Gas-dependent multi-sensor partial discharge response pathways in SF 6 and SF 6 /N 2 gas insulation

Le résumé fourni par la source

Abstract SF 6 /N 2 mixtures are regarded as an important transition solution for reducing the use of SF 6 in gas-insulated equipment, but the change in gas composition may also modify the partial discharge process and its measurable signatures. Understanding how the same defect is expressed in different sensing channels is therefore essential for reliable partial discharge diagnosis in SF 6 -reduced insulation systems. In this work, partial discharge characteristics of a gas-cavity defect in SF 6 and a 30%SF 6 /70%N 2 mixture were investigated using synchronized ultra-high-frequency (UHF), high-frequency current transformer (HFCT) and acoustic emission measurements, together with electron-collision calculations based on BOLSIG+. The results suggest that under the same defect geometry, pressure range and fixed measurement configuration, pure SF 6 and the 30%SF 6 /70%N 2 mixture exhibit distinct apparent multi-sensor response states. In SF 6 , the discharge current response is more temporally compact, leading to a shorter HFCT formative duration and a stronger UHF projection relative to the HFCT response. In the SF 6 /N 2 mixture, the current response becomes more extended, while the relative UHF projection decreases, indicating that SF 6 dilution changes the way in which the discharge is coupled into electromagnetic and current-response channels. Event-level coupling analysis further confirms that SF 6 and SF 6 /N 2 occupy distinct response states under gas-cavity defect. The calculation results show that the critical reduced electric field for the transition from attachment-dominated behaviour to net ionization decreases from approximately 354 Td in SF 6 to approximately 287 Td in SF 6 /N 2 , accompanied by changes in mean electron energy and electron drift velocity. These results provide supporting kinetic evidence for interpreting the observed response differences in terms of the ionization-attachment balance and electron transport behaviour, and offer a useful basis for gas-adaptive feature extraction and recognition in SF 6 /N 2 -insulated equipment.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.