A Decentralized State Estimator for Non-Linear Electric Power Systems

Abstract

State estimators are an integrating part of traditional power systems. Usually state estimators for electric power systems have been conceived as elements of a central control system receiving as input measured data gathered from the network and stored in a central data base. Furthermore, the traditional structure does not usually base the estimation on the integration of a dynamic model. This view could be considered as a limiting factor for modern power systems on board of ships or aircrafts. In order to solve both issues raised above the authors proposed in a recent publication to define the state estimation for a power system starting using a decentralized Kalman filter. Furthermore a novelty of the approach proposed concerns with the formalization of the estimation problem: in our approach the estimation is not exclusively based on the measurement data but also on a dynamic model of the power grid. This paper shows the performance of a decentralized Kalman filter applied to an electric power system for avionic and naval applications.