Upper Sweeping Frequency Selection for Cable Defect Location Based on STFT

• Published in: IEEE transactions on instrumentation and measurement Vol. 72; p. 1
• Main Authors: Huang, Jingtao, Zhou, Kai, Xu, Yefei, Meng, Pengfei, Tang, Zhirong, Liang, Zhongying
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.01.2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Analytical models; Attenuation; cable defect; Cross-linked polyethylene; Defects; Fourier transforms; Frequencies; Frequency domain reflection; Frequency ranges; Mathematical models; optimization positioning; Power cables; Reflection coefficient; short-time Fourier transform; Spatial data; Sweeping; Time-frequency analysis; upper sweeping frequency selection
• ISSN: 0018-9456; 1557-9662
• DOI: 10.1109/TIM.2023.3271718

A frequency domain reflection (FDR) method has been widely used to locate cable defects. Generally, the performance of the FDR method can be enhanced by extending the sweeping frequency range of the signal. However, with the increase in the signal frequency, misdiagnosis can easily occur owing to the attenuation of the signal and noise. Thus, this paper's authors propose the upper sweeping frequency (UF) selection method of FDR to improve the accuracy of cable defect location. First, the cable's three-dimensional location spectrum, including the frequency and spatial information, is obtained at a wide frequency range based on short-time Fourier transform with excellent time-frequency aggregation characteristics. Second, the reasonable UF is selected through the energy change characteristics of the cable end. Then, the optimal localization result of the cable defect is calculated by interpolating the sum of energy among the effective frequency bands. Finally, two single-and multi-defect cases are used for the YJV-8.7/10kV XLPE power cable in the experiment to demonstrate the performance of the proposed method. The experimental results show that with the UF selection method, the impact of high-frequency signal attenuation on defect location is eliminated, and cable defect can be localized with more accuracy.