GPU-enabled ensemble data assimilation for mesh-refined lattice Boltzmann method

Hasegawa, Yuta; Idomura, Yasuhiro; Onodera, Naoyuki

EPJ Web of Conferences, 302, p.03005_1 - 03005_9, 2024/10

https://doi.org/10.1051/epjconf/202430203005

We implemented the ensemble data assimilation (DA) method, the local ensemble transform Kalman filter (LETKF), into the mesh-refined lattice Boltzmann method (LBM) for turbulent flows. Both the LETKF and the mesh-refined LBM were fully implemented on GPUs, so that they are efficiently computed on modern GPU-based supercomputers. We examined the DA accuracy against the flow around a cylinder. The result showed that our method enabled accurate DA with spatially- and temporarily-sparse observation data; the error of the assimilated velocity field with the observation interval of and the observation resolution (1.56% of the total computational grids) was smaller than the amplitude of the observation noise, where is the period of the Krmn vortex and is diameter of the square cylinder.

Benchmarking GPU backend of longitudinal simulation code BLonD

Adachi, Kyosuke; Tamura, Fumihiko; Nomura, Masahiro; Shimada, Taihei; Miyakoshi, Ryosuke*; Okita, Hidefumi; Yoshii, Masahito*; Omori, Chihiro*; Seiya, Kiyomi*; Hara, Keigo*; et al.

Proceedings of 21st Annual Meeting of Particle Accelerator Society of Japan (Internet), p.543 - 546, 2024/10

https://www-linac.kek.jp/mirror/www.pasj.jp/web_publish/pasj2024/proceedings/PDF/THP0/THP003.pdf

no abstracts in English

Applying a deep generative model to mountain plot images

Nomura, Masahiro; Shimada, Taihei; Tamura, Fumihiko; Okita, Hidefumi; Miyakoshi, Ryosuke*; Seiya, Kiyomi*; Yoshii, Masahito*; Omori, Chihiro*; Hara, Keigo*; Hasegawa, Katsushi*; et al.

Proceedings of 21st Annual Meeting of Particle Accelerator Society of Japan (Internet), p.85 - 88, 2024/10

https://www.pasj.jp/web_publish/pasj2024/proceedings/PDF/THOT/THOT02.pdf

no abstracts in English

R&D status of digital technology on inverse estimation of radioactive source distributions and related source countermeasures; Fast Digital Twin Tech. in Decommissioning Field: 3D-ADRES-Indoor FrontEnd

Machida, Masahiko; Yamada, Susumu; Kim, M.; Tanaka, Satoshi*; Tobita, Yasuhiro*; Iwata, Ayako*; Aoki, Yuto; Aoki, Kazuhisa; Yanagisawa, Kenichi*; Yamaguchi, Takashi; et al.

RIST News, (70), p.3 - 22, 2024/09

Inside the Fukushima Daiichi Nuclear Power Plant (1F), there are many locations with high radiation levels due to contamination by radioactive materials that leaked from the reactor. These pose a significant obstacle to the smooth progress of decommissioning work. To help solve this issue, the Japan Atomic Energy Agency (JAEA), under a subsidy from the Ministry of Economy, Trade, and Industry's decommissioning and contaminated water management project, is conducting research and development on digital technologies to improve the radiation environment inside the decommissioning site. This project, titled "Development of Technology to Improve the Environment Inside Reactor Buildings (Enhancing Digital Technology for Environment and Source Distribution to Reduce Radiation Exposure)," began in April of FY 2023. In this project, the aim is to develop three interconnected systems: FrontEnd, Pro, and BackEnd. The FrontEnd system, based on the previously developed 3D-ADRES-Indoor (prototype) from FY 2021-2022, will be upgraded to a high-speed digital twin technology usable on-site. The Pro system will carry out detailed analysis in rooms such as the new office building at 1F, while the BackEnd system will serve as a database to centrally manage the collected and analyzed data. This report focuses on the FrontEnd system, which will be used on-site. After point cloud measurement, the system will quickly create a 3D mesh model, estimate the radiation source from dose rate measurements, and refine the position and intensity of the estimated source using recalculation techniques (re-observation instructions and re-estimation). The results of verification tests conducted on Unit 5 are also presented. Furthermore, the report briefly discusses the future research and development plans for this project.

Structure of drifting snow simulated by Lagrangian particle dispersion model coupled with large-eddy simulation using the lattice Boltzmann method

Watanabe, Tsutomu*; Ishikawa, Shuhei*; Kawashima, Masayuki*; Shimoyama, Ko*; Onodera, Naoyuki; Hasegawa, Yuta; Inagaki, Atsushi*

Journal of Wind Engineering and Industrial Aerodynamics, 250, p.105783_1 - 105783_17, 2024/07

https://doi.org/10.1016/j.jweia.2024.105783

This paper presents simulations of drifting snow using a Lagrangian particle dispersion model coupled with a large-eddy simulation code. The model accurately replicates observed features such as mass transport rate dependency on flow velocity and variations in particle size distribution. It also shows that the saltation layer height increases monotonically with flow velocity, contrary to conventional estimates. Additionally, the study confirms the transition from saltation to suspension near the estimated saltation layer height and finds that dense snow streamers are linked to small-scale low-speed streaks in near-surface flows.

Data assimilation experiment of flow around three-dimensional square cylinder using LBM-LETKF with small number of observation points

Hasegawa, Yuta; Idomura, Yasuhiro; Onodera, Naoyuki

Keisan Kogaku Koenkai Rombunshu (CD-ROM), 29, 4 Pages, 2024/06

We implemented the ensemble data assimilation (DA) of turbulence by using the mesh-refined lattice Boltzmann method with the local ensemble transform Kalman filter (LBM-LETKF). We examined the accuracy of the data assimilation against a turbulent flow around a three-dimensional square cylinder. The DA error was comparable or less than the observation noise when the observation interval was a half of the period of the Krmn vortex street and the number of observation points was 0.195% of computational grid points. The LBM-LETKF enables DA of turbulence with spatially- and temporally- sparse observations.

Optimization in granulation conditions for adsorbent of extraction chromatography

Hasegawa, Kenta; Goto, Ichiro*; Miyazaki, Yasunori; Ambai, Hiromu; Watanabe, So; Watanabe, Masayuki; Sano, Yuichi; Takeuchi, Masayuki

Mechanical Engineering Journal (Internet), 11(2), p.23-00407_1 - 23-00407_8, 2024/04

https://doi.org/10.1299/mej.23-00407

Effectiveness of fused LASSO for prediction of distribution of radioactive materials in reactor buildings

Yamada, Susumu; Yoshida, Toru*; Hasegawa, Yukihiro*; Machida, Masahiko

Proceedings of Waste Management Symposia 2024 (WM2024) (Internet), 15 Pages, 2024/03

In order to safely carry out the decommission of reactor buildings, it is extremely important to identify the radiation source distribution. It has been reported that when the structural model of the building is constructed by uniform cells, the source distribution can be estimated from the measured air dose rates by minimizing an evaluation function using the Least Absolute Shrinkage and Selection Operator (LASSO). Moreover, if cells are non-uniform, we can estimate the distribution using the fused LASSO which minimizes the evaluation function that takes account of the connectivity between the adjacent cells. However, when a group of some cells is considered disconnected from the surrounding ones due to the precision of the measured structural data, the concentration of the group can be singularly high. Therefore, in order to avoid the problem, we propose a new evaluation function that can prevent the singularity. We estimated the distribution for the test model using the proposed evaluation function and confirmed the validity of the function. Moreover, we succeeded in estimating the source distribution in the pool canal circulation system room in JMTR in the Japan Atomic Energy Agency by the fused LASSO for the new function more accurately than previous analysis.

Structural investigation of simulated waste glass surface drained in operation confirmation test of 3rd TVF glass melter

Nagai, Takayuki; Hasegawa, Takehiko*

JAEA-Research 2023-008, 41 Pages, 2023/12

JAEA-Research-2023-008.pdf:7.52MB

To reduce the risks posed by stored the high-level radioactive liquid waste (HAW), Tokai Vitrification Facility (TVF) is working to produce the HAW into vitrified bodies. With the aim of steady vitrification of HAW in TVF, the vitrification technology section has manufactured a new 3rd melter with an improved bottom structure and is working to verify the performance of this melter. In this study, solidified glass samples were taken from simulated vitrified bodies produced by flowing molten glass during the bottom drain-out test in the operation confirmation of the TVF 3rd melter. And the properties of the surface layer and fracture surface of the vitrified bodies were evaluated by using Raman spectroscopy, synchrotron radiation XAFS measurement, and laser ablation inductively coupled plasma atomic emission spectroscopy (LA ICP-AES) analysis. As a result of measuring the surface layer and fracture surface of the solidified samples produced on an actual scale, a slight difference was confirmed between the properties of the surface layer and those of the fracture surface. Since the chemical composition of these simulated vitrified bodies does not contain platinum group elements, it is expected that the glass structure of solidified glass samples is different from that of the actual vitrified body. However, this sample measuring was a valuable opportunity to evaluate samples produced by using the direct energized joule heating method. The properties of cullet used the operation confirmation of the TVF 3rd melter and the cullet of another production lot were measured and analyzed in the same manner under the measuring conditions of solidified glass samples. As a result, it was confirmed that cullet with different producing histories have different glass structures even with the same chemical composition, and that differences in glass structures remain in the glass samples after melting these cullet.

Continuous data assimilation of large eddy simulation by lattice Boltzmann method and local ensemble transform Kalman filter (LBM-LETKF)

Hasegawa, Yuta; Onodera, Naoyuki; Asahi, Yuichi; Ina, Takuya; Imamura, Toshiyuki*; Idomura, Yasuhiro

Fluid Dynamics Research, 55(6), p.065501_1 - 065501_25, 2023/11

https://doi.org/10.1088/1873-7005/ad06bd

We investigate the applicability of the data assimilation (DA) to large eddy simulations (LESs) based on the lattice Boltzmann method (LBM). We carry out the observing system simulation experiment of a two-dimensional (2D) forced isotropic turbulence, and examine the DA accuracy of the nudging and the local ensemble transform Kalman filter (LETKF) with spatially sparse and noisy observation data of flow fields. The advantage of the LETKF is that it does not require computing spatial interpolation and/or an inverse problem between the macroscopic variables (the density and the pressure) and the velocity distribution function of the LBM, while the nudging introduces additional models for them. The numerical experiments with grids and 10% observation noise in the velocity showed that the root mean square error of the velocity in the LETKF with observation points ( of the total grids) and 64 ensemble members becomes smaller than the observation noise, while the nudging requires an order of magnitude larger number of observation points to achieve the same accuracy. Another advantage of the LETKF is that it well keeps the amplitude of the energy spectrum, while only the phase error becomes larger with more sparse observation. From these results, it was shown that the LETKF enables robust and accurate DA for the 2D LBM with sparse and noisy observation data.