# Yuan, Watson and Watson find model complexity may not improve ROA fidelity [Canonical reading](https://escapemeridian.com/readings/yuan-watson-and-watson-find-model-complexity-may-not-improve-d9776fa7) Chenxi Yuan, Jeremy D. Watson and Neville R. Watson examine whether reduced-order models preserve the region of attraction (ROA)—the set of states from which a system returns to equilibrium. They study its use in certifying transient stability in grid-forming (GFM) and grid-following (GFL) converters. Their abstract says Sum-of-squares (SOS) programming can produce certified inner approximations of the ROA, but its computational cost grows rapidly with system order. Reduced-order models can lower that cost and enable analytical investigation, yet matching transient response or retaining more dynamics does not by itself ensure that the reduced model preserves the full-order system’s ROA. The authors compare SOS-based ROA estimates with time-domain ODE simulations, separating conservativeness introduced by SOS from distortion associated with model order. In the cases examined, more complex reduced-order models sometimes have closer agreement with the simulated ODE boundary in some regions, while achieving lower SOS coverage of the full-order reference ROA. The authors therefore argue that model selection should consider full-order ROA fidelity and certified coverage of the practical operating range, not model complexity alone. The abstract does not establish how broadly these results generalize beyond the cases examined. 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 arXiv eess.SY · Article Original source date: Nov 10, 2026 (2026-11-10). Date basis: Date basis not provided. Added to Escape Meridian: 2026-11-11T19:36:42.653Z. First observed: 2026-11-11T05:04:41.192Z. Retrieved: 2026-11-11T19:35:23.207Z. - [Impact of Reduced-Order Modelling on Sum-of-Squares Region-of-Attraction Certification for Grid-Forming and Grid-Following Converters](https://arxiv.org/abs/2611.13497) ## Limits This reading is based on the arXiv abstract, which describes results in the cases examined but does not detail those cases or quantify the reported coverage differences. The abstract does not establish how broadly the reported results apply beyond the cases examined. ## Open question How do the reported results vary across cases beyond those examined in the paper? ## Source-stated The authors compare SOS-based ROA estimates for reduced-order GFM and GFL models with simulations of both those models and their corresponding full-order models. > the SOS-based ROA of each reduced-order GFM and GFL model is assessed against both its own simulated ROA and that of the corresponding full-order model - [Supporting source](https://arxiv.org/abs/2611.13497) ## Source-stated In the cases examined, increasing reduced-order model complexity did not necessarily improve SOS-based ROA fidelity; more complex models had lower SOS coverage of the full-order reference ROA. > In the cases examined, more complex reduced-order models yield lower SOS coverage of the full-order reference ROA - [Supporting source](https://arxiv.org/abs/2611.13497) ## Source-stated The authors recommend selecting models by considering full-order ROA fidelity and certified coverage of the practical operating range, rather than complexity alone. > Model selection should therefore consider both full-order ROA fidelity and certified coverage of the practical operating range, rather than model complexity alone. - [Supporting source](https://arxiv.org/abs/2611.13497) ## Source and contributor context Yuan, Chenxi, Watson, Jeremy D., Watson, Neville R. · 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. - [Explore this context](/entities/arxiv) - [Reference](https://arxiv.org/) - [Reference](https://info.arxiv.org/help/api/index.html) ## TIME TIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services. - [Explore this context](/entities/time) - [Reference](https://time.com/) ## Editorial provenance Version 1. Publication ID: fe3f8eab-f18a-4a2b-8288-52b3f7fb3bce. Prepared by gpt-6-luna; grounding review by gpt-6-luna. Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-11T19:36:42.653Z. This checks fidelity to the source, not whether its claims are independently true. Editorial selection: accept; reasons: material\_development. Observation: 9b4bebfafc96798bbb6699776727810ba143cfa751aa2489325f3024d9f93ae4. Source text hash: 1ea21a0a5910c81385505766c04715f2b1b7259660440ee8aa7ddbf76d65e3b9. ## Revision and source-check history ### Version 1 — 2026-11-11T19:36:42.653Z Chenxi Yuan, Jeremy D. Watson and Neville R. Watson examine whether reduced-order models preserve the region of attraction (ROA)—the set of states from which a system returns to equilibrium. They study its use in certifying transient stability in grid-forming (GFM) and grid-following (GFL) converters. Their abstract says Sum-of-squares (SOS) programming can produce certified inner approximations of the ROA, but its computational cost grows rapidly with system order. Reduced-order models can lower that cost and enable analytical investigation, yet matching transient response or retaining more dynamics does not by itself ensure that the reduced model preserves the full-order system’s ROA. The authors compare SOS-based ROA estimates with time-domain ODE simulations, separating conservativeness introduced by SOS from distortion associated with model order. In the cases examined, more complex reduced-order models sometimes have closer agreement with the simulated ODE boundary in some regions, while achieving lower SOS coverage of the full-order reference ROA. The authors therefore argue that model selection should consider full-order ROA fidelity and certified coverage of the practical operating range, not model complexity alone. The abstract does not establish how broadly these results generalize beyond the cases examined. ## Original source arXiv eess.SY · Article Original source date: Nov 10, 2026 (2026-11-10). Date basis: Date basis not provided. Added to Escape Meridian: 2026-11-11T19:36:42.653Z. First observed: 2026-11-11T05:04:41.192Z. Retrieved: 2026-11-11T19:35:23.207Z. - [Impact of Reduced-Order Modelling on Sum-of-Squares Region-of-Attraction Certification for Grid-Forming and Grid-Following Converters](https://arxiv.org/abs/2611.13497) ## Limits This reading is based on the arXiv abstract, which describes results in the cases examined but does not detail those cases or quantify the reported coverage differences. The abstract does not establish how broadly the reported results apply beyond the cases examined. ## Open question How do the reported results vary across cases beyond those examined in the paper? ## Source-stated The authors compare SOS-based ROA estimates for reduced-order GFM and GFL models with simulations of both those models and their corresponding full-order models. > the SOS-based ROA of each reduced-order GFM and GFL model is assessed against both its own simulated ROA and that of the corresponding full-order model - [Supporting source](https://arxiv.org/abs/2611.13497) ## Source-stated In the cases examined, increasing reduced-order model complexity did not necessarily improve SOS-based ROA fidelity; more complex models had lower SOS coverage of the full-order reference ROA. > In the cases examined, more complex reduced-order models yield lower SOS coverage of the full-order reference ROA - [Supporting source](https://arxiv.org/abs/2611.13497) ## Source-stated The authors recommend selecting models by considering full-order ROA fidelity and certified coverage of the practical operating range, rather than complexity alone. > Model selection should therefore consider both full-order ROA fidelity and certified coverage of the practical operating range, rather than model complexity alone. - [Supporting source](https://arxiv.org/abs/2611.13497) ## Source and contributor context Yuan, Chenxi, Watson, Jeremy D., Watson, Neville R. · 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. - [Explore this context](/entities/arxiv) - [Reference](https://arxiv.org/) - [Reference](https://info.arxiv.org/help/api/index.html) ## TIME TIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services. - [Explore this context](/entities/time) - [Reference](https://time.com/) ## Editorial provenance Version 1. Publication ID: fe3f8eab-f18a-4a2b-8288-52b3f7fb3bce. Prepared by gpt-6-luna; grounding review by gpt-6-luna. Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-11T19:36:42.653Z. This checks fidelity to the source, not whether its claims are independently true. Editorial selection: accept; reasons: material\_development. Observation: 9b4bebfafc96798bbb6699776727810ba143cfa751aa2489325f3024d9f93ae4. Source text hash: 1ea21a0a5910c81385505766c04715f2b1b7259660440ee8aa7ddbf76d65e3b9. ### uncertain — 2026-11-11T19:36:42.653Z — anthropic/claude-opus-5.5 Publication: fe3f8eab-f18a-4a2b-8288-52b3f7fb3bce attribution — pass: The reading correctly credits Chenxi Yuan, Jeremy D. Watson and Neville R. Watson. It presents the findings and the model-selection recommendation as the authors' abstract claims, not as established facts. > Authors:Chenxi Yuan, Jeremy D. Watson, Neville R. Watson summary — uncertain: The summary is faithful to the abstract, and it usefully defines ROA. A non-specialist would still struggle with the central terms. Grid-forming and grid-following converters, reduced-order models, transient stability, SOS programming, system order, conservativeness and ODE simulations are named but never explained. The reading also never says plainly why this matters: power-electronic equipment such as converters is increasingly common on grids, and stability certificates depend on the model chosen. Smallest repair: add short plain-language glosses or term-library tooltips for these terms, plus one sentence of grid context. The question is acceptable but generic. > With the increasing penetration of power-electronic equipment, the transient stability of grid-forming (GFM) and grid-following (GFL) converters is increasingly characterized through the region of attraction (ROA) of the post-fault equilibrium. date — pass: The originalPublished date of 2026-11-10 matches the v1 arXiv submission date. The feed date of 2026-11-11 is correctly not used. > \[Submitted on 10 Nov 2026\] support — pass: Each claim matches the abstract. The draft keeps the qualifiers 'in the cases examined' and 'in some regions', and its limitations correctly note that the cases are not detailed and generalization is unknown. It adds no stakeholder benefits or costs beyond the source. > The results indicate that increasing reduced-order model complexity does not necessarily improve SOS-based ROA fidelity. media — pass: The selected media is the verified direct link to the original arXiv abstract page. That is the most direct way for readers to reach the source. ## Recorded references Explicit citations in public readings and public file attachments to this reading. Shared topics are not citations. Other artifact types need an explicit reference adapter. No public references recorded here yet. [Reference data (JSON)](/readings/d9776fa7fc0134151016bf69d4bf280466a5acd271abf6c18ff6e7bbe72f823c/references.json) ## Corrections and challenges [Send a correction or an opposing source](https://escapemeridian.com/corrections/) [JSON record](https://escapemeridian.com/readings/yuan-watson-and-watson-find-model-complexity-may-not-improve-d9776fa7.json)