Authors: Marek S. Weglowski
Addresses: Instytut Spawalnictwa, Bl.Czeslawa Street 16/18, Gliwice 44-100, Poland
Abstract: It has been presented the results of investigation spectrum of the electric arc light. The main goal of these researches was to check that welding arc light can be seen as a signal carrying essential information about the welding process and exploiting in the monitoring of the welding process. Designed and made experimental station. The advanced spectrophotometer was used to data acquisition of spectrum of arc light. The investigations were conducted on the automated TIG welding station. For each welding parameter the arc light spectrum was measured. Research results presented in this paper indicate that the welding arc light emission contains a lot of information concerning the course of the welding process. That signal is much more sensitive than the signals recorded in the electric circuit of the welding arc to the changes of welding conditions and should be used as a tool for monitoring of the TIG welding process. Spectrophotometer is based on reflection grating and CCD array with 1028 elements (pixels) in the visible spectral range of 340 to 860 nm. It is an expensive instrument and that can be used only as a complementary tool in sensing of welding processes. The investigation results are indicative of the next steps of researches on the welding arc radiation phenomenon and the possibilities of using this signal for online sensing of the welding process on fully automated and robotised stands. This monitoring system will be particularly attractive for welded-structures manufacturing industry because it could significantly reduce the cost for post weld analysis and repairs. Quite new approach to monitoring of welding processes replaces |trough the arc methods| of monitoring by non-conventional methods. Three fitting functions: Lorentz, Gauss and Voight were investigated as a means to simulate the spectrum distribution of the electric arc light. The mathematical-physical model of the arc light emission and neural networks were compared. [Received 15 November 2006; Accepted 13 March 2007]
Keywords: TIG welding; arc light intensity; arc length; spectrum; electric arc light emission; welding monitoring; simulation; neural networks; modelling.
International Journal of Computational Materials Science and Surface Engineering, 2007 Vol.1 No.6, pp.734 - 749
Published online: 23 Apr 2008 *Full-text access for editors Access for subscribers Purchase this article Comment on this article