Evaluating DC Partial Discharge With SF6 Decomposition Characteristics

This study mainly aims to obtain the decomposition characteristics of SF6 under various degrees of negative partial discharge (PD) and establish a method for evaluating the internal PD fault for gas-insulated equipment (GIE). Thus, metal protrusion, a frequent insulation defect in DC GIE, was used in the study of PD characteristics from the initial discharge to the near breakdown process. Then, seven parameters for characterizing the development stage of PD were extracted, and the fuzzy C-means clustering algorithm was used for the categorization of PD into three degrees, namely, slight, medium, and severe. We performed SF6 decomposition experiments on the PD degrees to obtain the decomposition characteristics. Results show that the SF6 decomposition characteristics (formation and content variation of five characteristic components, namely, CF4, CO2, SO2F2, SOF2, and SO2) are closely related to PD status. Fuzzy comprehensive evaluation theory was used for the establishment of a PD fault status assessment model, which was based on SF6 decomposition characteristics. Then, we evaluated the collected samples, and the accuracy of the evaluation method was up to 87.5%. This paper lays the foundation for future comprehensive assessment and fault diagnosis of insulation status of DC GIE on the basis of SF6 decomposition characteristics.