@article{oai:kyutech.repo.nii.ac.jp:00004277,
 author = {Ohtsuka, Shinya and 大塚, 信也 and 竹中, 賢司 and Kozako, Masahiro and 小迫, 雅裕 and Cho, Mengu and 趙, 孟佑 and 松久, 高士 and Hikita, Masayuki and 匹田, 政幸},
 issue = {6},
 journal = {電気学会論文誌. B, 電力・エネルギー部門誌},
 month = {Jun},
 note = {This paper describes insulation diagnosis of SF6 gas insulated switchgears (GIS) by detecting partial discharge (PD)-induced electromagnetic (EM) waves using UHF method. Sensitively detecting PD signals is effective to prevent breakdown in GIS because PD is a pre-breakdown phenomenon. Breakdown in SF6 gas under nonuniform field takes place when PD type proceeds from streamer to leader discharge. Therefore, breakdown prediction becomes possible by discriminating PD type, that is, by detecting generation of leader discharge. An amplitude of PD-induced EM wave depends on the time derivative of PD current. Experimental results revealed that the time derivative in rising part of PD current for streamer discharge is almost same as that for leader one while that in the falling part is quite different. Thus, one can recognize the generation of leader discharge by noting the ratio of the EM wave magnitude of the positive peak value to the negative one of the first incoming EM wave.In this study, an attempt was made to discriminate leader and streamer-type discharges on the basis of the amplitude ratio of first incoming EM wave measured with UHF sensors. Furthermore, we investigated effects of trigger level of a digital oscilloscope in the measurement of the EM wave and UHF sensor position on detection sensitivity and discrimination probability of PD type while changing the SF6 gas pressure and the applied ac voltage. As a result, it was found that leader discharge can be discriminated from streamer one with a probability as high as 70% using a criterion of amplitude ratio of first incoming EM wave VP+/VP-=0.7. When VP-=0.8 is used as criterion, the discrimination probability increases as high as 80%.},
 pages = {769--775},
 title = {UHF法で測定された電磁波第一波の振幅比に基づくGIS絶縁異常の診断},
 volume = {122},
 year = {2002},
 yomi = {オオツカ, シンヤ and コザコ, マサヒロ and チヨウ, メンウ and ヒキタ, マサユキ}
}