Abstract:
As autonomous technology rapidly evolves, the concept of machines making decisions without human intervention captivates the public imagination—whether it's a self-driving car navigating busy streets or an automated rig drilling miles below the earth’s surface. This presentation delves into the journey of drilling automation, drawing compelling parallels to the self-driving vehicles that are transforming personal transportation. Over a six-year period, advanced machine learning (ML) models for Rate of Penetration (ROP) prediction were developed, refined, and deployed, mirroring the path of self-driving cars from theory to reality. These models enable rigs to autonomously optimize drilling performance, feeding recommendations directly into the Driller’s interface, thus reducing manual intervention.
The presenter will take you on a journey from concept to deployment, highlighting technical challenges like model development, data integration, and teamwork across data science, petroleum engineering, and operations. The presentation will also distill insights from five SPE papers, published by the presenter, to offer a comprehensive overview of this project's milestones and achievements.
Key advantages of automated drilling technology include bridging the skill gap between new and experienced personnel for improved consistency, enabling the drilling of complex wells, providing a centralized dashboard, and reducing drilling costs.
By understanding the transformative power of drilling automation, the audience will gain a glimpse into the future of oil and gas operations, where data-driven ML solutions drive enhanced performance and efficiency, ushering in a new era of autonomous drilling technology.
|