Eigen Initial Conditions with NtAction

[Jeremyb8707bigbrain]

Trying to do a Transient with an Eigenvalue initial condition from a previous file.
Keeps erroring and says The initial condition 'init_g1' is being defined on a block that already has an initial condition defined with the same variable and state.

Is there a way to bypass the auto initialization of initial conditions or am I missing something entirely? If not do I have to manually setup my variables instead of using NtAction? Any help is appreciated!

flow_velocity = 21.3

[GlobalParams]
    num_groups = 4
    num_precursor_groups = 6
    account_delayed = true
    sss2_input = true
    temperature = 973.0
    use_exp_form = false
    pre_concs = 'pre1 pre2 pre3 pre4 pre5 pre6'
    group_fluxes = 'g_1 g_2 g_3 g_4'
[]

[Mesh]
    file = '/workspace/moltensaltsim/Meshes/30_reflector/meshgenerator_cluster_in.e'
[]

[Materials]
    [fuel_salt]
        type = MoltresJsonMaterial
        base_file = '/workspace/moltensaltsim/Material_Data/Moltres_JSONs/cluster_14cm.json'
        material_key = 'Molten_Salt'
        block = 'fuel_salt'
        interp_type = 'linear'
    []
    [graphite]
        type = MoltresJsonMaterial
        base_file = '/workspace/moltensaltsim/Material_Data/Moltres_JSONs/cluster_14cm.json'
        material_key = 'Graphite'
        block = 'graphite'
        interp_type = 'linear'
    []
[]

[Precursors]
  [delayed]
    var_name_base = 'pre'
    outlet_boundaries = '104'
    inlet_boundaries = '103'
    w_def = ${flow_velocity}
    v_def = 0
    u_def = 0
    block = 'fuel_salt'
    nt_exp_form = false
    family = MONOMIAL
    order = CONSTANT
    loop_precursors = false
  []
[]

[Nt]
    var_name_base = g_
    pre_blocks = 'fuel_salt'
    create_temperature_var = false
    eigen = false
    fission_blocks = 'fuel_salt'
    vacuum_boundaries = '10000'
[]

[UserObjects]
    [eigen_solution]
        type = SolutionUserObject
        mesh = '/workspace/moltensaltsim/Output/Eigen_Solutions/30_reflector/steady_14cm_cluster_out.e'
        system_variables = 'g_1 g_2 g_3 g_4'
        timestep = LATEST
    []
[]

[ICs]
    [init_g1]
        type = SolutionIC
        from_variable = 'g_1'
        variable = g_1
        solution_uo = eigen_solution
    []
    [init_g2]
        type = SolutionIC
        from_variable = 'g_2'
        variable = g_2
        solution_uo = eigen_solution
    []
    [init_g3]
        type = SolutionIC
        from_variable = 'g_3'
        variable = g_3
        solution_uo = eigen_solution
    []
    [init_g4]
        type = SolutionIC
        from_variable = 'g_4'
        variable = g_4
        solution_uo = eigen_solution
    []
[]

[Executioner]
  type = Transient
  solve_type = 'PJFNK'
  petsc_options = '-snes_converged_reason -ksp_converged_reason -snes_linesearch_monitor'
  petsc_options_iname = '-pc_type -pc_hypre_type'
  petsc_options_value = 'hypre boomeramg'

  automatic_scaling = true
  compute_scaling_once = false
  resid_vs_jac_scaling_param = 0.1
  line_search = none

  dt = 1e-3
  end_time = 5e-3
[]

[Outputs]
    exodus = true
[]

[smpark7]

You can refer to moltres/problems/2021-cnrs-benchmark/phase-2 for how I restarted the group fluxes from an exodus file. Essentially, you need to load your previous exodus solution using the FileMeshGenerator and indicate use_for_exodus_restart=true. Additionally you need to set init_nts_from_file=true in the Nt block to override the default flux=1 initial conditions.

Also, once you get your input file to run, it will converge to 0 or infinity depending on your k value because you have no thermal feedback to stabilize it.

[Jeremyb8707bigbrain]

This worked great thanks!

[Jeremyb8707bigbrain]

@smpark7 I was going back and I realized I could actually set precursors for my eigen, so I decided to mess around with that and im getting some confusing results. My outlet is the top of the mesh, so the salt would be flowing up in the W direction, but the results I get seem to indicate that the salt is flowing down. Below is pre1 and pre6, the precursor concentration moves down towards the center, which would make sense if it was entering from the top. I have quadruple checked my surfaces and done runs with the numbers flipped, both outlet and inlet boundaries defined, and with slower salt speed. None seem to fix this issue, any help would be great!

[Precursors]
  [pres]
    var_name_base = pre
    family = MONOMIAL
    order = CONSTANT
    block = 'fuel_salt'
    outlet_boundaries = '103'
    velocity_type = constant
    u_def = 0
    v_def = 0
    w_def = 21.7
    nt_exp_form = false
    loop_precursors = false
    transient = false
    eigen = true
  []
[]

[smpark7]

I dont see anything wrong. pre1 is longer lived so the peak is shifted upwards more than pre6.

[Jeremyb8707bigbrain]

Does pre1 not correspond to the first entry for delayed neutrons in the JSON file, I am basing my guesses off of the decay constant where the 6th entry has the longest?

I am still learning more about delayed neutrons with reactor work so forgive me for my confusion

Edit: I completely forgot about the unit being 1/s, forgive me thank you!