An Improved Method for Estimating Phasors and ROCOF Suitable for M Class PMUs

A method for accurate estimation of phasors for M class phasor measurement units (PMUs) is presented in this work. The M class PMUs require the filtering of the aliased signal. For this, the model order estimation of the signal is carried out for different data windows and sampling frequencies. The obtained model order is used to estimate the fundamental phasor, the first and the second derivatives of amplitude and phase of the signal using the second-order Taylor Series approximated total least square estimation of signal parameters via rotational invariance techniques (TLS-ESPRIT) method. The derivatives of the amplitude and the phase obtained, in this work, use a single data window of multiple cycle lengths. The estimated total vector error (TVE), frequency error (FE), and rate of change of frequency errors (RFE) are evaluated in the presence of noise for various test cases to comply with the synchrophasor standard and also compared with some of the recent work reported in the literature.

Automated summary

This work presents a method for accurately estimating phasors for M class phasor measurement units (PMUs). The method involves filtering the aliased signal by performing model order estimation for different data windows and sampling frequencies. The estimated model order is then used to estimate the fundamental phasor and the derivatives of amplitude and phase using the TLS-ESPRIT method. The performance of the estimation is evaluated in the presence of noise and compared with previous work in the literature.