Signature images for arc welding fault detection

Abstract The present work elucidates the signature image approach to welding fault detection, covering the calculation of signature image data objects from blocks of welding electrical data (voltage and current), the definition of appropriate vector operations, and the manipulation of the signatures to permit detection of welding faults. Detection of out of position welds in overlap joints is illustrated, and the relevant changes in signature features are related to the physics of the short arc welding process. Finally, it is shown how fault detection performance can be improved by optimising the way the signatures are calculated.