Mathematical model of influence of oxygen and moisture on feature concentration ratios of SF6 Decomposition Products

Partial discharge (PD) is triggered by insulation defects in a gas-insulated switchgear (GIS), leading to the decomposition of SF₆ gas. The decomposition products of SF₆ can be used to detect and recognize PD. Three concentration ratios of SF₆ decomposition products, namely, c(SO₂F₂)/c(SOF₂), c(CF₄)/c(CO₂) and c(CF₄+ CO₂)/c(SO₂F₂+SOF₂), have been used as feature parameters in PD recognition. However, the values of these concentration ratios are not only related to the type of PD, they are also affected by the oxygen and moisture contents in GIS. In order to improve the performance of PD recognition, it is essential to calibrate the values by the influence rules of oxygen and moisture on feature concentration ratios. In the current paper, an experimental platform was constructed to simulate SF₆ decomposition under PD in a GIS. The concentration data of CF₄, CO₂, SO₂F₂ and SOF₂ under different moisture contents and oxygen contents were obtained. The data show that the values of c(SO₂F₂)/c(SOF₂), c(CF₄)/c(CO₂) and c(CF₄+ CO₂)/c(SO₂F₂+SOF₂) decrease with the increase of moisture and oxygen contents. By analysis of chemical dynamic theory, it is found that the influence rules of oxygen and moisture contents on the concentration ratios can be expressed by power function. The mathematical model was then developed by nonlinear regression. The predicting results show that the model has an acceptable performance.