数据清洗与预警策略在桥梁健康监测系统中的应用

Application of Data Cleaning and Early Warning Strategy in Bridge Health Monitoring System

桥梁健康监测数据难以服务于桥梁实际安全运营,相当一部分原因是受到数据失真和预警策略不完善的影响。为了解决这一问题,分析总结了国内外常见数据失真类型和识别修复算法,建立了数据清洗策略;采用数据驱动和有限元建模多层级耦合的方法,制定一种基于多层神经网络的桥梁有限元模型损伤识别预警技术。经如皋焦港大桥健康监测系统历史监测数据比对分析,数据清洗策略识别准确率和灵敏度在95%以上;建立的三级预警体系,符合原始应变的变化规律,解决了系统因阈值设置不合理导致的误报警问题。

Bridge health monitoring data is difficult to serve the practical safety operation of bridges.A significant portion of this issue arises due to data distortion and inadequate warning strategies.In order to solve this problem,this paper analyzes and summarizes the common data distortion types and recognition and repair algorithms at home and abroad,establishes a data cleaning strategy,and formulates a damage recognition and early warning technology for bridge finite element model based on multi-layer neural network by using the multi-level coupling method of data-driven and finite element modeling.According to the comparison and analysis of the historical monitoring data of the health monitoring system of Rugao Jiaogang Bridge,the accuracy and sensitivity of the data cleaning strategy are more than 95%,and the three-level early warning system established is in line with the change law of the original strain,which solves the problem of false alarm caused by the unreasonable threshold setting of the system.