A high-efficient method to upscale H2 migration within heterogeneous saline aquifers

Abstract

Drawing upon a range of energy storage methods that cover daily, weekly, and seasonal dynamics presents a route to reduce carbon dioxide emissions per unit of energy delivered. The utilization of renewable energy to generate, store, and recycle hydrogen stands out as a particularly promising strategy to address the seasonal fluctuations in energy production and usage. Deep saline aquifers offer a feasible solution for large-scale hydrogen storage, tailored to meet the demands of long-term storage requirements. This research introduces an upscaling method based on percolation theory, aiming to reduce computational costs in simulating H2 migration across diverse saline aquifers with different correlation lengths. Two geological models of distinct correlation lengths showcase the method's effectiveness. In the model with a 1.2-m correlation length, percolation-based upscaling yields final H2 saturation errors of 8.76 % in the main area and 3.7 % in the sink area, reducing runtime nearly sevenfold. Similarly, the model with a 4.0-m correlation length demonstrates final H2 saturation errors of 10.7 % in the main area and 1.27 % in the sink area, decreasing runtime almost fivefold. To bolster the credibility of the proposed upscaling method, parameters derived from the Brooks-Corey and van Genuchten models are calibrated to align with experimentally acquired H2-water multiphase flow properties. The resulting coarse-scale model accurately replicates primary permeability and H2 migration behavior, maintaining errors below 5 %. Importantly, dominant H2 migration mechanisms during underground hydrogen storage in saline aquifers are preserved in upscaled models, enabling efficient H2 saturation predictions beneath caprock. This study enhances our understanding of fine-scale H2 migration in complex geological systems and sheds light on integrating small-scale capillary barrier characteristics during upscaling.