#include <LinearComptonCollisionFunc.H>

Public Member Functions
AMREX_GPU_HOST_DEVICE AMREX_INLINE void	operator() (index_type const I1s, index_type const I1e, index_type const I2s, index_type const I2e, index_type const AMREX_RESTRICT I1, index_type const AMREX_RESTRICT I2, const SoaData_type &soa_1, const SoaData_type &soa_2, GetParticlePosition< PIdx >, GetParticlePosition< PIdx >, amrex::ParticleReal const, amrex::ParticleReal const, amrex::ParticleReal const, amrex::ParticleReal const, amrex::Real const, amrex::ParticleReal const, amrex::ParticleReal const, amrex::ParticleReal const, amrex::ParticleReal const, amrex::Real const dt, amrex::Real const dV, index_type coll_idx, index_type const cell_start_pair, index_type AMREX_RESTRICT p_mask, index_type AMREX_RESTRICT p_pair_indices_1, index_type AMREX_RESTRICT p_pair_indices_2, amrex::ParticleReal AMREX_RESTRICT p_pair_reaction_weight, amrex::ParticleReal *, amrex::RandomEngine const &engine) const
	Executor of the LinearComptonCollisionFunc class. Performs linear Compton scattering at the cell level using the algorithm described in Higginson et al., Journal of Computational Physics 388, 439-453 (2019). Note that this function does not yet create the product particles, but instead fills an array p_mask that stores which collisions result in a scattering event.

Public Attributes
amrex::ParticleReal	m_event_multiplier
amrex::ParticleReal	m_probability_threshold
amrex::ParticleReal	m_probability_target_value
bool	m_isSameSpecies
bool	m_computeSpeciesDensities = false
bool	m_computeSpeciesTemperatures = false
bool	m_need_product_data = false

Member Function Documentation

◆ operator()()

AMREX_GPU_HOST_DEVICE AMREX_INLINE void LinearComptonCollisionFunc::Executor::operator()	(	index_type const	I1s,
		index_type const	I1e,
		index_type const	I2s,
		index_type const	I2e,
		index_type const *AMREX_RESTRICT	I1,
		index_type const *AMREX_RESTRICT	I2,
		const SoaData_type &	soa_1,
		const SoaData_type &	soa_2,
		GetParticlePosition< PIdx >	,
		GetParticlePosition< PIdx >	,
		amrex::ParticleReal const	,
		amrex::ParticleReal const	,
		amrex::ParticleReal const	,
		amrex::ParticleReal const	,
		amrex::Real const	,
		amrex::ParticleReal const	,
		amrex::ParticleReal const	,
		amrex::ParticleReal const	,
		amrex::ParticleReal const	,
		amrex::Real const	dt,
		amrex::Real const	dV,
		index_type	coll_idx,
		index_type const	cell_start_pair,
		index_type *AMREX_RESTRICT	p_mask,
		index_type *AMREX_RESTRICT	p_pair_indices_1,
		index_type *AMREX_RESTRICT	p_pair_indices_2,
		amrex::ParticleReal *AMREX_RESTRICT	p_pair_reaction_weight,
		amrex::ParticleReal *	,
		amrex::RandomEngine const &	engine ) const

inline

Executor of the LinearComptonCollisionFunc class. Performs linear Compton scattering at the cell level using the algorithm described in Higginson et al., Journal of Computational Physics 388, 439-453 (2019). Note that this function does not yet create the product particles, but instead fills an array p_mask that stores which collisions result in a scattering event.

Also note that there are three main differences between this implementation and the algorithm described in Higginson's paper:

1) The transformation from the lab frame to the center of mass frame is nonrelativistic in Higginson's paper. Here, we implement a relativistic generalization.
2) The behaviour when the estimated probability is greater than one is not specified in Higginson's paper. Here, we provide an implementation using two runtime dependent parameters (probability threshold and probability target value). See documentation for more details.
3) Here, we divide the weight of a particle by the number of times it is paired with other particles. This was not explicitly specified in Higginson's paper.

Parameters

[in]	I1s,I2s	is the start index for I1,I2 (inclusive).
[in]	I1e,I2e	is the stop index for I1,I2 (exclusive).
[in]	I1,I2	index arrays. They determine all elements that will be used.
[in]	soa_1,soa_2	contain the struct of array data of the two species
[in]	dt	time between collision calls (level-independent; collisions run once per coarse step).
[in]	dV	volume of the cell at the current refinement level (level-specific).
[in]	coll_idx	is the collision number of the cell
[in]	cell_start_pair	is the start index of the pairs in that cell.
[out]	p_mask	is a mask that will be set to true if a collision occurs for a given pair. It is only needed here to store information that will be used later on when actually creating the product particles.
[out]	p_pair_indices_1,p_pair_indices_2	arrays that store the indices of the particles of a given pair. They are only needed here to store information that will be used later on when actually creating the product particles.
[out]	p_pair_reaction_weight	stores the weight of the product particles. It is only needed here to store information that will be used later on when actually creating the product particles.
[in]	engine	the random engine.

Member Data Documentation

◆ m_computeSpeciesDensities

bool LinearComptonCollisionFunc::Executor::m_computeSpeciesDensities = false

◆ m_computeSpeciesTemperatures

bool LinearComptonCollisionFunc::Executor::m_computeSpeciesTemperatures = false

◆ m_event_multiplier

amrex::ParticleReal LinearComptonCollisionFunc::Executor::m_event_multiplier

◆ m_isSameSpecies

bool LinearComptonCollisionFunc::Executor::m_isSameSpecies

◆ m_need_product_data

bool LinearComptonCollisionFunc::Executor::m_need_product_data = false

◆ m_probability_target_value

amrex::ParticleReal LinearComptonCollisionFunc::Executor::m_probability_target_value

◆ m_probability_threshold

amrex::ParticleReal LinearComptonCollisionFunc::Executor::m_probability_threshold

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

/home/docs/checkouts/readthedocs.org/user_builds/warpx/checkouts/26.07/Source/Particles/Collision/BinaryCollision/LinearCompton/LinearComptonCollisionFunc.H

Public Member Functions

Public Attributes