摘要
对于大跨斜拉桥而言,拉索基频是拉索工作状态的重要体现,并且对整桥工作性能有重要影响,拉索基频监测是桥梁结构健康监测的极重要环节。斜拉桥温度场的变化将带来拉索基频的改变,斜拉桥温度数据与拉索基频有显著相关性。既有研究仅利用主梁温度搭建的拉索基频一元线性回归模型的误差较大,无法满足工程需求。本研究以某大跨斜拉桥为背景,在主梁平均温度基础上进行拓展,增加主梁竖向温差与索塔温度,运用属于机器学习的LSTM长短时记忆网络搭建模型,获得了极高的输出精度;然后采用主跨跨中和边跨跨中的拉索传感器进行了验证,为该桥斜拉索智能维养提供了重要依据。
For long-span cable-stayed bridges,the fundamental frequency of cables is an important embodiment of the working state of cables.At the same time,the fundamental frequency of cables has an important impact on the dynamic performance of the whole bridge.The monitoring of fundamental frequency of cables is a very important link in the health monitoring of bridge structures.The change of temperature field of cable-stayed bridge will change the fundamental frequency of cable.There is a significant correlation between the temperature data of cable-stayed bridge and the fundamental frequency of cable.If an accurate correlation model between cable and ambient temperature can be established,the regression value of cable temperature induced fundamental frequency can be output after the temperature data is input into the model.Taking the regression value as the benchmark,the measured value can be used to perceive whether the fundamental frequency of the cable is abnormal,so the accuracy of the regression model is very important.In the existing research,the univariate linear regression model of cable fundamental frequency built only by the temperature of the main beam has large error,which can not meet the engineering needs.After extending the average temperature of the main beam and increasing the vertical temperature difference of the main beam and the temperature of the cable tower,the LSTM long and short time memory network,which is a kind of machine learning,is used to build the model,and the high output accuracy is obtained.And the cable sensor in the middle of the main span and the side span is used to verify the model,which provides an important basis for the intelligent maintenance of the bridge cables.
作者
李永强
赵瀚玮
赵大成
陈斌
刘兴旺
LI Yong-qiang;ZHAO Han-wei;ZHAO Da-cheng;CHEN Bin;LIU Xing-wang(Liuzhou Urban Infrastructure Maintenance and Mangement Department,Liuzhou 545000,China;School of Civil Engineering,Southeast University,Nanjing 211189,China;China Railway Bridge&Tunnel Technologies Co.Ltd.,Nanjing 210061,China)
出处
《公路》
北大核心
2023年第5期102-109,共8页
Highway
基金
国家自然科学基金项目,项目编号52008099
江苏省自然科学基金项目,项目编号BK20200369
南京江北新区重点研发计划项目,项目编号ZDYF20200118。
关键词
斜拉桥
拉索基频
温度场
LSTM
机器学习
cable-stayed bridge
fundamental frequency of cable
temperature field
LSTM
machine learning