Application of Radial Basis Function Neural Network in Power Quality Events Classification

Ruchi Mahawar; Prof. Pramod Sharma; Shoyab Ali

Abstract

Power Quality concerns have become more important as the Smart Grid has evolved. Computers, microprocessor-controlled electronic loads, and power electronic devices are all used in urban development. Disturbances in power quality are caused by these electronic devices. Variations in the amplitude and frequency of system voltages and currents from their nominal levels identify PQ difficulties. We offer a hybrid technique to accomplishing this goal in this paper. To classify PQ events, different neural topologies have been used, such as Elman Backprop Neural Network (EBPNN), Feed Forward Backprop Neural Network (FFBPNN), Layer Recurrent Neural Network (LRNN), Cascade Forward Backprop Neural Network (CFBNN), Feed Forward Distributed Time Delay Neural Network (FFDTDNN), Nonlinear Autoregressive Exogenous Neural Network (NARX), Radial Basis Function Neural Network (RBFNN), The Radial Basis Function Neural Network (RBFNN) is shown to be the most efficient topology for performing the classification job in a meaningful comparison of different neural topologies. To illustrate the comparison of classifiers, different amounts of Additive White Gaussian Noise (AWGN) are applied to the input features, and it is observed that RBFNN performs well in noisy signals as well.