Network Topology Error Identification in Robust State Estimation for Power Systems

Abstract

The least squares methodology has been traditionally utilized in power systems for static state estimation. In power system operation and control it is one of the main routines. A set of reliable state variables is provided. One of the main disadvantages of the conventional state estimator based on the least squares methodology is zero robustness. This means that least squares methods are unable to process even one gross error. While some robust estinlation methods may even accept 50% of the number of measurements with gross errors and still render very acceptable results. A robust state estimation is developed based on the least median of squares. Several variatioas of the method are proposed and applied to power system state cstimation. In this paper the robust estimator is applied when the network topology is erroneously determined. The least median of squares method is theoretically able to manage errors in they direction, known as outliers, as well as in the x direction, known as leverage points. In this case one considers that a topological error is an error in the x direction. In state estimation a small system is more affected by gross errors than a large system. The tests are performed on the IEEE-14 bus system. The results obtained show that this method is still able to provide reliable results. Also a methodology to identify such topological errors is presented