Scholarlink

<p>Clogging of fine-grained soils remains a major source of uncertainty in earth pressure balance (EPB) tunnelling, often causing operational disruptions and increasing project costs. Although empirical charts provide useful first-order indicators of clogging potential, they do not account for the magnitude or uncertainty of soil–steel interaction forces governing adhesion-driven clogging mechanisms. This study introduces a probabilistic framework for quantifying clogging potential by integrating laboratory pull-out test–soil state correlations with empirical clogging susceptibility criteria. Plate pull-out tests were conducted on London, Malnome, and Viterbo clays, extending existing datasets and capturing variability in both soil state and measured interaction forces. Empirical correlations between pull-out force and consistency index were reformulated within a Bayesian framework using a Weibull likelihood, enabling the explicit treatment of aleatory variability and epistemic uncertainty while enforcing physically plausible force predictions. The probabilistic pull-out force model is then coupled with established clogging susceptibility charts to define a joint state–force exceedance contribution, characterizing how soil state occurrence, mechanical susceptibility, and force exceedance interact to become critical for clogging. The influence of extraction velocity on pull-out resistance was also systematically investigated over a range of 0.5–9&#xa0;mm/min. Results reveal a distinct non-monotonic strain-rate dependency, with minimum pull-out resistance observed near 1&#xa0;mm/min, challenging the commonly held assumption that the choice of standard test rate of 5&#xa0;mm/min does not necessarily have significant implications for the results produced. A parametric strain-rate normalization model is proposed to harmonize pull-out forces obtained at different extraction velocities to a common reference rate. The proposed framework advances the interpretation of laboratory adhesion measurements by explicitly incorporating uncertainty, soil state variability, and strain-rate effects, thereby improving the robustness and transferability of clogging assessments for EPB tunnelling in cohesive soils.</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

A probabilistic framework for clogging assessment in EPB tunnelling

  • 期刊论文

  • 发表日期：

    2026年3月3日

查看全文
  • Sara Mangifesta,
  • Stefano Collico,
  • Giovanni Spagnoli,
  • Diego Sebastiani

摘要

Clogging of fine-grained soils remains a major source of uncertainty in earth pressure balance (EPB) tunnelling, often causing operational disruptions and increasing project costs. Although empirical charts provide useful first-order indicators of clogging potential, they do not account for the magnitude or uncertainty of soil–steel interaction forces governing adhesion-driven clogging mechanisms. This study introduces a probabilistic framework for quantifying clogging potential by integrating laboratory pull-out test–soil state correlations with empirical clogging susceptibility criteria. Plate pull-out tests were conducted on London, Malnome, and Viterbo clays, extending existing datasets and capturing variability in both soil state and measured interaction forces. Empirical correlations between pull-out force and consistency index were reformulated within a Bayesian framework using a Weibull likelihood, enabling the explicit treatment of aleatory variability and epistemic uncertainty while enforcing physically plausible force predictions. The probabilistic pull-out force model is then coupled with established clogging susceptibility charts to define a joint state–force exceedance contribution, characterizing how soil state occurrence, mechanical susceptibility, and force exceedance interact to become critical for clogging. The influence of extraction velocity on pull-out resistance was also systematically investigated over a range of 0.5–9 mm/min. Results reveal a distinct non-monotonic strain-rate dependency, with minimum pull-out resistance observed near 1 mm/min, challenging the commonly held assumption that the choice of standard test rate of 5 mm/min does not necessarily have significant implications for the results produced. A parametric strain-rate normalization model is proposed to harmonize pull-out forces obtained at different extraction velocities to a common reference rate. The proposed framework advances the interpretation of laboratory adhesion measurements by explicitly incorporating uncertainty, soil state variability, and strain-rate effects, thereby improving the robustness and transferability of clogging assessments for EPB tunnelling in cohesive soils.