Calculation of Measurement Correlations Using Point Estimate

Abstract

Currents, voltages, and voltage–current phase angles are directly measured in substations and converted through current complex measurement systems into power injection and power-flow measurements. Since voltages, currents, and phase angles are directly measured, they are affected by errors that are statistically independent and generally Gaussian distributed. However, power injections and flows, which are fabricated out of currents, voltages, and phase angles, are affected by errors that are not generally independent. This paper describes a procedure to estimate the correlation matrix that characterizes the dependencies among all measurements within a substation. The proposed technique that relies on point estimate is accurate and computationally efficient. A realistic case study is used to compare the results obtained by using the proposed technique with those obtained using a cumbersome Monte Carlo algorithm.