DocumentCode :
2676052
Title :
Evaluation of algorithms for high impedance faults identification based on staged fault tests
Author :
Yang, Ming-Ta ; Gu, Jhy-Cherng ; Guan, Jin-Lung ; Cheng, Chau-Yuan
Author_Institution :
Dept. of Electr. Eng., St. John´´s Univ., Taipei
fYear :
0
fDate :
0-0 0
Abstract :
The main objective of this study is to develop an intelligent relay which is capable of protecting aerial lines from high impedance faults (HIFs). This investigation successfully develops a novel intelligent HIF detector that applies neutral line current to solve HIF problems. A self-turning scheme based on the chi-square distribution and 95 % confidence interval is first applied to set the threshold level automatically for the neutral line current variances examined. The feature extraction system based on wavelet transform and the pattern recognition technique found on neural networks are then applied to discriminate effectively between the HIFs and the switch operations. Two staged fault tests were undertaken to examine the feasibility of the proposed algorithm and measure its performance. The performance other un-intelligent relaying algorithms found in the literature was also compared with proposed intelligent HIF detector based on the staged fault records. Experimental results demonstrate that the proposed intelligent relay is feasible performance well
Keywords :
fault diagnosis; feature extraction; neural nets; power engineering computing; power overhead lines; power transmission faults; power transmission protection; relay protection; wavelet transforms; aerial lines protection; chi-square distribution; feature extraction system; high impedance faults identification; intelligent HIF detector; intelligent relay; neural networks; neutral line current; pattern recognition technique; self-turning scheme; staged fault tests; wavelet transform; Detectors; Fault diagnosis; Feature extraction; Impedance; Protection; Protective relaying; Relays; Switches; Testing; Wavelet transforms; Fault diagnosis; high impedance fault; neural networks; protection; statistical confidence; wavelet transform;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Power Engineering Society General Meeting, 2006. IEEE
Conference_Location :
Montreal, Que.
Print_ISBN :
1-4244-0493-2
Type :
conf
DOI :
10.1109/PES.2006.1709122
Filename :
1709122
Link To Document :
بازگشت