Prediction of springback in wipe-bending process of sheet metal using neural network

The wipe-bending is one of processes the most frequently used in the sheet metal product industry. Furthermore, the springback of sheet metal, which is defined as elastic recovery of the part during unloading, should be taken into consideration so as to produce bent sheet metal parts within acceptable tolerance limits. Springback is affected by the factors such as sheet thickness, tooling geometry, lubrication conditions, and material properties and processing parameters. In this paper, the prediction model of springback in wipe-bending process was developed using artificial neural network (ANN) approach. Here, several numerical simulations using finite element method (FEM) were performed to obtain the teaching data of neural network. The learned neural network is numerically tested and can be easily implemented springback prediction for new cases.