Xin Lin1,2,3, Kunpeng Zhu1,2,3,*
The International Conference on Computational & Experimental Engineering and Sciences, Vol.30, No.2, pp. 1-1, 2024, DOI:10.32604/icces.2024.012499
Abstract This paper reveals the interplay mechanism between melt pool, spattering and vapors, aiming to further improve the forming quality through in-situ monitoring with a CMOS camera. A Residual Network based on Convolutional Block Attention Module and Focal loss function is proposed to extract multi-scale features of single tracks and learn about their behavior changes. A t-SNE clustering analysis is utilized to analysis a large amount of time sequence data on the melt pool by collecting the schlieren photographs. It is found that patterns of unstable melt pool changing corelate to the defects in single tracks, More >