Online welding quality monitoring based on feature extraction of arc voltage signal

Arc sensing plays a significant role in the control and monitoring of welding quality for aluminum alloy pulsed gas touch argon welding (GTAW). A method for online quality monitoring was proposed based on the analysis of acquired arc voltage signal, through which two algorithms of feature extraction were developed in time and frequency domain, respectively. In time domain, the wavelet packet transform was carried out to eliminate the pulse interference of the feature parameter curve. In frequency domain, the other new algorithm was proposed based on the voltage power spectrum density (PSD) which was calculated by using the improved Welch algorithm and divided into five frequency bands before the statistic parameters were extracted. The correlation between the feature parameters in different frequency bands and welding defects were carefully analyzed to select a more sensitive one as the monitoring parameters. The proposed algorithms on this paper were verified to be capable of detecting lack of penetration, burn through, and the defect caused by lack of gas.