Scholarlink

<p>Estimating landslide risks is the premise of hazard prevention and reinforcement design. When the safety factors of different failure modes are close, but the failure risks are significantly different, the safety factor (<i>F</i><sub>s</sub>) can’t reflect the failure risk. Based on the generalized subset simulation (GSS) algorithm and the random finite difference method (RFDM), an approach was proposed to estimate the failure risk of slopes with multiple failure modes. The feasibility of the proposed method was demonstrated with two representative slopes, where the calculated failure risk is close to the MCS and subset simulation (SS) results. Further, the failure probabilities (<i>p</i><sub>f</sub>) of slope are compared when considering and not considering the spatial variability. Based on a double-layer clay slope, the computational efficiency of the GSS with or without solving safety factors is compared, and the effects of the vertical and horizontal autocorrelation distances (ACDs), autocorrelation functions (ACFs) on the <i>p</i><sub>f</sub> are discussed. The results show that the failure risk can be calculated based on the proposed method quickly; the failure risks of different modes are markedly diverse; the failure probability and the proportion of shallow sliding failure decrease with the vertical and horizontal ACDs; the failure probability corresponding to the simple exponential ACF is the smallest, comparing to those of the square exponential, second-order autoregressive and cosine exponential ACFs. This study could provide theoretical support for risk prevention and mitigation for slopes with multiple potential failure modes.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Risk estimation for earth slopes with multiple failure modes based on the generalized subset simulation

  • 期刊论文

  • 发表日期：

    2025年9月22日

查看全文
  • Zhongxin Zhang,
  • Shunchuan Wu,
  • Longqiang Han,
  • Pengbin He,
  • Min Ren

摘要

Estimating landslide risks is the premise of hazard prevention and reinforcement design. When the safety factors of different failure modes are close, but the failure risks are significantly different, the safety factor (Fs) can’t reflect the failure risk. Based on the generalized subset simulation (GSS) algorithm and the random finite difference method (RFDM), an approach was proposed to estimate the failure risk of slopes with multiple failure modes. The feasibility of the proposed method was demonstrated with two representative slopes, where the calculated failure risk is close to the MCS and subset simulation (SS) results. Further, the failure probabilities (pf) of slope are compared when considering and not considering the spatial variability. Based on a double-layer clay slope, the computational efficiency of the GSS with or without solving safety factors is compared, and the effects of the vertical and horizontal autocorrelation distances (ACDs), autocorrelation functions (ACFs) on the pf are discussed. The results show that the failure risk can be calculated based on the proposed method quickly; the failure risks of different modes are markedly diverse; the failure probability and the proportion of shallow sliding failure decrease with the vertical and horizontal ACDs; the failure probability corresponding to the simple exponential ACF is the smallest, comparing to those of the square exponential, second-order autoregressive and cosine exponential ACFs. This study could provide theoretical support for risk prevention and mitigation for slopes with multiple potential failure modes.