INVESTIGATING THE RELATIONSHIP BETWEEN PORE CHARACTERISTICS, FRACTAL DIMENSION, AND PERMEABILITY OF LIMESTONE USING HIGH-PRESSURE MERCURY INTRUSION, SEM ANALYSIS, AND BP NEURAL NETWORK

In this study, limestone samples from a coal mine in the North China region were selected for analysis. High Pressure Mercury Intrusion (HPMI) and Scanning Electron Microscopy (SEM) experiments were conducted to explore the impact of pore characteristics and fractal dimension of limestone on permeability. Additionally, regression analysis and a Backpropagation Neural Network (BPNN) were employed to predict permeability. The results of this study reveal that the pore-throat distribution of the limestone samples is non-uniform, indicating significant heterogeneity. The difference of pressure curve morphology affects the permeability. Utilizing multivariate regression analysis, a relationship was established between permeability and parameters such as mean radius, porosity, and fractal dimension. Furthermore, the BP neural network was effectively employed to predict permeability values, with small discrepancies between predicted and measured values. This study establishes a link between microstructural attributes and macroscopic permeability providing a robust theoretical foundation for permeability assessment and engineering applications pertaining to limestone.