Schlumberger Technology Corporation

Hui Xie

Ping Zhang

Tarek M Habashy

Dzevat Omeragic

John C Rasmus

Aria Abubakar

Roger Griffiths

Koji Ito

Christopher Edward Morriss

Shahzad Asif

Sushil Shetty

Helen Xiaoyan Zhong

Ettore Mirto

QingFeng Zhu

David Maggs

Yao Feng

Yan Kai Xu

A method for drilling includes obtaining formation-measurement pairings, training a machine-learning model using the formation-measurement pairings, receiving measurements obtained by a tool positioned in a well formed in a formation, and generating a formation model of at least a portion of the formation using the machine-learning model and the measurements. The formation model represents one or more physical parameters of the formation, one or more structural parameters of the formation, or both.

Geological interpretation with artificial intelligence

Systems and methods are disclosed for well logging using radiation detection and/or emission of gamma rays. A method according to the disclosure includes collecting data from the subterranean formation using a nuclear density tool, wherein the nuclear density tool is configured to collect data to form an azimuthal image. The method further includes characterizing a section of the subterranean formation comprising data and images acquired in a high angle wellbore section, a horizontal wellbore section, or a combination thereof. The method additionally includes performing a parallel inversion using apparent densities and volumetric photoelectric factor images to build a formation model, wherein the parallel inversion comprises a high angle workflow that models high angle wellbore sections and a horizontal workflow that models horizontal wellbore sections.

Inversion-based workflow for consistent interpretation of nuclear density images in horizontal wells

A method for transforming a 2D or 3D earth volume geometry into a 1D earth volume geometry includes performing a measurement using the measurement sensor in a wellbore. A layer boundary in the 2D or 3D earth volume geometry that is nearest to the measurement sensor is identified. A vector from the measurement sensor is generated toward the nearest layer boundary. A first intersection is identified between the vector and the nearest layer boundary, and a second intersection is identified between the vector and another layer boundary. Simulated boundaries that extend through the first and second intersections and are perpendicular to the vector are generated. The 1D earth volume geometry that is bounded by the first and second intersections is identified. A property value is extracted from the 2D or 3D earth volume geometry between the first and second intersections. The property value is assigned to the 1D earth geometry.

Method for determining petrophysical properties from logging measurements

Growing community of inventors

Katy, TX, United States of America

Hui Xie