Chevron U.s.a. Inc.

Bo Gong

Paul Hart

Paul Duke

Emmanuel Toumelin

Hanming Wang

Dustin J Keele

Simon Clinch

Embodiments of categorizing fractures in a subsurface formation are provided herein. One embodiment comprises obtaining at least one borehole image for a plurality of fractures in a subsurface formation and at least one resistivity log for the plurality of fractures of the surface formation. The embodiment comprises determining first aperture data based on the at least one borehole image for the plurality of fractures and determining second aperture data based on the at the least one resistivity log for the plurality of fractures. The embodiment comprises comparing the first aperture data based on the at least one borehole image and the second aperture data based on at least one resistivity log to determine whether to categorize each fracture of the plurality of fractures as a naturally occurring fracture or a non-naturally occurring fracture. The embodiment comprises generating an output indicating the categorization of each fracture and displaying the output.

Categorizing fractures in a subsurface formation

A method is described for training a model that refines estimated parameter values within core images is disclosed. The method includes receiving multiple training image pairs wherein each training image pair includes: (i) an unrefined core image of a rock sample to be used for estimating rock properties, and (ii) a refined core image of the same rock sample; generating a training dataset from the multiple training image pairs; receiving an initial core model; generating a conditioned core model by training, using the multiple training image pairs, the initial core model; and storing the conditioned core model in electronic storage. The conditioned core model may be applied to an initial target core image data set to generate a refined target sore image dataset. The method may be executed by a computer system.

System and method for estimation of rock properties from core images

A method is described for ways to generate a Fraction of Sand (F) estimate and net-to-gross (NTG) estimate of sand in a formation using a machine-learning algorithm such as a neural network based on borehole image logs. The method may use the Fand other information to estimate hydrocarbons in place in a subsurface formation. The method may be executed by a computer system.

Growing community of inventors

Houston, TX, United States of America

Bo Gong