Power system state estimation: Undetectable bad data

This paper proposes an index, called Undetectability Index (UI), to classify the measurements according to their characteristics of not reflecting their errors into the residuals of the weighted least squares state estimator (WLS) from a geometric analysis of this estimator. Gross errors in measurements with high UIs are very difficult to be detected by methods based on residual analysis, because the errors in those measurements are masked, i.e. they are not reflected in their residuals. The paper also presents a very simple algorithm to compute the UI of all available measurements in a metering system. This algorithm is very simple and uses routines already available in the existing WLS Estimator software. Two UI applications are investigated (i) to design more reliable metering systems for bad data processing, i.e. metering systems formed by measurements with UI lower than a pre-specified value and (ii) to estimate the measurement errors. Several simulation results (with IEEE-14 and 30 bus systems) have validated the propositions. Copyright (c) 2013 John Wiley & Sons, Ltd.