Based On Wavelet Transform And Fractal Theory Of Transmisson Line Fault Detection

With the continuous development of China's economy, and the increase of power systems scale, EHV transmission lines have grown rapidly in recent years. The fault detection of transmission line to ensure power system stability has become an important research project.Based on Fourier transform, Wavelet transform is developed in recent years, as a non-stationary signal processing method, wavelet transform has a good performance and time-frequency localization, and the modulus maxima characteristics of wavelet transform is very suitable for transient signal analysis and processing. In this paper, wavelet transform is used as phase selection algorithm for traveling-wave fault and the Fourier transform method is analyzed and compared. The Kailunbuer transform fault phase selection of the basic principles and methods are introduced in this paper. By constructing a suitable wavelet function, the modulus maxima detection of traveling wave current signal is proposed in this paper.Fractal theory has widespread application in recent years, to describe the nature of scientific laws of irregular things, its study object is non-linear systems generated by non-smooth and non-differentiable complex geometry. With the two basic features: self-similarity and scale invariance, it reveals intrinsic line between complex phenomena, local and the overall nature. Firstly the definition of fractal theory and characteristics is introduced in this paper, then the fractal dimension, fractal parameters to comment on the selection and to strike a narrative are put forward for the fractal theory in transmission line fault detection, and the critical parameters of the calculated.Finally, back-propagation algorithm (BP) neural network is used for the transmission line fault type identification, and the corresponding neural network model is proposed. An example is analyzed and calculated to verify the correctness.