Micro‐Parameter Sensitivity Analysis and Validation of the Mechanical Behaviour of Soft–Hard Composite Rock Based on 2D‐DEM

ABSTRACT Calibration of microparameters in numerical simulations is a critical factor affecting model accuracy. To determine the relationship between macro‐ and microparameters in composite rock masses and the influence of soft rock layer proportions on mechanical properties, this study employed PFC2D software to construct numerical models of composite rock masses with varying soft‐to‐hard layer thickness ratios. A systematic analysis was conducted to investigate the influence of microparameters on macro‐mechanical properties. For composite rock bodies with varying soft rock layer thicknesses, as the soft rock proportion increases, the failure mode gradually shifts from shear failure dominated by hard rock to foliated failure dominated by soft rock. Cracks propagate along bedding planes and are constrained by hard rock layers. When the soft rock layer thickness increases from 10% to 90%, the total number of cracks increases by approximately twofold, while the proportion of shear cracks decreases from 75% to 40%. The crack counting rate exhibits exponential growth with stress and synchronizes with stress amplitude, peaking at 180–200 times/s − 1 . Laboratory test results align with numerical simulations, demonstrating that the PFC model accurately predicts composite rock mass strength (error ≤3.7%) and elastic modulus (error ≤4.6%). This study provides theoretical support for correlating macro‐ and micro‐mechanical properties in composite rock masses of varying hardness, offering significant reference value for stability assessment and reinforcement design in underground engineering.