simulation

This study applies Monte Carlo simulation and an XGBoost regression model to assess the influence of various formations, geologic provinces, tectonic-plate types, and boundary conditions on hydrogen concentrations.

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 paper reviews fracturing-program design, completion technology, real-time data collection, data integration, and lessons learned for the Pikka development on the North Slope of Alaska.

This paper describes the integration of iterative torque/drag/buckling and hydraulic simulations for multiple tapered string combinations, the results of which guided the selection of a string configuration that deemed planned well total depths feasible.

This paper presents the development of an advanced simulation tool aimed at providing a better understanding of the complex fluid-displacement phenomena present in well-cementing processes.

Two examples from ONGC show how supervised AI-driven automation scaled well modeling across hundreds of offshore wells, saving more than 1,000 engineering hours.

The newest recipient of the title SPE Legend of Hydraulic Fracturing talks about his career, the evolution of fracture stimulation, the development of increasingly useful simulators, and the future of the oil and gas industry. The honor was given at the 2026 SPE Hydraulic Fracturing Technology Conference and Exhibition.

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.

In this study, forward simulation is executed by a commercial reservoir simulator while external code is developed for backward calculations.

In this study, the authors propose the use of a deep-learning reduced-order surrogate model that can lower computational costs significantly while still maintaining high accuracy for data assimilation or history-matching problems.