Abstract
A machine learning approach to damage detection is presented for a bridge structural health monitoring (SHM) system. The method is validated on the renowned Z24 bridge benchmark dataset where a sensor instrumented, three-span bridge was monitored for almost a year before being deliberately damaged in a realistic and controlled way. Several damage cases were successfully detected, making this a viable approach in a data-based bridge SHM system. The method addresses directly a critical issue in most data-based SHM systems, which is that the collected training data will not contain all natural weather events and load conditions. A SHM system that is trained on such limited data must be able to handle uncertainty in its predictions to prevent false damage detections. A Bayesian autoencoder neural network is trained to reconstruct raw sensor data sequences, with uncertainty bounds in prediction. The uncertainty-adjusted reconstruction error of an unseen sequence is compared to a healthy-state error distribution, and the sequence is accepted or rejected based on the fidelity of the reconstruction. If the proportion of rejected sequences goes over a predetermined threshold, the bridge is determined to be in a damaged state. This is a fully operational, machine learning-based bridge damage detection system that is learned directly from raw sensor data.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Ásgrímsson DS (2019) Quantifying uncertainty in structural condition with Bayesian deep learning—a study on the Z-24 bridge benchmark. KTH
Gonzalez I, Karoumi R (2015) BWIM aided damage detection in bridges using machine learning. J Civ Struct Health Monit 5(5):715–725
Neves AC, González I, Leander J, Karoumi R (2017) Structural health monitoring of bridges: a model-free ANN-based approach to damage detection. J Civ Struct Health Monit 7(5):689–702
Peeters B, De Roeck G (2001) One-year monitoring of the Z 24-bridge: environmental effects versus damage events. Earthquake Eng Struct Dyn 30(2):149–171
Gonzales I, Ülker-Kaustell M, Karoumi R (2013) Seasonal effects on the stiffness properties of a ballasted railway bridge. Eng Struct 57:63–72
Casas JR, Moughty JJ (2017) Bridge damage detection based on vibration data: past and new developments. Front Built Environ 3:4
Das S, Saha P, Patro SK (2016) Vibration-based damage detection techniques used for health monitoring of structures: a review. J Civ Struct Health Monit 6(3):477–507
Neves AC, González I, Karoumi R, Leander J (2020) The influence of frequency content on the performance of artificial neural network-based damage detection systems tested on numerical and experimental bridge data. Struct Health Monit 1475921720924320
Gomes GF, Mendéz YAD, Alexandrino PdaSL, da Cunha SS, Ancelotti AC (2018) The use of intelligent computational tools for damage detection and identification with an emphasis on composites—a review. Compos Struct
Teimouri H, Milani AS, Loeppky J, Seethaler R (2017) A Gaussian process-based approach to cope with uncertainty in structural health monitoring. Struct Health Monit 16(2):174–184
Simoen E, De Roeck G, Lombaert G (2015) Dealing with uncertainty in model updating for damage assessment: a review. Mech Syst Signal Process 56:123–149
Kelley HJ (1960) Gradient theory of optimal flight paths. ARS J 30(10):947–954
Bryson AE, Denham WF (1962) A steepest-ascent method for solving optimum programming problems. J Appl Mech 29(2):247–257
Rumelhart DE, Hinton GE, Williams RJ (1986) Learning representations by back-propagating errors. Nature 323(6088):533
Sutskever I, Martens J, Dahl G, Hinton G (2013) On the importance of initialization and momentum in deep learning. In: International conference on machine learning, pp 1139–1147
Kingma DP, Ba J (2014) Adam: a method for stochastic optimization. arXiv Preprint arXiv1412.6980
Leshno M, Lin VY, Pinkus A, Schocken S (1993) Multilayer feedforward networks with a nonpolynomial activation function can approximate any function. Neural Netw 6(6):861–867
Cybenko G (1989) Approximation by superpositions of a sigmoidal function. Math Control Signals Syst 2(4):303–314
Krizhevsky A, Sutskever I, Hinton GE (2012) Imagenet classification with deep convolutional neural networks. In: Advances in neural information processing systems, pp 1097–1105
Simonyan K, Zisserman A (2014) Very deep convolutional networks for large-scale image recognition. arXiv Preprint arXiv1409.1556
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 770–778
LeCun Y, Bengio Y, Hinton G (2015) Deep learning. Nature 521(7553):436
Hinton GE, Salakhutdinov RR (2006) Reducing the dimensionality of data with neural networks. Science (80-) 313(5786):504–507
Neal RM (1996) Bayesian learning for neural networks. Springer-Verlag New York Inc., New York
MacKay DJC (1992) A practical Bayesian framework for backpropagation networks. Neural Comput 4(3):448–472
Gal Y, Ghahramani Z (2016) Dropout as a Bayesian approximation: representing model uncertainty in deep learning. In: International conference on machine learning, pp 1050–1059
Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R (2014) Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res 15(1):1929–1958
Damianou A, Lawrence N (2013) Deep Gaussian processes. In: Artificial intelligence and statistics, pp 207–215
Gal Y (2016) Uncertainty in deep learning. University of Cambridge
Li Y, Gal Y (2017) Dropout inference in Bayesian neural networks with alpha-divergences. arXiv Preprint arXiv1703.02914
Gal Y, Hron J, Kendall A (2017) Concrete dropout. In: Advances in neural information processing systems, pp 3584–3593
Bergstra J, Bengio Y (2012) Random search for hyper-parameter optimization. J Mach Learn Res 13:281–305
Reynders E, De Roeck G (2009) Continuous vibration monitoring and progressive damage testing on the Z24 bridge. In: Encyclopedia of structural health monitoring
Reynders E, De Roeck G (2014) Vibration-based damage identification: the Z24 bridge benchmark. In: Encyclopedia of earthquake engineering, pp 1–8
Gonzalez I, Karoumi R (2014) Analysis of the annual variations in the dynamic behavior of a ballasted railway bridge using Hilbert transform. Eng Struct 60:126–132
Acknowledgements
We thank Dr. Edwin Reynders at KU Leuven for providing the Z24 benchmark dataset; without the contribution, this work would not have been possible.
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this chapter
Cite this chapter
Ásgrímsson, D.S., González, I., Salvi, G., Karoumi, R. (2022). Bayesian Deep Learning for Vibration-Based Bridge Damage Detection. In: Cury, A., Ribeiro, D., Ubertini, F., Todd, M.D. (eds) Structural Health Monitoring Based on Data Science Techniques. Structural Integrity, vol 21. Springer, Cham. https://doi.org/10.1007/978-3-030-81716-9_2
Download citation
DOI: https://doi.org/10.1007/978-3-030-81716-9_2
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-81715-2
Online ISBN: 978-3-030-81716-9
eBook Packages: EngineeringEngineering (R0)