arxiv: 2605.00769 · v1 · submitted 2026-05-01 · 📡 eess.SY · cs.SY

Voltage Ride-Through in Large Loads- A Dual PQ Approach

Amir Norouzi , Michael Morel This is my paper

Pith reviewed 2026-05-09 19:05 UTC · model grok-4.3

classification 📡 eess.SY cs.SY

keywords voltage ride-throughlarge loadsdata centersreactive power compensationdual PQ approachactive power supportgrid voltage dipspower system stability

0 comments p. Extension

The pith

Traditional reactive power compensation reaches hard limits during severe grid voltage dips in large loads, requiring non-grid resources with dynamic active and reactive power support.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper analyzes voltage ride-through for large loads such as AI data centers during power grid disturbances. It shows that the conventional method of reactive power compensation alone cannot maintain acceptable load voltage because of capacity limits in the load's distribution infrastructure and grid constraints. A dual PQ approach is introduced that incorporates dynamic active power from non-grid sources to extend ride-through capability. In extreme voltage dips the analysis indicates that acceptable load voltage may become unattainable, leading to possible disconnection. This matters for bulk power system stability because large loads can affect overall reliability if they trip offline.

Core claim

The central claim is that the traditional reactive-power-only approach for voltage ride-through is inadequate for large loads because capacity limits create a practical and theoretical ceiling on corrective reactive power; therefore a dual active-and-reactive-power (PQ) strategy is required in which non-grid resources supply dynamic P and Q, and that at sufficiently deep grid voltage dips maintaining load voltage within acceptable bounds becomes unattainable, which may force disconnection from the grid.

What carries the argument

The dual PQ approach, which augments reactive power compensation with dynamic active power injection from non-grid resources to overcome infrastructure capacity ceilings during transient voltage dips.

If this is right

Large loads will need to integrate or contract for non-grid dynamic active power sources to achieve reliable voltage ride-through.
Extreme grid voltage dips may result in unavoidable disconnection of large loads even with optimal compensation.
Grid stability analyses must account for the possibility that reactive power support reaches its limit before active power support is considered.
Design standards for data centers and similar loads may shift toward requiring dual P and Q capability rather than reactive compensation alone.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Data center operators may need to redesign on-site generation or storage systems to provide active power during dips rather than relying only on grid reactive support.
The same capacity-limit reasoning could apply to other large industrial loads, suggesting a broader re-examination of ride-through requirements across sectors.
Real-time monitoring of both P and Q margins at the load connection point could become necessary to predict when ride-through will fail.

Load-bearing premise

Capacity limits inside the load's power distribution infrastructure and grid constraints impose a hard ceiling on how much reactive power can be used for voltage correction.

What would settle it

A numerical simulation or field measurement that demonstrates whether reactive power compensation alone can keep load voltage above the minimum acceptable level during a specified deep voltage dip, such as 0.2 pu at the grid side, without any active power support.

Figures

Figures reproduced from arXiv: 2605.00769 by Amir Norouzi, Michael Morel.

**Figure 1.** Figure 1: A Real Case of Grid Disturbance and Voltage Dip. powered through Uninterruptible Power Supply (UPS) units [5]. UPS units have limited battery resources as emergency source of power and are designed to switch to these sources upon some defined disturbances on the power grid. From the perspective of the grid, switching of a computing load to non-grid resources is loss of that load, and depending on the size … view at source ↗

**Figure 2.** Figure 2: Representation of a Substation Serving a Large Load. power (𝑃), and reactive power (𝑄) are interrelated in loads is essential view at source ↗

**Figure 3.** Figure 3: Vector Diagram for view at source ↗

**Figure 4.** Figure 4: Load Side Power Circles at Constant Load Voltage. transformer’s reactance is unchanged, and hence circle diagrams remain applicable, under the conditions that voltage ride-through is reasonably expected to be achieved. B Reactive Power and Voltage: Conceptual and Practical Constraints The goal of the conventional reactive power compensation is power factor correction, i.e., making the load power factor as … view at source ↗

**Figure 5.** Figure 5: Apparent Power Versus |𝑉𝑆 | at Constant 𝑃 and 𝑉𝐿. of the non-grid active power depends on the extend of the dip in |𝑉𝑆 | as well as the power demand of the load. For instance, in view at source ↗

read the original abstract

This paper provides a detailed investigation of voltage ride-through in large loads, such as Artificial Intelligence data centers. Voltage ride-through capability of large loads during transient disturbances in the power grid is important because of the potential impact on the stability and reliability of the Bulk Power System. A mathematical analysis is presented and it is shown how the traditional approach, based on reactive power compensation, may not be adequate for voltage ride-through in large loads. Ultimately, due to capacity limits of the load's power distribution infrastructure and grid's constraints, there is a limit to using reactive power as a corrective tool. A new dual active and reactive power (PQ) approach is proposed in which non-grid resources with dynamic P and Q capabilities are shown to be needed to help with voltage ride-through. Additionally, the analysis illustrates that at extreme voltage dips in the power grid maintaining an acceptable level of load voltage can become practically or theoretically unattainable, which may lead to the load's disconnection from the grid. Analytical results are provided with practical numerical examples.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper argues reactive power alone hits infrastructure limits during voltage dips for big loads like data centers, so dual active-reactive control from external resources is needed, but the limits lack a clear model.

read the letter

The main point is that for very large loads, reactive compensation runs into hard ceilings from the distribution infrastructure and grid constraints, making it insufficient for voltage ride-through, and the authors push for non-grid resources that can adjust both P and Q to keep the load voltage acceptable. They back this with a mathematical analysis and numerical examples showing cases where maintaining load voltage becomes unattainable at extreme dips, potentially forcing disconnection. That flags a real issue as AI data centers scale up and affect bulk system stability. What works is the focus on practical limits rather than ideal compensation, plus the concrete examples that illustrate when things break. The analysis does engage with standard power flow ideas without obvious circularity. The soft spot is exactly the one in the stress-test: the claim that reactive-only methods fail rests on capacity limits acting as fixed ceilings, but the paper does not lay out an explicit model of those limits from conductor ratings, transformer impedances, or protection settings, nor does it test sensitivity to modest design changes like thicker feeders. Without that, it is hard to judge whether the dual PQ requirement is truly necessary or if better infrastructure would suffice. The non-grid resources also stay vague on what they actually are and how they avoid affecting the load itself. This is for power engineers working on grid codes or facility design for high-power consumers. A reader dealing with voltage stability in growing loads would pick up useful warnings and examples, even if the evidence is preliminary. It shows clear thinking on the problem and honest engagement with the literature on ride-through, so it deserves a serious referee to verify the derivations and push for the missing infrastructure details. I would send it to peer review with targeted revisions rather than desk reject.

Referee Report

2 major / 2 minor

Summary. The manuscript investigates voltage ride-through capability for large loads such as AI data centers during transient grid disturbances. It presents a mathematical analysis demonstrating that traditional reactive power compensation is inadequate due to capacity limits in the load's power distribution infrastructure and grid constraints. A dual PQ approach is proposed, requiring non-grid resources with dynamic active and reactive power capabilities to support voltage maintenance. The analysis further shows that extreme voltage dips can render acceptable load voltage levels practically or theoretically unattainable, potentially causing disconnection from the grid. Analytical results are illustrated with practical numerical examples.

Significance. If substantiated, the work would be significant for bulk power system stability by highlighting limitations of conventional reactive compensation for high-demand loads and motivating dual PQ strategies involving external resources. The analytical framing with numerical examples provides a foundation for further study, though the absence of open reproducible code or machine-checked derivations limits immediate verifiability.

major comments (2)

[Abstract and mathematical analysis] Abstract and the mathematical analysis section: The central claim that capacity limits of the load's power distribution infrastructure and grid constraints impose a hard ceiling on reactive power compensation (rendering the traditional approach inadequate) is load-bearing for the dual-PQ proposal. However, no explicit infrastructure model—such as equations deriving the limit from conductor ratings, transformer impedances, or protection settings—is provided, nor is a sensitivity analysis shown. If these bounds can be relaxed by modest design changes, the inadequacy conclusion does not follow.
[Analysis of extreme voltage dips] The section on extreme voltage dips: The illustration that maintaining an acceptable load voltage level can become practically or theoretically unattainable lacks a precise definition of the 'acceptable level' (e.g., a specific voltage tolerance band or duration) and the boundary conditions or equations leading to disconnection. This weakens the practical implications for grid reliability.

minor comments (2)

[Introduction] The term 'non-grid resources' is introduced in the abstract and proposal but is not explicitly defined or distinguished from standard grid-connected devices in the introduction or terminology section, which may cause ambiguity for readers.
[Numerical examples] Numerical examples in the results section would benefit from tabulated parameter values (e.g., assumed infrastructure ratings or dip depths) to allow independent verification of the analytical claims.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed feedback on our manuscript. The comments highlight important areas for improving clarity and rigor, and we address each major comment below. We will incorporate revisions to strengthen the presentation of the infrastructure limits and the definitions for extreme voltage dips.

read point-by-point responses

Referee: [Abstract and mathematical analysis] Abstract and the mathematical analysis section: The central claim that capacity limits of the load's power distribution infrastructure and grid constraints impose a hard ceiling on reactive power compensation (rendering the traditional approach inadequate) is load-bearing for the dual-PQ proposal. However, no explicit infrastructure model—such as equations deriving the limit from conductor ratings, transformer impedances, or protection settings—is provided, nor is a sensitivity analysis shown. If these bounds can be relaxed by modest design changes, the inadequacy conclusion does not follow.

Authors: We agree that an explicit derivation of the capacity limits would strengthen the central claim. In the revised manuscript, we will add a dedicated subsection to the mathematical analysis section that provides the infrastructure model, including equations for reactive power limits derived from conductor current ratings (ampacity), transformer thermal limits and impedance, and typical protection settings. We will also include a sensitivity analysis showing how these parameters affect the achievable compensation for large loads such as AI data centers. This analysis will demonstrate that, under realistic conditions for the load scales considered, the limits remain binding and cannot be relaxed sufficiently by modest design changes without major infrastructure investments, thereby supporting the need for the dual PQ approach. revision: yes
Referee: [Analysis of extreme voltage dips] The section on extreme voltage dips: The illustration that maintaining an acceptable load voltage level can become practically or theoretically unattainable lacks a precise definition of the 'acceptable level' (e.g., a specific voltage tolerance band or duration) and the boundary conditions or equations leading to disconnection. This weakens the practical implications for grid reliability.

Authors: We concur that precise definitions are essential for practical implications. In the revision, we will update the section on extreme voltage dips to define the acceptable load voltage level explicitly, referencing established standards such as the ITIC curve for voltage tolerance bands and durations. We will also present the explicit boundary conditions and governing equations that determine when the load voltage cannot be maintained within these tolerances, leading to disconnection. This will include the mathematical conditions under which even dual PQ resources become insufficient, thereby clarifying the implications for bulk power system reliability. revision: yes

Circularity Check

0 steps flagged

No circularity: claims rest on standard power-system limits without self-referential derivations or fitted predictions

full rationale

The paper presents a mathematical analysis showing limits of reactive-only compensation due to infrastructure capacity and proposes a dual PQ approach. No equations, parameter fits, or derivations appear in the provided text that reduce any result to its own inputs by construction. Capacity limits are asserted from domain knowledge of conductors, transformers, and grid constraints rather than fitted or self-defined within the paper. No self-citations, uniqueness theorems, or ansatzes are invoked in a load-bearing way. The derivation chain is self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

Based on abstract only; the central claim rests on standard power-system assumptions about reactive power limits and the availability of non-grid dynamic resources, with no free parameters or new entities quantified.

axioms (1)

domain assumption Reactive power compensation is limited by capacity of load power distribution infrastructure and grid constraints
Invoked to conclude that the traditional approach may not be adequate for large loads.

invented entities (1)

non-grid resources with dynamic P and Q capabilities no independent evidence
purpose: To provide active and reactive power support for voltage ride-through
Introduced as necessary for the dual PQ approach; no independent evidence or implementation details given in abstract.

pith-pipeline@v0.9.0 · 5473 in / 1465 out tokens · 47840 ms · 2026-05-09T19:05:11.091712+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

13 extracted references · 5 canonical work pages · 1 internal anchor

[1]

Characteristics and Risks of Emerg- ing Large Loads

NERC Large Load Task Force, “Characteristics and Risks of Emerg- ing Large Loads”, North American Electric Reliability Corporation, July 2025

2025
[2]

Electricity demand and grid impacts of ai data centers: Challenges and prospects,

Xin Chen,et al., “Electricity Demand and Grid Impacts of AI Data Centers: Challenges and Prospects”, https://arxiv.org/abs/2509.07218v4 [eess.SY], November 2025

work page arXiv 2025
[3]

Choukse, B

Esha Choukse,et al., “Power Stabilization for AI Training Data centers”, https://arxiv.org/abs/2508.14318v1 [cs.AR], August 2025

work page arXiv 2025
[4]

Texas Loads Ride Toward Grid Stability: Voltage Ride Through of Large Power Electronic Loads

J. Conto,et al., “Texas Loads Ride Toward Grid Stability: Voltage Ride Through of Large Power Electronic Loads”, inIEEE Power and Energy Magazine, vol. 23, no. 5, pp. 56-67, Sept.-Oct. 2025

2025
[5]

Review of Industry Efforts and Standards of Grid Readiness For Data Center Deployment

Data Centers Standards Needs Analysis and Recommendations Activity, “Review of Industry Efforts and Standards of Grid Readiness For Data Center Deployment”, vol., no., pp.1-62, 29 Jan. 2026. 9

2026
[6]

AI load dynamics–a power electronics perspective,

Yuzhuo Li, Yunwei Li, “AI Load Dynamics-A Power Electronics Perspec- tive”, https://arxiv.org/abs/2502.01647v2 [cs.AR], February 2025

work page arXiv 2025
[7]

Optimal County-Level Siting of Data Centers in the United States

Maria Vabson,et al., “Optimal County-Level Siting of Data Centers in the United States”, https://arxiv.org/abs/2601.16315v1 [eess.SY], January 2026

work page arXiv 2026
[8]

Transformers

A. E. Fitzgerald, Charles Kingsley, Jr., Stephen D. Umans, “Transformers”, inElectric Machinery, 6th ed., McGraw-Hill, 2003, ch. 2 [12]IEEE Application Guide for AC High-Voltage Circuit Breakers>1000 Vac Rated on a Symmetrical Current Basis, IEEE Std C37.010-2016 (Revision of IEEE Std C37.010-1999), April 2017

2003
[9]

Current and Voltage Relations on a Transmis- sion Line

William D. Stevenson, Jr., “Current and Voltage Relations on a Transmis- sion Line”, inElements of Power System Analysis, 4th ed., McGraw-Hill, 1982, ch. 5

1982
[10]

Line Model and Performance

Hadi Saadat, “Line Model and Performance”, inPower System Analysis, 2𝑛𝑑 ed., McGraw-Hill, 2004, ch. 5

2004
[11]

The Circle Diagram as a Teaching Tool

M. Siman, “The Circle Diagram as a Teaching Tool”, inIEEE Transactions on Education, vol. 11, no. 1, pp. 50-56, March 1968

1968
[12]

Basic Concepts

William D. Stevenson, Jr., “Basic Concepts”, inElements of Power System Analysis, 4th ed., McGraw-Hill, 1982, ch. 2

1982
[13]

Safe Reconnection Time for Large-Scale Data Center Loads: An Analytical Framework for Transient Stability Assessment

Ahmed Mesfer Alkhudaydi, Bai Cui, “Safe Reconnection Time for Large- Scale Data Center Loads: An Analytical Framework for Transient Stability Assessment”, https://arxiv.org/abs/2604.24009v1 [eess.SY], April 2026. 10

work page internal anchor Pith review Pith/arXiv arXiv 2026