CCSNet.ai: Advanced AI-Powered Modeling for CO2 Storage

CCSNet.ai is a cutting-edge deep learning modeling suite developed at Stanford University by Gege Wen under the guidance of Prof. Sally M. Benson, designed to predict CO2 storage behavior in saline reservoirs with unprecedented speed and accuracy. Tailored for researchers and professionals in carbon capture and storage (CCS), it offers interactive, high-resolution models for both 2D and 3D reservoir simulations, enabling rapid assessment of reservoir conditions, injection designs, rock properties, and permeability maps. Leveraging innovative neural network architectures like Nested FNO and U-FNO, CCSNet.ai accelerates complex spatial-temporal predictions by nearly 700,000 times compared to traditional simulators, making it a transformative tool for climate and energy research.

Key Features:

• Reservoir-scale-2D-isotropic Model: Predicts isotropic permeability using R-U-Net for radial 2D simulations.
• Reservoir-scale-2D-anisotropic Model: Models anisotropic permeability with U-FNO and synthetic heterogeneous maps.
• Basin-scale-3D Model: Delivers high-resolution spatial-temporal predictions of CO2 gas saturation plume and pressure buildup using Nested FNO.
• Interactive Simulations: Real-time, user-friendly models accessible via a Streamlit-powered web interface.
• Comprehensive Data Support: Handles reservoir conditions, injection designs, rock properties, and permeability maps.
• High-Speed Predictions: Nested FNO enables near real-time results, accelerating simulations by up to 700,000×.
• User Manual & Demo Cases: Available for download to support onboarding and practical application.

Pricing: CCSNet.ai is offered as a free tool to support research and innovation in carbon storage, with potential access to premium features or institutional support through collaboration with Stanford Center for Carbon Storage and ExxonMobil.

Conclusion: CCSNet.ai stands as a powerful, accessible, and scientifically robust AI solution for advancing CO2 storage research, combining academic excellence with transformative computational speed to support sustainable energy solutions.