Journals Articles Conferences News About CLP
Submit a paper Email Alert
Search
Search
Articles
Journals
News
Advanced Search
Top Searches
2D MaterialsTransformation opticsQuantum PhotonicsSmall lasersSemiconductor UV PhotonicsSupersymmetric microring laser arrays

Acta Optica Sinica, Volume. 45, Issue 10, 1012001(2025)

Pipeline Leak Signal Recognition Method Based on CNN-Transformer for DAS

Lang Li1,2, Zhongsheng Jiang1,2, Zhouchang Hu1,2, Shuang Yang2, Yuquan Tang2、*, Xingrong Jiang2,3, Miao Sun4, Zhirong Zhang1,2,3,5, Xiaoxia Qiu6, Shuai Wang6, and Chunfeng Hu6
Author Affiliations
  • 1University of Science and Technology of China, Hefei 230026, Anhui , China
  • 2Anhui Provincial Key Laboratory of Photonics Devices and Materials, Anhui Institute of Optics and Fine Mechanics, Hefei Institutes of Physical Science, Chinese Academy of Sciences, Hefei 230031, Anhui , China
  • 3School of Advanced Manufacturing Engineering, Hefei University, Hefei 230601, Anhui , China
  • 4School of Physics and Materials Engineering, Hefei Normal University, Hefei 230601, Anhui , China
  • 5Advanced Laser Technology Laboratory of Anhui Province, National University of Defense Technology, Hefei 230037, Anhui , China
  • 6Tangshan Xingbang Pipeline Engineering Equipment Co., Ltd., Tangshan 064106, Hebei , China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (11)
    Equations (0)
    References (25)
    Tools
    Paper Information
    Topics
    Figures & Tables(11)
    System architecture diagram

    Fig. 1. System architecture diagram

    Download full size
    Experimental site environment diagrams. (a) Layout of aerial and underground pipelines; (b) layout of underwater pipelines; (c) installation of sensor optical cable; (d) simulation of leakage holes; (e) site leveling; (f) water leakage experiment

    Fig. 2. Experimental site environment diagrams. (a) Layout of aerial and underground pipelines; (b) layout of underwater pipelines; (c) installation of sensor optical cable; (d) simulation of leakage holes; (e) site leveling; (f) water leakage experiment

    Download full size
    Time-domain waveform diagrams of signals under different leakage conditions. (a) Aerial gas leak; (b) aerial water leak; (c) underwater gas leak; (d) underwater water leak; (e) underground gas leak; (f) underground water leak

    Fig. 3. Time-domain waveform diagrams of signals under different leakage conditions. (a) Aerial gas leak; (b) aerial water leak; (c) underwater gas leak; (d) underwater water leak; (e) underground gas leak; (f) underground water leak

    Download full size
    Structure of proposed network

    Fig. 4. Structure of proposed network

    Download full size
    Network module details

    Fig. 5. Network module details

    Download full size
    Ablation experiment results

    Fig. 6. Ablation experiment results

    Download full size
    Comparative analysis of model validation effects. (a) Variation of validation accuracy for different models; (b) performance of 10-fold cross-validation

    Fig. 7. Comparative analysis of model validation effects. (a) Variation of validation accuracy for different models; (b) performance of 10-fold cross-validation

    Download full size
    Comparison of classification performance for different models

    Fig. 8. Comparison of classification performance for different models

    Download full size
    Comparative analysis of false positive rate and inference efficiency. (a) Comparison of false positive rates among different models; (b) inference time per batch for different models

    Fig. 9. Comparative analysis of false positive rate and inference efficiency. (a) Comparison of false positive rates among different models; (b) inference time per batch for different models

    Download full size
    t-SNE dimensionality reduction analysis. (a) Before extraction; (b) after extraction

    Fig. 10. t-SNE dimensionality reduction analysis. (a) Before extraction; (b) after extraction

    Download full size

    • Table 1. Experimental data details

      View table

      Table 1. Experimental data details

      TypeDescription of test scenarioType of test data
      Type 1Pipeline buried undergroundGas leak test
      Type 2Pipeline buried undergroundWater leak test
      Type 3Pipeline buried underwaterGas leak test
      Type 4Pipeline buried underwaterWater leak test
      Type 5Pipeline in an exposed settingGas leak test
      Type 6Pipeline in an exposed settingWater leak test
      Type 7Background noise in different settingsExcavator noise, human activity, etc.
    Tools

    Get Citation

    Copy Citation Text

    Lang Li, Zhongsheng Jiang, Zhouchang Hu, Shuang Yang, Yuquan Tang, Xingrong Jiang, Miao Sun, Zhirong Zhang, Xiaoxia Qiu, Shuai Wang, Chunfeng Hu. Pipeline Leak Signal Recognition Method Based on CNN-Transformer for DAS[J]. Acta Optica Sinica, 2025, 45(10): 1012001

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Set citation alerts for article
    Save article for my favorites
    Paper Information

    Category: Instrumentation, Measurement and Metrology

    Received: Jan. 10, 2025

    Accepted: Mar. 27, 2025

    Published Online: May. 14, 2025

    The Author Email: Yuquan Tang (laserway@aiofm.ac.cn)

    DOI:10.3788/AOS250471

    CSTR:32393.14.AOS250471

    Topics

    laser devices and laser physics
    Lasers and Laser Optics
    Laser physics
    laser manufacturing
    Instrumentation, Measurement and Metrology