Developing a control system for a fully automated CML system using a physical base model, AI & Digital Twin technologies in real-time

Prosjektforslaget omhandler et innovativt kontrollsystem for helautomatiserte bore- og sementeringsoperasjoner med kontrollert slamnivå i borestigerør(CML). Systemet, som er basert på en fysisk basismodell med AI og Digital Twin-teknologi, muliggjør beslutningstaking i sann tid. Løsningen vil forbedre sikkerhet, effektivitet og miljømessige forhold. Ved å opprettholde konstant bunnhullstrykk (BHP) og effektivt styre slamnivået i stigerøret, muliggjør CML-systemet automatisert boring gjennom utfordrende formasjoner, redusert væsketap og forbedret hullrensing samtidig som det reduserer sannsynlighet for fastsetting av borerør. I tillegg vil tidlig deteksjon av innstrømning av formasjonsvæske i borehullet (brønnspark) være mulig. En kan enkelt skifte over til et konvensjonelt boresystem om ønskelig. Med pågående innsats for å bruke maskinlæringsalgoritmer for tidlig oppdagelse av problemer, representerer dette automatiserte CML-systemet en betydelig forbedring av boreteknologi.

The proposal outlines a comprehensive approach to developing a control system for fully automated Controlled Mud Level (CML) operations, integrating a physical base model (FEM) with AI & Digital Twin technologies. The aim is to enable real-time decision-making and autonomous operation throughout various drilling stages, including drilling, connection, cementing, and testing. By combining these advanced technologies, the project seeks to revolutionize drilling practices, enhancing safety, efficiency, and environmental sustainability. The automated CML system, empowered by AI and a physical base model, offers benefits across drilling operations. Moreover, the system's ability to maintain constant BHP by controlling mud levels in the riser enables seamless drilling through formations with narrow mud windows and facilitates uninterrupted drilling for longer sections. Additionally, it effectively mitigates the effects of ECD and minimizes fluid loss in problematic zones, while improving hole cleaning and preventing stuck pipe issues. With the capability to detect kicks or influxes faster than traditional rig systems. Furthermore, it offers the flexibility to switch to conventional drilling if needed and improves well integrity during cementing operations. Additionally, extensive development efforts are focused on refining ML algorithms for early detection of kick and loss during connection, performing automated history matching of CML parameters during connection with previous connection data. In summary, the proposed control system for fully automated CML operations represents a significant advancement in drilling technology, with the potential to enhance safety, efficiency, and environmental sustainability in the oil and gas industry. The goal is to implement the CML system on the rig permanently so that the rig owners can use that in cases when it is required and the operation of the system becomes fully automated with the developed software.