Simulation of unstable rock failure under unloading conditions

Abstract

Rockburst is an unstable rock failure and one of the most hazardous problems in deep hard-rock mines. Before excavation, rocks are loaded under a polyaxial condition. Upon excavation, the rocks at the excavation boundaries are loaded in the tangential direction and unloaded in the radial direction. Understanding rock behaviour under this excavation loading condition is critical for developing measures to control rock failure in underground construction. In this paper, numerical simulation results of unstable rock failure using an explicit finite element tool are presented. Firstly, uniaxial compression tests were simulated to confirm the suitability of the adopted numerical tool for simulating unstable rock failures. Transferred energy ratio (TER) and loading system reaction intensity (LSRI) were proposed as indicators to distinguish between stable and unstable failures. Secondly, unstable rock failures under polyaxial unloading conditions were simulated. The influences of loading system stiffness (LSS), specimen's height to width ratio (H/W), and intermediate principal stress ( 2 ) on rock failure were investigated. The simulation results showed that rock failure was more violent when the loading system was softer, the specimen was taller, and the confinement was lower. The modelling approach presented in this paper can be useful for predicting unstable rock failure and estimating released kinetic energy, which is important for designing rock support in deep tunnels to control rockburst damage.Résumé : Les coups de toit sont des phénomènes de rupture de roche instable et constituent les plus grands dangers dans les mines en roche dure. Avant les travaux d'excavation, les roches sont soumises à un chargement multiaxial. Durant ces travaux, les roches situées sur les bords de la mine sont soumises à un chargement tangentiel et le relâchement de charge se fait selon un axe radial. Il est essentiel de bien comprendre le comportement de la roche lors d'une excavation dans de telles conditions de chargement afin de pouvoir mettre en place des mesures visant à limiter la rupture de la roche dans les chantiers souterrains. Dans le présent article, on présente les résultats de simulations numériques de mécanisme de rupture de roche instable, réalisées à l'aide d'une méthode claire, basée sur la méthode des éléments finis. Dans un premier temps, on a simulé des essais de compression uniaxiale pour confirmer la pertinence de la méthode numérique adoptée permettant de simuler les mécanismes de rupture de roche instable. Le taux d'énergie transférée (TET) et l'intensité de la réaction du système de chargement (IRSC) ont été proposés comme indicateurs permettant de faire la distinction entre les ruptures stables et instables. Dans un deuxième temps, on a simulé des ruptures de roche instable dans les conditions de relâchement de charge multiaxial. On a également étudié l'influence de la rigidité du système de chargement (RSC), le quotient de la hauteur du spécimen par la largeur de celui-ci (H/W) et la contrainte principale i...

Influence of strain energy released from a test machine on rock failure process

“…2D numerical modeling results using the Discrete Element Method found that the post-peak stress-strain curve of rock under the critical loading condition was close to that under a rigid loading condition (LSS = ∞) (Kias and Ozbay 2013). This observation was confirmed by Manouchehrian and Cai (2015) using a 3D Finite Element Method (FEM) numerical model.…”

“…However, ∆E in was not excluded in previous numerical models (Hemami and Fakhimi 2014;Kias and Ozbay 2013;Manouchehrian and Cai 2015) and the rock failure processes observed were the product of ∆E t plus ∆E in . If ∆E in is provided during the rock failure process, the rock failure type under the loading condition of LSS > λ depends upon the amount of ∆E in .…”

“…Experimental and numerical studies have demonstrated that rock failures are violent when LSS < λ (Cook 1965;Feng et al 2016;He et al 2015;Hudson et al 1972;Manouchehrian and Cai 2015;Zhao and Cai 2014). Unfortunately, ∆E in in these studies was not isolated to examine the influence of ∆E t on rock instability.…”

“…In this section, the rock failure process is interpreted in a more quantitative way using an indicator of unstable rock failure. According to the indicator proposed by Manouchehrian and Cai (2015) to distinguish between stable and unstable rock failures in laboratory testing, loading system reaction intensity (LSRI) is defined as the ratio of the maximum reaction velocity (V max ) of the loading platen (or test machine) at the rock specimenloading platen contact to the applied loading velocity (V 0 ) at the other end of the platen ( Figure 18). V 0 at the boundary is constant; however, V max at the rock specimen-loading platen contact D r a f t 27 varies (unless the loading platen is a rigid body), depending on the relation between LSS and the post-peak stiffness of the rock specimen.…”

“…Table 4 shows the relation between LSRI and LSS. Manouchehrian and Cai (2015) pointed out that LSRI is normally smaller than 2 for stable rock failure; LSRI of 2 to 4 can be loosely regarded as transition between the stable and the unstable rock failures. It is seen that even though ∆E in is isolated from the energy transfer in the rock failure process, the LSRI values can be used to identify rock failure type.…”