machine learning

This paper presents a physics-informed machine learning method that enhances the accuracy of pressure transient analysis, predicting reservoir properties to enhance waste slurry injection and waste disposal.

This paper presents a novel workflow with multiobjective optimization techniques to assess the integration of pressure-management methodologies for permanent geological carbon dioxide storage in saline aquifers.

Oil and gas experts encourage human/AI partnerships that can “supercharge” capabilities to create competitive advantages.

Deploying artificial intelligence across an enterprise requires thinking beyond the pilot.

This paper presents the development of a robust, physics-based, and data-driven workflow for modeling mud loss in fractured formations and predicting terminal mud loss volume and time, as well as equivalent hydraulic fracture aperture.

Traditionally, the drilling industry has relied on high-fidelity thermal simulators to predict downhole temperature for different operational scenarios. Though accurate, these models are too slow for real-time applications. To overcome this limitation, a deep-learning solution is proposed that enables fast, accurate prediction of downhole temperatures under a wide ran…

This paper presents a smart safety monitoring system to prevent accidents in environments with moving machinery at use on various global rigs.

An innovative approach uses a random-forest-based framework to link logging-while-drilling and multifrequencey seismic data to enable dynamic updates to lithology parameter predictions, enhancing efficiency and robustness of geosteering applications.

Foundation models are rapidly emerging as a transformative force across industries. While their effect on natural language processing and computer vision is well-documented, their potential in specialized engineering domains, particularly within the critical oil, gas, and broader energy sectors, is vast and increasingly recognized. This article explores how these powe…

This article presents a comparative study evaluating four machine-learning approaches, including three deep-learning methods, for forecasting gas and condensate production over a 5-year horizon.