| Author(s) |
Adam M. Krajewski, Arindam Debnath, Shuang Lin, Marcia Ahn, Hui Sun, Allison Beese, Wesley Reinhart, Zi-Kui Liu |
| Abstract Scope |
ULTERA database, developed under the ARPA-E's ULTIMATE program, is aimed at collecting literature data on high entropy alloys (HEAs) to facilitate rapid discovery of new ones using forward and inverse design, with the primary focus on creep behavior, yield stress, ductility, and hardness.
As of July 2022, ULTERA contains over 6,200 property-datapoints, corresponding to 2,485 unique HEAs, collected from 455 source DOIs. All data is available through a high-performance API, following FAIR principles, while statistics on it can be found at our phaseslab.com/ultera web page. The database architecture is designed to automatically integrate starting literature data in real-time with methods such as experiments, generative modeling, predictive modeling, and validations.
In this presentation, we showcase the database with particular emphasis on our methods developed to assure high quality of data and screen for abnormal entries; a step crucial to creating a dataset for data-driven alloy design. |