Physics of Fluids, Journal Year: 2024, Volume and Issue: 36(6)

Published: June 1, 2024

CO2 injection is a promising technology for enhancing gas recovery (CO2-EGR) that concomitantly reduces carbon emissions and aids the energy transition, although it has not yet been applied commercially at field scale. We develop an innovative workflow using raw data to provide effective approach in evaluating CH4 during CO2-EGR. A well-calibrated three-dimensional geological model generated validated actual data—achieving robust alignment between history simulation. visualize spread of plume quantitatively evaluate dynamic productivity single well. use three deep learning algorithms predict time histories rate feedback on production wells across various systems. The results indicate can enhance water-bearing reservoirs—CH4 increases with escalating. Specifically, increased diminishes breakthrough while concurrently expanding swept area. Deep exhibit superior predictive performance, gated recurrent unit being most reliable fastest among algorithms, particularly when accommodating series, as evidenced by its smallest values evaluation metrics. This study provides efficient method predicting before after injection, which exhibits speedup 3–4 orders magnitudes higher than traditional numerical Such models show promise advancing practical application CO2-EGR reservoir development.

Language: Английский