Machine Learning for Improved Directional Drilling
- Jacob Pollock (Oceanit Laboratories, Inc.) | Zachary Stoecker-Sylvia (Oceanit Laboratories, Inc.) | Vinod Veedu (Oceanit Laboratories, Inc.) | Neil Panchal (Shell International E&P Inc.) | Hani Elshahawi (Shell International E&P Inc.)
- Document ID
- Offshore Technology Conference
- Offshore Technology Conference, 30 April - 3 May, Houston, Texas, USA
- Publication Date
- Document Type
- Conference Paper
- 2018. Offshore Technology Conference
- 6.1 HSSE & Social Responsibility Management, 6 Health, Safety, Security, Environment and Social Responsibility, 1.5 drill Bits, 1.10 Drilling Equipment, 7.6 Information Management and Systems, 7.6.7 Neural Networks, 7 Management and Information, 7.6.6 Artificial Intelligence, 6.1.5 Human Resources, Competence and Training, 1.6 Drilling Operations, 1.6.6 Directional Drilling
- reinforcement learning, directional drilling, artificial intelligence, horizontal drilling, neural networks
- 48 in the last 30 days
- 985 since 2007
- Show more detail
- View rights & permissions
Directional drilling is a complex process involving the remote control of tool alignment and force application to a very long drill string subject to variable external forces. Controlling bit tool face orientation while ensuring adequate rate of penetration (ROP) is quite challenging, with aspects that have been described as more art than science. Improving this control helps preserve proper well trajectory and eliminate deviations that require corrective measures and add to well costs.
An artificial intelligence system was developed to learn from the actions of expert directional drillers and the mechanics of drilling simulations. Machine learning algorithms were employed to improve the efficiency of directional drilling: optimized ROP, less tortuous borehole, less personnel on board (POB), and consistency across operations. The system ingests historical and simulation data corresponding to the information used and actions taken by expert directional drillers and uses that data to generate decisions that result in efficient slide drilling.
To create a system for controlling tool face angle and guiding drill bit sliding during directional drilling, relevant historical data from directional drilling operations was gathered. Much of this data was recorded in the drilling logs, which the drilling operator traditionally uses to control drilling parameters. The collected data was then filtered and used to structure and train artificial neural networks and select appropriate hyperparameters. Reinforcement learning methods were used to refine the neural networks trained on historical data. A computational model for drill string physics was used to simulate the mechanics of directional drilling. A successfully trained network was considered one that minimized deviation from planned wellbore trajectory, minimized tortuosity, and maximized ROP.
The neural network developed could replicate the decisions of expert directional drillers within a small error (<3%). Reinforcement learning was then successfully used to improve network performance, particularly for conditions not previously considered.
Since the algorithm has demonstrated competence in the historical and simulated realms, it will be further tested as a real-time advisory system for control of directional drilling operations. The system will be tested in simulation with an expert directional driller before use in a field drilling operation. Ultimately, the algorithm can be directly integrated into drilling operations, enabling fully automated directional drilling.
|File Size||1 MB||Number of Pages||9|