Adaptive Weighted Least Squares-Based Algorithm to Estimate Synchronized Measurements over Wide Frequency Range

International audienceThis paper proposes a frequency-adaptive weighted least squares-based algorithm for accurate identification and estimation of synchronized measurements over wide frequency range. A new mathematical formulation is performed to estimate all synchronized components of a single-phase voltage or current input signal when spurious frequency as harmonics and off-nominal suchfrequency, within the range of 5 Hz around the fundamental frequency, are presented. The proposed algorithm decouples local system frequency estimation and synchronized components tracking, in which the static off-nominal frequency is estimated from the classical method based on the rate of change of phase angle relative to the offset from fundamental frequency whose outcome is used to calculate a new sampling angle that is utilized in the computation of the synchronized components. Several noisy signals are applied to evaluate the algorithm’s performance results obtained where the demonstrate the capability of the proposed algorithm in dealing efficiently with very corrupted signals. The algorithm is envisaged for M-Class Phasor Measurement Unit (PMU) aiming to improve the power quality issue of power distribution systems