Error Analysis of on-line Identification of Parameters of Synchronous Generators

The parameter identification of synchronous generator is closely related to the error of the measured data, but there is lack of systematic research on the error calculation of the identification results. A method to compute errors of on-line parameter identification of synchronous generators based on the Cramér-Ran lower bound is given in this paper. Firstly, the likelihood function expression of probability density function based on synchronous generator parameter identification is given by Bayesian formula. Secondly, the error of the identified parameters are computed via the Cramér-Ran lower bound and the trajectory sensitivity vectors. Thirdly, a method to combine identification results of different tests is given. Finally, simulation results of on-line data demonstrate the validity of the proposed algorithm.