System failure probability analysis of cohesive slope considering the spatial variability of undrained shear strength

Abstract

A system failure probability analysis method of cohesive slope considering the spatial variability of undrained shear strength is proposed. In this method, the local averaging parameter of the random field of undrained shear strength on slip surface is introduced as an equivalent parameter. The statistical characters of the equivalent parameter and the correlation coefficient between different equivalent parameters are formulated. Then, the reliability index of a single failure mode and the correlation coefficient between different failure modes are calculated based on the equivalent parameters. By considering both the reliability index and correlation coefficient between different failure modes, the representative slip surfaces are searched step by step, and the system failure probability is assessed using those representative slip surfaces. Finally, to verify this method, three slopes are analyzed as examples. The results show that the equivalent parameter obtained by local averaging along the circular slip surface is feasible to describe the spatial variability of the undrained shear strength, and the proposed method can assess the system failure probability of cohesive slope with small error. Meanwhile, the correlation coefficient between failure modes increases with the spatial correlation of the random field, thus the number of representative slip surfaces required to achieve convergence will reduce.

Keywords

• slope stability
• system failure probability
• spatial variability
• correlative failure mode
• equivalent parameter