Numerical simulation of the influence of small-scale defects on the true-triaxial strength of rock samples

A numerical model is used to investigate the influence of randomly located ‘defects’ on the strength of rock samples with a Mogi–Coulomb three-dimensional failure criterion that are subjected to polyaxial states of stress. A methodology is presented to compute ‘equivalent’ Mohr–Coulomb strength parameters as a function of the ‘instantaneous’ stress, and sensitivity analyses are employed to study the influence of the stress state, of the total percentage of defects, and of the type of spatial distribution of defects within the sample. Results indicate that increments of strength due to the intermediate principal stress are less relevant for samples with defects, hence suggesting that the influence of polyaxial stress states is less significant for rock masses (that incorporate ‘defects’ or ‘discontinuities’ into their structure) than for intact rock. Results also indicate that the strength of rock with defects can be modelled as a ‘homogeneous’ material with reduced strength properties. The reduction factors are a linear function of the percentage of defects, and they tend to be greater for “lumped” defects than for “isolated” defects.