Hybrid Data/Model-Driven Whole-System Admittance Identification via Single-Bus Injections

Identifying the whole-system admittance model of an interconnected power system with multiple inverter-based resources (IBRs) typically requires injecting small voltage perturbations at each bus connected with an IBR, which is impractical in both simulations and real-world systems. To overcome this challenge, this paper presents a hybrid data/model-driven approach to identifying the whole-system admittance model through combinations of the measured data and the theoretical network model. By introducing voltage perturbations at a single bus in the system and measuring current responses at all buses, the measured data can be integrated with the known network nodal admittance matrix to reconstruct the whole-system admittance model. The whole-system admittance model, which is a closed-loop model, encompasses the dynamic characteristics of the entire system and facilitates oscillation analysis, root cause tracing, as well as system strength assessment. The proposed new approach is validated through simulations and power hardware-in-the-loop experiments, demonstrating a practical solution for implementing a limited number of injection sources to identify potential oscillations and their root causes for the entire power system.