Evaluation Methods and Coupled Optimization at Macro- and Micro-Scales for Profiled Ring Rolling of Inconel718 Alloy

The forming quality of profiled ring rolling not only encompasses macroscopic accuracy but also emphasizes the microstructure. Due to the multiple process parameters and complex metal flow during profiled ring rolling, the various forming defects are difficult to control and difficult to study theoretically. The objective of this study is to establish a comprehensive method for evaluating the forming quality of profiled rings, which considers both the macroscopic forming accuracy and the microstructure. Firstly, the synthetic size factor was defined, and the evolutionary relation between the section forming rate and the diameter growth rate of E-section ring rolling was analyzed in detail. The synthetic size factor can be used to describe the dimensional evolution and evaluate the forming accuracy of the profiled ring rolling process. Taking into full consideration the features of intermittent deformation in local areas, a microstructure evolution model of the Inconel718 alloy during E-section ring rolling, which can accurately predict the recrystallization volume fraction and average grain size of the final ring, was established. Then, combined with finite element simulation, the influence of the rotation speed of the driving roll on the macro-size evolution and microstructure was systematically analyzed. The results indicate that there is often a discrepancy between dimensional accuracy and microstructure uniformity in the optimization trend. For instance, the higher the rotation speed of the driving roll is, the more uniform the microstructure is, but the more difficult it is for the section profile to form. Finally, combined with response surface methodology (RSM), multi-parameter optimization was carried out with section forming accuracy and grain uniformity as the optimization objectives. By using the optimal parameters, an E-section ring with a complete profile and a uniform microstructure was obtained, with a maximum prediction error of the recrystallization volume fraction lower than 5%. The results show that the macroscopic and microscopic quality evaluation methods proposed in this study, as well as the optimization method combining RSM, can be effectively applied to the process optimization of profiled ring rolling.