Loading...

A Practical Approach with Ensemble-Driven Rate of Penetration Prediction and Optimization

Kelishami, A. S ; Sharif University of Technology | 2024

151 Viewed
  1. Type of Document: Article
  2. DOI: 10.1109/ICTEM60690.2024.10631977
  3. Publisher: 2024
  4. Abstract:
  5. This research aims to predict and optimize the rate of penetration (ROP) in drilling operations using artificial intelligence. An ensemble machine learning model is implemented to forecast ROP based on drilling reports. The model is evaluated using metrics like RMSE and optimized with contour plots depicting expected trends based on weight on bit (WOB) and rotational speed per minute (RPM). Additionally, a multiple regression model is chosen to optimize the Rate of Penetration (ROP). This allows predicting and optimizing ROP using data-driven AI techniques, bridging novel technologies and traditional methods in the oil industry. The results demonstrate the feasibility of applying machine learning to enhance productivity in drilling operations. Further work can expand the models with additional parameters and more advanced algorithms. Overall, this research exemplifies integrating AI into vital industries to augment human expertise. © 2024 IEEE
  6. Keywords:
  7. Prediction ; Regression ; Adversarial machine learning ; Petroleum industry ; Contour plot ; Drilling operation ; Machine learning models ; Multiple regression modelling ; Neural-networks ; Optimisations ; Rate of penetration ; Rotational speed ; Weight on bits ; Prediction models
  8. Source: 2024 9th International Conference on Technology and Energy Management, ICTEM 2024 ; 2024 ; 979-835032979-7 (ISBN)
  9. URL: https://ieeexplore.ieee.org/document/10631977