A Rusanov/local Lax-Friedrichs Riemann solver. More...

#include <Rusanov.hpp>

Public Types
using	options = tmpl::list<>

using	dg_package_field_tags = tmpl::list< Tags::TildeD, Tags::TildeYe, Tags::TildeTau, Tags::TildeS< Frame::Inertial >, Tags::TildeB< Frame::Inertial >, Tags::TildePhi, ::Tags::NormalDotFlux< Tags::TildeD >, ::Tags::NormalDotFlux< Tags::TildeYe >, ::Tags::NormalDotFlux< Tags::TildeTau >, ::Tags::NormalDotFlux< Tags::TildeS< Frame::Inertial > >, ::Tags::NormalDotFlux< Tags::TildeB< Frame::Inertial > >, ::Tags::NormalDotFlux< Tags::TildePhi >, AbsCharSpeed >

using	dg_package_data_temporary_tags = tmpl::list< gr::Tags::Lapse< DataVector >, gr::Tags::Shift< DataVector, 3 > >

using	dg_package_data_primitive_tags = tmpl::list<>

using	dg_package_data_volume_tags = tmpl::list<>

using	dg_boundary_terms_volume_tags = tmpl::list<>

Public Types inherited from grmhd::ValenciaDivClean::BoundaryCorrections::BoundaryCorrection
using	creatable_classes = tmpl::list< Hll, Rusanov >

Public Member Functions
	Rusanov (const Rusanov &)=default

Rusanov &	operator= (const Rusanov &)=default

	Rusanov (Rusanov &&)=default

Rusanov &	operator= (Rusanov &&)=default

void	pup (PUP::er &p) override

std::unique_ptr< BoundaryCorrection >	get_clone () const override

Public Member Functions inherited from grmhd::ValenciaDivClean::BoundaryCorrections::BoundaryCorrection
	BoundaryCorrection (const BoundaryCorrection &)=default

BoundaryCorrection &	operator= (const BoundaryCorrection &)=default

	BoundaryCorrection (BoundaryCorrection &&)=default

BoundaryCorrection &	operator= (BoundaryCorrection &&)=default

virtual std::unique_ptr< BoundaryCorrection >	get_clone () const =0

Static Public Member Functions
static double	dg_package_data (gsl::not_null< Scalar< DataVector > * > packaged_tilde_d, gsl::not_null< Scalar< DataVector > * > packaged_tilde_ye, gsl::not_null< Scalar< DataVector > * > packaged_tilde_tau, gsl::not_null< tnsr::i< DataVector, 3, Frame::Inertial > * > packaged_tilde_s, gsl::not_null< tnsr::I< DataVector, 3, Frame::Inertial > * > packaged_tilde_b, gsl::not_null< Scalar< DataVector > * > packaged_tilde_phi, gsl::not_null< Scalar< DataVector > * > packaged_normal_dot_flux_tilde_d, gsl::not_null< Scalar< DataVector > * > packaged_normal_dot_flux_tilde_ye, gsl::not_null< Scalar< DataVector > * > packaged_normal_dot_flux_tilde_tau, gsl::not_null< tnsr::i< DataVector, 3, Frame::Inertial > * > packaged_normal_dot_flux_tilde_s, gsl::not_null< tnsr::I< DataVector, 3, Frame::Inertial > * > packaged_normal_dot_flux_tilde_b, gsl::not_null< Scalar< DataVector > * > packaged_normal_dot_flux_tilde_phi, gsl::not_null< Scalar< DataVector > * > packaged_abs_char_speed, const Scalar< DataVector > &tilde_d, const Scalar< DataVector > &tilde_ye, const Scalar< DataVector > &tilde_tau, const tnsr::i< DataVector, 3, Frame::Inertial > &tilde_s, const tnsr::I< DataVector, 3, Frame::Inertial > &tilde_b, const Scalar< DataVector > &tilde_phi, const tnsr::I< DataVector, 3, Frame::Inertial > &flux_tilde_d, const tnsr::I< DataVector, 3, Frame::Inertial > &flux_tilde_ye, const tnsr::I< DataVector, 3, Frame::Inertial > &flux_tilde_tau, const tnsr::Ij< DataVector, 3, Frame::Inertial > &flux_tilde_s, const tnsr::IJ< DataVector, 3, Frame::Inertial > &flux_tilde_b, const tnsr::I< DataVector, 3, Frame::Inertial > &flux_tilde_phi, const Scalar< DataVector > &lapse, const tnsr::I< DataVector, 3, Frame::Inertial > &shift, const tnsr::i< DataVector, 3, Frame::Inertial > &normal_covector, const tnsr::I< DataVector, 3, Frame::Inertial > &normal_vector, const std::optional< tnsr::I< DataVector, 3, Frame::Inertial > > &, const std::optional< Scalar< DataVector > > &normal_dot_mesh_velocity)

static void	dg_boundary_terms (gsl::not_null< Scalar< DataVector > * > boundary_correction_tilde_d, gsl::not_null< Scalar< DataVector > * > boundary_correction_tilde_ye, gsl::not_null< Scalar< DataVector > * > boundary_correction_tilde_tau, gsl::not_null< tnsr::i< DataVector, 3, Frame::Inertial > * > boundary_correction_tilde_s, gsl::not_null< tnsr::I< DataVector, 3, Frame::Inertial > * > boundary_correction_tilde_b, gsl::not_null< Scalar< DataVector > * > boundary_correction_tilde_phi, const Scalar< DataVector > &tilde_d_int, const Scalar< DataVector > &tilde_ye_int, const Scalar< DataVector > &tilde_tau_int, const tnsr::i< DataVector, 3, Frame::Inertial > &tilde_s_int, const tnsr::I< DataVector, 3, Frame::Inertial > &tilde_b_int, const Scalar< DataVector > &tilde_phi_int, const Scalar< DataVector > &normal_dot_flux_tilde_d_int, const Scalar< DataVector > &normal_dot_flux_tilde_ye_int, const Scalar< DataVector > &normal_dot_flux_tilde_tau_int, const tnsr::i< DataVector, 3, Frame::Inertial > &normal_dot_flux_tilde_s_int, const tnsr::I< DataVector, 3, Frame::Inertial > &normal_dot_flux_tilde_b_int, const Scalar< DataVector > &normal_dot_flux_tilde_phi_int, const Scalar< DataVector > &abs_char_speed_int, const Scalar< DataVector > &tilde_d_ext, const Scalar< DataVector > &tilde_ye_ext, const Scalar< DataVector > &tilde_tau_ext, const tnsr::i< DataVector, 3, Frame::Inertial > &tilde_s_ext, const tnsr::I< DataVector, 3, Frame::Inertial > &tilde_b_ext, const Scalar< DataVector > &tilde_phi_ext, const Scalar< DataVector > &normal_dot_flux_tilde_d_ext, const Scalar< DataVector > &normal_dot_flux_tilde_ye_ext, const Scalar< DataVector > &normal_dot_flux_tilde_tau_ext, const tnsr::i< DataVector, 3, Frame::Inertial > &normal_dot_flux_tilde_s_ext, const tnsr::I< DataVector, 3, Frame::Inertial > &normal_dot_flux_tilde_b_ext, const Scalar< DataVector > &normal_dot_flux_tilde_phi_ext, const Scalar< DataVector > &abs_char_speed_ext, dg::Formulation dg_formulation)

Static Public Attributes
static constexpr Options::String	help

Detailed Description

A Rusanov/local Lax-Friedrichs Riemann solver.

Let \(U\) be the state vector of evolved variables, \(F^i\) the corresponding fluxes, and \(n_i\) be the outward directed unit normal to the interface. Denoting \(F := n_i F^i\), the Rusanov boundary correction is

\begin{align*} G_\text{Rusanov} = \frac{F_\text{int} - F_\text{ext}}{2} - \frac{\text{max}\left(\{|\lambda_\text{int}|\}, \{|\lambda_\text{ext}|\}\right)}{2} \left(U_\text{ext} - U_\text{int}\right), \end{align*}

where "int" and "ext" stand for interior and exterior, and \(\{|\lambda|\}\) is the set of characteristic/signal speeds. The minus sign in front of the \(F_{\text{ext}}\) is necessary because the outward directed normal of the neighboring element has the opposite sign, i.e. \(n_i^{\text{ext}}=-n_i^{\text{int}}\). The characteristic/signal speeds are given in the documentation for grmhd::ValenciaDivClean::characteristic_speeds(). Since the fluid is travelling slower than the speed of light, the speeds we are interested in are

\begin{align*} \lambda_{\pm}&=\pm\alpha-\beta^i n_i, \end{align*}

which correspond to the divergence cleaning field.

Note

In the strong form the dg_boundary_terms function returns \(G - F_\text{int}\)
It may be possible to use the slower speeds for the magnetic field and fluid part of the system in order to make the flux less dissipative for those variables.

Member Function Documentation

◆ get_clone()

std::unique_ptr< BoundaryCorrection > grmhd::ValenciaDivClean::BoundaryCorrections::Rusanov::get_clone ( ) const

overridevirtual

Implements grmhd::ValenciaDivClean::BoundaryCorrections::BoundaryCorrection.

Member Data Documentation

◆ help

constexpr Options::String grmhd::ValenciaDivClean::BoundaryCorrections::Rusanov::help

staticconstexpr

Initial value:

= {
      "Computes the Rusanov or local Lax-Friedrichs boundary correction term "
      "for the GRMHD system."}

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

src/Evolution/Systems/GrMhd/ValenciaDivClean/BoundaryCorrections/Rusanov.hpp

Public Types

Public Member Functions

Static Public Member Functions

Static Public Attributes

Detailed Description

Member Function Documentation

◆ get_clone()

Member Data Documentation

◆ help