Weld joint forming prediction method based on complementary two-channel convolutional neural network

Abstract

The invention discloses a weld joint forming prediction method based on a complementary dual-channel convolutional neural network. Compared with a BP neural network, the convolutional neural network has the biggest characteristic that the extraction of molten pool characteristics is not needed, but the extraction of molten pool characteristic quantity is automatically carried out through a constructed multi-layer convolution kernel; the convolutional neural network takes the whole molten pool image as the input of the model, so that the time consumed for extracting the feature quantity of the molten pool is saved. Meanwhile, the loss of molten pool image information is avoided; compared with a common two-channel convolutional neural network laser welding seam forming prediction method, the method adopts two convolution modules to extract shallow layer features of the molten pool image to extract edge lines of the molten pool, and adopts a two-channel strategy, so that the obtained molten pool image features are more sufficient; laser welding process parameters are introduced by adopting a full-connection module to jointly predict the welding seam morphology of the T-shaped joint, so that the prediction performance of the model can be further improved.

Description

technical field [0001] The invention relates to a welding seam forming prediction method based on a complementary dual-channel convolutional neural network, which belongs to the technical field of material processing engineering. Background technique [0002] Lightweight, high-strength structural parts are widely used in aviation, aerospace, weaponry and other fields. Skin grid structure is an effective design form to achieve lightweight; skin grid structure has relatively high requirements on the quality of welded joints and the stability of welding process. High, destructive post-weld tests are difficult to meet the needs of real-time monitoring of welding quality; therefore, the quality monitoring of the welding process for non-visual T-joints of skin grids is the key to ensuring the consistency of joint quality. [0003] Laser welding is a technology that uses a pump source to excite the laser gain medium, and applies a high-energy-density laser heat source to the weldin...

AI Technical Summary

Problems solved by technology

[0006] The present invention aims to solve the existing problems of weld quality monitoring in the laser welding process, and proposes a laser welding seam shape prediction method based on complementary dual-channel convolutional neural network

Images