Compressed Sensing-Based Frequency Domain Reflection Method for Defect Localization in Power Cables

Frequency-domain reflection (FDR) is widely used for local defect localization in cables. However, spectral leakage and noise lead to sidelobes and spurious peaks in the FDR spectrum. Additionally, due to the difficulty in determining testing parameters, FDR often requires large amounts of data, making data transmission and storage inconvenient. To address this, this paper proposes a compressed sensing (CS)-based method for defect localization in power cables using FDR. The core idea is to treat the cable's time-domain reflection signal as a sparse signal. The method involves the following steps: first, preprocessing and compression of the raw signal; then, obtaining the sparse representation of the signal using the discrete Fourier transform (DFT) basis and a reconstruction algorithm. Next, the signal is reconstructed using the linear frequency modulation Z transform (CZT) basis, resulting in a sparse signal with improved localization accuracy. To validate the proposed method, the capacitance value was adjusted in the simulation to simulate different defects in the cable. In the field experiment, a <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$10 \text{kV} / 823$</tex> m power cable was selected, with local defects introduced at <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{4 2 0}$</tex> m and 620 m. Both simulation and experimental results confirm the method's effectiveness.