Minimization of uncertainties in analog measurements for use in state estimation

It is pointed out that the uncertainty in an analog measurement is a combination of systematic error and random error; The systematic error completely dominates the random error and varies in a recognizable pattern over the range of measurement. A procedure is presented for reducing the systematic errors and estimating the prevailing standard deviations. A range of analog measurements is obtained from a transmission station, spanning peak and light-load conditions. Then the redundancies in measurements are used to formulate several functions relating these measurements with their attending errors. Minimization of these functions yields the required correction coefficients which are used to minimize the systematic errors and to evaluate the actual random errors, both of which are required by the state estimation. The approach developed uses voltage, real, and reactive power measurements which generally are available and used by state estimators. Current measurement is limited to a critical line or transformer per substation. The effectiveness of the proposed approach is confirmed by an experiment involving a 500/230 kV transmission station.<
>