Hybrid deflection prediction on machining thin-wall monolithic aerospace components

Abstract

Thin-wall machining of monolithic parts allows better quality parts to be manufactured in less time. This brings advantages, particularly in inventory management and manufacturing efficiency. However, due to poor stiffness of thin-wall parts, deformation is more likely to occur in the machining process, which results in dimensional form errors. This paper describes a new methodology for prediction of wall deflection during machining thin-wall features with reduced analysis time from weeks to hours. The prediction methodology is based on a combination of the finite-element method and statistical analysis. It consists of a feature-based approach to parts creation, finite-element analysis of material removal, and statistical regression analysis of deflection associated with cutting parameters and component attributes. The prediction values have been validated by machining tests on titanium parts and show good agreement between simulation model and experimental data.