An innovative timestamp-based convolution integral method in synchrophasor estimation within digital relays

Abstract

In recent years, phasor measurement units (PMU) and wide-area monitoring systems (WAMS) techniques have been widely used in power systems to perform various monitoring, control and protection functions. It is a growing requirement for digital protective relays to include synchrophasor measurement functionalities for the apparent economical benefits. Modern digital protective relays typically use variable sampling rate (frequency tracking) so as to reduce measurement errors and avoid undesired trips during system frequency variations. However, frequency estimation may experience some inaccuracies during various power system dynamics, and in turn, it will result in sampling time interval errors because of the frequency tracking mechanism in the relay, and then result in synchrophasor estimation errors. An innovative timestamp-based convolution integral method has been developed to accurately estimate synchrophasors within digital relays, in which tracking frequency errors virtually do not have impacts to the phasor estimations. This method can be used for either variable sampling rate (frequency tracking) or fixed sampling rate. It can accurately estimate synchrophasors even with artificial extreme inaccuracies in tracking frequency estimations. In this paper, the theoretical derivations of the said method, the impact of the tracking frequency errors to phasor estimations, digital filtering and various tests to meet IEEE C37.118 steady state and dynamic performance requirements are discussed.