Determination of the amount of the spring-back in the sheet metal products using experimental and finite element methods and their estimation by means of neural networks

It is known that the sheet metals in the bending moulds are subject to some amount of spring-back depending on the shape deformation after bending. It is also a fact that the the mentioned spring-back amount is a parameter necessary for making the mould design in a short duration and reaching the desired dimensional fineness. In this project, a study is conducted upon the determination of the amount of the spring-back in the bending moulds both experimentally and using finite element method. Designing a modular free V bending mould to determine the spring-back amount, Dkp and stainless steel which are extensively used in the automotive industry, defence industries, and several industrial applications are selected as the materials for the experiments. As the thickness values of the specimens are chosen to be 1, 2, 3, and 4 mm, for each different thickness value, at least three bending operations are carried out for each specimen regarding the bending angles of 15, 30, 45, 60, 75, 90, 105 and 120 degrees with three different radius values corresponding to each angle. After checking the dimensional fineness of the produced specimens using an optical projection device and determination of the of the relevant spring-back quantities, these values are compared to the results of the analyses performed in the Marc-Mentat software used in the computer-aided finite element studies. Furthermore, the results of the constructed artificial neural network structures are also obtained and the yielded results are discussed. This way, the experimental and estimation results are contributed to the literature by means of this publication.