Measurement based analysis of electromechanical modes with Second Order Blind Identification

Abstract

The Measurement based modal analysis is an effective tool for obtaining essential information on the stability of power systems. This paper proposes a novel multivariate method for the measurement based modal analysis. The proposed method is based on a Blind Source Separation (BSS) technique called Second Order Blind Identification (SOBI), and it utilizes multiple signals measured from several locations in the power system. In addition to the SOBI algorithm, the proposed method uses the Random Decrement (RD) technique and the Eigensystem Realization Algorithm (ERA) to estimate the frequencies and damping ratios of the existing electromechanical modes. The performance of the SOBI–RD–ERA method is analyzed using both simulated data and measurement data collected from the Nordic power system. The results indicate that the frequencies and damping ratios of electromechanical modes can be estimated with high accuracy using the proposed method. Thus, the SOBI–RD–ERA method is a promising analysis tool for wide-area monitoring of electromechanical oscillations.