The Research On Partial Discharge Fault Diagnosis Technology In GIS

Gas insulated switchgear (GIS) has the advantages of compact structure, high reliability, good safety, installation period shortly, easy maintenance, is not influenced by the external environmental factors, and has been applied in electric power system widely. It is effectively diagnosis and evaluation of the running status of GIS by SF6 gas component, because F6 gas are insulation and arc-extinguishing medium; Partial discharge is an important characterization of GIS early insulation defects, and can predict and diagnose the early fault to avoid further deterioration which affects the reliability of power supply failure. Therefore, it is of great significance to study the fault diagnosis technology of GIS gas and partial discharge based on SF6.GIS SF6 gas effects of temperature, wind speed and operation of the GIS equipment of environmental factors on the SF6 gas leakage, puts forward a novel method of detecting GIS leakage. The method can be adapted to the current in the load fluctuation and environmental change, and gas detection SF6 switch equipment leakage rate. The text selects SF6 gas decomposition products group SOF2/SO2F2, CF4/CO2, (SOF2+SO2F2)/(CO2+CF4) for fault diagnosis of partial discharge, and uses the least square curve fitting method for SO2F2 gas ideal. SOF2 and SO2F2 can be used to diagnose partial discharge from metal tips, suspended solids or metal particles.Partial discharge in GIS is used to diagnose partial discharge from metal tips, suspended solids or metal particles, and summed up the discharge characteristics of UHF and ultrasonic detection method, which provides judgment for the internal discharge. The FDTD can simulate and calculate the UHF electromagnetic signals produced by the partial discharge, and compared to the actual measured spectrum.GIS fault diagnosis method based on least squares support vector machine, established the LS-SVM model and pattern recognition classifier, and recognized the pattern of partial discharge from metal tips, suspended solids or metal particles. Development of fault diagnosis procedure based on least squares support vector machine in Matlab is illustrated the correctness by exapmle.Synthetic fault diagnosis of equipment based on GIS optimal weights would combine the resource of ultrasonic detection method, UHF detection method and SF6 decomposition method. A new fault diagnosis method having optimal weights is combined by partial discharge from metal tips, suspended solids or metal particles. The C language program provides a good platform for the integrated fault diagnosis method based on the optimal weights of GIS equipment.