Bridge Monitoring Missing Data Reconstruction based on Bi-LSTM

DOI

10.2991/978-94-6463-728-1_71How to use a DOI?

Keywords

Health Monitoring; Data Reconstruction; Bi-LSTM; U-Net; Deep Learning

Abstract

The loss of monitoring data can affect the accurate assessment of bridge structural performance, making it necessary to reconstruct the missing monitoring data. Modern health monitoring systems are characterized by big data, and deep learning models are generally capable of handling large-scale data, making them highly efficient in addressing the problem of missing monitoring data. Therefore, this study presents an approach that utilizes the Bi-LSTM neural network to reconstruct missing bridge monitoring data. The employed network consists of forward LSTM layers and backward LSTM layers, enhancing the network’s predictive capability and accuracy by simultaneously considering the forward and backward information of the sequence. This paper aims to use incomplete multi-channel sensor data as input and employ the Bi-LSTM network to capture the spatiotemporal correlations between the data, ultimately outputting complete multi-channel sensor data. The experiment uses data from The 3rd International Competition for Structural Health Monitoring(IC-SHM 2022), designing various missing conditions to verify the effectiveness of the suggested reconstruction method. Experimental outcomes demonstrate that a network trained with only one type of missing scenario can still perform well under other missing scenarios. Furthermore, U-Net is introduced as a comparison network to additionally verify the efficiency of the proposed approach.

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  - Panping Liu
AU  - Zhiyi Tang
AU  - Wei Xu
PY  - 2025
DA  - 2025/05/19
TI  - Bridge Monitoring Missing Data Reconstruction based on Bi-LSTM
BT  - Proceedings of the 3rd International Conference on Green Building, Civil Engineering and Smart City (GBCESC 2024)
PB  - Atlantis Press
SP  - 759
EP  - 770
SN  - 2352-5401
UR  - https://doi.org/10.2991/978-94-6463-728-1_71
DO  - 10.2991/978-94-6463-728-1_71
ID  - Liu2025
ER  -