Yan, Xiao (China University of Mining and Technology) | Ma, Guowei (Hebei University of Technology) | Sun, Zizheng (Hebei University of Technology) | Li, Tuo (The University of Western Australia) | Chen, Yun (The University of Western Australia) | Song, Hongwei (China University of Mining and Technology)
ABSTRACT: The rock masses generally contains many discrete fracture networks, which play as channels for storage and migration of underground water, causing water inrush, water burst or collapse during tunnel excavation. Grouting in fractured rock masses has been regarded as an effective method to minimize groundwater flow and keep tunnel stability in tunnel engineering and mining projects. In this paper, an attempt has been made to propose numerical models to predict grout flow and penetration length in a 3D domain with fractures under constant injecting pressures using the Unified Pipe Network Method (UPNM). The effects of rock mass properties as joint aperture, roughness, dip and dip angle are considered. Barton-Bandies joint model is used to assign joint roughness (JRC) in the modeling. JRC are generated by a gaussian distribution model and the apertures of rough fracture surface are generated by a log-normal distribution model with different mean value and variance. Furthermore, the grout processes in static water and dynamic water are compared. The effect of JRC on the grout penetration length is large and the water flow will drive the grout flow.