Yuhong Zhao1*

EXTENDED ABSTRACT: Coupled with multi-scale phase field simulation and experimental design aided by integrated material calculation method, ultra-high specific strength magnesium alloy and high-strength and ductile copper alloy were prepared; Multi stage phase field simulation, multi physical field simulation and key experiments are used to replace repeated field experiments, design and optimize the liquid forming process of metal structures, guide the continuous casting process of continuous casting billets, including casting speed, superheat and pouring temperature, and greatly reduce cracking defects. Combined with machine learning, the multimodal mixed data including composition, melting and heat treatment of divergent grades are taken as features, and a random forest algorithm was used to find the non-linear pattern between the features and tensile strength. Then the chain equation multiple interpolation algorithms are used to predict the interpolation of the target missing data. The errors of both experimental and predicted values of tensile strength were within士5%; Additionally, the compositions are mapped to pseudo-2D periodic table, automatically extracting features via convolutional neural network for classification, simplifying prediction process and improving prediction accuracy, the prediction accuracy of intermetallic compounds exceeds 89%, solid solutions and amorphous phases exceed 98%.
Keywords:First-principle high-throughput calculation; Material genome engineering; Optoelectronic Semiconductor