Cable defect localization method based on windowed chirp-Z transform of broadband impedance spectroscopy

Accurate localization of defects in de-energized cross-linked polyethylene power cables is important for condition assessment and maintenance. Conventional frequency-domain reflectometry methods based on the inverse fast Fourier transform and discrete Fourier transform may generate redundant spatial components and interference peaks when processing densely sampled broadband impedance spectroscopy data. This paper proposes an offline defect localization method combining broadband impedance spectroscopy, the Chirp-Z Transform, and a Kaiser window. A distributed-parameter model is first established to characterize the broadband impedance response of a cable with a localized defect. The measured impedance spectrum is then converted directly into a spatial localization spectrum. The Chirp-Z Transform enables flexible control of the localization interval and output sample number, while the Kaiser window suppresses spectral leakage and sidelobe interference. For a 100 m simulated cable, the localization accuracy ranges from 99.48% to 99.91%. For a 15 m laboratory cable, it ranges from 99.29% to 99.84%. In a field test on a 450 m, 10 kV distribution cable, a defect at 236.1 m is located at 233.59 m, with an absolute error of 2.51 m and an accuracy of 98.93%. Compared with conventional methods, the proposed approach reduces redundant spatial components and improves localization-spectrum clarity. The measured broadband impedance spectrum also provides auxiliary information on dominant defect characteristics and relative severity. The method is intended for offline localization of defects that cause measurable broadband impedance variations.