Abstract
Nowadays firms are required to obtain high quality products in order to increase their competitiveness. The time required to obtain a new product is also essential to fight the concurrence. For manufacturers of bent parts, accurate prediction of the springback is very important. Therefore, this paper investigates the applicability of artificial neural networks (ANN) and adaptive neuro-fuzzy inference system (ANFIS) to predict the springback in the free cylindrical bending process of metallic sheets. The finite element method (FEM) was used to simulate the springback in the free cylindrical bending process and the results were used as training data for ANN and ANFIS. The finite element results were validated by comparison with experimental data. Statistic criteria were used to evaluate the performance of the developed ANN and ANFIS models. It was found that the predictions are in good agreement with the FEM data.
Highlights
Sheet metal bending operation is one of the most frequently used forming procedures in the production of metallic parts
∑ mean relative error (MRE)(%) = 1 N pi − ai ×100 N i=1 ai where: a is the actual value from experiments, p is the predicted value by models and N is the number of patterns
The accuracy of the developed artificial neural networks (ANN) and adaptive neuro-fuzzy inference system (ANFIS) models was evaluated by considering the Root Mean Square Error (RMSE) Eq (5), correlation coefficient (R) Eq (6) and Mean Relative Error (MRE) Eq (4)
Summary
Sheet metal bending operation is one of the most frequently used forming procedures in the production of metallic parts. One of the factors that affect the quality of bent parts is springback. Springback can be defined as an elastically-driven change of shape of a deformed product which takes place during removal of external loads. It is a complex physical phenomenon which is mainly governed by the stress state obtained at the end of a deformation [1]. The amount of springback depends on many factors including the material properties (Young’s modulus, Poisson’s ratio, strength coefficient, yield stress, strain hardening and anisotropy coefficients), bending operation (tooling geometry, pad force, clearance, initial bend angle and bend radius, contact friction condition etc). In order to obtain parts at required dimensions and tolerance limit is very important to determine as accurate as possible the springback amount
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
