Model-based synchrophasor estimation by exploiting the eigensystem realization approach

Abstract

This paper proposes an eigensystem realization (ER) approach to accurately provide synchrophasor estimates. The proposal improves the frequency computation and its tracking since the ER-based estimator works as a frequency adaptive method, reducing the total vector error (TVE) in the presence of dynamic compliances reported in the IEEE Synchrophasor Std. C37.118. Phasor estimates such as amplitude, phase, frequency and the rate of change of amplitude (ROCOA) can be provided in one-cycle. The new ER-based phasor estimator is evaluated under steady-state, dynamic and noisy conditions using both theoretical and actual signals stemming from a commercial PMU and a Digital Fault Recorder (DFR). Comparisons are established with the well known Discrete Fourier Transform (DFT), Prony method and the Interpolated DFT (IpDFT). Finally, the results exhibit that the proposed approach attains reliable estimates, even though under polluted conditions by high harmonic content, being able to track the changes in amplitude, phase, frequency, with enough precision. Thus, the eigensystem realization-based method becomes a class P phasor measurement algorithm.