Innovative AI-Powered Method for Advanced Reservoir Characterization
Author: subsurfaceAI

SubsurfaceAI Inc. & Eni Reveal

Innovative AI-Powered Method for Advanced Reservoir Characterization at the Fifth EAGE Borehole Geology Workshop

Calgary, Alberta, Canada – SubsurfaceAI Inc., is excited to announce a pioneering development in the field of geoscience, developed in collaboration with Eni. During the recent Fifth EAGE Borehole Geology Workshop, Eni presents a novel artificial intelligence technique for predicting petrophysical properties and sedimentary structures directly from core photographs. This state-of-the-art method represents a significant leap forward in the exploitation of digital core imagery for reservoir characterization.

The collaboration between SubsurfaceAI Inc. and Eni has culminated in the development of two cutting-edge methodologies leveraging Deep Learning, specifically through the application of the VGG19 Convolutional Neural Network enhanced by Transfer Learning. This approach allows for the detailed analysis of clastic reservoirs by extracting valuable insights from slabbed core photos, a resource that has been underutilized in the past.

Innovative Approach for Enhanced Reservoir Analysis

The first methodology focuses on the prediction of petrophysical properties, such as permeability, from core plug measurements and slabbed core images under various lighting conditions. The second methodology aims at identifying sedimentary structures, like massive and laminated intervals, crucial for understanding fluid dynamics within the reservoir. Both methods promise to revolutionize the way geoscientists interpret core data by providing continuous curves and zonations across cored intervals, thus enabling more accurate digital core model construction, log-based facies characterization, and synthetic petrophysical curve predictions.

Impressive Results from Offshore well application

Initial testing on offshore wells in clastic reservoirs has shown remarkable success, demonstrating the methodologies’ ability to accurately predict petrophysical parameters. Concerning the interpretation of sedimentary structures deposited in a deep-water depositional setting, the methodology shows encouraging results, being able to recognize a good number of depositional features. These advancements not only improve the resolution and reliability of reservoir characterization but also facilitate the upscale of data to meet various modeling requirements, thereby addressing the longstanding challenge of cross-scaling issues.

Future Directions and Applications

SubsurfaceAI Inc. plans to refine these methodologies for reservoir characterization. This breakthrough underscores the potential of AI in transforming geoscience research and exploration, paving the way for more efficient and detailed reservoir modeling and analysis.

SubsurfaceAI Inc. is committed to advancing the frontiers of geoscience, leveraging the power of artificial intelligence to unlock new possibilities in reservoir characterization.

For more information on this breakthrough technology and other developments, please visit [SubsurfaceAI Inc.’s website].

About ENI

Eni is a global energy company engaged in the exploration, development and extraction of natural gas and oil, power generation from traditional and renewable sources, refining and chemicals. The goal of sustainability incorporates business at every level.

About SubsurfaceAI Inc.

SubsurfaceAI Inc. is a leading provider of artificial intelligence solutions in the field of geoscience and reservoir engineering. Based in Calgary, Alberta, Canada, SubsurfaceAI is dedicated to developing cutting-edge technologies that enhance the accuracy and efficiency of subsurface exploration and characterization.