Simulation of Fluid Flow in Fractured Rocks Based on the Discrete Fracture Network Model Optimized by Measured Information

Abstract

The fluid flow of Jijicao block in Beishan, one of the main candidate sites for a Chinese high-level radioactive waste (HLW) repository, was studied by a simulation on the basis of the discrete fracture network (DFN) model that was optimized by measured information with a coupled stochastic-deterministic approach. The spatial distribution of the fracture network within a certain domain matches that in reality based on the deterministic information obtained from field investigation. The presentation started with the generation of stochastic DFN models, and then the geometric properties (location, density, etc.) of fracture inside the models are modified to reflect the reality as objectively as possible. Further, the inherent hierarchy and the intersection relationship of fracture networks were adjusted in a locally coupled manner to make the model further optimized. Finally, to obtain the directional permeability of Jijicao block the calculation of fluid flow was performed within two-dimensional (2D) planes [three orthogonal planes in three-dimensional (3D) space] based on the optimized model. Results show that the size of the representative elementary volume (REV) ranges from 8 to 12 m in 2D, and for K1-K′1, K2-K′2, and K3-K′3 planes they are about 8, 8–1, and 1–12 m, respectively. In addition, the directional permeabilities are all of −13 orders of magnitude. Specifically, the convergence values of permeability in the K1-K′1 plane are 5.33 × 1−13 m2 (kx) and 6.41 × 1−13 m2 (ky), in the K2-K′2 plane the values are 2.57 × 1−13 m2 (ky) and 4.38 × 1−13 m2 (kz), and in the K3-K′3 plane the values are 5.77 × 1−13 m2 (kz) and 4.25 × 1−13 m2 (kx), respectively. The fluid flow simulations in the HLW repository provide an effective way to evaluate the long-term impact on the environment induced by radionuclide migration associated with regional fluid flow.