YU J X, DAI D. Defect location method of three-core submarine cable based on improved frequency domain reflection methodJ. Wire & Cable, 2026, 69(8): 1-12. DOI: 10.16105/j.dxdl.1672-6901.20260038
    Citation: YU J X, DAI D. Defect location method of three-core submarine cable based on improved frequency domain reflection methodJ. Wire & Cable, 2026, 69(8): 1-12. DOI: 10.16105/j.dxdl.1672-6901.20260038

    Defect Location Method of Three-core Submarine Cable Based on Improved Frequency Domain Reflection Method

    • Accurate localization of defects in three-core submarine cables before fault occurrence is essential for the stable operation of offshore wind power systems. An improved frequency domain reflection–based defect localization method is therefore proposed. A multi-conductor coupling model of a three-core submarine cable considering seawater penetration into the polypropylene outer sheath is established to obtain the sending-end frequency-domain reflection coefficient spectrum. A windowless full-phase inverse fast Fourier transform method based on double-spectrum-line correction is then applied for frequency–time conversion, effectively suppressing spectral leakage and the picket-fence effect and enhancing defect identification and localization capability. Simulation studies on a 1000 m, 220 kV three-core submarine cable demonstrate that the proposed method accurately identifies and locates both main insulation aging and external damage defects, with an average absolute localization error of less than 0.5 m. Compared with the conventional frequency domain reflection method, higher sensitivity to weak defects and more than 50% reduction in localization error are achieved in single-point defect scenarios, while all ABC-phase defects are successfully detected under a 50 dB signal-to-noise ratio in multi-point defect scenarios, with relative localization errors below 0.1%
    • loading

    Catalog

      Turn off MathJax
      Article Contents

      /

      DownLoad:  Full-Size Img  PowerPoint
      Return
      Return