Physically Consistent Machine Learning for Melting Temperature Prediction of Refractory High-Entropy Alloys

TL;DR

This paper presents a physically consistent machine learning model using XGBoost to accurately predict the melting temperatures of refractory high-entropy alloys, capturing phase stability transitions without data leakage.

Contribution

The study introduces a novel ML approach that employs physically motivated features to predict Tm and phase stability in HEAs, avoiding data leakage and ensuring physical consistency.

Findings

Achieved R2 of 0.948 and 5% relative error in Tm prediction.

Successfully captured BCC-FCC phase transition at VEC ~6.87.

Validated model using known metallurgical principles.

Abstract

Predicting the melting temperature (Tm) of multi-component and high-entropy alloys (HEAs) is critical for high-temperature applications but computationally expensive using traditional CALPHAD or DFT methods. In this work, we develop a gradient-boosted decision tree (XGBoost) model to predict Tm for complex alloys based on elemental properties. To ensure physical consistency, we address the issue of data leakage by excluding temperature-dependent thermodynamic descriptors (such as Gibbs free energy of mixing) and instead rely on physically motivated elemental features. The optimized model achieves a coefficient of determination (R2) of 0.948 and a Mean Squared Error (MSE) of 9928 which is about 5% relative error for HEAs on a validation set of approximately 1300 compositions. Crucially, we validate the model using the Valence Electron Concentration (VEC) rule. Without explicit constraints during training, the model successfully captures the known stability transition between BCC and FCC phases at a VEC of approximately 6.87. These results demonstrate that data-driven models, when properly feature-engineered, can capture fundamental metallurgical principles for rapid alloy screening.