How to use previous timestep variable (scalar) value in FunctorMaterial or Function for ScalarTransport source term?

[Traiwit]

Check these boxes if you have followed the posting rules.

• Q&A Navier Stokes is the most appropriate category for my question
• I have consulted the posting Guidelines on the Discussions front page
• I have searched the Discussions forum and my question has not been asked before
• I have searched the MOOSE website and the documentation does not answer my question
• I have formatted my post following the posting guidelines (avoid screenshots if possible, triple back quotes before/after pasted text, etc)

Question

Hi all,

I’m working with the ScalarTransport module under NavierStokes to simulate radon transport using a passive scalar. I’d like to define a source term that depends on the previous timestep's value of the scalar variable (scalar) — specifically something like:

N_Ra(n+1) = N_Ra(n) + [ (M / 1000) * U * 12400 * 0.25 - 2.1e-6 * N_Ra(n) ]

Where N_Ra(n) is the scalar from the previous timestep.

I'm currently specifying the source term via a GenericFunctorMaterial:

[Physics/NavierStokes]
  [ScalarTransport]
    [Radon]
      coupled_flow_physics = 'flow'
      passive_scalar_names = 'scalar'
      passive_scalar_diffusivity = ${diffusion_coff}
      passive_scalar_source = 'radon_source'
      initial_scalar_variables = '0'
      passive_scalar_inlet_types = 'fixed-value'
      passive_scalar_inlet_functors = '0'
      block = '1 2 3'
    []
  []
[]

[FunctorMaterials]
  [radon_source]
    type = GenericFunctorMaterial
    prop_names = 'radon_source'
    prop_values = Radon_state_based
  []
[]

Q: Is there a built-in way to access the previous timestep value of a variable (like scalar) in a Function or GenericFunctorMaterial?

a quick note: Radon_state_based is from my function which take the value from element_based .csv to do some calculation

      [./Radon_state_based]
        type = StateBasedValueFromCSV
        read_prop_user_object = 'uobj_read_state_fromCsv'
        column_number = '0'
        read_type = element
        state_values = '1 2 3 4 5 6 7 8 9 10'
        replacement_values = '0 0 0 300 0 0 0 0 0 0'
      [../]

Thank you guys!

Additional information

Mesh size and type:
Reynolds number:
Discretization (finite element CG/DG, finite volume, etc):
Models (turbulence, porous media, etc):
Solver method (fully coupled, segregated, multiapps, etc):
Base input you started from:

[GiudGiud]

Hello

When using functors (when calling functor(space arg, state/ time arg), you can specify a state argument which can be:

• the current value
• previous iteration value
• previous or twice previous time step value

So the answer is you can do it, but you need to create an auxkernel or functor material to do this.
How are you planning to code this term?

N_Ra(n+1) = N_Ra(n) + [ (M / 1000) * U * 12400 * 0.25 - 2.1e-6 * N_Ra(n) ] 

in a Parsed-something? or in an object?

[GiudGiud]

ok use the functor system instead

see FunctorAux for an example of retrieving and storing a functor, and using it with the () operator

in the state argument, the second argument in the () operator, you can pass the iteration type and a 1 to say you want the old value
see MooseFunctorArguments.h for more info

[Traiwit]

  params.addClassDescription(
      "Evaluates a functor (variable, function or functor material property) on the current "
      "element, quadrature point, or node.");
  params.addRequiredParam<MooseFunctorName>("functor", "The functor to evaluate");

FunctorAux::FunctorAux(const InputParameters & parameters)
  : AuxKernel(parameters),
    _functor(getFunctor<Real>("functor")),
    _factor(getFunctor<Real>("factor")),

so _functor(getFunctor<Real>("functor")) can take variable?

and Real N_radon = _functor(elem_arg, 1) something like this?
im so confused

[GiudGiud]

so _functor(getFunctor("functor")) can take variable?

yes you can set functor = <vasriable name >

and Real N_radon = _functor(elem_arg, 1) something like this?

something. like this

#include MooseFunctorArguments.h
// where oldState is defined. If it does not work, create the Moose::StateArg manually

_functor(elem_arg, Moose::oldState());

[Traiwit]

Hi @GiudGiud,

I kinda gave up on this, but found a way to work around it:

[Physics/NavierStokes]
  [ScalarTransport]
    [Radon]
      # coupled_turbulence_physics = 'turbulence'
      coupled_flow_physics = 'flow'
      passive_scalar_names = 'scalar'

      passive_scalar_diffusivity = ${diffusion_coff}
      passive_scalar_source = 'radon_source'
      passive_scalar_coupled_source = 'scalar'
      passive_scalar_coupled_source_coeff = -2.1e-6

      initial_scalar_variables = '0'

      passive_scalar_inlet_types = 'fixed-value'
      passive_scalar_inlet_functors = '0'

    # for restart
    #   initialize_variables_from_mesh_file = true
      

      block =  '0 1 2 3 6'

    []
  []
[]

so radon_source is the constant term, then i used

      passive_scalar_coupled_source = 'scalar'
      passive_scalar_coupled_source_coeff = -2.1e-6

to deal with decay term

[GiudGiud]

ok this works! It's what we do for MSR delayed neutron precursor advection. It's a different numerical scheme though, it's not semi-explicit like you were showing, it's fully implicit

[Traiwit]

Hi @GiudGiud

I dont want to crate a new thread, but I have a quick question

if I do [ScalarTransport] with

[NavierStokes/flow] porous_medium_treatment = true

the scalar generated from the source term should be multiplied with porosity, right?

but it doesn't that right now, it generate at source term regardless of porosity

[GiudGiud]

the scalar generated from the source term should be multiplied with porosity, right?

when solving for the superficial concentration (e.g. epsilon * interstitial concentration), it should be. The passive_scalar_source_coefs let you do that though.

when it move from porous media into void element, it just follows the advection/diffusion

That is the intent yes when solving for the superficial concentration. We should not see a jump in value

[Traiwit]

Hi @GiudGiud

so i didnt manually do any multiplication of porosity at all when assign source term

the plot is concentration inside the stope (cave + air, source term is only active at cave material)

Orange - porosity of cave = 100%
blue/green - porosity of the cave = 40%

what does it mean? the code handle the porosity internally?

it looks like accumulation is also scaled by porosity

[GiudGiud]

it looks like accumulation is also scaled by porosity

it does not. Use Problem/type=DumpObjectsProblem to see the objects created by the action and how their parameters are set

To me, same source gives the same superficial concentration makes sense. We are solving dCi/dt = Q + advection

[Traiwit]

Hi @GiudGiud

I see, yeh the main problem is from the superficial velocity i think, the element/system with porosity has much higher velocity, so it push concentration out from the system faster > hence less concentration inside the stope.

thanks again, appreciated your patient :)

i have a few more q's re: convergence when there is a major different in drag coeff (tunnel and cave) AND spiky/roughness of the mesh, should i ask here or make a new thread

[GiudGiud]

let's open a new thread.

i have a few more q's re: convergence when there is a major different in drag coeff (tunnel and cave) AND spiky/roughness of the mesh, should i ask here or make a new thread

There are a couple treatments available:
for porosity jumps: porosity smoothing layers, bernouilli pressure drop imposition on the pressure variable
for friction jumps: body force reconstruction, and friction correction with consistent scaling