Use of Wavelet Transform and Generalized Regression Neural Network (GRNN) to the Characterization of Short-Duration Voltage Variation in Electric Power System.

This work presents the use of the wavelet transform and computational intelligence techniques to quantify voltage short-duration variation in electric power systems, with respect to time duration and magnitude. The wavelet transform is used to determine the event duration, as well as for obtaining a characteristic curve relating the signal norm as function of the number of cycles for a waveform without disturbance that is used as reference for the calculation of the magnitude of the event. A generalized regression neural network (GRNN) is used to interpolate not stored points of the characteristic curve. The method is part of a process to automate the post operation signal analysis in electric power systems, and it is used to quantify the voltage short-duration variation magnitude of previously selected signals. The method has been shown efficient, and some results obtained from the application of this method to power system real signals are presented.