HARMONIC ESTIMATION IN A POWER-SYSTEM USING ADAPTIVE PERCEPTRONS

The paper presents an adaptive neural network approach for the estimat ion of harmonic components of a power system. The neural estimator is based on the use of an adaptive perceptron comprising a linear adaptiv e neuron called Adaline. The learning parameters in the proposed algor ithm are adjusted to force the error between the actual and desired ou tputs to satisfy a stable difference error equation. The estimator tra cks the Fourier coefficients of the signal data corrupted with noise a nd decaying DC components very accurately. Adaptive tracking of harmon ic components of a power system can easily be done using this algorith m. Several numerical tests have been conducted for the adaptive estima tion of harmonic components of power system signals mixed with noise a nd decaying DC components. Data from a laboratory test is used to vali date the performance of this new approach.