Optimized Design of Cu‐Ni‐Co‐Si Alloy Combining Machine Learning and Experimental Feedback

ABSTRACT The machine learning method was employed to accelerate alloy design, and experimental feedback was provided to enhance predictive accuracy. A Cu‐2.7Ni‐1.0Co‐0.8Si alloy with superior properties was selected using a double‐objective optimization algorithm. The microstructure and properties of the designed alloy subjected to thermomechanical treatment were investigated systematically. The role of the key alloying element Co was discussed via first‐principles calculations. The results showed that the designed alloy exhibited favorable comprehensive properties, with a microhardness of 283 HV, a tensile strength of 1006 MPa, and an electrical conductivity of 41.3% IACS. The strength was mainly attributed to the combined effects of dislocation strengthening and precipitation strengthening, whereas the electrical conductivity was primarily limited by the solubility of solute atoms. The addition of an appropriate content of Co accelerated precipitation and increased the volume fraction of the precipitates, thereby synergistically enhancing the mechanical properties and electrical conductivity. This was attributed to the effect of Co on increasing the vacancy binding energy and decreasing the formation enthalpy of the precipitates.