Influence of Spatial Variability on Slope Failure Analysis by the Shear Strength Reduction Technique

Abstract eng:
In this paper, two-dimensional (2D) slope stability failure using random field analysis is considered using the shear strength reduction technique for several cases. The spatial variability is simulated using a self-complied MATLAB subroutine, in which both cohesion c and friction angle Ø are assumed to be normally distributed and correlated. Thus the analysis considered is in fact a 2-D correlated spatial variability problem. Present publications mainly use the predetermined spatial correlation length L, the dimensionless ratio of correlation length to the slope height L/H or the Markov chain approach to describe the tendency of elements spatially near each other to be correlated. In this paper, the randomness of the input model variables can be realized by the shape of the normal distribution curves and the fluctuations of c and Ø values among different elements are presumed to follow the normal distribution. Comparison between the probability of failure obtained by First-Order Reliability Method (FORM) and the proposed random field analysis by Monte Carlo simulation with many realizations of soil properties can provide better insights into the role of variability in design. The influences from the number of model elements, the coefficient of correlation between c and Ø and the number of realizations are also investigated. The results also indicate that consideration of spatial variability of input soil parameters influence the assessment of the global factor of safety of the soil slope.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket