A machine learning approach to predict austenite finish temperature in quaternary NiTiHfPd SMAs

Abstract

Machine learning (ML) has emerged as a promising tool for the design of multicomponent alloys due to their vast design spaces. Quaternary NiTiHfPd shape memory alloys (SMAs) possess unique potential to be employed in high-temperature actuation as well as damping systems. This study presents a machine learning approach using the currently available limited data regime to accelerate research on NiTiHfPd SMAs. To this end, a database of transformation temperatures of NiTiHfPd SMAs was compiled and expanded through compositional and post-processing features of the alloys. Various ML algorithms were utilized to predict the austenite finish temperature of NiTiHfPd SMAs and then validated through experiments.