neural networks

This study integrates physics-based constraints into machine-learning models, thereby improving their predictive accuracy and robustness.

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…

In this study, a deep-neural-network-based workflow with enhanced efficiency and scalability is developed for solving complex history-matching problems.

Whether it’s reviving inactive gas-condensate wells or identifying overlooked reserves in brownfields, operators are making the most of older wells and fields.

In this paper, the authors propose a regression machine-learning model to predict stick/slip severity index using sequences of surface measurements.

The objective of this study is to develop an explainable data-driven method using five different methods to create a model using a multidimensional data set with more than 700 rows of data for predicting minimum miscibility pressure.

The authors of this paper propose hybrid models, combining machine learning and a physics-based approach, for rapid production forecasting and reservoir-connectivity characterization using routine injection or production and pressure data.

In this study, artificial-intelligence techniques are used to estimate and predict well status in offshore areas using a combination of surface and subsurface parameters.

This paper investigates the use of machine-learning techniques to forecast drilling-fluid gel strength.

Energy efficiency is crucial for the oil and gas industry, where operational costs and environmental impact are under constant scrutiny. Predicting and managing electrical consumption and peak demand accurately, especially with the variability of weather conditions, is a significant challenge. This work presents a neural network model trained on historical weather and…