Covariance management based RLS algorithm for phasor estimation in severely noisy systems

This paper presents a new algorithm, derived from Recursive Least Square (RLS), to estimate power system voltage and current harmonics for digital relay purposes. The algorithm is based on the RLS covariance matrix management, by making use of the covariance resetting in a different way. The approach provides a kind of attractor that absorbs all nonmodeled dynamics. In order to demonstrate its effectiveness, the proposed algorithm is applied to simulated signals and its performance is compared with other conventional methods.