CCSNet: Subsurface intelligence for carbon storage

Credit: Photo by Joshua Sortino via Unsplash.com
A breakthrough for AI-based physics simulation, unlocking faster, safer, and cheaper carbon sequestration.
About the Technology
Revolutionizing speedup in subsurface modeling that enables better permanent storage of carbon dioxide underground.
Carbon sequestration, or storing CO2 in underground geologic formations, is a rapidly growing solution vital to global decarbonization. Hundreds of thousands of carbon storage projects will need to be developed to sequester the 10 billion tons of CO2 per year required to reach our climate targets by mid-century. However, project development is constrained by the slow speed of current numerical simulation tools, which is required during every step of the CO2 project development pipeline, including obtaining a Class VI well permit.
To combat this bottleneck, we have developed an AI-based physics simulation technology that provides end-to-end subsurface geological modeling 700,000x faster than the current state-of-the-art numerical simulators. This faster modeling speed allows (1) quickly screening large numbers of potential injection sites, (2) making probabilistic estimates to guide better decision-making and safer projects, and (3) designing optimal carbon injection schedules and strategies. Removing bottlenecks that previously required hundreds of employee hours increases the CO2 storage success rate while lowering project development costs.
Support from the Stanford University HIT Fund will be instrumental in commercializing our technology.
Team Members
Publications:
- Gege Wen, Zongyi Li, Kamyar Azizzadenesheli, Anima Anandkumar, Sally M Benson (2023). Real-time high-resolution CO2 geological storage prediction using nested Fourier neural operators. Energy & Environmental Science, Issue 4.
- Wen, Gege, Zongyi Li, Kamyar Azizzadenesheli, Anima Anandkumar, and Sally M. Benson (2022). U-FNO–An enhanced Fourier neural operator-based deep-learning model for multiphase flow. Advances in Water Resources: 104180.
- Wen, Gege, Catherine Hay, and Sally M. Benson (2021). CCSNet: A deep learning modeling suite for CO2 storage. Advances in Water Resources 155 (2021): 104009.
- Gege Wen, Meng Tang, Sally M Benson (2021). Towards a predictor for CO2 plume migration using deep neural networks. International Journal of Greenhouse Gas Control.