An image processing method for gas plug head characteristic parameters measurement and prediction in horizontal gas–liquid plug flow
Measurement Science and Technology · 29.05.2026 · 37 (22) · 225401
DOI: 10.1088/1361-6501/ae6dda ↗Kurz gefasst
Fachartikel in der Zeitschrift Measurement Science and Technology, online erschienen am 28.05.2026. Verfasst von Ning Zhao und 6 weiteren Autor:innen. Die Arbeit nennt 45 Literaturangaben. Gefördert von: Innovation Ability Cultivation Funding Project for Current Postgraduate Students, Hebei University, Cangzhou Natural Science Foundation, China, Hebei Natural Science Foundation, China.
Abstract (Originalsprache)
Gas–liquid plug flow is prevalent in various industrial processes, including oil extraction, and the flow state at the shale gas wellhead. In the petroleum industry, it is crucial to comprehend the flow structure characteristics of the gas–liquid two-phase plug flow, which influence the design of transportation and extraction systems and are closely tied to the safety and efficiency of the petroleum production process. A parameter measurement methodology utilizing phase interface segmentation is developed based on high-speed camera technology in the paper, enabling precise identification of the gas plug outline and extraction of gas plug characteristic parameters. By applying gradient-sensitive filtering technology, integrating Canny edge detection and the improved probabilistic Hough transform, the edge detection of the pipeline wall is achieved. A dynamic segmentation technology for the phase interface gas plug region is designed to achieve outline recognition and complete segmentation of the gas plug region. By converting the pixel points of the image, the characteristic parameters of the gas plug head, including length, height, and inclination angle, are measured. Based on decision tree regression, a prediction model for the length, height, and inclination angle of the gas plug head is established. The experimental results show that the mean absolute percentage errors of the prediction models for the length, height and inclination angle of the gas plug head are all less than 5%, while 90.91%, 98.18%, and 98.18% of the experimental measurement points fall within a ±5% relative error band respectively, achieving accurate prediction of the gas plug head characteristic parameters.
Förderung
- Innovation Ability Cultivation Funding Project for Current Postgraduate Students, Hebei University
- Cangzhou Natural Science Foundation, China
- Hebei Natural Science Foundation, China
- National Natural Science Foundation of China, China
Zitieren
APA
Zhao, N., Jia, H., Chen, K., Gao, L., Fang, L., Wen, J., & Zhao, Z. (2026). An image processing method for gas plug head characteristic parameters measurement and prediction in horizontal gas–liquid plug flow. Measurement Science and Technology, 37(22), 225401. https://doi.org/10.1088/1361-6501/ae6dda
Vancouver
Zhao N, Jia H, Chen K, Gao L, Fang L, Wen J, et al. An image processing method for gas plug head characteristic parameters measurement and prediction in horizontal gas–liquid plug flow. Measurement Science and Technology. 2026;37(22):225401. doi:10.1088/1361-6501/ae6dda
BibTeX
@article{Zhao2026,
author = {Zhao, Ning and Jia, Haodi and Chen, Keke and Gao, Lei and Fang, Lide and Wen, Jiaqi and Zhao, Zhiyue},
title = {An image processing method for gas plug head characteristic parameters measurement and prediction in horizontal gas–liquid plug flow},
journal = {Measurement Science and Technology},
year = {2026},
volume = {37},
number = {22},
pages = {225401},
doi = {10.1088/1361-6501/ae6dda}
}
EUVIMED