Electric arc additive manufacturing layer width and surplus height cooperative control method based on deep learning

Abstract

The invention relates to an electric arc additive manufacturing layer width and surplus height cooperative control method based on deep learning, and belongs to the technical field of precision welding. The method comprises the following steps of step 1, acquiring molten pool images through an electric arc additive manufacturing detection system, and extracting layer width and excess weld metal data of a welding seam; step 2, based on the layer width and excess weld metal data extracted in the step 1, constructing a weld layer width and excess weld metal regression model based on deep learning; and step 3, inputting the images of the positive and lateral molten pools collected in real time into the regression model to obtain the expected layer width and surplus height of the welding seam, and adjusting the welding current and controlling the layer width and surplus height of the welding seam through an active-disturbance-rejection control algorithm. The welding current in the step 3 is controlled by a welding power source, and the welding power source is connected with a computer through a control cabinet. According to the method, the cladding layer offset in the additive manufacturing process is quantitatively predicted by utilizing visual information of the molten pools, so that the actual position of a welding gun is adjusted in the additive manufacturing process, a good molten pool form is obtained, and the welding quality is improved.

Description

technical field [0001] The invention relates to a method for collaborative control of layer width and reinforcement in arc additive manufacturing based on deep learning, and belongs to the technical field of precision welding. Background technique [0002] Additive manufacturing technology has been widely used in the fields of automobile, aerospace, architectural design and biomedicine, etc. Arc additive manufacturing technology uses welding arc as fusion energy to melt metal wire, which has higher material utilization rate compared with other additive manufacturing methods High, high deposition efficiency, environmental protection and other characteristics have attracted extensive attention in recent years. However, in the process of arc additive manufacturing, as the number of cladding layers increases, there will be problems such as serious heat accumulation, poor heat dissipation conditions, overheating of the molten pool, difficulty in solidification, and difficulty in ...

AI Technical Summary

Problems solved by technology

However, in the process of arc additive manufacturing, as the number of cladding layers increases, there will be problems such as serious heat accumulation, poor heat dissipation conditions, overheating of the molten pool, difficulty in solidification, and difficulty in controlling the shape of the cladding layer. The robot monitors the welding situation in real time and adjusts the welding process parameters in time to improve the welding quality

In the current research, the real-time feedback control in the arc additive manufacturing process is still relatively insufficient.

Images