The high penetration of inverter-based resources (IBRs) represents both a challenge and an opportunity for modern power systems. While the reduced inertia and fast dynamics of IBRs complicate traditional stability paradigms, advancements in control strategies—such as virtual inertia, adaptive controls, and grid-forming inverters—offer promising paths to secure and resilient power system operation. However, the intermittent nature of renewable energy sources (RES) and the distinct characteristics of IBRs introduce new challenges in achieving stable, reliable, and efficient power system operation.
Hence, the evolution toward software-defined power systems (SDPS) is driven by the need to create a more resilient, flexible, and efficient grid capable of handling the complexities introduced by renewable energy integration and distributed generation. SDPS decouples control functions from the physical grid, enabling dynamic, real-time management through advanced software, high-speed communication, and centralized decision-making platforms. This thesis develops robust preventive control strategies that optimize system operation and ensure grid security and stability by harnessing the rapid, precise control of IBR units and leveraging advanced parameter tuning within a software-defined power system framework. To
achieve these goals, we investigate a diverse range of challenges spanning multiple domains. One key issue this thesis addresses is the challenge of system frequency and low inertia. We integrate data-driven based frequency dynamics into an algebraic optimization framework and effectively utilise frequency support from various sources. At the transmission level, our data driven approach optimizes system scheduling to minimize operating costs while ensuring frequency security by leveraging support from wind turbines adaptively.
Another issue tackled in this thesis is the stability challenge in inverter-dominated, carbonneutral power systems, where poorly damped oscillations occur. We propose an optimal power flow framework that integrates data-driven small-signal stability constraints and dynamically optimizes inverter control parameters. A participation-based sampling strategy is used to define stability boundaries efficiently.