Defect prediction and performance optimization of A413 vertical centrifugal castings through FEA-based simulation

Abstract

Abstract This study benefits from advanced simulation based on finite element analysis, where finite element analysis possesses robust numerical algorithms to confront critical industrial challenges, including prediction and optimization techniques. The primary goal is to predict the location and quantity of shrinkage porosity and air entrainment, and to mitigate them by optimizing key casting process factors while considering the geometric effects. Virtual experiments designed through the Taguchi method, with three parameters (aspect ratio, pouring temperature, and mold rotation speed) at five levels each, were conducted using the finite element analysis software ProCAST. Post-experimentation, comprehensive data analyses, including signal-to-noise ratio and response surface methods, were employed to determine optimal conditions. Additionally, a sophisticated regression model was developed to clarify the complex relationship between key parameters and defect aspects. Optimal conditions to reduce shrinkage porosity were identified within parameter ranges of (50–75) rpm, (750–800) °C, and (1.75–2) aspect ratio. Similarly, optimal parameters to reduce air entrainment were identified as 150 rpm, 800 °C, and (1) aspect ratio.