Robust State Estimation Against Cyber Attacks for Power Systems Measured by Synchro-Phasors

Abstract

TodayÍs power systems use the weighted least squares (WLS) estimator, which is a widely used and well-investigated method to solve the state estimation problem. Power systems are primarily measured by conventional measurements, namely bus power injection, branch power flow and bus voltage magnitude measurements, which are related to the system states non-linearly. Therefore, it is required to solve state estimation problem iteratively. Although Least Absolute Value (LAV) estimator is more robust compared to WLS estimators against random errors or intentional / coordinated manipulation of measurements constituting cyber attacks, it is not widely implemented due to its heavy computational cost and its vulnerability to the so-called leverage measurements. Phasor measurement units (PMUs), which are recently being deployed in large numbers in power systems, measure bus voltage phasors and line current phasors that are both linearly related to the system states. This linearity enables the solution to be obtained in a non-iterative fashion for both WLS and LAV estimators. Once the rapid increase in the number of synchro-phasors in power systems is considered, it is not unrealistic to expect to measure the power systems with only synchro-phasors in the future. In this work, we propose to use the LAV estimator for power systems measured by synchro-phasors, since (1) leveraging effect of the measurements can be eliminated easily, (2) LAV estimation is robust against cyber-attacks, and (3) it becomes computationally competitive with WLS estimation.