基于LSTM神经网络的多源数据融合桥梁变形重构方法

来源:用户上传      作者:张立奎 段大猷 王佐才

　　摘 要：桥梁变形是桥梁状态评估的重要指标，在测量桥梁位移时，直接测量方法如LVDT、RTK-GPS、LDV等在实际应用中有很多局限性。由于从所测得的动应变中获得实时的位移模态和应变模态较为困难，而且大多数应变测量的采样频率较低，基于应变的间接桥梁变形重构方法在应用中仍存在一定的不足。提出一种基于长短时记忆（LSTM）神经网络重构桥梁变形的方法，该网络融合应变和加速度数据在训后可以用作测量数据实时重构桥梁的变形。在应变模态、位移模态以及桥梁中性轴未知的情况下，该方法可以准确地重构变形。通过数值模拟和试验验证了重构结果的准确性，结果表明，基于LSTM的数据融合方法在不同条件下都可以实现高精度的桥梁变形重构。
　　关键词：位移重构;数据融合;长短时记忆网络;结构健康监测;位移监测
　　中图分类号：U446.3 文献标志码：A 文章编号：2096-6717（2022）03-0037-07
　　收稿日期：2021-04-23
　　基金项目：国家自然科学基金（51922036）;安徽省重点研发计划（1804a0802204）
　　作者简介：张立奎（1967- ），男，教授级高工，主要从事大跨桥梁设计及健康监测研究，E-mail：zhanglk49X@ahjkjt.com。
　　王佐才（通信作者），男，教授，博士生导师，E-mail：wangzuocai@hfut.edu.cn。
　　Received：2021-04-23
　　Foundation items：National Natural Science Foundation of China （No. 51922036）; Key Research and Development Plan of Anhui Province （No. 1804a0802204）
　　Author brief：ZHANG Likui （1967- ）， professorate senior engineer， main research interests： design and health monitoring of long span bridges， E-mail： zhanglk49X@ahjkjt.com.
　　WANG Zuocai （corresponding author）， professor， doctorial supervisor， E-mail： wangzuocai@hfut.edu.cn.
　　A multi-source data fusion method for bridge displacement reconstruction based on LSTM neural network
　　ZHANG Likui 1， DUAN Dayou2a，2b， WANG Zuocai2a，2c
　　（1. Anhui Transportation Holding Group Co.， Ltd.， Hefei 230088， P. R. China; 2a. School of Civil Engineering;
　　2b.Anhui Engineering Laboratory of Infrastructural Safety Inspection and Monitoring; 2c. Anhui Engineering Technology Research Center of Disaster Prevention and Mitigation in Civil Engineering， Hefei University of Technology， Hefei 230009， P. R. China）
　　Abstract： Deformation of the bridge is an important index of bridge condition assessment. The direct measurement methods for the bridge deformation such as LVDT， RTK-GPS and LDV have a lot of limitations. Because it is difficult to obtain the real-time displacement modes and strain modes from the measured dynamic strains while the sampling frequency of strain measurement is low， the indirect bridge deformation reconstruction method based on strain still has some shortcomings in engineering practice. This paper proposed a data fusion method to reconstruct the deformation of bridge structure based on LSTM neural network using the measured strain and acceleration data. In this paper， the accuracy of the reconstruction results is verified by numerical simulation and experiment without strain modes， displacement modes and neutral axis. The results show that data fusion method based on LSTM can achieve high accuracy under different conditions.
转载注明来源:https://www.xzbu.com/8/view-15424491.htm