Statistical analysis on the mechanical behaviour of non-persistent jointed rock masses using combined DEM and DFN

Due to the stochastic nature and random distribution of joints in the rock mass, statistical analyses can be employed as a scientific tool to improve our understanding regarding the complex behaviour of the jointed rock masses. This paper aims to investigate the effect of geometrical parameters of non-persistent joints on the mechanical behaviour of the rock masses under uniaxial loading condition. First, a validation study is carried out to compare the results of numerical discrete element models against physical experiments. Then, the effect of joint random geometrical parameters including joint orientation, joint length and joint density is investigated using the discrete fracture network model. In addition to conventional parametric analyses, experimental design methods of Taguchi orthogonal array and central composite design (CCD) are employed to statistically determine the significance of joint geometrical parameters on the strength and the deformation modulus. Results of this study show that the random nature of joints has a greater effect on the strength than the deformation modulus. Amongst the joint geometrical parameters, the joint orientation is the most effective parameter. The results also show that the increase in the joint density and the joint length results in a decrease in strength and deformation modulus. The combined statistical analysis using the CCD technique shows that the maximum strength occurs at the orientation angle of 0° while the minimum strength is expected to occur at the orientation angle in the range of 45°–60°.