An efficient prediction method for the dynamic deformation of thin-walled parts in flank milling

Abstract

Severe deformation may occur in the flank milling of thin-walled parts due to their low rigidity. Considering the varying stiffness resulting from the moving cutting force and the material removal effect, finite element-based methods have been developed to compute the deformation and generate compensated tool paths. However, they are time-consuming since they require solving the full-scale deformation equation at each cutting location. Therefore, an efficient deformation prediction method is firstly proposed in this study, in which the workpiece stiffness matrix is reduced and dynamically modified to avoid re-meshing meanwhile decreasing the computational scales of the solving process. Then, considering the coupling effect between the compensation value and the deformation error, a compensation strategy without the iteration process is developed to obtain the compensation value with high accuracy and efficiency. The flank milling of four comparison sets of thin-walled aluminum alloy blocks was conducted to validate the proposed method. The experiment results revealed that the proposed method obtained similar accuracy to the finite element methods, but cost less than 62.0% of their computational time. And the machining error ranges of four comparison sets were reduced by more than 58.7% after compensation. The proposed method shows great potential for efficiency and accuracy improvement in the 5-axis flank milling of thin-walled freeform surface parts. • A novel deformation prediction method for the milling of thin-walled parts is proposed to improve computational efficiency. • Compared with other FEM-based methods, the proposed method obtains similar accuracy but greatly decreases computation time. • Considering the coupling effect between the compensation values, a compensation strategy without iterations is developed. • The proposed method is validated through milling experiments, in which the machining errors were greatly reduced.