局所アンサンブル変換カルマンフィルタを用いた合金凝固シミュレーションのデータ同化

DOI Web Site

書誌事項

タイトル別名
  • Data Assimilation for Simulation of Alloy Solidification Using Local Ensemble Transform Kalman Filter

抄録

<p>We have applied the ensemble-based data assimilation (DA) methodology, the local ensemble transform Kalman filter (LETKF) to the three-dimensional phase-field simulation of dendritic growth in an alloy solidification for estimating not only the state of solidification (i.e. dendritic morphology and solute concentration) but also the model parameters involved in the phase-field model. LETKF enables us to estimate the state of solidification and the model parameters with lower computational resources than the conventional DA methodology, the ensemble Kalman filter (EnKF). In this study, we have performed the numerical experiments called twin experiments where the diffusion coefficient of Cu atom in liquid phase and the interfacial energy of solid/liquid interface were estimated from a synthetic observational data of dendritic morphology of Al-Cu binary alloy solidification. Through the twin experiments, we revealed that LETKF successfully estimated a priori assumed-true diffusion coefficient of Cu atom in liquid phase and interfacial energy of solid/liquid interface simultaneously.</p>

収録刊行物

キーワード

詳細情報 詳細情報について

問題の指摘

ページトップへ