Optimization of micro hardness of nanostructure Cu-Cr-Zr alloys prepared by the mechanical alloying using artificial neural networks and genetic algorithm

Cu–Cr-Zr alloys had wide applications in engineering applications such as electrical and welding industrial especially for their high strength, high electrical as well as acceptable thermal conductivities and melting points. It was possible to prepare the nano-structure of these age hardenable alloys using mechanical alloying method as a cheap and mass production technique to prepare the non-equilibrium materials such as solid solution structures. In this study, artificial neural networks (ANNs) program was developed to establish the relationship between the practical parameters of mechanical alloying, i.e., weight percentages of Cr and Zr as alloying element, milling times, milling speed, sintering time and temperature, on the micro hardness of prepared Cu-Cr-Zr nanostructure alloys. The results of sensitivity analysis showed that the alloying elements and sintering temperature had the highest and lowest effect on the micro hardness of products, respectively. Also, the optimum milling speed and sintering temperature proposed as 255-291 rpm and 530-590°C, respectively. The established models of ANN introduced to genetic algorithm (GA) for determination of the optimal condition. The results were evaluated using the confirmation experiments. Moreover, the optimal condition of nanostructures alloy preparation with the highest micro hardness had been proposed as 310 Hv with the root mean square error (RMSE) of lower than 3.4%.