About this Abstract |
| Meeting |
2027 TMS Annual Meeting & Exhibition
|
| Symposium
|
Computational Thermodynamics and Kinetics
|
| Presentation Title |
Navigating Composition-Heat Treatment Complexity in Nickel-based Superalloy-718 via Uncertainty-Aware Generative AI for Reliable Alloy Design |
| Author(s) |
Hasan Al Jame, Jixuan Dong, Bo Ni, S. Mohadeseh Taheri-Mousavi |
| On-Site Speaker (Planned) |
Hasan Al Jame |
| Abstract Scope |
Additive manufacturing of Nickel-based superalloy-718 presents a multidimensional design challenge, where compositional variability and heat-treatment (HT) pathways jointly govern microstructural evolution and qualification risk of mechanical properties. This study leverages a validated CALPHAD-based ICME framework to navigate the experimentally challenging, coupled composition-HT space by capturing the complex thermodynamic and kinetic drivers of γ′/γ″ phase evolution, revealing how alloy chemistry and HT-conditions cascade across the microstructure–property space. Yield-strength uncertainty is quantified and mapped across the design space, followed by sensitivity analysis to identify and rank the most critical parameters governing alloy-performance variability. Bayesian inference and generative AI techniques are then utilized to automate targeted inverse-design of these critical parameters within composition–HT windows, achieving target yield strength under specified reliability criteria for certifiable design and manufacturing of superalloy-718. Beyond this, the versatility of the proposed multimodal physics-based framework offers a transferable pathway toward uncertainty-aware, robust manufacturing of other complex alloy systems |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Additive Manufacturing, ICME, Copper / Nickel / Cobalt |