Development of accurate prediction method for solidification microstructure in additive manufacturing by high-performance phase-field simulation

2022 Fiscal Year Final Research Report

• Project/Area Number: 21K14041
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 18010:Mechanics of materials and materials-related
• Research Institution: Kyoto Institute of Technology
• Principal Investigator: Sakane Shinji 京都工芸繊維大学, 機械工学系, 助教 (70876755)
• Project Period (FY): 2021-04-01 – 2023-03-31
• Keywords: フェーズフィールド法 / 格子ボルツマン法 / 複数GPU並列計算 / 適合格子細分化法 / デンドライト・セル成長 / レーザー溶融法 / 金属積層造形 / 凝固

Outline of Final Research Achievements

In this study, to accurately predict the solidification microstructure formed by dendrite/cell growth during metal additive manufacturing (AM) in three dimensions (3D), a multi-physics model coupling the meso-scale thermal fluid flow model and the micro-scale phase-field (PF) model was developed. To enable the large-scale 3D PF simulations, a high-performance computing methods combining parallel computation using multiple GPUs and the adaptive mesh refinement method were implemented. Through the large-scale 3D PF simulation during AM, the 3D microstructure was successfully predicted, simulating the experimental microstructure well. The developed method is a powerful tool for revealing the mechanisms of dendrite/cell growth with multi-physics phenomena during AM.

Free Research Field

計算材料科学

Academic Significance and Societal Importance of the Research Achievements

積層造形の凝固組織は造形物の品質を決定づけるため，その高精度予測が重要であるが，積層造形中のマルチフィジックスを包括的に考慮した凝固組織予測モデル構築と，3次元組織予測を可能とする高性能計算手法開発が課題であった．本研究で構築した3次元組織予測手法は，フェーズフィールド組織予測のための高性能計算手法開発の指針となり，かつ，実験的にその場観察が難しい積層造形中の凝固組織形成過程を3次元的に再現・評価できるため，学術的な貢献が期待できる．また，高品質な積層造形品を製造するための知見を生み出す強力なツールとして，今後，産業面での貢献も期待できる．