Efficient prediction of milling distortion using inversely-identified inherent strains

Abstract

Abstract Predictive milling distortion has been widely used to optimize milling design and process. Inherent strains method (ISM) applies plastic strains calculated from small-scale thermo-mechanical simulations to large-scale models to assess residual deformation in less time. In this paper, ISM is proved feasible in milling process by theoretical and experimental analysis. Moreover, a novel method called inverse identification is proposed to obtain milling inherent strain efficiently. To verify the proposed method, two milling cases are presented to evaluate its accuracy in distortion prediction. In addition, comparisons with the thermo-mechanical model and traditional ISM model indicate that the inversely-identified inherent strains can significantly reduce the simulation time at the premise of similar precision, which is practical to be applied in industrial applications.