Effective elastic properties of 3D fractured systems
Abstract
Fractures have a significant impact on rock mass strength, which is a concern for rock engineering applications like excavation or repository design, support design, slope stability and caving in mines. Addressing this issue includes both the description of the fracturing pattern and the relationship between fracture characteristics and rock mass mechanical properties. While empirical knowledge is widely used in DFN (Discrete Fracture Network) engineering, theoretical relations between the DFN density and its effective elastic properties have been derived, to our knowledge, only for Poissonian (i.e. randomly distributed) low-density (i.e. where fractures are not almost intersecting) networks, with a narrow range of fracture radius. In the present study, the Young modulus of a solid containing a distribution of circular non cohesive frictionless fractures is calculated by means of distinct-element models and Green’s function methods. These methods allow the analysis of geologically realistic discrete fracture networks including non-Poissonian DFN with high densities, power-law length distributions, non-uniform orientation distributions and various mechanical properties. For a given range of DFN statistics, we found that the elastic modulus averaged over ten or more realizations decreases exponentially with the percolation parameter, and that the decay rate is controlled by fracture orientations and mechanical properties (stiffness). Large variability from the mean exists for heavy tailed fracture length distributions, when the probability of having fractures as large as the system size is not negligible. Based on these results and on the observed statistical distributions of fractures, we draw conclusions on the mechanical properties of rocks. In particular, we discuss some limitations of the GSI system and derive practical applications for the DFN modeling process.