A Statistical Investigation of PMU Errors in Current Measurements

Phasor Measurement Units (PMUs) are becoming widespread in power systems monitoring. Applications relying on synchrophasor measurements ask for models of measurement er-rors that can be integrated to describe the uncertainty of the data. In this paper, an assessment of PMU magnitude and phase-angle errors for current measurements is performed for the first time with the aim to address error distributions. In particular, two commercial PMUs are statistically characterized in a controlled environment to prevent non-stationarity issues. It is proven that Gaussianity appears a valid assumption for magnitude errors of both PMUs, whereas phase-angle errors might be non-Gaussian in an accurate PMU when the contribution of synchronization mechanisms prevails.