主管：中华人民共和国应急管理部
主办：应急管理部天津消防研究所
ISSN 1009-0029  CN 12-1311/TU
消防科学与技术

消防科学与技术 ›› 2025, Vol. 44 ›› Issue (10): 1540-1546.

• • 上一篇    下一篇

数字孪生电气火灾早期智能监测系统设计与应用

皮芳思1, 吴建彬1, 田泽2   

  1. （1.上海意静信息科技有限公司，上海 200120； 2.北京科技大学 资源与安全工程学院，北京 100083）
  • 收稿日期:2025-06-06 修回日期:2025-08-27 出版日期:2025-10-15 发布日期:2025-10-15
  • 作者简介:皮芳思，上海意静信息科技有限公司全栈开发工程师，主要从事智慧消防安全监测报警平台方面的研究，上海市自由贸易试验区郭守敬路351号2号楼A641-21室，200120， pifangsi@firedata.cn。
  • 基金资助:
    国家重点研发计划项目（2023YFC3009805）

Digital twin electrical fire intelligent early-warning monitoring system design and application

Pi Fangsi1, Wu Jianbin1, Tian Ze2   

  1. (1. Shanghai Aifire Information Technology Co., Ltd., Shanghai 200120, China; 2. School of Civil and Resource Engineering, University of Science and Technology Beijing, Beijing 100083, China)
  • Received:2025-06-06 Revised:2025-08-27 Online:2025-10-15 Published:2025-10-15

摘要： 针对电气火灾隐蔽性强、预警不及时等问题，设计了一种融合数字孪生技术的电气火灾早期智能监测系统。系统构建在“云-边-端”一体化架构的基础上，深入融合数字孪生五维模型理论，集成多维感知、人工智能与大数据分析等关键技术，实现从数据采集、边缘分析到云端融合的全流程智能化。通过前端设备采集电气系统的多维运行数据，边缘计算模块实现实时处理与预警响应，云端平台整合配电系统的数字模型、孪生数据、电气绝缘老化预测模型及风险评估模型，全面展现电气系统的运行态势与潜在风险，并通过“风险一张图”的形式直观地呈现火灾风险分布，实现三级配电系统数字孪生模型的精准刻画，增强了电气火灾的主动感知与智能防控能力，为电气系统提供全方位安全保障。

关键词: 电气火灾, 数字孪生, 预警系统, 五维模型, 火灾预防

Abstract: To address the issues of strong concealment and lack of timely warning for electrical fires, an intelligent early monitoring system for electrical fires integrating digital twin technology has been designed. The system is built upon a "cloud-edge-end" integrated architecture, and deeply incorporates the five-dimensional digital twin model theory. It integrates key technologies such as multi-dimensional sensing, artificial intelligence, and big data analytics to achieve full-process intelligence from data acquisition and edge analysis to cloud fusion. Front-end devices collect multi-dimensional operational data from the electrical system. The edge computing module enables real-time processing and early warning responses. The cloud platform integrates the digital model of the power distribution system, twin data, an electrical insulation aging prediction model, and a risk assessment model, comprehensively displaying the operational status and potential risks of the electrical system. Fire risk distribution is intuitively presented through a "unified risk map", achieving precise characterization of the digital twin model for a three-level power distribution system. This approach enhances proactive perception and intelligent prevention and control capabilities for electrical safety, providing comprehensive safety assurance for electrical systems.

Key words: electrical fire, digital twin, early-warning system, five-dimensional model, fire prevention