AN ADAPTIVE NEURAL-NETWORK APPROACH FOR THE ESTIMATION OF POWER-SYSTEM FREQUENCY

A new approach to the estimation of power system frequency using an ad aptive neural network is presented in this paper. This approach uses a linear adaptive neuron or an adaptive linear combiner called 'Adaline ' to identify the parameters of a discrete signal model of the power s ystem voltage. Here, the learning parameters are adjusted to force the error between the actual and the computed signal samples to satisfy a stable difference error equation, rather than to minimize an error fu nction. The proposed algorithm shows a high degree of robustness and e stimation accuracy over a wide range of frequency changes. The techniq ue is shown to be capable of tracking power system conditions and is i mmune to the effects of harmonics and random noise. (C) 1997 Elsevier Science S.A.