自然対流挙動のカルマンフィルターによるデータ同化解析

Data assimilation analysis of natural convection using Kalman filter

石垣 将宏* ; 廣瀬 意育  ; 安部 諭  ; 永井 亨*; 渡辺 正*

Ishigaki, Masahiro*; Hirose, Yoshiyasu; Abe, Satoshi; Nagai, Toru*; Watanabe, Tadashi*

To estimate thermal flow in a nuclear reactor during an accident, it is important to improve the accuracy of computational fluid dynamics (CFD) analysis. Temperature and flow velocity are not homogeneous and have large variations in a reactor containment vessel because of its very large volume. In addition, the importance of the influence of variations of initial and boundary conditions is pointed out by Kelm, et al(2016). Therefore, it is necessary to set the initial and boundary conditions that take into account the variations of these physical quantities. However, it is a difficult subject to set such complicated initial and boundary conditions. Then, we can obtain realistic initial and boundary conditions by data assimilation, and we can improve the accuracy of the simulation result. We applied the data assimilation by ensemble Kalman filter for natural convection simulation, and we discuss the improvement of the simulation result. By the data assimilation using only the temperature as the observed data, we succeeded in calculating the temperature and velocity distributions accurately.