porousmedia Goto Github PK

machine-learning-competition-2020

SPWLA PDDA’s 1st Petrophysical Data-Driven Analytics Contest -- Sonic Log Synthesis

machine-learning-competition-2021

micp_to_eptr

The notebook shows the principle behind the minimum incremental pore volume (MIPV) and estimating the effective pore-throat radius (EPTR) as outlined in steps in the paper above.

patchify_buki_mod

Slice and re-combine image data into 2D or 3D subvolumes of specific shape.

permeability_calculator

perm_ calculator is a tool for rapid calculation of permeability from raw and processed MICP data. This will help save time spent on manually obtaining permeability estimates using Dastidar, Winland, Swanson, Wells, and Kamath models See our paper for more details on the respective models: Ishola, O. and Vilcáez, J., Machine learning modeling of permeability in 3D heterogeneous porous media using a novel stochastic pore-scale simulation approach.

pore_microstructural_and_flow_modelling

porousmedia

Config files for my GitHub profile.

stack_3d_unet

unet_BukiMod is a modification to the U-net architecture for semantic segmentation of images.

star_ccm_flow_simulator

This STAR CCM+ java code takes in two csv files (pore bodies and pore throats), use them to generate pore microstructures, and simulate fluid flow through it. The output of this script is a csv file of flow properties through a set number of iterations.

stat453-deep-learning-ss21

STAT 453: Intro to Deep Learning @ UW-Madison (Spring 2021)

stochastic_pore_microstructural_generator

This code creates CSV files for pore bodies and pore throats of stochastically generated 3D pore microsturtures. The files are intended to make 3D images/ surface files and run simulations in STAR-CCM+.

xlearn

Deep learning toolbox for x-ray imaging

xray_to_pore_size_distribution

The notebook demonstrates the workflow for obtaining pore size distribution from binarized micro-CT images. The general principle involves identifying each pore, estimating the volume of each pore, and ultimately determining the radius of a sphere with an equivalent volume of each pore.