Prediction of plutonium content of BWR/MOX fuel using a neural network

Orgonasi, C.; Nishihara, Kenji; Luzieux, C.*; Abe, Takumi; Okamura, Tomohiro*; Nakase, Masahiko*

Annals of Nuclear Energy, 240, p.112713_1 - 112713_7, 2027/01

https://doi.org/10.1016/j.anucene.2026.112713

This study addresses the inaccuracy of NMB fuel cycle simulator due to the one-group cross-section as well as the computational burden by developing and integrating an Multi-Layer Perceptron (MLP) model into NMB. A custom PyTorch MLP was trained on a dataset of 5,000 depletion calculations generated with the SERPENT code for an ATRIUM-10 boiling water reactor geometry. The developed MLP, in conjunction with a dichotomy search algorithm, determined the plutonium content with an error below 1% and achieved calculation speed of approximately 60 times for determination of Pu enrichment and 4 times for overall calculation.

Demonstration of the inherent safety feature of HTGRs through the loss-of-forced-cooling test in the HTTR

Nagasumi, Satoru; Hasegawa, Toshinari; Iigaki, Kazuhiko; Nakagawa, Shigeaki; Kubo, Shinji; Shimazaki, Yosuke; Nakajima, Kunihiro; Sakurai, Yosuke; Shinohara, Masanori; Saito, Kenji; et al.

Nuclear Engineering and Design, 446(Part A), p.114542_1 - 114542_14, 2026/01

https://doi.org/10.1016/j.nucengdes.2025.114542

To demonstrate HTGR's safety features, a loss-of-forced-cooling (LOFC) test was conducted using the HTTR. In this test, the forced cooling in the reactor core was intentionally lost by shutting down all helium gas circulators (HGCs) without reactor scram. During steady-state operation at 100% reactor power (30 MW), after the LOFC, the reactor power spontaneously decreased. This power reduction occurred due to the negative reactivity feedback effect triggered by an increase in core temperature. The power stabilized at a lower value of 1.2% after re-criticality. Additionally, the measured radioactivity concentration in the primary coolant remained nearly unchanged during this LOFC operation and during an immediately subsequent HTTR operation. This indicates no failure of the coated particle fuel, even after the increase in core temperature associated with the LOFC event. These results provide experimental evidence of the safety features of HTGRs.

Impact of fast reactor fuel type on backend processes in the nuclear fuel cycle

Takeshita, Kenji*; Okamura, Tomohiro*; Nakase, Masahiko*; Abe, Takumi; Nishihara, Kenji

Progress in Nuclear Science and Technology (Internet), 8, p.52 - 57, 2025/09

https://doi.org/10.15669/pnst.8.52

Impact of nuclear fuel cycle operation factor uncertainty on nuclear power plant operation

Abe, Takumi; Nishihara, Kenji

Progress in Nuclear Science and Technology (Internet), 8, p.47 - 51, 2025/09

https://doi.org/10.15669/pnst.8.47

Achievement of safety demonstration tests using HTTR; Loss of forced cooling test at 100% reactor power (30 MW)

Nagasumi, Satoru; Hasegawa, Toshinari; Nakagawa, Shigeaki; Kubo, Shinji; Iigaki, Kazuhiko; Shinohara, Masanori; Saikusa, Akio; Nojiri, Naoki; Saito, Kenji; Furusawa, Takayuki; et al.

JAEA-Research 2025-005, 23 Pages, 2025/07

JAEA-Research-2025-005.pdf:2.68MB

A safety demonstration test under abnormal operating conditions using the HTTR (High Temperature Engineering Test Reactor) was conducted to demonstrate safety features of the HTGRs (High Temperature Gas-cooled Reactors). Under a simulation of a control rod shutdown failure, all primary helium gas circulators were intentionally stopped during a steady-state operation at 100% reactor thermal power (30 MW), temporal changes of the reactor power and temperatures around the reactor pressure vessel (RPV) were obtained after the complete loss of forced heat removal from the reactor core. After the event (primary coolant flow stopped), the reactor power quickly decreased due to the negative reactivity feedback associated with the core temperature rise, and then the reactor power spontaneously shifted to a stable state of low power (about 1.2%) even after a recriticality. Heat dissipation from RPV surface to a surrounding vessel cooling system (water-cooled panels) ensured the amount of heat removal required to maintain the reactor temperature constant in the low power state. In this way, the transition from the event occurrence to the stable and safety state, i.e., inherent safety features of HTGRs, were demonstrated in the case of core forced cooling loss without active shutdown operations.

Scenario analysis of future nuclear energy use in Japan, 1; Methodology of nuclear fuel cycle simulator: NMB4.0

Abe, Takumi; Oizumi, Akito; Nishihara, Kenji; Nakase, Masahiko*; Asano, Hidekazu*; Takeshita, Kenji*

Progress in Nuclear Science and Technology (Internet), 7, p.299 - 304, 2025/03

https://doi.org/10.15669/pnst.7.299

Currently, much research continues on stable energy sources that do not emit CO in order to achieve a carbon-neutral and sustainable society. Nuclear energy is one of the such sources, and various new reactors and reprocessing technologies are being developed. In order to implement the nuclear fuel cycle with these technologies, a nuclear fuel cycle simulator is required to quantitatively evaluate various quantities, such as the distribution of nuclear fuel materials and the scale of waste loading. For this purpose, NMB4.0 was developed in collaboration with Tokyo Institute of Technology and Japan Atomic Energy Agency. This code calculates the material balance of 179 nuclides including actinides and fission products (FPs) from the front-end to the back-end and simulates the nuclear fuel cycle in an integrated manner. Unlike other nuclear fuel cycle simulators, the code is capable of performing precise back-end analyses such as the number of radioactive wastes and the scale of the geological repository considering heat generation of waste package under diverse nuclear energy scenario, and is an open source code that runs on Microsoft Excel. By these features, it is possible to quantitatively study nuclear energy utilization strategies with various stakeholders. The presentation will detail the numerical model used in NMB4.0.

None

Nishihara, Kenji

Genshiryoku Nenkan 2025, p.75 - 80, 2024/10

no abstracts in English

Preliminary application of the nuclear fuel cycle code FAMILY-21 for SFR introduction in Japan

Kato, Atsushi; Takino, Kazuo; Yasumatsu, Naoto*; Nishihara, Kenji

Proceedings of International Conference on Nuclear Fuel Cycle (GLOBAL2024) (Internet), 4 Pages, 2024/10

This paper introduces application of the nuclear fuel cycle code FAMILY-21 which was developed in Fast reactor development division of Japan Atomic Energy Agency (JAEA). In the first part, benchmark comparison between FAMILY-21 and Nuclear Material Balance version 4.0 (NMB4.0) which is open-source code for integrated fuel cycle simulation developed in the sector of nuclear science research, JAEA is carried out. Second part shows preliminary application of FAMILY-21 for Japan.

Initial verification of Cyclus and NMB fuel cycle simulators

Bachmann, A. M.*; Richards, S.*; Feng, B.*; Nishihara, Kenji; Abe, Takumi

Proceedings of International Conference on Nuclear Fuel Cycle (GLOBAL2024) (Internet), 4 Pages, 2024/10

This work demonstrates the value of code verification as an initial step in utilizing fuel cycle simulation. Cyclus and NMB are open-source fuel cycle simulators that provide computational modeling of nuclear fuel cycle alternatives and were chosen by Argonne National Laboratory and Japan Atomic Energy Agency (JAEA), respectively, for a multi-year collaboration on fuel cycle benchmarks. Both are relatively new and can be improved after conducting a rigorous code-to-code comparison. Initial verification of these simulators was performed using a set of hypothetical scenarios for once-through and multi-recycle fuel cycles. The results of this work identify how differences in scenario definitions and the modeling methodologies of the two simulators lead to differences in results in material inventories, mass flows, and other important metrics for fuel cycle assessments.

Direct disposal concepts of spent mixed oxide fuel from light water reactors based on heat transfer calculations

Abe, Takumi; Nishihara, Kenji

Journal of Nuclear Science and Technology, 61(8), p.1048 - 1060, 2024/08

https://doi.org/10.1080/00223131.2023.2293106

Methods of direct disposal of mixed oxide spent fuel (MOXSF) consumed in light water reactors were investigated via heat transfer calculation. The temperature of the buffer material surrounding the waste is the most stringent limitation on the direct disposal of MOXSF. Therefore, the effects on the maximum temperature of the buffer material were examined by changing the occupied area, cooling term of MOXSF, and other parameters considering the direct disposal of uranium spent fuel. The results showed that it is necessary to change the cooling term and disposal depth in addition to the change of the occupied area. Accordingly, some disposal concepts that satisfy the limitation of the maximum buffer material temperature were derived, and estimates revealed that the occupied area per unit waste of the MOXSF is three to five times that of uranium spent fuel.