APPLICATION OF WAVELET NETWORKS IN POWER SYSTEM FAULT SIGNAL PROCESSING

According to different aspects of function approximation with wavelet transformations in wavelet analysis, three main wavelet networks are respectively outlined. The detailed presentation and comparison of the architectures, models and learning algorithms of the three wavelet networks are carried out and the essential differences among them are described. On this basis, the application of wavelet network in the classification of power system fault signals and the fault data compression is researched and corresponding digital simulations are given. The simulation results show that it is feasible to apply wavelet network to power system fault signal processing.