Geometric Modeling of Joints in 3D Discrete Fracture Network (Case Study: Liroo Dam construction site)

Nowadays, modeling of rock masses is widely used in order to accurately determine the properties of the rock mass and improve its behavior. In most cases, the geometric components of the fractures in the rock have a random nature.one of the most important methods to simulate the random nature of joints is the stochastic modeling of the fracture network of rock. The stochastic modeling of the fracture network makes the heterogeneous nature of rock mass with discrete elements, with geometrical properties and features that are statistically defined. In this study, the samples taken from the Liroo dam foundation are surveyed and analyzed. Then using the EasyFit that is a statistical software, the known distribution functions were determined by the maximum fitting on the surveyed geometric components and then their moments were determined. On the other hand, using the written code with Mathematica software, 3D-DFN three dimensional discrete fracture network method fractures simulated. Finally, the geometric model validated based on the comparison of the statistics of the distribution functions obtained from the model output with statistics of input distribution. The results in most cases indicate compliance of more than 90%. Also, P10, P21 and P32 values of the two-dimensional sections in a same strike with reality and in three-dimensional model were determined and compared with real values. The values of P10 and P32 showed more than 90% compliance and the P21 component was more than 75% consistent with actual data.