Al/Mg bimetallic composites have drawn considerable attention for their promising lightweight applications in sectors such as the aerospace and automotive industries. In these systems, the interfacial behavior critically governs the overall performance and reliability. In this research, the molecular dynamics (MD) simulation was employed to systematically study the effects of pouring temperatures (923 K, 973 K, and 1023 K) and preheating temperatures (373 K, 473 K, and 573 K) on the interfacial diffusion behavior and fracture mechanism of the polycrystalline Al/Mg bimetallic system. The results indicate that the influencing rule of pouring temperatures and preheating temperatures on the interfacial diffusion behavior is consistent. Specifically, the diffusion coefficient of Mg atoms is higher than that of Al atoms, while the diffusion distance of Al atoms is significantly greater than that of Mg atoms. As the temperature increases, the thickness of the interfacial transition layer correspondingly rises. However, the effects of these two parameters on tensile fracture behavior demonstrate notable discrepancies. Specifically, the fracture mode evolves with pouring temperature, transitioning from being mediated solely by dislocations to being co-mediated by twins and dislocations. In contrast, the fracture mechanism remains solely dislocation-controlled, regardless of the preheating temperature. In addition, all the models fractured at the interface between the diffusion layer and the Mg matrix. The optimal tensile strength of 1.850 GPa was achieved at a pouring temperature of 923 K and a preheating temperature of 473 K, representing an improvement of approximately 52% compared to the lowest value recorded in the study. This research offers significant theoretical insights for the rational optimization of preparation parameters and an in-depth understanding of fracture mechanisms in Al/Mg bimetallic systems.

This study investigates a heavy-duty CNC machine tool crossbeam casting manufactured by a leading heavy machine tool producer. A numerical simulation model for the demolding process was developed using proprietary Computer-Aided Engineering (CAE) software. The experimental validation of the residual stress was performed using the blind-hole method on the guide rail mounting surface. The simulation results were compared with experimental data, revealing that the post-demolding simulations exhibited smaller fluctuations than the pre-demolding predictions. The maximum principal stress prediction resulted in an absolute error of 11.8%, effectively reflecting the residual stress distribution for casting design and production optimization.