Optimization of Multi-Pass Laser Bending by means of Soft Computing Techniques

The relatively small angle produced after each laser-bending pass and the long time lapse between consecutive passes highly limits the laser bending process productivity. In the present study, different strategies have been carried out in order to overcome such limitations, including forced cooling of sheets as well as optimization of process parameters. A series of experimental tests were conducted on thin sheets made of AISI 304 stainless steel. The variation of the bending angle achieved by multi-pass laser experiments, the minimum interval time between consecutive scans and the maximum temperature reached during the process were modelled by means of an Artificial Neural Network as well as the bend angle achieved per pass. Finally, an iterative procedure was developed to determine the optimal processing conditions that allow the production of a given bending angle. From an industrial point of view, the proposed procedure represents a preliminary version of a modulus for laser bending process automation and control.