Research Of Partial Discharge Detect System In High-voltage Cable Accessories Based On UHF

With the continued rapid development of electric power industry in China, power cable for its superior performance in power system plays an increasingly important role in the power system, and its operation is closely related with the safety of the grid. Failure of cables and accessories will be a direct result of paralysis of the whole transmission system therefore the running state of the cable accessories is also directly related to the stable operation of the grid. The degree of partial discharge is closely related to insulation condition of cables and accessories for insulating materials, thus the accurate detection of cable and accessories is the most direct and reliable judgment of the insulation quality.This paper has studied XLPE insulation material of power cables and accessories, designed a ring capacitive sensor, and built the corresponding circuit model. Through researches and analysis, it is proved that the sensor is suited for detection of partial discharge signals in the cables and their accessories. Besides, some major factors affected the sensor have been researched to optimize the design. The paper also has studied the sensor signal processing circuit, applied high-frequency processing software Advanced Design System, selected the appropriate circuit elements and the corresponding circuit model to design a low-noise signal amplifier with the operating frequency of600MHz-800MHz and band-pass filter circuit. Some main parameters of the circuit, such as circuit gain, stability factor and reflection coefficient have been simulated, and the performance of the circuit signal amplifying and filtering processing has been optimized. To improve its SNR, down-conversion mixer has been designed to reduce the frequency after the signal through the amplification filtering processing. The frequency of the signal has been reduced after mix frequency, which facilitate signal for further processing.At the same time, through a series of pre-processing the partial discharge signals inevitably have some interference noise signal. The paper has analyzed the applicability of wavelet transform in noise filtering, used wavelet thresholding method to process partial discharge signals containing white noise, and analyzed the effect of several different wavelets in signal processing. Simulation results have shown that wavelet analysis is applicable and effective in the partial discharge signal noise filtering process, even in strong interference noise, choosing appropriate thresholds and wavelet bases can achieve better de-noising results.