End-to-end discriminative representation learning for fault diagnosis in safety-critical time series

Dong, F.*; Xiao, Y.*; Chen, S.*; Demachi, Kazuyuki*; Takaya, Shigeru; Yoshikawa, Masanori

Advanced Engineering Informatics, 69(Part D), p.104094_1 - 104094_23, 2026/01

https://doi.org/10.1016/j.aei.2025.104094

Improvement of a reinforcement learning model to assist operators in controlling a nuclear power plant under abnormal operating conditions

Yoshikawa, Masanori; Seki, Akiyuki*; Okita, Shoichiro; Takaya, Shigeru; Yan, X.

Nuclear Engineering and Design, 444, p.114350_1 - 114350_9, 2025/12

https://doi.org/10.1016/j.nucengdes.2025.114350

Multiple knowledge exploration functions for advanced reactor design and safety in knowledge management system implemented in the ARKADIA

Seki, Akiyuki; Kondo, Yuki; Hashidate, Ryuta; Yoshikawa, Masanori; Yokoyama, Kenji; Takaya, Shigeru; Enuma, Yasuhiro; Hazama, Taira; Wakai, Takashi; Asayama, Tai

Proceedings of 31st International Conference on Nuclear Engineering (ICONE31) (Internet), 7 Pages, 2024/11

https://doi.org/10.1115/ICONE31-131784

Development of a surrogate system of a plant dynamics simulation model and an abnormal situation identification system for nuclear power plants using deep neural networks

Seki, Akiyuki; Yoshikawa, Masanori; Nishinomiya, Ryota*; Okita, Shoichiro; Takaya, Shigeru; Yan, X.

Nuclear Technology, 210(6), p.1003 - 1014, 2024/06

https://doi.org/10.1080/00295450.2023.2273566

Two types of deep neural network (DNN) systems have been constructed with the intent to assist safety operation of a nuclear power plant. One is a surrogate system (SS) that can estimate physical quantities of a nuclear power plant in a computational time of several orders less than a physical simulation model. The other is an abnormal situation identification system (ASIS) that can estimate the state of the disturbance causing an anomaly from physical quantities of a nuclear power plant. Both systems are trained and tested using data obtained from the analytical code for incore and plant dynamics (ACCORD), which reproduces the steady and dynamic behavior of the actual high Temperature engineering test reactor (HTTR) under various scenarios. The DNN models are built by adjusting, the main hyperparameters. Through these procedures, these systems are shown able to perform with a high degree of accuracy.

Proposal of a novel AI-based plant operator support system for the safety of nuclear power plants

Takaya, Shigeru; Seki, Akiyuki; Yoshikawa, Masanori; Sasaki, Naoto*; Yan, X.

Mechanical Engineering Journal (Internet), 11(2), p.23-00408_1 - 23-00408_11, 2024/04

https://doi.org/10.1299/mej.23-00408

Enhancing the ability to manage abnormal situations is important for improvement of the safety of nuclear power plants. It is necessary to investigate potential risks thoroughly in advance, and prepare countermeasures against the identified risks. In case of an occurrence of an abnormal situation, plant operators are required to recognize the plant situation promptly and select a suitable countermeasure. This study develops a novel plant operator support system designed not only to estimate details of anomalies in a plant but also propose countermeasures adaptively by employing several AI technologies of deep neural network and reinforcement learning. The design and performance of the proposed system is illustrated using High Temperature engineering Test Reactor operated in Japan Atomic Energy Agency.

Comparison between fracture mechanics evaluation methods in ASME Boiler & Pressure Vessel Code, section XI and those in JSME leak-before-break evaluation guidelines for sodium-cooled fast reactors

Yada, Hiroki; Takaya, Shigeru; Machida, Hideo*

Proceedings of the ASME 2023 Pressure Vessels and Piping Conference (PVP 2023) (Internet), 8 Pages, 2023/09

ASME Boiler and Pressure Vessel code (BPVC), Section XI, Division 2 provides requirements for protecting passive components that affect reliability of the plant. It generally consists of technology-neutral common requirements, and additional ones for individual reactor types. Currently, an Appendix for sodium-cooled fast reactors (SFRs) is being developed based on Code Case N-875. In the Code Case, continuous leakage monitoring was employed as inspection method for components retaining liquid sodium. It is also important to introduce leak-before-break (LBB) assessment procedures in the Appendix because demonstration of LBB is necessary to show the adequacy of applying continuous leakage monitoring to the component of interest. However, LBB assessment method is not provided in ASME BPVCs. On the other hand, recently, LBB assessment guidelines for SFRs has been developed by the Japan Society of Mechanical Engineers (JSME). It could be used to prepare LBB assessment procedures for the Appendix, but it needs to confirm the consistency with ASME BPVC Sec. XI. In this study, fracture evaluation methods for pipes with through-wall crack are compared between JSME LBB assessment guidelines and applicable evaluation method in ASME BPVC Sec. XI, Div. 1.

Attention-based time series analysis for data-driven anomaly detection in nuclear power plants

Dong, F.*; Chen, S.*; Demachi, Kazuyuki*; Yoshikawa, Masanori; Seki, Akiyuki; Takaya, Shigeru

Nuclear Engineering and Design, 404, p.112161_1 - 112161_15, 2023/04

https://doi.org/10.1016/j.nucengdes.2023.112161

A Weakly supervised time series analysis framework for anomaly detection in nuclear power plants

Dong, F.*; Chen, S.*; Demachi, Kazuyuki*; Yoshikawa, Masanori; Seki, Akiyuki; Takaya, Shigeru

Proceedings of 29th International Conference on Nuclear Engineering (ICONE 29) (Internet), 7 Pages, 2022/08

https://doi.org/10.1115/ICONE29-91609

Development of ARKADIA for the innovation of advanced nuclear reactor design process (Overview of optimization process development in design optimization support tool, ARKADIA-Design)

Tanaka, Masaaki; Doda, Norihiro; Yokoyama, Kenji; Mori, Takero; Okajima, Satoshi; Hashidate, Ryuta; Yada, Hiroki; Oki, Shigeo; Miyazaki, Masashi; Takaya, Shigeru

Dai-26-Kai Doryoku, Enerugi Gijutsu Shimpojiumu Koen Rombunshu (Internet), 5 Pages, 2022/07

https://doi.org/10.1299/jsmepes.2022.26.A221

To assist conceptual studies of various reactor systems conducted by private sectors in nuclear power innovation, development of an innovative design system named ARKADIA (Advanced Reactor Knowledge- and AI-aided Design Integration Approach through the whole plant lifecycle) is undergoing to achieve the design of an advanced nuclear reactor as a safe, economic, and sustainable carbon-free energy source. In this paper, focusing on the ARKADIA-Design as a part of it, the progress in the development of optimization processes on the representative problems in the fields of the core design, the plant structure design, and the maintenance schedule planning are introduced.

Fundamental study on scheduling of inspection process for fast reactor plants

Suzuki, Masaaki*; Ito, Mari*; Hashidate, Ryuta; Takahashi, Keita; Yada, Hiroki; Takaya, Shigeru

2020 9th International Congress on Advanced Applied Informatics (IIAI-AAI 2020), p.797 - 801, 2021/07

https://doi.org/10.1109/IIAI-AAI50415.2020.00157