Design of a high-strength and high-temperature stress relaxation resistant copper alloy based on machine learning

Le résumé fourni par la source

Stress relaxation is a key factor contributing to connector failure in high-strength, high-elasticity copper alloys. However, the stress relaxation behavior is influenced by numerous complex factors, making it challenging to study effectively. In this study, a high-precision predictive model for the stress relaxation behavior of copper alloys was developed to accelerate stress relaxation prediction and optimize new alloys. Through algorithmic optimization, a Cu-Ni-Si-Mg-Mn alloy was developed that combines excellent mechanical properties with high resistance to stress relaxation. Experimental results show that the designed alloy contains a high density of dislocations and uniformly distributed nanoscale Ni 2 Si precipitate phases. After 100 h of service at 150 ℃ and 250 ℃, the stress relaxation rates of the Cu-Ni-Si-Mg-Mn alloy remained at 4.87% and 19.1%, respectively. The addition of Mg and Mn promotes the formation of Cottrell atmospheres and the Mn 6 Si 7 Ni 16 phase, which together exert a pronounced pinning effect on the motion of dislocations and substructures, thereby improving the stress relaxation resistance of the alloy. The high-precision stress relaxation behavior predictive model achieves an R 2 value of 0.96 and can effectively uncover the hidden nonlinear mathematical relationships in the stress relaxation process and is expected to partially replace traditional long-term stress relaxation experiments.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.