Abstract
For the problem of classification and identification of defects in polyethylene (PE) gas pipelines, this paper firstly performs preliminary screening of the acquired images and acquisition efficiency of defective image acquisition was improved. Images of defective PE gas pipelines were pre-processed. Then, edge detection of the defective images was performed using the improved Sobel algorithm and an adaptive threshold segmentation method was applied to segment the defects in the pipeline images. Finally, the defect images were morphologically processed to obtain binary images. The obtained binary images were applied with VGG16 to complete the training of the defect classifier. The experimental findings show that in the TensorFlow API environment, the test set’s highest accuracy reached 97%, which can achieve the identification of defect types of underground PE gas transmission pipelines.
Funder
National Natural Science Foundation of China
Xinjiang University Doctoral Start-up Foundation
Science and technology planning project of State Administration for Market Regulation
Subject
Fluid Flow and Transfer Processes,Computer Science Applications,Process Chemistry and Technology,General Engineering,Instrumentation,General Materials Science
Reference41 articles.
1. Investigation of Polyethylene Pipeline Behavior after 30 Years of Use in Gas Distribution Network;J. Mater. Eng. Perform.,2020
2. Fracture behavior of pre-cracked polyethylene gas pipe under foundation settlement by extended finite element method;Int. J. Press. Vessel. Pip.,2021
3. Velázquez, J.C., Hernández-Sánchez, E., Terán, G., Capula-Colindres, S., Diaz-Cruz, M., and Cervantes-Tobón, A. (2022). Probabilistic and Statistical Techniques to Study the Impact of Localized Corrosion Defects in Oil and Gas Pipelines: A Review. Metals, 12.
4. Modified leakage rate calculation models of natural gas pipelines;Math. Probl. Eng.,2020
5. Safety assessment of buried natural gas pipelines with corrosion defects under the ground settlement;Eng. Fail. Anal.,2021
Cited by
4 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献