https://mooseframework.inl.gov
Public Types | Public Member Functions | Static Public Member Functions | Public Attributes | Static Public Attributes | Protected Member Functions | Protected Attributes | List of all members
LinearAssemblySegregatedSolve Class Reference

Common base class for segregated solvers for the Navier-Stokes equations with linear FV assembly routines. More...

#include <LinearAssemblySegregatedSolve.h>

Inheritance diagram for LinearAssemblySegregatedSolve:
[legend]

Public Types

typedef DataFileName DataFileParameterType
 

Public Member Functions

 LinearAssemblySegregatedSolve (Executioner &ex)
 
virtual void linkRhieChowUserObject () override
 Fetch the Rhie Chow user object that is reponsible for determining face velocities and mass flux. More...
 
virtual bool solve () override
 Performs the momentum pressure coupling. More...
 
const std::vector< LinearSystem * > systemsToSolve () const
 Return pointers to the systems which are solved for within this object. More...
 
virtual void setInnerSolve (SolveObject &) override
 
void setupPressurePin ()
 Setup pressure pin if there is need for one. More...
 
virtual void checkIntegrity ()
 Check if the user defined time kernels. More...
 
virtual void initialSetup ()
 
virtual bool enabled () const
 
std::shared_ptr< MooseObject > getSharedPtr ()
 
std::shared_ptr< const MooseObject > getSharedPtr () const
 
MooseApp & getMooseApp () const
 
const std::string & type () const
 
const std::string & name () const
 
std::string typeAndName () const
 
MooseObjectParameterName uniqueParameterName (const std::string &parameter_name) const
 
MooseObjectName uniqueName () const
 
const InputParameters & parameters () const
 
const hit::Node * getHitNode () const
 
bool hasBase () const
 
const std::string & getBase () const
 
const T & getParam (const std::string &name) const
 
std::vector< std::pair< T1, T2 > > getParam (const std::string &param1, const std::string &param2) const
 
const T * queryParam (const std::string &name) const
 
const T & getRenamedParam (const std::string &old_name, const std::string &new_name) const
 
T getCheckedPointerParam (const std::string &name, const std::string &error_string="") const
 
bool isParamValid (const std::string &name) const
 
bool isParamSetByUser (const std::string &name) const
 
void connectControllableParams (const std::string &parameter, const std::string &object_type, const std::string &object_name, const std::string &object_parameter) const
 
void paramError (const std::string &param, Args... args) const
 
void paramWarning (const std::string &param, Args... args) const
 
void paramInfo (const std::string &param, Args... args) const
 
std::string messagePrefix (const bool hit_prefix=true) const
 
std::string errorPrefix (const std::string &) const
 
void mooseError (Args &&... args) const
 
void mooseDocumentedError (const std::string &repo_name, const unsigned int issue_num, Args &&... args) const
 
void mooseErrorNonPrefixed (Args &&... args) const
 
void mooseWarning (Args &&... args) const
 
void mooseWarningNonPrefixed (Args &&... args) const
 
void mooseDeprecated (Args &&... args) const
 
void mooseInfo (Args &&... args) const
 
void callMooseError (std::string msg, const bool with_prefix, const hit::Node *node=nullptr) const
 
std::string getDataFileName (const std::string &param) const
 
std::string getDataFileNameByName (const std::string &relative_path) const
 
std::string getDataFilePath (const std::string &relative_path) const
 
PerfGraph & perfGraph ()
 
bool isDefaultPostprocessorValue (const std::string &param_name, const unsigned int index=0) const
 
bool hasPostprocessor (const std::string &param_name, const unsigned int index=0) const
 
bool hasPostprocessorByName (const PostprocessorName &name) const
 
std::size_t coupledPostprocessors (const std::string &param_name) const
 
const PostprocessorName & getPostprocessorName (const std::string &param_name, const unsigned int index=0) const
 
const PostprocessorValue & getPostprocessorValue (const std::string &param_name, const unsigned int index=0) const
 
const PostprocessorValue & getPostprocessorValue (const std::string &param_name, const unsigned int index=0) const
 
const PostprocessorValue & getPostprocessorValueOld (const std::string &param_name, const unsigned int index=0) const
 
const PostprocessorValue & getPostprocessorValueOld (const std::string &param_name, const unsigned int index=0) const
 
const PostprocessorValue & getPostprocessorValueOlder (const std::string &param_name, const unsigned int index=0) const
 
const PostprocessorValue & getPostprocessorValueOlder (const std::string &param_name, const unsigned int index=0) const
 
virtual const PostprocessorValue & getPostprocessorValueByName (const PostprocessorName &name) const
 
virtual const PostprocessorValue & getPostprocessorValueByName (const PostprocessorName &name) const
 
const PostprocessorValue & getPostprocessorValueOldByName (const PostprocessorName &name) const
 
const PostprocessorValue & getPostprocessorValueOldByName (const PostprocessorName &name) const
 
const PostprocessorValue & getPostprocessorValueOlderByName (const PostprocessorName &name) const
 
const PostprocessorValue & getPostprocessorValueOlderByName (const PostprocessorName &name) const
 
const Parallel::Communicator & comm () const
 
processor_id_type n_processors () const
 
processor_id_type processor_id () const
 
UserObjectName getUserObjectName (const std::string &param_name) const
 
const T & getUserObject (const std::string &param_name, bool is_dependency=true) const
 
const T & getUserObjectByName (const UserObjectName &object_name, bool is_dependency=true) const
 
const UserObject & getUserObjectBase (const std::string &param_name, bool is_dependency=true) const
 
const UserObject & getUserObjectBaseByName (const UserObjectName &object_name, bool is_dependency=true) const
 
bool hasUserObject (const std::string &param_name) const
 
bool hasUserObject (const std::string &param_name) const
 
bool hasUserObject (const std::string &param_name) const
 
bool hasUserObject (const std::string &param_name) const
 
bool hasUserObjectByName (const UserObjectName &object_name) const
 
bool hasUserObjectByName (const UserObjectName &object_name) const
 
bool hasUserObjectByName (const UserObjectName &object_name) const
 
bool hasUserObjectByName (const UserObjectName &object_name) const
 

Static Public Member Functions

static InputParameters validParams ()
 
static void callMooseError (MooseApp *const app, const InputParameters &params, std::string msg, const bool with_prefix, const hit::Node *node)
 

Public Attributes

const ConsoleStream _console
 

Static Public Attributes

static const std::string type_param
 
static const std::string name_param
 
static const std::string unique_name_param
 
static const std::string app_param
 
static const std::string moose_base_param
 

Protected Member Functions

virtual std::vector< std::pair< unsigned int, Real > > solveMomentumPredictor () override
 Solve a momentum predictor step with a fixed pressure field. More...
 
virtual std::pair< unsigned int, Real > solvePressureCorrector () override
 Solve a pressure corrector step. More...
 
virtual std::pair< unsigned int, Real > correctVelocity (const bool subtract_updated_pressure, const bool recompute_face_mass_flux, const SolverParams &solver_params)
 Computes new velocity field based on computed pressure gradients. More...
 
std::pair< unsigned int, Real > solveAdvectedSystem (const unsigned int system_num, LinearSystem &system, const Real relaxation_factor, libMesh::SolverConfiguration &solver_config, const Real abs_tol, const Real field_relaxation=1.0, const Real min_value_limiter=std::numeric_limits< Real >::min())
 Solve an equation which contains an advection term that depends on the solution of the segregated Navier-Stokes equations. More...
 
std::pair< unsigned int, Real > solveSolidEnergy ()
 Solve an equation which contains the solid energy conservation. More...
 
void checkDependentParameterError (const std::string &main_parameter, const std::vector< std::string > &dependent_parameters, const bool should_be_defined)
 
PerfID registerTimedSection (const std::string &section_name, const unsigned int level) const
 
PerfID registerTimedSection (const std::string &section_name, const unsigned int level, const std::string &live_message, const bool print_dots=true) const
 
std::string timedSectionName (const std::string &section_name) const
 
virtual void addPostprocessorDependencyHelper (const PostprocessorName &) const
 
virtual void addUserObjectDependencyHelper (const UserObject &) const
 

Protected Attributes

std::vector< unsigned int > _momentum_system_numbers
 The number(s) of the system(s) corresponding to the momentum equation(s) More...
 
std::vector< LinearSystem * > _momentum_systems
 Pointer(s) to the system(s) corresponding to the momentum equation(s) More...
 
const unsigned int _pressure_sys_number
 The number of the system corresponding to the pressure equation. More...
 
LinearSystem & _pressure_system
 Reference to the nonlinear system corresponding to the pressure equation. More...
 
const unsigned int _energy_sys_number
 The number of the system corresponding to the energy equation. More...
 
LinearSystem * _energy_system
 Pointer to the nonlinear system corresponding to the fluid energy equation. More...
 
const unsigned int _solid_energy_sys_number
 The number of the system corresponding to the solid energy equation. More...
 
LinearSystem * _solid_energy_system
 Pointer to the nonlinear system corresponding to the solid energy equation. More...
 
std::vector< LinearSystem * > _passive_scalar_systems
 Pointer(s) to the system(s) corresponding to the passive scalar equation(s) More...
 
std::vector< LinearSystem * > _active_scalar_systems
 Pointer(s) to the system(s) corresponding to the active scalar equation(s) More...
 
std::vector< LinearSystem * > _turbulence_systems
 Pointer(s) to the system(s) corresponding to the turbulence equation(s) More...
 
RhieChowMassFlux * _rc_uo
 Pointer to the segregated RhieChow interpolation object. More...
 
std::vector< LinearSystem * > _systems_to_solve
 Shortcut to every linear system that we solve for here. More...
 
const std::vector< SolverSystemName > & _active_scalar_system_names
 The names of the active scalar systems. More...
 
const bool _has_active_scalar_systems
 Boolean for easy check if a active scalar systems shall be solved or not. More...
 
std::vector< unsigned int > _active_scalar_system_numbers
 
const std::vector< Real > _active_scalar_equation_relaxation
 The user-defined relaxation parameter(s) for the active scalar equation(s) More...
 
Moose::PetscSupport::PetscOptions _active_scalar_petsc_options
 Options which hold the petsc settings for the active scalar equation(s) More...
 
SIMPLESolverConfiguration _active_scalar_linear_control
 Options for the linear solver of the active scalar equation(s) More...
 
const Real _active_scalar_l_abs_tol
 Absolute linear tolerance for the active scalar equation(s). More...
 
const std::vector< Real > _active_scalar_absolute_tolerance
 The user-defined absolute tolerance for determining the convergence in active scalars. More...
 
const std::vector< SolverSystemName > & _momentum_system_names
 The names of the momentum systems. More...
 
SIMPLESolverConfiguration _momentum_linear_control
 Options for the linear solver of the momentum equation. More...
 
const Real _momentum_l_abs_tol
 Absolute linear tolerance for the momentum equation(s). More...
 
Moose::PetscSupport::PetscOptions _momentum_petsc_options
 Options which hold the petsc settings for the momentum equation. More...
 
const Real _momentum_equation_relaxation
 The user-defined relaxation parameter for the momentum equation. More...
 
const SolverSystemName & _pressure_system_name
 The name of the pressure system. More...
 
SIMPLESolverConfiguration _pressure_linear_control
 Options for the linear solver of the pressure equation. More...
 
const Real _pressure_l_abs_tol
 Absolute linear tolerance for the pressure equation. More...
 
Moose::PetscSupport::PetscOptions _pressure_petsc_options
 Options which hold the petsc settings for the pressure equation. More...
 
const Real _pressure_variable_relaxation
 The user-defined relaxation parameter for the pressure variable. More...
 
const bool _pin_pressure
 If the pressure needs to be pinned. More...
 
const Real _pressure_pin_value
 The value we want to enforce for pressure. More...
 
dof_id_type _pressure_pin_dof
 The dof ID where the pressure needs to be pinned. More...
 
const bool _has_energy_system
 Boolean for easy check if a fluid energy system shall be solved or not. More...
 
const Real _energy_equation_relaxation
 The user-defined relaxation parameter for the energy equation. More...
 
Moose::PetscSupport::PetscOptions _energy_petsc_options
 Options which hold the petsc settings for the fluid energy equation. More...
 
SIMPLESolverConfiguration _energy_linear_control
 Options for the linear solver of the energy equation. More...
 
const Real _energy_l_abs_tol
 Absolute linear tolerance for the energy equations. More...
 
const bool _has_solid_energy_system
 Boolean for easy check if a solid energy system shall be solved or not. More...
 
Moose::PetscSupport::PetscOptions _solid_energy_petsc_options
 Options which hold the petsc settings for the fluid energy equation. More...
 
SIMPLESolverConfiguration _solid_energy_linear_control
 Options for the linear solver of the energy equation. More...
 
const Real _solid_energy_l_abs_tol
 Absolute linear tolerance for the energy equations. More...
 
const std::vector< SolverSystemName > & _passive_scalar_system_names
 The names of the passive scalar systems. More...
 
const bool _has_passive_scalar_systems
 Boolean for easy check if a passive scalar systems shall be solved or not. More...
 
std::vector< unsigned int > _passive_scalar_system_numbers
 
const std::vector< Real > _passive_scalar_equation_relaxation
 The user-defined relaxation parameter(s) for the passive scalar equation(s) More...
 
Moose::PetscSupport::PetscOptions _passive_scalar_petsc_options
 Options which hold the petsc settings for the passive scalar equation(s) More...
 
SIMPLESolverConfiguration _passive_scalar_linear_control
 Options for the linear solver of the passive scalar equation(s) More...
 
const Real _passive_scalar_l_abs_tol
 Absolute linear tolerance for the passive scalar equation(s). More...
 
const std::vector< SolverSystemName > & _turbulence_system_names
 The names of the turbulence systems. More...
 
const bool _has_turbulence_systems
 Boolean for easy check if a turbulence scalar systems shall be solved or not. More...
 
std::vector< unsigned int > _turbulence_system_numbers
 
const std::vector< Real > _turbulence_equation_relaxation
 The user-defined relaxation parameter(s) for the turbulence equation(s) More...
 
std::vector< Real > _turbulence_field_relaxation
 The user-defined relaxation parameter(s) for the turbulence field(s) More...
 
std::vector< Real > _turbulence_field_min_limit
 The user-defined lower limit for turbulent quantities e.g. k, eps/omega, etc.. More...
 
Moose::PetscSupport::PetscOptions _turbulence_petsc_options
 Options which hold the petsc settings for the turbulence equation(s) More...
 
SIMPLESolverConfiguration _turbulence_linear_control
 Options for the linear solver of the turbulence equation(s) More...
 
const Real _turbulence_l_abs_tol
 Absolute linear tolerance for the turbulence equation(s). More...
 
const Real _momentum_absolute_tolerance
 The user-defined absolute tolerance for determining the convergence in momentum. More...
 
const Real _pressure_absolute_tolerance
 The user-defined absolute tolerance for determining the convergence in pressure. More...
 
const Real _energy_absolute_tolerance
 The user-defined absolute tolerance for determining the convergence in energy. More...
 
const Real _solid_energy_absolute_tolerance
 The user-defined absolute tolerance for determining the convergence in solid energy. More...
 
const std::vector< Real > _passive_scalar_absolute_tolerance
 The user-defined absolute tolerance for determining the convergence in passive scalars. More...
 
const std::vector< Real > _turbulence_absolute_tolerance
 The user-defined absolute tolerance for determining the convergence turbulence variables. More...
 
const unsigned int _num_iterations
 The maximum number of momentum-pressure iterations. More...
 
const bool _continue_on_max_its
 If solve should continue if maximum number of iterations is hit. More...
 
const bool _print_fields
 Debug parameter which allows printing the coupling and solution vectors/matrices. More...
 
Executioner & _executioner
 
FEProblemBase & _problem
 
DisplacedProblem * _displaced_problem
 
MooseMesh & _mesh
 
MooseMesh * _displaced_mesh
 
SystemBase & _solver_sys
 
AuxiliarySystem & _aux
 
SolveObject * _inner_solve
 
const bool & _enabled
 
MooseApp & _app
 
Factory & _factory
 
ActionFactory & _action_factory
 
const std::string & _type
 
const std::string & _name
 
const InputParameters & _pars
 
MooseApp & _pg_moose_app
 
const std::string _prefix
 
const Parallel::Communicator & _communicator
 

Detailed Description

Common base class for segregated solvers for the Navier-Stokes equations with linear FV assembly routines.

Once the nonlinear assembly-based routines are retired, this will be the primary base class instead of SIMPLESolveBase.

Definition at line 22 of file LinearAssemblySegregatedSolve.h.

Constructor & Destructor Documentation

◆ LinearAssemblySegregatedSolve()

LinearAssemblySegregatedSolve::LinearAssemblySegregatedSolve ( Executioner &  ex)

Definition at line 75 of file LinearAssemblySegregatedSolve.C.

76  : SIMPLESolveBase(ex),
77  _pressure_sys_number(_problem.linearSysNum(getParam<SolverSystemName>("pressure_system"))),
78  _pressure_system(_problem.getLinearSystem(_pressure_sys_number)),
79  _energy_sys_number(_has_energy_system
80  ? _problem.linearSysNum(getParam<SolverSystemName>("energy_system"))
81  : libMesh::invalid_uint),
82  _energy_system(_has_energy_system ? &_problem.getLinearSystem(_energy_sys_number) : nullptr),
83  _solid_energy_sys_number(
84  _has_solid_energy_system
85  ? _problem.linearSysNum(getParam<SolverSystemName>("solid_energy_system"))
86  : libMesh::invalid_uint),
87  _solid_energy_system(
88  _has_solid_energy_system ? &_problem.getLinearSystem(_solid_energy_sys_number) : nullptr),
89  _active_scalar_system_names(getParam<std::vector<SolverSystemName>>("active_scalar_systems")),
90  _has_active_scalar_systems(!_active_scalar_system_names.empty()),
91  _active_scalar_equation_relaxation(
92  getParam<std::vector<Real>>("active_scalar_equation_relaxation")),
93  _active_scalar_l_abs_tol(getParam<Real>("active_scalar_l_abs_tol")),
94  _active_scalar_absolute_tolerance(
95  getParam<std::vector<Real>>("active_scalar_absolute_tolerance"))
96 {
97  // We fetch the systems and their numbers for the momentum equations.
98  for (auto system_i : index_range(_momentum_system_names))
99  {
100  _momentum_system_numbers.push_back(_problem.linearSysNum(_momentum_system_names[system_i]));
101  _momentum_systems.push_back(&_problem.getLinearSystem(_momentum_system_numbers[system_i]));
102  _systems_to_solve.push_back(_momentum_systems.back());
103  }
104 
105  _systems_to_solve.push_back(&_pressure_system);
106 
107  if (_has_energy_system)
108  _systems_to_solve.push_back(_energy_system);
109 
110  if (_has_solid_energy_system)
111  _systems_to_solve.push_back(_solid_energy_system);
112  // and for the turbulence surrogate equations
113  if (_has_turbulence_systems)
114  for (auto system_i : index_range(_turbulence_system_names))
115  {
116  _turbulence_system_numbers.push_back(
117  _problem.linearSysNum(_turbulence_system_names[system_i]));
118  _turbulence_systems.push_back(
119  &_problem.getLinearSystem(_turbulence_system_numbers[system_i]));
120  }
121 
122  // and for the passive scalar equations
123  if (_has_passive_scalar_systems)
124  for (auto system_i : index_range(_passive_scalar_system_names))
125  {
126  _passive_scalar_system_numbers.push_back(
127  _problem.linearSysNum(_passive_scalar_system_names[system_i]));
128  _passive_scalar_systems.push_back(
129  &_problem.getLinearSystem(_passive_scalar_system_numbers[system_i]));
130  _systems_to_solve.push_back(_passive_scalar_systems.back());
131  }
132 
133  // and for the active scalar equations
134  if (_has_active_scalar_systems)
135  for (auto system_i : index_range(_active_scalar_system_names))
136  {
137  _active_scalar_system_numbers.push_back(
138  _problem.linearSysNum(_active_scalar_system_names[system_i]));
139  _active_scalar_systems.push_back(
140  &_problem.getLinearSystem(_active_scalar_system_numbers[system_i]));
141  _systems_to_solve.push_back(_active_scalar_systems.back());
142 
143  const auto & active_scalar_petsc_options =
144  getParam<MultiMooseEnum>("active_scalar_petsc_options");
145  const auto & active_scalar_petsc_pair_options = getParam<MooseEnumItem, std::string>(
146  "active_scalar_petsc_options_iname", "active_scalar_petsc_options_value");
147  Moose::PetscSupport::addPetscFlagsToPetscOptions(
148  active_scalar_petsc_options, "", *this, _active_scalar_petsc_options);
149  Moose::PetscSupport::addPetscPairsToPetscOptions(active_scalar_petsc_pair_options,
150  _problem.mesh().dimension(),
151  "",
152  *this,
153  _active_scalar_petsc_options);
154 
155  _active_scalar_linear_control.real_valued_data["rel_tol"] =
156  getParam<Real>("active_scalar_l_tol");
157  _active_scalar_linear_control.real_valued_data["abs_tol"] =
158  getParam<Real>("active_scalar_l_abs_tol");
159  _active_scalar_linear_control.int_valued_data["max_its"] =
160  getParam<unsigned int>("active_scalar_l_max_its");
161  }
162 
163  if (_active_scalar_equation_relaxation.size() != _active_scalar_system_names.size())
164  paramError("active_scalar_equation_relaxation",
165  "Should be the same size as the number of systems");
166 
167  // We disable the prefix here for the time being, the segregated solvers use a different approach
168  // for setting the petsc parameters
169  for (auto & system : _systems_to_solve)
170  system->system().prefix_with_name(false);
171 }
LinearAssemblySegregatedSolve::_active_scalar_equation_relaxation
const std::vector< Real > _active_scalar_equation_relaxation
The user-defined relaxation parameter(s) for the active scalar equation(s)
Definition: LinearAssemblySegregatedSolve.h:126
SolveObject::_problem
FEProblemBase & _problem
Moose::PetscSupport::addPetscFlagsToPetscOptions
void addPetscFlagsToPetscOptions(const MultiMooseEnum &petsc_flags, std::string prefix, const ParallelParamObject &param_object, PetscOptions &petsc_options)
libMesh::invalid_uint
const unsigned int invalid_uint
SolveObject::paramError
void paramError(const std::string &param, Args... args) const
SolveObject::getParam
const T & getParam(const std::string &name) const
SIMPLESolveBase::_has_energy_system
const bool _has_energy_system
Boolean for easy check if a fluid energy system shall be solved or not.
Definition: SIMPLESolveBase.h:127
Moose::PetscSupport::addPetscPairsToPetscOptions
void addPetscPairsToPetscOptions(const std::vector< std::pair< MooseEnumItem, std::string >> &petsc_pair_options, const unsigned int mesh_dimension, std::string prefix, const ParallelParamObject &param_object, PetscOptions &petsc_options)
LinearAssemblySegregatedSolve::_active_scalar_linear_control
SIMPLESolverConfiguration _active_scalar_linear_control
Options for the linear solver of the active scalar equation(s)
Definition: LinearAssemblySegregatedSolve.h:132
LinearAssemblySegregatedSolve::_passive_scalar_systems
std::vector< LinearSystem * > _passive_scalar_systems
Pointer(s) to the system(s) corresponding to the passive scalar equation(s)
Definition: LinearAssemblySegregatedSolve.h:100
SIMPLESolveBase::_has_turbulence_systems
const bool _has_turbulence_systems
Boolean for easy check if a turbulence scalar systems shall be solved or not.
Definition: SIMPLESolveBase.h:187
SIMPLESolveBase::_turbulence_system_numbers
std::vector< unsigned int > _turbulence_system_numbers
Definition: SIMPLESolveBase.h:190
LinearAssemblySegregatedSolve::_solid_energy_system
LinearSystem * _solid_energy_system
Pointer to the nonlinear system corresponding to the solid energy equation.
Definition: LinearAssemblySegregatedSolve.h:97
LinearAssemblySegregatedSolve::_has_active_scalar_systems
const bool _has_active_scalar_systems
Boolean for easy check if a active scalar systems shall be solved or not.
Definition: LinearAssemblySegregatedSolve.h:120
LinearAssemblySegregatedSolve::_active_scalar_l_abs_tol
const Real _active_scalar_l_abs_tol
Absolute linear tolerance for the active scalar equation(s).
Definition: LinearAssemblySegregatedSolve.h:136
libMesh::SolverConfiguration::real_valued_data
std::map< std::string, Real > real_valued_data
SIMPLESolveBase::_passive_scalar_system_names
const std::vector< SolverSystemName > & _passive_scalar_system_names
The names of the passive scalar systems.
Definition: SIMPLESolveBase.h:160
LinearAssemblySegregatedSolve::_solid_energy_sys_number
const unsigned int _solid_energy_sys_number
The number of the system corresponding to the solid energy equation.
Definition: LinearAssemblySegregatedSolve.h:94
LinearAssemblySegregatedSolve::_active_scalar_systems
std::vector< LinearSystem * > _active_scalar_systems
Pointer(s) to the system(s) corresponding to the active scalar equation(s)
Definition: LinearAssemblySegregatedSolve.h:103
LinearAssemblySegregatedSolve::_pressure_system
LinearSystem & _pressure_system
Reference to the nonlinear system corresponding to the pressure equation.
Definition: LinearAssemblySegregatedSolve.h:85
LinearAssemblySegregatedSolve::_momentum_system_numbers
std::vector< unsigned int > _momentum_system_numbers
The number(s) of the system(s) corresponding to the momentum equation(s)
Definition: LinearAssemblySegregatedSolve.h:76
LinearAssemblySegregatedSolve::_energy_system
LinearSystem * _energy_system
Pointer to the nonlinear system corresponding to the fluid energy equation.
Definition: LinearAssemblySegregatedSolve.h:91
LinearAssemblySegregatedSolve::_energy_sys_number
const unsigned int _energy_sys_number
The number of the system corresponding to the energy equation.
Definition: LinearAssemblySegregatedSolve.h:88
MooseMesh::dimension
virtual unsigned int dimension() const
libMesh::SolverConfiguration::int_valued_data
std::map< std::string, int > int_valued_data
SIMPLESolveBase::_turbulence_system_names
const std::vector< SolverSystemName > & _turbulence_system_names
The names of the turbulence systems.
Definition: SIMPLESolveBase.h:184
FEProblemBase::getLinearSystem
LinearSystem & getLinearSystem(unsigned int sys_num)
SIMPLESolveBase::_momentum_system_names
const std::vector< SolverSystemName > & _momentum_system_names
The names of the momentum systems.
Definition: SIMPLESolveBase.h:82
LinearAssemblySegregatedSolve::_turbulence_systems
std::vector< LinearSystem * > _turbulence_systems
Pointer(s) to the system(s) corresponding to the turbulence equation(s)
Definition: LinearAssemblySegregatedSolve.h:106
SIMPLESolveBase::SIMPLESolveBase
SIMPLESolveBase(Executioner &ex)
Definition: SIMPLESolveBase.C:365
LinearAssemblySegregatedSolve::_active_scalar_system_names
const std::vector< SolverSystemName > & _active_scalar_system_names
The names of the active scalar systems.
Definition: LinearAssemblySegregatedSolve.h:117
LinearAssemblySegregatedSolve::_systems_to_solve
std::vector< LinearSystem * > _systems_to_solve
Shortcut to every linear system that we solve for here.
Definition: LinearAssemblySegregatedSolve.h:112
FEProblemBase::mesh
virtual MooseMesh & mesh() override
FEProblemBase::linearSysNum
unsigned int linearSysNum(const LinearSystemName &linear_sys_name) const override
LinearAssemblySegregatedSolve::_active_scalar_petsc_options
Moose::PetscSupport::PetscOptions _active_scalar_petsc_options
Options which hold the petsc settings for the active scalar equation(s)
Definition: LinearAssemblySegregatedSolve.h:129
SIMPLESolveBase::_has_passive_scalar_systems
const bool _has_passive_scalar_systems
Boolean for easy check if a passive scalar systems shall be solved or not.
Definition: SIMPLESolveBase.h:163
LinearAssemblySegregatedSolve::_active_scalar_absolute_tolerance
const std::vector< Real > _active_scalar_absolute_tolerance
The user-defined absolute tolerance for determining the convergence in active scalars.
Definition: LinearAssemblySegregatedSolve.h:139
LinearAssemblySegregatedSolve::_active_scalar_system_numbers
std::vector< unsigned int > _active_scalar_system_numbers
Definition: LinearAssemblySegregatedSolve.h:123
LinearAssemblySegregatedSolve::_momentum_systems
std::vector< LinearSystem * > _momentum_systems
Pointer(s) to the system(s) corresponding to the momentum equation(s)
Definition: LinearAssemblySegregatedSolve.h:79
index_range
auto index_range(const T &sizable)
SIMPLESolveBase::_passive_scalar_system_numbers
std::vector< unsigned int > _passive_scalar_system_numbers
Definition: SIMPLESolveBase.h:166
SIMPLESolveBase::_has_solid_energy_system
const bool _has_solid_energy_system
Boolean for easy check if a solid energy system shall be solved or not.
Definition: SIMPLESolveBase.h:145
LinearAssemblySegregatedSolve::_pressure_sys_number
const unsigned int _pressure_sys_number
The number of the system corresponding to the pressure equation.
Definition: LinearAssemblySegregatedSolve.h:82

Member Function Documentation

◆ checkDependentParameterError()

void SIMPLESolveBase::checkDependentParameterError ( const std::string &  main_parameter,
const std::vector< std::string > &  dependent_parameters,
const bool  should_be_defined 
)
protectedinherited

Definition at line 600 of file SIMPLESolveBase.C.

Referenced by SIMPLESolveBase::SIMPLESolveBase(), and SIMPLESolveNonlinearAssembly::SIMPLESolveNonlinearAssembly().

603 {
604  for (const auto & param : dependent_parameters)
605  if (parameters().isParamSetByUser(param) == !should_be_defined)
606  paramError(param,
607  "This parameter should " + std::string(should_be_defined ? "" : "not") +
608  " be given by the user with the corresponding " + main_parameter +
609  " setting!");
610 }
SolveObject::paramError
void paramError(const std::string &param, Args... args) const
SolveObject::parameters
const InputParameters & parameters() const
SolveObject::isParamSetByUser
bool isParamSetByUser(const std::string &name) const

◆ checkIntegrity()

virtual void SIMPLESolveBase::checkIntegrity ( )
inlinevirtualinherited

Check if the user defined time kernels.

Reimplemented in SIMPLESolve, and SIMPLESolveNonlinearAssembly.

Definition at line 61 of file SIMPLESolveBase.h.

61 {}

◆ correctVelocity()

std::pair< unsigned int, Real > LinearAssemblySegregatedSolve::correctVelocity ( const bool  subtract_updated_pressure,
const bool  recompute_face_mass_flux,
const SolverParams &  solver_params 
)
protectedvirtual

Computes new velocity field based on computed pressure gradients.

Parameters
subtract_updated_pressureIf we need to subtract the updated pressure gradient from the right hand side of the system
recompute_face_mass_fluxIf we want to recompute the face flux too
solver_paramsDummy solver parameter object for the linear solve

Reimplemented in PIMPLESolve.

Definition at line 411 of file LinearAssemblySegregatedSolve.C.

Referenced by PIMPLESolve::correctVelocity(), and solve().

414 {
415  // Compute the coupling fields between the momentum and pressure equations.
416  // The first argument makes sure the pressure gradient is staged at the first
417  // iteration
418  _rc_uo->computeHbyA(subtract_updated_pressure, _print_fields);
419 
420  // We set the preconditioner/controllable parameters for the pressure equations through
421  // petsc options. Linear tolerances will be overridden within the solver.
422  Moose::PetscSupport::petscSetOptions(_pressure_petsc_options, solver_params);
423 
424  // Solve the pressure corrector
425  const auto residuals = solvePressureCorrector();
426 
427  // Compute the face velocity which is used in the advection terms. In certain
428  // segregated solver algorithms (like PISO) this is only done on the last iteration.
429  if (recompute_face_mass_flux)
430  _rc_uo->computeFaceMassFlux();
431 
432  auto & pressure_current_solution = *(_pressure_system.system().current_local_solution.get());
433  auto & pressure_old_solution = *(_pressure_system.solutionPreviousNewton());
434 
435  // Relax the pressure update for the next momentum predictor
436  NS::FV::relaxSolutionUpdate(
437  pressure_current_solution, pressure_old_solution, _pressure_variable_relaxation);
438 
439  // Overwrite old solution
440  pressure_old_solution = pressure_current_solution;
441  _pressure_system.setSolution(pressure_current_solution);
442 
443  // We recompute the updated pressure gradient
444  _pressure_system.computeGradients();
445 
446  // Reconstruct the cell velocity as well to accelerate convergence
447  _rc_uo->computeCellVelocity();
448 
449  return residuals;
450 }
LinearSystem::computeGradients
void computeGradients()
SIMPLESolveBase::_print_fields
const bool _print_fields
Debug parameter which allows printing the coupling and solution vectors/matrices. ...
Definition: SIMPLESolveBase.h:240
RhieChowMassFlux::computeFaceMassFlux
void computeFaceMassFlux()
Update the values of the face velocities in the containers.
Definition: RhieChowMassFlux.C:285
LinearSystem::setSolution
void setSolution(const NumericVector< Number > &soln)
LinearAssemblySegregatedSolve::_rc_uo
RhieChowMassFlux * _rc_uo
Pointer to the segregated RhieChow interpolation object.
Definition: LinearAssemblySegregatedSolve.h:109
SIMPLESolveBase::_pressure_variable_relaxation
const Real _pressure_variable_relaxation
The user-defined relaxation parameter for the pressure variable.
Definition: SIMPLESolveBase.h:113
RhieChowMassFlux::computeHbyA
void computeHbyA(const bool with_updated_pressure, const bool verbose)
Computes the inverse of the diagonal (1/A) of the system matrix plus the H/A components for the press...
Definition: RhieChowMassFlux.C:491
LinearAssemblySegregatedSolve::_pressure_system
LinearSystem & _pressure_system
Reference to the nonlinear system corresponding to the pressure equation.
Definition: LinearAssemblySegregatedSolve.h:85
RhieChowMassFlux::computeCellVelocity
void computeCellVelocity()
Update the cell values of the velocity variables.
Definition: RhieChowMassFlux.C:355
Moose::PetscSupport::petscSetOptions
void petscSetOptions(const PetscOptions &po, const SolverParams &solver_params, FEProblemBase *const problem=nullptr)
LinearAssemblySegregatedSolve::solvePressureCorrector
virtual std::pair< unsigned int, Real > solvePressureCorrector() override
Solve a pressure corrector step.
Definition: LinearAssemblySegregatedSolve.C:287
libMesh::System::current_local_solution
std::unique_ptr< NumericVector< Number > > current_local_solution
LinearSystem::solutionPreviousNewton
virtual const NumericVector< Number > * solutionPreviousNewton() const
NS::FV::relaxSolutionUpdate
void relaxSolutionUpdate(NumericVector< Number > &vec_new, const NumericVector< Number > &vec_old, const Real relaxation_factor)
Relax the update on a solution field using the following approach: $u = u_{old}+ (u - u_{old})$...
Definition: SegregatedSolverUtils.C:112
SIMPLESolveBase::_pressure_petsc_options
Moose::PetscSupport::PetscOptions _pressure_petsc_options
Options which hold the petsc settings for the pressure equation.
Definition: SIMPLESolveBase.h:110
LinearSystem::system
virtual System & system() override

◆ linkRhieChowUserObject()

void LinearAssemblySegregatedSolve::linkRhieChowUserObject ( )
overridevirtual

Fetch the Rhie Chow user object that is reponsible for determining face velocities and mass flux.

Implements SIMPLESolveBase.

Definition at line 174 of file LinearAssemblySegregatedSolve.C.

Referenced by SIMPLE::init(), and PIMPLE::init().

175 {
176  _rc_uo =
177  const_cast<RhieChowMassFlux *>(&getUserObject<RhieChowMassFlux>("rhie_chow_user_object"));
178  _rc_uo->linkMomentumPressureSystems(
179  _momentum_systems, _pressure_system, _momentum_system_numbers);
180 
181  // Initialize the face velocities in the RC object
182  if (!_app.isRecovering())
183  _rc_uo->initFaceMassFlux();
184  _rc_uo->initCouplingField();
185 }
RhieChowMassFlux
User object responsible for determining the face fluxes using the Rhie-Chow interpolation in a segreg...
Definition: RhieChowMassFlux.h:36
LinearAssemblySegregatedSolve::_rc_uo
RhieChowMassFlux * _rc_uo
Pointer to the segregated RhieChow interpolation object.
Definition: LinearAssemblySegregatedSolve.h:109
LinearAssemblySegregatedSolve::_pressure_system
LinearSystem & _pressure_system
Reference to the nonlinear system corresponding to the pressure equation.
Definition: LinearAssemblySegregatedSolve.h:85
LinearAssemblySegregatedSolve::_momentum_system_numbers
std::vector< unsigned int > _momentum_system_numbers
The number(s) of the system(s) corresponding to the momentum equation(s)
Definition: LinearAssemblySegregatedSolve.h:76
RhieChowMassFlux::linkMomentumPressureSystems
void linkMomentumPressureSystems(const std::vector< LinearSystem *> &momentum_systems, const LinearSystem &pressure_system, const std::vector< unsigned int > &momentum_system_numbers)
Update the momentum system-related information.
Definition: RhieChowMassFlux.C:111
SolveObject::_app
MooseApp & _app
LinearAssemblySegregatedSolve::_momentum_systems
std::vector< LinearSystem * > _momentum_systems
Pointer(s) to the system(s) corresponding to the momentum equation(s)
Definition: LinearAssemblySegregatedSolve.h:79
MooseApp::isRecovering
bool isRecovering() const
RhieChowMassFlux::initCouplingField
void initCouplingField()
Initialize the coupling fields (HbyA and Ainv)
Definition: RhieChowMassFlux.C:375
RhieChowMassFlux::initFaceMassFlux
void initFaceMassFlux()
Initialize the container for face velocities.
Definition: RhieChowMassFlux.C:198

◆ setInnerSolve()

virtual void SIMPLESolveBase::setInnerSolve ( SolveObject &  )
inlineoverridevirtualinherited

Reimplemented from SolveObject.

Definition at line 48 of file SIMPLESolveBase.h.

49  {
50  mooseError("Cannot set inner solve object for solves that inherit from SIMPLESolveBase");
51  }
SolveObject::mooseError
void mooseError(Args &&... args) const

◆ setupPressurePin()

void SIMPLESolveBase::setupPressurePin ( )
inherited

Setup pressure pin if there is need for one.

Definition at line 591 of file SIMPLESolveBase.C.

Referenced by SIMPLE::init(), SIMPLENonlinearAssembly::init(), and PIMPLE::init().

592 {
593  if (_pin_pressure)
594  _pressure_pin_dof = NS::FV::findPointDoFID(_problem.getVariable(0, "pressure"),
595  _problem.mesh(),
596  getParam<Point>("pressure_pin_point"));
597 }
SolveObject::_problem
FEProblemBase & _problem
FEProblemBase::getVariable
virtual const MooseVariableFieldBase & getVariable(const THREAD_ID tid, const std::string &var_name, Moose::VarKindType expected_var_type=Moose::VarKindType::VAR_ANY, Moose::VarFieldType expected_var_field_type=Moose::VarFieldType::VAR_FIELD_ANY) const override
SIMPLESolveBase::_pressure_pin_dof
dof_id_type _pressure_pin_dof
The dof ID where the pressure needs to be pinned.
Definition: SIMPLESolveBase.h:122
SIMPLESolveBase::_pin_pressure
const bool _pin_pressure
If the pressure needs to be pinned.
Definition: SIMPLESolveBase.h:116
NS::FV::findPointDoFID
dof_id_type findPointDoFID(const MooseVariableFieldBase &variable, const MooseMesh &mesh, const Point &point)
Find the ID of the degree of freedom which corresponds to the variable and a given point on the mesh...
Definition: SegregatedSolverUtils.C:197
FEProblemBase::mesh
virtual MooseMesh & mesh() override

◆ solve()

bool LinearAssemblySegregatedSolve::solve ( )
overridevirtual

Performs the momentum pressure coupling.

Returns
True if solver is converged.

Implements SolveObject.

Definition at line 540 of file LinearAssemblySegregatedSolve.C.

541 {
542  // Do not solve if problem is set not to
543  if (!_problem.shouldSolve())
544  return true;
545 
546  // Dummy solver parameter file which is needed for switching petsc options
547  SolverParams solver_params;
548  solver_params._type = Moose::SolveType::ST_LINEAR;
549  solver_params._line_search = Moose::LineSearchType::LS_NONE;
550 
551  // Initialize the SIMPLE iteration counter
552  unsigned int simple_iteration_counter = 0;
553 
554  // Assign residuals to general residual vector
555  const unsigned int no_systems = _momentum_systems.size() + 1 + _has_energy_system +
556  _has_solid_energy_system + _active_scalar_systems.size() +
557  _turbulence_systems.size();
558 
559  std::vector<std::pair<unsigned int, Real>> ns_residuals(no_systems, std::make_pair(0, 1.0));
560  std::vector<Real> ns_abs_tols(_momentum_systems.size(), _momentum_absolute_tolerance);
561  ns_abs_tols.push_back(_pressure_absolute_tolerance);
562 
563  // Push back energy tolerances
564  if (_has_energy_system)
565  ns_abs_tols.push_back(_energy_absolute_tolerance);
566  if (_has_solid_energy_system)
567  ns_abs_tols.push_back(_solid_energy_absolute_tolerance);
568  if (_has_active_scalar_systems)
569  for (const auto scalar_tol : _active_scalar_absolute_tolerance)
570  ns_abs_tols.push_back(scalar_tol);
571 
572  // Push back turbulence tolerances
573  if (_has_turbulence_systems)
574  for (const auto turbulence_tol : _turbulence_absolute_tolerance)
575  ns_abs_tols.push_back(turbulence_tol);
576 
577  bool converged = false;
578  // Loop until converged or hit the maximum allowed iteration number
579  while (simple_iteration_counter < _num_iterations && !converged)
580  {
581  simple_iteration_counter++;
582 
583  // We set the preconditioner/controllable parameters through petsc options. Linear
584  // tolerances will be overridden within the solver. In case of a segregated momentum
585  // solver, we assume that every velocity component uses the same preconditioner
586  Moose::PetscSupport::petscSetOptions(_momentum_petsc_options, solver_params);
587 
588  // Initialize pressure gradients, after this we just reuse the last ones from each
589  // iteration
590  if (simple_iteration_counter == 1)
591  _pressure_system.computeGradients();
592 
593  _console << "Iteration " << simple_iteration_counter << " Initial residual norms:" << std::endl;
594 
595  // Solve the momentum predictor step
596  auto momentum_residual = solveMomentumPredictor();
597  for (const auto system_i : index_range(momentum_residual))
598  ns_residuals[system_i] = momentum_residual[system_i];
599 
600  // Now we correct the velocity, this function depends on the method, it differs for
601  // SIMPLE/PIMPLE, this returns the pressure errors
602  ns_residuals[momentum_residual.size()] = correctVelocity(true, true, solver_params);
603 
604  // If we have an energy equation, solve it here.We assume the material properties in the
605  // Navier-Stokes equations depend on temperature, therefore we can not solve for temperature
606  // outside of the velocity-pressure loop
607  if (_has_energy_system)
608  {
609  // We set the preconditioner/controllable parameters through petsc options. Linear
610  // tolerances will be overridden within the solver.
611  Moose::PetscSupport::petscSetOptions(_energy_petsc_options, solver_params);
612  ns_residuals[momentum_residual.size() + _has_energy_system] =
613  solveAdvectedSystem(_energy_sys_number,
614  *_energy_system,
615  _energy_equation_relaxation,
616  _energy_linear_control,
617  _energy_l_abs_tol);
618  }
619  if (_has_solid_energy_system)
620  {
621  // We set the preconditioner/controllable parameters through petsc options. Linear
622  // tolerances will be overridden within the solver.
623  Moose::PetscSupport::petscSetOptions(_solid_energy_petsc_options, solver_params);
624  ns_residuals[momentum_residual.size() + _has_solid_energy_system + _has_energy_system] =
625  solveSolidEnergy();
626  }
627 
628  // If we have active scalar equations, solve them here in case they depend on temperature
629  // or they affect the fluid properties such that they must be solved concurrently with pressure
630  // and velocity
631  if (_has_active_scalar_systems)
632  {
633  _problem.execute(EXEC_NONLINEAR);
634 
635  // We set the preconditioner/controllable parameters through petsc options. Linear
636  // tolerances will be overridden within the solver.
637  Moose::PetscSupport::petscSetOptions(_active_scalar_petsc_options, solver_params);
638  for (const auto i : index_range(_active_scalar_system_names))
639  ns_residuals[momentum_residual.size() + 1 + _has_energy_system + _has_solid_energy_system +
640  i] = solveAdvectedSystem(_active_scalar_system_numbers[i],
641  *_active_scalar_systems[i],
642  _active_scalar_equation_relaxation[i],
643  _active_scalar_linear_control,
644  _active_scalar_l_abs_tol);
645  }
646 
647  // If we have turbulence equations, solve them here.
648  // The turbulent viscosity depends on the value of the turbulence surrogate variables
649  if (_has_turbulence_systems)
650  {
651  // We set the preconditioner/controllable parameters through petsc options. Linear
652  // tolerances will be overridden within the solver.
653  Moose::PetscSupport::petscSetOptions(_turbulence_petsc_options, solver_params);
654  for (const auto i : index_range(_turbulence_system_names))
655  {
656  ns_residuals[momentum_residual.size() + 1 + _has_energy_system + _has_solid_energy_system +
657  _active_scalar_system_names.size() + i] =
658  solveAdvectedSystem(_turbulence_system_numbers[i],
659  *_turbulence_systems[i],
660  _turbulence_equation_relaxation[i],
661  _turbulence_linear_control,
662  _turbulence_l_abs_tol,
663  _turbulence_field_relaxation[i],
664  _turbulence_field_min_limit[i]);
665  }
666  }
667 
668  _problem.execute(EXEC_NONLINEAR);
669 
670  converged = NS::FV::converged(ns_residuals, ns_abs_tols);
671  }
672 
673  // If we have passive scalar equations, solve them here. We assume the material properties in the
674  // Navier-Stokes equations do not depend on passive scalars, as they are passive, therefore we
675  // solve outside of the velocity-pressure loop
676  if (_has_passive_scalar_systems && (converged || _continue_on_max_its))
677  {
678  // The reason why we need more than one iteration is due to the matrix relaxation
679  // which can be used to stabilize the equations
680  bool passive_scalar_converged = false;
681  unsigned int ps_iteration_counter = 0;
682 
683  _console << "Passive scalar iteration " << ps_iteration_counter
684  << " Initial residual norms:" << std::endl;
685 
686  while (ps_iteration_counter < _num_iterations && !passive_scalar_converged)
687  {
688  ps_iteration_counter++;
689  std::vector<std::pair<unsigned int, Real>> scalar_residuals(
690  _passive_scalar_system_names.size(), std::make_pair(0, 1.0));
691  std::vector<Real> scalar_abs_tols;
692  for (const auto scalar_tol : _passive_scalar_absolute_tolerance)
693  scalar_abs_tols.push_back(scalar_tol);
694 
695  // We set the preconditioner/controllable parameters through petsc options. Linear
696  // tolerances will be overridden within the solver.
697  Moose::PetscSupport::petscSetOptions(_passive_scalar_petsc_options, solver_params);
698  for (const auto i : index_range(_passive_scalar_system_names))
699  scalar_residuals[i] = solveAdvectedSystem(_passive_scalar_system_numbers[i],
700  *_passive_scalar_systems[i],
701  _passive_scalar_equation_relaxation[i],
702  _passive_scalar_linear_control,
703  _passive_scalar_l_abs_tol);
704 
705  passive_scalar_converged = NS::FV::converged(scalar_residuals, scalar_abs_tols);
706  }
707 
708  // Both flow and scalars must converge
709  converged = passive_scalar_converged && converged;
710  }
711 
712  converged = _continue_on_max_its ? true : converged;
713 
714  return converged;
715 }
FEProblemBase::shouldSolve
bool shouldSolve() const
LinearAssemblySegregatedSolve::_active_scalar_equation_relaxation
const std::vector< Real > _active_scalar_equation_relaxation
The user-defined relaxation parameter(s) for the active scalar equation(s)
Definition: LinearAssemblySegregatedSolve.h:126
SolveObject::_problem
FEProblemBase & _problem
SIMPLESolveBase::_turbulence_linear_control
SIMPLESolverConfiguration _turbulence_linear_control
Options for the linear solver of the turbulence equation(s)
Definition: SIMPLESolveBase.h:205
SIMPLESolveBase::_energy_absolute_tolerance
const Real _energy_absolute_tolerance
The user-defined absolute tolerance for determining the convergence in energy.
Definition: SIMPLESolveBase.h:220
SIMPLESolveBase::_num_iterations
const unsigned int _num_iterations
The maximum number of momentum-pressure iterations.
Definition: SIMPLESolveBase.h:232
SIMPLESolveBase::_energy_linear_control
SIMPLESolverConfiguration _energy_linear_control
Options for the linear solver of the energy equation.
Definition: SIMPLESolveBase.h:136
LinearSystem::computeGradients
void computeGradients()
SIMPLESolveBase::_turbulence_petsc_options
Moose::PetscSupport::PetscOptions _turbulence_petsc_options
Options which hold the petsc settings for the turbulence equation(s)
Definition: SIMPLESolveBase.h:202
SIMPLESolveBase::_has_energy_system
const bool _has_energy_system
Boolean for easy check if a fluid energy system shall be solved or not.
Definition: SIMPLESolveBase.h:127
SolverParams::_line_search
Moose::LineSearchType _line_search
LinearAssemblySegregatedSolve::_active_scalar_linear_control
SIMPLESolverConfiguration _active_scalar_linear_control
Options for the linear solver of the active scalar equation(s)
Definition: LinearAssemblySegregatedSolve.h:132
SIMPLESolveBase::_turbulence_field_relaxation
std::vector< Real > _turbulence_field_relaxation
The user-defined relaxation parameter(s) for the turbulence field(s)
Definition: SIMPLESolveBase.h:196
SIMPLESolveBase::_passive_scalar_l_abs_tol
const Real _passive_scalar_l_abs_tol
Absolute linear tolerance for the passive scalar equation(s).
Definition: SIMPLESolveBase.h:179
LinearAssemblySegregatedSolve::solveSolidEnergy
std::pair< unsigned int, Real > solveSolidEnergy()
Solve an equation which contains the solid energy conservation.
Definition: LinearAssemblySegregatedSolve.C:351
LinearAssemblySegregatedSolve::_passive_scalar_systems
std::vector< LinearSystem * > _passive_scalar_systems
Pointer(s) to the system(s) corresponding to the passive scalar equation(s)
Definition: LinearAssemblySegregatedSolve.h:100
SIMPLESolveBase::_solid_energy_petsc_options
Moose::PetscSupport::PetscOptions _solid_energy_petsc_options
Options which hold the petsc settings for the fluid energy equation.
Definition: SIMPLESolveBase.h:148
SIMPLESolveBase::_has_turbulence_systems
const bool _has_turbulence_systems
Boolean for easy check if a turbulence scalar systems shall be solved or not.
Definition: SIMPLESolveBase.h:187
SIMPLESolveBase::_turbulence_system_numbers
std::vector< unsigned int > _turbulence_system_numbers
Definition: SIMPLESolveBase.h:190
LinearAssemblySegregatedSolve::solveAdvectedSystem
std::pair< unsigned int, Real > solveAdvectedSystem(const unsigned int system_num, LinearSystem &system, const Real relaxation_factor, libMesh::SolverConfiguration &solver_config, const Real abs_tol, const Real field_relaxation=1.0, const Real min_value_limiter=std::numeric_limits< Real >::min())
Solve an equation which contains an advection term that depends on the solution of the segregated Nav...
Definition: LinearAssemblySegregatedSolve.C:453
LinearAssemblySegregatedSolve::_has_active_scalar_systems
const bool _has_active_scalar_systems
Boolean for easy check if a active scalar systems shall be solved or not.
Definition: LinearAssemblySegregatedSolve.h:120
LinearAssemblySegregatedSolve::_active_scalar_l_abs_tol
const Real _active_scalar_l_abs_tol
Absolute linear tolerance for the active scalar equation(s).
Definition: LinearAssemblySegregatedSolve.h:136
LinearAssemblySegregatedSolve::correctVelocity
virtual std::pair< unsigned int, Real > correctVelocity(const bool subtract_updated_pressure, const bool recompute_face_mass_flux, const SolverParams &solver_params)
Computes new velocity field based on computed pressure gradients.
Definition: LinearAssemblySegregatedSolve.C:411
NS::FV::converged
bool converged(const std::vector< std::pair< unsigned int, Real >> &residuals, const std::vector< Real > &abs_tolerances)
Based on the residuals, determine if the iterative process converged or not.
Definition: SegregatedSolverUtils.C:229
SIMPLESolveBase::_pressure_absolute_tolerance
const Real _pressure_absolute_tolerance
The user-defined absolute tolerance for determining the convergence in pressure.
Definition: SIMPLESolveBase.h:217
SIMPLESolveBase::_momentum_absolute_tolerance
const Real _momentum_absolute_tolerance
The user-defined absolute tolerance for determining the convergence in momentum.
Definition: SIMPLESolveBase.h:214
SIMPLESolveBase::_passive_scalar_system_names
const std::vector< SolverSystemName > & _passive_scalar_system_names
The names of the passive scalar systems.
Definition: SIMPLESolveBase.h:160
LinearAssemblySegregatedSolve::_active_scalar_systems
std::vector< LinearSystem * > _active_scalar_systems
Pointer(s) to the system(s) corresponding to the active scalar equation(s)
Definition: LinearAssemblySegregatedSolve.h:103
SIMPLESolveBase::_passive_scalar_linear_control
SIMPLESolverConfiguration _passive_scalar_linear_control
Options for the linear solver of the passive scalar equation(s)
Definition: SIMPLESolveBase.h:175
FEProblemBase::execute
virtual void execute(const ExecFlagType &exec_type)
SIMPLESolveBase::_energy_petsc_options
Moose::PetscSupport::PetscOptions _energy_petsc_options
Options which hold the petsc settings for the fluid energy equation.
Definition: SIMPLESolveBase.h:133
LinearAssemblySegregatedSolve::_pressure_system
LinearSystem & _pressure_system
Reference to the nonlinear system corresponding to the pressure equation.
Definition: LinearAssemblySegregatedSolve.h:85
LinearAssemblySegregatedSolve::_energy_system
LinearSystem * _energy_system
Pointer to the nonlinear system corresponding to the fluid energy equation.
Definition: LinearAssemblySegregatedSolve.h:91
LinearAssemblySegregatedSolve::_energy_sys_number
const unsigned int _energy_sys_number
The number of the system corresponding to the energy equation.
Definition: LinearAssemblySegregatedSolve.h:88
SIMPLESolveBase::_passive_scalar_petsc_options
Moose::PetscSupport::PetscOptions _passive_scalar_petsc_options
Options which hold the petsc settings for the passive scalar equation(s)
Definition: SIMPLESolveBase.h:172
SIMPLESolveBase::_momentum_petsc_options
Moose::PetscSupport::PetscOptions _momentum_petsc_options
Options which hold the petsc settings for the momentum equation.
Definition: SIMPLESolveBase.h:92
SIMPLESolveBase::_passive_scalar_absolute_tolerance
const std::vector< Real > _passive_scalar_absolute_tolerance
The user-defined absolute tolerance for determining the convergence in passive scalars.
Definition: SIMPLESolveBase.h:226
SolverParams::_type
Moose::SolveType _type
SIMPLESolveBase::_turbulence_system_names
const std::vector< SolverSystemName > & _turbulence_system_names
The names of the turbulence systems.
Definition: SIMPLESolveBase.h:184
SIMPLESolveBase::_energy_l_abs_tol
const Real _energy_l_abs_tol
Absolute linear tolerance for the energy equations.
Definition: SIMPLESolveBase.h:140
Moose::PetscSupport::petscSetOptions
void petscSetOptions(const PetscOptions &po, const SolverParams &solver_params, FEProblemBase *const problem=nullptr)
SIMPLESolveBase::_turbulence_equation_relaxation
const std::vector< Real > _turbulence_equation_relaxation
The user-defined relaxation parameter(s) for the turbulence equation(s)
Definition: SIMPLESolveBase.h:193
LinearAssemblySegregatedSolve::_turbulence_systems
std::vector< LinearSystem * > _turbulence_systems
Pointer(s) to the system(s) corresponding to the turbulence equation(s)
Definition: LinearAssemblySegregatedSolve.h:106
SIMPLESolveBase::_passive_scalar_equation_relaxation
const std::vector< Real > _passive_scalar_equation_relaxation
The user-defined relaxation parameter(s) for the passive scalar equation(s)
Definition: SIMPLESolveBase.h:169
LinearAssemblySegregatedSolve::_active_scalar_system_names
const std::vector< SolverSystemName > & _active_scalar_system_names
The names of the active scalar systems.
Definition: LinearAssemblySegregatedSolve.h:117
SIMPLESolveBase::_energy_equation_relaxation
const Real _energy_equation_relaxation
The user-defined relaxation parameter for the energy equation.
Definition: SIMPLESolveBase.h:130
LinearAssemblySegregatedSolve::solveMomentumPredictor
virtual std::vector< std::pair< unsigned int, Real > > solveMomentumPredictor() override
Solve a momentum predictor step with a fixed pressure field.
Definition: LinearAssemblySegregatedSolve.C:188
SIMPLESolveBase::_turbulence_l_abs_tol
const Real _turbulence_l_abs_tol
Absolute linear tolerance for the turbulence equation(s).
Definition: SIMPLESolveBase.h:209
LinearAssemblySegregatedSolve::_active_scalar_petsc_options
Moose::PetscSupport::PetscOptions _active_scalar_petsc_options
Options which hold the petsc settings for the active scalar equation(s)
Definition: LinearAssemblySegregatedSolve.h:129
SIMPLESolveBase::_has_passive_scalar_systems
const bool _has_passive_scalar_systems
Boolean for easy check if a passive scalar systems shall be solved or not.
Definition: SIMPLESolveBase.h:163
LinearAssemblySegregatedSolve::_active_scalar_absolute_tolerance
const std::vector< Real > _active_scalar_absolute_tolerance
The user-defined absolute tolerance for determining the convergence in active scalars.
Definition: LinearAssemblySegregatedSolve.h:139
SolveObject::_console
const ConsoleStream _console
SIMPLESolveBase::_solid_energy_absolute_tolerance
const Real _solid_energy_absolute_tolerance
The user-defined absolute tolerance for determining the convergence in solid energy.
Definition: SIMPLESolveBase.h:223
LinearAssemblySegregatedSolve::_active_scalar_system_numbers
std::vector< unsigned int > _active_scalar_system_numbers
Definition: LinearAssemblySegregatedSolve.h:123
LinearAssemblySegregatedSolve::_momentum_systems
std::vector< LinearSystem * > _momentum_systems
Pointer(s) to the system(s) corresponding to the momentum equation(s)
Definition: LinearAssemblySegregatedSolve.h:79
SIMPLESolveBase::_continue_on_max_its
const bool _continue_on_max_its
If solve should continue if maximum number of iterations is hit.
Definition: SIMPLESolveBase.h:235
SIMPLESolveBase::_turbulence_field_min_limit
std::vector< Real > _turbulence_field_min_limit
The user-defined lower limit for turbulent quantities e.g. k, eps/omega, etc..
Definition: SIMPLESolveBase.h:199
SIMPLESolveBase::_turbulence_absolute_tolerance
const std::vector< Real > _turbulence_absolute_tolerance
The user-defined absolute tolerance for determining the convergence turbulence variables.
Definition: SIMPLESolveBase.h:229
index_range
auto index_range(const T &sizable)
SIMPLESolveBase::_passive_scalar_system_numbers
std::vector< unsigned int > _passive_scalar_system_numbers
Definition: SIMPLESolveBase.h:166
SIMPLESolveBase::_has_solid_energy_system
const bool _has_solid_energy_system
Boolean for easy check if a solid energy system shall be solved or not.
Definition: SIMPLESolveBase.h:145

◆ solveAdvectedSystem()

std::pair< unsigned int, Real > LinearAssemblySegregatedSolve::solveAdvectedSystem ( const unsigned int  system_num,
LinearSystem &  system,
const Real  relaxation_factor,
libMesh::SolverConfiguration &  solver_config,
const Real  abs_tol,
const Real  field_relaxation = 1.0,
const Real  min_value_limiter = std::numeric_limits<Real>::min() 
)
protected

Solve an equation which contains an advection term that depends on the solution of the segregated Navier-Stokes equations.

Parameters
system_numThe number of the system which is solved
systemReference to the system which is solved
relaxation_factorThe relaxation factor for matrix relaxation
solver_configThe solver configuration object for the linear solve
abs_tolThe scaled absolute tolerance for the linear solve
field_relaxation(optional) The relaxation factor for fields if relax_fields is true. Default value is 1.0.
min_value_limiter(optional) The minimum value for the solution field
Returns
The normalized residual norm of the equation.

Definition at line 453 of file LinearAssemblySegregatedSolve.C.

Referenced by solve().

460 {
461  _problem.setCurrentLinearSystem(system_num);
462 
463  // We will need some members from the implicit linear system
464  LinearImplicitSystem & li_system = libMesh::cast_ref<LinearImplicitSystem &>(system.system());
465 
466  // We will need the solution, the right hand side and the matrix
467  NumericVector<Number> & current_local_solution = *(li_system.current_local_solution);
468  NumericVector<Number> & solution = *(li_system.solution);
469  SparseMatrix<Number> & mmat = *(li_system.matrix);
470  NumericVector<Number> & rhs = *(li_system.rhs);
471 
472  // We need a vector that stores the (diagonal_relaxed-original_diagonal) vector
473  auto diff_diagonal = solution.zero_clone();
474 
475  // Fetch the linear solver from the system
476  PetscLinearSolver<Real> & linear_solver =
477  libMesh::cast_ref<PetscLinearSolver<Real> &>(*li_system.get_linear_solver());
478 
479  _problem.computeLinearSystemSys(li_system, mmat, rhs, true);
480 
481  // Go and relax the system matrix and the right hand side
482  NS::FV::relaxMatrix(mmat, relaxation_factor, *diff_diagonal);
483  NS::FV::relaxRightHandSide(rhs, solution, *diff_diagonal);
484 
485  if (_print_fields)
486  {
487  _console << system.name() << " system matrix" << std::endl;
488  mmat.print();
489  }
490 
491  // We compute the normalization factors based on the fluxes
492  Real norm_factor = NS::FV::computeNormalizationFactor(solution, mmat, rhs);
493 
494  // We need the non-preconditioned norm to be consistent with the norm factor
495  LibmeshPetscCall(KSPSetNormType(linear_solver.ksp(), KSP_NORM_UNPRECONDITIONED));
496 
497  // Setting the linear tolerances and maximum iteration counts
498  solver_config.real_valued_data["abs_tol"] = absolute_tol * norm_factor;
499  linear_solver.set_solver_configuration(solver_config);
500 
501  // Solve the system and update current local solution
502  auto its_res_pair = linear_solver.solve(mmat, mmat, solution, rhs);
503  li_system.update();
504 
505  if (_print_fields)
506  {
507  _console << " rhs when we solve " << system.name() << std::endl;
508  rhs.print();
509  _console << system.name() << " solution " << std::endl;
510  solution.print();
511  _console << " Norm factor " << norm_factor << std::endl;
512  }
513 
514  // Limiting scalar solution
515  if (min_value_limiter != std::numeric_limits<Real>::min())
516  NS::FV::limitSolutionUpdate(current_local_solution, min_value_limiter);
517 
518  // Relax the field update for the next momentum predictor
519  if (field_relaxation != 1.0)
520  {
521  auto & old_local_solution = *(system.solutionPreviousNewton());
522  NS::FV::relaxSolutionUpdate(current_local_solution, old_local_solution, field_relaxation);
523 
524  // Update old solution, only needed if relaxing the field
525  old_local_solution = current_local_solution;
526  }
527 
528  system.setSolution(current_local_solution);
529 
530  const auto residuals =
531  std::make_pair(its_res_pair.first, linear_solver.get_initial_residual() / norm_factor);
532 
533  _console << " Advected system: " << system.name() << " " << COLOR_GREEN << residuals.second
534  << COLOR_DEFAULT << " Linear its: " << residuals.first << std::endl;
535 
536  return residuals;
537 }
NS::FV::relaxMatrix
void relaxMatrix(SparseMatrix< Number > &matrix_in, const Real relaxation_parameter, NumericVector< Number > &diff_diagonal)
Relax the matrix to ensure diagonal dominance, we hold onto the difference in diagonals for later use...
Definition: SegregatedSolverUtils.C:25
SolveObject::_problem
FEProblemBase & _problem
SIMPLESolveBase::_print_fields
const bool _print_fields
Debug parameter which allows printing the coupling and solution vectors/matrices. ...
Definition: SIMPLESolveBase.h:240
NS::FV::computeNormalizationFactor
Real computeNormalizationFactor(const NumericVector< Number > &solution, const SparseMatrix< Number > &mat, const NumericVector< Number > &rhs)
Compute a normalization factor which is applied to the linear residual to determine convergence...
Definition: SegregatedSolverUtils.C:135
libMesh::NumericVector::zero_clone
virtual std::unique_ptr< NumericVector< T > > zero_clone() const=0
libMesh::LinearImplicitSystem::rhs
NumericVector< Number > * rhs
LinearSystem::setSolution
void setSolution(const NumericVector< Number > &soln)
libMesh::PetscLinearSolver::solve
virtual std::pair< unsigned int, Real > solve(SparseMatrix< T > &matrix_in, NumericVector< T > &solution_in, NumericVector< T > &rhs_in, const std::optional< double > tol=std::nullopt, const std::optional< unsigned int > m_its=std::nullopt) override
libMesh::LinearImplicitSystem::get_linear_solver
virtual LinearSolver< Number > * get_linear_solver() const override
NS::FV::relaxRightHandSide
void relaxRightHandSide(NumericVector< Number > &rhs_in, const NumericVector< Number > &solution_in, const NumericVector< Number > &diff_diagonal)
Relax the right hand side of an equation, this needs to be called once and the system matrix has been...
Definition: SegregatedSolverUtils.C:91
libMesh::SolverConfiguration::real_valued_data
std::map< std::string, Real > real_valued_data
FEProblemBase::computeLinearSystemSys
virtual void computeLinearSystemSys(libMesh::LinearImplicitSystem &sys, libMesh::SparseMatrix< libMesh::Number > &system_matrix, NumericVector< libMesh::Number > &rhs, const bool compute_gradients=true)
LinearSystem::name
virtual const std::string & name() const
libMesh::LinearImplicitSystem::solution
std::unique_ptr< NumericVector< Number > > solution
libMesh::NumericVector::print
virtual void print(std::ostream &os=libMesh::out) const
libMesh::PetscLinearSolver::set_solver_configuration
void set_solver_configuration(SolverConfiguration &solver_configuration)
Real
DIE A HORRIBLE DEATH HERE typedef LIBMESH_DEFAULT_SCALAR_TYPE Real
libMesh::LinearImplicitSystem::matrix
SparseMatrix< Number > * matrix
FEProblemBase::setCurrentLinearSystem
void setCurrentLinearSystem(unsigned int sys_num)
libMesh::LinearImplicitSystem::current_local_solution
std::unique_ptr< NumericVector< Number > > current_local_solution
LinearSystem::solutionPreviousNewton
virtual const NumericVector< Number > * solutionPreviousNewton() const
SolveObject::_console
const ConsoleStream _console
NS::FV::limitSolutionUpdate
void limitSolutionUpdate(NumericVector< Number > &solution, const Real min_limit=std::numeric_limits< Real >::epsilon(), const Real max_limit=1e10)
Limit a solution to its minimum and maximum bounds: $u = min(max(u, min_limit), max_limit)$.
Definition: SegregatedSolverUtils.C:123
NS::FV::relaxSolutionUpdate
void relaxSolutionUpdate(NumericVector< Number > &vec_new, const NumericVector< Number > &vec_old, const Real relaxation_factor)
Relax the update on a solution field using the following approach: $u = u_{old}+ (u - u_{old})$...
Definition: SegregatedSolverUtils.C:112
libMesh::SparseMatrix::print
void print(std::ostream &os=libMesh::out, const bool sparse=false) const
LinearSystem::system
virtual System & system() override

◆ solveMomentumPredictor()

std::vector< std::pair< unsigned int, Real > > LinearAssemblySegregatedSolve::solveMomentumPredictor ( )
overrideprotectedvirtual

Solve a momentum predictor step with a fixed pressure field.

Returns
A vector of (number of linear iterations, normalized residual norm) pairs for the momentum equations. The length of the vector equals the dimensionality of the domain.

Implements SIMPLESolveBase.

Definition at line 188 of file LinearAssemblySegregatedSolve.C.

Referenced by solve().

189 {
190  // Temporary storage for the (flux-normalized) residuals from
191  // different momentum components
192  std::vector<std::pair<unsigned int, Real>> its_normalized_residuals;
193 
194  LinearImplicitSystem & momentum_system_0 =
195  libMesh::cast_ref<LinearImplicitSystem &>(_momentum_systems[0]->system());
196 
197  PetscLinearSolver<Real> & momentum_solver =
198  libMesh::cast_ref<PetscLinearSolver<Real> &>(*momentum_system_0.get_linear_solver());
199 
200  // Solve the momentum equations.
201  // TO DO: These equations are VERY similar. If we can store the differences (things coming from
202  // BCs for example) separately, it is enough to construct one matrix.
203  for (const auto system_i : index_range(_momentum_systems))
204  {
205  _problem.setCurrentLinearSystem(_momentum_system_numbers[system_i]);
206 
207  // We will need the right hand side and the solution of the next component
208  LinearImplicitSystem & momentum_system =
209  libMesh::cast_ref<LinearImplicitSystem &>(_momentum_systems[system_i]->system());
210 
211  NumericVector<Number> & solution = *(momentum_system.solution);
212  NumericVector<Number> & rhs = *(momentum_system.rhs);
213  SparseMatrix<Number> & mmat = *(momentum_system.matrix);
214 
215  auto diff_diagonal = solution.zero_clone();
216 
217  // We assemble the matrix and the right hand side
218  _problem.computeLinearSystemSys(momentum_system, mmat, rhs, /*compute_grads*/ true);
219 
220  // Still need to relax the right hand side with the same vector
221  NS::FV::relaxMatrix(mmat, _momentum_equation_relaxation, *diff_diagonal);
222  NS::FV::relaxRightHandSide(rhs, solution, *diff_diagonal);
223 
224  // The normalization factor depends on the right hand side so we need to recompute it for this
225  // component
226  Real norm_factor = NS::FV::computeNormalizationFactor(solution, mmat, rhs);
227 
228  // Very important, for deciding the convergence, we need the unpreconditioned
229  // norms in the linear solve
230  LibmeshPetscCall(KSPSetNormType(momentum_solver.ksp(), KSP_NORM_UNPRECONDITIONED));
231  // Solve this component. We don't update the ghosted solution yet, that will come at the end
232  // of the corrector step. Also setting the linear tolerances and maximum iteration counts.
233  _momentum_linear_control.real_valued_data["abs_tol"] = _momentum_l_abs_tol * norm_factor;
234  momentum_solver.set_solver_configuration(_momentum_linear_control);
235 
236  // We solve the equation
237  auto its_resid_pair = momentum_solver.solve(mmat, mmat, solution, rhs);
238  momentum_system.update();
239 
240  // We will reuse the preconditioner for every momentum system
241  if (system_i == 0)
242  momentum_solver.reuse_preconditioner(true);
243 
244  // Save the normalized residual
245  its_normalized_residuals.push_back(
246  std::make_pair(its_resid_pair.first, momentum_solver.get_initial_residual() / norm_factor));
247 
248  if (_print_fields)
249  {
250  _console << " solution after solve " << std::endl;
251  solution.print();
252  _console << " matrix when we solve " << std::endl;
253  mmat.print();
254  _console << " rhs when we solve " << std::endl;
255  rhs.print();
256  _console << " velocity solution component " << system_i << std::endl;
257  solution.print();
258  _console << "Norm factor " << norm_factor << std::endl;
259  _console << Moose::stringify(momentum_solver.get_initial_residual()) << std::endl;
260  }
261 
262  // Printing residuals
263  _console << " Momentum equation:"
264  << (_momentum_systems.size() > 1
265  ? std::string(" Component ") + std::to_string(system_i + 1) + std::string(" ")
266  : std::string(" "))
267  << COLOR_GREEN << its_normalized_residuals[system_i].second << COLOR_DEFAULT
268  << " Linear its: " << its_normalized_residuals[system_i].first << std::endl;
269  }
270 
271  for (const auto system_i : index_range(_momentum_systems))
272  {
273  LinearImplicitSystem & momentum_system =
274  libMesh::cast_ref<LinearImplicitSystem &>(_momentum_systems[system_i]->system());
275  _momentum_systems[system_i]->setSolution(*(momentum_system.current_local_solution));
276  _momentum_systems[system_i]->copyPreviousNonlinearSolutions();
277  }
278 
279  // We reset this to ensure the preconditioner is recomputed new time we go to the momentum
280  // predictor
281  momentum_solver.reuse_preconditioner(false);
282 
283  return its_normalized_residuals;
284 }
NS::FV::relaxMatrix
void relaxMatrix(SparseMatrix< Number > &matrix_in, const Real relaxation_parameter, NumericVector< Number > &diff_diagonal)
Relax the matrix to ensure diagonal dominance, we hold onto the difference in diagonals for later use...
Definition: SegregatedSolverUtils.C:25
SolveObject::_problem
FEProblemBase & _problem
SIMPLESolveBase::_momentum_linear_control
SIMPLESolverConfiguration _momentum_linear_control
Options for the linear solver of the momentum equation.
Definition: SIMPLESolveBase.h:85
SIMPLESolveBase::_print_fields
const bool _print_fields
Debug parameter which allows printing the coupling and solution vectors/matrices. ...
Definition: SIMPLESolveBase.h:240
NS::FV::computeNormalizationFactor
Real computeNormalizationFactor(const NumericVector< Number > &solution, const SparseMatrix< Number > &mat, const NumericVector< Number > &rhs)
Compute a normalization factor which is applied to the linear residual to determine convergence...
Definition: SegregatedSolverUtils.C:135
SIMPLESolveBase::_momentum_equation_relaxation
const Real _momentum_equation_relaxation
The user-defined relaxation parameter for the momentum equation.
Definition: SIMPLESolveBase.h:95
libMesh::LinearImplicitSystem::rhs
NumericVector< Number > * rhs
SIMPLESolveBase::_momentum_l_abs_tol
const Real _momentum_l_abs_tol
Absolute linear tolerance for the momentum equation(s).
Definition: SIMPLESolveBase.h:89
libMesh::LinearImplicitSystem::get_linear_solver
virtual LinearSolver< Number > * get_linear_solver() const override
NS::FV::relaxRightHandSide
void relaxRightHandSide(NumericVector< Number > &rhs_in, const NumericVector< Number > &solution_in, const NumericVector< Number > &diff_diagonal)
Relax the right hand side of an equation, this needs to be called once and the system matrix has been...
Definition: SegregatedSolverUtils.C:91
libMesh::SolverConfiguration::real_valued_data
std::map< std::string, Real > real_valued_data
FEProblemBase::computeLinearSystemSys
virtual void computeLinearSystemSys(libMesh::LinearImplicitSystem &sys, libMesh::SparseMatrix< libMesh::Number > &system_matrix, NumericVector< libMesh::Number > &rhs, const bool compute_gradients=true)
LinearAssemblySegregatedSolve::_momentum_system_numbers
std::vector< unsigned int > _momentum_system_numbers
The number(s) of the system(s) corresponding to the momentum equation(s)
Definition: LinearAssemblySegregatedSolve.h:76
libMesh::SparseMatrix::zero_clone
virtual std::unique_ptr< SparseMatrix< T > > zero_clone() const=0
libMesh::LinearImplicitSystem::solution
std::unique_ptr< NumericVector< Number > > solution
libMesh::NumericVector::print
virtual void print(std::ostream &os=libMesh::out) const
Moose::stringify
std::string stringify(const T &t)
Real
DIE A HORRIBLE DEATH HERE typedef LIBMESH_DEFAULT_SCALAR_TYPE Real
libMesh::LinearImplicitSystem::matrix
SparseMatrix< Number > * matrix
FEProblemBase::setCurrentLinearSystem
void setCurrentLinearSystem(unsigned int sys_num)
libMesh::LinearImplicitSystem::current_local_solution
std::unique_ptr< NumericVector< Number > > current_local_solution
SolveObject::_console
const ConsoleStream _console
LinearAssemblySegregatedSolve::_momentum_systems
std::vector< LinearSystem * > _momentum_systems
Pointer(s) to the system(s) corresponding to the momentum equation(s)
Definition: LinearAssemblySegregatedSolve.h:79
index_range
auto index_range(const T &sizable)
libMesh::SparseMatrix::print
void print(std::ostream &os=libMesh::out, const bool sparse=false) const

◆ solvePressureCorrector()

std::pair< unsigned int, Real > LinearAssemblySegregatedSolve::solvePressureCorrector ( )
overrideprotectedvirtual

Solve a pressure corrector step.

Returns
The number of linear iterations and the normalized residual norm of the pressure equation.

Implements SIMPLESolveBase.

Definition at line 287 of file LinearAssemblySegregatedSolve.C.

Referenced by correctVelocity().

288 {
289  _problem.setCurrentLinearSystem(_pressure_sys_number);
290 
291  // We will need some members from the linear system
292  LinearImplicitSystem & pressure_system =
293  libMesh::cast_ref<LinearImplicitSystem &>(_pressure_system.system());
294 
295  // We will need the solution, the right hand side and the matrix
296  NumericVector<Number> & current_local_solution = *(pressure_system.current_local_solution);
297  NumericVector<Number> & solution = *(pressure_system.solution);
298  SparseMatrix<Number> & mmat = *(pressure_system.matrix);
299  NumericVector<Number> & rhs = *(pressure_system.rhs);
300 
301  // Fetch the linear solver from the system
302  PetscLinearSolver<Real> & pressure_solver =
303  libMesh::cast_ref<PetscLinearSolver<Real> &>(*pressure_system.get_linear_solver());
304 
305  _problem.computeLinearSystemSys(pressure_system, mmat, rhs, false);
306 
307  if (_print_fields)
308  {
309  _console << "Pressure matrix" << std::endl;
310  mmat.print();
311  }
312 
313  // We compute the normalization factors based on the fluxes
314  Real norm_factor = NS::FV::computeNormalizationFactor(solution, mmat, rhs);
315 
316  // We need the non-preconditioned norm to be consistent with the norm factor
317  LibmeshPetscCall(KSPSetNormType(pressure_solver.ksp(), KSP_NORM_UNPRECONDITIONED));
318 
319  // Setting the linear tolerances and maximum iteration counts
320  _pressure_linear_control.real_valued_data["abs_tol"] = _pressure_l_abs_tol * norm_factor;
321  pressure_solver.set_solver_configuration(_pressure_linear_control);
322 
323  if (_pin_pressure)
324  NS::FV::constrainSystem(mmat, rhs, _pressure_pin_value, _pressure_pin_dof);
325  pressure_system.update();
326 
327  auto its_res_pair = pressure_solver.solve(mmat, mmat, solution, rhs);
328  pressure_system.update();
329 
330  if (_print_fields)
331  {
332  _console << " rhs when we solve pressure " << std::endl;
333  rhs.print();
334  _console << " Pressure " << std::endl;
335  solution.print();
336  _console << "Norm factor " << norm_factor << std::endl;
337  }
338 
339  _pressure_system.setSolution(current_local_solution);
340 
341  const auto residuals =
342  std::make_pair(its_res_pair.first, pressure_solver.get_initial_residual() / norm_factor);
343 
344  _console << " Pressure equation: " << COLOR_GREEN << residuals.second << COLOR_DEFAULT
345  << " Linear its: " << residuals.first << std::endl;
346 
347  return residuals;
348 }
SolveObject::_problem
FEProblemBase & _problem
SIMPLESolveBase::_print_fields
const bool _print_fields
Debug parameter which allows printing the coupling and solution vectors/matrices. ...
Definition: SIMPLESolveBase.h:240
NS::FV::computeNormalizationFactor
Real computeNormalizationFactor(const NumericVector< Number > &solution, const SparseMatrix< Number > &mat, const NumericVector< Number > &rhs)
Compute a normalization factor which is applied to the linear residual to determine convergence...
Definition: SegregatedSolverUtils.C:135
libMesh::LinearImplicitSystem::rhs
NumericVector< Number > * rhs
LinearSystem::setSolution
void setSolution(const NumericVector< Number > &soln)
libMesh::PetscLinearSolver::solve
virtual std::pair< unsigned int, Real > solve(SparseMatrix< T > &matrix_in, NumericVector< T > &solution_in, NumericVector< T > &rhs_in, const std::optional< double > tol=std::nullopt, const std::optional< unsigned int > m_its=std::nullopt) override
libMesh::LinearImplicitSystem::get_linear_solver
virtual LinearSolver< Number > * get_linear_solver() const override
libMesh::SolverConfiguration::real_valued_data
std::map< std::string, Real > real_valued_data
FEProblemBase::computeLinearSystemSys
virtual void computeLinearSystemSys(libMesh::LinearImplicitSystem &sys, libMesh::SparseMatrix< libMesh::Number > &system_matrix, NumericVector< libMesh::Number > &rhs, const bool compute_gradients=true)
LinearAssemblySegregatedSolve::_pressure_system
LinearSystem & _pressure_system
Reference to the nonlinear system corresponding to the pressure equation.
Definition: LinearAssemblySegregatedSolve.h:85
SIMPLESolveBase::_pressure_linear_control
SIMPLESolverConfiguration _pressure_linear_control
Options for the linear solver of the pressure equation.
Definition: SIMPLESolveBase.h:103
SIMPLESolveBase::_pressure_l_abs_tol
const Real _pressure_l_abs_tol
Absolute linear tolerance for the pressure equation.
Definition: SIMPLESolveBase.h:107
libMesh::LinearImplicitSystem::solution
std::unique_ptr< NumericVector< Number > > solution
SIMPLESolveBase::_pressure_pin_dof
dof_id_type _pressure_pin_dof
The dof ID where the pressure needs to be pinned.
Definition: SIMPLESolveBase.h:122
libMesh::NumericVector::print
virtual void print(std::ostream &os=libMesh::out) const
SIMPLESolveBase::_pin_pressure
const bool _pin_pressure
If the pressure needs to be pinned.
Definition: SIMPLESolveBase.h:116
libMesh::PetscLinearSolver::set_solver_configuration
void set_solver_configuration(SolverConfiguration &solver_configuration)
Real
DIE A HORRIBLE DEATH HERE typedef LIBMESH_DEFAULT_SCALAR_TYPE Real
libMesh::LinearImplicitSystem::matrix
SparseMatrix< Number > * matrix
FEProblemBase::setCurrentLinearSystem
void setCurrentLinearSystem(unsigned int sys_num)
SIMPLESolveBase::_pressure_pin_value
const Real _pressure_pin_value
The value we want to enforce for pressure.
Definition: SIMPLESolveBase.h:119
libMesh::LinearImplicitSystem::current_local_solution
std::unique_ptr< NumericVector< Number > > current_local_solution
SolveObject::_console
const ConsoleStream _console
NS::FV::constrainSystem
void constrainSystem(SparseMatrix< Number > &mx, NumericVector< Number > &rhs, const Real desired_value, const dof_id_type dof_id)
Implicitly constrain the system by adding a factor*(u-u_desired) to it at a desired dof value...
Definition: SegregatedSolverUtils.C:180
libMesh::SparseMatrix::print
void print(std::ostream &os=libMesh::out, const bool sparse=false) const
LinearSystem::system
virtual System & system() override
LinearAssemblySegregatedSolve::_pressure_sys_number
const unsigned int _pressure_sys_number
The number of the system corresponding to the pressure equation.
Definition: LinearAssemblySegregatedSolve.h:82

◆ solveSolidEnergy()

std::pair< unsigned int, Real > LinearAssemblySegregatedSolve::solveSolidEnergy ( )
protected

Solve an equation which contains the solid energy conservation.

Definition at line 351 of file LinearAssemblySegregatedSolve.C.

Referenced by solve().

352 {
353  _problem.setCurrentLinearSystem(_solid_energy_sys_number);
354 
355  // We will need some members from the linear system
356  LinearImplicitSystem & system =
357  libMesh::cast_ref<LinearImplicitSystem &>(_solid_energy_system->system());
358 
359  // We will need the solution, the right hand side and the matrix
360  NumericVector<Number> & current_local_solution = *(system.current_local_solution);
361  NumericVector<Number> & solution = *(system.solution);
362  SparseMatrix<Number> & mmat = *(system.matrix);
363  NumericVector<Number> & rhs = *(system.rhs);
364 
365  // Fetch the linear solver from the system
366  PetscLinearSolver<Real> & solver =
367  libMesh::cast_ref<PetscLinearSolver<Real> &>(*system.get_linear_solver());
368 
369  _problem.computeLinearSystemSys(system, mmat, rhs, false);
370 
371  if (_print_fields)
372  {
373  _console << "Solid energy matrix" << std::endl;
374  mmat.print();
375  }
376 
377  // We compute the normalization factors based on the fluxes
378  Real norm_factor = NS::FV::computeNormalizationFactor(solution, mmat, rhs);
379 
380  // We need the non-preconditioned norm to be consistent with the norm factor
381  LibmeshPetscCall(KSPSetNormType(solver.ksp(), KSP_NORM_UNPRECONDITIONED));
382 
383  // Setting the linear tolerances and maximum iteration counts
384  _solid_energy_linear_control.real_valued_data["abs_tol"] = _solid_energy_l_abs_tol * norm_factor;
385  solver.set_solver_configuration(_solid_energy_linear_control);
386 
387  auto its_res_pair = solver.solve(mmat, mmat, solution, rhs);
388  system.update();
389 
390  if (_print_fields)
391  {
392  _console << " rhs when we solve solid energy " << std::endl;
393  rhs.print();
394  _console << " Solid energy " << std::endl;
395  solution.print();
396  _console << "Norm factor " << norm_factor << std::endl;
397  }
398 
399  _solid_energy_system->setSolution(current_local_solution);
400 
401  const auto residuals =
402  std::make_pair(its_res_pair.first, solver.get_initial_residual() / norm_factor);
403 
404  _console << " Solid energy equation: " << COLOR_GREEN << residuals.second << COLOR_DEFAULT
405  << " Linear its: " << residuals.first << std::endl;
406 
407  return residuals;
408 }
SolveObject::_problem
FEProblemBase & _problem
SIMPLESolveBase::_print_fields
const bool _print_fields
Debug parameter which allows printing the coupling and solution vectors/matrices. ...
Definition: SIMPLESolveBase.h:240
SIMPLESolveBase::_solid_energy_l_abs_tol
const Real _solid_energy_l_abs_tol
Absolute linear tolerance for the energy equations.
Definition: SIMPLESolveBase.h:155
NS::FV::computeNormalizationFactor
Real computeNormalizationFactor(const NumericVector< Number > &solution, const SparseMatrix< Number > &mat, const NumericVector< Number > &rhs)
Compute a normalization factor which is applied to the linear residual to determine convergence...
Definition: SegregatedSolverUtils.C:135
LinearAssemblySegregatedSolve::_solid_energy_system
LinearSystem * _solid_energy_system
Pointer to the nonlinear system corresponding to the solid energy equation.
Definition: LinearAssemblySegregatedSolve.h:97
libMesh::LinearImplicitSystem::rhs
NumericVector< Number > * rhs
LinearSystem::setSolution
void setSolution(const NumericVector< Number > &soln)
libMesh::PetscLinearSolver::solve
virtual std::pair< unsigned int, Real > solve(SparseMatrix< T > &matrix_in, NumericVector< T > &solution_in, NumericVector< T > &rhs_in, const std::optional< double > tol=std::nullopt, const std::optional< unsigned int > m_its=std::nullopt) override
libMesh::LinearImplicitSystem::get_linear_solver
virtual LinearSolver< Number > * get_linear_solver() const override
libMesh::SolverConfiguration::real_valued_data
std::map< std::string, Real > real_valued_data
FEProblemBase::computeLinearSystemSys
virtual void computeLinearSystemSys(libMesh::LinearImplicitSystem &sys, libMesh::SparseMatrix< libMesh::Number > &system_matrix, NumericVector< libMesh::Number > &rhs, const bool compute_gradients=true)
LinearAssemblySegregatedSolve::_solid_energy_sys_number
const unsigned int _solid_energy_sys_number
The number of the system corresponding to the solid energy equation.
Definition: LinearAssemblySegregatedSolve.h:94
libMesh::LinearImplicitSystem::solution
std::unique_ptr< NumericVector< Number > > solution
libMesh::NumericVector::print
virtual void print(std::ostream &os=libMesh::out) const
libMesh::PetscLinearSolver::set_solver_configuration
void set_solver_configuration(SolverConfiguration &solver_configuration)
Real
DIE A HORRIBLE DEATH HERE typedef LIBMESH_DEFAULT_SCALAR_TYPE Real
libMesh::LinearImplicitSystem::matrix
SparseMatrix< Number > * matrix
FEProblemBase::setCurrentLinearSystem
void setCurrentLinearSystem(unsigned int sys_num)
libMesh::LinearImplicitSystem::current_local_solution
std::unique_ptr< NumericVector< Number > > current_local_solution
SolveObject::_console
const ConsoleStream _console
SIMPLESolveBase::_solid_energy_linear_control
SIMPLESolverConfiguration _solid_energy_linear_control
Options for the linear solver of the energy equation.
Definition: SIMPLESolveBase.h:151
libMesh::SparseMatrix::print
void print(std::ostream &os=libMesh::out, const bool sparse=false) const
LinearSystem::system
virtual System & system() override

◆ systemsToSolve()

const std::vector<LinearSystem *> LinearAssemblySegregatedSolve::systemsToSolve ( ) const
inline

Return pointers to the systems which are solved for within this object.

Definition at line 38 of file LinearAssemblySegregatedSolve.h.

Referenced by PIMPLE::getTimeIntegrators(), and PIMPLE::relativeSolutionDifferenceNorm().

38 { return _systems_to_solve; }
LinearAssemblySegregatedSolve::_systems_to_solve
std::vector< LinearSystem * > _systems_to_solve
Shortcut to every linear system that we solve for here.
Definition: LinearAssemblySegregatedSolve.h:112

◆ validParams()

InputParameters LinearAssemblySegregatedSolve::validParams ( )
static

Definition at line 18 of file LinearAssemblySegregatedSolve.C.

Referenced by SIMPLESolve::validParams(), and PIMPLESolve::validParams().

19 {
20  InputParameters params = SIMPLESolveBase::validParams();
21 
22  params.addParam<std::vector<SolverSystemName>>(
23  "active_scalar_systems", {}, "The solver system for each active scalar advection equation.");
24 
25  /*
26  * Parameters to control the solution of each scalar advection system
27  */
28  params.addParam<std::vector<Real>>("active_scalar_equation_relaxation",
29  std::vector<Real>(),
30  "The relaxation which should be used for the active scalar "
31  "equations. (=1 for no relaxation, "
32  "diagonal dominance will still be enforced)");
33 
34  params.addParam<MultiMooseEnum>("active_scalar_petsc_options",
35  Moose::PetscSupport::getCommonPetscFlags(),
36  "Singleton PETSc options for the active scalar equation(s)");
37  params.addParam<MultiMooseEnum>(
38  "active_scalar_petsc_options_iname",
39  Moose::PetscSupport::getCommonPetscKeys(),
40  "Names of PETSc name/value pairs for the active scalar equation(s)");
41  params.addParam<std::vector<std::string>>(
42  "active_scalar_petsc_options_value",
43  "Values of PETSc name/value pairs (must correspond with \"petsc_options_iname\" for the "
44  "active scalar equation(s)");
45  params.addParam<std::vector<Real>>(
46  "active_scalar_absolute_tolerance",
47  std::vector<Real>(),
48  "The absolute tolerance(s) on the normalized residual(s) of the active scalar equation(s).");
49  params.addRangeCheckedParam<Real>("active_scalar_l_tol",
50  1e-5,
51  "0.0<=active_scalar_l_tol & active_scalar_l_tol<1.0",
52  "The relative tolerance on the normalized residual in the "
53  "linear solver of the active scalar equation(s).");
54  params.addRangeCheckedParam<Real>("active_scalar_l_abs_tol",
55  1e-10,
56  "0.0<active_scalar_l_abs_tol",
57  "The absolute tolerance on the normalized residual in the "
58  "linear solver of the active scalar equation(s).");
59  params.addParam<unsigned int>(
60  "active_scalar_l_max_its",
61  10000,
62  "The maximum allowed iterations in the linear solver of the turbulence equation.");
63 
64  params.addParamNamesToGroup(
65  "active_scalar_systems active_scalar_equation_relaxation active_scalar_petsc_options "
66  "active_scalar_petsc_options_iname "
67  "active_scalar_petsc_options_value active_scalar_petsc_options_value "
68  "active_scalar_absolute_tolerance "
69  "active_scalar_l_tol active_scalar_l_abs_tol active_scalar_l_max_its",
70  "Active Scalars Equations");
71 
72  return params;
73 }
Moose::PetscSupport::getCommonPetscKeys
MultiMooseEnum getCommonPetscKeys()
InputParameters::addParam
void addParam(const std::string &name, const std::initializer_list< typename T::value_type > &value, const std::string &doc_string)
SIMPLESolveBase::validParams
static InputParameters validParams()
Definition: SIMPLESolveBase.C:15
Moose::PetscSupport::getCommonPetscFlags
MultiMooseEnum getCommonPetscFlags()
Real
DIE A HORRIBLE DEATH HERE typedef LIBMESH_DEFAULT_SCALAR_TYPE Real

Member Data Documentation

◆ _active_scalar_absolute_tolerance

const std::vector<Real> LinearAssemblySegregatedSolve::_active_scalar_absolute_tolerance
protected

The user-defined absolute tolerance for determining the convergence in active scalars.

Definition at line 139 of file LinearAssemblySegregatedSolve.h.

Referenced by solve().

◆ _active_scalar_equation_relaxation

const std::vector<Real> LinearAssemblySegregatedSolve::_active_scalar_equation_relaxation
protected

The user-defined relaxation parameter(s) for the active scalar equation(s)

Definition at line 126 of file LinearAssemblySegregatedSolve.h.

Referenced by LinearAssemblySegregatedSolve(), and solve().

◆ _active_scalar_l_abs_tol

const Real LinearAssemblySegregatedSolve::_active_scalar_l_abs_tol
protected

Absolute linear tolerance for the active scalar equation(s).

We need to store this, because it needs to be scaled with a representative flux.

Definition at line 136 of file LinearAssemblySegregatedSolve.h.

Referenced by solve().

◆ _active_scalar_linear_control

SIMPLESolverConfiguration LinearAssemblySegregatedSolve::_active_scalar_linear_control
protected

Options for the linear solver of the active scalar equation(s)

Definition at line 132 of file LinearAssemblySegregatedSolve.h.

Referenced by LinearAssemblySegregatedSolve(), and solve().

◆ _active_scalar_petsc_options

Moose::PetscSupport::PetscOptions LinearAssemblySegregatedSolve::_active_scalar_petsc_options
protected

Options which hold the petsc settings for the active scalar equation(s)

Definition at line 129 of file LinearAssemblySegregatedSolve.h.

Referenced by LinearAssemblySegregatedSolve(), and solve().

◆ _active_scalar_system_names

const std::vector<SolverSystemName>& LinearAssemblySegregatedSolve::_active_scalar_system_names
protected

The names of the active scalar systems.

Definition at line 117 of file LinearAssemblySegregatedSolve.h.

Referenced by LinearAssemblySegregatedSolve(), and solve().

◆ _active_scalar_system_numbers

std::vector<unsigned int> LinearAssemblySegregatedSolve::_active_scalar_system_numbers
protected

Definition at line 123 of file LinearAssemblySegregatedSolve.h.

Referenced by LinearAssemblySegregatedSolve(), and solve().

◆ _active_scalar_systems

std::vector<LinearSystem *> LinearAssemblySegregatedSolve::_active_scalar_systems
protected

Pointer(s) to the system(s) corresponding to the active scalar equation(s)

Definition at line 103 of file LinearAssemblySegregatedSolve.h.

Referenced by SIMPLESolve::checkIntegrity(), LinearAssemblySegregatedSolve(), and solve().

◆ _continue_on_max_its

const bool SIMPLESolveBase::_continue_on_max_its
protectedinherited

If solve should continue if maximum number of iterations is hit.

Definition at line 235 of file SIMPLESolveBase.h.

Referenced by solve(), and SIMPLESolveNonlinearAssembly::solve().

◆ _energy_absolute_tolerance

const Real SIMPLESolveBase::_energy_absolute_tolerance
protectedinherited

The user-defined absolute tolerance for determining the convergence in energy.

Definition at line 220 of file SIMPLESolveBase.h.

Referenced by solve(), and SIMPLESolveNonlinearAssembly::solve().

◆ _energy_equation_relaxation

const Real SIMPLESolveBase::_energy_equation_relaxation
protectedinherited

The user-defined relaxation parameter for the energy equation.

Definition at line 130 of file SIMPLESolveBase.h.

Referenced by solve(), and SIMPLESolveNonlinearAssembly::solve().

◆ _energy_l_abs_tol

const Real SIMPLESolveBase::_energy_l_abs_tol
protectedinherited

Absolute linear tolerance for the energy equations.

We need to store this, because it needs to be scaled with a representative flux.

Definition at line 140 of file SIMPLESolveBase.h.

Referenced by solve(), and SIMPLESolveNonlinearAssembly::solve().

◆ _energy_linear_control

SIMPLESolverConfiguration SIMPLESolveBase::_energy_linear_control
protectedinherited

Options for the linear solver of the energy equation.

Definition at line 136 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveBase::SIMPLESolveBase(), SIMPLESolveNonlinearAssembly::solve(), and solve().

◆ _energy_petsc_options

Moose::PetscSupport::PetscOptions SIMPLESolveBase::_energy_petsc_options
protectedinherited

Options which hold the petsc settings for the fluid energy equation.

Definition at line 133 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveBase::SIMPLESolveBase(), SIMPLESolveNonlinearAssembly::solve(), and solve().

◆ _energy_sys_number

const unsigned int LinearAssemblySegregatedSolve::_energy_sys_number
protected

The number of the system corresponding to the energy equation.

Definition at line 88 of file LinearAssemblySegregatedSolve.h.

Referenced by solve().

◆ _energy_system

LinearSystem* LinearAssemblySegregatedSolve::_energy_system
protected

Pointer to the nonlinear system corresponding to the fluid energy equation.

Definition at line 91 of file LinearAssemblySegregatedSolve.h.

Referenced by SIMPLESolve::checkIntegrity(), LinearAssemblySegregatedSolve(), and solve().

◆ _has_active_scalar_systems

const bool LinearAssemblySegregatedSolve::_has_active_scalar_systems
protected

Boolean for easy check if a active scalar systems shall be solved or not.

Definition at line 120 of file LinearAssemblySegregatedSolve.h.

Referenced by SIMPLESolve::checkIntegrity(), LinearAssemblySegregatedSolve(), and solve().

◆ _has_energy_system

const bool SIMPLESolveBase::_has_energy_system
protectedinherited

Boolean for easy check if a fluid energy system shall be solved or not.

Definition at line 127 of file SIMPLESolveBase.h.

Referenced by SIMPLESolve::checkIntegrity(), SIMPLESolveNonlinearAssembly::checkIntegrity(), LinearAssemblySegregatedSolve(), SIMPLESolveBase::SIMPLESolveBase(), SIMPLESolveNonlinearAssembly::SIMPLESolveNonlinearAssembly(), SIMPLESolveNonlinearAssembly::solve(), and solve().

◆ _has_passive_scalar_systems

const bool SIMPLESolveBase::_has_passive_scalar_systems
protectedinherited

Boolean for easy check if a passive scalar systems shall be solved or not.

Definition at line 163 of file SIMPLESolveBase.h.

Referenced by SIMPLESolve::checkIntegrity(), SIMPLESolveNonlinearAssembly::checkIntegrity(), LinearAssemblySegregatedSolve(), SIMPLESolveBase::SIMPLESolveBase(), SIMPLESolveNonlinearAssembly::SIMPLESolveNonlinearAssembly(), SIMPLESolveNonlinearAssembly::solve(), and solve().

◆ _has_solid_energy_system

const bool SIMPLESolveBase::_has_solid_energy_system
protectedinherited

Boolean for easy check if a solid energy system shall be solved or not.

Definition at line 145 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveNonlinearAssembly::checkIntegrity(), LinearAssemblySegregatedSolve(), SIMPLESolveBase::SIMPLESolveBase(), SIMPLESolveNonlinearAssembly::solve(), and solve().

◆ _has_turbulence_systems

const bool SIMPLESolveBase::_has_turbulence_systems
protectedinherited

Boolean for easy check if a turbulence scalar systems shall be solved or not.

Definition at line 187 of file SIMPLESolveBase.h.

Referenced by SIMPLESolve::checkIntegrity(), LinearAssemblySegregatedSolve(), SIMPLESolveBase::SIMPLESolveBase(), and solve().

◆ _momentum_absolute_tolerance

const Real SIMPLESolveBase::_momentum_absolute_tolerance
protectedinherited

The user-defined absolute tolerance for determining the convergence in momentum.

Definition at line 214 of file SIMPLESolveBase.h.

Referenced by solve(), and SIMPLESolveNonlinearAssembly::solve().

◆ _momentum_equation_relaxation

const Real SIMPLESolveBase::_momentum_equation_relaxation
protectedinherited

The user-defined relaxation parameter for the momentum equation.

Definition at line 95 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveNonlinearAssembly::solveMomentumPredictor(), and solveMomentumPredictor().

◆ _momentum_l_abs_tol

const Real SIMPLESolveBase::_momentum_l_abs_tol
protectedinherited

Absolute linear tolerance for the momentum equation(s).

We need to store this, because it needs to be scaled with a representative flux.

Definition at line 89 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveNonlinearAssembly::solveMomentumPredictor(), and solveMomentumPredictor().

◆ _momentum_linear_control

SIMPLESolverConfiguration SIMPLESolveBase::_momentum_linear_control
protectedinherited

Options for the linear solver of the momentum equation.

Definition at line 85 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveBase::SIMPLESolveBase(), SIMPLESolveNonlinearAssembly::solveMomentumPredictor(), and solveMomentumPredictor().

◆ _momentum_petsc_options

Moose::PetscSupport::PetscOptions SIMPLESolveBase::_momentum_petsc_options
protectedinherited

Options which hold the petsc settings for the momentum equation.

Definition at line 92 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveBase::SIMPLESolveBase(), SIMPLESolveNonlinearAssembly::solve(), and solve().

◆ _momentum_system_names

const std::vector<SolverSystemName>& SIMPLESolveBase::_momentum_system_names
protectedinherited

The names of the momentum systems.

Definition at line 82 of file SIMPLESolveBase.h.

Referenced by LinearAssemblySegregatedSolve(), SIMPLESolveBase::SIMPLESolveBase(), and SIMPLESolveNonlinearAssembly::SIMPLESolveNonlinearAssembly().

◆ _momentum_system_numbers

std::vector<unsigned int> LinearAssemblySegregatedSolve::_momentum_system_numbers
protected

The number(s) of the system(s) corresponding to the momentum equation(s)

Definition at line 76 of file LinearAssemblySegregatedSolve.h.

Referenced by LinearAssemblySegregatedSolve(), linkRhieChowUserObject(), and solveMomentumPredictor().

◆ _momentum_systems

std::vector<LinearSystem *> LinearAssemblySegregatedSolve::_momentum_systems
protected

Pointer(s) to the system(s) corresponding to the momentum equation(s)

Definition at line 79 of file LinearAssemblySegregatedSolve.h.

Referenced by SIMPLESolve::checkIntegrity(), LinearAssemblySegregatedSolve(), linkRhieChowUserObject(), solve(), and solveMomentumPredictor().

◆ _num_iterations

const unsigned int SIMPLESolveBase::_num_iterations
protectedinherited

The maximum number of momentum-pressure iterations.

Definition at line 232 of file SIMPLESolveBase.h.

Referenced by solve(), and SIMPLESolveNonlinearAssembly::solve().

◆ _passive_scalar_absolute_tolerance

const std::vector<Real> SIMPLESolveBase::_passive_scalar_absolute_tolerance
protectedinherited

The user-defined absolute tolerance for determining the convergence in passive scalars.

Definition at line 226 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveBase::SIMPLESolveBase(), SIMPLESolveNonlinearAssembly::solve(), and solve().

◆ _passive_scalar_equation_relaxation

const std::vector<Real> SIMPLESolveBase::_passive_scalar_equation_relaxation
protectedinherited

The user-defined relaxation parameter(s) for the passive scalar equation(s)

Definition at line 169 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveBase::SIMPLESolveBase(), SIMPLESolveNonlinearAssembly::solve(), and solve().

◆ _passive_scalar_l_abs_tol

const Real SIMPLESolveBase::_passive_scalar_l_abs_tol
protectedinherited

Absolute linear tolerance for the passive scalar equation(s).

We need to store this, because it needs to be scaled with a representative flux.

Definition at line 179 of file SIMPLESolveBase.h.

Referenced by solve(), and SIMPLESolveNonlinearAssembly::solve().

◆ _passive_scalar_linear_control

SIMPLESolverConfiguration SIMPLESolveBase::_passive_scalar_linear_control
protectedinherited

Options for the linear solver of the passive scalar equation(s)

Definition at line 175 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveBase::SIMPLESolveBase(), SIMPLESolveNonlinearAssembly::solve(), and solve().

◆ _passive_scalar_petsc_options

Moose::PetscSupport::PetscOptions SIMPLESolveBase::_passive_scalar_petsc_options
protectedinherited

Options which hold the petsc settings for the passive scalar equation(s)

Definition at line 172 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveBase::SIMPLESolveBase(), SIMPLESolveNonlinearAssembly::solve(), and solve().

◆ _passive_scalar_system_names

const std::vector<SolverSystemName>& SIMPLESolveBase::_passive_scalar_system_names
protectedinherited

The names of the passive scalar systems.

Definition at line 160 of file SIMPLESolveBase.h.

Referenced by LinearAssemblySegregatedSolve(), SIMPLESolveBase::SIMPLESolveBase(), SIMPLESolveNonlinearAssembly::SIMPLESolveNonlinearAssembly(), and solve().

◆ _passive_scalar_system_numbers

std::vector<unsigned int> SIMPLESolveBase::_passive_scalar_system_numbers
protectedinherited

Definition at line 166 of file SIMPLESolveBase.h.

Referenced by LinearAssemblySegregatedSolve(), SIMPLESolveNonlinearAssembly::SIMPLESolveNonlinearAssembly(), SIMPLESolveNonlinearAssembly::solve(), and solve().

◆ _passive_scalar_systems

std::vector<LinearSystem *> LinearAssemblySegregatedSolve::_passive_scalar_systems
protected

Pointer(s) to the system(s) corresponding to the passive scalar equation(s)

Definition at line 100 of file LinearAssemblySegregatedSolve.h.

Referenced by SIMPLESolve::checkIntegrity(), LinearAssemblySegregatedSolve(), and solve().

◆ _pin_pressure

const bool SIMPLESolveBase::_pin_pressure
protectedinherited

If the pressure needs to be pinned.

Definition at line 116 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveBase::setupPressurePin(), SIMPLESolveNonlinearAssembly::solvePressureCorrector(), and solvePressureCorrector().

◆ _pressure_absolute_tolerance

const Real SIMPLESolveBase::_pressure_absolute_tolerance
protectedinherited

The user-defined absolute tolerance for determining the convergence in pressure.

Definition at line 217 of file SIMPLESolveBase.h.

Referenced by solve(), and SIMPLESolveNonlinearAssembly::solve().

◆ _pressure_l_abs_tol

const Real SIMPLESolveBase::_pressure_l_abs_tol
protectedinherited

Absolute linear tolerance for the pressure equation.

We need to store this, because it needs to be scaled with a representative flux.

Definition at line 107 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveNonlinearAssembly::solvePressureCorrector(), and solvePressureCorrector().

◆ _pressure_linear_control

SIMPLESolverConfiguration SIMPLESolveBase::_pressure_linear_control
protectedinherited

Options for the linear solver of the pressure equation.

Definition at line 103 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveBase::SIMPLESolveBase(), SIMPLESolveNonlinearAssembly::solvePressureCorrector(), and solvePressureCorrector().

◆ _pressure_petsc_options

Moose::PetscSupport::PetscOptions SIMPLESolveBase::_pressure_petsc_options
protectedinherited

Options which hold the petsc settings for the pressure equation.

Definition at line 110 of file SIMPLESolveBase.h.

Referenced by correctVelocity(), SIMPLESolveBase::SIMPLESolveBase(), and SIMPLESolveNonlinearAssembly::solve().

◆ _pressure_pin_dof

dof_id_type SIMPLESolveBase::_pressure_pin_dof
protectedinherited

The dof ID where the pressure needs to be pinned.

Definition at line 122 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveBase::setupPressurePin(), SIMPLESolveNonlinearAssembly::solvePressureCorrector(), and solvePressureCorrector().

◆ _pressure_pin_value

const Real SIMPLESolveBase::_pressure_pin_value
protectedinherited

The value we want to enforce for pressure.

Definition at line 119 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveNonlinearAssembly::solvePressureCorrector(), and solvePressureCorrector().

◆ _pressure_sys_number

const unsigned int LinearAssemblySegregatedSolve::_pressure_sys_number
protected

The number of the system corresponding to the pressure equation.

Definition at line 82 of file LinearAssemblySegregatedSolve.h.

Referenced by solvePressureCorrector().

◆ _pressure_system

LinearSystem& LinearAssemblySegregatedSolve::_pressure_system
protected

Reference to the nonlinear system corresponding to the pressure equation.

Definition at line 85 of file LinearAssemblySegregatedSolve.h.

Referenced by SIMPLESolve::checkIntegrity(), correctVelocity(), LinearAssemblySegregatedSolve(), linkRhieChowUserObject(), solve(), and solvePressureCorrector().

◆ _pressure_system_name

const SolverSystemName& SIMPLESolveBase::_pressure_system_name
protectedinherited

The name of the pressure system.

Definition at line 100 of file SIMPLESolveBase.h.

◆ _pressure_variable_relaxation

const Real SIMPLESolveBase::_pressure_variable_relaxation
protectedinherited

The user-defined relaxation parameter for the pressure variable.

Definition at line 113 of file SIMPLESolveBase.h.

Referenced by correctVelocity(), and SIMPLESolveNonlinearAssembly::solve().

◆ _print_fields

const bool SIMPLESolveBase::_print_fields
protectedinherited

Debug parameter which allows printing the coupling and solution vectors/matrices.

Definition at line 240 of file SIMPLESolveBase.h.

Referenced by correctVelocity(), SIMPLESolveNonlinearAssembly::solve(), SIMPLESolveNonlinearAssembly::solveAdvectedSystem(), solveAdvectedSystem(), SIMPLESolveNonlinearAssembly::solveMomentumPredictor(), solveMomentumPredictor(), SIMPLESolveNonlinearAssembly::solvePressureCorrector(), solvePressureCorrector(), solveSolidEnergy(), and SIMPLESolveNonlinearAssembly::solveSolidEnergySystem().

◆ _rc_uo

RhieChowMassFlux* LinearAssemblySegregatedSolve::_rc_uo
protected

Pointer to the segregated RhieChow interpolation object.

Definition at line 109 of file LinearAssemblySegregatedSolve.h.

Referenced by correctVelocity(), and linkRhieChowUserObject().

◆ _solid_energy_absolute_tolerance

const Real SIMPLESolveBase::_solid_energy_absolute_tolerance
protectedinherited

The user-defined absolute tolerance for determining the convergence in solid energy.

Definition at line 223 of file SIMPLESolveBase.h.

Referenced by solve(), and SIMPLESolveNonlinearAssembly::solve().

◆ _solid_energy_l_abs_tol

const Real SIMPLESolveBase::_solid_energy_l_abs_tol
protectedinherited

Absolute linear tolerance for the energy equations.

We need to store this, because it needs to be scaled with a representative flux.

Definition at line 155 of file SIMPLESolveBase.h.

Referenced by solveSolidEnergy(), and SIMPLESolveNonlinearAssembly::solveSolidEnergySystem().

◆ _solid_energy_linear_control

SIMPLESolverConfiguration SIMPLESolveBase::_solid_energy_linear_control
protectedinherited

Options for the linear solver of the energy equation.

Definition at line 151 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveBase::SIMPLESolveBase(), solveSolidEnergy(), and SIMPLESolveNonlinearAssembly::solveSolidEnergySystem().

◆ _solid_energy_petsc_options

Moose::PetscSupport::PetscOptions SIMPLESolveBase::_solid_energy_petsc_options
protectedinherited

Options which hold the petsc settings for the fluid energy equation.

Definition at line 148 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveBase::SIMPLESolveBase(), and solve().

◆ _solid_energy_sys_number

const unsigned int LinearAssemblySegregatedSolve::_solid_energy_sys_number
protected

The number of the system corresponding to the solid energy equation.

Definition at line 94 of file LinearAssemblySegregatedSolve.h.

Referenced by solveSolidEnergy().

◆ _solid_energy_system

LinearSystem* LinearAssemblySegregatedSolve::_solid_energy_system
protected

Pointer to the nonlinear system corresponding to the solid energy equation.

Definition at line 97 of file LinearAssemblySegregatedSolve.h.

Referenced by LinearAssemblySegregatedSolve(), and solveSolidEnergy().

◆ _systems_to_solve

std::vector<LinearSystem *> LinearAssemblySegregatedSolve::_systems_to_solve
protected

Shortcut to every linear system that we solve for here.

Definition at line 112 of file LinearAssemblySegregatedSolve.h.

Referenced by LinearAssemblySegregatedSolve(), and systemsToSolve().

◆ _turbulence_absolute_tolerance

const std::vector<Real> SIMPLESolveBase::_turbulence_absolute_tolerance
protectedinherited

The user-defined absolute tolerance for determining the convergence turbulence variables.

Definition at line 229 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveBase::SIMPLESolveBase(), and solve().

◆ _turbulence_equation_relaxation

const std::vector<Real> SIMPLESolveBase::_turbulence_equation_relaxation
protectedinherited

The user-defined relaxation parameter(s) for the turbulence equation(s)

Definition at line 193 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveBase::SIMPLESolveBase(), and solve().

◆ _turbulence_field_min_limit

std::vector<Real> SIMPLESolveBase::_turbulence_field_min_limit
protectedinherited

The user-defined lower limit for turbulent quantities e.g. k, eps/omega, etc..

Definition at line 199 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveBase::SIMPLESolveBase(), and solve().

◆ _turbulence_field_relaxation

std::vector<Real> SIMPLESolveBase::_turbulence_field_relaxation
protectedinherited

The user-defined relaxation parameter(s) for the turbulence field(s)

Definition at line 196 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveBase::SIMPLESolveBase(), and solve().

◆ _turbulence_l_abs_tol

const Real SIMPLESolveBase::_turbulence_l_abs_tol
protectedinherited

Absolute linear tolerance for the turbulence equation(s).

We need to store this, because it needs to be scaled with a representative flux.

Definition at line 209 of file SIMPLESolveBase.h.

Referenced by solve().

◆ _turbulence_linear_control

SIMPLESolverConfiguration SIMPLESolveBase::_turbulence_linear_control
protectedinherited

Options for the linear solver of the turbulence equation(s)

Definition at line 205 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveBase::SIMPLESolveBase(), and solve().

◆ _turbulence_petsc_options

Moose::PetscSupport::PetscOptions SIMPLESolveBase::_turbulence_petsc_options
protectedinherited

Options which hold the petsc settings for the turbulence equation(s)

Definition at line 202 of file SIMPLESolveBase.h.

Referenced by SIMPLESolveBase::SIMPLESolveBase(), and solve().

◆ _turbulence_system_names

const std::vector<SolverSystemName>& SIMPLESolveBase::_turbulence_system_names
protectedinherited

The names of the turbulence systems.

Definition at line 184 of file SIMPLESolveBase.h.

Referenced by LinearAssemblySegregatedSolve(), SIMPLESolveBase::SIMPLESolveBase(), and solve().

◆ _turbulence_system_numbers

std::vector<unsigned int> SIMPLESolveBase::_turbulence_system_numbers
protectedinherited

Definition at line 190 of file SIMPLESolveBase.h.

Referenced by LinearAssemblySegregatedSolve(), and solve().

◆ _turbulence_systems

std::vector<LinearSystem *> LinearAssemblySegregatedSolve::_turbulence_systems
protected

Pointer(s) to the system(s) corresponding to the turbulence equation(s)

Definition at line 106 of file LinearAssemblySegregatedSolve.h.

Referenced by SIMPLESolve::checkIntegrity(), LinearAssemblySegregatedSolve(), and solve().

The documentation for this class was generated from the following files:
Generated on Fri Aug 29 2025 20:23:40 for https://mooseframework.inl.gov by   doxygen 1.8.14