The global push for decarbonization is driving increased demand for geothermal energy and Carbon Capture, Utilization, and Storage (CCUS) projects, which often involve complex and high-risk drilling. Simultaneously, stringent safety regulations and rising operational costs in traditional oil and gas exploration necessitate advanced predictive analytics. This technology provides a critical solution to mitigate risks, optimize resource allocation, and maintain project schedules in an increasingly challenging and regulated global drilling landscape.
Provides High-Accuracy Stuck-Pipe Risk Prediction: Leverages 2D histograms from existing drilling parameters for real-time, high-precision risk assessment, surpassing conventional empirical rules and threshold alerts.
Enables High-Efficiency Analysis with Limited Data: Efficiently extracts multi-dimensional features using limited input data (2D histograms of bit depth and other drilling parameters), avoiding the need for extensive raw data.
Offers Strong Exclusive Advantage Through Robust IP: Secured through a meticulous patenting process by the University of Tokyo, overcoming examiner objections to establish a strong patent with recognized technical uniqueness among five prior art documents.
This patent protects the core elements of stuck-pipe prediction, including the selection of drilling parameters, the method for generating 2D histograms, and the application of the prediction model. It was granted after overcoming rigorous examiner objections and clarifying its scope through amendments, indicating a robust and defensible intellectual property with 12 claims.
This patent focuses on predictive analytics using existing data. Licensees could develop new sensor technologies for novel drilling parameters or integrate this prediction engine into fully autonomous drilling control systems for automated intervention, expanding beyond mere risk alerts.
A single stuck-pipe event during drilling is estimated to incur additional costs of ~$200K–$700K (AI est.) per incident, covering downtime, recovery operations, and material expenses. By reducing the average of 5 annual stuck-pipe events by 20% through this technology, an annual cost saving of ~$800K (AI est.) could be achieved ($400K/event (AI est.) × 5 events × 20%).
X: Real-time Prediction Accuracy
Y: Operational Cost Efficiency