Dynamic state estimation and prediction for real-time control and operation

Real-time control and operation are crucial to deal with increasing complexity of modern power systems. To effectively enable those functions, it is required a Dynamic State Estimation (DSE) function to provide accurate network state variables at the right moment and predict their trends ahead. This paper addresses the important role of DSE over the conventional static State Estimation in such new context of smart grids. DSE approaches normally based on Extended Kalman Filter (EKF) need to collect recursively time-historic data, to update covariance vectors, and to treat heavy computation matrices. Computation burden mitigates the state-of-the-art utilizations of DSE in real large-scale networks although DSE was introduced several decades ago. In this paper, an improvement of DSE by using Unscented Kalman Filter (UKF) to alleviate computation burden will be discussed. The UKF-based approach avoids using linearization procedure thus outperforms the EKF-based approach to cope with non-linear models. Performance of the method is investigated with a simulation on a 18-bus test network. Preliminary results have been gained through a case study that motivate further research on this approach