What question did this study set out to answer?

To evaluate and compare four ensemble machine learning algorithms for predicting drill bit rate of penetration using operational drilling and petrophysical formation parameters.

May 15, 2023SPE Western Regional Meeting

Ensemble Machine Learning for Data-Driven Predictive Analytics of Drilling Rate of Penetration (ROP) Modeling: A Case Study in a Southern Iraqi Oil Field

View Full Paper

Ask AI

Bookmark

Authors

DADhuha T. Al-Sahlanee RARaed H. AllawiThi Qar University WAWatheq J. Al‐MudhafarUniversity of Basrah

Discussion

Member takes

Implication

Machine learning study demonstrates accurate rate of penetration prediction across deep geological formations, indicating potential to optimize drilling operations and lower operational costs.

Key Points

To evaluate and compare four ensemble machine learning algorithms for predicting drill bit rate of penetration using operational drilling and petrophysical formation parameters.
Evaluated four ensemble machine learning algorithms—Random Forest, Gradient Boosting, Extreme Gradient Boost, and Adaptive Boosting—using subsurface data from three development wells spanning 19 geological formations to a depth of approximately 3200 m.
Trained predictive models on 14 operational and petrophysical parameters, including weight on bit, rotary speed, standpipe pressure, and logging metrics, dividing the dataset into an 85% training set and a 15% testing set.
Quantified model accuracy using root mean square prediction error and correlation coefficients across the testing subset and individual well depths.
Random Forest, Gradient Boosting, and Extreme Gradient Boost models achieved high accuracy and strong correlation between measured and predicted drilling rates across testing subsets and depth profiles, whereas Adaptive Boosting performed poorly.
Cross-validation models trained on combined data from three development wells yielded superior predictive performance compared to models trained on single-well reference datasets.

Cite This Study

Al-Sahlanee et al. (2023) studied this question.

synapsesocial.com/papers/69da991b85037e71b2684271 https://doi.org/10.2118/213043-ms

View Full Paper

Ask AI

Bookmark

May 15, 2023SPE Western Regional Meeting

Ensemble Machine Learning for Data-Driven Predictive Analytics of Drilling Rate of Penetration (ROP) Modeling: A Case Study in a Southern Iraqi Oil Field

View Full Paper

Ask AI

Bookmark

Authors

DADhuha T. Al-Sahlanee RARaed H. AllawiThi Qar University WAWatheq J. Al‐MudhafarUniversity of Basrah

Discussion

Member takes

Implication

Machine learning study demonstrates accurate rate of penetration prediction across deep geological formations, indicating potential to optimize drilling operations and lower operational costs.

Key Points

To evaluate and compare four ensemble machine learning algorithms for predicting drill bit rate of penetration using operational drilling and petrophysical formation parameters.
Evaluated four ensemble machine learning algorithms—Random Forest, Gradient Boosting, Extreme Gradient Boost, and Adaptive Boosting—using subsurface data from three development wells spanning 19 geological formations to a depth of approximately 3200 m.
Trained predictive models on 14 operational and petrophysical parameters, including weight on bit, rotary speed, standpipe pressure, and logging metrics, dividing the dataset into an 85% training set and a 15% testing set.
Quantified model accuracy using root mean square prediction error and correlation coefficients across the testing subset and individual well depths.
Random Forest, Gradient Boosting, and Extreme Gradient Boost models achieved high accuracy and strong correlation between measured and predicted drilling rates across testing subsets and depth profiles, whereas Adaptive Boosting performed poorly.
Cross-validation models trained on combined data from three development wells yielded superior predictive performance compared to models trained on single-well reference datasets.

Cite This Study

Al-Sahlanee et al. (2023) studied this question.

synapsesocial.com/papers/69da991b85037e71b2684271 https://doi.org/10.2118/213043-ms

Also Consider

Synapse has enriched 3 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark