|
Accession number;02A0423803
|
| Title;Feature extraction and pattern recognition of multi-source PD signals. |
| Author;
GAO K
(Tsinghua Univ., Beijing, Chn)
TAN K
(Tsinghua Univ., Beijing, Chn)
LI F
(Tsinghua Univ., Beijing, Chn)
GAO W
(Tsinghua Univ., Beijing, Chn)
|
Journal Title;Proceedings of the Symposium on Electrical and Electronic Insulating Materials and Applications in Systems
|
Journal Code:G0398B
|
ISSN:
|
|
VOL.33rd;NO.;PAGE.119-122(2001)
|
| Figure&Table&Reference;FIG.1, TBL.6, REF.2 |
| Pub. Country;Japan |
| Language;English |
| Abstract;Partial discharge signals from the industrial models of stator winding bars were used to compose the multi-source PD patterns, which were represented by 3-d charts. The tabulated data were calculated and served as characteristic features. Pattern recognition was performed using back-propagation(BP) network. Successful classification of multi-source PD signals showed that tabulated data were sensitive and stable to reflect different PD patterns. Suitability and robustness of BP network for classification was further verified by analyzing the untrained patterns' recognition. (author abst.) |
|
|
|
Related Articles;
|
