A Robust and Reduced-Order Power System Dynamic State Estimator for Grid-Forming Inverters

To date, dynamic state estimation (DSE) for grid-forming inverters (GFMs) remains an unexplored problem due to their timescale-coupled dynamics and unknown inner loop control strategy. To address these challenges, this article proposes a novel decentralized DSE framework that relies only on information on the outer loop control. In this framework, a reduced-order state-space model for DSE is derived, assuming immediate tracking of inner loop control and quasi-stationary LCL filter. By modeling the tracking offset of the inner loop control as a measurement outlier, an innovation saturation mechanism is developed to robustify DSE against it. This DSE framework is tested on both the 14-bus and 240-bus systems with integrated GFMs. Numerical results demonstrate that the proposed method is feasible and robust against unknown parameter settings of the inner loop control.