Validating Condition-Based Maintenance Algorithms Through Simulation

Authors: Marcel Chevalier, Léo Dupont, Sylvain Marié, Frédérique Roffet, Elena Stolyarova, William Templier, Costin Vasile

Abstract:

Industrial end users are currently facing an increasing need to reduce the risk of unexpected failures and optimize their maintenance. This calls for both short-term analysis and long-term ageing anticipation. At Schneider Electric, we tackle those two issues using both Machine Learning and First Principles models. Machine learning models are incrementally trained from normal data to predict expected values and detect statistically significant short-term deviations. Ageing models are constructed from breaking down physical systems into sub-assemblies, then determining relevant degradation modes and associating each one to the right kinetic law. Validating such anomaly detection and maintenance models is challenging, both because actual incident and ageing data are rare and distorted by human interventions, and incremental learning depends on human feedback. To overcome these difficulties, we propose to simulate physics, systems and humans – including asset maintenance operations – in order to validate the overall approaches in accelerated time and possibly choose between algorithmic alternatives.

Keywords: Degradation models, ageing, anomaly detection, soft sensor, incremental learning.

References:

[1] Chevalier, M., S. Marié, B. Boguslawski, M. Cercueil, F. Chupot, A. Vignon, and W. Youssef. 2021. “Combining First Principles and Machine Learning for optimal maintenance of electrical assets”. In CIGI QUALITA 2021. May 2021, Grenoble, France.
[2] Curreri F., G. Fiumara, and M. Xibilia. 2020. “Input Selection Methods for Soft Sensor Design: A Survey”. Future Internet vol. 12, no. 6: 97. https://doi.org/10.3390/fi12060097.
[3] Delange, M., R. Aljundi, M. Masana, S. Parisot, X. Jia, A. Leonardis, ... and T. Tuytelaars. 2021. “A continual learning survey: Defying forgetting in classification tasks”. IEEE Transactions on Pattern Analysis and Machine Intelligence. doi: 10.1109/TPAMI.2021.3057446
[4] Crawley D. B. and K. L. Lawrie. 2019. “Should We Be Using Just ‘Typical’ Weather Data in Building Performance Simulation?”. In 16th IBPSA Conference, 2019. https://climate.onebuilding.org
[5] Gao, T., B. Boguslawski, S. Marié, P. Béguery, S. Thebault, and S. Lecoeuche. 2019. “Data mining and data-driven modelling for Air Handling Unit fault detection”. In E3S Web Conf., Volume 111, CLIMA 2019 Congress. Bucharest, Romania.
[6] He, Y. and B. Sick. 2021. “CLeaR: An adaptive continual learning framework for regression tasks”. AI Perspectives vol. 3(1), pp. 1-16.
[7] Kadlec, P., R. Grbić and B. Gabrys. 2011. “Review of adaptation mechanisms for data-driven soft sensors”. Computers & chemical engineering vol.35(1), pp. 1-24.
[8] Kegel. L, M. Hahmann, and W. Lehner. 2017. “Generating What-If Scenarios for Time Series Data”. In Proceedings of the 29th International Conference on Scientific and Statistical Database Management (SSDBM '17). Association for Computing Machinery, New York, NY, USA, Article 3, pp. 1–12.
[9] Masterpact MTZ Maintenance guide, 2015 https://www.se.com/ww/en/download/document/0613IB1202/
[10] Parisi, G. I., R. Kemker, J. L. Part, C. Kanan and S. Wermter. 2019. “Continual lifelong learning with neural networks: A review”. Neural Networks vol. 113, pp. 54-71. ISSN 0893-6080
[11] Rizzi S. (2009) What-If Analysis. In: LIU L., ÖZSU M.T. (eds) Encyclopedia of Database Systems. Springer, Boston, MA. https://doi.org/10.1007/978-0-387-39940-9_466
[12] Schneider Electric, 2021, https://www.se.com/ww/en/work/services/ service-plan/ecostruxure-service-plan.jsp
[13] Souza, F. A., R. Araújo and J. Mendes. 2016. “Review of soft sensor methods for regression applications”. Chemometrics and Intelligent Laboratory Systems vol. 152, pp. 69-79.
[14] Carino, J. A., et al. 2018. "Fault Detection and Identification Methodology Under an Incremental Learning Framework Applied to Industrial Machinery," in IEEE Access, vol. 6, pp. 49755-49766.
[15] Yu, Y., Peng, M., Wang, H., Ma, Z., Cheng, S., & Renyi, X. 2021. “A continuous learning monitoring strategy for multi-condition of nuclear power plant”. Annals of Nuclear Energy, 164, 108544.