Optimizing the mechanical performance of A356–Sc–Sr alloy via combining machine learning and mechanical stirring under vacuum

In this study, a machine learning design system (MLDS) with a property-oriented optimization strategy was first established to predict the mechanical properties of the A356 alloys with adding Sc and Sr elements. Based on the experimental verification from the MLDS, the addition of 0.2 wt% Sc and 0.067 wt% Sr elements led to the refinement of α-Al grains and eutectic Si phases. Then, the vacuum–stirring was introduced to obtain the semi-solid microstructure of the A356–0.2Sc–0.067Sr alloy. The α-Al grains displayed the spherical morphology and the reduction in pores helped improve the mechanical properties of the alloy. In addition, the effects of stirring time and stirring temperature on the microstructure and mechanical properties of the alloy were investigated. The results demonstrated that the α-Al grains of the alloy were further spheroidized, resulting in the improved mechanical properties. The ultimate tensile strength and elongation of the alloy were 220 MPa and 6.0%, respectively, which were increased by 16.8% and 26.7% in comparison to those of the alloy without vacuum–stirring. The aim of this work is to provide a new method to prepare high-performance A356 alloy through composition design and microstructural control.