Data-Driven Estimation of Inertia for Multiarea Interconnected Power Systems Using Dynamic Mode Decomposition

Abstract

The refined estimation of inertia can provide a reliable basis for power system operation and control. In this article, a data-driven approach for the estimation of inertia is proposed, and it can estimate the effective inertia of different areas in the interconnected power systems. Based on eigenstructure analysis, the intrinsic relationships between inertia and the eigenvalue and eigenvector are analyzed using a linearized dynamic equation. Furthermore, detailed mathematical expressions between inertia and the eigenvalue and eigenvector are established. In addition, dynamic mode decomposition is introduced to extract eigenvalues and eigenvectors from the synchronized measurements to ensure that the scheme proposed in this article can estimate the effective inertia by using only the outputs measured by the phase measurement unit. The effectiveness of the proposed approach is demonstrated through numerical simulations on the IEEE 16-machine 5-area test system and the real measurements of an actual power system.