Partial discharge pattern recognition of transformer based on electric signal and ultrasonic comprehensive analysis

Author

Zhao, Shutao ; Li, Baoshu ; Wang, Yong

Author_Institution

State Key Lab. of Alternate Electr. Power, Syst. with Renewable Energy Sources North China Electr. Power Univ., Baoding, China

fYear

2012

fDate

19-20 May 2012

Firstpage

637

Lastpage

641

Abstract

According to the characteristics and disadvantages of PD electrical analysis method, an electro-acoustic and ultrasound combination partial discharge (PD) pattern recognition method is proposed. Firstly, the transformer PD test system´s wiring diagram is given, and anti-jamming measures are described. Secondly, Electrical pulse signals time-domain and ultrasound signal statistical features have been extracted to build feature vectors, and then the features is input to BP neural network, the typical PD pattern recognition has been derived. Lastly, the partial discharge analysis system software is developed, and experiments show that the analysis method is effective and software process is correct.

Keywords

acoustic signal processing; backpropagation; circuit diagrams; feature extraction; jamming; neural nets; partial discharge measurement; pattern recognition; power engineering computing; power system measurement; power transformer testing; time-domain analysis; ultrasonic applications; BP neural network; PD electrical analysis method; antijamming measures; electric signal; electrical pulse signal time-domain feature extraction; feature vectors; partial discharge analysis system software; partial discharge pattern recognition; transformer PD test system; ultrasonic comprehensive analysis; ultrasound signal statistical feature extraction; wiring diagram; Acoustics; Discharges (electric); Feature extraction; Partial discharges; Pattern recognition; Power transformer insulation; Feature Extracted; Pattern recognition; partial discharge; power transformer;

fLanguage

English

Publisher

ieee

Conference_Titel

Systems and Informatics (ICSAI), 2012 International Conference on

Conference_Location

Yantai

Print_ISBN

978-1-4673-0198-5

Type

conf

DOI

10.1109/ICSAI.2012.6223077

Filename

6223077