Finite element analysis of residual stresses and thin plate distortion after face milling

Abstract

Thin-walled structures are now widely used in aerospace structural components, turbines and many other applications, while the control of machining distortion has become one of the most difficult problems influencing their machining quality. Machining distortion is the result of many factors, among which the machining induced residual stress is an important issue. In this paper, a finite element analysis (FEA) approach of predicting machining induced residual stresses and the consequent distortion is presented, which includes the simulation of machining process, processing of the extracted residual stress data, loading of the fitted stress data and distortion analysis. Professional metal cutting simulation and optimization software, TWS AdvantEdge, was used for 3D machining process simulation. The constitutive model of work piece material was obtained from pressure bar tests and orthogonal cutting tests, and has been proved of good accuracy. General FEM software ABAQUS was used to create and mesh the thin-plate work piece. The meshing data outputted from ABAQUS and stress data calculated by AdvantEdge were processed with MATLAB, obtaining the stress distribution of each element, and then the stress distribution was input back into ABAQUS to calculate the distortion. Experiments were carried out, measuring the residual stress and distortion, to verify the FEM simulation. Under the designed simulation and experiment conditions, the machining induced residual stresses in feeding and peripheral directions are both tensile at the machined surface and fall off to a maximum compressive state with the increase in depth, after which the compressive stresses decrease and approach a steady value next to zero in near work piece substrate. Both experimental and FEA simulation results show that the release of the residual stresses make the thin plate to become convex. Yet there are still differences in conditions between simulation and experiment, and further will be carried out to improve the accuracy.