Prediction of distortion induced by machining residual stresses in thin-walled components

Machining of thin-walled components is standard practice in many fields such as spaceflight, aviation, automobile, medical equipment manufacturing, etc. When these thin-walled components are machined, however, part distortions arise from machining-induced stresses resulting from high cutting forces and temperatures. In this paper, a method of predicting distortion induced by machining residual stresses in thin-walled components is proposed, which includes an empirical model for predicting machining residual stresses with different cutting parameters and a modified FEM model for predicting the resulted distortion. On the basis of the measured residual stress results, an exponentially decaying sine function is fitted using the particle swarm optimization method and the coefficients of the fitting function are regressed with cutting parameters. General FEM software ABAQUS is used to create and mesh the thin-walled component. Standard parts of the same material with the experimental samples are machined to make modification to the predicted residual stress profiles under the arranged cutting conditions. The modified residual stress distributions are applied into ABAQUS to calculate the distortion of the experimental samples. Two experimental samples are machined to validate the prediction methodology. The results demonstrate that the proposed method can significantly improve the distortion prediction accuracy.