A study of rock pillar behaviors in laboratory and in-situ scales using combined finite-discrete element method models

With advances in numerical modeling techniques, the combined finite-discrete-element-method (FDEM) is increasingly being used to study the mechanical behavior and failure processes of brittle geomaterials under various loading conditions. The progressive rock fracturing process from initial formation to subsequent propagation can be simulated explicitly in FDEM models where the corresponding mechanical response in the simulations is governed by a set of microparameters. Previous studies have calibrated these microparameters by comparing the macroscopic results of simulated laboratory tests with those obtained in physical tests or by comparing larger-scale model results to observed field performance of engineered structures. Very few studies, however, have calibrated models at both the laboratory and field scales, and there is, therefore, a lack of understanding of the scale-dependency of FDEM material parameter inputs. To explore this scale issue, this study presents a series of calibrated laboratory-scale models of Creighton granite. Next, models of 8 m-wide pillars with different width-to-height ratios are calibrated against the strength trends predicted by empirical relationships. In the context of the pillar models developed, the effects of each input parameter on the macroscopic pillar stress-strain behavior are presented, considering the yield stress, peak strength, and post-yield behaviors. Ultimately, it is found that while the same set of input parameters can be used to reproduce both the expected laboratory specimen behavior and the expected pillar peak strengths at different width-to-height ratios, the post-yield behavior of the pillar models obtained using the laboratory parameters is much more brittle than would be expected in reality.