Bending has significant importance in the sheet metal product industry. Moreover, the springback of sheet metal should be taken into consideration in order to produce bent sheet metal parts within acceptable tolerance limits and solve geometrical variation for control of manufacturing process. The Air bending process offers the advantage that many less tool changes are required as compared with others bending processes, however the calculation of the required punch displacement presents some problems. In this paper, several numerical simulations using finite element method were performed to obtain the teaching data required for training the neural network by means of the back-propagation algorithm. In the predictive mode different process inputs from the ones used in the previous stage were considered, for each case the springback angle and the displacement required to achieve a certain angle after springback are predicted by the learned network. Fairly accurate results were achieved for the punch displacement and for the springback angle evens so the range considered for training the network is large. The neural network can be easily implemented in experiment or in real production to determine the punch displacement to achieve a certain bend angle within a narrow range around the desired angle.
|Number of pages||6|
|Journal||Life Science Journal|
|Publication status||Published - 2013|
- Artificial neural network
- Finite element simulation
ASJC Scopus subject areas
- Biochemistry, Genetics and Molecular Biology(all)