Kumar, Sunny, Parimi, Madhavi, Varma, Shashank, Wagh, Sushama, Stanković, Alexsandar, Singh, Navdeep · Oct 21, 2026
arXiv authors explain why the swing parameters are sloppy
Markdown · JSON · Original source
Sunny Kumar, Madhavi Parimi and their coauthors argue that inertia H and damping D are hard to identify from a synchronous generator’s late post-disturbance response because the rotor swing has already collapsed onto a two-dimensional normally attracting invariant manifold (NAIM). In that late window, they report, six of eleven parameters suffice to describe the response; when they evaluate it using a physical metric fixed by synchrophasor accuracy limits, five parameters are identifiable. Their geometric explanation is that the zeroth-order NAIM is independent of H, D, T_{d0}'' and T_{q0}''. The authors say neither a larger initial speed deviation nor a load change restores H or D in the late window, while the early window makes H the best-determined parameter. This makes the choice of observation window central to what parameter values the model can recover: a parameter difficult to estimate from late data may be better determined from early data. The abstract also reports that regression on principal components of whitened, sampled trajectories reached the Fisher-predicted recovery error in 113 of 114 parameter scenario pairs. The authors’ results concern this model and these analyses; their proof is stated to hold under smoothness and rank conditions.
Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims.
Original source
eess.SY · Article
Source details
Original source date: Oct 21, 2026 (2026-10-21).
Date basis: Date basis not provided.
Added to Escape Meridian: 2026-11-11T01:09:35.059Z. First observed: 2026-10-22T04:08:09.932Z. Retrieved: 2026-11-02T08:32:29.669Z.
Referenced by
Recorded citations and public file attachments. A reference alone does not establish agreement or influence.
No public references recorded here yet.
Limits
The results described in the abstract concern an infinite-bus synchronous generator and specified late and early post-disturbance windows; the abstract does not establish that they generalize to other models or experiments. The abstract refers to smoothness and rank conditions but does not specify them in the supplied text. The reported parameter identifiability and recovery results are model analyses; the abstract does not establish real-world reliability of parameter estimates.
Open question
What smoothness and rank conditions are required for the four limits to be boundaries of the model manifold?
Source-stated
The authors attribute the weak identifiability of inertia H and damping D in the late post-disturbance window to the rotor swing having collapsed onto a two-dimensional normally attracting invariant manifold (NAIM).
The rotor swing is a fast fibre that has collapsed onto a two-dimensional normally attracting invariant manifold (NAIM) of the transient fluxes before the window opens.
Source-stated
For the late post-disturbance window, the authors report that six of eleven parameters suffice to describe the response, while the physical metric they use yields five identifiable parameters.
where six of the eleven parameters suffice to describe the response.
Source-stated
The authors say the zeroth-order NAIM is independent of H, D, T_{d0}'' and T_{q0}'', which they identify as four of the five MBAM limits.
The zeroth-order NAIM is exactly independent of $H$, $D$, $T_{d0}''$ and $T_{q0}''$, four of the five MBAM limits.
Source-stated
The authors report that neither a larger initial speed deviation nor a load change restores H or D in the late window, whereas H is best determined in the early window.
neither a larger initial speed deviation nor a load change restores H or D in the late window, whereas the early window [0, 2]s makes H the best-determined parameter.
Source-stated
The authors report that regression on principal components of whitened, sampled trajectories attains the Fisher-predicted recovery error in 113 of 114 parameter scenario pairs.
regression on those components attains the Fisher-predicted recovery error in 113 of 114 parameter scenario pairs.
Source-stated
The authors state that their proof about the four limits depends on smoothness and rank conditions.
We prove, under stated smoothness and rank conditions, that each of the four limits is a boundary of the model manifold
Source and contributor context
Kumar, Sunny, Parimi, Madhavi, Varma, Shashank, Wagh, Sushama, Stanković, Alexsandar, Singh, Navdeep · Article author
These are credits stated by the source; they do not establish a profile match or authorship of every passage.
Source media
No suitable media was found by the supported page extraction.
Terms and references
Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.
arXiv
arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.
Editorial provenance
Version 1. Publication ID: 60675a57-56e1-4e20-8e29-306577b27459.
Prepared by gpt-6-luna; grounding review by gpt-6-luna.
Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-11T01:09:35.059Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: material_development.
Observation: 4cf4289889984ba8a269a35ee76bcda819cf9c2b3472ffa795d300b07b8fd1c8. Source text hash: 8c35f8f0bff9b38349e467c432f618cb0a6905cc5b1a22e12626845db83507d0.
Revision and source-check history
Version 1 · 2026-11-11T01:09:35.059Z
arXiv authors explain why the swing parameters are sloppy
Sunny Kumar, Madhavi Parimi and their coauthors argue that inertia H and damping D are hard to identify from a synchronous generator’s late post-disturbance response because the rotor swing has already collapsed onto a two-dimensional normally attracting invariant manifold (NAIM). In that late window, they report, six of eleven parameters suffice to describe the response; when they evaluate it using a physical metric fixed by synchrophasor accuracy limits, five parameters are identifiable. Their geometric explanation is that the zeroth-order NAIM is independent of H, D, T_{d0}'' and T_{q0}''. The authors say neither a larger initial speed deviation nor a load change restores H or D in the late window, while the early window makes H the best-determined parameter. This makes the choice of observation window central to what parameter values the model can recover: a parameter difficult to estimate from late data may be better determined from early data. The abstract also reports that regression on principal components of whitened, sampled trajectories reached the Fisher-predicted recovery error in 113 of 114 parameter scenario pairs. The authors’ results concern this model and these analyses; their proof is stated to hold under smoothness and rank conditions.
Original source
eess.SY · Article
Source details
Original source date: Oct 21, 2026 (2026-10-21).
Date basis: Date basis not provided.
Added to Escape Meridian: 2026-11-11T01:09:35.059Z. First observed: 2026-10-22T04:08:09.932Z. Retrieved: 2026-11-02T08:32:29.669Z.
Limits
The results described in the abstract concern an infinite-bus synchronous generator and specified late and early post-disturbance windows; the abstract does not establish that they generalize to other models or experiments. The abstract refers to smoothness and rank conditions but does not specify them in the supplied text. The reported parameter identifiability and recovery results are model analyses; the abstract does not establish real-world reliability of parameter estimates.
Open question
What smoothness and rank conditions are required for the four limits to be boundaries of the model manifold?
Source-stated
The authors attribute the weak identifiability of inertia H and damping D in the late post-disturbance window to the rotor swing having collapsed onto a two-dimensional normally attracting invariant manifold (NAIM).
The rotor swing is a fast fibre that has collapsed onto a two-dimensional normally attracting invariant manifold (NAIM) of the transient fluxes before the window opens.
Source-stated
For the late post-disturbance window, the authors report that six of eleven parameters suffice to describe the response, while the physical metric they use yields five identifiable parameters.
where six of the eleven parameters suffice to describe the response.
Source-stated
The authors say the zeroth-order NAIM is independent of H, D, T_{d0}'' and T_{q0}'', which they identify as four of the five MBAM limits.
The zeroth-order NAIM is exactly independent of $H$, $D$, $T_{d0}''$ and $T_{q0}''$, four of the five MBAM limits.
Source-stated
The authors report that neither a larger initial speed deviation nor a load change restores H or D in the late window, whereas H is best determined in the early window.
neither a larger initial speed deviation nor a load change restores H or D in the late window, whereas the early window [0, 2]s makes H the best-determined parameter.
Source-stated
The authors report that regression on principal components of whitened, sampled trajectories attains the Fisher-predicted recovery error in 113 of 114 parameter scenario pairs.
regression on those components attains the Fisher-predicted recovery error in 113 of 114 parameter scenario pairs.
Source-stated
The authors state that their proof about the four limits depends on smoothness and rank conditions.
We prove, under stated smoothness and rank conditions, that each of the four limits is a boundary of the model manifold
Source and contributor context
Kumar, Sunny, Parimi, Madhavi, Varma, Shashank, Wagh, Sushama, Stanković, Alexsandar, Singh, Navdeep · Article author
These are credits stated by the source; they do not establish a profile match or authorship of every passage.
Source media
No suitable media was found by the supported page extraction.
Terms and references
Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.
arXiv
arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.
Editorial provenance
Version 1. Publication ID: 60675a57-56e1-4e20-8e29-306577b27459.
Prepared by gpt-6-luna; grounding review by gpt-6-luna.
Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-11T01:09:35.059Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: material_development.
Observation: 4cf4289889984ba8a269a35ee76bcda819cf9c2b3472ffa795d300b07b8fd1c8. Source text hash: 8c35f8f0bff9b38349e467c432f618cb0a6905cc5b1a22e12626845db83507d0.
uncertain · 2026-11-11T01:09:35.059Z · anthropic/claude-opus-5.5
Publication: 60675a57-56e1-4e20-8e29-306577b27459
attribution — pass: The reading attributes the claims to Sunny Kumar, Madhavi Parimi and coauthors, matching the arXiv author list. It frames findings as the authors' own reports and arguments rather than as established fact.
Authors:Sunny Kumar, Madhavi Parimi, Shashank Varma, Sushama Wagh, Alexsandar Stanković, Navdeep Singh
summary — uncertain: The summary follows the abstract accurately, but a non-specialist cannot follow it. It never explains what a synchronous generator, inertia H, damping D, a post-disturbance window or parameter identifiability are. It also leaves NAIM, zeroth-order, T_{d0}'', MBAM limits, synchrophasor, Fisher and whitened principal components unexplained. It omits the infinite-bus model setting from the body, and the title's sloppy is unexplained. Smallest repair: add one plain sentence on fitting power-generator models to measured data and why recovering parameters matters, plus term-library tooltips for the central terms.
For an infinite-bus synchronous generator, both approaches show that inertia H and damping D are among the least identifiable parameters
date — pass: The originalPublished date of 2026-10-21 matches the v1 submission date. Later revisions, including the withdrawn v2 and v4, and the November feed date are correctly not used.
[Submitted on 21 Oct 2026 (v1), last revised 30 Oct 2026 (this version, v4)]
support — pass: Every claim matches the abstract, including the six-versus-five parameter counts, the four NAIM-independent parameters, the late-versus-early window results, 113 of 114 and the smoothness and rank conditions. The observation-window inference is hedged and grounded in the abstract's own experiment-dependence statement. The limitations correctly avoid claiming real-world reliability or generalization.
The same analysis predicts how identifiability changes with the experiment
media — pass: The selected media is the verified direct arXiv abstract page for this paper. It is the most direct way to access the source.
Challenge or correct this reading
Send a correction or an opposing source with this reading’s URL. Disagreement and uncertainty can remain unresolved.
arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.
Explore arXivarXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.
Explore arXiv