King Abdullah University of Science and Technology

Saudi Arabian Oil Company

Xupeng He

Hyung Tae Kwak

Marwah Mufid AlSinan

Hussein Ali Hoteit

Yiteng Li

Hussein Hoteit

Systems and methods for determining a 3D hydraulic aperture of a 3D fracture are disclosed. The method includes, obtaining a geometry of the 3D fracture, determining a fluid flow direction through the 3D fracture, and dividing the 3D fracture into a plurality of 2D cross-sections oriented substantially parallel to the fluid flow direction. The method further includes dividing each 2D cross-section into a plurality of Type I and Type II fracture segments based on a segment aspect ratio and a segment roughness ratio, determining a 2D segment hydraulic aperture for each of the plurality of Type I and Type II fracture segments, and determining the 3D hydraulic aperture of the 3D fracture based, at least in part, on the 2D segment hydraulic apertures of the plurality of Type I and Type II fracture segments.

Method to calculate the effective hydraulic aperture and permeability of rock fractures

A method may include obtaining fracture image data regarding a geological region of interest. The method may further include determining various fractures in the fracture image data using a first artificial neural network and a pixel-searching process. The method may further include determining a fracture model using the fractures, a second artificial neural network, and borehole image data. The method may further include determining various fracture permeability values using the fracture model and a third artificial neural network. The method may further include determining various matrix permeability values for the geological region of interest using core sample data. The method may further include generating a coarsened grid model for the geological region of interest using a fourth artificial neural network, the matrix permeability values, and the fracture permeability values.

Method and system for determining coarsened grid models using machine-learning models and fracture models

Techniques for determining reservoir formation properties include generating digital models of core samples taken from one or more reservoir formations based on corresponding images of the core samples; determining a pore throat size distribution of the core samples based on the generated digital models; determining corresponding capillary pressure curves and corresponding NMR value distributions of the core samples with one or more numerical simulations; generating one or more machine-learning (ML) models based on the pore throat size distribution, corresponding capillary pressure curves, and corresponding NMR value distributions of the core samples; adjusting the one or more ML models with one or more reservoir data of the one or more reservoir formations; generating adjusted capillary pressure curves and NMR value distributions from the one or more adjusted ML models; and determining a reservoir formation specific pore throat size distribution from at least one of the adjusted capillary pressure curves or the adjusted NMR value distributions.

Growing community of inventors

Thuwal, Saudi Arabia

Xupeng He