摘要
为实现对大跨桥梁支座性能劣化的准确判别,提出了一种考虑温致位移时滞的大跨桥梁支座性能劣化预警方法.首先,分析了温度效应下支座滑动的力学行为,揭示了支座性能劣化前后温致位移的变化特征;其次,建立了可考虑温致位移时滞的门控循环单元GRU网络模型,对支座温致位移进行预测,并提出了可剔除温度效应影响、凸显支座性能劣化的温致位移预测残差TDPE预警指标;最后,基于某大跨桥梁的监测数据,验证所提方法的有效性.结果表明:GRU网络模型可以自适应剔除支座温致位移的时滞效应,对支座位移具有较高的预测精度,预测误差在5 mm以内;TDPE预警指标可以实现支座温致位移4 mm以上的异常增量预警.
To realize the accurate identification of the performance deterioration of long-span bridge bearing,an early warning method of long-span bridge bearing performance deterioration considering the time lag of the thermal-induced displacement was proposed.Firstly,the mechanical behaviors of bearing sliding under the temperature effect was analyzed,and the variation characteristics of the thermal-induced displacement before and after bearing performance degradation were revealed.Secondly,a gated recurrent unit(GRU)network model considering the time lag effect of the thermal-induced displacement was established to predict the bearing thermal-induced displacement.A warning indicator of the thermal-induced displacement prediction residual error(TDPE)that can eliminate the influence of the temperature effect and highlight the performance degradation of the bearing was proposed.Finally,the effectiveness of the proposed method was verified based on the monitoring data of a long-span bridge.The results show that the GRU network model can eliminate adaptively the time lag effect of the bearing thermal-induced displacement.The bearing displacement can be predicted with high accuracy and the prediction error is within 5 mm.The warning indicator of the TDPE can realize the abnormal incremental warning of the bearing thermal-induced displacement more than 4 mm.
作者
杨东辉
孙家正
伊廷华
李宏男
李冲
李文杰
Yang Donghui;Sun Jiazheng;Yi Tinghua;Li Hongnan;Li Chong;Li Wenjie(School of Construction Engineering,Dalian University of Technology,Dalian 116024,China;CCCC Highway Bridges National Engineering Research Centre Co.,Ltd.,Beijing 100120,China;China Communications Construction Co.,Ltd.,Beijing 100088,China)
出处
《东南大学学报(自然科学版)》
EI
CAS
CSCD
北大核心
2024年第2期268-274,共7页
Journal of Southeast University:Natural Science Edition
基金
国家自然科学基金资助项目(52078102,52250011,52322807)。
关键词
桥梁健康监测
桥梁支座
支座损伤
温度效应
神经网络模型
bridge health monitoring
bridge bearing
bearing damage
temperature effect
neural network model