Power Cable Defects Location Based on Improved Time-frequency Domain Reflectometry

Power cables have been widely used in urban power distribution systems, and cable defects location has become a problem of concern to many scholars. For the traditional time-frequency domain reflectometry (TFDR) there are cross-term interference and other issues. This paper proposed an improved time-frequency domain reflectometry. The method selected a specific linear frequency modulation (LFM) reference signal and obtained a Wigner-Ville distribution (WVD) of the reflected signal to analyze the time-frequency characteristics. However, there was cross-item interference in the WVD, so a new cross-item suppression method was proposed in this paper. By analyzing the characteristic differences between the self-item and cross-item of the signal time-frequency distribution, the cross-item interference was eliminated without affecting the signal resolution by affine transformation (AT) and other signal processing methods, and the defects location were precisely determined by the time-frequency cross-correlation function of the signal. In this paper, the 10kV cross-linked polyethylene (XLPE) cable was simulated and analyzed, and the 105m XLPE cable with defects and the 500m XLPE cable with joints were experimented. The simulation and experimental results show that the proposed method is more accurate than the conventional method and has better robustness. © 2021 Chin. Soc. for Elec. Eng.