What question did this study set out to answer?

This research aims to estimate the delineation of radioactive anomalies using machine learning models based on well logs.

Machine Learning Algorithms to Predict Source and Reservoir Zones of Abu Roash Formation Using Spectral Gamma‐Ray Logs, Khalda Oil Field, Northern Western Desert, Egypt

View Full Paper

Ask AI

Bookmark

Authors

Randomized trial predicts radioactive anomalies in wells, indicating a cost-effective method for hydrocarbon exploration.

This research aims to estimate the delineation of radioactive anomalies using machine learning models based on well logs.
Developed a machine learning model incorporating geological constraints to predict DRAD.
Conducted preprocessing of logs including outlier removal and feature standardization.
Implemented two models: random forest and physics-informed neural network (PINN), using data from well KH-24.
PINN achieved R^2 values of 0.90, 0.82, and 0.85 for training, testing, and validation, increasing after smoothing to 0.92, 0.90, and 0.89.
Validation with wells KH-17 and KH-21 indicated that predicted DRAD profiles matched measured data satisfactorily.
Physics-informed learning outperformed empirical models in prediction stability and geological reliability.

Bakry et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark