Fluid seepage behavior in porous media is widely existed in many fields such as biomedicine, energy extraction and geotechnical engineering. As a key physical parameter to characterize the transport ability of porous media, the numerical value of permeability directly determines the seepage performance of fluid in the media. Therefore, a comprehensive understanding of the intrinsic relationship between permeability and the pore structure of porous media is crucial for advancing related fields.
Traditional studies on seepage mostly rely on macroscopic experiments or theoretical derivations, establishing the relationship between permeability and porosity by analyzing the influence of seepage conditions on seepage performance. However, such methods are difficult to reveal the intrinsic relationship between the microscopic pore structure and permeability. Due to the complexity of pore structure, even media with the same porosity may exhibit different permeability due to their structural differences. Therefore, this paper focuses on the influence mechanism of the microscopic structure of porous media on the permeability of the media. Based on the Quartet Structure Generation Set, three-dimensional anisotropic porous media models with different pore structures are constructed. The correlation between five structural characteristic parameters including anisotropy is extracted and analyzed. The Multiple-relaxation-time Lattice Boltzmann method is used to numerically simulate fluid seepage, visually presenting the seepage velocity and streamline distribution characteristics in different pore regions, and systematically analyzing the influence of the pore structure characteristics of porous media on the permeability of the medium.
The results show that in the microstructure characteristics of porous media, the anisotropy factor is independent of porosity, shape factor, specific surface area and tortuosity, and affects the seepage behavior in the medium. The significance of the influence of the five structural characteristic parameters on permeability from high to low is as follows: porosity, shape factor, tortuosity, anisotropy factor and specific surface area. Without considering anisotropy, the prediction accuracy of the KC model is only 0.8933. To enhance the accuracy of the porous medium model, this paper establishes a porous medium permeability prediction model composed of five structural characteristic parameters. The prediction accuracy is improved to 0.9982, expanding the application scope and prediction accuracy of the current KC model.