An Improved Rock Mass Characterization Method Using a Quantified Geological Strength Index and Synthetic Rock Mass Model
Yabing Zhang,F. Y. Ren,T. H. Yang,S. Y. Wang,W. F. Zhang,M. X. Yu
DOI: https://doi.org/10.1007/s00603-018-1532-z
IF: 6.518
2018-01-01
Rock Mechanics and Rock Engineering
Abstract:In this paper, an improved rock mass characterization method is presented. This method used a quantified geological strength index (GSI) and synthetic rock mass (SRM) model based on a discrete fracture network (DFN) and bonded particle model. A quantitative method is initially introduced to assist in the use of the conventional GSI chart, employing the rock block volume as the quantitative factor. The block volume for a given DFN model is calculated by a volumetric joint count, which is defined as the number of fractures intersecting a cube with unit length. The peak strength and deformation modulus of the rock mass are estimated based on the derived GSI, and then selected as references to constrain mechanical parameters of joints in the SRM model. A parametric study shows a positive correlation between the joint stiffness/friction coefficient and both the peak strength and deformation modulus. The joint friction coefficients of 0.5, 1.0, and 1.5 are suggested to correspond to the surface quality of ‘FAIR’, ‘GOOD’, and ‘VERY GOOD’ in the GSI chart. An advantage of the SRM over the GSI lies in the characterization of the post-peak strain softening behavior. The numerical results show that lower joint stiffness and friction coefficient result in a more ductile post-peak behavior. Additional SRM models with varied joint distributions show the complexities of rock mass characterization, where the rock mass strength, modulus, and post-peak strain softening rate may vary significantly with only one aspect of distributions of the joints, such as intensity or diameter. Triaxial testing is also conducted, and the SRM peak strengths are compared with the Hoek–Brown strength criterion based on the GSI system. The strengths of the regular triaxial models show general convergence with the Hoek–Brown criterion, but the strengths of the true triaxial models are less than those of the regular triaxial models due to deviatoric confining stresses. Meanwhile, ductile post-peak behavior is generally simulated in the triaxial testing. Cracking models are finally investigated and presented, behaving as additional information to explain the changes in the rock mass strength and deformability with different confining stresses. The method presented in this paper is meant to help geologists and engineers to quantify the GSI for jointed rock mass, and characterize the rock mass modulus and strength. Meanwhile, the synthetic rock mass method provides an alternative for rock mass characterization in addition to the traditional GSI method.