Springback in Sheet Metal U Bending-Fea and Neural Network Approach

All the forming processes are more or less prone to the springback depending upon different material properties and process parameters. Bending processes are very widely used in the manufacturing of sheet metal products, particularly in automobile industry. It is a well known fact that when sheet metal is formed, it does not form the desired shape because of the springback. In the present paper the springback is predicted by using finite element analysis, for various die radii, sheet thicknesses, R/t ratios and strength coefficients, for two different materials. The simple neural network is used to map the springback obtained from finite element analysis. The results obtained by FE simulations compared with simple neural network are found in good agreement.