Abstract
The flow characteristics of water and gas are closely linked to pore structure of porous media, which is of critical importance across various scientific and industrial fields. In this study, synthetic porous media with varying grain sizes and porosity were generated, and their corresponding pore structures were characterized using pore network modeling. Furthermore, the intrinsic permeability, water retention curve, water-gas relative permeability and relative gas diffusivity of the synthetic porous media were simulated via pore network modeling. The results demonstrate that the pore networks extracted from images can effectively distinguish pore structural characteristics. Specifically, the mean pore diameter, throat diameter, and throat length were larger in coarse-grained media compared to fine-grained media of the same porosity. In contrast, fine-grained media exhibited higher values for pore number, throat number, and coordination number. Additionally, the distributions of pore diameter, throat diameter, throat length and coordination were found to follow a lognormal distribution. Porous media with coarse grains and larger porosity exhibit greater intrinsic permeability and relative gas diffusivity compared to media composed of finer grains or lower porosity. The water-retention curves were fitting by van Genuchten model, revealing an exponential relationship between parameter α and throat diameter (or pore diameter). But the parameter n did not show a clear trend across various synthetic porous media, which is attributed to the relatively narrow range of pore size distribution. Similarly, for water-gas relative permeability, the critical water saturation did not vary significantly across different porous media. A strong correlation was observed among the pore structural parameters, irrespective of grain shape and size. Both intrinsic permeability and relative gas diffusivity exhibited a power-law relation with the porosity as well as with pore or throat radius. Moreover, the relationship between intrinsic permeability and relative gas diffusivity can be expressed as k = 166.51(Dp/D0)0.98, which provides a direct means of estimating relative gas diffusion from intrinsic permeability directly.