Case study on sampling techniques using machine learning and simplified physical model for simulation-based dynamic probabilistic risk assessment

Kubo, Kotaro; Zheng, X.; Ishikawa, Jun; Sugiyama, Tomoyuki; Jang, S.*; Takata, Takashi*; Yamaguchi, Akira*

Proceedings of Asian Symposium on Risk Assessment and Management 2020 (ASRAM 2020) (Internet), 11 Pages, 2020/11

Dynamic probabilistic risk assessment (PRA) enables a more realistic and detailed analysis than classical PRA. However, the trade-off for these improvements is the enormous computational cost associated with performing a large number of thermal-hydraulic (TH) analyses. In this study, based on machine learning (ML), we aim to reduce these costs by skipping the TH analysis. For the ML algorithm, we selected a support vector machine; we built it using a high-fidelity/high-cost detailed model and low-fidelity/low-cost simplified model. As a result, the computational costs could be reduced by approximately 80% without significantly decreasing the accuracy under the assumed conditions.

Analysis for the accident at unit 2 of the Fukushima Daiichi NPS with THALES2/KICHE code in BSAF2 project

Tamaki, Hitoshi; Ishikawa, Jun; Sugiyama, Tomoyuki; Maruyama, Yu

Proceedings of 18th International Topical Meeting on Nuclear Reactor Thermal Hydraulics (NURETH-18) (USB Flash Drive), p.100 - 111, 2019/08

JAEA participated in the OECD/NEA BSAF2 project with our integrated severe accident analysis code, THALES2/KICHE, in order to analyze and discuss the accident progression and source term of the accident at the Fukushima Daiichi NPS. One of important characteristics of THALES2/KICHE code is that it has the capability of predicting iodine chemistry based on reaction kinetics in the aqueous phase. JAEA performed the three week analysis for the accident at unit 2 on the basis of the boundary conditions and assumptions proposed by the BSAF2 project and our own assumptions. One of focusing points in the BSAF2 project was the trend of measured data of reactor vessel from 20:00 March 14 to 02:00 March 15. An assumption was made that the lower part of the suppression chamber failed to form a water leakage path. The released iodine and cesium within three weeks after the earthquake were predicted to be approximately 3% and 0.1% of the initial inventory, respectively.

Sensitivity analysis of source term in the accident of Fukushima Dai-ichi Nuclear Power Station Unit 1 using THALES2/KICHE

Tamaki, Hitoshi; Ishikawa, Jun; Sugiyama, Tomoyuki; Maruyama, Yu

Proceedings of Asian Symposium on Risk Assessment and Management 2018 (ASRAM 2018) (USB Flash Drive), 6 Pages, 2018/10

In the accidents at Fukushima Dai-ichi Nuclear Power Station, Tsunami caused loss of electric power supply and this event led to core melt and failure of Containment vessel. Finally, fission products were released to the environment. Currently, the activities for understanding of accident progressions are carried out based on the measured data during the accident, accident progression analysis using integrated severe accident analysis codes and investigation of inside of reactor buildings and containment vessels. On the other hand, there are some research activities with combination of accident progression analysis and accident consequence analysis. In Japan Atomic Energy Agency (JAEA), the research project of combination of these analyses using the computational simulation codes has been started. The results obtained from the combination analysis are expected to have broad width of uncertainty because of many uncertainty factors in this combined analysis. In order to perform the analysis efficiently, sensitivity analysis for failure location on containment vessel and its failure size were carried out by THALES2/KICHE developed by JAEA at first. This analysis was performed on unit 1, since it was the first plant to release radioactive materials to the environment during the accident and its consequence had no effect from other plants. The authors focused on the failure of containment vessel head flange, penetration seal and vacuum breaker pipe, and possibility of partial open of vent valve based on the investigations of reactor building inside performed by TEPCO. This paper presents the results obtained from this sensitivity analysis.

Source term analysis considering BC/steel interaction and oxidation during severe accidents

Ishikawa, Jun; Shiotsu, Hiroyuki; Sugiyama, Tomoyuki; Maruyama, Yu

Proceedings of 25th International Conference on Nuclear Engineering (ICONE-25) (CD-ROM), 7 Pages, 2017/07

Effectiveness evaluation of filtered containment venting system using THALES-2

Kondo, Masahiro*; Yoshimoto, Tatsuya*; Ishikawa, Jun; Okamoto, Koji*

Hozengaku, 15(4), p.79 - 85, 2017/01

https://ci.nii.ac.jp/naid/40021090097

no abstracts in English

Analysis for progression of accident at Fukushima Dai-ichi Nuclear Power Station with THALES2 code

Matsumoto, Toshinori; Ishikawa, Jun; Maruyama, Yu

Proceedings of 16th International Topical Meeting on Nuclear Reactor Thermal Hydraulics (NURETH-16) (USB Flash Drive), p.4033 - 4043, 2015/08

Influence of adsorption of molecular iodine onto aerosols on iodine source term in severe accident

Ishikawa, Jun; Ito, Hiroto; Maruyama, Yu

Proceedings of 23rd International Conference on Nuclear Engineering (ICONE-23) (DVD-ROM), 6 Pages, 2015/05

Development of source term analysis methodology considering FP chemistry

Ishikawa, Jun; Zheng, X.; Shiotsu, Hiroyuki; Sugiyama, Tomoyuki

no journal, , 

no abstracts in English

Bayesian optimization analysis of containment venting strategies in a severe accident

Zheng, X.; Ishikawa, Jun; Maruyama, Yu

no journal, , 

no abstracts in English