| dc.description.abstract | According to Burland (1987), the uncertainty of geotechnical analysis can be divided into three types of models: Geologic model, Ground model and Geotechnical model. The geologic model includes various geological characteristics of the study area, such as dip angle, groundwater and interface of stratum, etc. The ground model includes engineering parameters that are based on the geologic model, such as cohesion and friction angle. The geotechnical model is a tool for predicting and analyzing by adding design parameters to the ground model, such as the limit equilibrium method and numerical analysis models. This study takes the uncertainty of the geological model as the main discussion aspect and takes the sources of different dip angles and their critical groundwater pressure as the basis for the uncertainty. It is assumed that there are three types of failure modes: rail pile erosion, stability rail pile and in front of the rail pile with tension crack. The failure modes and the analysis scenarios of four combinations of dip angles and groundwater pressure uncertainties are analyzed using the limit equilibrium method and Monte Carlo simulation. Finally, the influence of failure probability on slope stability is discussed. The analysis result is that the failure probability in the dry state is inversely proportional to the precision of the standard deviation of the dip angles. The influence of different failure modes on the failure probability is that the failure probability of the mode in front of the rail pile with tension crack is the largest, which is consistent with the situation observed in the field investigation. The results under the condition of considering the critical groundwater pressure, if do not consider the difference in dip angle precision, the overall analysis results show that the sensitivity of dip angle variability to failure probability is higher than groundwater pressure variability. In this study, a three-dimensional numerical model is also used to explore the failure conditions of the study area under extreme conditions of changing the different combinations of discontinuity cohesion and friction angle. The results show that the overall study area is fairly stable under non-extreme conditions. | en_US |