基于大数据的桥梁健康监测数据存储及预警方法  

Data Storage and Early-warning Methods of Bridge Health Monitoring System Based on Big-data

作　　者：任普 丁幼亮[1,2] 李亚东 刘刚 蔡曙日[3] REN Pu;DING You-liang;LI Ya-dong;LIU Gang;CAI Shu-ri(Key Laboratory of Concrete and Prestressed Concrete Structures of Ministry of Education, Southeast University, Nanjing 210096, China;Jiangsu Building Mechanical and Electrical Seismic Research Institute, Nanjing 211200, China;Research Institute of Highway, Ministry of Transport, Beijing 100088, China)

机构地区：[1]东南大学混凝土及预应力混凝土结构教育部重点实验室,南京210096 [2]江苏建筑机电抗震研究院,南京211200 [3]交通运输部公路科学研究院,北京100088

出　　处：《科学技术与工程》2019年第12期266-270,共5页Science Technology and Engineering

基　　金：国家重点研发计划(2017YFC0840200)资助

摘　　要：提出一种基于大数据的桥梁健康监测系统平台,在全面考虑影响桥梁服役性能的各个因素下对桥梁安全状态进行评估和实时预警。该系统平台利用容错率高的分布式文件系统以及计算效率高的平行数据处理引擎,具有高可靠性、可用性和存储效率,且易于扩展。进而采用多因素分析方法充分挖掘桥梁各个传感器实时数据之间隐含的关联性,通过分析数据相关性准确建立桥梁服役性能评估模型,并对桥梁安全状况进行实时预警。此外,该系统平台为测试模型的有效性,分别采用多种模型验证方法进行评估,确保大数据分析方法的可靠性。With consideration of the whole factors that influence service performance of bridges,a platform of bridge structural health monitoring based on big-data technology is proposed,which can evaluate the safety condition of bridges and send real-time early-warning. This system platform is high reliability,usability,storage efficiency and easy to extend by means of high fault-tolerance distributed file systems and high computational-efficiency parallel data processing engines. In addition,the multiple factors analysis method is used to explore implicit correlations between real-time sensor data of bridges. Then evaluation models of service performance of bridges are built after analyzing relationships between data,which further send early-warning timely. To test the validation of models,this system arranges several model validation methods respectively to ensure the reliability of big-data analysis method.

关 键 词：桥梁健康监测 大数据 分布式文件系统 多因素分析 

分 类 号：TU311[建筑科学—结构工程]