Study on the Relationship Between Tortuosity Distribution Spectrum and Permeability Based on Digital Cores

This study conducts a systematic analysis of the negative correlation between pore tortuosity and permeability in the seepage characteristics of porous media within rocks, highlighting the theoretical limitations of traditional permeability prediction models based on average tortuosity. Due to the complexity of internal streamline networks in rocks, a single average tortuosity parameter fails to accurately characterize heterogeneous flow-field features. To address this, a refined numerical simulation method integrating 3D digital core reconstruction and multi-physics coupling is proposed. First, a 3D digital model of the pore structure of a real core is constructed using X-ray micro-CT scanning technology, and steady-state flow fields within the porous medium are simulated via the Finite Element Method (FEM). Subsequently, the Lattice Boltzmann Method (LBM) is employed to track fluid particle trajectories, reconstructing a 3D streamline network and quantitatively analyzing streamline length distributions. Numerical simulations on multiple core samples reveal that selecting the tortuosity parameter corresponding to the 95th percentile (95% of streamline lengths do not exceed this threshold) yields lower relative errors between calculated and simulated permeability values compared to traditional average tortuosity models, demonstrating significantly improved accuracy. These results indicate that the proposed tortuosity metric enhances the effectiveness of reservoir permeability predictions, offering critical application value for optimizing fluid migration efficiency assessments in hydrocarbon reservoir development.