Unscented Kalman Filter Based Dynamic State Estimation in Power Systems Using Complex Synchronized PMU Measurements

Abstract

This paper presents a novel scheme of dynamic state estimation (DSE) for power system states based on Unscented Kalman Filter (UKF). For the purpose of DSE using UKF, complex line current and complex bus voltage are utilized as a set of measurement vector that is obtained from PMUs, considering PMUs placed at all the buses (PPAB) and also PMUs placed at optimal location (PPOL). Unscented transformation (UT) technique has been utilized for the implementation of UKF that gives direct procedure of transforming mean and co-variance information. UKF has been compared with traditional weighted least square estimation (WLSE) to show the supremacy of UKF-based DSE. A major issue in power system state estimation is to tackle with highly nonlinear mathematical equation of power system network. This nonlinear equation is linearized utilizing Taylor series expansion in which derivative operator is involved. In view of this, a new approach of DSE has been formulated which is exempted from the derivative operator which is a major challenge in power system state estimation. The suggested DSE approach based on UKF utilizes linear relationship between complex PMU measurements and complex state variables and is derivative-free approach. The proposed method has been implemented on IEEE 6,14,30,57, and 118 bus systems. The obtained results show that the suggested approach is better compared to WLS-based SE scheme.