TL;DR
As grids increasingly rely on renewable energy sources, the traditional solution of increasing inertia to maintain stability is insufficient. This paper reveals that there's an optimal and maximum number of Virtual Synchronous Generators (VSGs) that can be safely integrated, and proposes "stability storage" as a new commodity to ensure grid stability in the short term.
What Problem Does This Paper Address?
The transition to renewable energy sources like solar and wind power is leading to a decrease in the inertia of power grids. Inertia, traditionally provided by large rotating synchronous generators, is crucial for maintaining grid stability by resisting frequency deviations during disturbances. To compensate for this loss, Virtual Synchronous Generators (VSGs) are being developed to emulate the inertial behavior of traditional generators. However, the paper questions whether simply increasing inertia is enough to guarantee grid stability when VSGs and synchronous generators coexist. Specifically, it investigates the complex interplay between VSGs and existing generators, the impact of their control parameters, and the potential for small-signal instabilities to arise even with high system inertia. This is particularly relevant as VSGs are often operated by prosumers with their own financial interests, and their decentralized nature makes traditional centralized control challenging.
Methods
The study employs frequency domain stability analysis using MATLAB to assess the small-signal stability of the Greater London power grid. They model the grid topology based on real-world data and simulate various future energy scenarios with different levels of System Non-Synchronous Penetration (SNSP) and numbers of VSGs. The stability is quantified by the damping ratio of the dominant oscillatory mode. A key novel contribution is the concept of "stability storage," which is implemented through virtual impedance control in VSGs. This allows VSGs to actively contribute to grid stability by emulating an additional output impedance, analogous to how energy storage devices provide power during imbalances. A metric called the Distributed Stability Metric (DSM) is introduced to quantify the individual contribution of each VSG to stability, enabling a decentralized market for this service.
Datasets
The power grid topology for the Greater London area was obtained from:
- Electric substation data from National Grid, UK.
- Road network and building data from OpenStreetMap.
Future energy scenarios and System Non-Synchronous Penetration (SNSP) projections were sourced from National Grid ESO.
Key Insights and Ideas
- Inertia alone is not enough: The paper's most striking finding is that simply increasing system inertia is insufficient to guarantee grid stability when VSGs are present. Small-signal instabilities can still emerge.
- Optimal and Maximum VSG Limits: There exists a critical optimal and maximum number of VSGs that can be safely integrated into the grid for a given inertia level and SNSP. Exceeding this limit can lead to instability.
- Symmetry of Instability: The heat maps illustrating grid stability show a symmetry along the anti-diagonal, suggesting a fundamental similarity in how VSGs and synchronous generators can contribute to instability when their control parameters (like droop gains) are mismatched.
- "Stability Storage" as a New Commodity: The paper introduces "stability storage" as a novel concept, analogous to energy storage. This service is provided by VSGs through virtual impedance control, actively contributing to grid stability.
- Decentralized Stability Provision: Stability storage can be provided in a decentralized manner by individual VSGs, making it scalable and suitable for prosumer participation.
- External Interventions are Insufficient: Adding a single, high-capacity VSG (even with "infinite inertia") does not necessarily improve stability in unstable scenarios, highlighting the need for a distributed approach.
Key Results
- The study demonstrates that grid instability can arise even with high inertia levels and low shares of inverter generation if VSGs are not properly sized.
- For a given inertia level, an optimal and a maximum number of VSGs that can be safely supported were identified. For instance, with H=4s and SNSP=50%, the Greater London grid can support a maximum of 35 VSGs.
- The proposed "stability storage" mechanism, implemented via virtual impedance control, was shown to effectively stabilize previously unstable grid scenarios. For example, in Scenario-1, stability storage brought the damping ratio from -0.2602 to 0.1232, making the grid stable.
- The effectiveness of stability storage increases with the number of participating inverters and the magnitude of impedance they contribute.
Limitations and Future Work
- Model Simplifications: The study uses simplified models for synchronous generators and assumes equal capacity for inverters. The real-world behavior of these components can be more complex.
- Grid Topology: While based on real data, the derived grid topology is a simplification. The actual network structure and its dynamic behavior might differ.
- Prosumer Participation: The study assumes that all VSGs can participate in stability storage. In reality, the willingness of prosumers to provide this service might vary, and the paper explores this by analyzing the impact of varying participation fractions.
- Future Work: The authors suggest that future work should focus on developing more sophisticated control strategies for VSGs, exploring the implications of diverse inverter control strategies, and investigating the economic incentives for stability storage provision in more detail. They also highlight the need for updated standards for grid-forming inverters that incorporate stabilizing functions beyond inertia and droop.
Summary updated