LinearFVTurbulentAdvection.C

Go to the documentation of this file.

//* This file is part of the MOOSE framework
//* https://mooseframework.inl.gov
//*
//* All rights reserved, see COPYRIGHT for full restrictions
//* https://github.com/idaholab/moose/blob/master/COPYRIGHT
//*
//* Licensed under LGPL 2.1, please see LICENSE for details
//* https://www.gnu.org/licenses/lgpl-2.1.html
 
#include "LinearFVTurbulentAdvection.h"
#include "LinearFVGradientManager.h"
#include "MooseLinearVariableFV.h"
#include "NavierStokesMethods.h"
#include "NS.h"
 
registerMooseObject("NavierStokesApp", LinearFVTurbulentAdvection);
 
InputParameters
LinearFVTurbulentAdvection::validParams()
{
  InputParameters params = LinearFVFluxKernel::validParams();
  params.addClassDescription("Represents the matrix and right hand side contributions of an "
                             "advection term for a turbulence variable.");
 
  params.addRequiredParam<UserObjectName>(
      "rhie_chow_user_object",
      "The rhie-chow user-object which is used to determine the face velocity.");
 
  params.addRequiredParam<InterpolationMethodName>(
      "advected_interp_method_name",
      "Name of the FVInterpolationMethod to use for the advected quantity.");
 
  params.addParam<std::vector<BoundaryName>>(
      "walls", {}, "Boundaries that correspond to solid walls.");
 
  return params;
}
 
LinearFVTurbulentAdvection::LinearFVTurbulentAdvection(const InputParameters & params)
  : LinearFVFluxKernel(params),
    FVInterpolationMethodInterface(this),
    _mass_flux_provider(getUserObject<RhieChowMassFlux>("rhie_chow_user_object")),
    _adv_interp_method(getFVAdvectedInterpolationMethod(
        getParam<InterpolationMethodName>("advected_interp_method_name"))),
    _gradient_field(_adv_interp_method.needsGradients()
                        ? &_var.requestCellGradients(_adv_interp_method.gradientMethodName())
                        : nullptr),
    _mass_face_flux(0.0),
    _wall_boundary_names(getParam<std::vector<BoundaryName>>("walls"))
{
  // The boundary contributions rely on the Rhie-Chow flux being oriented outward from this
  // variable's cell, which only holds if the variable's one-sided faces are one-sided for the
  // flow too.
  _mass_flux_provider.checkBlocks(_var);
}
 
void
LinearFVTurbulentAdvection::initialSetup()
{
  LinearFVFluxKernel::initialSetup();
  NS::getWallBoundedElements(
      _wall_boundary_names, _fe_problem, _subproblem, blockIDs(), _wall_bounded);
}
 
void
LinearFVTurbulentAdvection::addMatrixContribution()
{
  // Coumputing bounding map
  const Elem * elem = _current_face_info->elemPtr();
  const auto bounded_elem = _wall_bounded.find(elem) != _wall_bounded.end();
  const Elem * neighbor = _current_face_info->neighborPtr();
  const auto bounded_neigh = _wall_bounded.find(neighbor) != _wall_bounded.end();
 
  // If we are on an internal face, we populate the four entries in the system matrix
  // which touch the face
  if (_current_face_type == FaceInfo::VarFaceNeighbors::BOTH)
  {
    // The dof ids of the variable corresponding to the element and neighbor
    _dof_indices(0) = _current_face_info->elemInfo()->dofIndices()[_sys_num][_var_num];
    _dof_indices(1) = _current_face_info->neighborInfo()->dofIndices()[_sys_num][_var_num];
 
    // Compute the entries which will go to the neighbor (offdiagonal) and element
    // (diagonal).
    const auto elem_matrix_contribution = computeElemMatrixContribution();
    const auto neighbor_matrix_contribution = computeNeighborMatrixContribution();
 
    // Populate matrix
    if (hasBlocks(_current_face_info->elemInfo()->subdomain_id()) && !(bounded_elem))
    {
      _matrix_contribution(0, 0) = elem_matrix_contribution;
      _matrix_contribution(0, 1) = neighbor_matrix_contribution;
    }
 
    if (hasBlocks(_current_face_info->neighborInfo()->subdomain_id()) && !(bounded_neigh))
    {
      _matrix_contribution(1, 0) = -elem_matrix_contribution;
      _matrix_contribution(1, 1) = -neighbor_matrix_contribution;
    }
    (*_linear_system.matrix).add_matrix(_matrix_contribution, _dof_indices.get_values());
  }
  // We are at a block boundary where the variable is not defined on one of the adjacent cells.
  // We check if we have a boundary condition here
  else if (_current_face_type == FaceInfo::VarFaceNeighbors::ELEM ||
           _current_face_type == FaceInfo::VarFaceNeighbors::NEIGHBOR)
  {
    mooseAssert(_current_face_info->boundaryIDs().size() == 1,
                "We should only have one boundary on every face.");
 
    LinearFVBoundaryCondition * bc_pointer =
        _var.getBoundaryCondition(*_current_face_info->boundaryIDs().begin());
 
    if (bc_pointer || _force_boundary_execution)
    {
      if (bc_pointer)
        bc_pointer->setupFaceData(_current_face_info, _current_face_type);
      const auto matrix_contribution = computeBoundaryMatrixContribution(*bc_pointer);
 
      // We allow internal (for the mesh) boundaries too, so we have to check on which side we
      // are on (assuming that this is a boundary for the variable)
      if ((_current_face_type == FaceInfo::VarFaceNeighbors::ELEM) && !(bounded_elem))
      {
        const auto dof_id_elem = _current_face_info->elemInfo()->dofIndices()[_sys_num][_var_num];
        (*_linear_system.matrix).add(dof_id_elem, dof_id_elem, matrix_contribution);
      }
      else if ((_current_face_type == FaceInfo::VarFaceNeighbors::NEIGHBOR) && !(bounded_neigh))
      {
        const auto dof_id_neighbor =
            _current_face_info->neighborInfo()->dofIndices()[_sys_num][_var_num];
        (*_linear_system.matrix).add(dof_id_neighbor, dof_id_neighbor, matrix_contribution);
      }
    }
  }
}
 
void
LinearFVTurbulentAdvection::addRightHandSideContribution()
{
  // Coumputing bounding map
  const Elem * elem = _current_face_info->elemPtr();
  const auto bounded_elem = _wall_bounded.find(elem) != _wall_bounded.end();
  const Elem * neighbor = _current_face_info->neighborPtr();
  const auto bounded_neigh = _wall_bounded.find(neighbor) != _wall_bounded.end();
 
  // If we are on an internal face, we populate the two entries in the right hand side
  // which touch the face
  if (_current_face_type == FaceInfo::VarFaceNeighbors::BOTH)
  {
    // The dof ids of the variable corresponding to the element and neighbor
    _dof_indices(0) = _current_face_info->elemInfo()->dofIndices()[_sys_num][_var_num];
    _dof_indices(1) = _current_face_info->neighborInfo()->dofIndices()[_sys_num][_var_num];
 
    // Compute the entries which will go to the neighbor and element positions.
    const auto elem_rhs_contribution = computeElemRightHandSideContribution();
    const auto neighbor_rhs_contribution = computeNeighborRightHandSideContribution();
 
    // Populate right hand side
    if (hasBlocks(_current_face_info->elemInfo()->subdomain_id()) && !(bounded_elem))
      _rhs_contribution(0) = elem_rhs_contribution;
    if (hasBlocks(_current_face_info->neighborInfo()->subdomain_id()) && !(bounded_neigh))
      _rhs_contribution(1) = neighbor_rhs_contribution;
 
    (*_linear_system.rhs)
        .add_vector(_rhs_contribution.get_values().data(), _dof_indices.get_values());
  }
  // We are at a block boundary where the variable is not defined on one of the adjacent cells.
  // We check if we have a boundary condition here
  else if (_current_face_type == FaceInfo::VarFaceNeighbors::ELEM ||
           _current_face_type == FaceInfo::VarFaceNeighbors::NEIGHBOR)
  {
    mooseAssert(_current_face_info->boundaryIDs().size() == 1,
                "We should only have one boundary on every face.");
    LinearFVBoundaryCondition * bc_pointer =
        _var.getBoundaryCondition(*_current_face_info->boundaryIDs().begin());
 
    if (bc_pointer || _force_boundary_execution)
    {
      if (bc_pointer)
        bc_pointer->setupFaceData(_current_face_info, _current_face_type);
 
      const auto rhs_contribution = computeBoundaryRHSContribution(*bc_pointer);
 
      // We allow internal (for the mesh) boundaries too, so we have to check on which side we
      // are on (assuming that this is a boundary for the variable)
      if ((_current_face_type == FaceInfo::VarFaceNeighbors::ELEM) && !(bounded_elem))
      {
        const auto dof_id_elem = _current_face_info->elemInfo()->dofIndices()[_sys_num][_var_num];
        (*_linear_system.rhs).add(dof_id_elem, rhs_contribution);
      }
      else if ((_current_face_type == FaceInfo::VarFaceNeighbors::NEIGHBOR) && !(bounded_neigh))
      {
        const auto dof_id_neighbor =
            _current_face_info->neighborInfo()->dofIndices()[_sys_num][_var_num];
        (*_linear_system.rhs).add(dof_id_neighbor, rhs_contribution);
      }
    }
  }
}
 
Real
LinearFVTurbulentAdvection::computeElemMatrixContribution()
{
  const auto & coeffs = _adv_interp_result.weights_matrix;
  return coeffs.first * _mass_face_flux * _current_face_area;
}
 
Real
LinearFVTurbulentAdvection::computeNeighborMatrixContribution()
{
  const auto & coeffs = _adv_interp_result.weights_matrix;
  return coeffs.second * _mass_face_flux * _current_face_area;
}
 
Real
LinearFVTurbulentAdvection::computeElemRightHandSideContribution()
{
  return _adv_interp_result.rhs_face_value * _mass_face_flux * _current_face_area;
}
 
Real
LinearFVTurbulentAdvection::computeNeighborRightHandSideContribution()
{
  return -_adv_interp_result.rhs_face_value * _mass_face_flux * _current_face_area;
}
 
Real
LinearFVTurbulentAdvection::computeBoundaryMatrixContribution(const LinearFVBoundaryCondition & bc)
{
  const auto * const adv_bc = cast_ptr<const LinearFVAdvectionDiffusionBC *>(&bc);
  mooseAssert(adv_bc, "This should be a valid BC!");
 
  const auto boundary_value_matrix_contrib = adv_bc->computeBoundaryValueMatrixContribution();
 
  // Rhie-Chow mass fluxes on one-sided boundary faces are already oriented outward from the
  // active cell, even when the active cell is the FaceInfo neighbor.
  return boundary_value_matrix_contrib * _mass_face_flux * _current_face_area;
}
 
Real
LinearFVTurbulentAdvection::computeBoundaryRHSContribution(const LinearFVBoundaryCondition & bc)
{
  const auto * const adv_bc = cast_ptr<const LinearFVAdvectionDiffusionBC *>(&bc);
  mooseAssert(adv_bc, "This should be a valid BC!");
 
  const auto boundary_value_rhs_contrib = adv_bc->computeBoundaryValueRHSContribution();
  return -boundary_value_rhs_contrib * _mass_face_flux * _current_face_area;
}
 
void
LinearFVTurbulentAdvection::setupFaceData(const FaceInfo * face_info)
{
  LinearFVFluxKernel::setupFaceData(face_info);
 
  // Caching the mass flux on the face which will be reused in the advection term's matrix and right
  // hand side contributions
  _mass_face_flux = _mass_flux_provider.getMassFlux(*face_info);
 
  // Only internal faces need advected interpolation results; boundary contributions are handled
  // through the linear FV boundary conditions.
  if (_current_face_type != FaceInfo::VarFaceNeighbors::BOTH)
    return;
 
  const auto state = determineState();
  const auto & elem_info = *_current_face_info->elemInfo();
  const auto & neighbor_info = *_current_face_info->neighborInfo();
 
  const Real elem_value = _var.getElemValue(elem_info, state);
  const Real neighbor_value = _var.getElemValue(neighbor_info, state);
  if (_adv_interp_method.needsGradients())
  {
    mooseAssert(_gradient_field, "Gradient field should be registered when gradients are needed.");
    _elem_grad_storage = _gradient_field->gradient(elem_info);
    _neighbor_grad_storage = _gradient_field->gradient(neighbor_info);
  }
 
  _adv_interp_result = _adv_interp_method.advectedInterpolate(*_current_face_info,
                                                              elem_value,
                                                              neighbor_value,
                                                              &_elem_grad_storage,
                                                              &_neighbor_grad_storage,
                                                              _mass_face_flux);
}