Dynamic synchrophasor estimation using Smoothed Kalman filtering

Next-generation signal processing algorithms for Phasor Measurement Units (PMUs) must be able to estimate magnitude, phase, frequency and rate of change of frequency (ROCOF) of voltage or current waveforms with superior accuracy and speed, in order to meet the requirements of distribution systems afflicted by dynamic disturbances. In this paper, the potentialities and the limits of an algorithm based on a Smoothed Taylor-Kalman Filter (ST-KF) are presented and explained. The proposed algorithm relies on a variant of the so-called Taylor-Kalman Filter (T-KF), but with a reduced computational complexity thanks to the expression of complex quantities in rectangular coordinates and an iterated implementation. In addition, an output smoother provides a better frequency estimation in offnominal conditions. Some simulation results show the benefits of the proposed approach in dynamic conditions, as well as the performance limitation of Taylor-Kalman Filter solutions when large harmonics and inter-harmonics are considered.