A novel dimension reduction-based metamodel approach for efficient slope reliability analysis considering soil spatial variability

Xian Liu,Yadong Liu,Zhiyong Yang,Xueyou Li

DOI: https://doi.org/10.1016/j.compgeo.2024.106423

IF: 5.218

2024-05-18

Computers and Geotechnics

Abstract:The surrogate models become increasingly used in slope reliability problems, in which the expensive-to-evaluate slope deterministic analysis model is replaced with the cheap-to-evaluate surrogate model. However, the surrogate models are frequently criticized in spatially variable soil slopes where a large number of random variables are involved and establishment of the surrogate model requires extensive evaluations of the slope deterministic model. To address this issue, this paper proposed a novel dimension reduction-based metamodel approach for efficient slope reliability analysis, abbreviated as SBR-SRS-MGPR because it contains three steps: (a) using the novel slice method-based dimension reduction (SBR) to reduce the input variables' dimension; (b) using the strength reduction sampling (SRS) technique to expand the training samples; and (c) training the multiple Gaussian process regression (MGPR) model to approximate the slope deterministic model. The Monte Carlo simulation (MCS) is then used to estimate the probability of failure ( p f ) of slope system based on the well-constructed MGPR model. Two slope examples are employed to demonstrate the performance of the proposed method. The results show that the proposed method outperforms the previous surrogate model-based methods in terms of computational efficiency with fewer number of slope stability analysis and simultaneously gives smaller uncertainty of the estimated p f .

computer science, interdisciplinary applications,engineering, geological,geosciences, multidisciplinary