attributes inherited from wxm::app_base
static std::vector< int >	empty

Protected Member Functions inherited from wxm::app_base
virtual	~app_base ()=default

Constructor & Destructor Documentation

◆ DednerMixedGLM()

wxm::apps::mhd::divergence_cleaning::DednerMixedGLM::DednerMixedGLM ( )

◆ ~DednerMixedGLM()

wxm::apps::mhd::divergence_cleaning::DednerMixedGLM::~DednerMixedGLM ( )

override

Member Function Documentation

◆ getInputVariableIndexes()

const std::vector< int > & wxm::apps::mhd::divergence_cleaning::DednerMixedGLM::getInputVariableIndexes ( int flag ) const

inlineoverridevirtual

Reimplemented from wxm::app_base.

◆ getOutputVariableIndexes()

const std::vector< int > & wxm::apps::mhd::divergence_cleaning::DednerMixedGLM::getOutputVariableIndexes ( int flag ) const

inlineoverridevirtual

Reimplemented from wxm::app_base.

◆ internal_flux()

real wxm::apps::mhd::divergence_cleaning::DednerMixedGLM::internal_flux	(	const real *	q,
		const real *	aux,
		const solverVariables_t *	pSV,
		std::vector< std::vector< real > > &	internalFlux
	)		const

overridevirtual

Reimplemented from WmApplication.

◆ numerical_flux()

real wxm::apps::mhd::divergence_cleaning::DednerMixedGLM::numerical_flux	(	const real *	q_l,
		const real *	q_r,
		const real *	aux_l,
		const real *	aux_r,
		const solverVariables_t *	pFV,
		real *	numericalFlux
	)		const

overridevirtual

Reimplemented from WmApplication.

◆ setup()

void wxm::apps::mhd::divergence_cleaning::DednerMixedGLM::setup ( const WxCryptSet & wxc )

overridevirtual

Reimplemented from WmApplication.

◆ source()

real wxm::apps::mhd::divergence_cleaning::DednerMixedGLM::source	(	const real *	q,
		const real *	aux,
		const elementGeometry_t *	pEG,
		real *	source
	)		const

overridevirtual

Reimplemented from WmApplication.

Member Data Documentation

◆ _ch_factor

real wxm::apps::mhd::divergence_cleaning::DednerMixedGLM::_ch_factor

protected

Multiplier on max mhd wave speed (fast magnetosonic speed, cf), to calculate ch = cf * _ch_factor \(\in (0, \infty)\).

Default: 1. Dedner paper uses 1. Probably should be at minimum 1 so that divergence errors advect faster than fastest MHD wave.

◆ _cr

real wxm::apps::mhd::divergence_cleaning::DednerMixedGLM::_cr

protected

Factor for calculating cp = sqrt(cr * ch) \(\in (0, \infty)\), which constributes to the strength of the source term adding the parabolic component of the divergence cleaning.

Default: 0.18. This is the value that Dedner uses.

◆ _gas_gamma

real wxm::apps::mhd::divergence_cleaning::DednerMixedGLM::_gas_gamma

protected

◆ _input_variable_idcs

std::vector<int> wxm::apps::mhd::divergence_cleaning::DednerMixedGLM::_input_variable_idcs

protected

◆ _min_density_floor

real wxm::apps::mhd::divergence_cleaning::DednerMixedGLM::_min_density_floor

protected

◆ _min_pressure_floor

real wxm::apps::mhd::divergence_cleaning::DednerMixedGLM::_min_pressure_floor

protected

◆ _output_variable_idcs

std::vector<int> wxm::apps::mhd::divergence_cleaning::DednerMixedGLM::_output_variable_idcs

protected

◆ _use_psi_source

bool wxm::apps::mhd::divergence_cleaning::DednerMixedGLM::_use_psi_source

protected

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

src/apps/mhd/divergence_cleaning/mixed_glm.h

Detailed Description

Public Member Functions

Protected Attributes

Additional Inherited Members

Constructor & Destructor Documentation

◆ DednerMixedGLM()

◆ ~DednerMixedGLM()

Member Function Documentation

◆ getInputVariableIndexes()

◆ getOutputVariableIndexes()

◆ internal_flux()

◆ numerical_flux()

◆ setup()

◆ source()

Member Data Documentation

◆ _ch_factor

◆ _cr

◆ _gas_gamma

◆ _input_variable_idcs

◆ _min_density_floor

◆ _min_pressure_floor

◆ _output_variable_idcs

◆ _use_psi_source