Improving Accuracy in Aluminum Incremental Sheet Forming of Complex Geometries Using Iterative Learning Control

Incremental Sheet Forming is a flexible process characterized by low costs and higher process times with respect to traditional forming technologies. It is therefore suitable for prototypes, small series or custom mass productions. Its flexibility derives from the use of a hemispherical punch that is moved by a CNC machine and gradually deforms the sheet in presence, or not, of a counter die. As a consequence, the sheet clamping is reduced and the part accuracy is lower than traditional sheet forming process as stamping. Therefore, the improvement of the part accuracy in Incremental Sheet Forming is a relevant research topic and solutions for error reduction are required for improving the process quality. The present paper describes the use of an Iterative Learning Control (ILC) algorithm for compensating the ISF part geometrical error. In particular, it iteratively corrects the part geometry on the basis of the error map obtained as the difference between formed and target part geometries. The ILC uses the target geometry to form a first trial part, it measures the obtained geometry and estimates the geometrical error map. Then the error map is used to modify the target geometry and another part is formed. This procedure gets iterated until the desired geometrical tolerance is achieved. The correction algorithm was experimentally tested in forming both axisymmetric and not axisymmetric parts using aluminum sheets. Results showed that in few iteration steps it was possible to significantly improve the part accuracy and to achieve geometrical tolerances comparable with the traditional sheet forming processes.