Inertia Estimation Method for VSG-Dominated Power System Based on Improved Extended Kalman Filter

This paper proposes a method to estimate the in-ertia of power system dominated by virtual synchronous gen-erators (VSGs) using an improved extended Kalman filter (EKF). Initially, the swing equation is reformulated into a dis-crete state-space equation incorporating parameters and noise effects. Then, during power system disturbances, signal pre-processing is applied to active power and frequency data measured by phasor measurement units (PMUs) at the VSG terminal bus. Finally, the grey wolf optimization (GWO) algorithm and U-D Filtering algorithm are integrated into the
$\textbf{EKF}$
algorithm. Compared to traditional methods of inertia constant estimation, this approach not only accurately esti-mates the virtual inertia constant of VSGs but also enables the estimation of virtual damping coefficients. Moreover, relative to direct parameter estimation using EKF, the algorithm en-hances identification accuracy and robustness. The effective-ness of the method is validated through simulations on a VSG-dominated IEEE 9-bus power system.