深度学习在桥梁响应预测与健康监测中的应用

田壮; 樊启武; 王昌杰

PDF(707 KB)

铁道工程学报 ›› 2021, Vol. 38 ›› Issue (6) : 47-52.

长大干线: 桥梁工程

深度学习在桥梁响应预测与健康监测中的应用

田壮^1**, 樊启武², 王昌杰²

作者信息 +

Application of Deep Learning in Bridge Structure Response Prediction and Health Monitoring System

TIAN Zhuang¹, FAN Qiwu², WANG Changjie²

Author information +

文章历史 +

摘要

研究目的:随着基于信息化、智能化技术“新基建”的发展,大跨桥梁健康监测系统的应用更加广泛。本文依托某大跨全钢结构斜拉桥,通过深度融合深度学习算法、自动化实时检算以及BIM驱动的信息载体等技术,探索桥梁运营维护的智能化决策实践,提升桥梁智慧化养护水平和资产管理质量。
研究结论:(1)研发的系统可实现Midas civil全功能、全过程的自动化操作,基于云服务,可极大提高实时分析的计算效率;(2)基于BP神经网络算法创建大跨径桥梁结构响应预测元模型,准确率均超过90%,可替代有限元计算实现快速、有效的结构响应分析和状态评估;(3)提出了将元代理模型集成于BIM驱动的桥梁健康监测系统为实时结构性能评估和应急响应提供技术支撑;(4)本研究成果可为桥梁智能化监测运营提供借鉴与参考。

Abstract

Research purposes:With the development of "new infrastructure" based on information and intelligent technology, the application of long-span bridge health monitoring system is more extensive. Based on a long-span cable-stayed bridge with steel structure, the intelligent decision-making practice of bridge operation and maintenance is explored through depth learning integration of algorithm, automatic real-time checking and BIM driven information carrier, to improve the intelligent maintenance level and asset management quality of the bridge.
Research conclusions:(1) The developed system can realize the full function and automatic operation of Midas civil. Based on cloud services, it can greatly improve the computing efficiency of real-time analysis. (2) Based on BP neural network algorithm to create long-span bridge structure response prediction element model, the accuracy is more than 90%, which can replace the finite element calculation to achieve fast and effective structural response analysis and state evaluation. (3) The meta-agent model is integrated into BIM driven bridge health monitoring system to provide technical support for real-time structural performance evaluation and emergency response. (4) The research results can provide reference for intelligent monitoring operation of bridge.

导出引用

田壮, 樊启武, 王昌杰. 深度学习在桥梁响应预测与健康监测中的应用[J]. 铁道工程学报, 2021, 38(6): 47-52

TIAN Zhuang, FAN Qiwu, WANG Changjie. Application of Deep Learning in Bridge Structure Response Prediction and Health Monitoring System[J]. Journal of Railway Engineering Society, 2021, 38(6): 47-52

中图分类号： TP399

参考文献

[1] 蒋燕芳. 基于图像处理与深度学习的RC桥梁表观病害识别[D]. 重庆:重庆大学,2019.
Jiang Yanfang.Identification of Apparent Defects in RC Bridge with Image Processing and Deep Learning[D]. Chongqing:Chongqing University,2019.
[2] Yin T, Zhu H P. An Efficient Algorithm for Architecture Design of Bayesian Neural Network in Structural Model Updating[J]. Computer-Aided Civil and Infrastructure Engineering, 2020(4):354-372.
[3] 袁灿. 基于深度学习的桥梁健康监测传感器优化布置方法研究[D]. 重庆:重庆交通大学,2018.
Yuan Can. Research on Sensor Placement Optimization for Bridge Health Monitoring Based on Deep Learning[D]. Chongqing:Chongqing Jiaotong University,2018.
[4] 侯晋祥. 基于大数据分析的桥梁结构健康状态评估[D]. 重庆:重庆大学,2018.
Hou Jinxiang .The Bridge Structural Health Evaluation Based on Big Data Analysis[D].Chongqing:Chongqing University,2018.
[5] Sorensen S, Nielsen H, Jgensen H J. Influence of Harmonic Voltages on Single Line to Ground Faults in Distribution Networks with Isolated Neutral or Resonant Earthing[C]//International Conference and Exhibition on Electricity Distribution. IET, 2005:1-4.
[6] 朱庆,朱军,黄华平,等.实景三维空间信息平台与数字孪生川藏铁路[J].高速铁路技术,2020(2):46-53.
Zhu Qing, Zhu Jun, Huang Huaping,etc. Real 3D Spatial Information Platform and Digital Twin Sichuan-Tibet Railway[J]. High Speed Railway Technology,2020(2):46-53.
[7] 张贵忠,赵维刚,张浩. 沪通长江大桥数字化运维系统的设计研发[J]. 铁道学报,2019(5): 16-26.
Zhang Guizhong, Zhao Weigang, Zhang Hao. Design and Development of Digital Operation and Maintenance System for Hutong Yangtze River Bridge[J]. Journal of the China Railway Society, 2019(5):16-26.
[8] 顾津申.石济客专济南黄河桥健康监测系统总体设计[J].铁道工程学报,2019(4):54-59.
Gu Jinshen. General Design of Health Monitoring System for the Yellow River Bridge of Shijiazhuang - Jinan Passenger Dedicated Line[J]. Journal of Railway Engineering Society,2019(4):54-59.