Quantitative analysis of the influence of regularity of SF6 decomposition characteristics with trace O2 under partial discharge

To obtain the influence regularity of trace O₂ on SF₆ decomposition characteristics under partial discharge (PD) and establish the foundation of using SF₆ decomposed components to diagnose the insulation status of SF6 electrical equipment, first, a series of PD decomposition experiments with different O₂ concentrations (%)were conducted on a special experiment platform. The variation regularity between the concentrations of characteristic components and the different content of O2 were obtained quantitatively. Second, the correlation coefficient r testing method was applied to identify the associated influence of trace O₂ on the decomposed characteristic components and characteristic ratios. Based on this, an exponential regression model of the related characteristics was developed. Finally, the influence mechanism of trace O₂ on SF₆ decomposition progress was explained reasonably. The results show that under the PD caused by metal protrusion insulation fault without organic solid insulated materials, CO₂, SO₂F₂, and SOF₂ were the mainly decomposed components in the specific range of trace O₂. Although no obvious influences of the trace O₂ on the concentrations of CO₂ and SOF₂ as well as the ratio c(CO₂)/c(SO₂F₂+SOF₂), it is notable that the trace O₂ is negatively and positively related to the concentration of SO2F2 and the ratio c(SOF₂)/c(SO₂F₂), respectively. The exponential regression model can minimize the effect of trace O2 on the fault diagnostic result when using SF6 decomposed components.