Dynamic phasor estimation using adaptive artificial neural network

The variation in amplitude and phase of voltage and current signals of a power system under dynamics is often the basis for inaccurate phasor estimation. Further, the frequency estimator derived from estimated phasor is being affected under such dynamic characteristics. This paper presents a method to compute the magnitude and phase angle of the signal during the dynamics using an artificial neural network. First the modeling of signal during dynamics is presented then a multi-layered feed-forward neural network algorithm is developed to estimate the unknown parameters of the model. Subsequently, the magnitude and phase angle and frequency of the signal are calculated using the estimated parameters. The performance of the proposed method is evaluated using standard test signals as per the synchrophasor standard.