Quality Evaluation Method of Anchor Chain Flash Butt Welding Based on Deep Learning

(0 User reviews)   530   134
Jiahe Gao, Haibo Wen, Shenao Zhu, Shihui Dong, Shijie Su, Jian Zhang
School of Mechanical Engineering, Jiangsu University of Science and Technology, Zhenjiang, China

International Journal of Mechanical Engineering and Applications (Science Publishing Group) 2023
11 : 1
1-8
10.11648/j.ijmea.20231101.11
English
Jiahe Gao, Haibo Wen, Shenao Zhu, Shihui Dong, Shijie Su, Jian Zhang. Quality Evaluation Method of Anchor Chain Flash Butt Welding Based on Deep Learning, International Journal of Mechanical Engineering and Applications. Volume 11, Issue 1, February 2023 , pp. 1-8. doi: 10.11648/j.ijmea.20231101.11. Share Research.
Abstract
Flash butt welding, a mainstream welding method employed in producing anchor chains, is a critical manufacturing process affecting the quality of anchor chains. Ultrasonic and load testing are used to evaluate the welding quality of anchor chains, but the cost of checking and replacing unqualified chain links is high. A deep learning-based quality evaluation method for flash butt welding is proposed to reduce the cost of detecting and replacing substandard chain links. First, displacement and current sensors collect electrode position and current signals during welding. Second, since the number of qualified anchor links is much larger than that of unqualified ones, a new data synthesis method is proposed: nearest-neighbor splicing sampling, which achieves the enhancement of minority samples by segmenting and combining existing data samples according to the features of anchor chain welding. Then, a piecewise linear interpolation method is used to handle the varying data length problem, thus satisfying the input requirements of the convolutional neural network (CNN). Finally, a CNN model is established, and dropout is used to reduce the over-fitting phenomenon. The experimental results show that the accuracy of the under-sampling method, over-sampling method, and nearest-neighbor splicing sampling method are 93.8%, 95.9%, and 96.3%, respectively, and the sensitivity, specificity, and accuracy of the CNN model are 95.7%, 93%, and 94.3%, respectively, which are better than those of the support vector machine (SVM).
Deep Learning, Quality Evaluation, Anchor Chain, Flash Butt Welding, Nearest-Neighbor Splicing Sampling

Flash butt welding, a mainstream welding method employed in producing anchor chains, is a critical manufacturing process affecting the quality of anchor chains. Ultrasonic and load testing are used to evaluate the welding quality of anchor chains, but the cost of checking and replacing unqualified chain links is high. A deep learning-based quality evaluation method for flash butt welding is proposed to reduce the cost of detecting and replacing substandard chain links. First, displacement and current sensors collect electrode position and current signals during welding. Second, since the number of qualified anchor links is much larger than that of unqualified ones, a new data synthesis method is proposed: nearest-neighbor splicing sampling, which achieves the enhancement of minority samples by segmenting and combining existing data samples according to the features of anchor chain welding. Then, a piecewise linear interpolation method is used to handle the varying data length problem, thus satisfying the input requirements of the convolutional neural network (CNN). Finally, a CNN model is established, and dropout is used to reduce the over-fitting phenomenon. The experimental results show that the accuracy of the under-sampling method, over-sampling method, and nearest-neighbor splicing sampling method are 93.8%, 95.9%, and 96.3%, respectively, and the sensitivity, specificity, and accuracy of the CNN model are 95.7%, 93%, and 94.3%, respectively, which are better than those of the support vector machine (SVM).
disclaimer

There are no reviews for this Publication.

0
0 out of 5 (0 User reviews )

Add a Review

Your Rating *
There are no comments for this Publication.
You must log in to post a comment.
Log in

Related Publications