Spatial Variability Modelling of Structural Surface Mechanical Parameters for Reliability Assessment of Rock Slopes

The spatial variability modelling of mechanical parameters for the highly fractured weathered rock masses and those dominated by a single structural plane has been extensively studied. However, few attempts have been made to characterize the spatial variability of mechanical parameters for the wedge rock masses. In this paper, a method for modelling the spatial variability of the structural surface mechanical parameters for the wedge rock masses is proposed. The function relationship between the factor of safety of the rock slope and uncertain input parameters is constructed using the back-propagation neural network. The non-intrusive stochastic finite difference method is adopted to explore the influence of the spatial variability of the structural surface mechanical parameters on the probability of slope failure. The proposed method can provide an effective and robust means for the characterization of the spatial variability of structural surface mechanical parameters in complex rock slope reliability analysis. Ignoring the spatial variability of the structural surface mechanical parameters can overestimate the probability of slope failure, resulting in conservative slope reinforcement design schemes.