Excavation disturbance analysis based on crack tensor model and virtual fracture model for predicting the rock mass behavior during excavation of shafts and research galleries at the Mizunami Underground Research Laboratory

The Japan Atomic Energy Agency (JAEA) is conducting the Mizunami Underground Research Laboratory (MIU) Project in order to develop comprehensive geological investigation and engineering techniques for deep underground applications (e.g. repository of HLW). The purpose of this study is to contribute to the evaluation of the mechanical stability of a research drift and to plan the future studies. The MIU Project is now on the Construction Phase (Phase II). The rock mechanics studies in Phase II comprehend laboratory testing of core specimens for obtaining rock mechanical properties and, hydraulic fracturing testing in boreholes for measuring the in-situ stresses at sub-stages. Studies of the Excavation Disturbed Zones are also planned in the shafts at a depth of around -500 m from the ground surface. A 3D Crack Tensor Model analysis was carried out to predict the rock mass behavior of the complex (articulated) section at -500 m. Properties were chosen based on the Rock Mechanics Model constructed in the Surface-based Investigation Phase (Phase I). The Virtual Fracture Model was applied to analyze the change the hydraulic conductivity. The results of this study can be summarized as follows: 1) As rock quality is getting low, the convergence of the shaft and drift increases. Also the maximum shear stress in the rock mass and, the stress in the support system increase, while the change of the distribution of the safety factor and hydraulic conductivity are almost negligible. 2) For the convergence of the shaft, an increase of about 10% was found at the intersection between shaft and gallery. The domain of influence of the convergence from the intersection ranged over twice as much the typical diameter of the drift. In the drift, the maximum value of the convergence at the intersection was also about twice the value away from the intersection. 3) Compared to numerical result that disregard the Excavation Damaged Zone, numerical results that consider the Excavation Damaged Zone showed increased convergence and stresses in the support system. In particular, the maximum value of the hydraulic conductivity increased remarkably. 4) 3D analysis was more effective than 2D analysis in predicting the behavior of a rock masses with many joints with various orientations. (author)