Prediction of spring-back and bend force in air bending of electro-galvanised steel sheets using artificial neural networks

Abstract

Air bending is a frequently used process for sheet metal fabrication. Spring-back and bend force are the two essential parameters to be predicted for the design of tooling and selection of press. In air bending of electro-galvanised steel sheets, these two parameters are affected by the factors such as strain hardening exponent, coating thickness, die opening, die radius, punch radius, punch travel and punch velocity. This paper presents the development of a predictive model of spring-back and bend force using artificial neural network (ANN) approach. A central composite design of 88 experiments is conducted to obtain the data required for training the network. A new set of data combinations, not belonging to the training data, is used to assess the generalisation performance of the neural network. The network is validated with a set of validation data and it is found that ANN is capable of predicting the two parameters.