Page 11 - Read Online

Page 4 of 28 Zhang et al. J Mater Inf 2024;4:34 https://dx.doi.org/10.20517/jmi.2024.64

Table 1. The chemical composition table of titanium alloy TA2
Material Fe C N H O Ti
TA2 0.027 0.0063 0.0047 < 0.001 0.10 Margin

Figure 1. The technical roadmap of AI agent assistant smart welding consisted of five basic sections, including large language model,
auto-coding via AI agent, image processing, image mosaic and machine learning for welding defect detection. AI: Artificial intelligence.

The intelligent system is equipped with the TPR 2000 welding machine, 500A programmable power supply,
real-time data acquisition system and vision system . In the vision system, the Xiris XVC-1100 high
[56]
dynamic range (HDR) camera is employed to observe the welding process with a maximum frame rate of 55
fps . High-speed imaging provides a direct approach to capturing the melting dynamics of the welding
[21]
wire and the flow behavior of the molten pool [57,58] . The camera position is refined through multiple
experiments to capture high-quality images of the molten pool . The input/output channels of the camera
[59]
are equipped with photoelectric isolation to protect against electromagnetic noise. As a full factorial
experimental design requires testing all possible parameter combinations which is both time-intensive and
costly, the orthogonal experimental method is employed to efficiently limit the number of tests while
ensuring experimental validity . Table 2 shows the parameter values for each experimental sample,
[60]
including pulse current (I ), welding speed (V ), pulse width (t ), duty cycle ratio (δ) and weld seam
p
s
p
thickness at the arc starting point (H).
Deep learning methods
BPNNs constitute a class of multi-layer feedforward neural networks trained through the error
[61]
backpropagation algorithm . As depicted in Figure 3, the typical architecture of BPNNs comprises three