Deflection prediction and early warning method of main girder of suspension bridge based on CNN-LSTM-GD

PDF(2458 KB)

Journal of Vibration and Shock ›› 2025, Vol. 44 ›› Issue (14) : 84-95.

VIBRATION THEORY AND INTERDISCIPLINARY RESEARCH

Deflection prediction and early warning method of main girder of suspension bridge based on CNN-LSTM-GD

XIAO Xinhui1，LIU Xian1，ZHANG Haiping*1，WANG Zepeng1，CHEN Fanghuai1，LUO Yuan1，LIU Yang2

Author information +

History +

Abstract

As a flexible bridge, the deflection control of the main girder is particularly important during the operation of a suspension bridge. To predict the vertical deflection of the main girder of an existing suspension bridge under the combined effects of random traffic flow and environmental temperature, this paper establishes an integrated deflection interval prediction method based on Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM) networks, probability density estimation layer, and bridge monitoring data. Using health monitoring data from the Nanxi Yangtze River Bridge, a time series training set of environmental temperature, vehicle load, and deflection monitoring data was established. The combined CNN-LSTM layers captured local features and long-term memory in the time series. A Gaussian distribution was used as the probability density function, and the parameters of the Gaussian distribution were evaluated using the maximum likelihood method, resulting in optimal deflection prediction values and probability intervals. Based on this, a method for identifying abnormal deflection and warning thresholds for the main girder of existing suspension bridges was proposed. The study shows that compared to LSTM and CNN-LSTM models, the CNN-LSTM-GD model has better predictive capabilities for small deflection fluctuations and extreme deflections, with deflection prediction values closely matching the monitoring data. Over a 24-hour time scale, compared to the traditional LSTM model, the CNN-LSTM-GD model improved the Root Mean Square Error (RMSE) and the coefficient of determination (R2) by 54.40% and 10.22%, respectively. Compared to the CNN-LSTM model, the improvements in RMSE and R2 were 38.43% and 5.31%, respectively.

Key words

bridge engineering / structural health monitoring / probabilistic deep learning / bridge deflection / vehicle load / environmental temperature

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

XIAO Xinhui1, LIU Xian1, ZHANG Haiping1, WANG Zepeng1, CHEN Fanghuai1, LUO Yuan1, LIU Yang2. Deflection prediction and early warning method of main girder of suspension bridge based on CNN-LSTM-GD[J]. Journal of Vibration and Shock, 2025, 44(14): 84-95

References

[1] 施洲, 顾家昌, 余万庆, 等. 大跨度斜拉桥双向曲面混合桥塔钢-混结合段受力性能研究 [J]. 中国公路学报, 2022, 35(06): 73-85.
SHI zhou, GU jia-chang, YU wan-qing, et al. Study on the Mechanical Behavior of Bi-directionally Curved Composite Pylon Steel-concrete Joints for a Long-span Cable-stayed Bridge [J]. China Journal of Highway and Transport, 2022, 35(06): 73-85.
[2] 余加勇, 彭志豪, 孟晓林, 等. 极端台风下特大悬索桥的GNSS实时振动监测及动力响应分析 [J]. 中国公路学报: 1-14.
YU jia-yong, PENG zhi-hao, MENG xiao-lin, et al. GNSS-based Real-time Vibration Monitoring and Dynamic Response Analysis of a Long-span Suspension Bridge under an Extreme Typhoon Event [J]. China Journal of Highway and Transport: 1-14
[3] HUANG Y, WANG Y, FU J, et al. Measurement of the real-time deflection of cable-stayed bridge based on cable tension variations [J]. Measurement, 2018, 119: 218-28.
[4] XIN J, JIANG Y, ZHOU J, et al. Bridge deformation prediction based on SHM data using improved VMD and conditional KDE [J]. Engineering Structures, 2022, 261.
[5] 李喜梅, 王新, 母渤海, 等. 车辆荷载作用下双工字钢-混组合连续梁桥振动控制研究 [J]. 中国公路学报, 2022, 35(07): 173-83.
LI xi-mei, WANG xin, MU bo-hai, et al. Research on Vibration Control of Twin-I Steel-concrete Composite Continuous Girder Bridge Under Vehicle Load [J]. China Journal of Highway and Transport, 2022, 35(07): 173-83.
[6] ZHOU J, SUN Z, WEI B, et al. Deflection-based multilevel structural condition assessment of long-span prestressed concrete girder bridges using a connected pipe system [J]. Measurement, 2021, 169.
[7] 曾国良, 邓扬, 马斌, 等. 基于监测数据的在役大跨度桥梁竖向挠度可靠性评估 [J]. 中南大学学报(自然科学版), 2021, 52(10): 3636-46.
ZENG guo-liang, DENG yang, MA bin, et al. Monitoring-based reliability assessment of vertical deflection for in-service long-span bridges [J]. Journal of Central South University(Science and Technology), 2021, 52(10): 3636-46.
[8] XU X, REN Y, HUANG Q, et al. Thermal response separation for bridge long-term monitoring systems using multi-resolution wavelet-based methodologies [J]. Journal of Civil Structural Health Monitoring, 2020, 10(3): 527-41.
[9] HAN Q, MA Q, XU J, et al. Structural health monitoring research under varying temperature condition: a review [J]. Journal of Civil Structural Health Monitoring, 2020, 11(1): 1-25.
[10] 谭冬梅, 姚欢, 吴浩, 等. 基于EWT-FastICA的斜拉桥监测挠度温度效应分离 [J]. 振动测试与诊断, 2022, 42(05): 980-7+1038-9.
TAN dong-mei, YAO huan, WU hao, et al. Separation of Cable⁃Stayed Bridge Monitoring Deflection Temperature Effect Based on EWT⁃FastICA [J]. Journal of Vibration, Measurement & Diagnosis, 2022, 42(05): 980-7+1038-9.
[11] LI M, REN W X, HUANG T I, et al. Extraction of the Bridge Temperature Strain Effect Based on EMD and IMF Energy [J]. Journal of Highway & Transportation Research & Development, 2016, 10(1): 41-8.
[12] 谭冬梅, 刘晓飞, 柳勇, 等. 基于改进PCA和EEMD的桥梁挠度监测中温度效应分离 [J]. 武汉理工大学学报, 2018, 40(03): 28-34.
TAN dong-mei, LIU xiao-fei, LIU yong, et al. Study on Separation of Bridge Deflection Temperature Effect Based on Improved PCA and EEMD. [J]. Journal of Wuhan University of Technology, 2018, 40(03): 28-34.
[13] LI S, XIN J, JIANG Y, et al. Temperature-induced deflection separation based on bridge deflection data using the TVFEMD-PE-KLD method [J]. Journal of Civil Structural Health Monitoring, 2023, 13(2-3): 781-97.
[14] 李明, 钟继卫, 严凤. 结构温度与悬索桥主梁挠度的关联性分析 [J]. 振动与冲击, 2018, 37(11): 237-45.
LI ming, ZHONG ji-wei, YAN feng. Correlation analysis between structure temperature and deflection of a suspension bridge girder [J]. Journal of Vibration and Shock, 2018, 37(11): 237-45.
[15] XU X, HUANG Q, REN Y, et al. Modeling and Separation of Thermal Effects from Cable-Stayed Bridge Response [J]. Journal of Bridge Engineering, 2019, 24(5).
[16] WANG M, DING Y, ZHAO H. Digital prediction model of temperature-induced deflection for cable-stayed bridges based on learning of response-only data [J]. Journal of Civil Structural Health Monitoring, 2022, 12(3): 629-45.
[17] 郭永刚, 张美霞, 王凯, 等. 基于CEEMDAN-VMD-PSO-LSTM模型的桥梁挠度预测 [J]. 安全与环境工程, 2024, 31(03): 150-9.
GUO yong-gang, ZHANG mei-xia, WANG kai, et al. CEEMDAN-VMD-PSO-LSTM model for bridge deflection prediction [J]. Safety and Environmental Engineering, 2024, 31(03): 150-9.
[18] YUE Z-X, DING Y-L, ZHAO H-W. Deep Learning-Based Minute-Scale Digital Prediction Model of Temperature-Induced Deflection of a Cable-Stayed Bridge: Case Study [J]. Journal of Bridge Engineering, 2021, 26(6).
[19] YUE Z, DING Y, ZHAO H, et al. Ultra-high precise Stack-LSTM-CNN model of temperature-induced deflection of a cable-stayed bridge for detecting bridge state driven by monitoring data [J]. Structures, 2022, 45: 110-25.
[20] WANG H, ZHANG Y-M, MAO J-X, et al. Modeling and forecasting of temperature-induced strain of a long-span bridge using an improved Bayesian dynamic linear model [J]. Engineering Structures, 2019, 192: 220-32.
[21] 贺文宇, 武骥元, 任伟新. 基于车致桥梁响应和L1正则化的损伤识别研究 [J]. 中国公路学报, 2021, 34(04): 61-70.
HE wen-yu, WU ji-yuan, REN wei-xin. Bridge Damage Detection Based on the Moving-vehicle-induced Response and L1 Regularization [J]. China Journal of Highway and Transport, 2021, 34(04): 61-70.
[22] 李锦华, 张耀, 张焕涛, 等. 基于移动荷载谱的桥梁竖向自由振动最大响应研究 [J]. 振动与冲击, 2021, 40(14): 261-7.
LI jin-hua, ZHANG yao, ZHANG huan-tao, et al. Investigation of maximum vertical free vibration response of a bridge based on moving load spectrum [J]. Journal of Vibration and Shock, 2021, 40(14): 261-7.
[23] ZHAO H-W, DING Y-L, NAGARAJAIAH S, et al. Behavior Analysis and Early Warning of Girder Deflections of a Steel-Truss Arch Railway Bridge under the Effects of Temperature and Trains: Case Study [J]. Journal of Bridge Engineering, 2019, 24(1).
[24] 鞠翰文, 邓扬, 李爱群. 桥梁结构挠度-温度-车辆荷载监测数据相关性模型 [J]. 振动与冲击, 2023, 42(06): 79-89.
JU han-wen, DENG yang, LI ai-qun. Correlation model of deflection-temperature-vehicle load monitoring data for bridge structures [J]. Journal of Vibration and Shock, 2023, 42(06): 79-89.
[25] 刘凡, 李利祥, 赵岩. 移动荷载作用下具有不确定参数桥梁动力响应分析 [J]. 应用数学和力学, 2023, 44(03): 241-7.
LIU fan, LI li-xiang, ZHAO yan. Dynamic Responses Analysis of Bridges With Uncertain Parameters Under Moving Loads [J]. Applied Mathematics and Mechanics, 2023, 44(03): 241-7.
[26] GUO Q, SUN Y, MI T. Assessment on long-term deflection of concrete beam bridges based on uncertainty quantification method [J]. Structures, 2021, 34: 3013-27.
[27] NI Y-C, ZHANG Q-W, LIU J-F. Dynamic Property Evaluation of a Long-Span Cable-Stayed Bridge (Sutong Bridge) by a Bayesian Method [J]. International Journal of Structural Stability and Dynamics, 2018, 19(01).
[28] HICKEY J, LANGLEY R. Alternative metrics for design decisions based on separating aleatory and epistemic probabilistic uncertainties [J]. Mechanical Systems and Signal Processing, 2022, 181.
[29] WINDSCHITL P D, MILLER J E, PARK I, et al. The desirability bias in predictions under aleatory and epistemic uncertainty [J]. Cognition, 2022, 229: 105254.
[30] SOUSA H, SANTOS L O, CHRYSSANTHOPOULOS M. Quantifying monitoring requirements for predicting creep deformations through Bayesian updating methods [J]. Structural Safety, 2019, 76: 40-50.
[31] BAI S, KOLTER J Z, KOLTUN V. An empirical evaluation of generic convolutional and recurrent networks for sequence modeling [J]. arXiv preprint arXiv:180301271, 2018.
[32] DENG Y, JU H, ZHAI W, et al. Correlation model of deflection, vehicle load, and temperature for in‐service bridge using deep learning and structural health monitoring [J]. Structural Control and Health Monitoring, 2022, 29(12).
[33] 赵维刚, 王新敏, 杜彦良, 等. 常用跨度铁路桥梁运行状态分布式监测与预警技术 [J]. 上海交通大学学报, 2015, 49(07): 1046-51.
ZHAO wei-gang, WANG xin-min, DU yan-liang, et al. Distributed Running State Monitoring and Pre-Warning of Common Span Railway Bridges [J]. Journal of Shanghai Jiaotong University, 2015, 49(07): 1046-51.
[34] 苏成, 廖威, 袁昆, 等. 桥梁健康监测在线预警指标研究 [J]. 桥梁建设, 2015, 45(03): 44-50.
SU cheng, LIAO wei, YUAN kun, et al. Research of Online Early Warning Indices for Bridge Health Monitoring [J]. Bridge Construction, 2015, 49(07): 1046-51.
[35] FAN Z, HUANG Q, REN Y, et al. Real-Time Dynamic Warning on Deflection Abnormity of Cable-Stayed Bridges Considering Operational Environment Variations [J]. Journal of Performance of Constructed Facilities, 2021, 35(1).