Hybrid deep fault detection and isolation: Combining deep neural networks and system performance models

##plugins.themes.bootstrap3.article.main##

##plugins.themes.bootstrap3.article.sidebar##

Manuel Arias Chao Chetan Kulkarni Kai Goebel Olga Fink

Abstract

With the increased availability of condition monitoring data on the one hand and the increased complexity of explicit system physics-based models on the other hand, the application of data-driven approaches for fault detection and isolation has recently grown. While detection accuracy of such approaches is generally very good, their performance on fault isolation often suffers from the fact that fault conditions affect a large portion of the measured signals thereby masking the fault source. To overcome this limitation, we propose a hybrid approach combining physical performance models with deep learning algorithms. Unobserved process variables are inferred with a physics-based performance model to enhance the input space of a data-driven diagnostics model. The resulting increased input space gains representation power enabling more accurate fault detection and isolation. To validate the effectiveness of the proposed method, we generate a condition monitoring dataset of an advanced gas turbine during flight conditions under healthy and four faulty operative conditions based on the Aero-Propulsion System Simulation (C-MAPSS) dynamical model. We evaluate the performance of the proposed hybrid methodology in combination with two different deep learning algorithms: deep feed forward neural networks and Variational Autoencoders, both of which demonstrate a significant improvement when applied within the hybrid fault detection and diagnostics framework. The proposed method is able to outperform pure data-driven solutions, particularly for systems with a high variability of operating conditions. It provides superior results both for fault detection as well as for fault isolation. For the fault isolation task, it overcomes the smearing effect that is commonly observed in pure data-driven approaches and enables a precise isolation of the affected signal. We also demonstrate that deep learning algorithms provide a better performance on the fault detection task compared to the traditional machine
learning algorithms.

Abstract 46 | PDF Downloads 58

##plugins.themes.bootstrap3.article.details##

Keywords

Fault detection and isolation, deep learning, Variational Auto-Encoders (VAE), Calibration-Based Hybrid Diagnostics

References
Arias Chao, M., Lilley, D. S., Math´e, P., & Schloßhauer, V. (2015). Calibration and Uncertainty Quantification of Gas Turbine Performance Models. In Proceedings of the asme turbo expo (Vol. 7A, p. V07AT29A001). doi: 10.1115/gt2015-42392
Baraldi, P., Di Maio, F., Turati, P., & Zio, E. (2015). Robust signal reconstruction for condition monitoring of industrial components via a modified auto associative kernel regression method. Mechanical Systems and Signal Processing, 60, 29–44.
Borguet, S. (2012). Variations on the Kalman Filter for Enhanced Performance Monitoring of Gas Turbine Engines (PhD Thesis). Universit´e de Li`ege.
Brunell, B. J., Mathews, J. H. K., & Aditya Kumar. (2004). United States Patent Design of an Adaptive Model-Based Control for Controlling a Gas Turbine
(Vol. 121). DASHlink - Flight Data For Tail 687. (2012). Retrieved 2019-01-29, from https://c3.nasa.gov/dashlink/Doersch, C. (2016, jun). Tutorial on Variational Autoencoders (Tech. Rep.). Retrieved from http://arxiv.org/abs/1606.05908
Dourado, A., & Viana, F. A. C. (2019, sep). Physics- Informed Neural Networks for Corrosion-Fatigue Prognosis. Proceedings of the Annual Conference
of the PHM Society, 11(1). doi: 10.36001/PHMCONF. 2019.V11I1.814
Frank, S., Heaney, M., Jin, X., Robertson, J., Cheung, H., Elmore, R., & Henze, G. P. (2016). Hybrid Model-Based and Data-Driven Fault Detection and Diagnostics for Commercial Buildings. Proceedings of the ACEEE Summer Study on Energy Efficiency in Buildings, Aug 21-26, 12.1–12.14.
Frederick, D. K., Decastro, J. A., & Litt, J. S. (2007). User’s Guide for the Commercial Modular Aero-Propulsion System Simulation (C-MAPSS) (Tech. Rep.). Retrieved from http://www.sti.nasa.gov
Glorot, X., & Bengio, Y. (2010). Understanding the difficulty of training deep feedforward neural networks (Tech. Rep.). Retrieved from http://www.iro.umontreal.
Hanachi, H., Yu, W., Kim, I. Y., & Mechefske, C. K. (2017). Hybrid Physics-Based and Data-Driven PHM. Canadian Machinery Vibration Association (CMVA) Annual Conference, Edmonton, Alberta, Canada.
Hu, Y., Palm´e, T., & Fink, O. (2017). Fault detection based on signal reconstruction with auto-associative extreme learning machines. Engineering Applications of Artificial Intelligence, 57, 105–117.
Jia, X., Karpatne, A., Willard, J., Steinbach, M., Read, J., Hanson, P. C., . . . Kumar, V. (2018, oct). Physics Guided Recurrent Neural Networks
For Modeling Dynamical Systems: Application to Monitoring Water Temperature And Quality In Lakes (Tech. Rep.). Retrieved from http://arxiv.org/abs/1810.02880
Julier, S. J., & Uhlmann, J. K. (1997). New extension of the Kalman filter to nonlinear systems. In Signal processing, sensor fusion, and target recognition vi (Vol. 3068, p. 182). doi: 10.1117/12.280797
Khan, S., & Yairi, T. (2018, jul). A review on the application of deep learning in system health management (Vol. 107). Academic Press. doi:
10.1016/j.ymssp.2017.11.024
Kingma, D. P., & Ba, J. L. (2015). Adam: A method for stochastic optimization. In 3rd international conference on learning representations, iclr 2015 - conference track proceedings.
Kingma, D. P., & Welling, M. (2014). Auto-encoding variational bayes. In 2nd international conference on learning representations, iclr 2014 - conference track proceedings.
LeCun, Y. A., Bottou, L., Orr, G. B., & M¨uller, K. R. (2012). Efficient backprop. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 7700 LECTU, 9–48. doi: 10.1007/978-3-642-35289-8-3
Michau, G., Hu, Y., Palm´e, T., & Fink, O. (2017). Feature Learning for Fault Detection in High-Dimensional Condition-Monitoring Signals. Submitted for a possible publication in IEEE Transactions on Cybenetics.
Michau, G., Palm´e, T., & Fink, O. (2017). Deep feature learning network for fault detection and isolation. In Phm 2017, st. petersburg, usa, 2-5 october 2017 (pp. 108–118).
Moya, M. M., & Hush, D. R. (1996). Network constraints and multi-objective optimization for one-class classification. Neural Networks, 9(3), 463–474. doi: 10.1016/0893-6080(95)00120-4
Nascimento, R. G., & Viana, F. A. (2019). Fleet prognosis with physics-informed recurrent neural networks. In Structural health monitoring 2019: Enabling intelligent life-cycle health management for industry internet of things (iiot) - proceedings of the 12th international workshop on structural health monitoring (Vol. 2, pp.1740–1747).
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., . . . Duchesnay, E. (2011). Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12, 2825–2830.
Rausch, R. T., Goebel, K. F., Eklund, N. H., & Brunell, B. J. (2005). Integrated In-Flight Fault Detection and Accommodation: A Model-Based Study. In Volume 1: Turbo expo 2005 (pp. 561–569). ASME. doi: 10.1115/GT2005-68300
Sarkar, D., Bali, R., & Ghosh, T. (2018). Hands-on transfer learning with Python : implement advanced deep learning and neural network models using TensorFlow and Keras.
Scheirer, W. J., de Rezende Rocha, A., Sapkota, A., & Boult, T. E. (2013). Toward Open Set Recognition. IEEE Transactions on Pattern Analysis
and Machine Intelligence, 35(7), 1757–1772. doi: 10.1109/TPAMI.2012.256
Sch¨olkopf, B., Williamson, R., Smola, A., Shawe-Taylor, J., & Piatt, J. (2000). Support vector method for novelty detection. In Advances in neural information processing systems (pp. 582–588).
Wang, Q., Michau, G., & Fink, O. (2019). Domain adaptive transfer learning for fault diagnosis. arXiv preprint arXiv:1905.06004.
Wang, S., Minku, L. L., & Yao, X. (2013). Online class imbalance learning and its applications in fault detection. International Journal of Computational Intelligence and Applications, 12(04), 1340001.
Xu, L., Chow, M.-Y., & Taylor, L. S. (2007). Power distribution fault cause identification with imbalanced data using the data mining-based fuzzy classification e-algorithm. IEEE Transactions on Power Systems, 22(1), 164–171.
Yucesan, Y. A., & Viana, F. A. C. (2019). Wind Turbine Main Bearing Fatigue Life Estimation with Physicsinformed Neural Networks. In Phm 2019 (Vol. 11, pp. 1–14). doi: 10.36001/PHMCONF.2019.V11I1.807
Zhang, Y., Li, X., Gao, L., Wang, L., & Wen, L. (2018). Imbalanced data fault diagnosis of rotating machinery using synthetic oversampling and feature learning. Journal of manufacturing systems, 48, 34–50.
Zhao, R., Yan, R., Chen, Z., Mao, K., Wang, P., & Gao, R. X. (2016, dec). Deep Learning and Its Applications to Machine Health Monitoring:
A Survey (Tech. Rep.). Retrieved from http://arxiv.org/abs/1612.07640
Zhao, S., Song, J., & Ermon, S. (2019, jul). InfoVAE: Balancing Learning and Inference in Variational Autoencoders. In Proceedings of the aaai conference on artificial intelligence (Vol. 33, pp. 5885–5892). doi: 10.1609/aaai.v33i01.33015885
Zhu, W., Miao, J., Qing, L., & Huang, G.-B. (2015). Hierarchical Extreme Learning Machine for unsupervised representation learning. In 2015 international joint conference on neural networks (ijcnn) (pp. 1–8). IEEE. doi: 10.1109/IJCNN.2015.7280669
Section
Technical Papers