Nuclear Engineering and Technology

Korean Nuclear Society (한국원자력학회)
권호 표지
DOI QR코드

DOI QR Code

A machine learning informed prediction of severe accident progressions in nuclear power plants

  • JinHo Song (Department of Nuclear Engineering, Hanyang University) ;
  • SungJoong Kim (Department of Nuclear Engineering, Hanyang University)
  • Received : 2023.11.06
  • Accepted : 2024.01.24
  • Published : 2024.06.25
Download PDF
⟨ Previous Next ⟩

Abstract

A machine learning platform is proposed for the diagnosis of a severe accident progression in a nuclear power plant. To predict the key parameters for accident management including lost signals, a long short term memory (LSTM) network is proposed, where multiple accident scenarios are used for training. Training and test data were produced by MELCOR simulation of the Fukushima Daiichi Nuclear Power Plant (FDNPP) accident at unit 3. Feature variables were selected among plant parameters, where the importance ranking was determined by a recursive feature elimination technique using RandomForestRegressor. To answer the question of whether a reduced order ML model could predict the complex transient response, we performed a systematic sensitivity study for the choices of target variables, the combination of training and test data, the number of feature variables, and the number of neurons to evaluate the performance of the proposed ML platform. The number of sensitivity cases was chosen to guarantee a 95 % tolerance limit with a 95 % confidence level based on Wilks' formula to quantify the uncertainty of predictions. The results of investigations indicate that the proposed ML platform consistently predicts the target variable. The median and mean predictions were close to the true value.

Keywords

Acknowledgement

This work was supported by the National Research Foundation of Korea (NRF) grant (No. RS-2022-00144202) and by the Innovative Small Modular Reactor Development Agency grant (No. RS-2023-00259516) funded by the Korean government.

References

  1. IAEA, The Fukushima Daiichi Accident, 1 - 4, 2015.
  2. TEPCO, Information Portal for the Fukushima Daiichi Accident Analysis and Decommissioning Activities, 2022. https://fdada.info/en/home2/. (Accessed 31 August 2023).
  3. USNRC, ML11171A416 - Westinghouse AP1000 Design Control Document Rev. 19 - Tier 2, 2011 (Chapter 19) - Probabilistic Risk Assessment - Appendix 19D Equipment Survivability Assessment (36 page(s), 6/13/2011, https://www.nrc.gov/docs/ML1117/ML11171A500.html. (Accessed 31 August 2023).
  4. V. Sobes, B. Hiscox, E. Popov, et al., AI-based design of a nuclear reactor core, Sci. Rep. 11 (2021) 19646, https://doi.org/10.1038/s41598-021-98037-1.
  5. G.P. Choi, et al., Estimation of LOCA break size using cascaded fuzzy neural networks, Nucl. Eng. Technol. 49 (3) (2017) 495-503. https://doi.org/10.1016/j.net.2016.11.001
  6. M.I. Radaideh, et al., Neural-based time series forecasting of loss of coolant accidents in nuclear power plants, Expert Syst. Appl. 160 (2020) 113699, https://doi.org/10.1016/j.eswa.2020.113699.
  7. Y.H. Chae, S.G. Kim, H. Kim, J.T. Kim, P.H. Seong, A methodology for diagnosing FAC induced pipe thinning using accelerometers and deep learning models, Ann. Nucl. Energy 143 (2020) 107501.
  8. J. Bae, J.W. Park, S.J. Lee, Limit surface/states searching algorithm with a deep neural network and Monte Carlo dropout for nuclear power plant safety assessment, Appl. Soft Comput. 124 (2022) 109007.
  9. Y. Chae, C. Lee, S. Han, P. Seong, Graph neural network based multiple accident diagnosis in nuclear power plants: data optimization to represent the system configuration, Nucl. Eng. Technol. (2022) 54.
  10. G. Lee, S.J. Lee, C. Lee, A convolutional neural network model for abnormality diagnosis in a nuclear power plant, Appl. Soft Comput. 99 (2021).
  11. S. Ryu, B. Jeon, H. Seo, M. Lee, J.-W. Shin, Y. Yu, Development of Deep Autoencoder-Based Anomaly Detection System for HANARO, Nuclear Engineering and Technology, 2022.
  12. S. Ryu, H. Kim, S.G. Kim, K. Jin, J. Cho, J. Park, Probabilistic deep learning model as a tool for supporting the fast simulation of a thermal-hydraulic code, Expert Syst. Appl. (2022) 200.
  13. J. She, et al., Diagnosis and prediction for loss of coolant accidents in nuclear power plants using deep learning methods, Front. Energy Res. (2021), https://doi.org/10.3389/fenrg.2021.665262.
  14. J.H. Shin, J.M. Kim, S.J. Lee, Abnormal state diagnosis model tolerant to noise in plant data, Nucl. Eng. Technol. 53 (2021) 1181-1188. https://doi.org/10.1016/j.net.2020.09.025
  15. J.S. Kang, S.J. Lee, Concept of an intelligent operator support system for initial emergency responses in nuclear power plants, Nucl. Eng. Technol. 54 (2022) 2453-2466. https://doi.org/10.1016/j.net.2022.02.010
  16. K. Hossny, W. Villanueva, H.D. Wang, Distinctive physical insights driven from machine learning modelling of nuclear power plant severe accident scenario propagation, Sci. Rep. 13 (2023) 930, https://doi.org/10.1038/s41598-023-28205-y.
  17. J.H. Song, K.S. Ha, A simulation and machine learning informed diagnosis of the severe accidents, Nucl. Eng. Des. 395 (2022) 111881, https://doi.org/10.1016/j.nucengdes.2022.111881.
  18. S. Hochreiter, J. Schmidhuber, Long short-term memory, Neural Comput. 9 (8) (1997) 1735-1780. https://doi.org/10.1162/neco.1997.9.8.1735
  19. B. Lindemann, et al., A survey on long short-term memory networks for time series prediction, Procedia CIRP 99 (2021) 650-655, https://doi.org/10.1016/j.procir.2021.03.088.
  20. A. Sagheer, M. Kotb, Time series forecasting of petroleum production using deep LSTM recurrent networks, Neurocomputing 323 (2019) 203-213, https://doi.org/10.1016/j.neucom.2018.09.082.
  21. SNL, SAND2018-13560 O, MELCOR Computer Code Manuals 2 (2018). Reference Manual Version 2.2.11932.
  22. M. Pellegrini, et al., Main findings, remaining Uncertainties, and lessons learned from the OECD/NEA BSAF project, Nucl. Technol. (2020), https://doi.org/10.1080/00295450.2020.1724731.
  23. T. Sevon, A melcor model of Fukushima Daiichi unit 3 accident, Nucl. Eng. Des. 284 (2015) 80-90, https://doi.org/10.1016/j.nucengdes.2014.11.038.
  24. S.I. Kim, et al., Analysis of Fukushima unit 2 accident considering the operating conditions of RCIC system, Nucl. Eng. Des. 298 (2016) 183-191. https://doi.org/10.1016/j.nucengdes.2015.12.024
  25. S. He, et al., A deep-learning reduced-order model for thermal hydraulic characteristics rapid estimation of steam generators, Int. J. Heat Mass Tran. 198 (2022) 123424.
  26. N.W. Porter, Wilks' formula applied to computational tools: a practical discussion and verification, Ann. Nucl. Energy 133 (2019) 129-137. https://doi.org/10.1016/j.anucene.2019.05.012
  27. Z. Che, S. Purushotham, K. Cho, et al., Recurrent neural networks for multivariate time series with missing values, Sci. Rep. 8 (2018) 6085, https://doi.org/10.1038/s41598-018-24271-9.
  28. F. Emmert-Streib, et al., An introductory review of deep learning for prediction models with big data, Front. Artif. Intell. (2020), https://doi.org/10.3389/frai.2020.00004.
  29. M. Abadi, et al., Tensorflow: Large-Scale Machine Learning on Heterogeneous Distributed Systems, 2016 arXiv preprint arXiv:1603.04467.
  30. Diederik P. Kingma, Jimmy Ba. Adam: A Method for Stochastic Optimization, 2015. https://doi.org/10.48550/arXiv.1412.6980. Published as a conference paper at ICLR 2015.
  31. M. Hosoda, S. Tokonami, H. Tazoe, et al., Activity concentrations of environmental samples collected in Fukushima Prefecture immediately after the Fukushima nuclear accident, Sci. Rep. 3 (2013) 2283, https://doi.org/10.1038/srep02283.