Optimized Process Parameters for Zn Alloy Die Casting via AI
摘要
Focusing on common issues in the zinc alloy vehicle-casting process, such as insufficient filling, shrinkage porosity, deformation, burrs, cold shuts, entrapped gas, oxides, and surface defects of cracks, this study takes the casting of zinc alloy toy streetcars as the object. It adopts CAE technology (FLOW-3D software) to simulate the casting process and combines the Central Composite Design (CCD) response surface methodology to optimize process parameters. Through the CCD response surface methodology, a response surface model was established between the shrinkage porosity rate, deformation amount, and the four process parameters—pouring temperature, mold temperature, filling speed, and holding pressure (T-T-v-P). While the CCD method can identify key influencing factors and verify the model significance via analysis of variance (ANOVA), it fails to achieve the optimal effect for multiple defects simultaneously. Therefore, further integrated analysis of the defects identified by the CCD method is required. Based on this, an AI model was established for verification to obtain the optimal parameter combination, and a production-guiding program template for quickly acquiring the optimal parameter combination was developed. After dual verification via CAE simulation and practical production, the parameters were confirmed to be reasonable and consistent. The optimization results show that under the optimal process parameters, the shrinkage porosity rate decreased by 33.3%, the deformation amount reduced by 45.1%, the product qualification rate increased by 14%, and the comprehensive performance reached the optimal level. The results of this study can provide references for mold designers and industrial production.