多源数据融合的滑坡监测预警评估  

Landslide monitoring and early warning evaluation based on multi-source data fusion

作　　者：王霄[1] 张译 邵健 黄玲 胡娟 WANG Xiao;ZHANG Yi;SHAO Jian;HUANG Ling;HU Juan(College of Electrical Engineering,Guizhou University,Guiyang 550025,China)

机构地区：[1]贵州大学电气工程学院,贵州贵阳550025

出　　处：《现代电子技术》2022年第10期124-130,共7页Modern Electronics Technique

基　　金：国家自然科学基金项目(61861007);国家自然科学基金项目(61640014);贵州省工业攻关项目(黔科合支撑[2019]2152);黔科合人才团队((2015)4014);物联网理论与应用案例库(KCALK201708);自动化专业卓越工程师计划(ZYS 2015004)。

摘　　要：针对现有监控山体滑坡设备价格高昂、采集数据单一化,单个传感器数据存在不确定性,同时难以将滑坡监测区域的空间预测和时间预测预报信息多维度融合等问题,文中采用两级并联型融合方式,通过GIS技术、传感器技术、无线和通信技术采集滑坡监测区域的空间与时间信息,并将时间信息实时远程传送至Tlink物联网云服务器。首先利用BP神经网络对获取的数据进行第一级融合,以得到同类数据的最佳值;然后,使用多层模糊综合评价法进行决策层融合,第一层采用AHP确定权重,第二层采用灾害熵确定权重,以实现预警评估的自适应能力并保护滑坡因子在模糊运算后信息不被掩盖。运行结果表明,通过多维信息能够获取更加丰富有效的监测区域信息,消除单个滑坡条件因子对滑坡等级评价的局限性,并提升整体滑坡等级评价的可靠性和准确性。In allusion to the problems of the existing landslide monitoring equipment,such as high price,acquisition data simplification,the data of single sensor is uncertain,and it is difficult to multi-dimensional fusion of the spatial and temporal prediction information in the landslide monitoring area,two-stage parallel fusion is used to collect spatial and temporal information in landslide monitoring area by means of GIS technology,sensor technology and wireless communication technology,and the temporal information is remotely transmitted to Tlink Internet of Things cloud server in real time. BP neural network is used to fuse the acquired data at the first level to obtain the best value of homogeneous data, and multi-layer fuzzy comprehensive evaluation method is used for the decision-making level fusion. AHP is used to determine the weight of the first layer,and hazard entropy is used to determine the weight of the second layer,so as to realize the adaptive ability of early warning evaluation and protect the information of landslide factor from being concealed after fuzzy operation. The operation results show that multi-dimensional information can be used to obtain more abundant and effective monitoring regional information, eliminate the limitation of single landslide condition factor on landslide grade evaluation, and improve the reliability and accuracy of overall landslide grade evaluation.

关 键 词：滑坡监测 预警评估 多源数据融合 远程传输 BP神经网络 权重计算 结果分析 

分 类 号：TN919-34[电子电信—通信与信息系统] TP311[电子电信—信息与通信工程]