Webinar
MACHINE LEARNING SERIES
High-throughput CALPHAD calculations for screening and machine learning of refractory complex concentrated alloys
Refractory complex, concentrated alloys (also called high entropy alloys) are very high-melting-temperature alloys with significant concentrations of many different alloying elements, and they represent a paradigm shift in alloy design strategies by motivating researchers to explore the entirety of phase diagrams. Using relatively coarse composition grids across 10 potential alloying elements, one can easily generate more than 100,000 unique compositions.
Thanks to highly parallelized processes, the phase equilibria and thermo-physical properties of 100,000 or more compositions can be readily calculated and screened to examine only those exhibiting favorable characteristics, such as single-phase body centered cubic at elevated temperatures. In this webinar, attendees will see a workflow for performing high-throughput CALPHAD calculations using TC-Python implemented in the Thermo-Calc(R) software to screen initially large and high dimensional composition space.
This webinar showcases how machine learning and active learning can reduce the number of experiments required to optimize a set of properties from ~5,000 down to just ~20. These efforts simultaneously increase potential useful composition space and dramatically the pace of accelerated alloy design.
This webinar is a part of Thermo-Calc Software’s Machine Learning Webinar Series, taking place between November 2024 and February 2025. Find more information about the other webinars below.
[ON-DEMAND] CALPHAD by Machine Learning and for Machine Learning
[ON-DEMAND] Alloy Design Based on Artificial Intelligence and Machine Learning
[ON-DEMAND] Bayesian Frameworks for Accelerated Alloy Discovery