Search

After analyzing the various interferences when monitoring the Gas Insulated Substation (GIS) partial discharge, this paper puts forward a hierarchical sequential model, which can reduce the disturbance of partial discharge signal. The hierarchical sequential method, based on the Fourier series and wavelet transform method, can effectively and accurately suppress the white noise and periodic narrowband interference, so as to extract the useful partial discharge signal from the strong interference background. The method is simple, flexible and easy to apply to the other occasions. In the meanwhile, the subsequent data and simulation both verify the accuracy of the method, which will have wide range of application.

The converter transformer is a key hub device connecting DC and AC in the converter substation, and its safe operation is directly related to the reliability level of the power grid. According to a report issued by CIGRE, the failure rate of converter transformers has increased year by year. Insulation failure of converter internal connections, outlet bushings and valve side windings accounted for 65% of the total failure rate. Once the converter transformer fails, it will inevitably lead to the blocking of the converter valve, causing local or even large-scale power outages, will inevitably cause incalculable serious economic losses. The main performance of the internal insulation failure of the converter is the partial discharge failure. The failure of the casing is mainly the failure of the internal insulation of the casing, the oil leakage of the casing and the grounding of the casing end screen. In this paper, using the integrated online monitoring technology of converter and casing based on multi-sensor fusion and polarity identification method to realize the monitoring and early warning of converter and casing failure, and to discover the delay fault and defect hidden danger of casing in advance, so as to improve the safety and reliability of transformer equipment.