Paper
25 September 2023 Research on fault location method for distribution network based on traveling wave full waveform matching
Sixu Feng, Feng Deng, Yarui Zu
Author Affiliations +
Abstract
Because of the complex structure of distribution network, the traveling wave at different fault sections shows different energy distribution in the time and frequency domain. In this paper, a fault location method for distribution network based on traveling wave full waveform (TWFW) matching is proposed. First, transmission processes of traveling wave are analysed at different fault sections theoretically. The time and frequency distribution of TWFW is quantified by wavelet energy entropy. The mapping relationship between the fault section and the energy distribution of traveling wave is discussed, whereby the matching relationship is established. Then, TWFW is taken as the input sample to build a convolutional neural network (CNN) with a category classifier. CNN extracts the energy distribution of TWFW in the time and frequency domain to mine the fault section information. The proposed method requires less location device and makes full use of fault information in the full frequency band. Extensive numerical experiments show that this method achieves accurate fault section location on the test set containing a large number of fault samples with the fault impedance greater than 1kΩ.
(2023) Published by SPIE. Downloading of the abstract is permitted for personal use only.
Sixu Feng, Feng Deng, and Yarui Zu "Research on fault location method for distribution network based on traveling wave full waveform matching", Proc. SPIE 12788, Second International Conference on Energy, Power, and Electrical Technology (ICEPET 2023), 127885H (25 September 2023); https://doi.org/10.1117/12.3004535
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Convolution

Signal to noise ratio

Head

Wavelets

Signal attenuation

Matrices

Mining

Back to Top