Revision 96815658fb3df52d570cadbf0ce551062490dde6 authored by Jan Thiedau on 31 January 2023, 11:44:42 UTC, committed by Jan Thiedau on 31 January 2023, 11:44:42 UTC
[PL/CT] Extend residual output for more than one component and to Picard solver See merge request ogs/ogs!4451
ThermoRichardsFlowProcess.h
/**
* \file
* \copyright
* Copyright (c) 2012-2023, OpenGeoSys Community (http://www.opengeosys.org)
* Distributed under a Modified BSD License.
* See accompanying file LICENSE.txt or
* http://www.opengeosys.org/project/license
*
*/
#pragma once
#include "LocalAssemblerInterface.h"
#include "ProcessLib/Process.h"
#include "ThermoRichardsFlowProcessData.h"
namespace ProcessLib
{
namespace ThermoRichardsFlow
{
/**
* \brief Global assembler for the monolithic scheme of the non-isothermal
* Richards flow.
*
* <b>Governing equations without vapor diffusion</b>
*
* The energy balance equation is given by
* \f[
* (\rho c_p)^{eff}\dot T -
* \nabla (\mathbf{k}_T^{eff} \nabla T)+\rho^l c_p^l \nabla T \cdot
* \mathbf{v}^l
* = Q_T
* \f]
* with\f$T\f$ the temperature, \f$(\rho c_p)^{eff}\f$ the effective
* volumetric heat
* capacity, \f$\mathbf{k}_T^{eff} \f$
* the effective thermal conductivity, \f$\rho^l\f$ the density of liquid,
* \f$c_p^l\f$ the specific heat capacity of liquid, \f$\mathbf{v}^l\f$ the
* liquid velocity, and \f$Q_T\f$ the point heat source. The effective
* volumetric heat can be considered as a composite of the contributions of
* solid phase and the liquid phase as
* \f[
* (\rho c_p)^{eff} = (1-\phi) \rho^s c_p^s + S^l \phi \rho^l c_p^l
* \f]
* with \f$\phi\f$ the porosity, \f$S^l\f$ the liquid saturation, \f$\rho^s \f$
* the solid density, and \f$c_p^s\f$ the specific heat capacity of solid.
* Similarly, the effective thermal conductivity is given by
* \f[
* \mathbf{k}_T^{eff} = (1-\phi) \mathbf{k}_T^s + S^l \phi k_T^l \mathbf I
* \f]
* where \f$\mathbf{k}_T^s\f$ is the thermal conductivity tensor of solid, \f$
* k_T^l\f$ is the thermal conductivity of liquid, and \f$\mathbf I\f$ is the
* identity tensor.
*
* The mass balance equation is given by
* \f{eqnarray*}{
* \left(S^l\beta - \phi\frac{\partial S}{\partial p_c}\right) \rho^l\dot p
* - S \left( \frac{\partial \rho^l}{\partial T}
* +\rho^l(\alpha_B -S)
* \alpha_T^s
* \right)\dot T\\
* +\nabla (\rho^l \mathbf{v}^l) + S \alpha_B \rho^l \nabla \cdot \dot {\mathbf
* u}= Q_H
* \f}
* where \f$p\f$ is the pore pressure, \f$p_c\f$ is the
* capillary pressure, which is \f$-p\f$ under the single phase assumption,
* \f$\beta\f$ is a composite coefficient by the liquid compressibility and
* solid compressibility, \f$\alpha_B\f$ is the Biot's constant,
* \f$\alpha_T^s\f$ is the linear thermal expansivity of solid, \f$Q_H\f$
* is the point source or sink term, \f$H(S-1)\f$ is the Heaviside function, and
* \f$ \mathbf u\f$ is the displacement. While this process does not contain a fully
* mechanical coupling, simplfied expressions can be given to approximate the latter
* term under certain stress conditions.
* The liquid velocity \f$\mathbf{v}^l\f$ is
* described by the Darcy's law as
* \f[
* \mathbf{v}^l=-\frac{{\mathbf k} k_{ref}}{\mu} (\nabla p - \rho^l \mathbf g)
* \f]
* with \f${\mathbf k}\f$ the intrinsic permeability, \f$k_{ref}\f$ the relative
* permeability, \f$\mathbf g\f$ the gravitational force.
*/
class ThermoRichardsFlowProcess final : public Process
{
public:
ThermoRichardsFlowProcess(
std::string name,
MeshLib::Mesh& mesh,
std::unique_ptr<ProcessLib::AbstractJacobianAssembler>&&
jacobian_assembler,
std::vector<std::unique_ptr<ParameterLib::ParameterBase>> const&
parameters,
unsigned const integration_order,
std::vector<std::vector<std::reference_wrapper<ProcessVariable>>>&&
process_variables,
ThermoRichardsFlowProcessData&& process_data,
SecondaryVariableCollection&& secondary_variables,
bool const use_monolithic_scheme);
//! \name ODESystem interface
//! @{
bool isLinear() const override { return false; }
//! @}
private:
using LocalAssemblerIF = LocalAssemblerInterface;
void initializeConcreteProcess(
NumLib::LocalToGlobalIndexMap const& dof_table,
MeshLib::Mesh const& mesh,
unsigned const integration_order) override;
void setInitialConditionsConcreteProcess(std::vector<GlobalVector*>& x,
double const t,
int const /*process_id*/) override;
void assembleConcreteProcess(const double t, double const dt,
std::vector<GlobalVector*> const& x,
std::vector<GlobalVector*> const& xdot,
int const process_id, GlobalMatrix& M,
GlobalMatrix& K, GlobalVector& b) override;
void assembleWithJacobianConcreteProcess(
const double t, double const dt, std::vector<GlobalVector*> const& x,
std::vector<GlobalVector*> const& xdot, int const process_id,
GlobalMatrix& M, GlobalMatrix& K, GlobalVector& b,
GlobalMatrix& Jac) override;
void postTimestepConcreteProcess(std::vector<GlobalVector*> const& x,
double const t, double const dt,
const int process_id) override;
private:
std::vector<MeshLib::Node*> _base_nodes;
std::unique_ptr<MeshLib::MeshSubset const> _mesh_subset_base_nodes;
ThermoRichardsFlowProcessData _process_data;
std::vector<std::unique_ptr<LocalAssemblerIF>> _local_assemblers;
/// Sparsity pattern for the flow equation, and it is initialized only if
/// the staggered scheme is used.
GlobalSparsityPattern _sparsity_pattern_with_linear_element;
void computeSecondaryVariableConcrete(double const t, double const dt,
std::vector<GlobalVector*> const& x,
GlobalVector const& x_dot,
int const process_id) override;
std::vector<NumLib::LocalToGlobalIndexMap const*> getDOFTables(
const int number_of_processes) const;
MeshLib::PropertyVector<double>* _heat_flux = nullptr;
MeshLib::PropertyVector<double>* _hydraulic_flow = nullptr;
};
} // namespace ThermoRichardsFlow
} // namespace ProcessLib
![swh spinner](/static/img/swh-spinner.gif)
Computing file changes ...