Using physics-informed graph neural networks for modal identification of a population of structures
Xudong Jian, Wei Liu, Kiran Bacsa, Eleni Chatzi
Abstract. Modal identification is crucial for almost every downstream task related to structural health monitoring (SHM), in the sense that it contains information that is vital across the levels of the SHM hierarchy, including damage identification and prediction of future performance. This study proposes a deep learning model that is built on the Transformer module and the GraphSAGE module for modal identification across a population of structures. The model processes structural dynamic measurements and topology of structural systems to output the decomposed modal responses and corresponding mode shapes of monitored structures. Based on the model input (structural dynamic measurements) and output (modal responses and mode shapes), we exploit the modal decomposition theory and independence of the structural modes, to train the model in a physics-informed manner. We perform numerical simulation to verify the proposed model. Results show that the proposed model can decompose dynamic response for structural configurations from both the training set and the unseen testing set, demonstrating its accuracy and generalization ability in terms of modal decomposition. The decomposed modal responses can further be used to identify natural frequencies and damping ratios.
Keywords
Population-Based Structural Health Monitoring, Modal Identification, Physics Informed Deep Learning, Graph Neural Network, Transformer
Published online 3/25/2025, 8 pages
Copyright © 2025 by the author(s)
Published under license by Materials Research Forum LLC., Millersville PA, USA
Citation: Xudong Jian, Wei Liu, Kiran Bacsa, Eleni Chatzi, Using physics-informed graph neural networks for modal identification of a population of structures, Materials Research Proceedings, Vol. 50, pp 172-179, 2025
DOI: https://doi.org/10.21741/9781644903513-21
The article was published as article 21 of the book Structural Health Monitoring
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 license. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
References
[1] Worden, K., Bull, L. A., Gardner, P., Gosliga, J., Rogers, T. J., Cross, E. J., … & Dervilis, N. (2020). A brief introduction to recent developments in population-based structural health monitoring. Frontiers in Built Environment, 6, 146. https://doi.org/10.3389/fbuil.2020.00146
[2] Zhao, M., Taal, C., Baggerohr, S., & Fink, O. (2024). Graph neural networks for virtual sensing in complex systems: Addressing heterogeneous temporal dynamics. arXiv preprint arXiv:2407.18691. https://doi.org/10.2139/ssrn.4941745
[3] Haywood-Alexander, M., Liu, W., Bacsa, K., Lai, Z., & Chatzi, E. (2023). Discussing the Spectra of Physics-Enhanced Machine Learning via a Survey on Structural Mechanics Applications. arXiv preprint arXiv:2310.20425. https://doi.org/10.1017/dce.2024.33
[4] Liu, D., Tang, Z., Bao, Y., & Li, H. (2021). Machine‐learning‐based methods for output‐only structural modal identification. Structural Control and Health Monitoring, 28(12), e2843. https://doi.org/10.1002/stc.2843
[5] Shu, J., Zhang, C., Gao, Y., & Niu, Y. (2023). A multi-task learning-based automatic blind identification procedure for operational modal analysis. Mechanical Systems and Signal Processing, 187, 109959. https://doi.org/10.1016/j.ymssp.2022.109959
[6] Bao, Y., Liu, D., & Li, H. (2024). A mechanics-informed neural network method for structural modal identification. Mechanical Systems and Signal Processing, 216, 111458. https://doi.org/10.1016/j.ymssp.2024.111458
[7] Jian, X., Xia, Y., Duthé, G., Bacsa, K., Liu, W., & Chatzi, E. (2024). Using Graph Neural Networks and Frequency Domain Data for Automated Operational Modal Analysis of Populations of Structures. arXiv preprint arXiv:2407.06492.
[8] Hamilton, W., Ying, Z., & Leskovec, J. (2017). Inductive representation learning on large graphs. Advances in neural information processing systems, 30.
[9] Hembert, P., Ghnatios, C., Cotton, J., & Chinesta, F. (2024). Assessing Sensor Integrity for Nuclear Waste Monitoring Using Graph Neural Networks. Sensors, 24(5), 1580. https://doi.org/10.3390/s24051580
[10] Jian, X. (2024). Github repository [https://github.com/JxdEngineer].
