Education, Science, Technology, Innovation and Life
Open Access
Sign In

Machine learning-based real-time detection of residual wall thickness in industrial pipelines

Download as PDF

DOI: 10.23977/autml.2024.050201 | Downloads: 2 | Views: 253

Author(s)

Yingtong Wan 1, Haozhe Jin 2, Songhang Wu 1, Junhao Zhao 1, Kan Zhu 1

Affiliation(s)

1 Qixin Honors School, Zhejiang Sci-Tech University, Hangzhou, 310018, China
2 School of Mechanical Engineering, Zhejiang Sci-Tech University, Hangzhou, 310018, China

Corresponding Author

Yingtong Wan

ABSTRACT

Far-field eddy current detection technology is a special eddy current technology that utilizes electromagnetic effect to detect the pipe through the wall. This paper introduces the principle of the far-field eddy current detection technology, including the two propagation modes of direct coupling and indirect coupling, and explains the division of different regions and their effects. The design and composition of the far-field eddy current detection device are then presented. In addition, this paper discusses the application of machine learning in analyzing far-field eddy current detection data, including the basic principles of machine learning, data preprocessing, data segmentation and model building steps. Finally, the evaluation indexes of machine learning models are introduced, as well as specific algorithms and evaluation indexes that may be used in far-field eddy current detection.

KEYWORDS

Far-field eddy current detection technology, Machine learning, Industrial Piping

CITE THIS PAPER

Yingtong Wan, Haozhe Jin, Songhang Wu, Junhao Zhao, Kan Zhu, Machine learning-based real-time detection of residual wall thickness in industrial pipelines. Automation and Machine Learning (2024) Vol. 5: 1-8. DOI: http://dx.doi.org/10.23977/autml.2024.050201.

REFERENCES

[1] Wu Weixin, Hou Huiwen, Shi Leyi. Research on industrial control intrusion detection based on deep learning and federated learning [J/OL]. Microelectronics and Computers, 1-9[2024-03-13]. 
[2] Jun Wang, Changfu Si,Kaipeng Wang et al. Intrusion detection method based on integrated learning and improved feature selection of PSO-GA algorithm [J/OL]. Journal of Jilin University (Engineering Edition), 1-9[2024-03-13]. 
[3] Wan Xuefeng. Analysis of damage forms of pressure pipelines for petrochemical industry and its non-destructive testing [J]. Manufacturing and Upgrading Today, 2023, (11): 188-190.
[4] Xihui Mou, Sixin Wang, Linsen Liao, et al. Discussion on on-line inspection and testing technology for non-stop transmission of industrial pipelines [J]. China Equipment Engineering, 2023, (19): 207-209.
[5] Liu Haizhao, Hu Shengzhong, Liang Haiming et al. Application of pulsed eddy current detection technology in industrial pipeline inspection [J]. China Special Equipment Safety, 2023, 39 (S2): 81-85.
[6] Hou Yining. Research on online inspection technology of industrial pipelines [J]. Product Reliability Report, 2023, (08): 119-121.
[7] Chen Youqiang. Application of ultrasonic testing in periodic inspection of industrial pipelines [J]. China Petroleum and Chemical Standards and Quality, 2023, 43 (14): 50-52.
[8] Guo Yue. Research on intrusion detection in industrial control networks based on improved CNN [J]. Mechanical Design and Manufacturing Engineering, 2023, 52 (06): 103-108.
[9] Liu Yuewen,Sun Ziwen. An enhanced model for IWSN intrusion detection against defense [J/OL]. Small Microcomputer Systems, 1-7 [2024-03-13]. 
[10] C.F. Li, H.T. Li, G.H. Gao. Damage patterns and inspection methods for high-pressure pipelines [J]. China Special Equipment Safety, 2023, 39 (03): 59-62.

Downloads: 2380
Visits: 92773

Sponsors, Associates, and Links


All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.