Modal Analysis of Spatial Load Correlation in AI Data Center-Dominated Power Systems

arXiv:2606.13847 · eess.SY, cs.SY · Submitted 2026-06-11 · Read on arXiv

Listen

Transcript

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

Rosa: Today's paper: "Modal Analysis of Spatial Load Correlation in AI Data Center-Dominated Power Systems".

Dev: The gist: This paper proposes a data-driven modal framework based on Dynamic Mode Decomposition (DMD) to analyze nonstationary spatial load correlation in AI data center-dominated power systems,

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

Paper summary: Rosa: So, we're talking about this paper called "Modal Analysis of Spatial Load Correlation in AI Data Center-Dominated Power Systems". It looks at how loads in these massive data center grids behave when they aren't behaving the way classical models expect them to.

Dev: Basically, the authors are pointing out that the correlations between power fluctuations at different buses aren't steady; they change over time and space in ways that time-averaged methods just miss.

Rosa: They claim these spatial and temporal correlations in AI data center grids are episodic and nonstationary, which means we need a way to look at what’s happening in real time instead of just looking at the average.

Dev: The core idea is using Dynamic Mode Decomposition, or DMD, to build a state representation from raw recordings that lets them analyze these transient structures without assuming everything is constant.

Taro: I'm interested in how this moves beyond standard analysis because classical planning assumes loads are independent and scale with the square root of the number of elements when they aren't actually correlated.

Rosa: Right, so the paper sets up a correlation state-space using raw RTDS recordings and then uses spectral characterization combined with DMD to pull out these dominant spatial correlation modes and figure out what they mean physically.

Dev: They define this pairwise Pearson correlation over a window of length Tw using equation one, which essentially isolates the spatial coherence structure from just individual load changes or ramp transients.

Taro: That state vector construction seems key because it’s designed to be the natural way to analyze aggregate fluctuation statistics and contingency exposure when you have these kinds of complex dependencies.

Rosa: Then they apply DMD, seeking a linear operator A such that X prime is approximated by AX, but they smart about this by avoiding direct computation of A and instead using a low-rank projection from the economy Singular Value Decomposition.

Dev: They project the reduced operator onto the dominant Proper Orthogonal Decomposition subspace to get Ae equals U top X prime V Sigma minus one which then allows them to find eigenvalues and eigenvectors that give them continuous-time quantities like oscillation frequency fk and growth or decay rate sigma k <ref:2606.13847#pg2>.

Taro: The physical interpretation of those eigenvalues is what really matters here because they link the math back to the physics of the system.

Rosa: Exactly. They say the position of an eigenvalue on a complex plane tells you something direct about what’s happening physically in the grid dynamics.

Paper summary: Dev: For example, an eigenvalue sitting on the unit circle with sigma k being zero means there's sustained oscillatory coherence, like from settled HVAC cycling, and an eigenvalue inside the unit circle with a negative sigma k indicates a coherence burst that's naturally decaying.

Taro: And if it’s outside the unit circle with a positive sigma k, that signals intensifying coherence in the system.

Rosa: They also map the frequency fk to specific physical coupling mechanisms, which is important because you get two different ways to understand the underlying dynamics by looking at both fk and phi k.

Dev: They use a sliding-window portrait approach where they run DMD over windows of length T DMD advanced in steps of delta t, and plotting the dominant mode frequency and energy against the window index gives them a time-frequency portrait of how these correlation dynamics evolve.

Taro: A specific indicator they found is that when µ(n)k is greater than one that acts as a precursor, detectable before those pairwise coefficients actually peak up, which suggests it could be used for streaming operational alerts <ref:2606.13847#pg2>.

Rosa: They quantify the lead time from when that precursor flag appears to the subsequent peak of the dominant pairwise correlation as being between four point zero and eleven seconds, with a mean around four point four seconds.

Dev: This is interesting because it means we can get an early warning signal about these intermittent intensification events caused by workload orchestration that global analysis might miss.

Taro: But I have to ask, how does this all translate to real-world stability issues for someone just listening who doesn't work in a lab?

Rosa: Well, the paper validates this finding through cross-validation with RTDS voltage coherence, showing that both the load-domain DMD portrait and the voltage coherence findings reflect genuine network-level coupling.

Dev: The cross-validation specifically looks at the magnitudesquared coherence gamma 2ij omega of voltage deviations during flagged or sparse episodes <ref:2606.13847#pg2>.

Taro: And what we see is that during a flagged episode, the dominant peak hits zero point three six six Hz with a gamma two value of zero point nine six two, which falls in the workload orchestration band, while in sparse episodes it shifts up to sixteen point six Hz with a gamma two of zero point nine nine two in the converter band.

Rosa: That shift between those frequencies really shows how different types of coupling—orchestration versus converter dynamics—are happening at different times during these events.

Dev: The analysis also showed that the slow/thermalband peak moves from a gamma two of zero point eight zero seven at zero point zero six one Hz during flagged episodes up to a gamma two of zero point nine five four at zero point zero nine two Hz in sparse episodes, which directly matches the load-domain DMD portrait they found earlier, linking the two domains together.

Paper summary: Taro: So what this means for someone just listening is that these massive AI data center systems aren't just big power plants; they have these dynamic internal structures driven by how workloads are orchestrated across all those facilities simultaneously.

Rosa: It confirms that the nonstationary nature of load correlation isn't just some noise; it’s a structured phenomenon you can detect with this modal framework, and it helps us understand the physical coupling mechanisms at play.

Dev: The paper establishes a data-driven modal framework based on Dynamic Mode Decomposition to analyze this nonstationary spatial load correlation in AI data center-dominated power systems, which moves past the limitations of classical methods that assume stationarity.

Taro: It’s important to remember that the authors also point out a limitation: their rank-three model doesn't produce eigenvalues in either the workload orchestration band or the converter control band, and they explain this is because fast converter dynamics are effectively decoupled from slower thermal behavior, and workload orchestration is driven by independent semi-Markov state machines with no inter-facility coordination.

Rosa: That’s a big caveat; so while the DMD method finds these modes, it doesn't fully capture the dynamics within those specific operational bands because of how fast the converter controls are separated from the slower thermal load effects.

Dev: So, to wrap up this discussion on "Modal Analysis of Spatial Load Correlation in AI Data Center-Dominated Power Systems", we see a way to use DMD to extract physical insights from complex power system recordings that change over time.

Rosa: So, in simple terms, this paper uses Dynamic Mode Decomposition on load correlations over time to find patterns—sustained coherence, decaying transients, or intensifying events—that you can detect in real-time using a sliding window portrait and a precursor indicator based on the eigenvalue magnitude.

Dev: The main implication is that these dynamic correlations are episodic and nonstationary, which means standard methods fail, so we need this approach to characterize the physical mechanisms producing inter-bus coherence in these AI data center grids.

Taro: It changes how we think about stability because it shows that workload orchestration contributes intermittent intensification events at the transmission scale that global analysis often misses.

Rosa: And the authors confirm this coupling is real, not just an artifact of the modeling, because they cross-validated their findings against actual RTDS voltage coherence data.

Dev: This paper offers a viable operational diagnostic tool for monitoring these complex systems without needing to assume stationarity, providing a concrete way to look at the transient structure of AI power fluctuations.

Conclusion: Rosa: So, we're looking at this paper about "Modal Analysis of Spatial Load Correlation in AI Data Center-Dominated Power Systems." Basically, they're using Dynamic Mode Decomposition to figure out how power loads across different parts of these massive data center grids behave when things aren't steady.

Dev: Yeah, the authors are showing that classical methods that assume everything is constant just don't work for this kind of nonstationary stuff. They’re not looking at averages anymore; they’re tracking the actual structure in time.

Taro: What does it mean for someone just listening who doesn't work on a grid? It suggests that these massive AI power systems have these specific, dynamic ways they link up between different buildings or buses over time.

Rosa: Exactly. They found that the way loads correlate isn't random noise; it’s structured, episodic behavior driven by things like workload orchestration.

Dev: The numbers show them tracking this using a sliding window portrait and an indicator based on eigenvalue magnitudes that lets you spot these intensification events before they get really big.

Taro: And they did cross-validate this against real voltage data, which is important because it proves the correlation they found in the load domain actually shows up in the actual electrical signals.

Rosa: So, it moves beyond just saying "it's correlated" to showing *how* that correlation evolves—whether it's growing, decaying, or oscillating—and linking those mathematical modes to physical things like HVAC cycling or converter control.

Dev: It gives engineers a way to monitor these systems in real time by looking for these specific frequency shifts and growth rates rather than just waiting for a failure.

Taro: But the authors also noted a limitation, which is important. Their model doesn't fully capture dynamics in the fast converter control band because those things are decoupled from the slower thermal load effects they were tracking.

Rosa: Right, so it’s a powerful tool for understanding the large-scale load structure and its episodic behavior, even if it can't fully map every tiny detail of every single component.

Dev: It offers a diagnostic way to see the health of these systems based on how their internal connections are behaving dynamically rather than just static measurements.

Taro: We’ll look at how this kind of real-time modal analysis might help us predict stability issues when these AI clusters get really busy or stressed.

Michigan State University · Sandia National Laboratories

eess.SY, cs.SY

Submitted: 2026-06-11

Updated: 2026-06-30

Comments: To appear in proceedings of 8th International Conference on Smart Energy Systems and Technologies, September 2-4, 2026 | Ciudad Real, Spain

DOI: 10.1109/SEST67798.2026.11712308

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

Importance score: 88/100

The gist: The gist: This paper proposes a data-driven modal framework based on Dynamic Mode Decomposition (DMD) to analyze nonstationary spatial load correlation in AI data center-dominated power systems,

Key concepts

Spatial Load Correlation: This refers to the statistical dependence between active power fluctuations at different locations in a grid. When loads are correlated, aggregate fluctuations grow faster than expected under standard assumptions, meaning traditional planning and reserve sizing based on independence fail to account for the actual risk.
Dynamic Mode Decomposition (DMD): DMD is an algorithm used here to find linear operators that best describe how the system's states evolve over time. It helps extract dominant spatial correlation modes from complex data by finding eigenvalues and eigenvectors, which reveal oscillation frequencies and growth/decay rates of the system's dynamics.
Correlation State-Space Formulation: This method creates a mathematical state vector by calculating pairwise Pearson correlations over specific time windows. This isolates the structure of spatial coherence from individual load changes, making it an ideal input for modal analysis to study aggregate fluctuation statistics and contingency exposure.

Terminology

Spatial Load Correlation — 15× in this paper · explained in 2 episodes
Dynamic Mode Decomposition — 14× in this paper
Dynamic Mode Decomposition (DMD) — 6× in this paper · explained in 3 episodes
Proper Orthogonal Decomposition — 4× in this paper
Fidelity — 2× in this paper · explained in 2 episodes
Real-Time Digital Simulator — 2× in this paper
Reconstruction Error — 2× in this paper · explained in 2 episodes
Alignment — 1× in this paper · explained in 1 episode
Anti-correlation — 1× in this paper · explained in 3 episodes
Benchmark — 1× in this paper · explained in 2 episodes
Core — 1× in this paper · explained in 2 episodes
Dynamical regimes — 1× in this paper · explained in 2 episodes
Heating Ventilation and Air Conditioning — 1× in this paper
Non-stationarity — 1× in this paper · explained in 2 episodes
Ornstein–Uhlenbeck Process — 1× in this paper · explained in 3 episodes
Spectral Gap — 1× in this paper · explained in 5 episodes
Temporal Correlations — 1× in this paper · explained in 2 episodes

Summary

The gist: This paper proposes a data-driven modal framework based on Dynamic Mode Decomposition (DMD) to analyze nonstationary spatial load correlation in AI data center-dominated power systems, which addresses the limitations of classical methods that assume stationarity.

Spatial Load Correlation

Spatial load correlation refers to the statistical dependence among active power fluctuations at different transmission buses Under classical planning assumptions, loads at geographically separated buses vary independently so that aggregate fluctuations scale with the square root of the number of demand elements. When loads are correlated, aggregate fluctuations scale faster, diversity factors understate variance, and reserve margins sized for independence are insufficient. The characterization of this phenomenon in AI data center-dominated grids, the physical mechanisms that produce inter-bus coherence, and their implications for voltage and frequency stability are established in prior work.

Methodology: Correlation State-Space Formulation

The proposed method constructs a correlation state-space from raw RTDS recordings and, through spectral characterization and DMD, extracts dominant spatial correlation modes and their physical interpretations. The pairwise Pearson correlation over a window of length Tw is defined by the equation. This state vector isolates the spatial coherence structure from individual load and single-bus ramp transients, and is the natural choice for modal analysis of aggregate fluctuation statistics and correlated contingency exposure.

Methodology: Dynamic Mode Decomposition (DMD)

The DMD algorithm seeks the best-fit linear operator A such that X′ ≈ AX. Direct computation of A ∈ R Np×Np is avoided to favor the low-rank projection from the economy SVD of X. The reduced operator projected onto the dominant Proper Orthogonal Decomposition (POD) subspace is Ae = U⊤X′V Σ−1 ∈ R r×r. Eigendecomposition AWe = WΛ yields eigenvalues and eigenvectors, from which continuous-time quantities are recovered: oscillation frequency fk and growth/decay rate σk.

Physical Interpretation of Eigenvalues

The position of µk on the complex plane carries direct physical meaning. An eigenvalue on the unit circle (µk = 1, σk = 0) corresponds to sustained oscillatory coherence driven by an active periodic mechanism such as settled HVAC cycling. An eigenvalue strictly inside the unit circle (µk < 1, σk < 0) indicates a coherence burst in natural decay. An eigenvalue outside the unit circle (µk > 1, σk > 0) signals intensifying coherence. The frequency fk maps to the physical coupling mechanism via Table I.

Sliding-Window Portrait and Analysis

The proposed approach applies DMD over windows of length T DMD w advanced in steps of δt. The dominant mode frequency and energy plotted against window index constitute a time-frequency portrait of the correlation dynamics. The µ(n)k > 1 criterion constitutes a precursor indicator, detectable before pairwise coefficients reach their peak, suitable for deployment as a streaming operational alert. The median lead from the flag onset to the subsequent peak of the dominant pairwise correlation is 4.0 s, with a mean of 4.4 s and a range of 0–11 s.

Conclusion and Validation

The results establish that the modal growth indicator is a viable operational diagnostic for the episodic inter-bus coherence that characterizes AI data center clusters at transmission scale. Cross-validation with RTDS voltage coherence confirms that both findings reflect genuine network-level coupling. The sliding-window portrait reveals that workload orchestration contributes intermittent intensification events that global analysis alone would miss.

The paper is a data-driven modal framework based on Dynamic Mode Decomposition (DMD) to analyze nonstationary spatial load correlation in AI data center-dominated power systems, which addresses the limitations of classical methods that assume stationarity. The gist: This paper proposes a data-driven modal framework based on Dynamic Mode Decomposition (DMD) to analyze nonstationary spatial load correlation in AI data center-dominated power systems, which addresses the limitations of classical methods that assume stationarity.

How it works

The method constructs a correlation state-space from raw RTDS recordings and, through spectral characterization and DMD, extracts dominant spatial correlation modes and their physical interpretations.
The pairwise Pearson correlation over a window of length Tw is defined by the equation. This state vector isolates the spatial coherence structure from individual load and single-bus ramp transients, and is the natural choice for modal analysis of aggregate fluctuation statistics and correlated contingency exposure.
The DMD algorithm seeks the best-fit linear operator A such that X′ ≈ AX. Direct computation of A ∈ R Np×Np is avoided to favor the low-rank projection from the economy SVD of X.
The reduced operator projected onto the dominant Proper Orthogonal Decomposition (POD) subspace is Ae = U⊤X′V Σ−1 ∈ R r×r. Eigendecomposition AWe = WΛ yields eigenvalues and eigenvectors, from which continuous-time quantities are recovered: oscillation frequency fk and growth/decay rate σk.
The proposed approach applies DMD over windows of length T DMD w advanced in steps of δt. The dominant mode frequency and energy plotted against window index constitute a time-frequency portrait of the correlation dynamics.
The µ(n)k > 1 criterion constitutes a precursor indicator, detectable before pairwise coefficients reach their peak, suitable for deployment as a streaming operational alert.
Cross-validation with RTDS voltage coherence confirms that both findings reflect genuine network-level coupling.

Limitations of Global DMD

The rank-3 model does not produce eigenvalues in either the workload orchestration band or the converter control band. Two mechanisms explain this absence: fast converter dynamics are effectively decoupled from the slower thermal behavior, and workload orchestration is driven by independent semi-Markov state machines with no inter-facility coordination.

Cross-Validation Against RTDS Voltage Signals

The cross-validation is performed through magnitudesquared coherence γ2ij (ω) of voltage deviations at flagged and sparse episodes. The flagged-episode dominant peak at 0.366 Hz (γ2 = 0.962) falls in the workload orchestration band; during sparse episodes the peak shifts to 16.6 Hz (γ2 = 0.992) in the converter band. The Slow/Thermalband peak shifts from γ2 = 0.807 at 0.061 Hz (flagged) to γ2 = 0.954 at 0.092 Hz (sparse), directly cross-validating the load-domain DMD portrait <ref:2606.

Improvements for AI systems

Improved early-warning signal for correlation intensification: The proposed modal growth indicator provides an early-warning signal of correlation intensification, with a lead of of about 4 s before pairwise coherence reaches its peak. This allows operators to detect impending instability before the pairwise coefficients reach their peak.
Enhanced operational diagnostics via sliding-window analysis: The method produces a time-frequency portrait where the dominant mode energy exhibits rapid transitions between high-concentration episodes (Ek → 1) and diffuse episodes (Ek ≈ 0.5), identifying specific dynamical regimes like a phase-locked regime where one mode captures nearly all state-vector variance.
Predictive reserve pre-positioning: An autoregressive model of the dominant mode amplitude can be used to predict spatial concentration index exceedance over a short horizon, giving operators lead time for reserve pre-positioning and correlation-aware dispatch.