Vision-based melt pool monitoring for wire-arc additive manufacturing using deep learning method

Wire-arc additive manufacturing (WAAM) technology has been widely recognized as a promising alternative for fabricating large-scale components, due to its advantages of high deposition rate and material utilization rate. However, some anomalies may occur during the process, such humping, spattering, robot suspend, pores, cracking so on. This study proposed apply deep learning in visual monitoring diagnose different WAAM process. The melt pool images were collected training validation by system. classification performance several representative CNN (convolutional neural network) architectures, including ResNet, EfficientNet, VGG-16 GoogLeNet, investigated compared. accuracy 97.62%, 97.45%, 97.15% 97.25% was achieved each model. results proved that models are effective classifying types WAAM. Our is applicable beyond should benefit other or arc welding techniques.