Abstract

The bond calibration method (BCM) was designed as a calibration procedure to determine the effective Young's modulus (E) of a compact aggregate of discrete bonded particles. By systematically varying the Young's modulus of each particle–particle link (EB), the BCM adjusts the E of the aggregate, disregarding additional macroscopic parameters such as the Poisson's ratio (ν) and the ultimate compressive strength (UCS). In this study, an extension of the BCM is presented in which E, ν, and UCS are the target properties in the micro-characterization of the material. This new approach relies on parametric analysis of the influence of the bond microproperties on the macroscopic mechanical indices of the aggregate obtained from discrete element simulations of uniaxial compression tests (UCTs). The proposed methodology includes three main steps: (i) determination of the influence of particle–particle bond shear modulus (GB) on ν; (ii) calibration of the particle–particle bond Young's modulus (EB) based on UCT simulations; (iii) setting an appropriate value for the friction angle (ϕB), and performing discrete element method simulations to establish the influence of the cohesion microparameter (cB) on UCS. The reliability of the BCM was tested through DEM simulations of UCTs of Dokoohak limestone specimens with different particle radius distributions (5≤[Formula presented]≤10). The mean relative errors between the target macroproperties and the DEM simulations with microproperties obtained from BCM were 9.74%, 5.17%, 1.55%, for ν (0.25), E (26GPa), and UCS (52MPa) respectively.