
info@juzhikan.asia
广东酒店管理职业技术学院,广东东莞,523960;
摘要:针对电动自行车充电环境的火灾安全隐患,本文设计了一种基于多传感器融合的电动自行车充当环境智能火灾预警系统。该系统采用YOLOv8n深度学习模型实现了视觉火焰检测,在测试集上的火焰检测准确率达到98.1%。系统基于火灾场景多传感器模拟时序数据,采用规则引擎结合决策级加权融合的双决策机制,进行火灾风险分级评估。实验结果表明,系统面向多传感器数据的预警响应时长小于3秒、火灾预警检测准确率100%,综合风险分数位于0.423~0.566的区间,能够稳定触发早期预警。本研究为电动自行车充电环境火灾早期预警提供了可执行的技术方案。
关键词:多传感器融合;电动自行车火灾预警;YOLOv8n;决策级融合
参考文献
[1]Jin L, Gu J Y, Jiang S L, et al. Fire analysis and control countermeasures of electric bicycles[J]. Chemical Industry Times, 2024, 38(2): 40-42. DOI:10.16597/j.cnki.issn.1002-154x.2024.02.010.
[2]Deng B B, Xu Y B, Yao W H, et al. Research on fire risk prevention and control countermeasures of electric bicycles[J]. Labour Protection, 2023, (9): 84-86.
[3]吴海潮.多传感器信息融合的汽车火灾探测报警系统设计研究[J].内燃机与配件,2021,(21):212-213.DOI:10.19475/j.cnki.issn1674-957x.2021.21.101.
[4]孙继平,李月.基于双目视觉的矿井外因火灾感知与定位方法[J].工矿自动化,2021,47(06):12-16+78.DOI:10.13272/j.issn.1671-251x.17766.
[5]林水生,卫伯言,杨海芬,等.引入新数据源的D-S融合检测方法[J].电子科技大学学报,2021,50(06):861-867.
[6]Mahbub, Mobasshir , M. M. Hossain , and M. S. A. Gazi . "Cloud-Enabled IoT-based embedded system and software for intelligent indoor lighting, ventilation, early stage fire detection and prevention." Computer Networks 184 (15 January 2021)(2020):1-7.
[7]Baek, Jaeseung , et al. "Intelligent Multi-Sensor Detection System for Monitoring Indoor Building Fires." IEEE sensors journal 21-24(2021).
[8]Verma, Neetu , and D. Singh . "Analysis of cost-effective sensors: Data Fusion approach used for Forest Fire Application." Materials today: proceedings 24(2020):2283-2289.