Study on the Metal Tip Partial Discharge in GIS with C4F7N/CO2/O2 Mixed Gas

Gas-Insulated Systems (GIS) have been extensively employed in the electrical industry due to their high reliability and low maintenance costs. However, traditional GIS predominantly uses SF6 as its insulating medium, which has drawn widespread attention due to its greenhouse gas effect on the environment. Consequently, the eco-friendly insulating gas, C4F7N, is progressively being regarded as a viable alternative. Nevertheless, research concerning the Partial Discharge (PD) characteristics of a mixture of C4F7N with CO2/O2 remains somewhat limited. In this context, this study establishes an experimental platform based on a complete GIS setup to investigate the time-domain and spectral features of ultra-high-frequency (UHF) electromagnetic and acoustic signals generated by tip defects within the C4F7N/CO2/O2 mixed gas GIS. Further, the distribution characteristics of PD signal features are explored. Experimental results reveal that compared to SF6-filled GIS, the PD signals from C4F7N/CO2/O2 mixed gas GIS display noticeable differences, particularly with respect to peak locations in the spectrum. From the distribution of feature parameters, a discernible distinction emerges between the PD signals in SF6 gas and those in C4F7N/CO2/O2 mixed gas, thereby establishing an effective means to distinguish discharge signals between the two gases. These findings offer invaluable guidance for optimizing the PD signal pattern recognition algorithms specific to C4F7N/CO2/O2 mixed gas GIS, enhancing the accuracy and monitoring efficiency of pattern recognition.