micromodel_2d Namespace Reference

Classes
class	Health

Functions
	posts (seed=3)
	post_sdf (P)
	box_dimension (mask, sizes=(1, 2, 4, 8, 16, 32))
	front_stats (inv)
	run (theta_deg, ca, sdf, steps, budget=None, npy_dir="")
	main ()

Variables
	NX
	NY
	NZ
	NCOL
	NROW
	X0
	DX
	DY = NY / NROW
float	R_POST = 6.5
float	JITTER = 1.5
float	MIN_GAP = 6.0
float	X_IN = 10.0
float	X_BT = 118.0
float	SIGMA = 100.0
	RHO_L
	RHO_G
	MU_L
	MU_G
float	CAP_CFL = 0.5
int	PRESS_CAP = 400
int	PRESS_RTOL = 1e-8
int	MG_LEVELS = 5
int	VEL_SWEEPS = 20

Detailed Description

WO-V7 case 3 — a ZHAO-2019-STYLE 2-D MICROMODEL: wettability and the invasion pattern.

A disordered array of ~56 SDF cylinders (a jittered lattice of posts, the geometry class Zhao,
MacMinn & Juanes, *PNAS* 116:13799 (2019) used) fills a quasi-2D cell that is gas-filled at t = 0.
Liquid is injected through the -x face at a fixed capillary number and leaves at +x.  The sweep is
over the contact angle only: theta = 45 (imbibition), 90 (neutral), 135 (drainage).

Zhao et al.'s qualitative result, which this page reproduces or fails to:
**wettability controls the pattern**.  Strong imbibition gives a COMPACT, high-saturation
displacement (cooperative pore filling, corner films run ahead); drainage gives a FINGERED,
low-saturation one (capillary fingering / invasion percolation).  The quantitative read-outs are
the saturation at breakthrough and the box-counting fractal dimension of the invaded region.

CONVENTIONS AND CAVEATS (WO-V7 preamble; stated on every run):
  * STATIC contact angle (V5b).  Ca is APPARENT: mu_l U_inlet / sigma.
  * density ratio 100; inflow/outflow (never a periodic net force); the posts are clear of both
    open faces; a capped pressure solve makes the run INVALID.
  * **corner films are NOT represented.**  VOF_PLAN 4/V7 names this as the lesson to design
    against: the strong-imbibition end of Zhao's benchmark is decided by films in the corners of
    the posts, which a 5.7-cell-radius post on a unit-spacing grid cannot resolve.  Expect the
    compact/fingered TREND, not the saturations.

Usage:
    PYTHONPATH=$PWD/build_cuda python tests/study/pore_scale/micromodel_2d.py [--quick]
        [--theta 45,90,135] [--ca 1e-3] [--budget 1800] [--npy DIR]

Function Documentation

posts()

micromodel_2d.posts

(

seed = 3

)

A staggered lattice of post centres with bounded jitter, rejected until every pair is at
least MIN_GAP apart at the surfaces (y periodic).

Definition at line 74 of file micromodel_2d.py.

References mean().

Referenced by main().

post_sdf()

micromodel_2d.post_sdf

(

P

)

Definition at line 95 of file micromodel_2d.py.

Referenced by main().

box_dimension()

micromodel_2d.box_dimension	(		mask,
			sizes = (1, 2, 4, 8, 16, 32)
	)

Box-counting dimension of a 2-D boolean set; returns (D, [(eps, N)]).

Definition at line 134 of file micromodel_2d.py.

Referenced by run().

front_stats()

micromodel_2d.front_stats

(

inv

)

Roughness of the invasion front, the discriminator Zhao's experiment is really about.

`inv` is the boolean invaded set of the array region in the mid-plane, indexed (x, y).  For
every transverse row y that the liquid reached, the front is the furthest x it reached; the
MEAN of that is how far the displacement has got and its STANDARD DEVIATION is how ragged it
is.  A compact (cooperative-filling) displacement has a small std and reaches every row; a
fingered one has a large std, a large max/mean ratio, and leaves rows untouched.

Definition at line 151 of file micromodel_2d.py.

References mean().

Referenced by run().

run()

micromodel_2d.run	(		theta_deg,
			ca,
			sdf,
			steps,
			budget = None,
			npy_dir = ""
	)

Definition at line 171 of file micromodel_2d.py.

References box_dimension(), front_stats(), mean(), and slice().

Referenced by main().

main()

micromodel_2d.main

(

)

Definition at line 306 of file micromodel_2d.py.

References main(), mean(), post_sdf(), posts(), and run().

Referenced by main().

Variable Documentation

NX

micromodel_2d.NX

Definition at line 40 of file micromodel_2d.py.

NY

micromodel_2d.NY

Definition at line 40 of file micromodel_2d.py.

NZ

micromodel_2d.NZ

Definition at line 40 of file micromodel_2d.py.

NCOL

micromodel_2d.NCOL

Definition at line 54 of file micromodel_2d.py.

NROW

micromodel_2d.NROW

Definition at line 54 of file micromodel_2d.py.

X0

micromodel_2d.X0

Definition at line 55 of file micromodel_2d.py.

DX

micromodel_2d.DX

Definition at line 55 of file micromodel_2d.py.

DY

micromodel_2d.DY = NY / NROW

Definition at line 56 of file micromodel_2d.py.

R_POST

float micromodel_2d.R_POST = 6.5

Definition at line 57 of file micromodel_2d.py.

JITTER

float micromodel_2d.JITTER = 1.5

Definition at line 58 of file micromodel_2d.py.

MIN_GAP

float micromodel_2d.MIN_GAP = 6.0

Definition at line 59 of file micromodel_2d.py.

X_IN

float micromodel_2d.X_IN = 10.0

Definition at line 61 of file micromodel_2d.py.

X_BT

float micromodel_2d.X_BT = 118.0

Definition at line 62 of file micromodel_2d.py.

SIGMA

float micromodel_2d.SIGMA = 100.0

Definition at line 64 of file micromodel_2d.py.

RHO_L

micromodel_2d.RHO_L

Definition at line 65 of file micromodel_2d.py.

RHO_G

micromodel_2d.RHO_G

Definition at line 65 of file micromodel_2d.py.

MU_L

micromodel_2d.MU_L

Definition at line 66 of file micromodel_2d.py.

MU_G

micromodel_2d.MU_G

Definition at line 66 of file micromodel_2d.py.

CAP_CFL

float micromodel_2d.CAP_CFL = 0.5

Definition at line 67 of file micromodel_2d.py.

PRESS_CAP

int micromodel_2d.PRESS_CAP = 400

Definition at line 68 of file micromodel_2d.py.

PRESS_RTOL

int micromodel_2d.PRESS_RTOL = 1e-8

Definition at line 69 of file micromodel_2d.py.

MG_LEVELS

int micromodel_2d.MG_LEVELS = 5

Definition at line 70 of file micromodel_2d.py.

VEL_SWEEPS

int micromodel_2d.VEL_SWEEPS = 20

Definition at line 71 of file micromodel_2d.py.