.png){kind=icon}
.png){kind=icon}
Return
Optimizing hydrogen storage in the subsurface using a reservoir-simulation-based and deep-learning-accelerated optimization method
.png){kind=icon}
PRE
AI
DOI：10.1016/j.ijhydene.2025.03.031.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
This paper addresses the challenge of optimizing subsurface hydrogen storage in porous media, a crucial component for advancing energy transition. The multifaceted nature of this challenge stems from the complex physics governing the process, coupled with operational limitations, and subsurface geological uncertainties. The objective is to maximize recoverable hydrogen while maintaining constraints, with the hydrogen deliverability index as the objective function, calculated using a compositional reservoir simulator. A deep learning-accelerated gradient (DLAG) method is applied to the Brugge field case study across two scenarios. In the first, we optimize the placement of eight storage wells in a single subsurface realization, comparing the results with and without DLAG. In the second, we extend the analysis to include five different subsurface realizations and impose specific location constraints. The results demonstrate that the DLAG method achieved faster convergence on average. The optimization method proved to be effective and practical in improving the hydrogen storage efficiency in the subsurface.
Keywords:
Underground hydrogen storage
Numerical optimization
Reservoir simulation
Journal
IF:
8.3
Papers:
5.4W
Citations:
23.1W