Neue Veröffentlichung: „Data-driven Adaptive Surface Control for Automated Directional Drilling“

Felix Häusser, Andreas Himmel, Kai S. Karvinen, Rolf Findeisen


Data-driven Adaptive Surface Control for Automated Directional Drilling


Directional drilling is the process of drilling boreholes that adhere to a specified reference well plan as closely as possible while ensuring safe operation. This process is typically controlled by repeatedly manipulating the steering inputs of an underground bottom-hole assembly. In contrast, the surface inputs, such as the mud flow rate, the drill string rotation speed, or the hook load, are only occasionally adjusted, requiring expert knowledge. This is due to the fact that it is challenging to model the influence of the surface inputs on the drilling process – especially the downhole activity. To tackle this challenge, a data-driven adaptive two degree of freedom control concept is proposed. This control concept applies a simplified data-based surrogate model of the directional drilling process. The surrogate model is subsequently used in an adaptive two degree of freedom control concept, splitting the tasks into reference tracking and disturbance rejection. The control concept is tested in a virtual directional drilling test environment to automatically adjust the drill string rotation speed and hook load to achieve a desired trajectory curvature.

DOI: 10.1109/CCTA54093.2023.10252434