Xiao Wu, Qiang Wang, Zhimin Zhang, Jian Xu, Mei Cheng, Xianwei Ren
In this study, a 15 wt% B 4 C/7093Al composite was fabricated via powder metallurgy, and its hot deformation behavior and microstructural evolution were systematically investigated at temperatures ranging from 300 °C to 500 °C and strain rates between 0.001s −1 and 10 s −1 . The composite exhibited pronounced sensitivity to both temperature and strain rate. At a given temperature, dynamic recovery dominated at lower strain rates, while dynamic recrystallization was markedly enhanced at higher rates. A back-propagation neural network optimized using the Sparrow Search Algorithm (SSA–BP) was established and demonstrated excellent predictive accuracy for flow stress (R 2 = 0.996). Electron backscatter diffraction (EBSD) analysis revealed a distinct transition from recovery-dominated to recrystallization-dominated mechanisms with increasing temperature. The B 4 C particles exerted a dual regulatory effect—promoting recrystallization through particle-stimulated nucleation (PSN) while simultaneously inhibiting grain growth via the Zener pinning effect—resulting in a refined grain size range of 2.07–7.58 μm. Furthermore, the incorporation of B 4 C significantly weakened texture intensity (f(g) < 2.5), leading to a more randomized orientation distribution. Transmission electron microscopy (TEM) characterization confirmed the dynamic precipitation of η' (Al 4 Mg 2 Zn 3 , AlZnMgCu) and η (MgZn 2 ) phases during deformation. These findings provide a scientific basis for optimizing hot working parameters and establish a solid theoretical foundation for the industrial application of B 4 C-reinforced aluminum matrix composites.