Weld Defect Detection of a CMT Arc-Welded Aluminum Alloy Sheet Based on Arc Sound Signal Processing

The cold metal transfer (CMT) process is widely used in thin plate welding because of its characteristics low heat input and stable arc. In actual production, a larger weld gap, misalignment, or other problems due to assembly error lead serious defects, such as burn-through lack fusion. arc sound contains wealth information related the quality weld. This work analyzes mechanism CMT generation, well correlation between time–frequency spectrum signal quality. paper studies extraction multi-channel an inputs it custom convolutional neural network for defect identification aluminum alloy plates. experimental result shows that average accuracy proposed model 91.49% arc-welded sheet, which higher than single-channel traditional classification models.