pw Namespace Reference

Functions
bool	run_deltaspin_lambda_loop (const int iter, const double drho, const Input_para &inp)
	Run the inner lambda loop for DeltaSpin method to constrain atomic magnetic moments.
void	check_deltaspin_oscillation (const int iter, const double drho, Charge_Mixing *p_chgmix, const Input_para &inp)
	Check if SCF oscillation occurs for DeltaSpin method.
void	iter_init_dftu_pw (const int iter, const int istep, Plus_U &dftu, const void *psi, const ModuleBase::matrix &wg, const UnitCell &ucell, const Input_para &inp)
template<typename T , typename Device >
void	setup_pot (const int istep, UnitCell &ucell, const K_Vectors &kv, Structure_Factor &sf, elecstate::ElecState *pelec, const Parallel_Grid &para_grid, const Charge &chr, pseudopot_cell_vl &locpp, pseudopot_cell_vnl &ppcell, Plus_U &dftu, VSep *vsep_cell, psi::Psi< T, Device > *kspw_psi, hamilt::HamiltBase *p_hamilt, ModulePW::PW_Basis_K *pw_wfc, const ModulePW::PW_Basis *pw_rhod, const Input_para &inp)
void	setup_pwrho (UnitCell &ucell, const bool double_grid, bool &pw_rho_flag, ModulePW::PW_Basis *&pw_rho, ModulePW::PW_Basis *&pw_rhod, ModulePW::PW_Basis_Big *&pw_big, const std::string &classname, const Input_para &inp)
void	teardown_pwrho (bool &pw_rho_flag, const bool double_grid, ModulePW::PW_Basis *&pw_rho, ModulePW::PW_Basis *&pw_rhod)
void	teardown_pwwfc (ModulePW::PW_Basis_K *&pw_wfc)
void	setup_pwwfc (const Input_para &inp, const UnitCell &ucell, const ModulePW::PW_Basis &pw_rho, K_Vectors &kv, ModulePW::PW_Basis_K *&pw_wfc)
void	update_cell_pw (const UnitCell &ucell, pseudopot_cell_vnl &ppcell, const K_Vectors &kv, ModulePW::PW_Basis_K *pw_wfc, const Input_para &inp)

Function Documentation

check_deltaspin_oscillation()

void pw::check_deltaspin_oscillation	(	const int	iter,
		const double	drho,
		Charge_Mixing *	p_chgmix,
		const Input_para &	inp
	)

Check if SCF oscillation occurs for DeltaSpin method.

This function checks if the SCF iteration is oscillating and sets the mixing restart step if oscillation is detected. This is used to increase the precision of magnetization calculation.

Parameters

iter	The current iteration number (1-indexed).
drho	The current charge density difference.
p_chgmix	Pointer to the Charge_Mixing object.
inp	The input parameters.

Return if DeltaSpin is not enabled

Get the singleton instance of SpinConstrain

Check if higher magnetization precision is needed

Detect SCF oscillation

If oscillation detected, set mixing restart step for next iteration

Here is the call graph for this function:

Here is the caller graph for this function:

iter_init_dftu_pw()

void pw::iter_init_dftu_pw	(	const int	iter,
		const int	istep,
		Plus_U &	dftu,
		const void *	psi,
		const ModuleBase::matrix &	wg,
		const UnitCell &	ucell,
		const Input_para &	inp
	)

Here is the call graph for this function:

Here is the caller graph for this function:

run_deltaspin_lambda_loop()

bool pw::run_deltaspin_lambda_loop	(	const int	iter,
		const double	drho,
		const Input_para &	inp
	)

Run the inner lambda loop for DeltaSpin method to constrain atomic magnetic moments.

This function is used in the PW basis SCF iteration to optimize lambda parameters for constraining atomic magnetic moments to target values using the DeltaSpin method.

Parameters

iter	The current iteration number (0-indexed).
drho	The current charge density difference.
inp	The input parameters.

Returns

true if the solver should be skipped (lambda loop was executed), false otherwise.

Return false if DeltaSpin is not enabled

Get the singleton instance of SpinConstrain

Case 1: Magnetic moments not yet converged and SCF is close to convergence. This is the first time we enter the lambda loop after SCF is nearly converged.

Optimize lambda to get target magnetic moments

Case 2: Magnetic moments already converged in previous iteration. Continue to refine lambda in subsequent SCF iterations.

Default: run the normal solver

Here is the call graph for this function:

Here is the caller graph for this function:

setup_pot()

template<typename T , typename Device >

void pw::setup_pot	(	const int	istep,
		UnitCell &	ucell,
		const K_Vectors &	kv,
		Structure_Factor &	sf,
		elecstate::ElecState *	pelec,
		const Parallel_Grid &	para_grid,
		const Charge &	chr,
		pseudopot_cell_vl &	locpp,
		pseudopot_cell_vnl &	ppcell,
		Plus_U &	dftu,
		VSep *	vsep_cell,
		psi::Psi< T, Device > *	kspw_psi,
		hamilt::HamiltBase *	p_hamilt,
		ModulePW::PW_Basis_K *	pw_wfc,
		const ModulePW::PW_Basis *	pw_rhod,
		const Input_para &	inp
	)

0) DFT-1/2 calculations, sep potential need to generate

1) Renew local pseudopotential

2) Symmetrize the charge density (rho)

Symmetry_rho should behind init_scf, because charge should be initialized first. liuyu comment: Symmetry_rho should be located between init_rho and v_of_rho?

3) Calculate the effective potential with rho

liuyu move here 2023-10-09 D in uspp need vloc, thus behind init_scf() calculate the effective coefficient matrix for non-local pseudopotential projectors

4) Onsite projectors

5) Spin-constrained algorithms

6) DFT+U algorithm

Here is the call graph for this function:

Here is the caller graph for this function:

setup_pwrho()

void pw::setup_pwrho	(	UnitCell &	ucell,
		const bool	double_grid,
		bool &	pw_rho_flag,
		ModulePW::PW_Basis *&	pw_rho,
		ModulePW::PW_Basis *&	pw_rhod,
		ModulePW::PW_Basis_Big *&	pw_big,
		const std::string &	classname,
		const Input_para &	inp
	)

initialie the plane wave basis for rho

for OFDFT calculations

initialize the FFT grid

initialize the double grid (for uspp) if necessary

Here is the call graph for this function:

Here is the caller graph for this function:

setup_pwwfc()

void pw::setup_pwwfc	(	const Input_para &	inp,
		const UnitCell &	ucell,
		const ModulePW::PW_Basis &	pw_rho,
		K_Vectors &	kv,
		ModulePW::PW_Basis_K *&	pw_wfc
	)

setup pw_wfc

new plane wave basis, fft grids, etc.

initialize the number of plane waves for each k point

Here is the call graph for this function:

Here is the caller graph for this function:

teardown_pwrho()

void pw::teardown_pwrho	(	bool &	pw_rho_flag,
		const bool	double_grid,
		ModulePW::PW_Basis *&	pw_rho,
		ModulePW::PW_Basis *&	pw_rhod
	)

Here is the caller graph for this function:

teardown_pwwfc()

void pw::teardown_pwwfc

(

ModulePW::PW_Basis_K *&

pw_wfc

)

Here is the caller graph for this function:

update_cell_pw()

void pw::update_cell_pw	(	const UnitCell &	ucell,
		pseudopot_cell_vnl &	ppcell,
		const K_Vectors &	kv,
		ModulePW::PW_Basis_K *	pw_wfc,
		const Input_para &	inp
	)

Here is the call graph for this function:

Here is the caller graph for this function: