润湿
流离失所(心理学)
多孔介质
多孔性
材料科学
机械
流量(数学)
氢
原位
体积热力学
接触角
岩土工程
复合材料
计算机模拟
残余油
垂直位移
地质学
GSM演进的增强数据速率
多相流
芯(光纤)
残余物
矿物学
工作(物理)
石油工程
流体力学
计算
作者
Daiying Feng,Rui Song,Jiajun Peng,K. Liu,Chunhe Yang
摘要
ABSTRACT The heterogeneous in situ wettability of porous rocks exerts a critical influence on both the effective hydrogen storage capacity and the efficiency of hydrogen injection and production in underground hydrogen storage (UHS). To improve the realism and predictive reliability of pore‐scale hydrogen–water displacement simulations for underground hydrogen storage in depleted oil and gas reservoirs, it is necessary to consider in situ heterogeneous wettability rather than using a single averaged contact angle. Nevertheless, most existing studies oversimplified this complexity by assigning a single averaged value of contact angles, which failed to represent the intrinsic pore‐scale heterogeneity of wettability. To address this limitation, a series of visualised multi‐cycle hydrogen–water displacement experiments were conducted, yielding 1354 in situ contact angle measurements at the pore scale. Based on these data, a digital rock model was constructed based on the pore structure of the natural rock samples and the spatial distribution of in situ contact angles. This model was then employed in numerical simulations of the hydrogen–formation water displacement process. The simulation results revealed that incorporating in situ wettability distributions enables more accurate reproduction of the experimentally observed displacement fronts and residual gas configurations. In contrast, neglecting wettability heterogeneity oversimplified the residual gas morphology and systematically overestimated the working gas volume in UHS. Overall, this study bridges the gap between macroscopic averaging and pore‐scale in situ characterisation of contact angles, providing a robust theoretical framework and modelling strategy for elucidating hydrogen flow mechanisms and optimising the design of UHS systems.
科研通智能强力驱动
Strongly Powered by AbleSci AI