Abstract
The condition monitoring (CM) for the gas insulated switchgear (GIS) requires an accurate and reliable identification of the defective equipments in it for maintenance purposes. In this paper, a feature extraction procedure is explored, which is based on the power spectra density (PSD) of the de-noised partial discharges (PDs) emanating from the defective equipments in GIS. Furthermore, artificial intelligence techniques, in particular, the self organizing map (SOM) are investigated for the role as classifier to precisely identify these defective equipments, based on the PSD feature patterns. The performance of the SOM based classifier is ascertained by using the PDs acquired from practical GISs on South Korean 154 kV EHV transmission networks.
| Original language | English |
|---|---|
| Pages | 362-368 Vol.1 |
| Publication status | Published - 2004 |
| Event | Power Engineering Society General Meeting, 2004. IEEE - Duration: 1 Jan 2004 → … |
Conference
| Conference | Power Engineering Society General Meeting, 2004. IEEE |
|---|---|
| Period | 1/01/04 → … |
Keywords
- power spectra density
- partial discharge
- EHV transmission network
- feature extraction
- power engineering computing
- self-organising feature maps
- gas insulated switchgear
- 154 kV
- partial discharges
- defective equipments diagnosis
- artificial intelligence
- condition monitoring
- transmission networks
- self organizing map
- GIS
- artificial intelligence technique