Aluminum alloy welding defect online monitoring method

Abstract

The invention discloses an aluminum alloy welding defect online monitoring method, and belongs to the technical field of intelligent welding manufacturing. The device and the method aim at the on-line rapid monitoring requirement of the welding quality in manufacturing of high-end equipment large aluminum alloy components. A high-speed passive vision sensing system is used for collecting a molten pool image, a semantic segmentation method based on deep learning is used for simultaneously detecting feature areas, not covered by an oxide layer, of the edge and the surface of the molten pool, and on this basis, global features and local texture features of the morphology of the molten pool are extracted respectively; and early detection and identification of typical defects are realized through a machine learning model. On one hand, early detection and early warning of welding defects can be achieved, the defects can be positioned, and accidents such as shutdown caused by waste products due to the defects are avoided; on one hand, the welding seam forming quality can be recognized in real time to provide a basis for online feedback control of process parameters. The method can be widely applied to a thick plate and medium plate aluminum alloy helium arc welding process, and is particularly suitable for welding occasions of components such as large spaceflight aluminum alloy storage tanks and aluminum alloy liquid tanks.

Description

technical field [0001] The invention belongs to the technical field of intelligent welding manufacturing. The invention relates to an on-line monitoring method for aluminum alloy welding defects, which can be widely used in the welding manufacture of thick and medium-thick aluminum alloys of large-scale equipment such as aerospace propellant storage tanks and aluminum alloy liquid tanks. Background technique [0002] Helium tungsten arc welding is an important tungsten inert gas shielded welding technology. Compared with conventional argon arc welding, the high temperature and concentrated heat of helium arc under the same conditions enable helium arc welding to obtain greater penetration and melting efficiency. Therefore, helium tungsten arc welding is widely used in the welding of thick and medium plate aluminum alloys in the field of aerospace equipment and other high-end equipment manufacturing, such as the welding and manufacturing of large propellant tanks. [0003] ...

AI Technical Summary

Problems solved by technology

[0006] To sum up, none of the existing welding defect monitoring technologies at home and abroad is aimed at detecting defects in aluminum alloy helium tungsten arc welding, but only involves visual sensing for specific scenes and the use of traditional image segmentation algorithms to extract conventional geometric feature parameters combined with machine learning. The automatic detection technology of the classification model, there is no public report on the online monitoring method of aluminum alloy welding defects based on semantic segmentation combined with joint characteristic parameters of molten pool

Images