Bayesian approach in the modal analysis of electromechanical oscillations

This paper presents a Bayesian approach for identifying the modal parameters (i.e., frequency, damping ratio, and modal excitation) of electromechanical modes. The proposed approach identifies the modal parameters and calculates their uncertainty using ambient phasor measurement unit (PMU) measurements from a power system. The method is applied in the frequency domain on a selected frequency band, which significantly simplifies the identification. The performance of the method is studied with simulated data from the IEEE New England test system and the Nordic power system simulation model. In addition, measured PMU data from the Nordic power system are used. The results indicate that the modal parameters of electromechanical modes can be identified reliably and their identification uncertainty can be fundamentally calculated using the proposed method. Thus, the Bayesian approach is a promising identification method for wide-area monitoring of electromechanical oscillations.