Emission spectra and diagnosis of C4F7N/CO2 discharge faults

The C 4 F 7 N/CO 2 mixed gas is regarded as one of the most promising environmental-friendly alternatives to SF 6 . Studying its discharge spectral characteristics under different voltages and fault types is of great significance for insulation condition monitoring and fault diagnosis. Based on a 126 kV real-scale GIL experimental platform, this paper conducted DC breakdown, DC surface flashover, and AC surface flashover experiments for 10% C 4 F 7 N/90% CO 2 mixed gas at 0.1 MPa. By constructing a spectral data preprocessing method to eliminate the random interference of discharge, the emission spectral differences between C 4 F 7 N/CO 2 and SF 6 , as well as pure CO 2 gases, were systematically compared, and the spectral evolution laws under different conditions were analyzed. The experimental results show that, unlike the "low, flat, wide" characteristic of the SF 6 spectrum, the C 4 F 7 N/CO 2 mixed gas spectrum presents a "high, thin, steep" form, with energy concentrated in the ultraviolet and near-infrared bands. In the DC breakdown condition, the spectrum is mainly dominated by electrode evaporation (Al resonance line) and gas decomposition (CN, C, O spectral lines); in the DC surface flashover condition, due to the participation of insulator solid materials in discharge decomposition, the CN radical spectral band (385.0 - 387.6 nm) and CII ion spectral lines are significantly enhanced, while the Al and O spectral lines are relatively weakened; in the AC surface flashover condition, the reciprocating motion of charged particles changes the energy transfer mechanism, causing the spectrum to redshift towards the near-infrared region, with the O atomic characteristic peak (777.3 nm) becoming dominant, while the CN characteristic peak is weakened due to the high-frequency thermal accumulation effect. The study reveals the microscopic luminescence mechanism of the C 4 F 7 N/CO 2 mixed gas under the combined action of electrode evaporation, gas ionization, and material decomposition, confirming that changes in specific spectral line intensities can be used as criteria for distinguishing fault types and voltage levels, providing theoretical support for the optical diagnostic technology of environmentally friendly gas-insulated equipment.