Comparison of effectiveness of various mother wavelet functions in the detection of actual 3-phase voltage sags

Conventional methods for analyzing power quality disturbances are primarily based on visual inspection of the rms value and Fourier Transform (FT) of the voltage and current waveforms that were recorded by power quality recorders. By analyzing these waveforms power utilities' engineers can evaluate the condition of the network and identify any potential degrading trends in the electrical system. However, to perform manual analyses on all the voltage events recorded in the networks is time consuming. There are also questions posed on the accuracy of the rms and FT in the detection of non-stationary waveforms. To overcome these two deficiencies, an automated technique comprising of signal processing and artificial intelligence techniques is proposed. Signal processing techniques such as Short Time Fourier transform (STFT), S-transform and wavelet transform (WT) are widely used for analyzing voltage events. In the WT approach, the original signal is multiplied with a function known as the mother wavelet. There are many mother wavelet functions to be selected for generating the daughter wavelets and it is important to determine the best mother wavelet function for accurate detection of power quality disturbances. In this paper, evaluations were performed to evaluate the effectiveness of five mother wavelet functions in the detection of voltage sags. The results of the evaluations are presented in this paper.