Scholarlink

<p>Utilization of underground space is rapidly increasing in urban areas in order to compensate for the shortage of land. Underground space utilization in construction requires deep excavations, which are challenging in the highly populated urban areas. In such a scenario, micropiles are useful to stabilize the excavation. This study deals with the numerical analysis of a micropile-stabilized excavation system, followed by probabilistic reliability analysis. A three-dimensional (3D) finite element model is developed for the excavation stabilized with micropiles. Reliability analysis is performed by generating data for input variables using Latin Hypercube Sampling (LHS), and the same data is used to develop the functional relationships between input and output variables with the help of Multivariate Adaptive Regression Splines (MARS). The Monte Carlo simulation (MCS) technique is used to estimate the probability of failure and the reliability index of the excavation system. The sensitivity analysis is also performed to identify the dominant input variables of the system. The probability of failure of the system provided more quantitative information on stability compared to the factor of safety. Results of the present study are useful in understanding the behavior of excavation stabilized with micropile walls and developing a digital twin/Building Information Modelling (BIM) for the excavation system to monitor the performance during pre- and post-construction periods.</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

Reliability Analysis of Micropile-Supported Deep Excavations Using Multivariate Adaptive Regression Splines

  • 期刊论文

  • 发表日期：

    2026年1月16日

查看全文
  • K. Menaka,
  • Sougata Mukherjee,
  • G. L. Sivakumar Babu,
  • G. R. Dodagoudar

摘要

Utilization of underground space is rapidly increasing in urban areas in order to compensate for the shortage of land. Underground space utilization in construction requires deep excavations, which are challenging in the highly populated urban areas. In such a scenario, micropiles are useful to stabilize the excavation. This study deals with the numerical analysis of a micropile-stabilized excavation system, followed by probabilistic reliability analysis. A three-dimensional (3D) finite element model is developed for the excavation stabilized with micropiles. Reliability analysis is performed by generating data for input variables using Latin Hypercube Sampling (LHS), and the same data is used to develop the functional relationships between input and output variables with the help of Multivariate Adaptive Regression Splines (MARS). The Monte Carlo simulation (MCS) technique is used to estimate the probability of failure and the reliability index of the excavation system. The sensitivity analysis is also performed to identify the dominant input variables of the system. The probability of failure of the system provided more quantitative information on stability compared to the factor of safety. Results of the present study are useful in understanding the behavior of excavation stabilized with micropile walls and developing a digital twin/Building Information Modelling (BIM) for the excavation system to monitor the performance during pre- and post-construction periods.