| About this Abstract |
| Meeting |
2021 TMS Annual Meeting & Exhibition
|
| Symposium
|
Additive Manufacturing: Solid-State Phase Transformations and Microstructural Evolution
|
| Presentation Title |
Probabilistic Machine Learning Assisted Study of Directed Energy Deposited Alloys |
| Author(s) |
Soumya Nag, Yiming Zhang, Sreekar Karnati, Lee Kerwin, Eric MacDonald, Neil Johnson, Sathyanarayanan Raghavan, Dora Cheung, Alex Kitt, Changjie Sun, Genghis Khan, Chris Williams, Thomas Broderick, Mark Benedict, Brandon Ribic |
| On-Site Speaker (Planned) |
Soumya Nag |
| Abstract Scope |
Additive Manufacturing is truly a “complexity for free” fabrication process. However, the complex interaction of design, materials and manufacturing often lead to long iterative evaluation cycles. Therefore, materials and manufacturing strategies need to be aligned with Materials Genome Initiative to develop, produce and deploy high throughput components. To achieve this, coupling experimental and computational tools, as well as application of novel processing techniques are essential.
In the current study, powder blown Directed Energy Deposition (DED) AM modality was employed to fabricate structural Titanium and Nickel alloys. The overarching goal was to set up subscale build DOEs that encompass critical feature and/or compositional space. Subsequently, physics-based model that generated response surfaces was used to predict and in turn optimize build pathways tailored toward targeted build strategies. It is important to note that this paradigm may be universally applied to materials and AM modalities beyond the current scope of this work. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Titanium, Additive Manufacturing, Machine Learning |