Research on the Identification of Bridge Structure damage Based on Multi-source Sensor Data and Graphical Convolutional Network

DOI

10.2991/978-94-6463-856-1_36How to use a DOI?

Keywords

Bridge damage identification; Graph convolutional network; Multi-source data fusion; Structural health monitoring; Real-time monitoring

Abstract

This paper proposes an intelligent bridge structure damage identification method based on multi-source sensor data fusion and Graph Convolutional Networks (GCN). As bridges age, structural damage monitoring becomes crucial for ensuring traffic safety. Traditional structural health monitoring methods struggle to simultaneously utilize the spatial distribution characteristics of sensors and the complementary advantages of multiple data types. To address this issue, this research constructs a GCN model architecture that incorporates sensor topological relationships by representing each sensor node and its spatial connections as a graph structure, effectively extracting damage features. Experiments were conducted on the Z24 bridge dataset and a self-designed cable-stayed bridge experimental model. Results show that the proposed multi-source data fusion GCN method achieves a damage classification accuracy of 95.1%, representing improvements of 7.1% and 5.6% over traditional 1D-CNN and LSTM methods, respectively. Although GCN’s single inference time is slightly higher than comparative methods, it can still achieve millisecond-level real-time monitoring with hardware acceleration. Case studies further confirm the method’s high recognition capability and good robustness for multiple types of bridge damage in complex environments, providing a new technical approach for intelligent and real-time bridge safety monitoring.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Zhongyu Shi
AU  - Boren Yang
AU  - Xi Ling
AU  - Sijie Zhou
AU  - Hexiao Wang
AU  - Jiachen Yi
PY  - 2025
DA  - 2025/09/22
TI  - Research on the Identification of Bridge Structure damage Based on Multi-source Sensor Data and Graphical Convolutional Network
BT  - Proceedings of the 2025 International Conference on Resilient City and Safety Engineering (ICRCSE 2025)
PB  - Atlantis Press
SP  - 386
EP  - 394
SN  - 2352-5401
UR  - https://doi.org/10.2991/978-94-6463-856-1_36
DO  - 10.2991/978-94-6463-856-1_36
ID  - Shi2025
ER  -