Welding with laser beams is an innovative optical technique, which leads to higher penetration depth and a narrower seam compared to conventional welding techniques. Significant criteria of the quality of a junction besides detected faults are the penetration depth and the seam width. Within this article optical sensors for process monitoring as well as a predictive control scheme based on these are presented. In the closed loop control the process´ inputs laser power and focal position are optimised by taking the future welding speed into account. For modelling the physical demanding, non-linear process an Artificial Neural Network with external dynamics is applied. First results of the application on a real laser welding system are described.
Print ISSN: 0178-2312
Volume: 53, 10/2005
Pages: 513 - 521