본문 바로가기
KERD
논문KCI 등재

A reliable intelligent diagnostic assistant for nuclear power plants using explainable artifi cial intelligence of GRU-AE, LightGBM and SHAP

이상현|조혜선|오상원|박지훈|나만균
한국원자력학회 (Korean Nuclear Society)|2022.04|Nuclear Engineering and Technology|제54권 제4호|pp. 1271-1287|DOI https://doi.org/10.1016/j.net.2021.10.024

2

조회

0

다운로드

5

인용

0

참고문헌

키워드

Explainable artificial intelligenceDiagnostic algorithmAbnormal operating conditionsDiagnostic tasksReliability of artificial intelligenceReliable intelligent diagnostic assistant
주제분류원자력공학

초록

When abnormal operating conditions occur in nuclear power plants, operators must identify theoccurrence cause and implement the necessary mitigation measures. Accordingly, the operator mustrapidly and accurately analyze the symptom requirements of more than 200 abnormal scenarios fromthe trends of many variables to perform diagnostic tasks and implement mitigation actions rapidly. However, the probability of human error increases owing to the characteristics of the diagnostic tasksperformed by the operator. Researches regarding diagnostic tasks based on Artificial Intelligence (AI)have been conducted recently to reduce the likelihood of human errors; however, reliability issues due tothe black box characteristics of AI have been pointed out. Hence, the application of eXplainable ArtificialIntelligence (XAI), which can provide AI diagnostic evidence for operators, is considered. In conclusion,the XAI to solve the reliability problem of AI is included in the AI-based diagnostic algorithm. A reliableintelligent diagnostic assistant based on a merged diagnostic algorithm, in the form of an operatorsupport system, is developed, and includes an interface to efficiently inform operators.

Abstract

When abnormal operating conditions occur in nuclear power plants, operators must identify theoccurrence cause and implement the necessary mitigation measures. Accordingly, the operator mustrapidly and accurately analyze the symptom requirements of more than 200 abnormal scenarios fromthe trends of many variables to perform diagnostic tasks and implement mitigation actions rapidly. However, the probability of human error increases owing to the characteristics of the diagnostic tasksperformed by the operator. Researches regarding diagnostic tasks based on Artificial Intelligence (AI)have been conducted recently to reduce the likelihood of human errors; however, reliability issues due tothe black box characteristics of AI have been pointed out. Hence, the application of eXplainable ArtificialIntelligence (XAI), which can provide AI diagnostic evidence for operators, is considered. In conclusion,the XAI to solve the reliability problem of AI is included in the AI-based diagnostic algorithm. A reliableintelligent diagnostic assistant based on a merged diagnostic algorithm, in the form of an operatorsupport system, is developed, and includes an interface to efficiently inform operators.

참고문헌 (31)

  1. "Operational Performance Information System for Nuclear Power Plant, Nuclear Accident and Failure Status" (2020)[URL]
  2. 김재민, "Abnormality diagnosis model for nuclear power plants using two-stage gated recurrent units", Nuclear Engineering and Technology, 52(9), pp.2009~2016 (2020), 한국원자력학회[DOI]
  3. J. Yang, "Accident diagnosis algorithm with untrained accident identification during power-increasing operation", Reliab. Eng. Syst. Saf, 202, pp.107032~ (2020)
  4. H. Kim, "Development of a diagnostic algorithm for abnormal situations using long short-term memory and variational autoencoder", Ann. Nucl. Energy, 153, pp.108077~ (2021)
  5. 유쾌환, "Smart support system for diagnosing severe accidents in nuclear power plants", Nuclear Engineering and Technology, 50(4), pp.562~569 (2018), 한국원자력학회
  6. Abiodun Ayodeji, "Support vector ensemble for incipient fault diagnosis in nuclear plant components", Nuclear Engineering and Technology, 50(8), pp.1306~1313 (2018), 한국원자력학회
  7. A. Ayodeji, "Knowledge base operator support system for nuclear power plant fault diagnosis", Prog. Nucl. Energy, 105, pp.42~50 (2018)
  8. Y. -k. Liu, "A cascade intelligent fault diagnostic technique for nuclear power plants", J. Nucl. Sci. Technol, 55(3), pp.254~266 (2018)
  9. Min-jun Peng, "An intelligent hybrid methodology of on-line system-level fault diagnosis for nuclear power plant", Nuclear Engineering and Technology, 50(3), pp.396~410 (2018), 한국원자력학회
  10. R. L. Boring, "Computerized operator support systems to aid decision making in nuclear power plants", Procedia Manuf, 3, pp.5261~5268 (2015)
  11. M. A. Kramer, "Nonlinear principle component analysis using autoassociative neural networks", AIChE J, 37(2), pp.233~243 (1991)
  12. M. Sakurada, "Anomaly detection using autoencoders with nonlinear dimensionality reduction", Proceedings of the MLSDA 2014 2nd Workshop on Machine Learning for Sensory Data Analysis, pp.4~11 (2014)
  13. R. Chalapathy, "Anomaly Detection Using One-Class Neural Networks", arXiv preprint arXiv: 1802.06360, ~ (2018)
  14. S. Hochreiter, "Long short-term memory", Neural Comput, 9(8), pp.1735~1780 (1997)
  15. K. Cho, "Learning Phrase Representations Using RNN Encoder-Decoder for Statistical Machine Translation", arXiv preprint arXiv : 1406. 1078, ~ (2014)
  16. J. H. Friedman, "Greedy function approximation : a gradient boosting machine", Ann. Stat, 29(5), pp.1189~1232 (2001)
  17. J. H. Friedman, "Stochastic gradient boosting", Comput. Stat. Data Anal, 38(4), pp.367~378 (2002)
  18. Ke Guolin, "LightGBM: a highly efficient gradient boosting decision tree", Proc. Advances in neural information processing systems, pp.3146~3154 (2017)
  19. A.B. Arrieta, "Explainable Artificial Intelligence (XAI): concepts, taxonomies, opportunities and challenges toward responsible AI", Inf. Fusion, 58, pp.82~115 (2020)
  20. J.S. Bridle, "Probabilistic interpretation of feedforward classification network outputs, with relationships to statistical pattern recognition", Neurocomputing, pp.227~236 (1990)
  21. Explainable Artificial Intelligence (XAI), "D. Gunning", Defense Advanced Research Projects Agency, ~ (2017)
  22. S. M. Lundberg, "Consistent Individualized Feature Attribution for Tree Ensembles", arXiv preprint arXiv : 1802. 03888, ~ (2018)
  23. S. M. Lundberg, "From local explanation to global understanding with explainable AI for trees", Nat Mach Intell, 2(1), pp.2522~5839 (2020)
  24. L. S. Shapley, "The Shapley Value: Essays in Honor of Lloyd" (1988), Cambridge University Press
  25. Hayes-Roth, Frederick, "Rule-based systems", Commun. ACM, 28(9), pp.921~932 (1985)
  26. Equipment and Performance Upgrade of Compact Nuclear Simulator, "J. -c. Park", KAERI, ~, pp.KAERI (1999), RR-1967[DOI]
  27. J. Miettinen, "Development and assessment of the SBLOCA code SMABRE", Proceedings of the CSNI Specialists' Meeting on Small Break LOCA Analyses in LWRs, pp.481~495 (1985)
  28. C. Tang, "An efficient intrusion detection method based on LightGBM and autoencoder", Symmetry, 12(9), pp.1458~ (2020)
  29. H. S. Hota, "Time series data predicting using sliding window based RBF neural network", Int. J. Comput. Intell. Res, 13(5), pp.1145~1156 (2017)
  30. 박성호, "Receiver Operating Characteristic (ROC) Curve: Practical Review for Radiologists", Korean Journal of Radiology, 5(1), pp.11~18 (2004), 대한영상의학회
  31. J. N. Mandrekar, "Receiver operating characteristic curve in diagnostic test assessment", J. Thorac. Oncol, 5(9), pp.1315~1316 (2010)

관련 논문AI추천

Development of an AI-based remaining trip time prediction system for nuclear power plants나만균, 박지훈, 오상원, 조혜선한국원자력학회 · Nuclear Engineering and Technology2024 · 제56권 제8호Explainable machine-learning tools for predictive maintenance of circulating water systems in nuclear power plantsAgarwal Vivek, Lin Linyu, Walker Cody한국원자력학회 · Nuclear Engineering and Technology2025 · 제57권 제9호Concept of understandable diagnostic cause visualization with explainable AI and multilevel flow modeling강정성, 김재민, 신지현, 이승준한국원자력학회 · Nuclear Engineering and Technology2025 · 제57권 제8호Consistency check algorithm for validation and re-diagnosis to improve the accuracy of abnormality diagnosis in nuclear power plants김근희, 김재민, 신지현, 이승준한국원자력학회 · Nuclear Engineering and Technology2022 · 제54권 제10호Approach to diagnosing multiple abnormal events with single-event training data구서룡, 신지현, 이승준, 조승규한국원자력학회 · Nuclear Engineering and Technology2024 · 제56권 제2호Long-term prediction of safety parameters with uncertainty estimation in emergency situations at nuclear power plants김종현, 김효진한국원자력학회 · Nuclear Engineering and Technology2023 · 제55권 제5호

이 간행물의 다른 논문 (15건)

Classification of ultrasonic signals of thermally aged cast austenitic stainless steel (CASS) using machine learning (ML) models장창희, 강성식, 김진겸pp. 1167-1174Fractal kinetic characteristics of uranium leaching from low permeability uranium-bearing sandstoneKaixuan Tan, Sheng Zeng, Shuwen Zhang, Wenhao Ye, Yuan Shen, Bing Sunpp. 1175-1184Adsorption behaviour of film-forming amine on pre-oxidized carbon steel surfaceLiu Canshuai, Fang Jun, Song Lijun, Lin Genxian, Liu Bin, Sun Yunpp. 1185-1194Study of concrete de-bonding assessment technique for containment liner plates in nuclear power plants using ultrasonic guided wave approach이용희, 조윤호, 윤현민pp. 1221-1229A study on classification of the security controls for the effective implementation to nuclear power plant채영호, 성풍현, 한상민, 이찬영pp. 1245-1252
원문 보기