Scholarlink

<p>The complexity of a building's indoor environment exacerbates its fire safety issues, making them particularly problematic. Traditional fire protection systems only detect fires when their characteristics are already visible, and fire recognition delays have become a significant problem. This paper presents a digital twin framework for managing building fire safety in real time. Sensor data and a Bert_C model are used to analyze the transition of a fire from smoldering to flaming combustion. The results show that the model achieved 96.1% accuracy and 96.2% F1 score, and the framework significantly improves the timeliness and accuracy of fire warnings and visualizes 3D fire scenarios, providing powerful technical support for firefighting and emergency rescue operations. This study confirms the feasibility of real-time fire safety management using 3D environments and digital twins, providing an innovative solution for fire safety management in the built environment.</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

Digital twin-enabled deep learning for real-time fire situation awareness

  • 期刊论文

  • 发表日期：

    2025年9月8日

查看全文
  • Tinglong Tang,
  • Chunli Zhao,
  • XinQiong Liu,
  • ShuiFa Sun

摘要

The complexity of a building's indoor environment exacerbates its fire safety issues, making them particularly problematic. Traditional fire protection systems only detect fires when their characteristics are already visible, and fire recognition delays have become a significant problem. This paper presents a digital twin framework for managing building fire safety in real time. Sensor data and a Bert_C model are used to analyze the transition of a fire from smoldering to flaming combustion. The results show that the model achieved 96.1% accuracy and 96.2% F1 score, and the framework significantly improves the timeliness and accuracy of fire warnings and visualizes 3D fire scenarios, providing powerful technical support for firefighting and emergency rescue operations. This study confirms the feasibility of real-time fire safety management using 3D environments and digital twins, providing an innovative solution for fire safety management in the built environment.