Abstract
Data driven SHM methodologies take raw signals obtained from sensor networks, and process them to obtain features representative of the condition of the structure. New measurements are then compared with baselines to detect damage. Because damage-sensitive features also exhibit variation due to environmental and operational changes, these comparisons are not always straightforward and an automated, probabilistic approach is necessary, particularly for large-scale sensor networks. In this paper an automated novelty detection methodology based on one-class support vector machines (OCSVM) is proposed and tested on an instrumented experimental steel frame structure. OCSVMs are an advanced machine learning method which can classify new data points based only on data from one class. This enables training of a classifier for damage detection based only on information from a baseline structure. OCSVMs can suffer from over-fitting, a problem which is usually ameliorated by cross-validation. In the absence of any data from the damaged state cross-validation is not possible. In this paper the over-fitting problem is combated by the use of three different recently proposed parameter selection heuristics. These strategies are tested for various damage scenarios of the laboratory structure and the results compared.
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
Scholkopf B, Platt JC, Shawe-Taylor J, Smola AJ, Williamson RC (2001) Estimating the support of a high dimensional distribution. Neural Comput 13(7):1443–1471
Das S, Srivastava AN, Chattopadhyay A (2007) Classification of damage signatures in composite plates using one-class SVMs. In: Aerospace conference, 2007 IEEE, 3–10 March 2007, pp 1, 19
Unnthorsson R, Runarsson RT, Johnson TM (2003) Model selection in one class nu-SVMs using RBF Kernels. In: 16th conference on condition monitoring and diagnostic engineering management, 27–29 August 2003
Khazai S, Homayouni S, Safari A, Mojaradi B (2011) Anomaly detection in hyperspectral images based on an adaptive support vector method. Geosci Remote Sens Lett IEEE 8(4):646–650
Wang H, Zhang L, Xiao Y, Xu W (2012) An approach to choosing Gaussian kernel parameter for one-class SVMs via tightness detecting. In: 2012 4th international conference on intelligent human-machine systems and cybernetics (IHMSC), vol 2, 26–27 August 2012, pp 318, 323
Hofmann T, Schölkopf B, Smola AJ (2008) Kernel methods in machine learning. Ann Stat 36:1171–1220
Schwarz B, Richardson MH (2004) Measurements required for displaying operating deflection shapes. In: Proceedings of IMAC XXII, 26 January 2004
Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, Blondel M, Prettenhofer P, Weiss R, Dubourg V, Vanderplas J, Passos A, Cournapeau D (2011) Scikit-learn: machine learning in Python. J Mach Learn Res 12:2825–2830
Acknowledgements
The authors acknowledge the support provided by Royal Dutch Shell through the MIT Energy Initiative, and thank chief scientists Dr. Dirk Smit, Dr. Sergio Kapusta, project manager Dr. Yile Li, and Shell-MIT liaison Dr. Jonathan Kane for their oversight of this work. Thanks are also due to Dr. Michael Feng and his team from Draper Laboratory for their collaboration in the development of the laboratory structural model and sensor systems. Finally, the authors would like to express their gratitude to Edward H. Linde (1962) and the Linde Family Foundation for their generosity and support through the Presidential Graduate Fellowship awarded to J. Long in 2012.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendix
Appendix
The results discussed in Sect. 14.6 above are shown here in table form.
Rights and permissions
Copyright information
© 2014 The Society for Experimental Mechanics, Inc.
About this paper
Cite this paper
Long, J., Buyukozturk, O. (2014). Automated Structural Damage Detection Using One-Class Machine Learning. In: Catbas, F. (eds) Dynamics of Civil Structures, Volume 4. Conference Proceedings of the Society for Experimental Mechanics Series. Springer, Cham. https://doi.org/10.1007/978-3-319-04546-7_14
Download citation
DOI: https://doi.org/10.1007/978-3-319-04546-7_14
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-04545-0
Online ISBN: 978-3-319-04546-7
eBook Packages: EngineeringEngineering (R0)