Wind turbine structure health monitoring through zero-shot learning with supervised variational autoencoders
Kiran Bacsa, Gregory Duthé, Wei Liu, Xudong Jian, Eleni Chatzi
Abstract. Real-world Structural Health Monitoring (SHM) applications can be classified to Zero-Shot Learning (ZSL) tasks, since structural data hardly ever reveal or label the full extent of damage that may be incurred to the system. This aligns with the definition of ZSL tasks, where part of the classes (or even all but one class) of the problem are hidden during the training time. Thus, the model must be able to generalize to classes that it has not been exposed to during the training. In SHM, this translates to a model of the studied structure generalizing the states/damages that are not present within the training data. This type of generalization is crucial to design resilient digital twins of infrastructure. Variational inference models are a prominent set of models for engineering features for ZSL tasks. This is because they learn latent features for the model with regularization for independence. Thus new clusters corresponding to unseen data would be more clearly identifiable in such a latent space. Specifically, Conditional Variational Autoencoders (CVAE) are a popular method used in ZSL tasks, as they allow the conditioning of the latent variables on external information pertaining to the data samples. Recent works have shown that Supervised VAEs (SVAE) can learn features that generalize just as well as CVAEs, yet they do not require the labels as an input during deployment. The SVAE conditions the latent space with an auxiliary tasks, in this case the GZSL task. Both the latent features and the GSZL task are learned jointly. Thus the labels are only needed during the training phase. We validate our method on a synthetic dataset consisting of simulated measurements on a wind turbine subject to stiffness degradation. In this case, the different classes correspond to different levels of erosion of the blades. Part of these damage levels are removed from the training set. We show that, when the unseen classes can be expressed by an interpolation of the seen classes, that the SVAE is able to learn global features. Moreover, our study demonstrates this generalization by tackling the damage detection task in a ZSL setting.
Keywords
Deep Learning, Zero-Shot Learning, Structure Health Monitoring, Wind Turbines
Published online 3/25/2025, 12 pages
Copyright © 2025 by the author(s)
Published under license by Materials Research Forum LLC., Millersville PA, USA
Citation: Kiran Bacsa, Gregory Duthé, Wei Liu, Xudong Jian, Eleni Chatzi, Wind turbine structure health monitoring through zero-shot learning with supervised variational autoencoders, Materials Research Proceedings, Vol. 50, pp 189-200, 2025
DOI: https://doi.org/10.21741/9781644903513-23
The article was published as article 23 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] Mishnaevsky L., Hasager C., Bak C. et al. (2021), Leading edge erosion of wind turbine blades: Understanding, prevention and protection, Renewable Energies, 169, pp. 953-969 https://doi.org/10.1016/j.renene.2021.01.044
[2] Keegan M. H. (2014), Wind Turbine Blade Leading Edge Erosion: An Investigation of Rain Droplet and Hailstone Impact Induced Damage Mechanisms
[3] Barber, S., Deparday, J., Marykovskiy, Y., Chatzi, E., Abdallah, I. and Duthé, G., Magno, M., Polonelli, T., Fischer, R., Mueller, H. (2022), Development of a wireless, non-intrusive, MEMS-based pressure and acoustic measurement system for large-scale operating wind turbine blades, Wind Energy Science, 7(4), pp 1383 – 1398 https://doi.org/10.5194/wes-7-1383-2022
[4] Simpson T., Dervilis N. and Chatzi E. (2021), Machine Learning Approach to Model Order Reduction of Nonlinear Systems via Autoencoder and LSTM Networks, Journal of Engineering Mechanics, 147, 04021061 https://doi.org/10.1061/(ASCE)EM.1943-7889.0001971
[5] Li Z., Xu P., Xing J. et al. (2022), SDFormer: A Novel Transformer Neural Network for Structural Damage Identification by Segmenting the Strain Field Map, Sensors (Basel, Switzerland), 22(6), 2358 https://doi.org/10.3390/s22062358
[6] Kingma D. P. and Welling M. (2013), Auto-Encoding Variational Bayes
[7] Bacsa K., Liu W., Abdallah I., Chatzi E., Structural Dynamics Feature Learning Using a Supervised Variational Autoencoder, Journal of Engineering Mechanics 151 (2), 04024106 https://doi.org/10.1061/JENMDT.EMENG-7635
[8] Pourpanah F., Abdar M., Luo Y., Zhou X., Wang R., Lim C. P., Wang X. Z., Wu Q. M. J. (2022), A review of generalized zero-shot learning methods, IEEE transactions on pattern analysis and machine intelligence https://doi.org/10.1109/TPAMI.2022.3191696
[9] Sohn K., Lee H., and Yan X. (2015), Learning structured output representation using deep conditional generative models, in Proc. Adv. Neural Informat. Process. Syst., pp. 3483-3491
[10] Mishra A., Krishna R. S., Mittal A, and Murthy H. A. (2018), A generative model for zero shot learning using conditional variational autoencoders, in Proc. IEEE Conf. Comput. Vis. Pattern Recognit. Workshops, pp. 2188-2196 https://doi.org/10.1109/CVPRW.2018.00294
[11] Huang H., Wang C., Yu P. S., and Wang C. D. (2019), Generative dual adversarial network for generalized zero-shot learning, in Proc. IEEE Conf. Comput. Vis. Pattern Recognit., pp. 801-810 https://doi.org/10.1109/CVPR.2019.00089
[12] Abdallah I., Natarajan A. and Sørensen J. (2015), Impact of uncertainty in airfoil characteristics on wind turbine extreme loads, Renew. Energy, 75, pp. 283-300 https://doi.org/10.1016/j.renene.2014.10.009
[13] Ji T., Vuppala S. T., Chowdhary G. et al. (2020), Multi-Modal Anomaly Detection for Unstructured and Uncertain Environments, in Conference on Robot Learning (CoRL)
[14] Kingma D. P. and Ba J. (2015), Adam: A Method for Stochastic Optimization, ICLR (Poster)
[15] Van der Maaten L. and Hinton H. (2008), Visualizing Data using t-SNE, JMLR 9(86), pp. 2579−2605
