Data-driven prediction of rate of penetration (ROP) in drilling operations using advanced machine learning models
Abstract Predicting the rate of penetration (ROP) is critical for optimizing drilling performance, yet it remains a complex task due to the interplay of multiple geological and operational parameters. This study comprehensively evaluates machine learning models, utilizing a real-time, high-resolutio...
Saved in:
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2025-06-01
|
| Series: | Journal of Petroleum Exploration and Production Technology |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s13202-025-02018-3 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Abstract Predicting the rate of penetration (ROP) is critical for optimizing drilling performance, yet it remains a complex task due to the interplay of multiple geological and operational parameters. This study comprehensively evaluates machine learning models, utilizing a real-time, high-resolution dataset from drilling operations in southeast Iraq. Among the models tested, the Random Forest algorithm demonstrated outstanding performance, achieving an R2 of 0.955, a Mean Squared Error (MSE) of 0.119, and an Average Absolute Relative Error (AARE%) of 7.683, highlighting its reliability and robustness in predicting ROP. Sensitivity analysis and SHAP (Shapley Additive Explanations) also identified fracture pressure, kinematic viscosity, and rotary speed (RPM) as the most influential parameters affecting ROP. While alternative methods like Decision Tree and AdaBoost showed signs of overfitting, the results emphasize the Random Forest model’s superiority in balancing accuracy and generalizability. This research underscores the potential of advanced machine learning techniques in enhancing drilling performance, offering significant implications for real-world applications. |
|---|---|
| ISSN: | 2190-0558 2190-0566 |