Today numerical modeling along with empirical methods has an important role in the design of underground excavations in rock. Numerical analysis can provide reasonable illustration of the field conditions taking into consideration major fundamentals of the excavation such as the geomechanical features of the ground and the in situ stress condition. This research presents the comparison between support system designs of the Gate Chamber structure by empirical methods (Q-system and RMR) and numerical analysis using 3DEC. The structure is a part of intake structure to control the water level in Sardasht dam. The surrounding rock mass of the structure consists of highly weathered and metamorphic slate. The model has been run in 3DEC and it was observed that the required rock support to stabilize this structure were conservative with respect to empirical methods. The rockbolt loading capacity was determined by pull-out test in the field and compared with axial load of rockbolt in 3DEC. The rock support resulted from the empirical method is weaker than that from the numerical method. By excavation progress and repetitious collapses it was concluded that the rock support design using empirical method must be carried out in conjunction with numerical analysis to prevent catastrophic events.
There are various methods for evaluating the stability of underground openings. These methods are widely used for designing tunnels, underground spaces, slopes and etc. In rock mechanics different methods of numerical analyses such as FEM, BEM and DEM has been developed rapidly during the recent decade (Negro & Queiroz, 1999). These methods are broadly employed (Cundall, 1971) in engineering works for designing structures in rock like tunnels, underground openings, slopes, dam foundations and etc. Determination of geological structures, insitu geomechanical properties of rock, initial insitu stresses, ground water elevation and permeability are arduous tasks.