Performance comparison of Taylor-Fourier and optimal synchrophasor estimators

Abstract

In this paper, we present a detailed comparison of synchrophasor estimators obtained with two state-of-the-art approaches for synchrophasor estimation: the well-known Taylor-Fourier (TF) filters and the recently introduced convex semi-infinite programming (CSIP) filters. It is shown that for short filters and relatively mild constraints, the optimal CSIP estimators are equivalent to traditional DFT estimators. In general, the CSIP filters are shown to be more robust to noise than the TF filters. Moreover, the uniform performance of the CSIP filters in terms of the tests established in the IEEE Std. C37.118.1-2011 is always superior to that of the TF filters. Finally, from the designer point of view, the CSIP filters are much easier to design to comply with a set of prescribed time and frequency domain performance requirements.