Skip to main content
Version: Latest

Fluid injection into a poroelastic borehole

Fluid at pressure enters a borehole wall: it loads the wall and diffuses into the rock, which changes the stresses around the hole as it goes. The axisymmetric part of Detournay and Cheng’s borehole problem, on a finite annulus.

On the 0.05 a mesh: pressure within 0.06% of at every node for 1.05 a ≤ r ≤ 4 a and all four times; total stress rms 0.16% (hoop) with a largest element error of 1.06%.

At a glance
Dimension2D
Buildmake ndims=2 from the poromechanics source (not yet merged; DynEarthSol #134)
Inputsborehole.zip
Reproduce withDynEarthSol 3774df0

Problem setup​

Setup

Geometry. A quarter annulus m, , plane strain, with symmetry rollers on both cuts; meshes refined out to .

Loading. At the wall takes fluid at MPa: pore pressure (a 1 ms ramp) and normal traction . The outer arc is sealed and traction-free. The rock starts unstressed at ; gravity is off.

Material. MPa, MPa, , , ; diffusivity .

What is compared. Nodal pressure and element total radial and hoop stress for at 9, 18, 27 and 36 s, relative to .

Input and how to run​

Every file this example needs is in borehole.zip. Run from that folder:

# Meshes the quarter annulus from annulus.poly (shipped alongside) at 0.05 m inner spacing and runs 144 steps of 0.25 s (36 s), saving hole.save.* every 36 steps (9, 18, 27 and 36 s) with hole.info listing the frames.
dynearthsol2d input.cfg

Settings that define this example​

SettingValueWhat it does here
mesh.meshing_option9090: mesh the polygon read from poly_filename
mesh.poly_filenameannulus.polyannulus.poly: quarter annulus a = 1 m to R = 20 m with a refined ring to 5a; flags: vertical cut x0, borehole wall x1, horizontal cut z0, outer arc z1
mesh.resolution0.050.05 m element size near the wall, the finest mesh of the study
bc.hbc_x12type 2: pore pressure prescribed on the borehole wall from hbc_pressure
bc.hbc_pressure[0,1000000.0,0,0,0,0]the x1 (wall) entry is the fluid pressure p_w = 1 MPa
bc.stress_bc_x14type 4: normal stress on the curved wall
bc.stress_val_x1-1000000.0−p_w: the wall fluid pushes on the rock (tension positive)
bc.num_hbc_schedule2two schedule points for the wall pressure
bc.hbc_schedule_time_in_yr[0, 3.168876453507857e-11] yr = [0, 0.001] s0 and 1 ms: the wall pressure ramps on almost instantly
bc.hbc_schedule_ratio[0, 1]0 to 1, then held at p_w
mat.hydraulic_perm[1e-12]k = 1e-12 m²; with water viscosity 1e-3 Pa s the mobility k/μ is 1e-9 m²/(Pa s)
mat.fluid_bulk_modulus[200000000.0]K_f = 200 MPa; with porosity 0.2 the storage 1/M = φ/K_f is 1e-9 /Pa
mat.bulk_modulus[200000000.0]K = 200 MPa (shear modulus 100 MPa); together they give diffusivity c = 0.25 m²/s
sim.output_step_interval3636 steps: frames at 9, 18, 27 and 36 s, the four compared times

input.cfg holds all 57 settings that differ from DES’s defaults; dynearthsol2d -h lists every option with its default.

Output​

See Running the examples for output files, the Python reader and ParaView conversion.

FieldWhat to look at
pore pressurenodal pressure, binned by radius and compared with the reference in panel (a)
stresselement total stress, rotated to radial and hoop components in panel (b)

Result: DES and the analytical solution​

Result

(a) Pore pressure, scaled by the wall pressure p_w, against distance from the borehole centre at the four saved times: lines, the finite-annulus poroelastic solution; symbols, DES. (b) Total radial and hoop stress, scaled by p_w, at the last saved time. (c) Largest pressure and stress errors against mesh spacing.

This checks the coupled poroelastic response around an opening: diffusion from a curved boundary, the pressure-driven hoop stress and the wall load together.

To reproduce the whole study behind this figure, from the DES source folder:

python3 benchmarks/poromechanics/borehole_reference.py
python3 tests/functional/hydromechanics/check_borehole.py dynearthsol2d OUT/mesh --spacings 0.2 0.1 0.05
python3 tests/functional/hydromechanics/check_borehole.py dynearthsol2d OUT/steps --spacings 0.1 --steps 576

The study’s full numbers: record.json.

Analytical reference​

Tension positive. The radial displacement is , so the volumetric strain is ; and follow from the two traction conditions and are linear in and .

Derivation​

Radial symmetry makes the displacement irrotational, so equilibrium integrates once: the volumetric strain is plus a uniform term . Fluid mass, , then gives the diffusion equation above.

On a finite annulus and depends on , which couples the pressure to its own integral; it moves the pressure by about 1% of here. In Laplace space the pressure is a Bessel (, , , ) solution plus a uniform part , with from one scalar equation; Stehfest inversion agrees with an independent Crank–Nicolson solution to of .

Detournay and Cheng (1988), International Journal of Rock Mechanics and Mining Sciences 25(3), 171–182, axisymmetric loading modes; the finite annulus is solved in borehole_reference.py.

Original reference.

Validity and limits​

Axisymmetric loading only (no deviatoric in-situ stress), linear poroelasticity, small strain. The finite annulus is part of the reference; the infinite-medium Detournay–Cheng solution is its limit.