基于分布式光纤振动传感技术的地下管廊入侵监测系统  

Intrusion Monitoring System for Utility Tunnel Based on Distributed Optical Fiber Vibration Sensing Technology

作　　者：包恬 李楚瑞[1] 邱旭 黄祖煜 BAO Tian;LI Churui;QIU Xu;HUANG Zuyu(Department of Materials Science,Fudan University,Shanghai 200433,China;Zhuhai Da Hengqin Utility Tunnel Management Co.,Ltd.,Zhuhai Guangdong 519000,China)

机构地区：[1]复旦大学材料科学系,上海200433 [2]珠海大横琴城市综合管廊运营管理有限公司,广东珠海519000

出　　处：《传感技术学报》2023年第12期1974-1980,共7页Chinese Journal of Sensors and Actuators

摘　　要：针对目前地下综合管廊的安全防控需求,设计研制了一种基于分布式光纤振动传感技术的地下管廊入侵监测系统。系统采用分布式光纤马赫曾德尔-萨格纳克混合干涉结构,结合振动传递装置将井盖开关行为转化成对光缆的振动冲击,再利用一维卷积神经网络算法对振动冲击信号做模式识别,通过监测井盖开关实现对地下管廊人为入侵行为的监测。对所设计的系统进行实地测试,采集1573组扰动数据进行模式识别,模式识别正确率在99%以上,识别实时性高,表明该系统能够满足地下综合管廊实际应用的需求,有利于拓展光纤传感技术在地下管廊领域的应用。In view of the security requirements of the utility tunnel,an intrusion monitoring system for utility tunnel based on distributed optical fiber vibration sensing technology is developed.The system adopts the distributed optical fiber Mach Zehnder-Sagnac interference structure,utilizes a vibration transmission device to convert the switch of the manhole cover into the vibration impact on the optical cable,uses the one-dimensional convolutional neural network algorithm to recognize the pattern of impact signals,so the effect of monitoring the intrusion behavior of the utility tunnel is realized by monitoring the switch of the manhole cover.The designed system is tested on the spot and 1573 groups of disturbance data are collected for pattern recognition.The correct rate of pattern recognition is more than 99%,and the recognition real-time is high.It shows that the system can meet the actual application requirements of the utility tunnel,and is conducive to expanding the application of optical fiber sensing technology in the field of underground pipe gallery.

关 键 词：光纤传感 入侵监测系统 模式识别 卷积神经网络 地下综合管廊 

分 类 号：TP212.9[自动化与计算机技术—检测技术与自动化装置]