A Robust approach for the identification of synchronous machine parameters and dynamic states based on PMU data

In this paper an approach for on-line identification of parameters and dynamic states of synchronous generator, is proposed. Synchrophasors data and Hybrid Dynamic Simulation provide the trajectories deviations caused by parameters errors. Simulated variables are combined with measurements to obtain Hybrid Trajectory Sensibility Functions (HTSF), which are used in the identification process based on Nonlinear Least Squares. This approach simplifies the calculation of HTSF. Constraints on the parameters range are enforced and the resulting constrained optimization problems is solved by a Primal-Dual Interior Points method. This solution is discussed in the paper. The method is applied to synthetic data and to a large generator of the Itaipu power plant.