ABSTRACT
The detection of keyhole-induced pore positions is a critical procedure for assessing laser welding quality. Considering the detection error due to pore migration and noise interference, this research proposes a regional prediction model based on the time-frequency-domain features of the laser plume. The original plume signal was separated into several signal segments to construct the morphological sequences. To suppress the mode mixing caused by environmental noise, variational modal decomposition (VMD) was utilized to process the signals. The time-frequency features extracted from the decomposed signals were acquired as the input of a backpropagation (BP) neural network to predict the pore locations. To reduce the prediction error caused by pore migration, the effect of the length of the signal segments on the prediction accuracy was investigated. The results show that the optimal signal segment length was 0.4 mm, with an accuracy of 97.77%. The 0.2 mm signal segments failed to eliminate the negative effects of pore migration. The signal segments over 0.4 mm resulted in prediction errors of small and dense pores. This work provides more guidance for optimizing the feature extraction of welding signals to improve the accuracy of welding defect identification.
ABSTRACT
The currently available high-power laser shows promising opportunities for the welding of thick plates in a single pass. However, weld-root defect frequently occurs when a high-power laser is used to join thick plates in a full-penetration mode, which has a significantly adverse effect on the serviceability of the weld joint. The purpose of this work is to understand the defect formation mechanism and reduce these defects through controlling welding parameters. In this study, the characteristics of weld root defects were investigated using a 10 kW fiber laser using a program of experiment and theoretical analysis. The corresponding defect formation mechanisms were discussed based on the bottom molten pool behaviors observed by the high-speed camera. The results showed that there were four types of weld-root appearances as follows with an increase of linear heat input from 300 J/mm to 1000 J/mm: weld-root humping (30 mm/s), sound weld (25 mm/s), weld sagging (20 mm/s) and excessive weld sagging. The remedies for reducing weld-root defects were also presented to obtain sound weld bead by optimizing welding parameters. Weld-root humping was formed due to the quasi-full-penetration keyhole. Weld sagging resulted from the imbalance of the hydrostatic pressure and surface tension in the condition of a through keyhole. It was also found that the sound weld was formed when a through keyhole and a proper molten pool size were obtained. Thus, the state of the keyhole and molten pool geometry were the major factors that affect weld-root defects.
