CCSNet: a deep learning modeling suite for CO$_2$ storage
This addresses computational bottlenecks for researchers and engineers in subsurface flow simulations, specifically for carbon capture and storage, though it is incremental as it applies existing deep learning methods to a new domain.
The authors tackled the computational challenges of numerical simulation for carbon capture and storage by developing CCSNet, a deep-learning modeling suite that provides outputs like saturation distributions and pressure buildup, achieving results 10^3 to 10^4 times faster than conventional simulators.
Numerical simulation is an essential tool for many applications involving subsurface flow and transport, yet often suffers from computational challenges due to the multi-physics nature, highly non-linear governing equations, inherent parameter uncertainties, and the need for high spatial resolutions to capture multi-scale heterogeneity. We developed CCSNet, a general-purpose deep-learning modeling suite that can act as an alternative to conventional numerical simulators for carbon capture and storage (CCS) problems where CO$_2$ is injected into saline aquifers in 2d-radial systems. CCSNet consists of a sequence of deep learning models producing all the outputs that a numerical simulator typically provides, including saturation distributions, pressure buildup, dry-out, fluid densities, mass balance, solubility trapping, and sweep efficiency. The results are 10$^3$ to 10$^4$ times faster than conventional numerical simulators. As an application of CCSNet illustrating the value of its high computational efficiency, we developed rigorous estimation techniques for the sweep efficiency and solubility trapping.