2025-04-01 トロント大学
＜関連情報＞
- https://www.utoronto.ca/news/ai-driven-approach-3d-printing-metal-game-changer-manufacturing-researchers
- https://www.sciencedirect.com/science/article/pii/S2214860425001009
レーザーエネルギー蒸着における正確な逆プロセス最適化フレームワーク Accurate inverse process optimization framework in laser directed energy deposition
Xiao Shang, Ajay Talbot, Evelyn Li, Haitao Wen, Tianyi Lyu, Jiahui Zhang, Yu Zou
Additive Manufacturing Available online: 12 March 2025
DOI:https://doi.org/10.1016/j.addma.2025.104736
Graphical Abstract
Highlights
- Inversely identifies optimal process parameters from customizable objectives.
- Accurately predicts melt pool geometries directly from process parameters.
- Finds optimal hatch spacing and layer thickness to make fully dense prints.
- Transferable to new materials systems and optimization objectives with a small amount of extra data.
Abstract
In additive manufacturing (AM), particularly in laser-based metal AM, process optimization is crucial to the quality of products and the efficiency of production. The identification of optimal process parameters out of a vast parameter space, however, is a daunting task. Despite advances in simulations, the process optimization for specific materials and geometries is developed through a sequential and time-consuming trial-and-error approach and often lacks the versatility to address multiple optimization objectives. Machine learning (ML) provides a powerful tool to accelerate the optimization process, but most current studies focus on simple single-track prints, which hardly translate to manufacturing 3D bulk components for engineering applications. In this study, we develop an Accurate Inverse process optimization framework in laser Directed Energy Deposition (AIDED), based on machine learning models and a genetic algorithm, to aid the process optimization in laser DED. Using AIDED, we demonstrate the following: (i) Accurate prediction of the area of single-track melt pool (R2 score 0.995), the tilt angle of multi-track melt pool (R2 score 0.969), and the cross-sectional geometries of multi-layer melt pool (1.75 % and 12.04 % errors in width and height, respectively) directly from process parameters; (ii) Determination of appropriate hatch spacing and layer thickness for fabricating fully dense (density > 99.9 %) multi-track and multi-layer prints; (iii) Inverse identification of optimal process parameters directly from customizable application objectives within 1–3 hours. We also validate the effectiveness of the AIDED experimentally by solving a multi-objective optimization problem to identify the optimal process parameters for achieving high print speeds with small effective track widths. Furthermore, we show the transferability of the framework from stainless steel to pure nickel using a small amount of additional data on pure nickel. With such transferability in AIDED, we pave a new way for “aiding” the process optimization of the laser-based AM processes that applies to a wide range of materials.
