Quantifying Grid-Forming Behavior: Bridging Device-level Dynamics and System-Level Strength

arXiv:2503.24152 · eess.SY, cs.SY · Submitted 2025-03-31 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Robotics Radio. Generated commentary on the latest robotics and control papers.

Rosa: Today's paper: "Quantifying Grid-Forming Behavior".

Dev: Grid-forming (GFM) technology is widely regarded as a promising solution for future power systems dominated by power electronics,

Rosa: First, who's behind it and why it matters.

Title and authors: Dev: Now that we’ve talked about the big picture of what the paper is trying to achieve, let's really dig into how they summarize their findings in "Quantifying Grid-Forming Behavior: Bridging Device-level Dynamics and System-Level Strength."

Rosa: I think the summary boils down to introducing two key metrics—the Forming Index at the device level and system strength at the system level—to solve that lack of a universal definition for GFM behavior.

Taro: So, to summarize, they are moving away from just looking at various control architectures and instead providing a metric, FI(jω), to quantify the converter’s response to grid voltage fluctuations.

Dev: Correct; this index is formally defined as F I(j omega) =

S v(j omega): , which directly relates to the converter's voltage source behavior, where an FI less than one is what they call a stronger GFM capability <ref:2503.24152#pg0>.

Rosa: And on the system side, they introduce system strength as a measure of multi-bus voltage stiffness, defined by kappa(j omega) = sigma-one

ZCl(j omega): = sigma

YCl(j omega): <ref:2503.24152#pg2>.

Taro: This system strength metric captures how sensitive the multi-bus voltage vector is to current or power disturbances, and they further break it down into grid strength alpha(j omega) and bus strength kappa i(omega).

Dev: The central idea they are hammering home is the formal proof that a GFM converter enhances system strength, which means linking the two indices together through specific propositions.

Rosa: So, if I'm following along, they’re showing that when you reduce the FI of a connected device, it demonstrably increases both the lower bound of system strength kappa(j omega) and the bus strength kappa n+one(omega) <ref:2503.24152#pg0>.

Taro: That linkage is really important because it provides a concrete mechanism for how we can use device-level design choices to influence network-wide stability.

Dev: It moves the field from qualitative assessment to quantitative assessment, giving us a unified benchmark for designing power electronics.

Rosa: That unified benchmark sounds like exactly what we need if we're trying to standardize how engineers evaluate different control schemes across different applications.

Taro: And as an autonomy researcher, I see this as a way to design resilient systems where the failure of one component doesn't cascade into total system instability.

The paper's summary: Rosa: Moving on to what the paper suggests for improvements, it seems they are suggesting these metrics can be used practically in control design and physical placement optimization.

Dev: That’s right; they propose using the Forming Index as a "principled cost function" for GFM control design within an H-infinity robust control framework, specifically referencing Equation twenty-two.

Taro: So, this means designers can use the FI directly in their optimization problem to make sure the resulting controller is inherently robust against grid voltage variations across various timescales.

Rosa: That sounds like a way to ensure that even under difficult transient conditions, the converter maintains that stiff voltage source response they are aiming for.

Dev: Beyond control design, they also suggest using system strength indices to guide physical placement optimization by formulating an H-infinity norm optimization to maximize system strength.

Taro: So, this means we can use the bus strength metric kappa i(omega) to decide exactly where to put GFM converters in the topology—prioritizing weak buses for placement.

Rosa: That’s a powerful concept; it suggests that placement isn't just about proximity, but about optimizing for stability based on how much that specific location contributes to overall system stiffness.

Dev: They also suggest a heuristic approach where GFM devices should be placed preferentially at weak buses with lower bus strength, which is a practical guideline for system enhancement.

Taro: It’s interesting because it gives us a principled way to tackle the uncertainty in placement decisions that usually plague power systems design.

The paper's improvements: Rosa: So, wrapping up this discussion on "Quantifying Grid-Forming Behavior: Bridging Device-level Dynamics and System-Level Strength," it seems the authors have established a lot about linking device dynamics to system properties.

Dev: They've successfully introduced FI and system strength as quantifiable measures to bridge the gap between device behavior and grid performance, proving that GFM converters enhance system strength through formal propositions.

Taro: The main implication for me is that this framework provides a rigorous way to approach stability assessment in complex systems where we can actually predict how these components influence the network dynamics.

Rosa: It sounds like this paper gives us a solid toolkit for both designing robust controls and strategically placing equipment based on quantifiable system strength.

Dev: Indeed, the utility of the Forming Index and system strength indices is that they allow operators to move beyond just reacting to simple frequency deviations and instead proactively monitor the evolution of these metrics for potential instability.

Taro: My final thought is that this work sets a new standard for how we can approach stability analysis by grounding it in measurable, measurable quantities derived from the paper "Quantifying Grid-Forming Behavior: Bridging Device-level Dynamics and System-Level Strength."

Conclusion: Rosa: So we’ve been talking about "Quantifying Grid-Forming Behavior: Bridging Device-level Dynamics and System-Level Strength," which is really about tying device metrics to overall grid stability. It sounds like the authors have given us a much clearer way to benchmark how good a converter actually is at providing that necessary grid support.

Dev: Yeah, I agree, Rosa; moving from just looking at control structures to using something like the Forming Index as a quantifiable metric for voltage source behavior makes sense for assessing loop rates and latency issues.

Taro: From my side, it’s the way they formally prove that enhancing grid strength actually helps lower the system strength bound; that linkage between device action and network performance is what gets me interested.

Rosa: It really does give us a unified benchmark, which means instead of guessing how good a design is, we have these specific indices to measure.

Dev: And for the engineering side, having the FI as a cost function for H-infinity control design gives us a concrete objective to minimize when we're tuning those virtual impedances and droop coefficients.

Taro: It’s powerful because it helps us understand how placement decisions, guided by system strength, translate into real-world resilience against disturbances when the world starts acting unpredictably.

Rosa: Exactly; if we can use these metrics to guide placement at weak buses, it means we can proactively design systems that are inherently more robust before they even hit a major issue.

Dev: I'm still thinking about how fast this has to run in real-time; if the measurement of system strength needs high frequency updates, how do we keep the latency low enough for these metrics to be useful in a fast power system?

Taro: That’s a good point, Dev; if the monitoring happens too slowly, we miss the transient events where grid behavior changes rapidly and could see that system strength drop below critical levels.

Rosa: Well, I think this research really solidifies how we can systematically approach the design of power electronics for a more resilient grid.

Dev: It certainly gives us a rigorous foundation for testing failure modes; it moves the discussion past just "it works" to "how much strength does it add?"

Taro: And looking forward, I think we need to see how this framework extends into scenarios where multiple devices are interacting in highly dynamic, uncertain environments. I think this paper really gives us a solid toolkit for both designing robust controls and strategically placing equipment based on quantifiable system strength.

eess.SY, cs.SY

Submitted: 2025-03-31

Updated: 2026-10-05

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 83/100

The gist: Grid-forming (GFM) technology is widely regarded as a promising solution for future power systems dominated by power electronics, yet a universally accepted definition and precise quantification of

Key concepts

Forming Index (FI)
The FI is a metric applied at the device level to measure how well a converter follows or rejects grid voltage fluctuations. A lower FI value indicates stronger GFM capability, with an ideal voltage source having an FI of zero. It quantifies the converter's response to grid variations.
System Strength
System strength is a measure of how stiff the multi-bus voltage vector is against current or power disturbances. A larger system strength value means smaller voltage variations occur under disturbances, indicating better disturbance rejection capabilities in the power system.
Grid Strength ($\alpha(j\omega)$)
Grid strength represents the inherent stiffness of the electrical network itself, defined by the singular value of its admittance matrix. Enhancing this grid strength provides a lower bound for overall system strength, suggesting that improving network topology helps stabilize the system.
Bridging Device and System Levels
This concept describes using device-level metrics (like FI) to predict and quantify improvements at the system level (like $\kappa(j\omega)$). The core finding is that a converter with good device-level behavior (low FI) translates directly into enhanced system strength, creating a unified assessment tool.

Terminology

Summary

Grid-forming (GFM) technology is widely regarded as a promising solution for future power systems dominated by power electronics, yet a universally accepted definition and precise quantification of its behavior remain elusive, creating a significant disconnect between device and system levels. This paper introduces novel metrics—the Forming Index (FI) at the device level and system strength at the system level—and formally proves that GFM converters enhance system strength, providing a unified benchmark for converter design, placement, and stability assessment.

The gist: The proposed framework bridges device-level dynamics and system-level strength by introducing the Forming Index (FI) to quantify a converter’s response to grid voltage fluctuations and defining system strength as the sensitivity of the multi-bus voltage vector to current or power disturbances.

Device Level Quantification via Forming Index (FI)

The paper introduces the Forming Index (FI) as a novel metric at the device level to quantify a converter’s response to grid voltage fluctuations, moving beyond enumerating various control architectures. The FI is defined based on the concept of frequency smoothing and is formally defined as the maximum singular value of Sv(jω), i.e., F I(jω) = ¯σ[Sv(jω)] (Definition II.1). This index quantifies the extent to which a converter either follows or rejects grid variations, reflecting its voltage source behavior. A key finding is that "F I(jω) < 1 quantitatively defines GFM capability: a smaller F I(jω) corresponds to a stronger GFM capability, with F I(jω) = 0 representing an ideal voltage source." Furthermore, Lemma II.2 proves that the robust stability margin is given by∥Sv(s)∥∞:= max∀ω∈[0,∞) σ¯ [Sv(jω)] = max∀ω∈[0,∞) F I(jω).

System Level Quantification via System Strength

At the system level, a new quantitative measure of system strength is proposed to capture the multi-bus voltage stiffness. This metric is defined as the sensitivity of the multi-bus voltage vector to multi-bus current (or power) disturbances, mathematically expressed as κ(jω) = σ−1[ZCl(jω)] = σ[YCl(jω)] (Definition III.1). A larger κ corresponds to smaller voltage variations under disturbances, indicating stronger disturbance rejection. The paper further defines grid strength α(jω) as the singular value of the network admittance matrix, and bus strength κi(ω) as the maximum singular value derived from the sensitivity function ZCl,ij (s). Proposition III.3 establishes a formal link between system strength and grid strength: enhancing grid strength can improve a lower bound for system strength.

Bridging Device and System Levels

The central contribution of the paper is formally proving that GFM converters enhance system strength. This linkage is established through the following findings:

  1. A converter exhibiting GFM behavior at the device level enhances system strength.

  2. Proposition IV.2 states: "A GFM converter with F I(jω) < 1 connected to bus n+1 enhances the grid strength α(jω) according to αnew(jω) ≥ αold(jω) + const. × (1 − F I)."

  3. This enhancement is quantified by showing that reducing F I(jω) increases both the lower bound of system strength κ(jω) and the bus strength κn+1(jω).

Application and Design Potential

The proposed metrics provide a unified benchmark for various aspects of GFM converter design. The FI can serve as a principled cost function for GFM control design within an H∞ robust control framework (Equation 22). Additionally, the system strength indices are used to guide placement optimization:

  1. Placement of GFM converters can be formulated via H∞ norm optimization to maximize system strength.

  2. A heuristic approach suggests placing GFM devices preferentially at weak buses with lower bus strength.

Case Results and Validation

The utility of the proposed indices is validated using case studies on the IEEE 39-bus and 68-bus systems. The results demonstrate that:

  1. Grid strength and system strength are improved by enhancing (i.e., lowering) the FI of a connected bus, consistent with Proposition IV.2.

  2. GFM converters (VSG, VOC, droop, and PLL-VAC) all contribute to enhancing system strength, whereas GFL converters reduce it.

  3. The analysis effectively predicts the impact of different device types on grid strength and time-domain stability; for instance, connecting a VSG significantly increases its bus strength.

Improvements for AI systems

Here are the specific improvements that can be made to AI systems, based on the concepts introduced in this scientific paper, and what those improved systems could achieve:

) Control Design of GFM Converters (using Forming Index - FI):

The AI system can incorporate a loss function derived from the Forming Index, specifically by minimizing the weighted H-infinity norm of the sensitivity function, as formulated in Equation (22). This allows for the design of robust control architectures that explicitly aim to achieve a desired level of Grid-Forming Mode (GFM) behavior.

The improved AI system can:

  1. Design GFM control laws (e.g., for PLL-based converters) that are inherently robust against variations in grid voltage and frequency, ensuring the converter maintains a stiff voltage source response across sub-transient to transient timescales, even under varying Short Circuit Ratios (SCRs).

  2. Optimize control parameters (like virtual impedance or droop coefficients) not just for nominal performance, but specifically to minimize the Forming Index at critical frequencies (e.g., near 50Hz resonance peaks), thereby suppressing undesirable oscillations and increasing the robust stability margin.

) Placement of GFM Converters (using System Strength):

The AI system can be equipped with an optimization module that uses the concept of System Strength, defined by Definition III.1, to determine the optimal physical location for new GFM converters within a power grid topology. This involves formulating a constrained optimization problem where the objective is to maximize overall system strength or minimize the worst-case voltage offset.

The improved AI system can:

  1. Identify weak buses (those with low bus strength, defined by Definition III.4) and prioritize the placement of GFM converters there to achieve maximal enhancement of overall system stability.

  2. Formulate a heuristic or principled placement strategy that places converters where they yield the largest increase in grid strength and bus strength, leading to more resilient power systems that can withstand larger disturbances.

) Real-Time System Monitoring and Stability Assessment:

The AI system can use the derived metrics (FI, Grid Strength, Bus Strength) as real-time diagnostic indicators for grid health. Instead of relying solely on static SCR or simple frequency deviation checks, the AI can continuously calculate these indices based on measured voltage and current data.

The improved AI system can:

  1. Provide an immediate GFM Capability Score (based on FI) for every connected converter, allowing operators to instantly assess whether a specific device is behaving as a desired GFM source or a Grid-Following (GFL) element during transient events.

  2. Predict potential instability by monitoring the evolution of the System Strength metric; if it drops below critical thresholds (e.g., < 0.5), the AI can trigger automated corrective actions or alert operators to imminent voltage limit violations, providing a proactive measure against system collapse that traditional metrics might miss.

) Enhanced Control Strategy Classification:

The paper demonstrates that different control structures (PLL-PQ, VSG, dVOC, PLL-VAC) can yield similar GFM behavior if tuned correctly (i.e., F I < 1). The AI system can use the FI as a universal classifier to distinguish between these architectures based on dynamic response rather than just the control structure.

The improved AI system can:

  1. Automate the classification of converter control strategies by calculating their respective Forming Indices across a frequency spectrum, providing an objective metric to determine if a specific implementation is truly GFM or GFL in practice.

  2. Suggest optimal tuning parameters for existing converters that maximize their FI, effectively re-engineering the control behavior to achieve superior system performance without requiring a complete hardware redesign.

Sources

Related papers