Data & Analytics

Transfer learning workflow showing the use of the physics-based trained backbone as the starting point for feature transfer (adjust only end layers and keep initial parameters frozen) and fine tuning (retrain entire network) with new data.

Transfer-Learning Validation Improves Hydraulic Fracture Design

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

Drilling bit and positive displacement motor in front of drilling rig in black & white

Drilling/completion fluids

Fluid-Advisory System Offers Recommendations for Various Systems

This research aims to develop a fluid-advisory system that provides recommendations for optimal amounts of chemical additives needed to maintain desired fluid properties in various drilling-fluid systems.

Flow assurance

AI Powers Hydrate Detection, Prevention in North Kuwait Jurassic Gas Field

This paper discusses a comprehensive hybrid approach that combines machine learning with a physics-based risk-prediction model to detect and prevent the formation of hydrates in flowlines and separators.