About this Abstract |
Meeting |
2025 TMS Annual Meeting & Exhibition
|
Symposium
|
Additive Manufacturing Modeling, Simulation and Machine Learning
|
Presentation Title |
A data driven framework to predict and bridge multiscale mechanical phenomenon in additively manufactured component |
Author(s) |
Abhijeet Dhal, Dilip Banerjee, Rajiv Mishra |
On-Site Speaker (Planned) |
Abhijeet Dhal |
Abstract Scope |
Complex multiscale variations in microstructure and mechanical behavior have made the qualification of additively manufactured (AM) components a significant challenge. To address this, a conceptual model has been developed to delineate mechanical behavior across multiple length scales and establish crucial process-structure-property correlations for AM components. First, a microstructure predictive model was created, considering the combined thermal history of the primary solidification cycle and reheating stages. Classical strengthening models were then applied to determine spatial variations in strength, hardness, and modulus based on the predicted microstructural changes. Second, nanoindentation mapping was conducted in critical regions to validate the model predictions. Finally, machine learning was utilized to bridge various length scale phenomena and predict the most suitable mechanical properties of the AM components. |
Proceedings Inclusion? |
Planned: |
Keywords |
Additive Manufacturing, Computational Materials Science & Engineering, Machine Learning |