摘要
从城市轨道交通关键设备预警状态出发,以城市轨道交通桥梁设备为例,依据欧氏距离对同质传感器数据融合,分析了桥梁各监测传感器4种预警指标,建立桥梁预警状态模型,并在此基础上,提出了一种改进的D-S证据理论方法,对桥梁异质传感器进行数据融合,实现桥梁的综合监测预警.通过仿真测试,验证了该方法的可行性,同时将改进后的融合算法与原D-S算法作比较,结果表明,利用传统D-S证据理论融合数据的预警概率是0.6824,改进后的算法预警概率为0.8386.这表明改进后的方法具有更好的性能,可提升融合的可信度,并得到更为合理的预警结果,可为城市轨道交通其他设备预警提供借鉴.
Starting from the early warning state of key equipment in urban rail transit,taking urban rail transit bridge equipment as an example,this paper first fuses the homogeneous sensor data based on Euclidean distance.Four early warning indicators of each monitoring sensor of the bridge are then analyzed.A bridge early warning state model is established.On this basis,an improved D-S evidence theory method is proposed to fuse bridge heterogeneous sensors to realize comprehensive monitoring and early warning of bridges.Through the simulation test,the feasibility of the method is verified.At the same time,the improved fusion algorithm is compared with the original D-S combination rule.The results show that the early warning probability of data fusion using the traditional D-S evidence theory is 0.6824,and the early warning probability of the improved algorithm is 0.8386.This shows that the improved method has better performance,can improve the credibility of the fusion,and obtain more reasonable early warning results,which can provide reference for early warning of other equipment in urban rail transit.
作者
郑胜洁
刘留
胡祖翰
石先明
刘利平
徐余明
ZHENG Shengjie;LIU Liu;HU Zuhan;SHI Xianming;LIU Liping;XU Yuming(Electronic Information Engineering,Beijing Jiaotong University,Beijing 100044,China;China Railway Siyuan Survey and Design Group Co.,Ltd.,Wuhan 430063,China)
出处
《交通工程》
2022年第6期28-35,41,共9页
Journal of Transportation Engineering
基金
中铁第四勘察设计院集团有限公司重点研究项目:城市轨道交通智能运维管理系统方案研究(项目编号:2020K175)。
关键词
城市轨道交通
预警
数据融合
D-S证据理论
人工蜂群算法
urban rail transit
early warning
data fusion
D-S evidence theory
artificial bee colony algorithm
