Fuzzy systems and wavelet transform techniques for evaluating the quality of the electric power system waveforms

Power quality evaluation becomes essential especially while the electric power system is undergoing deregulation. Accurate electric power quality evaluation requires accurate power quality indices (PQIs) formulation in order to quantify qualitatively and quantitatively the electric power system under different operating conditions. Power quality indices and power components according to the IEEE Standard 1459-2000 are based on the Fourier foundation which gives accurate results only in case of stationary disturbances. However, due to the widespread use of nonlinear loads, power system faults and transients, the voltage and current waveforms may contains stationary and nonstationary distortions and therefore Fourier transform become unsuitable in these situations.;In order to test the validity, accuracy and performance of the new approach it has been applied to different case studies considering single and three-phase system in the presence of stationary and nonstationary power quality disturbances. The results show that the new approach outperforms other traditional approaches plus the new index (FWPTPQI) is very sensitive to any changes in any other input power quality index and successfully quantifies any power quality disturbance. The thesis also includes a comparative study of applying different wavelet basis functions to the developed fuzzy-wavelet-based modules and the results show that Daubechies with order 10 could be the wise choice that could satisfy good accuracy while requiring less memory size and computational time.;In this thesis, power components and power quality indices are reformulated using the wavelet packet transform (WPT) since it proves to be more suitable for analyzing nonstationary waveforms while preserving both time and frequency information plus providing uniform frequency bands. On the other hand, electric power system operation is associated with many sources of uncertainties, therefore fuzzy systems have been used to develop fuzzy-based modules in order to amalgamate the reformulated wavelet packet transform-based power quality indices and collapsing to a single universal index called fuzzy-wavelet packet transform based power quality index (FWPTPQI). The proposed approach is developed for single-phase systems then using the time-frequency symmetrical component concept and the equivalent voltage and current root mean square, the approach is extended to balanced and unbalanced three-phase systems.