LAUSR

Conventional forging combined with subsequent heat treatments is a promising method for TA15 Ti-alloy to obtain a tri-modal microstructure with excellent mechanical properties. In this paper, a prediction model based on an improved back-propagation neural network was adopted to investigate the combinations of deformation temperature and degree at different strain rates and post-forging cooling modes. There exist reasonable combinations under a strain rate of 0.01 s−1 and air cooling or a strain rate of 0.1 s−1 and water quenching. The dependence of final microstructural feature parameters on forging parameters was obtained for the two cases. Targeting ideal tri-modal microstructure feature parameters, the allowable ranges of the forging parameters were obtained in reverse. The results show that the allowable ranges under a strain rate of 0.01 s−1 and air cooling are wider. This provides a guide to obtain a tri-modal microstructure by conventional forging combined with subsequent heat treatment during actual production.