Scholarlink

<p>This review explores the integration of deep learning in tunnel and underground engineering, focusing on areas like parameter inversion, machinery optimization, risk assessment, and safety monitoring. The study highlights how deep learning has advanced tunnel parameters' prediction and optimization, improving construction machinery's accuracy and efficiency. Additionally, integrating multi-source data has enabled innovative risk control solutions and dynamic construction management. Advanced image processing and computer vision, combined with deep learning, have enhanced real-time defect detection, significantly improving safety and maintenance. Despite advancements, challenges remain in deep learning applications in tunnel engineering, particularly in data sharing, the absence of standardized metrics, and real-time data processing in complex geological conditions. Looking forward, future research should focus on refining deep learning algorithms for more accurate and robust predictions, improving system integration, and developing intelligent disaster prevention systems for tunnel. Ultimately, deep learning is poised to be a cornerstone in the evolution of smart, sustainable, and resilient tunnel construction practices, aligning with global objectives for intelligent infrastructure development.</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 comprehensive review on advancements and challenges of deep learning in tunnel and underground engineering

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年4月16日

查看全文
  • Chunlei Zhang,
  • Md Habibullah,
  • Jingke Wu,
  • Mahabub Hasan Sajib,
  • Arifuggaman Arif,
  • Mohammed Alip,
  • Boyina Manohar,
  • Golam Mostakim Shikhon,
  • Mingjie Feng,
  • Ye Zhang,
  • Md Sohel Rana

摘要

This review explores the integration of deep learning in tunnel and underground engineering, focusing on areas like parameter inversion, machinery optimization, risk assessment, and safety monitoring. The study highlights how deep learning has advanced tunnel parameters' prediction and optimization, improving construction machinery's accuracy and efficiency. Additionally, integrating multi-source data has enabled innovative risk control solutions and dynamic construction management. Advanced image processing and computer vision, combined with deep learning, have enhanced real-time defect detection, significantly improving safety and maintenance. Despite advancements, challenges remain in deep learning applications in tunnel engineering, particularly in data sharing, the absence of standardized metrics, and real-time data processing in complex geological conditions. Looking forward, future research should focus on refining deep learning algorithms for more accurate and robust predictions, improving system integration, and developing intelligent disaster prevention systems for tunnel. Ultimately, deep learning is poised to be a cornerstone in the evolution of smart, sustainable, and resilient tunnel construction practices, aligning with global objectives for intelligent infrastructure development.