High-throughput directed energy deposition-based manufacturing combined with machine learning to fabricate gradient-composition Cu-Fe-Cr alloys

Abstract Fabrication of gradient-composition alloys is a significant challenge because the adjustment of chemical composition and physical performance is essential. The directed energy deposition manufacturing provided a high-throughput fabrication method for gradient Cu-Fe-Cr alloys, on a master alloy substrate. The liquid diffusion of Fe and Cr contributed to the gradient variation in microstructure and composition. Followed decomposition precipitation improved the hardness of as-fabricated Cu-Fe-Cr alloys, which increased from 84.6 HV to 100 HV. Machine learning was employed to predict the hardness of Cu-18.6Fe-5Cr alloy at different annealing time, the experimental results agreed well with the prediction. These findings indicated that high-throughput manufacturing combined with machine learning can accelerate the design and fabrication of gradient-composition copper alloys.