This class implements the leveling thinning algorithm as defined in Muraviev, A., et al. arXiv:2006.08593 (2020). The main steps of the algorithm are the following: for every cell we calculate a level weight, defined by the average weight of the species particles in that cell multiplied by the target ratio. Then, particles with a weight lower than the level weight are either removed, with a probability 1 - particle_weight/level_weight, or have their weight set to the level weight. More...

#include <LevelingThinning.H>

Inheritance diagram for LevelingThinning:

Public Member Functions
	LevelingThinning ()=default
	Default constructor of the LevelingThinning class. More...

	LevelingThinning (const std::string species_name)
	Constructor of the LevelingThinning class. More...

void	operator() (WarpXParIter &pti, const int lev, WarpXParticleContainer *const pc) const override final
	A method that performs leveling thinning for the considered species. More...

Public Member Functions inherited from ResamplingAlgorithm
virtual	~ResamplingAlgorithm ()=default
	Virtual destructor of the abstract ResamplingAlgorithm class. More...

Private Attributes
amrex::Real	m_target_ratio = amrex::Real(1.5)

int	m_min_ppc = 1

Detailed Description

This class implements the leveling thinning algorithm as defined in Muraviev, A., et al. arXiv:2006.08593 (2020). The main steps of the algorithm are the following: for every cell we calculate a level weight, defined by the average weight of the species particles in that cell multiplied by the target ratio. Then, particles with a weight lower than the level weight are either removed, with a probability 1 - particle_weight/level_weight, or have their weight set to the level weight.

Constructor & Destructor Documentation

◆ LevelingThinning() [1/2]

LevelingThinning::LevelingThinning ( )

default

Default constructor of the LevelingThinning class.

◆ LevelingThinning() [2/2]

LevelingThinning::LevelingThinning ( const std::string species_name )

Constructor of the LevelingThinning class.

Parameters

[in] species_name the name of the resampled species

Member Function Documentation

◆ operator()()

void LevelingThinning::operator()	(	WarpXParIter &	pti,
		const int	lev,
		WarpXParticleContainer *const	pc
	)		const

finaloverridevirtual

A method that performs leveling thinning for the considered species.

Parameters

[in]	pti	WarpX particle iterator of the particles to resample.
[in]	lev	the index of the refinement level.
[in]	pc	a pointer to the particle container.

Implements ResamplingAlgorithm.

Member Data Documentation

◆ m_min_ppc

int LevelingThinning::m_min_ppc = 1

private

◆ m_target_ratio

amrex::Real LevelingThinning::m_target_ratio = amrex::Real(1.5)