PorousFlowSquarePulsePointEnthalpySource

Point source that adds heat energy at a constant mass flux rate for times between the specified start and end times.

Description

PorousFlowSquarePulsePointEnthalpySource adds a point contribution to the energy (temperature) equation representing the heat carried by fluid injected at a single point, at a constant mass flux mass_flux and a fixed inlet temperature T_in, but only during the time window $var element = document.getElementById("moose-equation-c8d71d3d-9a30-4764-82f8-ccc38f606f07");katex.render("[", element, {displayMode:false,throwOnError:false});$ start_time $var element = document.getElementById("moose-equation-b5d5cc40-1ec5-4218-b930-cfc410991d28");katex.render(",", element, {displayMode:false,throwOnError:false});$ end_time $var element = document.getElementById("moose-equation-491230f8-bd68-4fcd-9351-09a639c273b8");katex.render("]", element, {displayMode:false,throwOnError:false});$ ("square pulse"):

$var element = document.getElementById("moose-equation-d8494a39-4ec5-4af2-b199-fd26f5b15117");katex.render("\\dot{Q} = \\mathrm{factor}(t) \\, \\dot{m} \\, h(P, T_{in})", element, {displayMode:true,throwOnError:false});$

where $var element = document.getElementById("moose-equation-f01d61f6-2663-4e47-8520-913dab25823d");katex.render("h", element, {displayMode:false,throwOnError:false});$ is evaluated by the fp fluid properties user object at the local porepressure and the fixed T_in, and $var element = document.getElementById("moose-equation-034ad128-61b9-4e68-aa88-9e6230258e9a");katex.render("\\mathrm{factor}(t) \\in [0, 1]", element, {displayMode:false,throwOnError:false});$ ramps the mass flux linearly on and off across whichever timestep(s) straddle start_time or end_time, so that the total mass (and hence energy) injected over the pulse is exact regardless of the timestep size used – see PorousFlowSquarePulsePointEnthalpySink for the full enumeration of the ramp cases. Unlike that sink counterpart, this object always uses the fixed T_in rather than the local solution temperature, and its residual carries the opposite sign convention (it adds, rather than removes, heat energy for a positive mass_flux).

The constructor errors if end_time is not strictly greater than start_time.

Example Input Syntax

[DiracKernels<<<{"href": "../../syntax/DiracKernels/index.html"}>>>]
  [source_h]
    type = PorousFlowSquarePulsePointEnthalpySource<<<{"description": "Point source that adds heat energy at a constant mass flux rate for times between the specified start and end times.", "href": "PorousFlowSquarePulsePointEnthalpySource.html"}>>>
    variable<<<{"description": "The name of the variable that this residual object operates on"}>>> = temperature
    mass_flux<<<{"description": "The mass flux at this point in kg/s (positive is flux in, negative is flux out)"}>>> = 0.01
    fp<<<{"description": "The name of the user object used to calculate the fluid properties of the injected fluid"}>>> = simple_fluid
    pressure<<<{"description": "Pressure used to calculate the injected fluid enthalpy (measured in Pa)"}>>> = pressure
    T_in<<<{"description": "The incoming fluid temperature"}>>> = 320
    point<<<{"description": "The x,y,z coordinates of the point source (sink)"}>>> = '0.5 0.5 0'
    start_time<<<{"description": "The time at which the source will start (Default is 0)"}>>> = 250
    end_time<<<{"description": "The time at which the source will end (Default is 1e30)"}>>> = 650
  []
[]

Input Parameters

T_inThe incoming fluid temperature
C++ Type:Real
Unit:(no unit assumed)
Controllable:No
Description:The incoming fluid temperature
fpThe name of the user object used to calculate the fluid properties of the injected fluid
C++ Type:UserObjectName
Controllable:No
Description:The name of the user object used to calculate the fluid properties of the injected fluid
mass_fluxThe mass flux at this point in kg/s (positive is flux in, negative is flux out)
C++ Type:Real
Unit:(no unit assumed)
Controllable:No
Description:The mass flux at this point in kg/s (positive is flux in, negative is flux out)
pointThe x,y,z coordinates of the point source (sink)
C++ Type:libMesh::Point
Controllable:No
Description:The x,y,z coordinates of the point source (sink)
pressurePressure used to calculate the injected fluid enthalpy (measured in Pa)
C++ Type:std::vector<VariableName>
Unit:(no unit assumed)
Controllable:No
Description:Pressure used to calculate the injected fluid enthalpy (measured in Pa)
variableThe name of the variable that this residual object operates on
C++ Type:NonlinearVariableName
Unit:(no unit assumed)
Controllable:No
Description:The name of the variable that this residual object operates on

Required Parameters

allow_moving_sourcesFalseIf true, allow Dirac sources to move, even if the mesh does not move, during the simulation.
Default:False
C++ Type:bool
Controllable:No
Description:If true, allow Dirac sources to move, even if the mesh does not move, during the simulation.
blockThe list of blocks (ids or names) that this object will be applied
C++ Type:std::vector<SubdomainName>
Controllable:No
Description:The list of blocks (ids or names) that this object will be applied
end_time1e+30The time at which the source will end (Default is 1e30)
Default:1e+30
C++ Type:Real
Unit:(no unit assumed)
Controllable:No
Description:The time at which the source will end (Default is 1e30)
point_not_found_behaviorIGNOREBy default (IGNORE), it is ignored if an added point cannot be located in the specified subdomains. If this option is set to ERROR, this situation will result in an error. If this option is set to WARNING, then a warning will be issued.
Default:IGNORE
C++ Type:MooseEnum
Options:ERROR, WARNING, IGNORE
Controllable:No
Description:By default (IGNORE), it is ignored if an added point cannot be located in the specified subdomains. If this option is set to ERROR, this situation will result in an error. If this option is set to WARNING, then a warning will be issued.
start_time0The time at which the source will start (Default is 0)
Default:0
C++ Type:Real
Unit:(no unit assumed)
Controllable:No
Description:The time at which the source will start (Default is 0)

Optional Parameters

absolute_value_vector_tagsThe tags for the vectors this residual object should fill with the absolute value of the residual contribution
C++ Type:std::vector<TagName>
Controllable:No
Description:The tags for the vectors this residual object should fill with the absolute value of the residual contribution
extra_matrix_tagsThe extra tags for the matrices this Kernel should fill
C++ Type:std::vector<TagName>
Controllable:No
Description:The extra tags for the matrices this Kernel should fill
extra_vector_tagsThe extra tags for the vectors this Kernel should fill
C++ Type:std::vector<TagName>
Controllable:No
Description:The extra tags for the vectors this Kernel should fill
matrix_onlyFalseWhether this object is only doing assembly to matrices (no vectors)
Default:False
C++ Type:bool
Controllable:No
Description:Whether this object is only doing assembly to matrices (no vectors)
matrix_tagssystemThe tag for the matrices this Kernel should fill
Default:system
C++ Type:MultiMooseEnum
Options:nontime, system
Controllable:No
Description:The tag for the matrices this Kernel should fill
vector_tagsnontimeThe tag for the vectors this Kernel should fill
Default:nontime
C++ Type:MultiMooseEnum
Options:nontime, time
Controllable:No
Description:The tag for the vectors this Kernel should fill

Contribution To Tagged Field Data Parameters

control_tagsAdds user-defined labels for accessing object parameters via control logic.
C++ Type:std::vector<std::string>
Controllable:No
Description:Adds user-defined labels for accessing object parameters via control logic.
drop_duplicate_pointsTrueBy default points added to a DiracKernel are dropped if a point at the same locationhas been added before. If this option is set to false duplicate points are retainedand contribute to residual and Jacobian.
Default:True
C++ Type:bool
Controllable:No
Description:By default points added to a DiracKernel are dropped if a point at the same locationhas been added before. If this option is set to false duplicate points are retainedand contribute to residual and Jacobian.
enableTrueSet the enabled status of the MooseObject.
Default:True
C++ Type:bool
Controllable:Yes
Description:Set the enabled status of the MooseObject.
implicitTrueDetermines whether this object is calculated using an implicit or explicit form
Default:True
C++ Type:bool
Controllable:No
Description:Determines whether this object is calculated using an implicit or explicit form
search_methodnearest_node_connected_sidesChoice of search algorithm. All options begin by finding the nearest node in the primary boundary to a query point in the secondary boundary. In the default nearest_node_connected_sides algorithm, primary boundary elements are searched iff that nearest node is one of their nodes. This is fast to determine via a pregenerated node-to-elem map and is robust on conforming meshes. In the optional all_proximate_sides algorithm, primary boundary elements are searched iff they touch that nearest node, even if they are not topologically connected to it. This is more CPU-intensive but is necessary for robustness on any boundary surfaces which has disconnections (such as Flex IGA meshes) or non-conformity (such as hanging nodes in adaptively h-refined meshes).
Default:nearest_node_connected_sides
C++ Type:MooseEnum
Options:nearest_node_connected_sides, all_proximate_sides
Controllable:No
Description:Choice of search algorithm. All options begin by finding the nearest node in the primary boundary to a query point in the secondary boundary. In the default nearest_node_connected_sides algorithm, primary boundary elements are searched iff that nearest node is one of their nodes. This is fast to determine via a pregenerated node-to-elem map and is robust on conforming meshes. In the optional all_proximate_sides algorithm, primary boundary elements are searched iff they touch that nearest node, even if they are not topologically connected to it. This is more CPU-intensive but is necessary for robustness on any boundary surfaces which has disconnections (such as Flex IGA meshes) or non-conformity (such as hanging nodes in adaptively h-refined meshes).
seed0The seed for the master random number generator
Default:0
C++ Type:unsigned int
Controllable:No
Description:The seed for the master random number generator
use_displaced_meshFalseWhether or not this object should use the displaced mesh for computation. Note that in the case this is true but no displacements are provided in the Mesh block the undisplaced mesh will still be used.
Default:False
C++ Type:bool
Controllable:No
Description:Whether or not this object should use the displaced mesh for computation. Note that in the case this is true but no displacements are provided in the Mesh block the undisplaced mesh will still be used.

Advanced Parameters

prop_getter_suffixAn optional suffix parameter that can be appended to any attempt to retrieve/get material properties. The suffix will be prepended with a '_' character.
C++ Type:MaterialPropertyName
Unit:(no unit assumed)
Controllable:No
Description:An optional suffix parameter that can be appended to any attempt to retrieve/get material properties. The suffix will be prepended with a '_' character.
use_interpolated_stateFalseFor the old and older state use projected material properties interpolated at the quadrature points. To set up projection use the ProjectedStatefulMaterialStorageAction.
Default:False
C++ Type:bool
Controllable:No
Description:For the old and older state use projected material properties interpolated at the quadrature points. To set up projection use the ProjectedStatefulMaterialStorageAction.

Material Property Retrieval Parameters

Input Files

(test/tests/dirackernels/square_pulse_point_enthalpy_source/inp.i)
(test/tests/dirackernels/jacobian/inp_scaled.i)
(test/tests/dirackernels/jacobian/inp.i)

(test/tests/dirackernels/square_pulse_point_enthalpy_source/inp.i)

# Analytic check for the PorousFlowSquarePulsePointEnthalpySource DiracKernel.
#
# This class adds heat energy at a constant mass flux and fixed inlet
# temperature T_in, but only during [start_time, end_time], with the mass
# flux ramped smoothly across the timestep(s) that straddle start_time or
# end_time so that the total added mass (and hence energy) over the whole
# pulse is exact regardless of the timestep size. There are six cases for
# how (start_time, end_time) relate to the current step (t-dt, t) -- see the
# comment block in PorousFlowSquarePulsePointEnthalpySource::computeQpResidual.
# Denoting that fraction "factor" (0 <= factor <= 1):
#   1) t < start_time                        : factor = 0            ("before")
#   2) t-dt < start_time <= t <= end_time     : factor = (t-start_time)/dt   ("ramp up")
#   3) t-dt < start_time and t > end_time     : factor = (end_time-start_time)/dt  (pulse
#                                                entirely inside this one step)
#   4) start_time <= t-dt and t <= end_time   : factor = 1            ("fully inside")
#   5) start_time <= t-dt < end_time < t      : factor = (end_time-(t-dt))/dt ("ramp down")
#   6) t-dt >= end_time                       : factor = 0            ("after")
# Cases 1,2,4,5,6 are exercised directly below by choosing a constant dt=100
# with start_time=250, end_time=650 (so the pulse spans several steps).
# Case 3 shares the same "t-dt < start_time" branch as case 2 (differing only
# in the trailing else), so is not separately re-derived here; its formula is
# identical in form to case 2's, just with the pulse's own (end_time -
# start_time) in place of (t - start_time).
#
# Using the same energy-conservation argument as point_enthalpy_sink/inp.i,
# and since T_in is a fixed input (h = cv*T_in is exactly known, see
# EnthalpySink.md), the exact change in the domain's total heat energy over
# the timestep ending at t_n is:
#   +mass_flux * cv * T_in * factor(t_n) * dt
# (the sign, opposite to PointEnthalpySink/PorousFlowSquarePulsePointEnthalpySink,
# was confirmed empirically: this class's computeQpResidual has an explicit
# minus sign that the Sink classes do not have, and positive mass_flux here
# increases the domain's heat energy, consistent with the class being a
# "Source".)
#
# "residual" = abs(delta_E - mass_flux*cv*T_in*factor(t)*dt) should be ~0 (to
# solver/roundoff tolerance) at every timestep except t=0 (see
# point_enthalpy_sink/inp.i for why the "initial" row is an exception).

[Mesh]
  type = GeneratedMesh
  dim = 2
  nx = 1
  ny = 1
[]

[GlobalParams]
  PorousFlowDictator = dictator
[]

[Variables]
  [pressure]
    initial_condition = 1e6
  []
  [temperature]
    initial_condition = 300
  []
[]

[FluidProperties]
  [simple_fluid]
    type = SimpleFluidProperties
    bulk_modulus = 2e9
    density0 = 1000
    thermal_expansion = 0
    cv = 4000
    porepressure_coefficient = 0 # forces h_from_p_T(P,T) = cv*T exactly
  []
[]

[PorousFlowUnsaturated]
  porepressure = pressure
  temperature = temperature
  coupling_type = ThermoHydro
  gravity = '0 0 0'
  fp = simple_fluid
  add_darcy_aux = false
[]

[Materials]
  [thermal_conductivity]
    type = PorousFlowThermalConductivityIdeal
    dry_thermal_conductivity = '2.5 0 0  0 2.5 0  0 0 2.5'
  []
  [permeability]
    type = PorousFlowPermeabilityConst
    permeability = '1e-15 0 0  0 1e-15 0  0 0 1e-15'
  []
  [rock_heat]
    type = PorousFlowMatrixInternalEnergy
    specific_heat_capacity = 830.0
    density = 2750
  []
  [porosity]
    type = PorousFlowPorosityConst
    porosity = 0.1
  []
[]

[DiracKernels]
  [source_h]
    type = PorousFlowSquarePulsePointEnthalpySource
    variable = temperature
    mass_flux = 0.01
    fp = simple_fluid
    pressure = pressure
    T_in = 320
    point = '0.5 0.5 0'
    start_time = 250
    end_time = 650
  []
[]

[Postprocessors]
  [heat_energy]
    type = PorousFlowHeatEnergy
    phase = '0'
    execute_on = 'initial timestep_end'
  []
  [delta_E]
    type = ChangeOverTimePostprocessor
    postprocessor = heat_energy
    execute_on = 'initial timestep_end'
  []
  [dt_pp]
    type = TimestepSize
    execute_on = 'initial timestep_end'
  []
  [factor]
    type = ParsedPostprocessor
    pp_names = 'dt_pp'
    use_t = true
    expression = 'if(t < 250, 0,
                  if(t - dt_pp >= 650, 0,
                  if(t - dt_pp < 250,
                    if(t <= 650, (t-250)/dt_pp, (650-250)/dt_pp),
                    if(t <= 650, 1, (650-(t-dt_pp))/dt_pp))))'
    execute_on = 'initial timestep_end'
  []
  [residual]
    type = ParsedPostprocessor
    pp_names = 'delta_E factor dt_pp'
    expression = 'abs(delta_E - 0.01 * 4000 * 320 * factor * dt_pp)'
    execute_on = 'initial timestep_end'
  []
[]

[Preconditioning]
  [preferred]
    type = SMP
    full = true
    petsc_options_iname = '-pc_type'
    petsc_options_value = ' lu     '
  []
[]

[Executioner]
  type = Transient
  solve_type = Newton
  nl_abs_tol = 1e-10
  [TimeStepper]
    type = ConstantDT
    dt = 100
  []
  end_time = 800
[]

[Outputs]
  csv = true
  execute_on = 'initial timestep_end'
[]

(test/tests/dirackernels/square_pulse_point_enthalpy_source/inp.i)

# Analytic check for the PorousFlowSquarePulsePointEnthalpySource DiracKernel.
#
# This class adds heat energy at a constant mass flux and fixed inlet
# temperature T_in, but only during [start_time, end_time], with the mass
# flux ramped smoothly across the timestep(s) that straddle start_time or
# end_time so that the total added mass (and hence energy) over the whole
# pulse is exact regardless of the timestep size. There are six cases for
# how (start_time, end_time) relate to the current step (t-dt, t) -- see the
# comment block in PorousFlowSquarePulsePointEnthalpySource::computeQpResidual.
# Denoting that fraction "factor" (0 <= factor <= 1):
#   1) t < start_time                        : factor = 0            ("before")
#   2) t-dt < start_time <= t <= end_time     : factor = (t-start_time)/dt   ("ramp up")
#   3) t-dt < start_time and t > end_time     : factor = (end_time-start_time)/dt  (pulse
#                                                entirely inside this one step)
#   4) start_time <= t-dt and t <= end_time   : factor = 1            ("fully inside")
#   5) start_time <= t-dt < end_time < t      : factor = (end_time-(t-dt))/dt ("ramp down")
#   6) t-dt >= end_time                       : factor = 0            ("after")
# Cases 1,2,4,5,6 are exercised directly below by choosing a constant dt=100
# with start_time=250, end_time=650 (so the pulse spans several steps).
# Case 3 shares the same "t-dt < start_time" branch as case 2 (differing only
# in the trailing else), so is not separately re-derived here; its formula is
# identical in form to case 2's, just with the pulse's own (end_time -
# start_time) in place of (t - start_time).
#
# Using the same energy-conservation argument as point_enthalpy_sink/inp.i,
# and since T_in is a fixed input (h = cv*T_in is exactly known, see
# EnthalpySink.md), the exact change in the domain's total heat energy over
# the timestep ending at t_n is:
#   +mass_flux * cv * T_in * factor(t_n) * dt
# (the sign, opposite to PointEnthalpySink/PorousFlowSquarePulsePointEnthalpySink,
# was confirmed empirically: this class's computeQpResidual has an explicit
# minus sign that the Sink classes do not have, and positive mass_flux here
# increases the domain's heat energy, consistent with the class being a
# "Source".)
#
# "residual" = abs(delta_E - mass_flux*cv*T_in*factor(t)*dt) should be ~0 (to
# solver/roundoff tolerance) at every timestep except t=0 (see
# point_enthalpy_sink/inp.i for why the "initial" row is an exception).

[Mesh]
  type = GeneratedMesh
  dim = 2
  nx = 1
  ny = 1
[]

[GlobalParams]
  PorousFlowDictator = dictator
[]

[Variables]
  [pressure]
    initial_condition = 1e6
  []
  [temperature]
    initial_condition = 300
  []
[]

[FluidProperties]
  [simple_fluid]
    type = SimpleFluidProperties
    bulk_modulus = 2e9
    density0 = 1000
    thermal_expansion = 0
    cv = 4000
    porepressure_coefficient = 0 # forces h_from_p_T(P,T) = cv*T exactly
  []
[]

[PorousFlowUnsaturated]
  porepressure = pressure
  temperature = temperature
  coupling_type = ThermoHydro
  gravity = '0 0 0'
  fp = simple_fluid
  add_darcy_aux = false
[]

[Materials]
  [thermal_conductivity]
    type = PorousFlowThermalConductivityIdeal
    dry_thermal_conductivity = '2.5 0 0  0 2.5 0  0 0 2.5'
  []
  [permeability]
    type = PorousFlowPermeabilityConst
    permeability = '1e-15 0 0  0 1e-15 0  0 0 1e-15'
  []
  [rock_heat]
    type = PorousFlowMatrixInternalEnergy
    specific_heat_capacity = 830.0
    density = 2750
  []
  [porosity]
    type = PorousFlowPorosityConst
    porosity = 0.1
  []
[]

[DiracKernels]
  [source_h]
    type = PorousFlowSquarePulsePointEnthalpySource
    variable = temperature
    mass_flux = 0.01
    fp = simple_fluid
    pressure = pressure
    T_in = 320
    point = '0.5 0.5 0'
    start_time = 250
    end_time = 650
  []
[]

[Postprocessors]
  [heat_energy]
    type = PorousFlowHeatEnergy
    phase = '0'
    execute_on = 'initial timestep_end'
  []
  [delta_E]
    type = ChangeOverTimePostprocessor
    postprocessor = heat_energy
    execute_on = 'initial timestep_end'
  []
  [dt_pp]
    type = TimestepSize
    execute_on = 'initial timestep_end'
  []
  [factor]
    type = ParsedPostprocessor
    pp_names = 'dt_pp'
    use_t = true
    expression = 'if(t < 250, 0,
                  if(t - dt_pp >= 650, 0,
                  if(t - dt_pp < 250,
                    if(t <= 650, (t-250)/dt_pp, (650-250)/dt_pp),
                    if(t <= 650, 1, (650-(t-dt_pp))/dt_pp))))'
    execute_on = 'initial timestep_end'
  []
  [residual]
    type = ParsedPostprocessor
    pp_names = 'delta_E factor dt_pp'
    expression = 'abs(delta_E - 0.01 * 4000 * 320 * factor * dt_pp)'
    execute_on = 'initial timestep_end'
  []
[]

[Preconditioning]
  [preferred]
    type = SMP
    full = true
    petsc_options_iname = '-pc_type'
    petsc_options_value = ' lu     '
  []
[]

[Executioner]
  type = Transient
  solve_type = Newton
  nl_abs_tol = 1e-10
  [TimeStepper]
    type = ConstantDT
    dt = 100
  []
  end_time = 800
[]

[Outputs]
  csv = true
  execute_on = 'initial timestep_end'
[]

(test/tests/dirackernels/jacobian/inp_scaled.i)

# Companion to inp.i: the same eight RTES enthalpy DiracKernels, but on a deliberately
# non-dimensionalised problem so that EVERY block of the Jacobian is O(1).
#
# Why a second input exists. PetscJacobianTester gates on ratio_tol, i.e.
# ||J - Jfd||_F / ||J||_F. In inp.i the full PorousFlowUnsaturated action is present, and
# PorousFlowMatrixInternalEnergy (density = 2750, specific_heat_capacity = 830) contributes an
# energy time-derivative diagonal of order 1e5 per node. That single term sets ||J||_F, so the
# absolute error budget at ratio_tol = 1e-7 is ~1e-2 -- while the pressure-block entries these
# kernels actually contribute are mass_flux * dh/dp ~ 1e-5. Three to four orders of magnitude
# separate the two, so a wrong (or missing) pressure off-diagonal is invisible there:
#
#   * the dh/dT diagonals of PointEnthalpySink, PorousFlowSquarePulsePointEnthalpySink and
#     PointEnthalpySourceSinkFromFunction perturb the norm by ~8      -> caught by inp.i
#   * the leading minus of PorousFlowSquarePulsePointEnthalpySource   -> ~6e-6, invisible there
#   * the dh/dp chain rule of EnthalpySink / FunctionEnthalpySink     -> ~6e-6, invisible there
#
# This input follows the pattern of MOOSE's own
# modules/porous_flow/test/tests/jacobian/line_sink02.i: it keeps the PorousFlowDictator and the
# handful of materials the DiracKernels genuinely require, but replaces the action's bulk mass and
# energy kernels with a plain TimeDerivative on each variable, and scales the fluid properties,
# mass fluxes and initial conditions to O(1). With no term dominating, ||J||_F is O(0.1) and the
# pressure-block entries are of the same order, so each defect above moves the ratio by O(1).
#
# porepressure_coefficient is left at its physical default of 1: SimpleFluidProperties computes
# h = e + porepressure_coefficient * p / rho, so setting it to 0 (as the analytic residual tests
# elsewhere in test/tests/dirackernels/ do) would zero dh/dp everywhere and silently remove the
# very term under test.

[Mesh]
  type = GeneratedMesh
  dim = 3
  nx = 1
  ny = 1
  nz = 1
[]

[GlobalParams]
  PorousFlowDictator = dictator
[]

[Variables]
  [pressure]
  []
  [temperature]
  []
[]

[ICs]
  # O(1) ranges, as in line_sink02.i -- these are non-dimensional values chosen to condition the
  # Jacobian, not physical pressures and temperatures.
  [pressure_ic]
    type = RandomIC
    variable = pressure
    min = 0.2
    max = 1.0
  []
  [temperature_ic]
    type = RandomIC
    variable = temperature
    min = 1.0
    max = 2.0
  []
[]

[Kernels]
  # Plain time derivatives in place of the PorousFlow action's mass/energy kernels: they keep the
  # diagonal O(1) instead of O(1e5), which is the whole point of this input.
  [dummy_pressure]
    type = TimeDerivative
    variable = pressure
  []
  [dummy_temperature]
    type = TimeDerivative
    variable = temperature
  []
[]

[FluidProperties]
  [simple_fluid]
    type = SimpleFluidProperties
    bulk_modulus = 1.5
    density0 = 1
    thermal_expansion = 0
    viscosity = 1
    cv = 1.1
    # porepressure_coefficient left at its default of 1 -- see header comment.
  []
[]

[UserObjects]
  [dictator]
    type = PorousFlowDictator
    porous_flow_vars = 'pressure temperature'
    number_fluid_phases = 1
    number_fluid_components = 1
  []
  [enthalpy_sink_uo]
    type = PorousFlowSumQuantity
  []
  [function_enthalpy_sink_uo]
    type = PorousFlowSumQuantity
  []
  [source_sink_fn_uo]
    type = PorousFlowSumQuantity
  []
  [mass_source_sink_fn_uo]
    type = PorousFlowSumQuantity
  []
[]

[Materials]
  [temperature]
    type = PorousFlowTemperature
    temperature = temperature
  []
  [ppss]
    type = PorousFlow1PhaseFullySaturated
    porepressure = pressure
  []
  [massfrac]
    type = PorousFlowMassFraction
  []
  [simple_fluid]
    type = PorousFlowSingleComponentFluid
    fp = simple_fluid
    phase = 0
  []
  [permeability]
    type = PorousFlowPermeabilityConst
    permeability = '1 0 0  0 1 0  0 0 1'
  []
  [porosity]
    type = PorousFlowPorosityConst
    porosity = 0.1
  []
[]

[Functions]
  [t_in_fn]
    type = ParsedFunction
    expression = '1.4 + 0.2*y'
  []
  [mass_flux_fn]
    type = ParsedFunction
    expression = '0.1'
  []
  [temperature_fn]
    type = ParsedFunction
    expression = '0'
  []
[]

[Postprocessors]
  [mass_flux_pp]
    type = ConstantPostprocessor
    value = 0.1
  []
[]

[DiracKernels]
  [enthalpy_sink]
    type = EnthalpySink
    variable = temperature
    fp = simple_fluid
    pressure = pressure
    T_in = 1.5
    fluid_phase = 0
    line_base = '1 0.1 0.1 0.1'
    line_direction = '0 1 0'
    line_length = 0.2
    # Deliberately NON-flat: PorousFlowPolyLineSink builds its outflow derivative from
    # _sink_func.sampleDerivative(), which is identically 0 for a flat table -- and that would
    # multiply the h * outflowp half of the chain-rule fix by literal zero.
    p_or_t_vals = '0 2'
    fluxes = '0.04 0.06'
    SumQuantityUO = enthalpy_sink_uo
  []
  [function_enthalpy_sink]
    type = FunctionEnthalpySink
    variable = temperature
    fp = simple_fluid
    pressure = pressure
    function = t_in_fn
    fluid_phase = 0
    line_base = '1 0.6 0.1 0.1'
    line_direction = '0 1 0'
    line_length = 0.2
    p_or_t_vals = '0 2'
    fluxes = '0.04 0.06'
    SumQuantityUO = function_enthalpy_sink_uo
  []
  [point_enthalpy_sink]
    type = PointEnthalpySink
    variable = temperature
    mass_flux = mass_flux_pp
    fp = simple_fluid
    pressure = pressure
    point = '0.2 0.8 0.2'
  []
  [square_pulse_sink]
    type = PorousFlowSquarePulsePointEnthalpySink
    variable = temperature
    mass_flux = 0.1
    fp = simple_fluid
    pressure = pressure
    point = '0.5 0.8 0.2'
  []
  [square_pulse_source]
    type = PorousFlowSquarePulsePointEnthalpySource
    variable = temperature
    mass_flux = 0.1
    fp = simple_fluid
    pressure = pressure
    T_in = 1.7
    point = '0.8 0.8 0.2'
  []
  [source_sink_from_function]
    type = PointEnthalpySourceSinkFromFunction
    variable = temperature
    SumQuantityUO = source_sink_fn_uo
    mass_flux_function = mass_flux_fn
    temperature_function = temperature_fn
    pressure = pressure
    fp = simple_fluid
    point = '0.2 0.2 0.8'
  []
  [enthalpy_source_from_function]
    type = PointEnthalpySourceFromFunction
    variable = temperature
    mass_flux = mass_flux_pp
    fp = simple_fluid
    pressure = pressure
    function = t_in_fn
    point = '0.5 0.2 0.8'
  []
  [mass_source_sink_from_function]
    type = PointSourceSinkFromFunction
    variable = pressure
    mass_flux_function = mass_flux_fn
    SumQuantityUO = mass_source_sink_fn_uo
    point = '0.8 0.2 0.8'
  []
[]

[Preconditioning]
  [check]
    type = SMP
    full = true
  []
[]

[Executioner]
  type = Transient
  solve_type = Newton
  dt = 1
  end_time = 1
[]

[Outputs]
  exodus = false
[]

(test/tests/dirackernels/jacobian/inp.i)

# Jacobian test for the RTES enthalpy DiracKernels reviewed for missing/incorrect Jacobian terms.
#
# How this is run: the PetscJacobianTester in ./tests injects -snes_test_jacobian itself; it is NOT
# supplied by the [Preconditioning] block below, and -snes_type=test has not existed since PETSc
# 3.9. Because run_sim defaults to False the tester also adds
#   -snes_type ksponly -ksp_type preonly -pc_type none -snes_convergence_test skip
# so there is NO nonlinear solve: with Executioner/num_steps=1, exactly one Jacobian is checked, at
# the initial state. One consequence is that only factor = 1.0 of the SquarePulse kernels' six-case
# pulse ramp is ever exercised -- which is why that ramp now lives in a single pulseFactor() helper
# per class rather than being copied into each of computeQpResidual/Jacobian/OffDiagJacobian.
#
# The defects under test:
#   - EnthalpySink, FunctionEnthalpySink: missing computeQpBaseOutflowJacobian override (were
#     missing both the fluid-enthalpy factor and its dh/dp chain-rule term).
#   - PorousFlowSquarePulsePointEnthalpySource: off-diagonal Jacobian was missing a leading minus
#     relative to its residual.
#   - PointEnthalpySink, PorousFlowSquarePulsePointEnthalpySink, PointEnthalpySourceSinkFromFunction
#     (extraction branch): computeQpJacobian() returned 0, discarding a genuine dh/dT diagonal
#     term (these are applied to the temperature variable, and h depends on the local
#     PorousFlow_temperature_qp).
#
# All of the analytic residual tests elsewhere in test/tests/dirackernels/ deliberately set
# SimpleFluidProperties' porepressure_coefficient=0 to get closed-form residuals -- but that also
# zeroes dh/dp everywhere, and (combined with most of those tests using solve=false) means none of
# them exercise a Jacobian at all. This test leaves porepressure_coefficient at its physical
# default of 1.0 so dh/dp is live.
#
# This input keeps the full PorousFlowUnsaturated physics, which makes it representative but also
# means PorousFlowMatrixInternalEnergy's ~1e5 energy diagonal dominates ||J||_F, leaving too coarse
# an absolute budget to see the O(1e-5) pressure-block entries. The three dh/dT diagonals are
# caught here; the pressure-block terms are covered by the companion inp_scaled.i, which is the
# same eight kernels on an O(1)-scaled problem. See that file's header.

[Mesh]
  type = GeneratedMesh
  dim = 3
  nx = 1
  ny = 1
  nz = 1
[]

[GlobalParams]
  PorousFlowDictator = dictator
[]

[Variables]
  [pressure]
  []
  [temperature]
  []
[]

[ICs]
  [pressure_ic]
    type = RandomIC
    variable = pressure
    min = 1e5
    max = 2e6
  []
  [temperature_ic]
    type = RandomIC
    variable = temperature
    min = 280
    max = 350
  []
[]

[FluidProperties]
  [simple_fluid]
    type = SimpleFluidProperties
    bulk_modulus = 2e9
    density0 = 1000
    thermal_expansion = 0
    cv = 4000
    # porepressure_coefficient is left at its physical default (1.0) -- see header comment.
  []
[]

[PorousFlowUnsaturated]
  porepressure = pressure
  temperature = temperature
  coupling_type = ThermoHydro
  gravity = '0 0 0'
  fp = simple_fluid
  add_darcy_aux = false
[]

[Materials]
  [thermal_conductivity]
    type = PorousFlowThermalConductivityIdeal
    dry_thermal_conductivity = '2.5 0 0  0 2.5 0  0 0 2.5'
  []
  [permeability]
    type = PorousFlowPermeabilityConst
    permeability = '1e-15 0 0  0 1e-15 0  0 0 1e-15'
  []
  [rock_heat]
    type = PorousFlowMatrixInternalEnergy
    specific_heat_capacity = 830.0
    density = 2750
  []
  [porosity]
    type = PorousFlowPorosityConst
    porosity = 0.1
  []
[]

[UserObjects]
  [enthalpy_sink_uo]
    type = PorousFlowSumQuantity
  []
  [function_enthalpy_sink_uo]
    type = PorousFlowSumQuantity
  []
  [source_sink_fn_uo]
    type = PorousFlowSumQuantity
  []
  [mass_source_sink_fn_uo]
    type = PorousFlowSumQuantity
  []
[]

[Functions]
  [t_in_fn]
    type = ParsedFunction
    expression = '310 + 5*y'
  []
  [mass_flux_fn]
    type = ParsedFunction
    expression = '0.01'
  []
  [temperature_fn]
    type = ParsedFunction
    expression = '0'
  []
[]

[DiracKernels]
  [enthalpy_sink]
    type = EnthalpySink
    variable = temperature
    fp = simple_fluid
    pressure = pressure
    T_in = 320
    fluid_phase = 0
    line_base = '1 0.1 0.1 0.1'
    line_direction = '0 1 0'
    line_length = 0.2
    # Deliberately non-flat: a flat table makes _sink_func.sampleDerivative() identically 0, which
    # multiplies the h * outflowp half of the chain-rule fix by literal zero.
    p_or_t_vals = '0 1e10'
    fluxes = '0.04 0.06'
    SumQuantityUO = enthalpy_sink_uo
  []
  [function_enthalpy_sink]
    type = FunctionEnthalpySink
    variable = temperature
    fp = simple_fluid
    pressure = pressure
    function = t_in_fn
    fluid_phase = 0
    line_base = '1 0.6 0.1 0.1'
    line_direction = '0 1 0'
    line_length = 0.2
    p_or_t_vals = '0 1e10'
    fluxes = '0.04 0.06'
    SumQuantityUO = function_enthalpy_sink_uo
  []
  [point_enthalpy_sink]
    type = PointEnthalpySink
    variable = temperature
    mass_flux = mass_flux_pp
    fp = simple_fluid
    pressure = pressure
    point = '0.2 0.8 0.2'
  []
  [square_pulse_sink]
    type = PorousFlowSquarePulsePointEnthalpySink
    variable = temperature
    mass_flux = 0.01
    fp = simple_fluid
    pressure = pressure
    point = '0.5 0.8 0.2'
  []
  [square_pulse_source]
    type = PorousFlowSquarePulsePointEnthalpySource
    variable = temperature
    mass_flux = 0.01
    fp = simple_fluid
    pressure = pressure
    T_in = 330
    point = '0.8 0.8 0.2'
  []
  [enthalpy_source_from_function]
    # Has a hand-written nonzero pressure off-diagonal and a hard-coded return 0. diagonal; neither
    # was verified by any test before this one.
    type = PointEnthalpySourceFromFunction
    variable = temperature
    mass_flux = mass_flux_pp
    fp = simple_fluid
    pressure = pressure
    function = t_in_fn
    point = '0.5 0.2 0.8'
  []
  [mass_source_sink_from_function]
    # Contributes no Jacobian at all (the residual is a pure Function of time and position); this
    # pins that down rather than leaving it asserted only in a comment.
    type = PointSourceSinkFromFunction
    variable = pressure
    mass_flux_function = mass_flux_fn
    SumQuantityUO = mass_source_sink_fn_uo
    point = '0.8 0.2 0.8'
  []
  [source_sink_from_function]
    type = PointEnthalpySourceSinkFromFunction
    variable = temperature
    SumQuantityUO = source_sink_fn_uo
    mass_flux_function = mass_flux_fn
    temperature_function = temperature_fn
    pressure = pressure
    fp = simple_fluid
    point = '0.2 0.2 0.8'
  []
[]

[Postprocessors]
  [mass_flux_pp]
    type = ConstantPostprocessor
    value = 0.01
  []
[]

[Preconditioning]
  [check]
    type = SMP
    full = true
  []
[]

[Executioner]
  type = Transient
  solve_type = Newton
  dt = 1
  end_time = 2
[]

[Outputs]
  exodus = false
[]

Description
Example Input Syntax
Input Parameters
Input Files

Theory

Usage

Development