Scholarlink

<p>This study proposes an innovative predictive model for ground deformation induced by shield tunneling. The model is developed using an analytical solution based on the semi-infinite saturated soil under concentrated force–initial solution framework and incorporates Karhunen–Loève expansion to construct a stochastic field of soil parameters. Sobol global sensitivity analysis is employed to evaluate the relative importance of input variables, while coupled sparse polynomial chaos expansion (SPCE) and the Monte Carlo simulation approach is adopted to predict ground deformation during tunneling. Comparison with field monitoring data reveals that the model effectively captures the spatial evolution of surface deformation, the effect of soil parameter variability, and the probability of engineering failure. The key findings are as follows: (1) Surface deformation exhibits a distinct spatial distribution, with maximum heave and settlement occurring approximately 3Rs ahead and 8Rs behind the tunnel face, respectively. Theoretical predictions align well with observed data. (2) Sobol sensitivity analysis identifies soil elastic modulus as the most critical parameter that influences prediction accuracy, while Poisson’s ratio plays a lesser role. (3) The SPCE method demonstrates excellent predictive accuracy across varying conditions, particularly when the variability of input parameters is low. (4) An increase in the coefficient of variation of soil parameters leads to greater dispersion in settlement and significantly raises the likelihood of exceeding design thresholds. Overall, the proposed approach provides a robust theoretical foundation for risk assessment and safety management in shield tunneling operations.</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

Probabilistic Prediction of Ground Deformation Induced by Shield Tunneling

  • 期刊论文

  • 发表日期：

    2026年1月6日

查看全文
  • Chen-Yang He,
  • Zhi Ding,
  • Feng Huang

摘要

This study proposes an innovative predictive model for ground deformation induced by shield tunneling. The model is developed using an analytical solution based on the semi-infinite saturated soil under concentrated force–initial solution framework and incorporates Karhunen–Loève expansion to construct a stochastic field of soil parameters. Sobol global sensitivity analysis is employed to evaluate the relative importance of input variables, while coupled sparse polynomial chaos expansion (SPCE) and the Monte Carlo simulation approach is adopted to predict ground deformation during tunneling. Comparison with field monitoring data reveals that the model effectively captures the spatial evolution of surface deformation, the effect of soil parameter variability, and the probability of engineering failure. The key findings are as follows: (1) Surface deformation exhibits a distinct spatial distribution, with maximum heave and settlement occurring approximately 3Rs ahead and 8Rs behind the tunnel face, respectively. Theoretical predictions align well with observed data. (2) Sobol sensitivity analysis identifies soil elastic modulus as the most critical parameter that influences prediction accuracy, while Poisson’s ratio plays a lesser role. (3) The SPCE method demonstrates excellent predictive accuracy across varying conditions, particularly when the variability of input parameters is low. (4) An increase in the coefficient of variation of soil parameters leads to greater dispersion in settlement and significantly raises the likelihood of exceeding design thresholds. Overall, the proposed approach provides a robust theoretical foundation for risk assessment and safety management in shield tunneling operations.