Limiting conditions prediction using machine learning for loss of condenser vacuum event
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
We implement machine learning regression models to predict peak pressures of primary and secondary systems, a major safety concern in Loss Of Condenser Vacuum (LOCV) accident. We selected the Multi-dimensional Analysis of Reactor Safety-KINS standard (MARS-KS) code to analyze the LOCV accident, and the reference plant is the Korean Optimized Power Reactor 1000MWe (OPR1000). eXtreme Gradient Boosting (XGBoost) is selected as a machine learning tool. The MARS-KS code is used to generate LOCV accident data and the data is applied to train the machine learning model. Hyperparameter optimization is performed using a simulated annealing. The randomly generated combination of initial conditions within the operating range is put into the input of the XGBoost model to predict the peak pressure. These initial conditions that cause peak pressure with MARS-KS generate the results. After such a process, the error between the predicted value and the code output is calculated. Uncertainty about the machine learning model is also calculated to verify the model accuracy. The machine learning model presented in this paper successfully identifies a combination of initial conditions that produce a more conservative peak pressure than the values calculated with existing methodologies
Abstract
We implement machine learning regression models to predict peak pressures of primary and secondary systems, a major safety concern in Loss Of Condenser Vacuum (LOCV) accident. We selected the Multi-dimensional Analysis of Reactor Safety-KINS standard (MARS-KS) code to analyze the LOCV accident, and the reference plant is the Korean Optimized Power Reactor 1000MWe (OPR1000). eXtreme Gradient Boosting (XGBoost) is selected as a machine learning tool. The MARS-KS code is used to generate LOCV accident data and the data is applied to train the machine learning model. Hyperparameter optimization is performed using a simulated annealing. The randomly generated combination of initial conditions within the operating range is put into the input of the XGBoost model to predict the peak pressure. These initial conditions that cause peak pressure with MARS-KS generate the results. After such a process, the error between the predicted value and the code output is calculated. Uncertainty about the machine learning model is also calculated to verify the model accuracy. The machine learning model presented in this paper successfully identifies a combination of initial conditions that produce a more conservative peak pressure than the values calculated with existing methodologies
참고문헌 (32)
- Safety Margins of Operating Reactors-Analysis of Uncertainties and Implications for Decision Making, "I. A. E. Agency", International Atomic Energy Agency, ~ (2003)
- A. International Atomic Energy, "Deterministic Safety Analysis for Nuclear Power Plants : Specific Safety Guide" (2019), IAEA
- Korea Non-LOCA Analysis Package, "KHNP", KHNP, ~, pp.TR-KHNP-0009 (2007)
- M.G. Genton, "Classes of kernels for machine learning: a statistics perspective", J. Mach. Learn. Res, 2(2), pp.299~312 (2002)
- H. Mannila, "Data mining: machine learning, statistics, and databases", Eighth International Conference on Scientific and Statistical Database Systems, Proceedings, pp.2~9 (1996)
- V. Gudivada, "Data quality considerations for big data and machine learning:Going beyond data cleaning and transformations", Int. J. Adv. Software, 10(1), pp.1~20 (2017)
- X. Shi, "An accident prediction approach based on XGBoost", 2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE) (2017), IEEE
- Y. Wang, "A hybrid ensemble method for pulsar candidate classification", Astrophys. Space Sci, 364(8) (2019)
- T. Chen, "XGboost: a scalable tree boosting system", Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (2016)
- S. S. Dhaliwal, "Effective intrusion detection system using XGBoost", Information, 9(7), pp.149~ (2018)
- H. T. Bang, "Application of machine learning methods to predict a thermal conductivity model for compacted bentonite", Ann. Nucl. Energy, 142, pp.107395~ (2020)
- J. Cai, "An assembly-level neutronic calculation method based on LightGBM algorithm", Ann. Nucl. Energy, 150, pp.107871~ (2021)
- Chaoliang Xu, "A study of predicting irradiation-induced transition temperature shift for RPV steels with XGBoost modeling", Nuclear Engineering and Technology, 53(8), pp.2610~2615 (2021), 한국원자력학회[DOI]
- MARS CODE MANAUAL VOLUME IV-Developmental Assessment Report, "B. D. Chung", Korea Atomic Energy Research Institute, ~ (2010)
- "NUREG-0800, Standard review plan for the review of safety analysis reports for nuclear power plants, Rev", ~ (2007)
- (Chapter 15), June, APR1400 Standard Safety Analysis Report, , ~ (2002)
- Application of a Combined Safety Approach for the Evaluation of Safety Margin during a Loss of Condenser Vacuum Event, "D. H. Shin", Nuclear Engineering and Technology, ~ (2021)
- R. Cuocolo, "Current applications of big data and machine learning in cardiology", J. Geriatr. Cardiol. : JGC, 16(8), pp.601~ (2019)
- X. Ying, "Journal of Physics:Conference Series" (2019), IOP Publishing
- A.N. Richter, "Learning curve estimation with large imbalanced datasets", 2019 18th IEEE International Conference on Machine Learning and Applications (ICMLA) (2019), IEEE
- C. D, J. R, "Statistics without Maths for Psychology", pp.175~ (2011), Prentice Hall
- A. G´omez-Ríos, "International Conference on Hybrid Artificial Intelligence Systems" (2017), Springer
- S. Sun, "A survey of optimization methods from a machine learning perspective", IEEE Trans. Cybern, 50(8), pp.3668~3681 (2019)
- S. Kirkpatrick, "Optimization by simulated annealing", Science, 220(4598), pp.671~680 (1983)
- "dmlc XGBoost stable"[URL]
- R. Poli, "Particle swarm optimization", Swarm Intell, 1(1), pp.33~57 (2007)
- RETRAN-3D–A Program for Transient Thermal-Hydraulic Analysis of Complex Fluid Flow Systems, "M. Paulsen", ~, pp.NP-7450 1(4) (1996)
- CESEC Digital Simulation of a Combustion Engineering Nuclear Steam Supply System, "Combustion Engineering Inc", ~, pp.LD-82-001 (1981)
- C. Liao, "A tolerance interval for the normal distribution with several variance components", Stat. Sin, pp.217~229 (2004)
- H. Arsham, "Bartlett’s test", Int. encycl. Stat. Sci, 1, pp.87~88 (2011)
- M. G. Vangel, "One-sided nonparametric tolerance limits", Commun. Stat. Simulat. Comput, 23(4), pp.1137~1154 (1994)
- D. S. Young, "Tolerance : an R package for estimating tolerance intervals", J. Stat. Software, 36, pp.1~39 (2010)