Feature extraction and pattern recognition of signals radiated from partial discharge

Author

Liu Weidong ; Liu Shanghe ; Hu Xiaofeng

Author_Institution

Electrostatic & Electromagn. Protection Res. Inst., Mech. Eng. Coll., Shijiazhuang, China

fYear

2009

fDate

16-20 Sept. 2009

Firstpage

114

Lastpage

117

Abstract

The artificial neural networks based BP algorithm is used to recognize two typical discharge patterns, corona and spark. In order to have a comparison, feature extraction based on waveform parameter and time-frequency analysis were used separately to provide the training input. The results show that the highest average recognition rate based on waveform parameter reaches 92.5%, while this based on time-frequency is 95%. On the contrary, the lowest average recognition rate based on waveform parameter is 70%, while this based on time-frequency is 90%. This indicates that time-frequency analysis is more effective and more suitable for discharge pattern recognition.

Keywords

backpropagation; corona; feature extraction; neural nets; partial discharges; sparks; BP algorithm; artificial neural networks; corona; discharge patterns; feature extraction; partial discharge; pattern recognition; spark; time-frequency analysis; waveform parameter; Artificial neural networks; Corona; Electromagnetic radiation; Feature extraction; Partial discharge measurement; Partial discharges; Pattern recognition; Sparks; Time frequency analysis; Voltage; BP network; feature extraction; partial discharge; pattern recognition;

fLanguage

English

Publisher

ieee

Conference_Titel

Environmental Electromagnetics, 2009. CEEM 2009. 5th Asia-Pacific Conference on

Conference_Location

Xian

Print_ISBN

978-1-4244-4344-4

Type

conf

DOI

10.1109/CEEM.2009.5304189

Filename

5304189