Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
Maruyama, Shuhei; Endo, Tomohiro*; Yamamoto, Akio*
Journal of Nuclear Science and Technology, 60(11), p.1372 - 1385, 2023/11
Times Cited Count：1 Percentile：68.31(Nuclear Science & Technology)Katano, Ryota; Oizumi, Akito; Fukushima, Masahiro; Pyeon, C. H.*; Yamamoto, Akio*; Endo, Tomohiro*
Nuclear Science and Engineering, 20 Pages, 2023/00
Times Cited Count：0 Percentile：0.01(Nuclear Science & Technology)In this study, we have demonstrated that data assimilation using lead and bismuth sample reactivities measured in the Kyoto University Critical Assembly A-core can successfully reduce the uncertainty of the coolant void reactivity in accelerator-driven systems derived from inelastic-scattering cross-sections of lead and bismuth. We re-evaluated and highlighted the experimental uncertainties and correlations of the sample reactivities for the data assimilation formula. We used the MCNP6.2 code to evaluate the sample reactivities and their uncertainties, and performed data assimilation using the reactor analysis code system MARBLE. The high-sensitivity coefficients of the sample reactivities to lead and bismuth allowed us to reduce the cross-section-induced uncertainty of the void reactivity of the accelerator-driven system from 6.3% to 4.8%, achieving a provisional target accuracy of 5% in this study. Furthermore, we demonstrated that the uncertainties arising from other dominant factors, such as minor actinides and steel, can be effectively reduced by using integral experimental data sets for the unified cross-section dataset ADJ2017.
Hasegawa, Yuta; Imamura, Toshiyuki*; Ina, Takuya; Onodera, Naoyuki; Asahi, Yuichi; Idomura, Yasuhiro
Proceedings of 13th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Heterogeneous Systems (ScalAH22) (Internet), p.10 - 17, 2022/00
The ensemble data assimilation of computational fluid dynamics simulations based on the lattice Boltzmann method (LBM) and the local ensemble transform Kalman filter (LETKF) is implemented and optimized on a GPU supercomputer based on NVIDIA A100 GPUs. To connect the LBM and LETKF parts, data transpose communication is optimized by overlapping computation, file I/O, and communication based on data dependency in each LETKF kernel. In two dimensional forced isotropic turbulence simulations with the ensemble size of 
and the number of grid points of 
, the optimized implementation achieved 
speedup from the naive implementation, in which the LETKF part is not parallelized. The main computing kernel of the local problem is the eigenvalue decomposition (EVD) of 
real symmetric dense matrices, which is computed by a newly developed batched EVD in EigenG. The batched EVD in EigenG outperforms that in cuSolver, and 
speedup was achieved.
