A Defect Monitoring Method for Power Cables Based on EWT in Frequency Domain

Power cables have been widely used in urban distribution network systems, and the detection of cable defects has attracted much attention of the researchers. The detection signal of the cable has the problem of mode mixing caused by the discontinuous time-frequency scale, which leads to the spurious peak of the traditional frequency domain reflectometry (FDR), and makes the detection of defects cause interference. In this paper, a cable defect state detection method based on empirical wavelet transform (EWT) is proposed. First, EWT is used to adaptively decompose FDR signals, because it has good identification characteristics for modal signals. Then analyze the correlation between the decomposed signal and the original signal, and use the threshold to reconstruct the original signal, so as to weaken the influence of spurious peaks on the detection. Finally, the effectiveness of the method proposed in this paper is verified by simulation and experiment under multi-reflection signals.