Failure Mechanism and Factor of Safety for Spatially Variable Undrained Soil Slope

Liang Li,Xuesong Chu

doi:10.1155/2019/8575439

Abstract

This paper aims to investigate the differences in factor of safety (FS) and failure mechanism (FM) for spatially variable undrained soil slope between using finite element method (FEM) , finite difference method (FDM), and limit equilibrium method (LEM). The undrained shear strength of cohesive soil slope is modeled by a one-dimensional random field in the vertical direction. The FS and FM for a specific realization of random field are determined by SRT embedded in FEM- and FDM-based software (e.g., Phase26.0 and FLAC) and LEM, respectively. The comparative study has demonstrated that the bishop method (with circular failure surface) exhibits performance as fairly good as that of SRT both in FS and FM for the undrained slope cases where no preferable controlling surfaces such as hydraulic tension crack and inclined weak seams dominate the failure mechanism. It is, however, worthwhile to point out that unconservative FM is provided by the Bishop method from the aspect of failure consequence (i.e., the failure consequence indicated by the FM from the Bishop method is smaller than that from SRT). The rigorous LEM (e.g., M-P and Spencer method with noncircular failure surface) is not recommended in the stability analysis of spatially variable soil slopes before the local minima and failure to converge issues are fully addressed. The SRT in combination with FEM and/or FDM provides a rigorous and powerful tool and is highly preferable for slope reliability of spatially variable undrained slope.

Highlights

Both the failure probability and the respective failure consequence are involved in the landslide risk assessment [1,2,3,4,5].e shallow slide and deep slide as shown in Figure 1 clearly demonstrate the difference in failure consequence
E objective of this paper is to investigate the factor of safety (FS) and failure mechanism (FM) obtained from different numerical procedures of finite element method (FEM), finite difference method (FDM), and limit equilibrium method (LEM) as well under specific realizations of random field for undrained shear strength in cohesive soil slope. e comparison is expected to provide some insights into the applicability of abovementioned numerical procedures in reliability analysis for spatially variable undrained soil slope. e paper starts with
It is 0.999 for Figure 10(a) case, and it implies a trivial dispersion in FS between FEM and FDM. e similar trend and correlation coefficient r have been noticed from Figure 10(b) for the data pairs of FS between FEM and Bishop method

Summary

Introduction

Both the failure probability and the respective failure consequence are involved in the landslide risk assessment [1,2,3,4,5]. The extensive computation effort within the implementation of FEM and FDM in slope reliability analysis always enforces the geotechnical practitioners to the use of simplified LEM. It is, of practical interest and importance to know the difference in factor of safety (FS) and failure mechanism (FM) among the aforementioned three methods especially for the probabilistic slope stability problem taking into account the spatial variability of soil properties. The conclusions and discussion are given, and some insights into the applicability of FEM, FDM, and LEM in reliability analysis of spatially variable undrained soil slope are provided

Probabilistic Characteristics of Spatially Random Soil

Illustrative Example

Results

Discussion and Conclusions