Classification of Power Quality Disturbance Signals Using FFT, STFT, Wavelet Transforms and Neural Networks - A Comparative Analysis

classification using frequency domain, time domain transforms and Artificial Neural Networks(ANN).Different power system events(disturbances) like sag, swell, transients, chirp, notch, spikes are generated through matlab. The spectral coefficients of the signals are obtained by means of DFT,FFT,STFT and DWT. Along with the spectral analysis the statistical parameters such as mean, autocorrelation, norm, standard deviation, energy and variance are also found out. Finally a comparison is made between the various transforms. Further these parameters are given as inputs to the ANN for classifying different power system events accordingly. The training and testing data required to develop the ANN model is generated through simulation. Keywords: Discrete Fourier(DFT),Fast Fourier Transform(FFT),Short Time Fourier Transform(STFT),Wavelet Transform(WT),Artificial Neural Networks(ANN),Multi Resolution Analysis(MRA)