Dynamic displacement reconstruction of bridge based on physics-informed recurrent neural network

Abstract

It is extremely challenging to directly measure the dynamic displacement which is essential in bridge state evaluation. The indirect physical-driven displacement reconstruction methods are restricted by the deviation existing between mechanism model and actual bridge, while indirect data-driven methods are restricted by requirement for a large amount of data. This paper proposes a physics-informed recurrent neural network (PI-RNN) based dynamic displacement reconstruction method. Firstly, the recurrent neural network is established, and the physical equation between data of measured points and target points are derived. Then, the derived physical equation is represented as physical information and added in the loss function of the network. Thus, the loss function contains a physical-based regularization term, which can guide the training direction of the network model, alleviate the over-fitting problem, and improve the generalization ability of the network. Subsequently, the displacement response reconstruction procedure based on PI-RNN is provided in detail. Finally, the effectiveness and superiority of the proposed method are verified by numerical and experimental examples. The results indicate that the PI-RNN is superior to RNN in terms of accuracy and efficiency in reconstruction bridge displacement.

Keywords

Physics-informed neural network / Displacement reconstruction / Physical-driven / Data-driven

References

[1] Ba Y, Zhao G, KAdambi A (2019) Blending Diverse Physical Priors with Neural Networks. CoRR abs/1910.00201. https://doi.org/10.48550/arXiv.1910.00201.

[2] BaoY, LiH, SunX, et al. . Compressive sampling–based data loss recovery for wireless sensor networks used in civil structural health monitoring. Struct Health Monit, 2013, 12178-95.

[3] CrossE, KooK, BrownjohnJ, et al. . Long-term monitoring and data analysis of the Tamar Bridge. Mech Syst Signal Process, 2013, 351–216-34.

[4] FanG, LiJ, HaoH. Lost data recovery for structural health monitoring based on convolutional neural networks. Struct Control Health Monit, 2019, 2610E2433.

[5] GonzálezA, OBrien E, McGetrick P,. Identification of damping in a bridge using a moving instrumented vehicle. J Sound Vib, 2012, 331184115-4131.

[6] HaoT, ScottM, YebinW, et al. . Integrating physics-based modeling with machine learning for lithium-ion batteries. Appl Energy, 2023, 329: 120289.

[7] JiaF, LeiY, LinJ, et al. . Deep neural networks: A promising tool for fault characteristicmining and intelligent diagnosis of rotating machinery with massive data. Mechanical Systems and Signal Processing, 2016, 72: 303-315.

[8] KarimpouliS, TahmasebiP. Physics informed machine learning: Seismic wave equation. Geosci Front, 2020, 1161993-2001.

[9] Kingma D, Ba J, (2014) Adam: A method for stochastic optimization. Arxiv preprint arxiv 1412: 6980. https://doi.org/10.48550/arXiv.1412.6980.

[10] KooYK, BrownjohnWMJ, ListID, et al. . Structural health monitoring of the Tamar suspension bridge. Structural Control and Health Monitoring, 2013, 204609-625.

[11] KwanBO, BrankoG, YousokK, et al. . Convolutional neural network–based data recovery method for structural health monitoring. Struct Health Monit, 2020, 1961821-1838.

[12] LeiX, SiringoringoDM, SunZ, et al. . Displacement response estimation of a cable-stayed bridge subjected to various loading conditions with one-dimensional residual convolutional autoencoder method. Struct Health Monit, 2023, 2231790-1806.

[13] LiJ, LawSS, DingY. Damage Detection of a Substructure Based on Response Reconstruction in Frequency Domain. Key Eng Mater, 2013, 2579569–570823-830.

[14] BaoY, LiH, et al. . Machine learning paradigm for structural health monitoring. Struct Health Monit, 2021, 2041353-1372.

[15] LiH, ZhangZ, et al. Mechanical Systems and Signal Processing, 2024, 209: 111120.

[16] PengJ, HuaY, et al. . Data and physics-driven modeling for fluid flow with a physics-informed graph convolutional neural network. Ocean Eng, 2024, 301: 117551.

[17] QiY, TanZ, SketchSegNet+,. An End-to-End Learning of RNN for Multi-Class Sketch Semantic Segmentation. IEEE Access, 2019, 7: 102717-102726.

[18] RaiR, SahuKC. A Review of Hybrid Physics Guided Machine Learning Techniques With Cyber-Physical System (CPS) Focus. IEEE Access, 2020, 8: 71050-71073.

[19] RibeiroA, SilvaJ, MaiaN. On the generalisation of the transmissibility concept. Mech Syst Signal Process, 2000, 14129-35.

[20] SalvatoreC, SchianoVCD, FabioG, et al. . Scientific Machine Learning Through Physics-Informed Neural Networks: Where we are and What’s Next. J Sci Comput, 2022, 92388.

[21] SharmaS, KumarV. A comprehensive review on multi-objective optimization techniques: Past, present and future. Archives of Computational Methods in Engineering, 2022, 2975605-5633.

[22] ShresthaA, MahmoodA. Review of deep learning algorithms and architectures. IEEE Access, 2019, 7: 53040-53065.

[23] WanZ, LiS, HuangQ, et al. . Structural response reconstruction based on the modal superposition method in the presence of closely spaced modes. Mech Syst Signal Process, 2014, 421–214-30.

[24] Zaremba W, Sutskever I, Vinyals O (2014) Recurrent Neural Network Regularization. CoRR abs/1409.2329. https://doi.org/10.48550/arXiv.1409.2329.

Funding

National Natural Science Foundation of China(52378298)

Natural Science Fund for Distinguished Young Scholars of Anhui Province(2208085J20)

RIGHTS & PERMISSIONS

The Author(s)