Rockburst tendency prediction in a deeply buried tunnel based on numerical simulations

Deeply buried mountain tunnels are often exposed to the risk of rock bursts, which always cause serious damage to the supporting structures and threaten the safety of the engineers. Due to the limited data available, a suitable approach to predict the rockburst tendency at the preliminary stage becomes very important. In this study, an integrated methodology combining 3D initial stress inversion and rockburst tendency prediction was developed and subsequently applied to a case study of the Sangzhuling Tunnel on the Sichuan–Tibet Railway. The numerical modelling involved inverting the initial stress field using a multiple linear regression method. The tunnel excavation was simulated separately by FDM and DEM, based on a stress boundary condition from the inverted stress field. The comparative analysis demonstrates that the rockburst ratio calculated using DEM (76.70%) exhibits a slight increase compared to FDM (75.38%), and the rockburst location is consistent with the actual situation. This suggests that DEM is more suitable for simulating the stress redistribution during excavation in a jointed rock mass. The numerical simulation combined with the deviatoric stress approach effectively predicts rockburst tendency, meeting the engineering requirements. Despite its limitations, numerical simulation remains a reliable method for predicting rock bursts.