Classes
struct	Arrhenius
	Modified Arrhenius rate k = A T^b exp(-Ea / (R T)), SI units with kmol. More...

struct	ConstantVolumeReactor
	Adiabatic, constant-volume reactor at density rho with state y = (Y_1 .. Y_Ns, T): dY_k/dt = W_k omega_k / rho, dT/dt = -sum_k u_k omega_k / (rho cv), with u_k the molar internal energies. The ODE system of integrate(). More...

struct	DenseLU
	Dense linear solver of integrate(): A = diagonal I - J factored by lu_factor in n * n doubles (LU, like J in the lanes' layout) and n integers (pivot). More...

struct	Dimension
	Exponents of a quantity's dimension in the basis in which a YAML file sets its default units: mass, length, time, temperature, quantity, energy, pressure (e.g. a molar enthalpy is energy^1 quantity^-1). More...

struct	IgnitionObserver
	Ignition delay as the time of the maximum of dT/dt: the vertex of the parabola through the largest sample and its two neighbors. The reactor has ignited once dT/dt has since fallen below half that maximum: the thermal runaway is over, however small the temperature rise (very lean mixtures). More...

struct	KineticsTable
	Reactions of a mechanism as flat device tables (compressed rows per reaction), in MemorySpace. All quantities SI with kmol, in double precision in every build. Concentrations C_k = rho Y_k / W_k. More...

struct	MassFractions
	Mass fractions held in a contiguous array. More...

struct	Mechanism
	One ideal-gas phase of a Cantera YAML file: its elements, species and reactions. More...

struct	NoObserver
	An observer that ignores the trajectory. More...

struct	OpticallyThinRadiation
	Optically thin radiative loss q = 4 sigma sum_i p_i a_i(T) (T^4 - T_amb^4), p_i the partial pressures [atm] of the radiating species present. More...

struct	Reaction
	A gas-phase reaction: elementary, three-body or falloff (Lindemann, Troe, SRI), reversible (reverse rate from the equilibrium constant) or not, with optional non-integer orders. More...

struct	ReactionDerivatives
	Per-reaction derivatives of the rates of progress, in one block of KineticsTable::derivatives_size() doubles: d q / dT at fixed concentrations; d q / d C of each forward (reactant) and reverse (product) mass-action term; d q / d[M] of third-body and falloff reactions; and the d q / d C_j shared by every species through the pressure of PLOG and Chebyshev reactions. More...

struct	ReactorOptions

struct	Rodas
	Coefficients of RODAS (Hairer & Wanner, Solving ODEs II, IV.7; rodas.f), a stiffly accurate, L-stable Rosenbrock method of order 4 with an embedded order-3 solution, in the form with A = I / (gamma h) - J. More...

struct	RosenbrockOptions

struct	RosenbrockResult

struct	SerialLanes
	One thread does all the work of a cell. More...

struct	SparseLU
	The linear solver of integrate() for SparseLUPattern: factor() gathers diagonal I - J_s from the compact Jacobian and factors it, then z = A_s^-1 u and beta = 1 - v . z; solve() is A_s^-1 b plus the Sherman-Morrison correction z (v . A_s^-1 b) / beta. More...

struct	SparseLUPattern
	The pattern of a reactor Jacobian in (Y_1 .. Y_Ns, T) without its rank-one part, an elimination order, the fill of its LU factors and the operations of a factorization without pivoting, on the device. More...

struct	Species

struct	SpeciesThermo
	Species thermodynamics as piecewise polynomials in NASA-9 form, one per temperature range: cp/R = a0/T^2 + a1/T + a2 + a3 T + a4 T^2 + a5 T^3 + a6 T^4, h/(RT) = -a0/T^2 + a1 ln(T)/T + a2 + a3 T/2 + a4 T^2/3 + a5 T^3/4 + a6 T^4/5 + a7/T, s/R = -a0/(2 T^2) - a1/T + a2 ln(T) + a3 T + a4 T^2/2 + a5 T^3/3 + a6 T^4/4 + a8. More...

struct	SpeciesTransport
	Gas transport data of a species (Cantera's "gas" transport model), in SI units. More...

struct	TeamLanes
	All threads and vector lanes of a team (TeamPolicy(cells, T, V), T * V lanes) share the work of a cell. Scalar code runs on every lane with the same values; sums accumulate SUM_SHARES strided shares in index order and then the shares in order, so a team's results do not depend on its width. Needs scratch_bytes(T * V) of level-0 team scratch. More...

struct	ThermalDiffusionFits
	Pair data of Cantera's mixture-averaged thermal diffusion: its fit of the reduced collision integral C* as a polynomial of degree 8 in ln T* (Monchick & Mason's table, with the fits in the reduced dipole moment), and the pair's well depth epsilon_kj [J], per pair (k, j >= k), k-major. More...

struct	ThermoTable
	Species thermodynamics of a mechanism on the device, in double precision in every build: per species its NASA-9 form ranges (SpeciesThermo), and the ideal-gas mixture properties per unit mass. More...

struct	TransportFits
	Pure-species transport fits of a mechanism, as Cantera computes them for its mixture-averaged model: polynomials of degree 4 in ln T of sqrt(mu_k / sqrt(T)) (viscosity), lambda_k / sqrt(T) (conductivity) and p D_kj / T^1.5 (binary diffusion coefficients), fitted over the mechanism's common thermo temperature range from Lennard-Jones collision integrals (Monchick & Mason) with Cantera's polar corrections and Parker's rotational relaxation. More...

struct	TransportTable
	Transport properties of a mechanism's mixtures on the device, in double precision in every build: Wilke's viscosity, the conductivity of Mathur, Tondon & Saxena, and the species diffusion coefficients D_k of the model (Cantera's "mixture-averaged" and "unity-Lewis-number" models, and constant Lewis numbers), to be used with mole-fraction gradients and a correction velocity; with the mixture-averaged model optionally also Cantera's thermal diffusion coefficients (Soret effect). More...

class	UnitSystem
	Default units of a YAML file (its `units:` map; Cantera's defaults otherwise) and conversion of values to Mallard's units: SI with quantities in kmol, as Cantera uses internally. More...

Enumerations
enum class	ThermoModel { NASA7 , NASA9 , CONSTANT_CP }

enum class	MoleculeGeometry { ATOM , LINEAR , NONLINEAR }

enum class	ReactionType { ELEMENTARY , THREE_BODY , FALLOFF , PLOG , CHEBYSHEV }

enum class	FalloffType { LINDEMANN , TROE , SRI }

enum class	RadiatingSpecies { H2O = 0 , CO2 = 1 , CO = 2 , CH4 = 3 }

enum class	RosenbrockStatus : uint8_t { SUCCESS = 0 , TOO_MANY_STEPS = 1 , STEP_SIZE_UNDERFLOW = 2 }

enum class	TransportModel { MIXTURE_AVERAGED , UNITY_LEWIS , CONSTANT_LEWIS }

Functions
template<typename MemorySpace = Kokkos::DefaultExecutionSpace::memory_space>
KineticsTable< MemorySpace >	make_kinetics_table (const Mechanism &mechanism, const double C_reg=KineticsTable< MemorySpace >::C_REG)
	Copy a mechanism's reactions to MemorySpace.

template<bool ColumnMajor>
KOKKOS_INLINE_FUNCTION constexpr uint32_t	dense_index (const uint32_t n, const uint32_t r, const uint32_t c)
	Index of entry (r, c) of an n x n matrix stored by rows, or by columns.

double	atomic_weight (const std::string &symbol)
	Atomic weight in kg/kmol of an element symbol (Cantera's table; case-insensitive), or a negative value if unknown.

Mechanism	read_mechanism (const std::string &file, const std::string &phase="")
	Read a phase of a Cantera YAML file.

KOKKOS_INLINE_FUNCTION double	planck_mean_absorption (const RadiatingSpecies s, const double T)
	Planck-mean absorption coefficient, 1 / (m atm), of species s at T (fits valid over 300-2500 K).

bool	use_sparse_lu (const ReactorOptions &options, const Mechanism &mechanism)
	Whether a mechanism's reactors use the sparse LU: as options.sparse says, or automatically from 30 species when the factors of the static pattern fill at most 60% of the dense matrix (GRI-3.0 53%: sparse; on an A100 the sparse LU is about 3x faster at 100 species).

template<typename MemorySpace >
KOKKOS_INLINE_FUNCTION uint32_t	reactor_fast_size (const KineticsTable< MemorySpace > &kinetics, const SparseLUPattern< MemorySpace > *sparse=nullptr)
	Doubles of the part of advance_reactor()'s work memory that can be in fast (team scratch) memory: the state, the factors and the integrator's vectors, the most used in the factorization and solves.

template<typename MemorySpace >
KOKKOS_INLINE_FUNCTION uint32_t	reactor_work_size (const KineticsTable< MemorySpace > &kinetics, const SparseLUPattern< MemorySpace > *sparse=nullptr)
	Doubles of work memory advance_reactor() needs, the linear solver's included: dense, or sparse with a SparseLUPattern.

template<bool Sparse = false, typename Lanes , typename MemorySpace , typename Observer = NoObserver>
KOKKOS_INLINE_FUNCTION RosenbrockResult	advance_reactor (const Lanes &lanes, const ThermoTable< MemorySpace > &thermo, const KineticsTable< MemorySpace > &kinetics, const double rho, const double dt, double Y, double &T, double &h, const ReactorOptions &options, double work, uint32_t pivot, Observer &&observer=Observer(), const SparseLUPattern< MemorySpace > sparse=nullptr, double fast=nullptr, const double forcing=nullptr)
	Advance one adiabatic constant-volume reactor over dt: integrate (Y, T), then clip negative mass fractions, renormalize, and take T from the initial internal energy (so energy is exact). Negative initial mass fractions are clipped the same way before integrating.

template<typename MemorySpace , typename Observer = NoObserver>
KOKKOS_INLINE_FUNCTION RosenbrockResult	advance_reactor (const ThermoTable< MemorySpace > &thermo, const KineticsTable< MemorySpace > &kinetics, const double rho, const double dt, double Y, double &T, double &h, const ReactorOptions &options, double work, uint32_t pivot, Observer &&observer=Observer(), const SparseLUPattern< MemorySpace > sparse=nullptr)
	advance_reactor() by one thread.

template<typename Lanes >
KOKKOS_INLINE_FUNCTION bool	lu_factor (const Lanes &lanes, const uint32_t n, double A, uint32_t pivot)
	Dense LU factorization with partial pivoting of the n x n matrix A (by rows, or by columns for lanes with Lanes::column_major), in place; pivot[i] is the row swapped with row i. Rows below the pivot are eliminated one per lane, each in the same order as by one thread, so the factors do not depend on the lane count.

KOKKOS_INLINE_FUNCTION bool	lu_factor (const uint32_t n, double A, uint32_t pivot)

template<typename Lanes >
KOKKOS_INLINE_FUNCTION void	lu_solve (const Lanes &lanes, const uint32_t n, const double LU, const uint32_t pivot, double *b)
	Solve A x = b in place with the factors of lu_factor, by columns: each column's updates go one row per lane.

KOKKOS_INLINE_FUNCTION void	lu_solve (const uint32_t n, const double LU, const uint32_t pivot, double *b)

KOKKOS_INLINE_FUNCTION constexpr uint32_t	rosenbrock_vectors_size (const uint32_t n)
	Doubles of the vectors integrate() needs for a system of size n (besides the dense J).

template<typename Lanes , typename System , typename Solver >
KOKKOS_INLINE_FUNCTION void	rodas_step (const Lanes &lanes, const System &system, const Solver &solver, const uint32_t n, const double h, const double y0, const double f0, double const k[6], double y, double *f_tmp)
	One RODAS step of size h from y0 with f0 = f(y0), given the factored I / (gamma h) - J.

template<typename Lanes , typename System , typename Solver , typename Observer = NoObserver>
KOKKOS_INLINE_FUNCTION RosenbrockResult	integrate (const Lanes &lanes, const System &system, const Solver &solver, const double t_start, const double t_end, double y, double &h, const RosenbrockOptions &options, double J, double *vectors, Observer &&observer=Observer())
	Integrate dy/dt = f(y) from t_start to t_end with adaptive RODAS steps and error control sqrt(mean((err_i / (atol_i + rtol max(\|y0_i\|, \|y1_i\|)))^2)) <= 1.

template<typename MemorySpace = Kokkos::DefaultExecutionSpace::memory_space>
SparseLUPattern< MemorySpace >	make_sparse_lu_pattern (const Mechanism &mechanism)
	Build the pattern for a mechanism's reactor Jacobian (see SparseLUPattern).

template<typename MemorySpace = Kokkos::DefaultExecutionSpace::memory_space>
ThermoTable< MemorySpace >	make_thermo_table (const Mechanism &mechanism)
	Copy a mechanism's species thermodynamics to MemorySpace.

TransportFits	fit_transport (const Mechanism &mechanism)
	Fit the transport properties of every species of a mechanism.

ThermalDiffusionFits	fit_thermal_diffusion (const Mechanism &mechanism)
	Fit the thermal diffusion data of a mechanism.

template<typename MemorySpace = Kokkos::DefaultExecutionSpace::memory_space>
TransportTable< MemorySpace >	make_transport_table (const Mechanism &mechanism, const TransportModel model, const std::vector< double > &lewis={}, const bool thermal_diffusion=false)
	The transport table of a mechanism for a model.

Variables
constexpr double	AVOGADRO = 6.02214076e26

constexpr double	BOLTZMANN = 1.380649e-23

constexpr double	GAS_CONSTANT = AVOGADRO * BOLTZMANN

constexpr double	ELEMENTARY_CHARGE = 1.602176634e-19

constexpr double	ONE_ATM = 101325.0

constexpr double	LIGHT_SPEED = 299792458.0

constexpr double	EPSILON_0 = 8.854187812773345e-12

constexpr int	N_RADIATING_SPECIES = 4

constexpr double	STEFAN_BOLTZMANN = 5.670374419e-8

Enumeration Type Documentation

◆ FalloffType

enum class chemistry::FalloffType

strong

Enumerator
LINDEMANN
TROE
SRI

◆ MoleculeGeometry

enum class chemistry::MoleculeGeometry

strong

Enumerator
ATOM
LINEAR
NONLINEAR

◆ RadiatingSpecies

enum class chemistry::RadiatingSpecies

strong

Enumerator
H2O
CO2
CO
CH4

◆ ReactionType

enum class chemistry::ReactionType

strong

Enumerator
ELEMENTARY
THREE_BODY
FALLOFF
PLOG
CHEBYSHEV

◆ RosenbrockStatus

enum class chemistry::RosenbrockStatus : uint8_t

strong

Enumerator
SUCCESS
TOO_MANY_STEPS
STEP_SIZE_UNDERFLOW

◆ ThermoModel

enum class chemistry::ThermoModel

strong

Enumerator
NASA7
NASA9
CONSTANT_CP

◆ TransportModel

enum class chemistry::TransportModel

strong

Enumerator
MIXTURE_AVERAGED
UNITY_LEWIS
CONSTANT_LEWIS

Function Documentation

◆ advance_reactor() [1/2]

template<bool Sparse = false, typename Lanes , typename MemorySpace , typename Observer = NoObserver>

KOKKOS_INLINE_FUNCTION RosenbrockResult chemistry::advance_reactor	(	const Lanes &	lanes,
		const ThermoTable< MemorySpace > &	thermo,
		const KineticsTable< MemorySpace > &	kinetics,
		const double	rho,
		const double	dt,
		double *	Y,
		double &	T,
		double &	h,
		const ReactorOptions &	options,
		double *	work,
		uint32_t *	pivot,
		Observer &&	observer = `Observer()`,
		const SparseLUPattern< MemorySpace > *	sparse = `nullptr`,
		double *	fast = `nullptr`,
		const double *	forcing = `nullptr`
	)

Advance one adiabatic constant-volume reactor over dt: integrate (Y, T), then clip negative mass fractions, renormalize, and take T from the initial internal energy (so energy is exact). Negative initial mass fractions are clipped the same way before integrating.

Parameters

Y	Mass fractions (n_species), T temperature: updated in place.
h	Sub-step size: first guess in, proposal for the next call out.
work	reactor_work_size(kinetics, sparse) doubles (with fast, only the Jacobian and the reactor's scratch are used).
pivot	n_species + 1 integers (dense LU).
sparse	The pattern of the sparse LU (Sparse = true; else unused).
fast	Null, or reactor_fast_size(kinetics, sparse) doubles of faster memory for the rest of the work memory.
forcing	Null, or the constant sources g_1 .. g_Ns, g_e per unit initial density (see ConstantVolumeReactor): Y then holds the mass fractions at the end, at density rho (1 + dt sum_k g_k), and T follows from the internal energy e + dt g_e per unit initial mass.

◆ advance_reactor() [2/2]

template<typename MemorySpace , typename Observer = NoObserver>

KOKKOS_INLINE_FUNCTION RosenbrockResult chemistry::advance_reactor	(	const ThermoTable< MemorySpace > &	thermo,
		const KineticsTable< MemorySpace > &	kinetics,
		const double	rho,
		const double	dt,
		double *	Y,
		double &	T,
		double &	h,
		const ReactorOptions &	options,
		double *	work,
		uint32_t *	pivot,
		Observer &&	observer = `Observer()`,
		const SparseLUPattern< MemorySpace > *	sparse = `nullptr`
	)

advance_reactor() by one thread.

◆ atomic_weight()

double chemistry::atomic_weight ( const std::string & symbol )

Atomic weight in kg/kmol of an element symbol (Cantera's table; case-insensitive), or a negative value if unknown.

◆ dense_index()

template<bool ColumnMajor>

KOKKOS_INLINE_FUNCTION constexpr uint32_t chemistry::dense_index	(	const uint32_t	n,
		const uint32_t	r,
		const uint32_t	c
	)

constexpr

Index of entry (r, c) of an n x n matrix stored by rows, or by columns.

◆ fit_thermal_diffusion()

ThermalDiffusionFits chemistry::fit_thermal_diffusion ( const Mechanism & mechanism )

Fit the thermal diffusion data of a mechanism.

Exceptions

std::runtime_error naming a species without gas transport data.

◆ fit_transport()

TransportFits chemistry::fit_transport ( const Mechanism & mechanism )

Fit the transport properties of every species of a mechanism.

Exceptions

std::runtime_error naming a species without gas transport data.

◆ integrate()

template<typename Lanes , typename System , typename Solver , typename Observer = NoObserver>

KOKKOS_INLINE_FUNCTION RosenbrockResult chemistry::integrate	(	const Lanes &	lanes,
		const System &	system,
		const Solver &	solver,
		const double	t_start,
		const double	t_end,
		double *	y,
		double &	h,
		const RosenbrockOptions &	options,
		double *	J,
		double *	vectors,
		Observer &&	observer = `Observer()`
	)

Integrate dy/dt = f(y) from t_start to t_end with adaptive RODAS steps and error control sqrt(mean((err_i / (atol_i + rtol max(|y0_i|, |y1_i|)))^2)) <= 1.

The System provides size(), rhs(lanes, y, f), rhs_jacobian(lanes, y, f, J) (dense J in the lanes' layout, see dense_index), atol(i) and admissible(lanes, y); steps to inadmissible states (e.g. negative mass fractions) are rejected. The Solver factors diagonal I - J (factor) and solves with it (solve), e.g. DenseLU.

Parameters

y	State, advanced in place.
h	First sub-step size if positive; on return the proposed next one.
J	n * n doubles for the Jacobian (kept over rejected sub-steps).
vectors	rosenbrock_vectors_size(n) doubles.
observer	Called as observer(t, y, f) at every accepted state from which a step starts (t_start included, t_end excluded).

◆ lu_factor() [1/2]

template<typename Lanes >

KOKKOS_INLINE_FUNCTION bool chemistry::lu_factor	(	const Lanes &	lanes,
		const uint32_t	n,
		double *	A,
		uint32_t *	pivot
	)

Dense LU factorization with partial pivoting of the n x n matrix A (by rows, or by columns for lanes with Lanes::column_major), in place; pivot[i] is the row swapped with row i. Rows below the pivot are eliminated one per lane, each in the same order as by one thread, so the factors do not depend on the lane count.

Returns: False if A is singular.

◆ lu_factor() [2/2]

KOKKOS_INLINE_FUNCTION bool chemistry::lu_factor	(	const uint32_t	n,
		double *	A,
		uint32_t *	pivot
	)

◆ lu_solve() [1/2]

template<typename Lanes >

KOKKOS_INLINE_FUNCTION void chemistry::lu_solve	(	const Lanes &	lanes,
		const uint32_t	n,
		const double *	LU,
		const uint32_t *	pivot,
		double *	b
	)

Solve A x = b in place with the factors of lu_factor, by columns: each column's updates go one row per lane.

◆ lu_solve() [2/2]

KOKKOS_INLINE_FUNCTION void chemistry::lu_solve	(	const uint32_t	n,
		const double *	LU,
		const uint32_t *	pivot,
		double *	b
	)

◆ make_kinetics_table()

template<typename MemorySpace = Kokkos::DefaultExecutionSpace::memory_space>

KineticsTable< MemorySpace > chemistry::make_kinetics_table	(	const Mechanism &	mechanism,
		const double	C_reg = `KineticsTable<MemorySpace>::C_REG`
	)

Copy a mechanism's reactions to MemorySpace.

Parameters

C_reg Concentration of the regularization of orders 0 < n < 1 (KineticsTable::C_REG by default).

◆ make_sparse_lu_pattern()

template<typename MemorySpace = Kokkos::DefaultExecutionSpace::memory_space>

SparseLUPattern< MemorySpace > chemistry::make_sparse_lu_pattern ( const Mechanism & mechanism )

Build the pattern for a mechanism's reactor Jacobian (see SparseLUPattern).

◆ make_thermo_table()

template<typename MemorySpace = Kokkos::DefaultExecutionSpace::memory_space>

ThermoTable< MemorySpace > chemistry::make_thermo_table ( const Mechanism & mechanism )

Copy a mechanism's species thermodynamics to MemorySpace.

◆ make_transport_table()

template<typename MemorySpace = Kokkos::DefaultExecutionSpace::memory_space>

TransportTable< MemorySpace > chemistry::make_transport_table	(	const Mechanism &	mechanism,
		const TransportModel	model,
		const std::vector< double > &	lewis = `{}`,
		const bool	thermal_diffusion = `false`
	)

The transport table of a mechanism for a model.

Parameters

lewis	Constant Lewis numbers per species (CONSTANT_LEWIS only); empty: 1.
thermal_diffusion	Also the data of the thermal diffusion coefficients (MIXTURE_AVERAGED only).

◆ planck_mean_absorption()

KOKKOS_INLINE_FUNCTION double chemistry::planck_mean_absorption	(	const RadiatingSpecies	s,
		const double	T
	)

Planck-mean absorption coefficient, 1 / (m atm), of species s at T (fits valid over 300-2500 K).

◆ reactor_fast_size()

template<typename MemorySpace >

KOKKOS_INLINE_FUNCTION uint32_t chemistry::reactor_fast_size	(	const KineticsTable< MemorySpace > &	kinetics,
		const SparseLUPattern< MemorySpace > *	sparse = `nullptr`
	)

Doubles of the part of advance_reactor()'s work memory that can be in fast (team scratch) memory: the state, the factors and the integrator's vectors, the most used in the factorization and solves.

◆ reactor_work_size()

template<typename MemorySpace >

KOKKOS_INLINE_FUNCTION uint32_t chemistry::reactor_work_size	(	const KineticsTable< MemorySpace > &	kinetics,
		const SparseLUPattern< MemorySpace > *	sparse = `nullptr`
	)

Doubles of work memory advance_reactor() needs, the linear solver's included: dense, or sparse with a SparseLUPattern.

◆ read_mechanism()

Mechanism chemistry::read_mechanism	(	const std::string &	file,
		const std::string &	phase = `""`
	)

Read a phase of a Cantera YAML file.

Parameters

file	Path of the YAML file.
phase	Name of the phase; empty for the first one.

Exceptions

std::runtime_error naming the offending entry for unsupported or malformed input.

◆ rodas_step()

template<typename Lanes , typename System , typename Solver >

KOKKOS_INLINE_FUNCTION void chemistry::rodas_step	(	const Lanes &	lanes,
		const System &	system,
		const Solver &	solver,
		const uint32_t	n,
		const double	h,
		const double *	y0,
		const double *	f0,
		double *const	k[6],
		double *	y,
		double *	f_tmp
	)

One RODAS step of size h from y0 with f0 = f(y0), given the factored I / (gamma h) - J.

Parameters

k	Six stage vectors (k[5] receives the error estimate).
y	Result; f_tmp is scratch.

◆ rosenbrock_vectors_size()

KOKKOS_INLINE_FUNCTION constexpr uint32_t chemistry::rosenbrock_vectors_size ( const uint32_t n )

constexpr

Doubles of the vectors integrate() needs for a system of size n (besides the dense J).

◆ use_sparse_lu()

bool chemistry::use_sparse_lu	(	const ReactorOptions &	options,
		const Mechanism &	mechanism
	)

inline

Whether a mechanism's reactors use the sparse LU: as options.sparse says, or automatically from 30 species when the factors of the static pattern fill at most 60% of the dense matrix (GRI-3.0 53%: sparse; on an A100 the sparse LU is about 3x faster at 100 species).

Variable Documentation

◆ AVOGADRO

constexpr double chemistry::AVOGADRO = 6.02214076e26

constexpr

◆ BOLTZMANN

constexpr double chemistry::BOLTZMANN = 1.380649e-23

constexpr

◆ ELEMENTARY_CHARGE

constexpr double chemistry::ELEMENTARY_CHARGE = 1.602176634e-19

constexpr

◆ EPSILON_0

constexpr double chemistry::EPSILON_0 = 8.854187812773345e-12

constexpr

◆ GAS_CONSTANT

constexpr double chemistry::GAS_CONSTANT = AVOGADRO * BOLTZMANN

constexpr

◆ LIGHT_SPEED

constexpr double chemistry::LIGHT_SPEED = 299792458.0

constexpr

◆ N_RADIATING_SPECIES

constexpr int chemistry::N_RADIATING_SPECIES = 4

constexpr

◆ ONE_ATM

constexpr double chemistry::ONE_ATM = 101325.0

constexpr

◆ STEFAN_BOLTZMANN

constexpr double chemistry::STEFAN_BOLTZMANN = 5.670374419e-8

constexpr

Classes

Enumerations

Functions

Variables

Enumeration Type Documentation

◆ FalloffType

◆ MoleculeGeometry

◆ RadiatingSpecies

◆ ReactionType

◆ RosenbrockStatus

◆ ThermoModel

◆ TransportModel

Function Documentation

◆ advance_reactor() [1/2]

◆ advance_reactor() [2/2]

◆ atomic_weight()

◆ dense_index()

◆ fit_thermal_diffusion()

◆ fit_transport()

◆ integrate()

◆ lu_factor() [1/2]

◆ lu_factor() [2/2]

◆ lu_solve() [1/2]

◆ lu_solve() [2/2]

◆ make_kinetics_table()

◆ make_sparse_lu_pattern()

◆ make_thermo_table()

◆ make_transport_table()

◆ planck_mean_absorption()

◆ reactor_fast_size()

◆ reactor_work_size()

◆ read_mechanism()

◆ rodas_step()

◆ rosenbrock_vectors_size()

◆ use_sparse_lu()

Variable Documentation

◆ AVOGADRO

◆ BOLTZMANN

◆ ELEMENTARY_CHARGE

◆ EPSILON_0

◆ GAS_CONSTANT

◆ LIGHT_SPEED

◆ N_RADIATING_SPECIES

◆ ONE_ATM

◆ STEFAN_BOLTZMANN