UKF based estimation of synchronous generator electromechanical parameters from phasor measurements

Abstract

The paper proposes an Unscented Kalman Filter (UKF)-based algorithm to estimate the electromechanical parameters and states of synchronous machines (rotor angle, q-axis reactance, inertia, damping and mechanical power). The algorithm uses observations or measurements available at the output terminal of the machine polluted by colored noise. Testing of the algorithm was conducted against a model of the same complexity and a model of a higher complexity. The contribution of this paper is twofold: 1) the algorithm is able to estimate electromechanical parameters such as inertia H and damping D which have not been investigated in other machine estimation papers. Modeling error is modeled as colored noise to enhance the estimation capability; and 2) a dual UKF filter is set up to carry out the estimation where the estimator of q-axis reactance is separated from the other states and parameters. Such set up solves the difficulty of UKF based estimation. Two case studies demonstrate the feasibility of the estimation.