Nuclear Engineering and Technology

Korean Nuclear Society (한국원자력학회)
권호 표지
DOI QR코드

DOI QR Code

Identification of primary input parameters affecting evacuation in ventilated main control room through CFAST simulations and application of a machine learning algorithm to replace CFAST model

  • Sumit Kumar Singh (Department of Mechanical Engineering, Chungnam National University) ;
  • Jinsoo Bae (School of Industrial and Management Engineering, Korea University) ;
  • Yu Zhang (Department of Mechanical Engineering, Chungnam National University) ;
  • Saerin Lim (School of Industrial and Management Engineering, Korea University) ;
  • Jongkook Heo (School of Industrial and Management Engineering, Korea University) ;
  • Seoung Bum Kim (School of Industrial and Management Engineering, Korea University) ;
  • Weon Gyu Shin (Department of Mechanical Engineering, Chungnam National University)
  • Received : 2023.06.02
  • Accepted : 2024.04.13
  • Published : 2024.09.25
Download PDF
⟨ Previous Next ⟩

Abstract

Accurately predicting evacuation time in a ventilated main control room (MCR) during fire emergencies is crucial for ensuring the safety of personnel at nuclear power plants. This study proposes to use neural networks alongside consolidated fire and smoke transport (CFAST) simulations to serve as a surrogate model for physics-based simulation tools. Our neural networks can promptly predict the evacuation time in MCRs, proving to be a valuable asset in fire emergencies and eliminating the need for time-consuming rollouts of the CFAST simulations. The CFAST model simulates fire and evacuation scenarios in a ventilated MCR with variations in input parameters such as door conditions, ventilation flow rate, leakage area, and fire propagation time. Target output parameters, such as hot gas layer temperature (HGLT), heat flux (HF), and optical density (OD), are used alongside standardized evacuation variables to train a machine learning model for predicting evacuation time. The findings suggest that high ventilation flow rates help to dilute smoke and discharge hot gas, leading to lower target output parameters and quicker evacuation. Standardized evacuation variables exceed the required abandonment criteria for all door conditions, indicating the importance of proper evacuation procedures. The results show that neural networks can generate evacuation times close to those obtained from CFAST simulations.

Keywords

Acknowledgement

This work was supported by the National Research Foundation of Korea(NRF) grant funded by the Korea government (MSIT) (No. RS-2022-00144190).

References

  1. H.S. Han, J.O. Lee, C.H. Hwang, J. Kim, S. Lee, Assessment of the habitability for a cabinet fire in the main control room of nuclear power plant using sensitivity analysis, Fire Science and Engineering 31 (2017) 52-60.
  2. B.G. Kim, H.S. Lim, Y.S. Lee, M.S. Kim, Effect of the HVAC conditions on the smoke ventilation performance and habitability for a main control room fire in nuclear power plant, Fire Science and Engineering 30 (2016) 74-81.
  3. D.I. Kang, K. Kim, S.C. Jang, S.Y. Yoo, Fire simulations for the abandonment risk assessment of main control room fire in domestic nuclear power plant, J. Korean Surg. Soc. 29 (2014) 199-207.
  4. D.I. Kang, K. Kim, H.G. Lim, Evaluation logic of main control board fire risk, Transactions of the Korean Nuclear Society Spring Meeting 47 (08) (2015) 1-4.
  5. J.M. Jyung, Y.S. Chang, Electrical fire simulation in control room of an AGN reactor, Nucl. Eng. Technol. 53 (2021) 466-473.
  6. U.S. NRC, Nuclear Power Plant Fire Modeling Analysis Guidelines, NUREG-1934, United States Nuclear Regulatory Commission, Washington, D.C., 2012.
  7. M.J. Hurley, et al., SFPE Handbook of Fire Protection Engineering, fifth ed., Springer, 2015 https://doi.org/10.1007/978-1-4939-2565-0.
  8. U.S. NRC, EPRI/NRC-RES Fire PRA Methodology for Nuclear Power Facilities Detailed Methodology (No. NUREG/CR-6850, EPRI 1011989 Final Report), vol. 2, U.S. NRC-RES, 2005.
  9. Y.H. Jung, D.I. Kang, A Study on Fire Modeling of Main Control Benchboard Fire Scenarios for Evaluation of Main Control Room Habitability Conditions, Transactions of the Korean Nuclear Society Autumn Meeting, 2019, pp. 9-10.
  10. D.I. Kang, K. Kim, S.C. Jang, S.Y. Yoo, Risk assessment of main control board fire using fire dynamics simulator, Nucl. Eng. Des. 289 (2015) 195-207.
  11. A.M. Bayomy, Q. Chen, K. Podila, L. Sun, T. Beuthe, Smoke and evacuation modelling of multi-compartment building for nuclear applications, Heat Mass Tran. 58 (2022) 383-394.
  12. B. Manescau, L. Courty, L. Acherar, B. Coudour, H.Y. Wang, J.P. Garo, Effects of ventilation conditions and procedures during a fire in a reduced-scale room, Process Saf. Environ. Protect. 144 (2020) 263-272.
  13. B. Kim, J. Lee, S. Kim, W.G. Shin, Habitability evaluation considering various input parameters for main control benchboard fire in the main control room, Nucl. Eng. Technol. 54 (2022) 4195-4208.
  14. U.S.NRC-RES and EPRI, Refining and Characterizing Heat Release Rates from Electrical Enclosures during Fire (RACHELLE-FIRE), Volume 2: Fire Modeling Guidance for Electrical Cabinets, Electric Motors, Indoor Dry Transformers, and the Main Control Board, NUREG-2178 Vol.vol. 2 and EPRI 3002016052, U.S.NRC-RES and EPRI, 2019..
  15. B.Y. Lattimer, J.L. Hodges, A.M. Lattimer, Using machine learning in physics-based simulation of fire, Fire Saf. J. 114 (2020) 102991.
  16. H. Fang, S.M. Lo, Y. Zhang, Y. Shen, Development of a machine-learning approach for identifying the stages of fire development in residential room fires, Fire Saf. J. 126 (2021) 103469.
  17. N. Iqbal, M.H. Salley, "Fire Dynamics Tools (FDTs); Quantitative Fire Hazard Analysis Methods for the U.S. Nuclear Regulatory Commission Fire Protection Inspection Program", 2004. NUREG-1805, U.S.NRC,.
  18. U.S. NRC, Verification and Validation of Selected Fire Models for Nuclear Power Plant Applications-Final Report (No. NUREG-1824, Supplement 1), 2016.
  19. D.P. Kingma, J. Ba, Adam: A Method for Stochastic Optimization, 2014 arXiv preprint arXiv:1412.6980.
  20. Alex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton, Imagenet classification with deep convolutional neural networks, Adv. Neural Inf. Process. Syst. 25 (2012).
  21. Ilya Sutskever, Oriol Vinyals, Quoc V. Le, Sequence to sequence learning with neural networks, Adv. Neural Inf. Process. Syst. 27 (2014).
  22. Volodymyr Mnih, et al., Human-level control through deep reinforcement learning, Nature 518 (7540) (2015) 529-533.
  23. Claudio Angione, Eric Silverman, Elisabeth Yaneske, Using machine learning as a surrogate model for agent-based simulations, PLoS One 17 (2) (2022) e0263150.
  24. Adam Paszke, et al., Pytorch: an imperative style, high-performance deep learning library, Adv. Neural Inf. Process. Syst. 32 (2019).