Processing time, temperature, and initial chemical composition prediction from materials microstructure by deep network for multiple inputs and fused data

Prediction of the chemical composition and processing history from microstructure morphology can help in material inverse design. In this work, we propose a fused-data deep learning framework that predict microstructure. We used Fe-Cr-Co alloys as model material. The developed is able to heat treatment time, temperature, initial compositions by reading Fe distribution its concentration. results show trained neural network has highest accuracy for chemistry then time temperature. identified two scenarios inaccurate predictions; 1) There are several paths an identical microstructure, 2) Microstructures reach steady-state morphologies after long aging. error analysis shows majority wrong predictions indeed not wrong, but other right answers. validated successfully with experimental transmission electron microscopy micrograph.