Scholarlink

<p>In New Austrian Tunneling Method (NATM) construction, the determination of critical values for surrounding rock instability in complex geological environments is challenging, rendering efforts in data collection and information transmission ineffective. Based on numerous domestic tunnel-related specifications, this study employs data classification statistics and envelope regression analysis to establish an instability identification system that integrates cumulative displacement criteria and deformation rate criteria incorporating deformation acceleration, adhering to the principle of multi-dimensional identification indicators. Furthermore, the instability identification methods for the vault and arch waist are distinguished, enabling targeted and refined instability prevention and control. Additionally, this study optimizes Variational Mode Decomposition (VMD) using the Sparrow Search Algorithm (SSA) to denoise time-series monitoring data of surrounding rock displacement, thereby enhancing the learning efficiency and prediction accuracy of subsequent models. Moreover, chaotic mapping, mutation strategies, and dynamic perturbation terms are introduced to improve the Aquila Optimizer (APO), leading to the development of an SSA-VMD-TCN-BIGRU-IAPO time-series prediction model for surrounding rock deformation. This model demonstrates high prediction accuracy and, when combined with diversified surrounding rock instability criteria, provides multiple safeguards for safe and efficient tunnel construction, offering valuable insights for construction early warning and management in related engineering projects.</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

Criteria for Monitoring and Predicting Instability in NATM Tunnel Construction Under Complex Geology

  • 期刊论文

  • 发表日期：

    2025年6月27日

查看全文
  • Shijiang Pu,
  • Shunchuan Wu,
  • Xiaolong Yang,
  • Haiyong Cheng,
  • Tianbing Xiang,
  • Tao Wang

摘要

In New Austrian Tunneling Method (NATM) construction, the determination of critical values for surrounding rock instability in complex geological environments is challenging, rendering efforts in data collection and information transmission ineffective. Based on numerous domestic tunnel-related specifications, this study employs data classification statistics and envelope regression analysis to establish an instability identification system that integrates cumulative displacement criteria and deformation rate criteria incorporating deformation acceleration, adhering to the principle of multi-dimensional identification indicators. Furthermore, the instability identification methods for the vault and arch waist are distinguished, enabling targeted and refined instability prevention and control. Additionally, this study optimizes Variational Mode Decomposition (VMD) using the Sparrow Search Algorithm (SSA) to denoise time-series monitoring data of surrounding rock displacement, thereby enhancing the learning efficiency and prediction accuracy of subsequent models. Moreover, chaotic mapping, mutation strategies, and dynamic perturbation terms are introduced to improve the Aquila Optimizer (APO), leading to the development of an SSA-VMD-TCN-BIGRU-IAPO time-series prediction model for surrounding rock deformation. This model demonstrates high prediction accuracy and, when combined with diversified surrounding rock instability criteria, provides multiple safeguards for safe and efficient tunnel construction, offering valuable insights for construction early warning and management in related engineering projects.