A neural network model for high entropy alloy design

A neural network model is developed to search vast compositional space of high entropy alloys (HEAs). The model predicts the mechanical properties of HEAs better than several other models. It's because the special structure of the model helps the model understand the characteristics of constituent elements of HEAs. In addition, thermodynamics descriptors were utilized as input to the model so that the model predicts better by understanding the thermodynamic properties of HEAs. A conditional random search, which is good at finding local optimal values, was selected as the inverse predictor and designed two HEAs using the model. We experimentally verified that the HEAs have the best combination of strength and ductility and this proves the validity of the model and alloy design method. The strengthening mechanism of the designed HEAs is further discussed based on microstructure and lattice distortion effect. The present alloy design approach, specialized in finding multiple local optima, could help researchers design an infinite number of new alloys with interesting properties.

Automated summary

A neural network model is developed to search the vast compositional space of high entropy alloys (HEAs) and predicts their mechanical properties better than other models due to its special structure and the utilization of thermodynamics descriptors as input. A conditional random search is selected as the inverse predictor and used to design two HEAs, which are experimentally verified to have the best combination of strength and ductility. The alloy design approach could help researchers design numerous new alloys with interesting properties.