Scholarlink

<p>As underground infrastructure expands into increasingly complex geological environments, ensuring construction safety against unforeseen hazards has become a paramount engineering priority. In tunnel engineering, water-bearing geological anomalies represent the primary cause of severe water and mud inrush disasters. While advanced geological prediction using the direct current (DC) resistivity method is a crucial preventative measure, its accuracy is often severely compromised by electromagnetic interference from widely distributed steel arch supports. To address this, this study proposes a novel Bayesian inversion imaging approach. Based on 3D forward modeling, we quantitatively establish that steel arch-induced interference approximately follows a zero-mean Gaussian distribution. By embedding the likelihood function of this interference directly into the inversion objective function using a maximum a posteriori (MAP) framework, we eliminate the need for traditional preliminary denoising steps. Numerical simulations verify that the proposed method accurately reconstructs the geometry and location of low-resistivity anomalies at distances of 10&#xa0;m and 20&#xa0;m ahead of the face, significantly outperforming conventional scaling coefficient methods. Field application in the Xianglu Mountain No. 2 Tunnel of the Central Yunnan Water Diversion Project successfully identified potential water-bearing structures within 20&#xa0;m ahead of the tunnel face. The quantitative predictions were highly consistent with the actual excavation records, indicating that the proposed method can significantly improve the safety, efficiency, and reliability of resistivity-based advanced geological prediction.</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

Electrical resistivity tomography for water-bearing structures incorporating the interference distribution pattern of steel arch in tunnels

  • 期刊论文

  • 发表日期：

    2026年7月28日

查看全文
  • Lichao Nie,
  • Yu Zhou,
  • Zhicheng Song,
  • Zhi-Qiang Li,
  • Guoan Chu,
  • Shuo Zhang

摘要

As underground infrastructure expands into increasingly complex geological environments, ensuring construction safety against unforeseen hazards has become a paramount engineering priority. In tunnel engineering, water-bearing geological anomalies represent the primary cause of severe water and mud inrush disasters. While advanced geological prediction using the direct current (DC) resistivity method is a crucial preventative measure, its accuracy is often severely compromised by electromagnetic interference from widely distributed steel arch supports. To address this, this study proposes a novel Bayesian inversion imaging approach. Based on 3D forward modeling, we quantitatively establish that steel arch-induced interference approximately follows a zero-mean Gaussian distribution. By embedding the likelihood function of this interference directly into the inversion objective function using a maximum a posteriori (MAP) framework, we eliminate the need for traditional preliminary denoising steps. Numerical simulations verify that the proposed method accurately reconstructs the geometry and location of low-resistivity anomalies at distances of 10 m and 20 m ahead of the face, significantly outperforming conventional scaling coefficient methods. Field application in the Xianglu Mountain No. 2 Tunnel of the Central Yunnan Water Diversion Project successfully identified potential water-bearing structures within 20 m ahead of the tunnel face. The quantitative predictions were highly consistent with the actual excavation records, indicating that the proposed method can significantly improve the safety, efficiency, and reliability of resistivity-based advanced geological prediction.