Abstract
Because welded structures such as automotive bodies have become smaller and lighter, a strong emphasis is being placed on the quality of the welds. Over the years many quality estimation systems have been developed by several researchers in order to ensure that welds are of high quality. The dynamic resistance signal, which is closely related to the nugget formation of the weld, has been used very effectively for this purpose, together with the electrode displacement signal. However, in previous research, the dynamic resistance was calculated using process parameters that were measured in the secondary circuit of the welding machine, causing many in-process problems. In this study, the process variables, which were monitored in the primary circuit of the welding machine, are used to obtain the variation of the dynamic resistance across electrodes. This allows the dynamic resistance monitoring system to be applied to the in-process system without any extra monitoring devices in the secondary circuit. Also, in order to test the reliability of such a system, an artificial intelligence algorithm to estimate the weld quality using the primary dynamic resistance is proposed.
Export citation and abstract BibTeX RIS