cs_runge_kutta_integrator.cpp File Reference

#include "base/cs_defs.h"
#include <assert.h>
#include <stdarg.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include "bft/bft_error.h"
#include "bft/bft_printf.h"
#include "base/cs_dispatch.h"
#include "base/cs_field_default.h"
#include "base/cs_field_pointer.h"
#include "base/cs_runge_kutta_integrator.h"
#include "base/cs_runge_kutta_integrator_param.h"

Include dependency graph for cs_runge_kutta_integrator.cpp:

Functions
int	cs_runge_kutta_integrator_create (cs_runge_kutta_scheme_t scheme, const char *name, const cs_real_t *dt, int dim, cs_lnum_t n_elts)
	Create a RK integrator. More...
void	cs_runge_kutta_integrators_initialize ()
	Create RK integrator structures. More...
cs_runge_kutta_integrator_t *	cs_runge_kutta_integrator_by_id (int rk_id)
	Return a Runge-Kutta integrator by id. More...
void	cs_runge_kutta_integrators_destroy ()
	Clean all Runge-Kutta integrators. More...
void	cs_runge_kutta_staging_potential (cs_dispatch_context &ctx, cs_runge_kutta_integrator_t *rk, cs_real_t *phi_stage)
	Perform one Runge-Kutta staging for the potential in the NS system. More...
void	cs_runge_kutta_stage_complete_scalar_rhs (cs_dispatch_context &ctx, cs_runge_kutta_integrator_t *rk, int idtvar, int f_id, int imucpp, cs_equation_param_t *eqp, cs_field_bc_coeffs_t *bc_coeffs, const cs_real_t i_massflux[], const cs_real_t b_massflux[], const cs_real_t i_visc[], const cs_real_t b_visc[], cs_real_t viscel[][6], const cs_real_t weighf[][2], const cs_real_t weighb[], int icvflb, const int icvfli[], cs_real_t pvar[], const cs_real_t xcpp[])
	prepare and complete rhs per stage for Runge-Kutta integrator. Align with legacy equations' building sequence. 1) Collect initial rhs done in cs_solve_navier_stokes/cs_solve_equation_scalar 2) Complete rhs with adding explicit part of the convection/diffusion balance More precisely, the right hand side is updated as follows: More...

Variables
cs_runge_kutta_integrator_t **	_rk_lst = nullptr
int	_n_rk_integrators = 0

Function Documentation

cs_runge_kutta_integrator_by_id()

cs_runge_kutta_integrator_t * cs_runge_kutta_integrator_by_id

(

int

rk_id

)

Return a Runge-Kutta integrator by id.

cs_runge_kutta_integrator_create()

int cs_runge_kutta_integrator_create	(	cs_runge_kutta_scheme_t	scheme,
		const char *	name,
		const cs_real_t *	dt,
		int	dim,
		cs_lnum_t	n_elts
	)

Create a RK integrator.

Parameters

[in]	scheme	RK scheme type (RK_NONE, RK1, RK2, RK3, RK4)
[in]	name	associated equation or field's name
[in]	dt	time step
[in]	dim	variable dimension
[in]	n_elts	number of computational elements

return the RK integrator's id in the RK list

cs_runge_kutta_integrators_destroy()

void cs_runge_kutta_integrators_destroy

(

)

Clean all Runge-Kutta integrators.

cs_runge_kutta_integrators_initialize()

void cs_runge_kutta_integrators_initialize

(

)

Create RK integrator structures.

cs_runge_kutta_stage_complete_scalar_rhs()

void cs_runge_kutta_stage_complete_scalar_rhs	(	cs_dispatch_context &	ctx,
		cs_runge_kutta_integrator_t *	rk,
		int	idtvar,
		int	f_id,
		int	imucpp,
		cs_equation_param_t *	eqp,
		cs_field_bc_coeffs_t *	bc_coeffs,
		const cs_real_t	i_massflux[],
		const cs_real_t	b_massflux[],
		const cs_real_t	i_visc[],
		const cs_real_t	b_visc[],
		cs_real_t	viscel[][6],
		const cs_real_t	weighf[][2],
		const cs_real_t	weighb[],
		int	icvflb,
		const int	icvfli[],
		cs_real_t	pvar[],
		const cs_real_t	xcpp[]
	)

prepare and complete rhs per stage for Runge-Kutta integrator. Align with legacy equations' building sequence. 1) Collect initial rhs done in cs_solve_navier_stokes/cs_solve_equation_scalar 2) Complete rhs with adding explicit part of the convection/diffusion balance More precisely, the right hand side is updated as follows:

Warning:

• has already been initialized before calling cs_balance_vector!
• mind the sign minus

Parameters

[in]	ctx	Reference to dispatch context
[in,out]	rk	pointer to a Runge-Kutta integrator
[in]	idtvar	indicator of the temporal scheme
[in]	f_id	field id (or -1)
[in]	imucpp	indicator 0 do not multiply the convectiv term by Cp 1 do multiply the convectiv term by Cp
[in]	eqp	pointer to a cs_equation_param_t structure which contains variable calculation options
[in]	bc_coeffs	boundary condition structure for the variable
[in]	i_massflux	mass flux at interior faces
[in]	b_massflux	mass flux at boundary faces
[in]	i_visc	at interior faces for the r.h.s.
[in]	b_visc	at boundary faces for the r.h.s.
[in]	viscel	symmetric cell tensor
[in]	weighf	internal face weight between cells i j in case of tensor diffusion
[in]	weighb	boundary face weight for cells i in case of tensor diffusion
[in]	icvflb	global indicator of boundary convection flux 0 upwind scheme at all boundary faces 1 imposed flux at some boundary faces
[in]	icvfli	boundary face indicator array of convection flux 0 upwind scheme 1 imposed flux
[in]	pvar	solved variable (current time step)
[in]	xcpp	array of specific heat (Cp)

cs_runge_kutta_staging_potential()

void cs_runge_kutta_staging_potential	(	cs_dispatch_context &	ctx,
		cs_runge_kutta_integrator_t *	rk,
		cs_real_t *	phi_stage
	)

Perform one Runge-Kutta staging for the potential in the NS system.

Parameters

[in]	ctx	Reference to dispatch context
[in,out]	rk	pointer to a Runge-Kutta integrator
[in,out]	phi_stage	high level variable array along stages

Variable Documentation

_n_rk_integrators

int _n_rk_integrators = 0

_rk_lst

cs_runge_kutta_integrator_t** _rk_lst = nullptr