Research on Structural Damage Identification Based on Multiple Model Deep ResNet

DOI

10.2991/978-94-6463-728-1_80How to use a DOI?

Keywords

structural health monitoring; structural damage identification; deep learning; CNN; ResNet

Abstract

Damage identification of structural components has always been one of the key issues in the field of structural health monitoring, which is of great significance to ensure the stability and service life of components. Although various machine learning and deep learning algorithms have given many feasible approaches in the field of structural damage identification, it is still a key technical challenge to find the key information of component damage in the large amount of data provided by various sensors. This paper proposes a 1D structural damage recognition algorithm based on residual neural network (ResNet), by increasing the batch standard normalization layer and residual network module, increasing the model learning depth at the same time can ensure that the model training difficulty does not increase, combined with the Convolutional Neural Networks (CNN) convolutional layer, activation layer and pooling layer, a better result conclusion is obtained. On this basis, the feature data classification is carried out by combining the fully connected layer with Softmax function, and then the structural damage localization and quantitative effective identification are carried out. Modal analysis using numerical simulation of steel truss structure is carried out to verify the above frame structure. The results show that compared with the general convolutional neural network, the accuracy of structural component damage identification based on ResNet is significantly improved, and the damage classification accuracy reaches more than 90.0%, which proves that the method has a high recognition accuracy for structural damage identification of beams.

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  - Haixin Xia
AU  - Xiaomin Huang
PY  - 2025
DA  - 2025/05/19
TI  - Research on Structural Damage Identification Based on Multiple Model Deep ResNet
BT  - Proceedings of the 3rd International Conference on Green Building, Civil Engineering and Smart City (GBCESC 2024)
PB  - Atlantis Press
SP  - 868
EP  - 887
SN  - 2352-5401
UR  - https://doi.org/10.2991/978-94-6463-728-1_80
DO  - 10.2991/978-94-6463-728-1_80
ID  - Xia2025
ER  -