EUVIMEDEUVIMED entdecken
Crossref

Pipeline Leakage Detection With Multisensor Fusion Based on Support Vector Machine

Xinyao Liu, Mingye Yin, Mengxiao Zhang, Rui Yang, Yanbing Li

IEEE Sensors Journal · 2026-08 · 26 (15) · 23049-23063

DOI: 10.1109/jsen.2026.3704869 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Sensors Journal, 2026-08. Verfasst von Xinyao Liu und 4 weiteren Autor:innen. Die Arbeit nennt 39 Literaturangaben. Gefördert von: National Key Laboratory of Space Target Awareness, National Radar Signal Processing Laboratory.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Pipeline leakage, often induced by corrosion or environmental factors, poses significant risks of severe environmental pollution and substantial economic loss. To improve the reliability and robustness of leakage detection in complex industrial environments, this study proposes a multisensor feature-level fusion method based on a support vector machine (SVM). Initially, a discrete wavelet transform (DWT) is applied to denoise the raw monitoring signals collected from acoustic, pressure, and flow sensors. Subsequently, a sliding window technique is utilized to segment these signals, from which a diverse set of representative features is extracted from both the time and frequency domains. These extracted features are then consolidated into a comprehensive fused feature set, which is used to train the SVM classifier for accurate leak identification. Evaluated on a dataset derived from actual pipeline operations, the experimental results indicate the effectiveness of the proposed method in leak detection. Detailed comparative analyses with representative single-sensor algorithms and other multisensor fusion models demonstrate that the proposed approach provides more stable and discriminative detection performance.

Förderung

  • National Key Laboratory of Space Target Awareness
  • National Radar Signal Processing Laboratory

Zitieren

APA

Liu, X., Yin, M., Zhang, M., Yang, R., & Li, Y. (2026). Pipeline Leakage Detection With Multisensor Fusion Based on Support Vector Machine. IEEE Sensors Journal, 26(15), 23049-23063. https://doi.org/10.1109/jsen.2026.3704869

Vancouver

Liu X, Yin M, Zhang M, Yang R, Li Y. Pipeline Leakage Detection With Multisensor Fusion Based on Support Vector Machine. IEEE Sensors Journal. 2026;26(15):23049-23063. doi:10.1109/jsen.2026.3704869

BibTeX
@article{Liu2026,
  author = {Liu, Xinyao and Yin, Mingye and Zhang, Mengxiao and Yang, Rui and Li, Yanbing},
  title = {Pipeline Leakage Detection With Multisensor Fusion Based on Support Vector Machine},
  journal = {IEEE Sensors Journal},
  year = {2026},
  volume = {26},
  number = {15},
  pages = {23049--23063},
  doi = {10.1109/jsen.2026.3704869}
}
Kostenlos starten →