Analysis Of Characteristics Of Metallic Plume During High-power Disk Laser Welding

With the rapid development of modern industrial manufacture, the laser welding become widely used in joint industry for its rapid, environmental and high-effective. Contrast to the traditional laser equipment, the high power disk laser has distinct advantages at the beam quality, processing efficiency, the weight and the volume of the equipment.During high-power disk laser welding, the surface of the weldment is gasified strongly by the laser irradiation. Meanwhile, the metallic plume is generated and changed into the plasma state. This plume can transmit through the laser beam reversely and has a significant shielding effect on the laser beam. Therefore, the laser power which should be transferred inside the weldment is decreased, the welding efficiency and welding quality are seriously affected. By studying and exploring the relation between the characteristics and variation rules of plume and the welding quality, the state of laser welding could be monitored in real-time by the plume characteristics.In this paper, the high-power disk laser bead on plate welding of Type304austenitic stainless steel plates at a continuous laser power of10kW, an extraviolet and visible sensitive high speed color camera was used to capture the metallic plume dynamic images. These digital images were transfered to the HSI (Hue-Saturation-Intensity) color spaces from the RGB color spaces. Also, the area of metallic plume and the path length which the laser beam runs through the metallic plume were segmented and defined as the plume eigenvalues. The fluctuations of weld bead width was used to evaluate the stability of welding quality. By analyzing the statistics of the variance and mean values of plume eigenvalues, the experimental results have confirmed that metallic vapor can reflect the laser energy absorption rate, the bigger metallic plume is the better welded quality is, and the path length can reflect the energy reduction, the longer path length is the worse welded quality is. At the same time, Time-frequency character recognition and hierarchical clustering analysis were used to study the relationship between metallic plume and weld bead width. The short-time Fourier transform was applied to obtain the time-frequency characteristics of plume images, and the hierarchical cluster was analyzed for the characteristics. Welding experimental results showed that there existed a relationship between the metallic plume and welding quality, the fluctuation trend of6th fitting of area clustering curve was similar to the6th fitting curve of weld seam width and the fluctuation trend of6th fitting of path length clustering curve was opposed to the6th fitting curve of weld seam width. Finally, the ratio of the metallic plume area and path length was analyzed, result showed that use median to filter the ratio data, it can achieve real-time monitoring of weld quality by setting certain threshold of the welded seam width and ratio date.