# Transformer-guided Variable Projection for Phasor and Fault-Time Estimation Under DDC Offset [Canonical reading](https://escapemeridian.com/readings/transformer-guided-variable-projection-for-phasor-and-fault-0b92c96f) The paper’s authors propose a transformer-guided variable-projection estimator for measuring electrical signals after a fault, when decaying DC (DDC) offsets can distort phasor estimates. An offline-trained transformer supplies a signal structure and decay rate; variable projection then refines the parameters and estimates fault inception time, including in windows that contain the fault without knowing when it occurred. The authors report lower mean maximum overshoot than the discrete Fourier transform (DFT) in their tested scenarios: at most 2.7% on post-fault windows and 2.2% during online operation, versus 10–14% for the DFT. They also report that the post-fault phasor settles within 5% approximately 12.5 ms after inception, and that circuit-level validation on a 230-kV two-source system yielded 1.6% mean maximum overshoot, versus 9.5% for the DFT. These results suggest a way to combine a learned signal representation with parameter refinement to estimate fault-related quantities quickly, but the abstract reports results from specified scenarios and one circuit-level system; it does not establish performance in other settings. 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 2, 2026 (2026-11-02). Date basis: Date basis not provided. Added to Escape Meridian: 2026-11-10T07:15:55.903Z. First observed: 2026-11-03T08:51:38.777Z. Retrieved: 2026-11-03T17:02:11.558Z. - [Transformer-guided Variable Projection for Phasor and Fault-Time Estimation Under DDC Offset](https://arxiv.org/abs/2611.03416) ## Limits The evidence provided is the arXiv abstract, which reports results for specified single- and dual-DDC scenarios and circuit-level validation on a 230-kV two-source system; it does not provide the paper’s full methods or results. ## Open question How does the estimator perform in settings beyond the single 230-kV two-source system described in the abstract? ## Source-stated The authors propose an estimator that uses an offline-trained transformer to provide signal structure and decay rate, then uses variable projection to refine parameters and estimate fault inception time. > An offline-trained transformer provides the signal structure and decay rate, while variable projection refines the parameters - [Supporting source](https://arxiv.org/abs/2611.03416) ## Source-stated The authors report mean maximum overshoot no greater than 2.7% on post-fault windows and 2.2% during online operation, compared with 10–14% for the DFT. > the mean maximum overshoot (MO) is at most 2.7% on post-fault windows and 2.2% during online operation, where the DFT reaches 10--14%. - [Supporting source](https://arxiv.org/abs/2611.03416) ## Source-stated In circuit-level validation on a 230-kV two-source system, the authors report 1.6% mean maximum overshoot, compared with 9.5% for the DFT. > Circuit-level validation on a 230-kV two-source system yields 1.6% mean MO, compared with 9.5% for the DFT. - [Supporting source](https://arxiv.org/abs/2611.03416) ## Escape Meridian interpretation The abstract’s reported circuit-level validation is specific to one 230-kV two-source system; whether the results extend to other settings remains an open question. > Circuit-level validation on a 230-kV two-source system yields 1.6% mean MO - [Supporting source](https://arxiv.org/abs/2611.03416) ## Source and contributor context Wada, Marcus Chen I, Mohammadi, Sina, Bui, Van-Hai, Su, Wencong · 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. ## 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/) ## 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) ## Editorial provenance Version 1. Publication ID: 3d9a3979-a913-4505-803e-f515c7028570. Prepared by gpt-6-luna; grounding review by gpt-6-luna. Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-10T07:15:55.903Z. This checks fidelity to the source, not whether its claims are independently true. Editorial selection: accept; reasons: material\_development. Observation: a458a82849291bc02ea6de43e24c5d48fa610f571b007abb566a2d374f8ad75b. Source text hash: cf4d693c434d0fe302592cf0a2a8b9ce905430287b2e2db88b80b89644562731. ## Revision and source-check history ### Version 1 — 2026-11-10T07:15:55.903Z The paper’s authors propose a transformer-guided variable-projection estimator for measuring electrical signals after a fault, when decaying DC (DDC) offsets can distort phasor estimates. An offline-trained transformer supplies a signal structure and decay rate; variable projection then refines the parameters and estimates fault inception time, including in windows that contain the fault without knowing when it occurred. The authors report lower mean maximum overshoot than the discrete Fourier transform (DFT) in their tested scenarios: at most 2.7% on post-fault windows and 2.2% during online operation, versus 10–14% for the DFT. They also report that the post-fault phasor settles within 5% approximately 12.5 ms after inception, and that circuit-level validation on a 230-kV two-source system yielded 1.6% mean maximum overshoot, versus 9.5% for the DFT. These results suggest a way to combine a learned signal representation with parameter refinement to estimate fault-related quantities quickly, but the abstract reports results from specified scenarios and one circuit-level system; it does not establish performance in other settings. ## Original source arXiv eess.SY · Article Original source date: Nov 2, 2026 (2026-11-02). Date basis: Date basis not provided. Added to Escape Meridian: 2026-11-10T07:15:55.903Z. First observed: 2026-11-03T08:51:38.777Z. Retrieved: 2026-11-03T17:02:11.558Z. - [Transformer-guided Variable Projection for Phasor and Fault-Time Estimation Under DDC Offset](https://arxiv.org/abs/2611.03416) ## Limits The evidence provided is the arXiv abstract, which reports results for specified single- and dual-DDC scenarios and circuit-level validation on a 230-kV two-source system; it does not provide the paper’s full methods or results. ## Open question How does the estimator perform in settings beyond the single 230-kV two-source system described in the abstract? ## Source-stated The authors propose an estimator that uses an offline-trained transformer to provide signal structure and decay rate, then uses variable projection to refine parameters and estimate fault inception time. > An offline-trained transformer provides the signal structure and decay rate, while variable projection refines the parameters - [Supporting source](https://arxiv.org/abs/2611.03416) ## Source-stated The authors report mean maximum overshoot no greater than 2.7% on post-fault windows and 2.2% during online operation, compared with 10–14% for the DFT. > the mean maximum overshoot (MO) is at most 2.7% on post-fault windows and 2.2% during online operation, where the DFT reaches 10--14%. - [Supporting source](https://arxiv.org/abs/2611.03416) ## Source-stated In circuit-level validation on a 230-kV two-source system, the authors report 1.6% mean maximum overshoot, compared with 9.5% for the DFT. > Circuit-level validation on a 230-kV two-source system yields 1.6% mean MO, compared with 9.5% for the DFT. - [Supporting source](https://arxiv.org/abs/2611.03416) ## Escape Meridian interpretation The abstract’s reported circuit-level validation is specific to one 230-kV two-source system; whether the results extend to other settings remains an open question. > Circuit-level validation on a 230-kV two-source system yields 1.6% mean MO - [Supporting source](https://arxiv.org/abs/2611.03416) ## Source and contributor context Wada, Marcus Chen I, Mohammadi, Sina, Bui, Van-Hai, Su, Wencong · 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. ## 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/) ## 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) ## Editorial provenance Version 1. Publication ID: 3d9a3979-a913-4505-803e-f515c7028570. Prepared by gpt-6-luna; grounding review by gpt-6-luna. Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-10T07:15:55.903Z. This checks fidelity to the source, not whether its claims are independently true. Editorial selection: accept; reasons: material\_development. Observation: a458a82849291bc02ea6de43e24c5d48fa610f571b007abb566a2d374f8ad75b. Source text hash: cf4d693c434d0fe302592cf0a2a8b9ce905430287b2e2db88b80b89644562731. ### uncertain — 2026-11-10T07:15:55.903Z — anthropic/claude-opus-5.5 Publication: 3d9a3979-a913-4505-803e-f515c7028570 attribution — pass: The proposal, results and comparisons are attributed to the paper's authors and framed as reported findings. The arXiv page lists the authors and the method matches the abstract. > Authors:Marcus Chen I Wada, Sina Mohammadi, Van-Hai Bui, Wencong Su summary — uncertain: The numbers are faithful, but a non-specialist cannot follow the central terms. Phasor, transformer, variable projection, maximum overshoot and fault inception time are not explained. The reading also never says this concerns power-system faults, so a reader may not see why first-cycle errors matter. It drops the paper's stated motive: an interpretable analytical expression without the high computational cost of earlier approaches. Smallest repairs: add term-library tooltips or one plain-language framing sentence, restore the interpretability and cost point, and label the 'These results suggest' sentence as Escape Meridian interpretation. > Recent approaches have sought interpretable waveform representations, but identifying analytical expressions can incur substantial computational costs. date — pass: The originalPublished date of 2026-11-02 matches the v1 arXiv submission date. The feed date is later and is correctly not used. > \[Submitted on 2 Nov 2026\] support — pass: The reported figures, the handling of windows that contain the fault without knowing the fault time, and the 12.5 ms settling time all match the abstract. The limitation and the open question about other settings are reasonable and stay within the source's scope. The phrase 'quickly' is loosely supported by the 12.5 ms settling figure. > variable projection refines the parameters and handles windows containing the fault without knowledge of time of fault media — pass: The selected media is the verified direct arXiv abstract page for this paper. ## 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/0b92c96f91ac11b95eeb56a29bad9a134d20fbe490e01bd07f9b4272ed069c48/references.json) ## Corrections and challenges [Send a correction or an opposing source](https://escapemeridian.com/corrections/) [JSON record](https://escapemeridian.com/readings/transformer-guided-variable-projection-for-phasor-and-fault-0b92c96f.json)