Data driven prediction of thermal hysteresis in NiTi-based high entropy shape memory alloys

This study provides a data-driven approach to predict thermal hysteresis in NiTi-based high-entropy shape memory alloys (HESMAs), using a dataset of NiTi-based shape memory alloys that includes the four transformation temperatures (M s , M f , A s , A f ). Machine learning models based on extremely randomized trees were proposed to predict M s and the thermal hysteresis (∆T = A f - M s ), using 11 physical material descriptors as inputs. M s prediction is dominated by mixing enthalpy and the valence electron ratio, while hysteresis depends on 5 additional balanced features such as the melting temperature, electronegativity, atomic radius, and the mixing entropy. The predictions models are highly reliable with mean absolute errors of 38 °C for M s and 14 °C for ΔT. Finally, a design strategy is proposed for (NiCuCo)(TiZrHf) HESMAs that optimises both M s and ΔT with targeted values. • Database of NiTi-like high entropy and shape memory alloys of the transformation temperatures. • Machine learning algorithm for thermal hysteresis and martensitic transformation temperature prediction. • Tool for high entropy shape memory alloys design with desired properties. • Exploration of the compositional space of (NiCuCo)(TiZrHf) HESMAs.