Affordable Access

A Non-linear Approach for On-line Parameter Estimation of Synchronous Machines

Publication Date


Badge A NON-LINEAR APPROACH FOR ON-LINE PARAMETER ESTIMATION OF SYNCHRONOUS MACHINES Gustavo Valverde Elias Kyriakides Vladimir Terzija The University of Manchester The University of Cyprus The University of Manchester Manchester, United Kingdom Nicosia, Cyprus Manchester, United Kingdom [email protected] [email protected] [email protected] Abstract – This paper presents a non-linear parameter estimator to update the model parameters of synchronous machines. It is presented how recorded measurements from the stator and field winding are used to successfully estimate the machine parameters during transient condi- tions. The high accuracy of this effective non-linear ap- proach, based on the Unscented Kalman Filter, is vali- dated when updating the parameters of a 325 MVA syn- chronous machine operating during transient conditions. Keywords: Kalman filtering, non-linear recursive estimation, parameter estimation, synchronous ma- chines. 1 INTRODUCTION Reliable stability studies require that power system elements are modeled in such a way that they truly represent the transient behavior of the system. The mod- els used for stability studies are well known and ac- cepted; the problem appears when the parameters of those models do not correspond to the actual condition of the element under consideration. In the case of syn- chronous generators, the parameters of the machine can change, with respect to the manufacturer data, due to aging processes of the machine and saturation condi- tions [1]. Even when the model parameters can be updated from off-line tests based on frequency response tests, this option may not be the most appropriate for the financial interests of generation companies. Recent efforts have been concentrated on extracting the ma

There are no comments yet on this publication. Be the first to share your thoughts.


Seen <100 times