|About this Abstract
||2019 TMS Annual Meeting & Exhibition
||Computational Approaches for Big Data, Artificial Intelligence and Uncertainty Quantification in Computational Materials Science
||Impact of Uncertainty Quantification in Automated CALPHAD Modeling on the Design of Additively Manufactured Functionally-graded Alloys
||Brandon Bocklund, Lourdes D. Bobbio, Richard A. Otis, ShunLi Shang, Allison Beese, Zi-Kui Liu
|On-Site Speaker (Planned)
There has been renewed interest in uncertainty quantification of CALPHAD models as they become more widely used in ICME approaches and as software tools to develop and use CALPHAD models with quantified uncertainty have become more mature. Functionally-graded materials (FGMs) that vary in composition and properties can be developed using additive manufacturing, however deleterious phases prevent directly joining many alloys with different principle components. Because multicomponent CALPHAD databases are based on constituent subsystems, few multicomponent databases exist that accurately describe phase relations across the entire composition range. Progress will be presented on a Cr-Fe-Ni-Ti-V CALPHAD database developed using the ESPEI software package. ESPEI is a tool for CALPHAD modeling and uncertainty quantification using a two-step model selection and optimization method. The presentation will discuss uncertainty in the Gibbs energies of individual phases, propagation of uncertainty to the phase diagram, and the impact of uncertainty on the design of additively-manufactured FGMs.
||Planned: Supplemental Proceedings volume