Robust Remaining Useful Life Prediction Using Jacobian Feature Regression-Based Model Adaptation
##plugins.themes.bootstrap3.article.main##
##plugins.themes.bootstrap3.article.sidebar##
Abstract
The accurate and robust prediction of remaining useful life (RUL) is critical for enabling the proactive mitigation of fault effects rather than reacting to them. For RUL prediction, one must model nominal and faulty system behaviors and how different faults progress over time. Complex data-driven machine learning (ML) models may capture both nominal and fault progression by updating the model parameters at different stages. As new data are observed, these model parameters can be updated to keep the system model always accurate. However, complete retraining of these models is both data- and computation-intensive and unsuitable for dynamic, fast-changing environments requiring quick recalibration. This calls for efficiently adapting the model to new operating conditions or the system’s current state. One such efficient way to recalibrate model parameters to newly observed data using Jacobian feature regression (JFR) is presented in Forgione, Muni, Piga, and Gallieri (2023), where a recurrent neural network (RNN) models the current behavior of the dynamic system. Then, any subsequent deviation of observed measurements and the RNN model is attributed to an “unacceptable degradation of the nominal model performance.” To update the RNN model, Forgione et al. (2023) propose augmenting the current model with additive correction terms learned by implementing JFR on observed “perturbed system” data. In this paper, we propose an automated online framework to adapt the model efficiently to always reflect the system’s current state and use it for accurate RUL prediction and select JFR as one such adaptation technique. We extend the implementation of JFR-based model adaptation to hybrid models and demonstrate JFR to be more sustainable than the other retraining methods. Finally, we showcase the application of this approach to the oil and gas industry. A testbed that simulates a digital synthetic oilfield is used to show the effectiveness of this adaptation-based RUL prediction technique.
How to Cite
##plugins.themes.bootstrap3.article.details##
Transfer Learning, RUL Prediction, Hybrid Modeling, Domain Adaptation
da Costa, P. R. d. O., Akcay, A., Zhang, Y., & Kaymak, U. (2020). Remaining Useful Lifetime Prediction via Deep Domain Adaptation. Reliability Engineering & System Safety, 195, 106682.
Ding, Y., Ding, P., Zhao, X., Cao, Y., & Jia, M. (2022). Transfer Learning for Remaining Useful Life Prediction Across Operating Conditions Based on Multi-source Domain Adaptation. IEEE/ASME Transactions on Mechatronics, 27(5), 4143-4152.
Forgione, M., Muni, A., Piga, D., & Gallieri, M. (2023). On the Adaptation of Recurrent Neural Networks for System Identification. Automatica, 155, 111092.
Huang, Z., Xu, Z., Wang, W., & Sun, Y. (2015). Remaining Useful Life Prediction for a Nonlinear Heterogeneous Wiener Process Model With an Adaptive Drift. IEEE Transactions on Reliability, 64(2), 687-700.
Karpatne, A., Watkins, W., Read, J., & Kumar, V. (2017). Physics-Guided Neural Networks (PGNN): An Application in Lake Temperature Modeling. arXiv preprint arXiv:1710.11431, 2.
Lei, Y., Li, N., Gontarz, S., Lin, J., Radkowski, S., & Dybala, J. (2016). A Model-Based Method for Remaining Useful Life Prediction of Machinery. IEEE Transactions on Reliability, 65(3), 1314-1326.
Liu, J., Saxena, A., Goebel, K., Saha, B., & Wang, W. (2010). An Adaptive Recurrent Neural Network for Remaining Useful Life Prediction of Lithium-ion Batteries. In Annual conference of the PHM Society (Vol. 2).
Liu, L., Guo, Q., Liu, D., & Peng, Y. (2019). Data-Driven Remaining Useful Life Prediction Considering Sensor Anomaly Detection and Data Recovery. IEEE Access, 7, 58336-58345.
Ma, M., & Mao, Z. (2021). Deep-Convolution-Based LSTM Network for Remaining Useful Life Prediction. IEEE Transactions on Industrial Informatics, 17(3), 1658- 1667.
Pan, D., Li, H., & Wang, S. (2022). Transfer Learning-Based Hybrid Remaining Useful Life Prediction for Lithium-Ion Batteries Under Different Stresses. IEEE Transactions on Instrumentation and Measurement, 71, 1-10.
Sheth, P., Roychoudhury, I., Chatar, C., & Celaya, J. (2022). A Hybrid Physics-Based and Machine-Learning Approach for Stick/Slip Prediction. In SPE/IADC Drilling Conference and Exhibition.
Si, X.-S., Hu, C.-H., Chen, M.-Y., & Wang, W. (2011). An Adaptive and Nonlinear Drift-based Wiener Process for Remaining Useful Life Estimation. In 2011 Prognostics and System Health Management Conference (p. 1- 5).
Siahpour, S., Li, X., & Lee, J. (2022). A Novel Transfer Learning Approach in Remaining Useful Life Prediction for Incomplete Dataset. IEEE Transactions on Instrumentation and Measurement, 71, 1-11.
Sun, C., Ma, M., Zhao, Z., Tian, S., Yan, R., & Chen, X. (2019). Deep Transfer Learning Based on Sparse Autoencoder for Remaining Useful Life Prediction of Tool in Manufacturing. IEEE Transactions on Industrial Informatics, 15(4), 2416-2425.
Wang, Y., Zhao, Y., & Addepalli, S. (2020). Remaining Useful Life Prediction using Deep Learning Approaches: A Review. Procedia Manufacturing, 49, 81-88.
Zhang, A., Wang, H., Li, S., Cui, Y., Liu, Z., Yang, G., & Hu, J. (2018). Transfer Learning with Deep Recurrent
Neural Networks for Remaining Useful Life Estimation. Applied Sciences, 8(12).
Zhang, Y., Xiong, R., He, H., & Liu, Z. (2017). A LSTM-RNN Method for the Lithuim-ion Battery Remaining Useful Life Prediction. In 2017 Prognostics and System Health Management Conference (PHM-Harbin) (p. 1-4).
Zhang, Y., Yang, Y., Xiu, X., Li, H., & Liu, R. (2021). A Remaining Useful Life Prediction Method in the Early Stage of Stochastic Degradation Process. IEEE Transactions on Circuits and Systems II: Express Briefs, 68(6), 2027-2031.
This work is licensed under a Creative Commons Attribution 3.0 Unported License.
The Prognostic and Health Management Society advocates open-access to scientific data and uses a Creative Commons license for publishing and distributing any papers. A Creative Commons license does not relinquish the author’s copyright; rather it allows them to share some of their rights with any member of the public under certain conditions whilst enjoying full legal protection. By submitting an article to the International Conference of the Prognostics and Health Management Society, the authors agree to be bound by the associated terms and conditions including the following:
As the author, you retain the copyright to your Work. By submitting your Work, you are granting anybody the right to copy, distribute and transmit your Work and to adapt your Work with proper attribution under the terms of the Creative Commons Attribution 3.0 United States license. You assign rights to the Prognostics and Health Management Society to publish and disseminate your Work through electronic and print media if it is accepted for publication. A license note citing the Creative Commons Attribution 3.0 United States License as shown below needs to be placed in the footnote on the first page of the article.
First Author et al. This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 United States License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.