CityTransformer; A Transformer-based model for contaminant dispersion prediction in a realistic urban area

Asahi, Yuichi; Onodera, Naoyuki; Hasegawa, Yuta; Shimokawabe, Takashi*; Shiba, Hayato*; Idomura, Yasuhiro

Boundary-Layer Meteorology, 34 Pages, 2023/01

https://doi.org/10.1007/s10546-022-00777-8

We develop a Transformer-based deep learning model to predict the plume concentrations in the urban area under uniform flow conditions. Our model has two distinct input layers: Transformer layers for sequential data and convolutional layers in convolutional neural networks (CNNs) for image-like data. Our model can predict the plume concentration from realistically available data such as the time series monitoring data at a few observation stations and the building shapes and the source location. It is shown that the model can give reasonably accurate prediction with orders of magnitude faster than CFD simulations. It is also shown that the exactly same model can be applied to predict the source location, which also gives reasonable prediction accuracy.

Theoretical evaluation of neutron thermal scattering laws of heavy water for JENDL-5

Ichihara, Akira; Abe, Yutaka*

JAEA-Conf 2022-001, p.175 - 180, 2022/11

https://doi.org/10.11484/jaea-conf-2022-001

Thermal neutron scattering law data were calculated for the heavy water molecule toward the fifth version of the Japanese evaluated nuclear data library, JENDL-5. The scattering laws for deuterium and oxygen atoms were computed using the molecular dynamics simulations. The simulations have been performed in the temperature range from 283.6 K to 600 K. The scattering law data have been evaluated in the neutron incident energies between 0.01 meV and 10 eV. With the obtained scattering laws, we calculated the cross sections for the heavy water molecule, and confirmed that the experimental data at room temperature were well reproduced. Moreover, in the computed temperature range, the cross sections were almost consistent with the ENDF/B-VIII.0 evaluations.

Measurement of fragments of a wall-impinging liquid jet in a shallow pool

Horiguchi, Naoki; Yoshida, Hiroyuki; Kaneko, Akiko*; Abe, Yutaka*

Proceedings of 12th Japan-Korea Symposium on Nuclear Thermal Hydraulics and Safety (NTHAS12) (Internet), 6 Pages, 2022/10

For safety evaluation of nuclear reactors in severe accidents, it is important to estimate physical quantities of fragments generated from the molten fuel jet, which falls in a pool and breaks up. The evaluation method has been developed for the behavior as liquid jet with hydrodynamic interaction including fuel coolant interaction (FCI). In case of a shallow pool assumed in ex-vessel, the molten fuel jet is assumed to behave as wall-impinging liquid jet and to form liquid film flow spreading on the floor with/without fragmentation. In our research, focusing on hydrodynamic interaction and the transient 3-dimensional spreading on the floor, we have developed the evaluation method by numerical simulation using the two-phase flow simulation code with interface tracking method (TPFIT) developed by JAEA and, the experimental method using the 3D-LIF method in liquid-liquid system for the validation data. In our previous studies, we investigated the wall-impinging liquid jet behavior with fragmentation and observed that the liquid film flow had some characteristic parts transiently. Since it indicates that the quantities change depending on the parts and affect the safety evaluation, it is important to measure the quantities of the fragments generated from each part. This paper explains the measurement of the physical quantities of the fragments generated from each part of the wall-impinging liquid jet in a shallow pool for the validation of the numerical simulation. We conducted an experiment with the 3D-LIF method and segmented the experimental data based on the fragmentation point over the liquid film flow using the dispersed phase tracking method, developed by JAEA. Then, we measured the diameter and amount of the fragments from the segmented experimental data and investigated their changing trend.

Structure, stability, and actinide leaching of simulated nuclear fuel debris synthesized from UO, Zr, and stainless-steel

Kirishima, Akira*; Akiyama, Daisuke*; Kumagai, Yuta; Kusaka, Ryoji; Nakada, Masami; Watanabe, Masayuki; Sasaki, Takayuki*; Sato, Nobuaki*

Journal of Nuclear Materials, 567, p.153842_1 - 153842_15, 2022/08

https://doi.org/10.1016/j.jnucmat.2022.153842

To understand the chemical structure and stability of nuclear fuel debris consisting of UO, Zr, and Stainless Steel (SUS) generated by the Fukushima Daiichi Nuclear Power Plant accident in Japan in 2011, simulated debris of the UO-SUS-Zr system and other fundamental component systems were synthesized and characterized. The simulated debris were synthesized by heat treatment for 1 to 12 h at 1600C, in inert (Ar) or oxidative (Ar + 2% O) atmospheres. Np and Am tracers were doped for the leaching tests of these elements and U from the simulated debris. The characterization of the simulated debris was conducted by XRD, SEM-EDX, Raman spectroscopy, and Mssbauer spectroscopy, which provided the major uranium phase of the UO -SUS-Zr debris was the solid solution of UO (s.s.) with Zr(IV) and Fe(II) regardless of the treatment atmosphere. The long-term immersion test of the simulated debris in pure water and that in seawater revealed the macro scale crystal structure of the simulated debris was chemically very stable in the wet condition for a year or more. Furthermore, the leaching test results showed that the actinide leaching ratios of U, Np, Am from the UO-SUS-Zr debris were very limited and less than 0.08 % for all the experiments in this study.

Experimental study of liquid spreading and atomization due to jet impingement in liquid-liquid systems

Yamamura, Sota*; Fujiwara, Kota*; Honda, Kota*; Yoshida, Hiroyuki; Horiguchi, Naoki; Kaneko, Akiko*; Abe, Yutaka*

Physics of Fluids, 34(8), p.082110_1 - 082110_13, 2022/08

https://doi.org/10.1063/5.0100340

Liquid spreading and atomization due to jet impingement in liquid-liquid systems are considered to be crucial for understanding the cooling behavior of high-temperature molten material in a shallow water pool. This phenomenon takes place when a liquid jet enters a pool filled with other immiscible liquid. The jet spreads radially after impinging on the floor while forming a thin liquid film and atomizing droplets. In this paper, we explain the result to quantify the unsteady three-dimensional behavior of the spreading jet by the employment of 3D-LIF measurements and 3-dimensional reconstruction. Under high flow velocity conditions, the phenomena of hydraulic jump and atomization of the liquid film occurred along with the spreading. To evaluate the spreading behavior, a comparison of the jump radius position of the liquid-liquid system as the representative value was made with the one calculated by the existing theory of a gas-liquid system. As the result, the spreading of the liquid film in the liquid-liquid system was suppressed compared with that in the gas-liquid system. Furthermore, the PTV method was successfully used to measure the velocity boundary layer and velocity profile in the liquid film, which are important factors that affect the spreading mechanism of the liquid film. These results revealed that in liquid-liquid systems, shear stress at the liquid-liquid interface causes a decrease in the flow velocity and suppressed the development of the velocity boundary layer. Also, to evaluate the atomization behavior, the number and diameter distribution of the droplets were measured from the acquired 3-dimensional shape data of the jet. As the result, the number of droplets increased with the flow velocity. Based on these results, we concluded that the spreading of the liquid film is affected by such atomization behavior.

Fuel unloading work in decommissioning of the prototype fast breeder reactor Monju; Fuel treatment and storage work in 2018 and 2019

Shiota, Yuki; Yabe, Takanori; Murakami, Makio; Isobe, Yuta; Sato, Masami; Hamano, Tomoharu; Takagi, Tsuyohiko; Nagaoki, Yoshihiro

JAEA-Technology 2022-001, 117 Pages, 2022/07

JAEA-Technology-2022-001.pdf:25.55MB

In the first stage of Monju decommissioning project, fuel unload work began to be carried out. There are two tasks in this work. One is fuel treatment and storage work that gets rid of sodium on the fuel assemblies unloaded from Ex-Vessel fuel Storage Tank (EVST) and carries it in the storage pool, and the other is fuel unloading that the fuel assemblies in the reactor core is replaced with dummy fuels and stored in EVST. Fuel treatment and storage work and fuel unloading work are performed alternately, and 370 bodies in the core and 160 pieces in EVST are all carried to the storage pool. 86 fuel assemblies was carried to the storage pool in fuel treatment and storage work in 2018 and 76 dummy fuels were stored in EVST for fuel unloading work. During the work, 86 types and 232 alarms / malfunctions occurred, but there was no impact on safety. There was one equipment's failure at gripper's claw open / close clutch of ex-vessel fuel transfer machine B, but it was repaired and restarted. Also it was eliminated the cause of problem or concession that the equipment failure due to the sticking of the sodium compound and continuous use of the equipment. Some problems related to system control occurred, but the work was done after checking the safety. With estimation of various troubles, reduction of frequency of trouble occurrence and minimization of impacts on schedule performed.

First demonstration of a novel single-end readout type position-sensitive optical fiber radiation sensor based on wavelength-resolved photon counting

Terasaka, Yuta; Watanabe, Kenichi*; Uritani, Akira*

Nuclear Instruments and Methods in Physics Research A, 1034, p.166793_1 - 166793_6, 2022/07

https://doi.org/10.1016/j.nima.2022.166793

Performance measurement of an urban wind simulation code with the Locally Mesh-Refined Lattice Boltzmann Method over NVIDIA and AMD GPUs

Asahi, Yuichi; Onodera, Naoyuki; Hasegawa, Yuta; Shimokawabe, Takashi*; Shiba, Hayato*; Idomura, Yasuhiro

Keisan Kogaku Koenkai Rombunshu (CD-ROM), 27, 5 Pages, 2022/06

We have ported the GPU accelerated Lattice Boltzmann Method code "CityLBM" to AMD MI100 GPU. We present the performance of CityLBM achieved on NVIDIA P100, V100, A100 GPUs and AMDMI100 GPU. Using the host to host MPI communications, the performance on MI100 GPU is around 20% better than on V100 GPU. It has turned out that most of the kernels are successfully accelerated except for interpolation kernels for Adaptive Mesh Refinement (AMR) method.

-clustering in atomic nuclei from first principles with statistical learning and the Hoyle state character

Otsuka, Takaharu; Abe, Takashi*; Yoshida, Toru*; Tsunoda, Yusuke*; Shimizu, Noritaka*; Itagaki, Naoyuki*; Utsuno, Yutaka; Vary, J. P.*; Maris, P.*; Ueno, Hideki*

Nature Communications (Internet), 13, p.2234_1 - 2234_10, 2022/04

https://doi.org/10.1038/s41467-022-29582-0

no abstracts in English

A Storm-induced flood and associated nearshore dispersal of the river-derived suspended Cs

Uchiyama, Yusuke*; Tokunaga, Natsuki*; Azuma, Kohei*; Kamidaira, Yuki; Tsumune, Daisuke*; Iwasaki, Toshiki*; Yamada, Masatoshi*; Tateda, Yutaka*; Ishimaru, Takashi*; Ito, Yukari*; et al.

Science of the Total Environment, 816, p.151573_1 - 151573_13, 2022/04

https://doi.org/10.1016/j.scitotenv.2021.151573

no abstracts in English