An eigensystem realization algorithm based data-driven approach for extracting electromechanical oscillation dynamic patterns from synchrophasor measurements in bulk power grids

Abstract

In power systems, electromechanical oscillatory dynamics, dominant modes, mode shapes, participation factors and coherent groups of generators are all important modal parameters. Most of the existing measurement-based methods focus on estimating one or two aspects among dominant modes, mode shapes, participation factors, and coherent groups. Nevertheless, none of existing methods explore all the four aspects from synchrophasor measurements at the same time. In this work, an eigensystem realization algorithm (ERA) based data-driven approach is developed to estimate dominant modes, mode shapes, participation factors and coherent groups from synchrophasor measurements in a holistic framework. In the proposed approach, the reduced power system dynamic model is first estimated by ERA from multichannel synchrophasor measurements. Then, based on the estimated power system dynamic model, electromechanical oscillation modes, corresponding mode shapes and the left eigenvalue vectors are solved. Next, using the solved mode shapes and left eigenvalue vectors, the participation factors of the generators associated with the electromechanical oscillation modes are calculated. Further, the direction cosines among the generators representing coherent strength of generators are calculated by using the achieved mode shapes. Finally, in accordance with the calculated direction cosines, the coherent generators are then identified. The proposed approach is demonstrated with the simulation data of the 16-machine, 68-bus test system and the field Phasor Measurement Units (PMUs) data collected in Liaoning Electric Power Grid. The results demonstrate that the proposed data-driven approach captures dominant modes, mode shapes, participation factors, and coherent groups of generators from synchrophasor measurements with high accuracy and efficiency.