Research on hot deformation behavior of Cu-Ti alloy based on machine learning algorithms and microalloying

High strength and high elasticity copper alloys have good application prospects due to their excellent mechanical properties. The deformation and heat treatment in its preparation process are important steps to ensure the service performance. In this work, Cu-Ti based alloys were employed and hot compression experiments were conducted at 500-800℃ under four different strain rates. Their constitutive equations and thermal processing maps were constructed. The influence of microalloying elements on activation energy was also analyzed. The microstructure evolution and recrystallization behavior under hot deformation conditions were studied based on the results of TEM. At the same time, Arrhenius, back propagation neural network (BPNN) and support vector machine (SVM) algorithms were used to predict the flow stress of the alloy. The results indicate that the relationship between flow stress and the input factors (strain rate, temperature and deformation) can be well predicted using the support vector machine algorithm. Microalloying can increase the activation energy of copper alloys, and the sensitivity of activation energy to temperature changes is related to the interaction between precipitates and dislocations.