A novel method based on wavelet network and fractal theory for detecting short-circuit fault is proposed. To increase the signal-noise-ratio of the fault signal, the signal denoising technology using the statistic rule is brought forward to determine the threshold and the decomposition level of each order of wavelet space. On the basis of the inter relationship of wavelet transform and fractal theory, the whole and local fractal exponents obtained as features are presented for extracting fault signals. The effectiveness of the new algorithm used to extract the characteristic signal is described, which can be achieved by different fractal dimensions values of those types of short-circuit fault. In accordance with the threshold value of each type of short-circuit fault in each frequency band, the correlation between the type of short-circuit and the fractal dimensions can be figured to perform extraction. This detection model incorporates the advantages of morphological filter and multi-scale wavelet transform to extract the feature of faults meanwhile restraining various noises. The simulation results demonstrates that the proposed model is verified effectively.
[1]
Y. Sheng,et al.
Decision Trees and Wavelet Analysis for Power Transformer Protection
,
2002,
IEEE Power Engineering Review.
[2]
Q. H. Wu,et al.
Transient Positional Protection of Transmission Lines Using Complex Wavelet Analysis
,
2002,
IEEE Power Engineering Review.
[3]
Q. H. Wu,et al.
Ultra-High-Speed Directional Protection of Transmission Lines Using Mathematical Morphology
,
2002,
IEEE Power Engineering Review.
[4]
Karen L. Butler-Purry,et al.
Characterization of transients in transformers using discrete wavelet transforms
,
2003
.
[5]
L.J. Hadjileontiadis,et al.
Detection of explosive lung and bowel sounds by means of fractal dimension
,
2003,
IEEE Signal Processing Letters.
[6]
Leontios J. Hadjileontiadis,et al.
Wavelet-based enhancement of lung and bowel sounds using fractal dimension thresholding-part II: application results
,
2005,
IEEE Transactions on Biomedical Engineering.