Real-time Estimation of Break Sizes during LOCA in Nuclear Power Plants Using NARX Neural Network

다국어 초록 (Multilingual Abstract)

This paper deals with break size estimation of loss of coolant accidents (LOCA) using a nonlinearautoregressive with exogenous inputs (NARX) neural network. Previous studies used static approaches,requiring time-integrated parameters and independent firing algorithms. NARX neural network is able todirectly deal with time-dependent signals for dynamic estimation of break sizes in real-time. The casestudied is a LOCA in the primary system of Bushehr nuclear power plant (NPP). In this study, number ofhidden layers, neurons, feedbacks, inputs, and training duration of transients are selected by performingparametric studies to determine the network architecture with minimum error. The developed NARXneural network is trained by error back propagation algorithm with different break sizes, covering 5%e100% of main coolant pipeline area. This database of LOCA scenarios is developed using RELAP5thermal-hydraulic code. The results are satisfactory and indicate feasibility of implementing NARX neuralnetwork for break size estimation in NPPs. It is able to find a general solution for break size estimationproblem in real-time, using a limited number of training data sets. This study has been performed in theframework of a research project, aiming to develop an appropriate accident management support tool forBushehr NPP.

참고문헌 (Reference)

1 L. Tsungnan, "What to remember: how memory order affects the performance of NARX neural networks" 2 : 1051-1056, 1998

2 Takeshi Takeda, "Uncertainty analysis of ROSA/LSTF test by RELAP5 code and PKL counterpart test concerning PWR hot leg break LOCAs" 한국원자력학회 50 (50): 829-841, 2018

3 A. Petruzzi, "Thirty years’ experience in RELAP5 applications at GRNSPG & NINE" 193 : 47-87, 2016

4 S. H. Park, "Smart sensing of the RPV water level in NPP severe accidents using a GMDH algorithm" 61 : 931-938, 2014

5 EPRI, "Severe Accident Management Guidance Technical Basis Report, Volumes 2: The Physics of Accident Progression" 2012

6 Z. -Q. Wang, "Remaining useful life prediction of degrading systems subjected to imperfect maintenance : application to draught fans" 100 : 802-813, 2018

7 Takeshi Takeda, "ROSA/LSTF test and RELAP5 code analyses on PWR 1% vessel upper head small-break LOCA with accident management measure based on core exit temperature" 한국원자력학회 50 (50): 1412-1420, 2018

8 D. Y. Kim, "Prediction of leak flow rate using fuzzy neural networks in severe post-LOCA circumstances" 61 : 3644-3652, 2014

9 D. Roverso, "Plant diagnostics by transient classification : the aladdin approach" 17 : 767-790, 2002

10 L. Tsungnan, "Learning long-term dependencies in NARX recurrent neural networks" 7 : 1329-1338, 1996

11 M. Saghafi, "Introduction of a research project on development of accident management support tool for BNPP (WWER-1000) based on the lessons learned from Fukushima accident" 2015

12 M. Boroushaki, "Identification of a nuclear reactor core(VVER)using recurrent neural networks" 29 : 1225-1240, 2002

13 M. Marseguerra, "Identification of a line break by a neural network methodology" 21 : 249-258, 1994

14 M. G. Na, "Estimation of break location and size for loss of coolant accidents using neural networks" 232 : 289-300, 2004

15 S. H. Lee, "Diagnostics of loss of coolant accidents using SVC and GMDH models" 58 : 267-276, 2011

16 T. V. Santosh, "Diagnostic system for identification of accident scenarios in nuclear power plants using artificial neural networks" 94 : 759-762, 2009

17 M. Saghafi, "Development and qualification of a thermal-hydraulic nodalization for modeling station blackout accident in PSBVVER test facility" 303 : 109-121, 2016

18 M. Saghafi, "Determination of PAR configuration for PWR containment design : a hydrogen mitigation strategy" 42 : 7104-7119, 2017

19 M. G. Na, "Detection and diagnostics of loss of coolant accidents using support vector machines" 55 : 628-636, 2008

20 J.M. Izquierdo, "Current Status and Applications of Integrated Safety Assessment and Simulation Code System for ISA" 한국원자력학회 49 (49): 295-305, 2017

21 S. J. Lee, "Computational Intelligence in Nuclear Applications : lessons Learned and Recent DevelopmentsA dynamic neural network based accident diagnosis advisory system for nuclear power plants" 46 : 268-281, 2005

22 P. Baraldi, "Clustering for unsupervised fault diagnosis in nuclear turbine shut-down transients" 58-59 : 160-178, 2015

23 M. Saghafi, "Application of FFTBM with signal mirroring to improve accuracy assessment of MELCOR code" 308 : 238-251, 2016

24 이성욱, "Analysis of Uncertainty Quantification Method by Comparing Monte-Carlo Method and Wilks' Formula" 한국원자력학회 46 (46): 481-488, 2014

25 M. Saghafi, "Accident management support tools in nuclear power plants : a post-Fukushima review" 92 : 1-14, 2016

26 K. Mo, "A dynamic neural network aggregation model for transient diagnosis in nuclear power plants" 49 : 262-272, 2007