neural networks

This work describes a study in which distributed data parallel training, paired with a node-local caching pipeline, enabled efficient multigraphics-processing-unit scaling for a CO₂-storage graph-neural-network surrogate while maintaining generalization.

This work uses a novel pseudosteady-state-based simulation to reduce training-data-generation cost while maintaining high-performance predictions of data-driven proxy models for carbon-sequestration projects.

This study explores the use of autoencoder models with convolutional neural networks to present a framework and prototype for early and accurate kick detection during offshore oilwell drilling.

The authors propose a deep-learning-based approach enabling near-real-time CO2-plume visualization and rapid data assimilation incorporating multiple geological realizations for predicting future CO2 plume evolution and area-of-review determination.

This paper presents a self-supervised approach for training a seismic foundation model and demonstrates scenarios in which it is used for seismic data conditioning, interpretation, and inversion through six real-use cases.

This work presents the development of fast predictive models and optimization methodologies to evaluate the potential of carbon-dioxide EOR and storage operations quickly in mature oil fields.

In this third work in a series, the authors conduct transfer-learning validation with a robust real-field data set for hydraulic fracturing design.

This paper explores the use of machine learning in predicting pump statuses, offering probabilistic assessments for each dynacard, automating real-time analysis, and facilitating early detection of pump damage.

This paper focuses on developing a model that can be used in an automated, end-to-end flare-smoke detection, alert, and distribution-control solution that leverages existing flare closed-circuit television cameras at manufacturing facilities.

This paper presents a case study highlighting the demonstration, refinement, and implementation of a machine-learning algorithm to optimize multiple electrical-submersible-pump wells in the Permian Basin.