Online Dynamic Assessment of System Stability in Power Systems Using the Unscented Kalman Filter

Abstract

Dynamic state estimation on of power systems stability based on PhasorMeasurement Units (PMUs) data is a requirement in order to assess power system security. This paper addresses the state estimation problem from a strong mathematical point of view to be applied for estimation on real time. Specifically, this paper formulates a framework based on the Unscented Kalman Filter (UFK) in order to estimate in real-time if the angle stability is compromised or approaching instability. This paper proposes a predicting window as a time interval to forecast the rotor angle using real-time information. The framework uses the generator rotor angles and the electrical angular velocity as state variables, given that the power output of the generators is measured by PMUs and that the rotor movement equations are separated from the network ones. The estimation of the angle stability can provide meaningful information needed to evaluate and enhance the power system security. In order to note the performance for the UKF to assess and forecast online system stability for power systems, the UKF is tested when there are not measurements available. In this case, the UKF performs a forecasting on the angle in a onesecond window without available data. Emulated PMUs sampling data have been used for carrying the simulations and have been validated using the 9-IEEE buses test system. The results confirm the method and the performance of the estimation. The proposed approach provides an effective tool for real-time environment for security.