Abstract
This study aimed to optimize the grain structure of complex thin-walled nickel-based superalloy castings by investigating the influence of key casting parameters using both cellular automaton-finite element (CAFE) simulations and experimental validation. The main problem addressed was the inhomogeneous grain morphology arising from complex mold geometries and uneven thermal conditions during investment casting. The solidification process was simulated using the ProCAST software, incorporating the CAFE method to model temperature fields and grain growth dynamics. The results revealed that the molten metal flow pattern during mold filling significantly affected the local temperature field and subsequent grain formation. Specifically, simultaneous bidirectional filling minimized thermal gradients and suppressed coarse columnar grain formation, promoting finer, more uniform equiaxed grains. Lowering the pouring temperature (to 1430 °C) in combination with reduced shell temperature (600-800 °C) enhanced nucleation and improved grain uniformity in thin-walled regions. Higher cooling rates also refined the grain structure by increasing undercooling and limiting grain growth. Experimental castings confirmed these simulation outcomes, demonstrating that the proposed optimization strategies can significantly improve grain homogeneity in critical structural areas. These findings provide a practical approach for controlling microstructure in large, intricate superalloy components through targeted process parameter tuning.