Feature extraction method and neural network pattern recognition on time-resolved partial discharge signals

Author

Tho, Nguyen Thi Ngoc ; Chakrabarty, Chandan Kumar ; Siah, Yap Keem ; Ghani, Ahmad Basri Abd

Author_Institution

Dept. of Electron. & Commun. Eng., Univ. Tenaga Nasional, Darul Ehsan, Malaysia

fYear

2011

fDate

25-28 Sept. 2011

Firstpage

237

Lastpage

240

Abstract

Magnetic sensor is a relatively new method to collect time-resolved partial discharge (PD) signals in XLPE cables. This paper proposes a simple yet effective method to recognize patterns of PD signals obtained from the magnetic sensor. The technique consists of wavelet transformation to de-noise the signals, statistical analysis to extract features and multi-layer perceptron back propagation (MLP BP) neural network to classify different types of PD signals. The result is elaborated in this paper.

Keywords

XLPE insulation; backpropagation; cable insulation; feature extraction; magnetic sensors; multilayer perceptrons; partial discharges; pattern recognition; signal denoising; statistical analysis; wavelet transforms; PD signal; XLPE cable; feature extraction method; magnetic sensor; multilayer perceptron backpropagation neural network; neural network pattern recognition; signal denoising; statistical analysis; time resolved partial discharge signal; wavelet transformation; Cable insulation; Feature extraction; Partial discharges; Pattern recognition; Signal to noise ratio; Wavelet transforms; neural network; partial discharge; pattern recognition; statistical method; time-resolved signals; wavelet de-noising;

fLanguage

English

Publisher

ieee

Conference_Titel

Open Systems (ICOS), 2011 IEEE Conference on

Conference_Location

Langkawi

Print_ISBN

978-1-61284-931-7

Type

conf

DOI

10.1109/ICOS.2011.6079231

Filename

6079231