Multiscale Casting Simulation for Aluminum Alloy Defect Prediction

The patent divides the casting system into multiple hierarchical scales (macroscopic casting geometry, mesoscopic dendritic structure, microscopic precipitates and defects). Each scale is modeled separately with appropriate simplifications, and the results are integrated through coupling relationships. This segmentation allows computationally efficient modeling at each scale while maintaining overall prediction accuracy by capturing scale-specific phenomena without requiring prohibitively complex full-scale models.

The patent merges component design and process modeling into an integrated computational framework where both aspects are simulated and optimized simultaneously. The coupled model allows iterative refinement of both component geometry and casting parameters together, enabling designers to evaluate how design changes affect microstructure and defects, and how process changes affect component performance. This integration eliminates the sequential approach, reducing development cycles while maintaining manageable complexity through systematic coupling of the design and process models.

The patent applies different levels of modeling detail to different regions and scales within the casting system. At the macroscopic level, continuum mechanics models capture overall heat and mass transfer. At the mesoscopic level, dendritic growth models capture tree-like solidification structures with appropriate detail. At the microscopic level, precipitate formation and defect nucleation are modeled with high detail only where relevant. This local quality approach ensures high microstructural accuracy where needed while avoiding prohibitive computational complexity by using coarser models in regions where fine detail is less critical.

The patent performs preliminary computational simulations of the casting process and microstructure formation before actual prototype fabrication. The integrated model predicts microporosity, dendritic structure, and defect distribution based on proposed casting parameters and geometry, allowing designers to identify and correct potential problems virtually. This preliminary computational action reduces the need for costly and time-consuming trial-and-error iterations in the foundry, improving both efficiency and reliability by ensuring designs are optimized before manufacturing.