基于非色散红外原理的 GIS 内部SO₂ 含量高精度检测方法

If faults such as overheating, partial discharge, or arc interruption occur during the operation of gas insulated switchgear (GIS) equipment, SF₆ gas will react with other substances or decompose. The concentration of SO₂ generated by different discharge faults decomposition is common between 0～1 000 μL/L. Therefore, it is feasible to use SO₂ as a detection gas and judge the status or defects or faults of SF₆ equipment based on the concentration of SO₂. However, commercial SO₂ sensors with different sensing principles can not take into account the detection accuracy, manufacturing cost, equipment volume and other aspects. In order to solve the above problems, this paper designs a portable SO₂ detection system with short response time, high sensitivity, detection limit width, low cross sensitivity, and temperature compensation, which can detect SO₂ gas in the SF₆ background inside GIS. This article first explores the temperature drift correction method under air background, measures the peak to peak output voltage of the detector at concentrations of 0～1 000 μL/L and temperatures of 0～40℃, proposes the concept of correction coefficient at different temperatures, and establishes a relationship model between the correction coefficient, detector output peak to peak, and SO₂ concentration. The average prediction error of the model in the air background is 1.2%, and compared with before temperature drift correction, the corrected sensor detection error can be reduced by 92.1%. Due to the high cost of conducting long-term SO₂ temperature drift compensation experiments under the SF₆ background, and the fact that the greenhouse effect of SF₆ is 23 500 times that of CO₂ gas, it severely limits the practical application and promotion of this technology in terms of economy and environmental protection. Therefore, the next consideration is to migrate the temperature compensation method under the air background to the SF₆ background for detection. In the experiment, it was found that even if the absorption spectra of SF₆ and SO₂ overlap, the pyroelectric sensor can still generate further response and be used for SO₂ concentration calibration due to the saturation of high-purity SF₆ absorption. The average error of the SO₂ concentration calibration formula is only 2.82% when the compensation coefficient is transferred from the air background to the SF₆ background. Compared with before temperature correction, the corrected sensor detection error is reduced by 71.0%, and the temperature drift correction effect is significant. In addition, H₂S and CO are important components specified in the maintenance regulations for switchgear, in order to ensure that the SO₂ detection device will not be affected by the above gases, cross sensitivity experiments were conducted on the three gases: H₂S, CO, and SO₂. The average response error of SO₂ before and after mixing a single interfering gas was 4.5%, and the average response error of SO₂ before and after mixing two interfering gases was 5.1%. Finally, on-site discharge experiments were simulated, Set a breakdown discharge voltage of 15 kV for GIS equipment, collect gas samples from 20, 40, 60, and 80 discharges, and use non-dispersive infrared (NDIR) sensors and gas chromatographs to detect the measured gas separately. The NDIR sensor produces a significant step response, and the response result is inverted into the concentration of SO₂. Compared with the detection result of the gas chromatograph, the average error of the two is only ±3.8 μL/L.