Stability Analysis in IBR-Dominated Grids: Model-Based and Data-Driven Methodologies

The gist The instability mechanisms of static, small-signal and large-signal phenomena are conceptually explained, and the capability of different modeling and analysis tools to investigate them is examined.

Model-Based Methodologies for Stability Analysis

The paper discusses model-based methodologies for stability analysis of IBR-dominated grids, which involve considering the interplay among different aspects of grid stability The instability mechanisms of static, small-signal and large-signal phenomena are conceptually explained, and the capability of different modeling and analysis tools to investigate them is examined.

Model-based analysis considers a nonlinear state-space model of the IBR in a synchronous reference frame. Large-signal static instability is characterized by the absence of a steadystate equilibrium point for the IBR, which is linked with the unsolvability of IBR state-space equation in static condition. Small-signal stability analysis assesses the IBR local stability in the neighborhood of ¯x in the time- or frequency-domain. Large-signal dynamic stability analysis is used to inspect nonlinear dynamics and instability phenomena that are conceptually reconducted to the situation shown in Fig. 1(c), with a stable equilibrium point with limited region of attraction.

Small-signal and Large-signal Stability Analysis

The paper elaborates on the nature of small-signal and large-signal instability phenomena typical of IBR-based grids, illustrated by the analogy of a rolling ball on a curved surface. Small-signal analysis is carried out on the IBR model linearized around x¯, computed through the static analysis in subsection II-B. Large-signal dynamic stability analysis is used to inspect nonlinear dynamics and instability phenomena that are conceptually reconducted to the situation shown in Fig. 1(c), with a stable equilibrium point with limited region of attraction.

The paper examines how different factors affect system stability, noting that some IBR dynamics affect multiple aspects of the stability in a contrasting way. For example, control tunings that improve one aspect may deteriorate another one. Small-signal analysis can be performed using either time-domain (e.g. eigenvalue analysis) or frequency-domain methods (e.g. Nyquist criterion).

Data-Driven Methodologies for Stability Analysis

Data-driven methodologies for stability analysis and control of IBR-dominated power systems are gaining increasing attention. The paper discusses the challenges on data completeness, model generalization under untrained conditions, physical interpretability, and uncertainty awareness.

The paper introduces several data-driven tools to capture the interplay between static (i.e. IBR operating point) and dynamic stability aspects (e.g. small-signal stability). The challenges related to data requirements, generalization under untrained conditions, physical interpretability, and uncertainty awareness are discussed.

Measurement and Gray-Box Modeling for IBRs

To protect the sensitive intellectual property (IP) of the vendors [110], data-driven models identified from simulation or experiment measurements have become the trend for industry standard. A surrogate linear state-space model has recently been proposed, where the state-space matrix is transformed into diagonal form using an encrypted transformation matrix.

Graybox models based on identification techniques have utilized Hammerstein-Wiener parametrization to capture large-signal dynamics of IBRs. Impedance models based on measurements at different operating points can preserve the IP of different vendors while still allowing model identification to accurately capture system behavior for small-signal stability analysis, control and integration.

Black-box Nonlinear Model of IBRs for Large-Signal Analyses

Recent research has increasingly employed data-driven techniques to construct nonlinear dynamic models of IBRs considering unavailable control details. A feedforward neural network is trained to approximate the full nonlinear behavior directly from measurement data.

Developing such a data-driven nonlinear model of IBRs is considerably more challenging than a small-signal linearized model, as large-signal modeling must capture full nonlinear dynamics across wide operating ranges including current limitation. Large-signal transients span multiple time scales simultaneously, making the data-driven models far less generalizable and more expensive to develop.

Data-Driven Stability Analysis and Data-Driven Control

Data-driven stability analysis has emerged as an effective complement to modeling approaches for IBRs, particularly in scenarios where large-scale operating condition screening is required. Instead of learning impedance models, neuralnetworks could directly map OPs to stability labels.

For small-signal analysis, linear systems can be represented directly by the input-output trajectories via Willems’ Fundamental Lemma, enabling stability analysis directly from data. For large-signal stability, physics-informed neural networks (PINNs) offer a promising route by building Lyapunov functions from data. Data-driven control design often favors direct (modelfree) approaches because the best model fitting the data may not be the best model for the intended control design task.

Conclusion and Future Trends

The authors identify several research gaps that deserve further investigation, including analytical IBR models used for large-signal analysis being often oversimplified with respect to their small-signal counterpart. Decentralized criteria to ensure small-signal stability and interoperability among IBRs in multi-vendor power systems are currently under investigation. While data-driven small-signal analysis is reaching a good level of maturity, data-driven models for largesignal analysis leave key aspects unaddressed. The lack of physical knowledge indicates that purely data-driven models may violate fundamental physical laws or constraints in real IBR operation.

These findings suggest a recent shift from blackbox linearized IBR representations toward direct nonlinear system identification, where neural networks approximate the underlying differential equations of IBR dynamics. The paper concludes by highlighting the need for further investigation into multi-IBR systems and the interaction mechanism between different control strategies.

--- Page 18 ---

IEEE TRANSACTIONS ON POWER ELECTRONICS

The paper has discussed the current challenges and recent advancements in model-based and data-driven stability analysis of power electronics-dominated power grids. Three kinds of converter-driven instability – namely static, small-signal, and large-signal dynamic instability – have fist been defined, and the main factors influencing each of them have been discussed (see Table II). The manuscript has demonstrated the strict interdependence among these three types of instability, and stressed the necessity of studying them concurrently through proper stability analysis tools. In the light of this result, model-based tools for static, small-signal and large-signal stability analysis have been examined, and their capability of investigating different instability scenarios has been assessed (see Figs. 4 and 5). Afterwards, data-driven tools for stability analysis and control have been explored, motivated by the unattainability of detailed IBRs and power grid models in the presence of IP-protection, and their application potential and research gaps have been debated (see Table III). To conclude, the authors identify the following research gaps that deserve further investigation.

--- Page 19 ---

IEEE TRANSACTIONS ON POWER ELECTRONICS

The paper has discussed the current challenges and recent advancements in model-based and data-driven stability analysis of power electronics-dominated power grids. Three kinds of converter-driven instability – namely static, small-signal, and large-signal dynamic instability – have fist been defined, and the main factors influencing each of them have been discussed (see Table II). The manuscript has demonstrated the strict interdependence among these three types of instability, and stressed the necessity of studying them concurrently through proper stability analysis tools. In the light of this result, model-based tools for static, small-signal and large-signal stability analysis have been examined, and their capability of investigating different instability scenarios has been assessed (see Figs. 4 and 5). Afterwards, data-driven tools for stability analysis and control have been explored, motivated by the unattainability of detailed IBRs and power grid models in the presence of IP-protection, and their application potential and research gaps have been debated (see Table III). To conclude, the authors identify the following research gaps that deserve further investigation.

--- Page 20 ---

IEEE TRANSACTIONS ON POWER ELECTRONICS

The paper has discussed the current challenges and recent advancements in model-based and data-driven stability analysis of power electronics-dominated power grids. Three kinds of converter-driven instability – namely static, small-signal, and large-signal dynamic instability – have fist been defined, and the main factors influencing each of them have been discussed (see Table II). The manuscript has demonstrated the strict interdependence among these three types of instability, and stressed the necessity of studying them concurrently through proper stability analysis tools. In the light of this result, model-based tools for static, small-signal and large-signal stability analysis have been examined, and their capability of investigating different instability scenarios has been assessed (see Figs. 4 and 5).

1. Improved AI system can perform combined static/dynamic stability analysis by utilizing nonlinear state-space models to capture this interplay among different aspects of grid stability as stated in Section II.A.

2. Improved AI system can assess potential instability regions using numerical methods, as suggested by the finding that "Static instability occurs for E < 0.8 with SCR = 1.5 in Fig. 4(a), independently on the CLM."

3. Improved AI system can dynamically detect small-signal oscillations through eigenvalue analysis or Nyquist criterion, which is used to characterize the damping ratio of the different IBR oscillations modes.

4. Improved AI system can predict stability boundaries for unseen operating conditions by employing a feature-driven neural network approach, as described in Section III.B, where the model is trained on a dataset covering a wide range of OPs and can predict impedance without the need for repeated measurements through perturbation injection.

5. Improved AI system can perform large-signal transient stability assessment by utilizing feedforward neural networks trained to approximate the full nonlinear behavior directly from measurement data, preserving the continuous-time structure of IBR system.

6. Improved AI system can provide real-time stability assessment by employing dynamic mode decomposition on measured signals to realize fast extraction of modal frequency and damping information even under black-box conditions.

Stability issues in converter-dominated grids have a multifold nature, where static, small-signal and large-signal phenomena can influence each other. Capturing the interplay among these different stability aspects requires a rethinking of conventional modeling and analysis tools for power systems. As a further complication, Intellectual Property protection among manufacturers and system operators makes detailed converters and grid models unattainable. This motivates the use of data-driven approaches for stability analysis and control, nevertheless achieving an insightful representation of static, small-signal and large-signal phenomena from data still leaves many open questions. This paper discusses model-based and data-driven methodologies for stability analysis of converter-dominated grids. The instability mechanisms of static, small-signal and large-signal phenomena are conceptually explained, and the capability of different modeling and analysis tools to investigate them is examined. The effectiveness of data-driven methodologies to analyze instability phenomena in their multifold nature is debated, and the challenges on data completeness, model generalization under untrained conditions, physical interpretability, and uncertainty awareness are discussed.