Air bending of sheet metal based on the grey prediction model

Owing to the intercoupling of multiple factors, strong nonlinearity exists during the bending of sheet metals, enabling the inevitable springback of sheet metal parts after bending unloading. Consequently, the bending forming angle becomes difficult to control, thereby seriously affecting the accuracy of the bending forming parts. A prediction model for the press amount of the upper mold and the bending forming angle was proposed in this study to address the shortage of bending accuracy and reveal the relationship between the bending forming angle and the press amount of the upper bending mold. First, elliptic fitting was used to improve the insufficient fitting of the background value of the original grey prediction GM (1,1) model and reduce the intermediate fitting error of the grey prediction model. Second, the modeling method was adopted at the head and the tail to reduce the long-term fitting error of the model and establish the mathematical model of the press amount of the upper mold and the forming angle in segmented bending based on the experimental bending data and in accordance with the actual working conditions. Finally, the feasibility of the model was verified through field experiment and ABAQUS simulation. Results show that the maximum fitting error of the prediction model for the bending angle obtained through the aforementioned method is only 0.7%, proving the model's capacity in accurately predicting the bending springback angle. The study provides a new method for controlling the bending springback of sheet metals. © 2019 Eastern Macedonia and Thrace Institute of Technology.