|About this Abstract
||2018 TMS Annual Meeting & Exhibition
||Computational Design and Simulation of Materials (CDSM 2018): Computational Design of Materials
||Effect of Stability of Critical Phases in Nickel-based Superalloys: Combined Machine Learning and CALPHAD Approach
||Rajesh Jha, George S Dulikravich
|On-Site Speaker (Planned)
This work focused on designing the chemistry of Nickel-based superalloys while simultaneously optimizing their multiple bulk properties. We used concepts of artificial intelligence and multi-objective optimization to obtain a set of compositions for alloys that were exposed to identical heat treatment. Thereafter, we studied phase transformations in this system by CALPHAD approach on a few compositions that were experimentally verified. From meta-modeling we also identified elements that have negligible effect on the bulk properties. Then, we studied variation of critical phases responsible for superior tensile and creep properties for a range of temperatures that these alloys are exposed to. We also studied the compositions that were experimentally verified but the predicted properties differed from those measured. For these alloys, we performed uncertainity analysis to find compositions in the vicinity of the predicted compositions from machine learning that will help us to obtain these critical phases for the same heat treatment.
||Planned: Supplemental Proceedings volume