Date of Graduation
Statler College of Engineering and Mineral Resources
Petroleum and Natural Gas Engineering
Huyesin I Bilgesu
Drilling automation has focused on developing predictive controls based on existing formation and well sites for which abundant data is available. These methods are not suitable in new locations where there is little information and where drilling data has not been recorded. This study focuses on a proof-of-concept to allow drilling in locations with little or no data available by determining drilling parameters via an artificial intelligence algorithm. The methods used were tested for use in the second annual Drillbotics competition sponsored by the Drilling Systems Automation Technical Section (DSATS) of the Society of Petroleum Engineers (SPE).;This study addresses the difficulties and challenges faced in adapting artificial intelligence optimization algorithms for use with real-world applications. Furthermore, the limitations of such a system are examined and the breakdown between the algorithm and operational limitations are explored. A review of past drilling automation attempts and research was conducted and existing problems identified.;This research was completed on a pilot-scale drilling rig used by West Virginia University in the Drillbotics competition. The rig was used with multiple samples made in-house in order to provide a variety of materials, inclinations, and drilling conditions. The review of the test was subject to professional judges to provide an unbiased decision and serve to advance this study.
Cox, Zachary, "Adaptation of Simulated Annealing for Rate of Penetration Optimization in Automated Drilling" (2017). Graduate Theses, Dissertations, and Problem Reports. 5405.