Stability evaluations of three-layered soil slopes based on extreme learning neural network

Abstract

ABSTRACT Slope stability analysis is one of the most critical and important topics for geotechnical engineers and thus various stability charts have been developed in order to provide quick first assessments of slope stability. However, general chart solutions are often found to be insufficient for purpose when variables to be considered increase. Motivated by that, this study aims to adopt an artificial neural network (ANN) with the extreme learning machine algorithm to develop a convenient and efficient tool for assessing three-layered soil slope stability. The neural network is trained using the obtained solutions from the finite element upper and lower bound limit analysis methods. The proposed tool in this study is capable of providing a quick first assessment of three-layered soil slope stability. The results showed that the slope stability estimation using the proposed ANN-based tool can still maintain good accuracy and convenience.