Prediction of critical fraction of solid in low-pressure die casting of aluminum alloys using artificial neural network

Casting simulation programs are computer programs that digitally model the casting of an alloy in the sand, shell, or permanent mold and, then, the cooling and solidification processes. However, obtaining consistent results out of casting modeling depends on the incorporation of many accurate parameters and boundary conditions. Critical Fraction of Solid (CFS), which is one of the most important of these parameters, is defined as the point where solid dendrites do not allow any flow of the liquid metal in the mushy zone. Since the CFS value varies based on many factors, inconsistent results can be experienced in the modeling applications. In this study, the CFS value obtained during the solidification of various commercial aluminum alloys' casting process, carried out using low-pressure die casting method, is predicted by using Artificial Neural Network (ANN) method based on alloy type, grain refiner and modifier additions, initial mold temperature, and pressure level parameters. In the scope of the study, 162 experiments are conducted. The results obtained from the low-pressure die casting experiments using a special model designed for the study are validated by using SOLIDCast casting simulation. The CFS values are obtained in this validation range of 33% to 61%. These CFS values are used in the development of ANN models. Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and Mean Square Error (MSE) are used to assess the prediction accuracy of the ANN models. Calculated values of MAE, MAPE, and MSE are 0.0188, 7.06%, and 0.0006, respectively. The results show that the proposed ANN model predicts a CFS value with high accuracy.