Background
In total 500 meters have been drilled autonomously at our full-scale offshore type of rig Ullrigg. By autonomous drilling we mean a system capable of taking its own decisions by evaluating the current conditions and adapting to them while considering multiple horizon strategies to fulfill the drilling operation goal. Markov decision process (AI method that has been customized for this purpose) has been used for decisions. To make a safe operation it is important that the driller is up to date.
Innovation
Develop an explainability module to so that the driller has a better overview of the operation and why the system has made the decisions that have been made. This will enable the driller to take over manually if necessary.
Value
Automated drilling has made the drilling operation more consistent and independent of the drilling crews. Statistics show that incidents have been reduced. It is challenging to foresee the additional value for autonomous drilling before having any statistical material. The aim is to continue this pathway with reduced incidents and potentially also NPT. If autonomous drilling also can reduce manning offshore remains to be seen.
Status
The autonomous drilling solution has been made D-WIS compatible, and it is ongoing work with the explainability module.
Next steps
- Demonstrate Autonomous Drilling included explainability using the D-WIS framework both in OpenLab and at the full-scale rig Ullrigg without heave
- Similar simulator testing in OpenLab with heave
- Offshore testing of Autonomous Drilling included explainability using the D-WIS framework at the semi-sub Deepsea Stavanger.
Reference paper for levels of autonomous drilling systems:
Many industry actors claim to have autonomous solutions for drilling and well construction. For the DigiWells center and the international drilling community it is important to have a common understanding and avoid inflation of the word autonomous. By working on a definition together with important international players, our goal has been to have a reference paper which was achieved on the 2024 IADC/SPE Drilling Conference: Taxonomy Describing Levels of Autonomous Drilling Systems: Incorporating Complexity, Uncertainty, Sparse Data, With Human Interaction, John de Wardt, Eric Cayeux, Rodica Mihai, John D. Macpherson, Pradeep Annaiyappa and D. Pirovolou. https://doi.org/10.2118/217754-MS
This work is part of a spin-off project linked to the centres workpackage 4 and 5. In a project supported by the RCN, Autonomous drilling was demonstrated on Ullrigg. In DigiWells Innovation program the focus has been on explainability, and Autonomous Drilling has been adapted to the D-WIS infrastructure
The first time ever, autonomous drilling was demonstrated on a full-scale offshore type drilling rig at the NORCE Ullrigg Test Center. The autonomous system decides and executes by itself every action that shall be taken and this without human intervention. The goal of autonomous drilling is to drill as fast as possible in a safe and efficient way, while taking actions that help preventing and avoiding problems.
Eric Cayeux, Benoit Daireaux, Adrian Ambrus, Rodica Mihai, Liv Carlsen: “Autonomous Decision-Making While Drilling” / Energies / 2021. https://doi.org/10.3390/en14040969
Rodica Mihai, Eric Cayeux, Benoit Daireaux, Liv Carlsen, Adrian Ambrus, Per Simensen, Morten Welmer, Matthew Jackson: “Demonstration of Autonomous Drilling on a Full-Scale Test Rig” /SPE Annual Technical Conference and Exhibition, Houston, Texas, USA, 2022/ SPE-210229-MS. https://doi.org/10.2118/210229-MS
R. Mihai, B. Daireaux, A. Ambrus, E. Cayeux and L. Carlsen, “On Transitions Functions Model for Decision-Making in Offshore Operations,” 2022 IEEE 17th International Conference on Control & Automation (ICCA), 2022, pp. 309- 314, https://doi.org/10.1109/ICCA54724.2022.9831911
