MultiApp + PorousFlowPeacemanBorehole: 1D fracture–2D matrix flow nonconvergence — how to define borehole length? Any example for transferring fracture porepressure as normal stress to the matrix to incorporate the mechanical effects?

[Yulongggggg]

I’m trying to simulate flow through a 1D fracture embedded in a 2D domain, and I’m confused about how to choose the “borehole length” for PorousFlowPeacemanBorehole in a 2D setup.

In my fracture app, the injection borehole is defined by injection.bh:
0.2 0 -250 10
0.2 0 -250 0

I intended this to represent a 10 m borehole segment, but the simulation fails to converge with this file. If I change the segment length to 1 m (i.e., use z=1 instead of z=10), the model converges.

Could you clarify how the borehole length/segment length should be interpreted when using PorousFlowPeacemanBorehole in a 2D (embedded-fracture) simulation? Below is my fracture app input file, and i am also wondering how to transferring fracture porepressure as normal stress to the matrix to incorporate the mechanical effects:

[GiudGiud]

hello

did you read this doco on peaceman boreholes?
https://mooseframework.inl.gov/modules/porous_flow/sinks.html#6a71b05a-b0b0-45aa-bd59-778f3880980c

before we tag the authors

[Yulongggggg]

Are these points inside an element in your meshed domain: 0 -250 10 0 -250 0

Do you have 2D elements in your mesh or just the 1D fracture elements.

Hi, I have 2D elements as a matrix and 1D elements as fractures, in the fracture app.i I have an injection point located at (0, -250, 0)

[lynnmunday]

You also have an injection point at (0 -250 10), is that point in the domain?

[Yulongggggg]

You also have an injection point at (0 -250 10), is that point in the domain?

No, it's not. Is it ok for me to use one point in the BH file for this 2D case? Do you know how to represent the wellbore length in this case?

[lynnmunday]

I use this:

[Postprocessors]
  [mass_flux_src_stage_9_7]
    type = FunctionValuePostprocessor
    function = mass_flux_src
    scale_factor = ${inj_ratio_stage_10}
    execute_on = 'initial timestep_end'
  []
[]
  [source_9_7]
    type = PorousFlowPointSourceFromPostprocessor
    variable = porepressure
    mass_flux = mass_flux_src_stage_9_7
    point = '-31.5796530000 235.5353800000 406.9108400000'
    point_not_found_behavior = WARNING
    # block = 100
  []