pith. sign in

arxiv: 2604.07777 · v1 · submitted 2026-04-09 · 📡 eess.SY · cs.SY

Differences in Small-Signal Stability Boundaries Between Aggregated and Granular DFIG Models

Leyou Zhou , Mucheng Li , Xiaojie Shi , Meng Zhan , Juanjuan Wang , Dan Wu This is my paper

Pith reviewed 2026-05-10 18:03 UTC · model grok-4.3

classification 📡 eess.SY cs.SY
keywords DFIGsmall-signal stabilityaggregated modelsgranular modelsstability boundarieswind farmsD-decomposition
0
0 comments X

The pith

Aggregated models of wind turbine generators produce stability regions that differ in shape and critical modes from detailed multi-unit models.

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

The paper investigates the impact of model aggregation on small-signal stability analysis for doubly-fed induction generator wind farms. It constructs detailed models with one, two, or three units and compares them to their aggregated single-unit equivalents. Using a D-decomposition-based ray-extrapolation technique, the authors map out stability boundaries across planes of control parameters and operating conditions. They find that aggregation alters the basic shape of these regions, the associated critical modes, and how the regions evolve, which can lead to incorrect judgments about stability margins. This matters because many existing studies of wind farm oscillations rely on aggregated models, and discrepancies could affect the reliability of stability predictions as wind power penetration grows.

Core claim

Aggregated models of DFIGs yield stability regions within parameter planes of control settings and operating conditions that differ from granular models in terms of basic shape, critical modes, and evolution patterns, which poses a risk of misjudging stability margins.

What carries the argument

The D-decomposition-related ray-extrapolation method for characterizing the small-signal stability region of nonlinear DFIG models by delineating boundaries under numerous parameter combinations.

If this is right

  • Aggregated models may identify different critical modes than granular models for the same conditions.
  • The shape of stability boundaries changes with the level of aggregation.
  • Stability margins could be misjudged, leading to either overly conservative or insufficiently cautious operating points.
  • Evolution patterns of stability regions under varying parameters diverge between model types.

Where Pith is reading between the lines

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

  • Past analyses of broadband oscillations in wind farms that used aggregated models may require re-examination for accuracy.
  • Wind farm operators might need to incorporate more detailed modeling when assessing stability under specific control settings.
  • Extending the comparison to larger numbers of units could show whether the discrepancies increase or saturate with farm size.

Load-bearing premise

The D-decomposition-related ray-extrapolation method accurately delineates the small-signal stability boundaries of the nonlinear DFIG models without introducing artifacts or missing critical behaviors under the tested parameter combinations.

What would settle it

Time-domain simulations or full eigenvalue analysis of both aggregated and granular DFIG models at boundary points in the parameter space that show identical stability status and modes would indicate the differences are not present.

Figures

Figures reproduced from arXiv: 2604.07777 by Dan Wu, Juanjuan Wang, Leyou Zhou, Meng Zhan, Mucheng Li, Xiaojie Shi.

Figure 1
Figure 1. Figure 1: Schematic diagram of DFIGs systems into the validity of small-signal stability assessment for DFIG wind farm models. II. DYNAMIC MODEL OF DFIG SYSTEM A. Wind System Configuration We consider the tree structure network which is shown in Fig.1. The turbine circuit and control diagrams of the DFIG unit are shown in Fig.2. Each unit comprises a wind turbine, drive-train, induction generator, rotor-side convert… view at source ↗
Figure 3
Figure 3. Figure 3: Validation of EMT and state-space model responses. [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Morphological evolution of the stability boundary in [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗
Figure 6
Figure 6. Figure 6: Coupling characteristics at the stability boundary in [PITH_FULL_IMAGE:figures/full_fig_p005_6.png] view at source ↗
read the original abstract

Broadband oscillations in wind farms have been widely reported in recent years. Past studies have examined various types of oscillations in wind farms, relating small-signal stability to control settings, operating conditions, and electrical parameters. However, most analyses are performed on aggregated single-unit models, which may deviate from the true behavior, leading to misleading stability assessments. To investigate how aggregation affects stability conclusions, this paper develops detailed single-, two-, and three-unit doubly-fed induction generator (DFIG) models and their aggregated counterparts. Then, a D-decomposition-related ray-extrapolation method is proposed to characterize the small-signal stability region of nonlinear DFIG models in the parameter space, delineating stability boundaries under numerous parameter combinations. The study reveals that aggregated models stability regions within the parameter planes of control settings and operating conditions differ from those of granular models in terms of basic shape, critical modes, and evolution patterns, posing a risk of misjudging stability margins.

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.

Referee Report

2 major / 2 minor

Summary. The paper constructs detailed single-, two-, and three-unit DFIG models together with their aggregated single-unit equivalents. It introduces a D-decomposition-related ray-extrapolation procedure to trace small-signal stability boundaries in planes of control gains and operating-point parameters. The central claim is that the resulting stability regions for aggregated models differ from those of the granular models in basic shape, the identity of the critical modes, and their evolution patterns, creating a risk of misjudging stability margins when aggregated models are used.

Significance. If the reported differences are shown to be free of methodological artifacts, the work would demonstrate that aggregation can change not only the size but the topology and critical-mode structure of stable operating regions in DFIG wind farms. The ray-extrapolation technique provides a practical means of mapping stability boundaries across many parameter planes without exhaustive gridding. Such results would directly affect control-tuning practices and the reliability of stability assessments based on reduced-order models.

major comments (2)
  1. [§4] §4 (ray-extrapolation procedure): The manuscript describes the D-decomposition ray-extrapolation algorithm but provides no cross-validation against dense gridding of the parameter plane or against time-domain simulation of the nonlinear model at points near the traced boundary. Without such checks, it remains possible that non-convex lobes or abrupt critical-mode switches are missed or distorted by the extrapolation step, directly undermining the claim that shape and mode differences are model-intrinsic rather than procedural.
  2. [Results section] Results section (stability-region figures): The reported boundaries for aggregated versus granular models are presented without accompanying error bounds, sensitivity to extrapolation step size, or confirmation that the same critical eigenvalue is tracked across the entire contour. These omissions are load-bearing because the headline differences in shape and critical modes rest on the fidelity of the traced contours.
minor comments (2)
  1. [Abstract] The abstract states that boundaries are delineated 'under numerous parameter combinations,' yet the main text does not specify the exact number of planes examined or the systematic selection criteria used.
  2. [Model section] Notation for the aggregated versus granular state-space matrices is introduced without an explicit side-by-side comparison table, making it harder to verify that the only structural difference is the aggregation step.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback, which has helped us improve the rigor of our validation procedures. We address each major comment below and have revised the manuscript to incorporate additional cross-checks and sensitivity analyses.

read point-by-point responses

  1. Referee: [§4] §4 (ray-extrapolation procedure): The manuscript describes the D-decomposition ray-extrapolation algorithm but provides no cross-validation against dense gridding of the parameter plane or against time-domain simulation of the nonlinear model at points near the traced boundary. Without such checks, it remains possible that non-convex lobes or abrupt critical-mode switches are missed or distorted by the extrapolation step, directly undermining the claim that shape and mode differences are model-intrinsic rather than procedural.

    Authors: The D-decomposition method locates the stability boundary exactly for the linearised system by solving for purely imaginary roots of the characteristic equation; the ray-extrapolation step is simply an efficient numerical continuation along rays in the parameter plane. Nevertheless, we agree that explicit cross-validation strengthens the claim that observed differences are model-intrinsic. In the revised §4 we now include (i) direct comparisons of ray-extrapolated contours against dense uniform gridding for representative parameter planes, confirming agreement within numerical tolerance, and (ii) time-domain simulations of the full nonlinear model at test points immediately inside and outside the traced boundaries, verifying the predicted stability transitions. These additions confirm that non-convex features or mode switches are not artifacts of the extrapolation. revision: yes

  2. Referee: [Results section] Results section (stability-region figures): The reported boundaries for aggregated versus granular models are presented without accompanying error bounds, sensitivity to extrapolation step size, or confirmation that the same critical eigenvalue is tracked across the entire contour. These omissions are load-bearing because the headline differences in shape and critical modes rest on the fidelity of the traced contours.

    Authors: We acknowledge that quantitative measures of numerical fidelity are necessary to support the reported topological and modal differences. The revised Results section now reports (i) sensitivity of the traced boundaries to extrapolation step size, together with derived error bounds, (ii) the eigenvalue-tracking algorithm employed (continuous monitoring of the eigenvalue with largest real part combined with proximity checks between successive points), and (iii) explicit verification that the same critical mode is followed along each contour. These additions demonstrate that the differences in shape and critical-mode identity between aggregated and granular models are robust to the numerical settings used. revision: yes

Circularity Check

0 steps flagged

No significant circularity; models and numerical method are independently constructed

full rationale

The paper builds granular single-, two-, and three-unit DFIG models and their aggregated counterparts from standard machine equations, then applies a proposed D-decomposition ray-extrapolation procedure to trace stability boundaries in control and operating-point planes. The central claim of differing region shapes, critical modes, and evolution patterns follows directly from comparing the outputs of this procedure on the two model classes. No equation reduces the reported differences to a fitted quantity drawn from the same data, no stability boundary is defined in terms of itself, and no load-bearing premise rests on a self-citation chain. The derivation remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

The central claim rests on the validity of the proposed ray-extrapolation method for nonlinear systems and on the assumption that the constructed DFIG models faithfully represent real-device dynamics; no new physical entities are postulated.

free parameters (1)
  • control gains and operating-point parameters
    These are swept to trace stability boundaries; specific numerical values are not reported in the abstract.
axioms (2)
  • standard math Small-signal stability can be assessed via linearization around operating points
    Standard assumption in power-system oscillation studies invoked for the DFIG models.
  • domain assumption D-decomposition combined with ray extrapolation correctly locates stability boundaries for the nonlinear DFIG system
    Core methodological premise introduced to characterize the regions.

pith-pipeline@v0.9.0 · 5478 in / 1378 out tokens · 58501 ms · 2026-05-10T18:03:11.421367+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

15 extracted references · 15 canonical work pages

  1. [1]

    Real-world subsynchronous oscillation events in power grids with high penetrations of inverter-based resources,

    Y . Cheng, L. Fan, J. Rose, S.-H. Huang, J. Schmall, X. Wang, X. Xie, J. Shair, J. R. Ramamurthy, N. Modi, C. Li, C. Wang, S. Shah, B. Pal, Z. Miao, A. Isaacs, J. Mahseredjian, and J. Zhou, “Real-world subsynchronous oscillation events in power grids with high penetrations of inverter-based resources,”IEEE Transactions on Power Systems, vol. 38, no. 1, pp...

    work page 2023

  2. [2]

    Quantitative ssr analysis of series-compensated dfig-based wind farms using aggregated rlc circuit model,

    H. Liu, X. Xie, C. Zhang, Y . Li, H. Liu, and Y . Hu, “Quantitative ssr analysis of series-compensated dfig-based wind farms using aggregated rlc circuit model,”IEEE Transactions on Power Systems, vol. 32, no. 1, pp. 474–483, 2017

    work page 2017

  3. [3]

    Investigation of ssr in practical dfig-based wind farms connected to a series-compensated power system,

    L. Wang, X. Xie, Q. Jiang, H. Liu, Y . Li, and H. Liu, “Investigation of ssr in practical dfig-based wind farms connected to a series-compensated power system,”IEEE Transactions on Power Systems, vol. 30, no. 5, pp. 2772–2779, 2015

    work page 2015

  4. [4]

    Modeling of dfig-based wind farms for ssr analysis,

    L. Fan, R. Kavasseri, Z. L. Miao, and C. Zhu, “Modeling of dfig-based wind farms for ssr analysis,”IEEE Transactions on Power Delivery, vol. 25, no. 4, pp. 2073–2082, 2010

    work page 2073

  5. [5]

    Stability analysis for dfig-based wind farm grid-connected system under all wind speed conditions,

    C. Du, X. Du, C. Tong, Y . Li, and P. Zhou, “Stability analysis for dfig-based wind farm grid-connected system under all wind speed conditions,”IEEE Transactions on Industry Applications, vol. 59, no. 2, pp. 2430–2445, 2023

    work page 2023

  6. [6]

    Small-signal stability analysis of type-4 wind in series-compensated networks,

    Y . Xu, M. Zhang, L. Fan, and Z. Miao, “Small-signal stability analysis of type-4 wind in series-compensated networks,”IEEE Transactions on Energy Conversion, vol. 35, no. 1, pp. 529–538, 2020

    work page 2020

  7. [7]

    High-frequency resonance damp- ing of dfig-based wind power system under weak network,

    Y . Song, X. Wang, and F. Blaabjerg, “High-frequency resonance damp- ing of dfig-based wind power system under weak network,”IEEE Transactions on Power Electronics, vol. 32, no. 3, pp. 1927–1940, 2017

    work page 1927

  8. [8]

    Impedance-based high-frequency resonance analysis of dfig system in weak grids,

    ——, “Impedance-based high-frequency resonance analysis of dfig system in weak grids,”IEEE Transactions on Power Electronics, vol. 32, no. 5, pp. 3536–3548, 2017

    work page 2017

  9. [9]

    Participation factors weak robustness analysis and miti- gation of direct-drive wind farms connected to the parallel-compensated grid,

    B. Shao, Z. Miao, L. Wang, W. Ma, X. Meng, Q. Xiao, Z. Chen, and F. Blaabjerg, “Participation factors weak robustness analysis and miti- gation of direct-drive wind farms connected to the parallel-compensated grid,”IEEE Transactions on Energy Conversion, vol. 38, no. 3, pp. 1924–1936, 2023

    work page 1924

  10. [10]

    Overview of dfig-based wind power system resonances under weak networks,

    Y . Song and F. Blaabjerg, “Overview of dfig-based wind power system resonances under weak networks,”IEEE Transactions on Power Elec- tronics, vol. 32, no. 6, pp. 4370–4394, 2017

    work page 2017

  11. [11]

    The adequacy of the present practice in dynamic aggregated modeling of wind farm systems,

    L. P. Kunjumuhammed, B. C. Pal, C. Oates, and K. J. Dyke, “The adequacy of the present practice in dynamic aggregated modeling of wind farm systems,”IEEE Transactions on Sustainable Energy, vol. 8, no. 1, pp. 23–32, 2016

    work page 2016

  12. [12]

    Impedance analysis and phil demonstration of reactive power oscillations in a wind power plant using a 4-mw wind turbine,

    P. Koralewicz, S. Shah, V . Gevorgian, R. Wallen, K. Jha, D. Mashtare, K. V . R. Gadiraju, and A. Tiwari, “Impedance analysis and phil demonstration of reactive power oscillations in a wind power plant using a 4-mw wind turbine,”Frontiers in Energy Research, vol. 8, p. 156, 2020

    work page 2020

  13. [13]

    An equivalent model for sub-synchronous oscillation analysis in direct-drive wind farms with vsc-hvdc systems,

    B. Shao, S. Zhao, B. Gao, Y . Yang, and F. Blaabjerg, “An equivalent model for sub-synchronous oscillation analysis in direct-drive wind farms with vsc-hvdc systems,”International Journal of Electrical Power & Energy Systems, vol. 125, p. 106498, 2021

    work page 2021

  14. [14]

    Search for the parameter values that make automatic control system stable,

    Y . I. Neimark, “Search for the parameter values that make automatic control system stable,”Avtom. Telemekh, vol. 9, no. 3, pp. 190–203, 1948

    work page 1948

  15. [15]

    Adaptive passivity- based control of forced subsynchronous oscillation in dfig-based wind farms,

    D. Sun, Z. Qian, F. Meng, Y . Lv, and Q. Chen, “Adaptive passivity- based control of forced subsynchronous oscillation in dfig-based wind farms,”IEEE Transactions on Sustainable Energy, vol. 16, no. 2, pp. 904–918, 2025

    work page 2025