Title :
Effect of pulse propagation characteristics in power transformer on partial discharge recognition
Author :
Gao, Wensheng ; Tan, Kexiong
Author_Institution :
Dept. of Electr. Eng., Tsinghua Univ., Beijing, China
Abstract :
An experimental method is utilized to investigate the propagation characteristic of discharge pulse and its influence on the recognition of partial discharge in this paper. Firstly, the discharge model tests of common defect in oil-paper insulation are finished. A series of tests are conducted to study the propagation characteristic of discharge pulse along the winding. Thus the identification of discharge waveform could be implemented. The restoring method of the original wave and the extracting method of wave characteristic are also put forward. Finally the backpropagation (BP) neural network is utilized to recognize the wave characteristic of discharge signal. The validity of recognition system is illustrated
Keywords :
backpropagation; impregnated insulation; neural nets; paper; partial discharge measurement; pattern recognition; power engineering computing; power transformer insulation; power transformer testing; transformer oil; back propagation neural network; common defect; discharge model tests; discharge pulse; discharge waveform identification; oil-paper insulation; partial discharge recognition; pattern recognition; power transformer; pulse propagation characteristic; restoring method; wave characteristic; winding; Backpropagation; Character recognition; Insulation testing; Neural networks; Oil insulation; Partial discharges; Power transformer insulation; Power transformers; Pulse transformers; Signal restoration;
Conference_Titel :
Electrical Insulating Materials, 2001. (ISEIM 2001). Proceedings of 2001 International Symposium on
Conference_Location :
Himeji
Print_ISBN :
4-88686-053-2
DOI :
10.1109/ISEIM.2001.973803