Probabilistic stability analyses of slopes using the ANN-based response surface

摘要: Abstract Slope stability analysis is a geotechnical engineering problem characterized by many sources of uncertainty. Some these are connected to the uncertainties soil properties involved in analysis. In this paper, numerical procedure for integrating commercial finite difference method into probabilistic slope presented. Given that limit state function cannot be expressed an explicit form, artificial neural network (ANN)-based response surface adopted approximate function, thereby reducing number calculations. A trained ANN model used calculate probability failure through first- and second-order reliability methods Monte Carlo simulation technique. Probabilistic assessments hypothetical two-layer as well Cannon Dam Missouri, USA performed verify application potential proposed method.