Efficient Dynamic Phasor and Frequency Estimation Using Evolving Order Based Arctangent Affine Projection Algorithm

Phasor Measurement Units (PMUs) find wider applications in modern power systems and compliance of the performance metrics is mandated. Under dynamic conditions, PMU algorithms must ensure accurate measurement with short latencies and high reporting rates. The existing estimators mostly require at least two cycles to meet the P Class standards specified in the IEEE/IEC 60255-118-1a:2018. In this article a synchrophasor estimation algorithm is proposed that relies on dynamic signal modelling with arctangent cost function minimized using the evolving order affine projection (E-ATAPA) algorithm to maximize precision within one-cycle interval. Additionally, it considers a new signal model during critical phase change for estimating frequency and rate of change of frequency (ROCOF). Moreover, the evolving order compares the instantaneous power of output error to a threshold determined by the computed steady-state mean-square error (MSE) of E-ATAPA, thereby enabling variable projection orders that reduces overall computational complexity. This enables implementation on GPS synchronized digital signal processors (DSP) with reduced turnaround time.