p.3171
p.3175
p.3180
p.3186
p.3190
p.3194
p.3199
p.3203
p.3207
Smart Detection Technology of Serial Arc-Fault on Low Voltage Power Lines
Abstract:
It is important to use electric power safely. Electric arcing faults on the low voltage power lines must be detected and power sources have to be shut down before fire hazards. In this paper, it suppose a method to analyze and detect arc fault characteristics. The current waveforms are different in arc fault and normal operation conditions. The discrete wavelet transform is used to obtain the spectrum characteristics of experiment data and the spectrum energy statistics can reflect the characteristics. Additionally, the values of spectrum energy are input of the back-propagation neural networks (BPNN), which can identify the arc fault efficiently. The test results have demonstrated that the proposed method is effective.
Info:
Periodical:
Pages:
3190-3193
Citation:
Online since:
August 2013
Authors:
Price:
Permissions: