Wang, Jinsong, Loparo, Kenneth A. · Mar 6, 2023
Jinsong Wang and Kenneth A. Loparo on Wind Turbine Gearbox Fault Detection
Markdown · JSON · Original source
Jinsong Wang and Kenneth A. Loparo investigate a two-stage method for detecting wind turbine gearbox faults from vibration measurements. Sparse filtering learns compact features from the measurements; a knowledge-based graph links those features to gearbox components, sensors, and health states. The authors compare graph models, including one that produces structured sequences linking predicted states with component and sensor information. In experiments on a National Renewable Energy Laboratory gearbox , their sparse-filtering gated graph neural network reports higher classification accuracy than its manually engineered feature counterpart. model also reports results for one- and two-fault-signal cases. The authors say these results support using learned vibration representations within a structured sensor graph. Their evidence is limited to the evaluated benchmark and operating condition; independent component localization, transfer across turbines, and deployment still require validation. The work therefore presents a benchmark result, not evidence that the method will perform similarly across turbines or in deployment.
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: Mar 6, 2023 (2023-03-06).
Date basis: Date basis not provided.
Added to Escape Meridian: 2026-11-10T09:10:59.648Z. First observed: 2026-11-03T08:51:18.951Z. Retrieved: 2026-11-03T19:05:43.760Z.
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 authors say the evidence is restricted to the evaluated and operating condition; independent component localization, transfer across turbines, and deployment require further validation. The abstract does not report results from independent tests across turbines or deployment latency.
Open question
How does the method perform when independently tested for component localization, transfer across turbines, and deployment ?
Source-stated
The authors investigate a two-stage approach combining sparse filtering and graph neural networks to detect wind turbine gearbox faults from vibration measurements.
We investigate a two-stage framework that combines sparse filtering with graph neural networks for wind turbine gearbox fault detection.
Source-stated
The knowledge-based graph connects learned features with gearbox components, sensors, and health-state entities.
a knowledge-based graph associates these features with gearbox components, sensors, and health-state entities.
Source-stated
On the stated benchmark, the sparse-filtering gated graph neural network with 100 learned features reports 93.74% classification accuracy, compared with 89.28% for its manually engineered feature counterpart.
With 100 learned features, the sparse-filtering gated graph neural network achieves a reported classification accuracy of 93.74%, compared with 89.28% for its manually engineered feature counterpart.
Source-stated
The sequence model reports accuracies of 90.73% and 90.36% in the one- and two-fault-signal cases, respectively.
The sequence model achieves reported accuracies of 90.73% and 90.36% in the one- and two-fault-signal cases, respectively.
Source-stated
The authors restrict their evidence to the evaluated benchmark and operating condition, and identify further validation needs.
The evidence is restricted to the evaluated benchmark and operating condition; independent component localization, transfer across turbines, and deployment latency require further validation.
Source and contributor context
Wang, Jinsong, Loparo, Kenneth A. · 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.
Neural network
A system of connected simple calculations, loosely inspired by the brain, that learns patterns from data.
Benchmark
A defined set of tests used to compare systems. Results depend on the tasks and measures selected.
Latency
The delay between sending a request and receiving the result.
arXiv
arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.
The Sequence
The Sequence publishes machine-learning analysis and interviews whose guest answers retain their own attribution.
Editorial provenance
Version 1. Publication ID: 70fb1b4c-f8c5-4f29-8cf5-5bb767b3631d.
Prepared by gpt-6-luna; grounding review by gpt-6-luna.
Source check: passed by anthropic/claude-opus-5.5 on 2026-11-10T09:10:59.648Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: material_development.
Observation: 15f99d6bae29a249a71123f418aa0b6f7c9b90d18bc92e2403e0c0fecd793ea0. Source text hash: 7b0b5d577020a6f31f8ce80bd32b7abc36b2afb068f97083fd99cf52a8087f64.
Revision and source-check history
Version 1 · 2026-11-10T09:10:59.648Z
Jinsong Wang and Kenneth A. Loparo on Wind Turbine Gearbox Fault Detection
Jinsong Wang and Kenneth A. Loparo investigate a two-stage method for detecting wind turbine gearbox faults from vibration measurements. Sparse filtering learns compact features from the measurements; a knowledge-based graph links those features to gearbox components, sensors, and health states. The authors compare graph models, including one that produces structured sequences linking predicted states with component and sensor information. In experiments on a National Renewable Energy Laboratory gearbox , their sparse-filtering gated graph neural network reports higher classification accuracy than its manually engineered feature counterpart. model also reports results for one- and two-fault-signal cases. The authors say these results support using learned vibration representations within a structured sensor graph. Their evidence is limited to the evaluated benchmark and operating condition; independent component localization, transfer across turbines, and deployment still require validation. The work therefore presents a benchmark result, not evidence that the method will perform similarly across turbines or in deployment.
Original source
eess.SY · Article
Source details
Original source date: Mar 6, 2023 (2023-03-06).
Date basis: Date basis not provided.
Added to Escape Meridian: 2026-11-10T09:10:59.648Z. First observed: 2026-11-03T08:51:18.951Z. Retrieved: 2026-11-03T19:05:43.760Z.
Limits
The authors say the evidence is restricted to the evaluated and operating condition; independent component localization, transfer across turbines, and deployment require further validation. The abstract does not report results from independent tests across turbines or deployment latency.
Open question
How does the method perform when independently tested for component localization, transfer across turbines, and deployment ?
Source-stated
The authors investigate a two-stage approach combining sparse filtering and graph neural networks to detect wind turbine gearbox faults from vibration measurements.
We investigate a two-stage framework that combines sparse filtering with graph neural networks for wind turbine gearbox fault detection.
Source-stated
The knowledge-based graph connects learned features with gearbox components, sensors, and health-state entities.
a knowledge-based graph associates these features with gearbox components, sensors, and health-state entities.
Source-stated
On the stated benchmark, the sparse-filtering gated graph neural network with 100 learned features reports 93.74% classification accuracy, compared with 89.28% for its manually engineered feature counterpart.
With 100 learned features, the sparse-filtering gated graph neural network achieves a reported classification accuracy of 93.74%, compared with 89.28% for its manually engineered feature counterpart.
Source-stated
The sequence model reports accuracies of 90.73% and 90.36% in the one- and two-fault-signal cases, respectively.
The sequence model achieves reported accuracies of 90.73% and 90.36% in the one- and two-fault-signal cases, respectively.
Source-stated
The authors restrict their evidence to the evaluated benchmark and operating condition, and identify further validation needs.
The evidence is restricted to the evaluated benchmark and operating condition; independent component localization, transfer across turbines, and deployment latency require further validation.
Source and contributor context
Wang, Jinsong, Loparo, Kenneth A. · 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.
Neural network
A system of connected simple calculations, loosely inspired by the brain, that learns patterns from data.
Benchmark
A defined set of tests used to compare systems. Results depend on the tasks and measures selected.
Latency
The delay between sending a request and receiving the result.
arXiv
arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.
The Sequence
The Sequence publishes machine-learning analysis and interviews whose guest answers retain their own attribution.
Editorial provenance
Version 1. Publication ID: 70fb1b4c-f8c5-4f29-8cf5-5bb767b3631d.
Prepared by gpt-6-luna; grounding review by gpt-6-luna.
Source check: passed by anthropic/claude-opus-5.5 on 2026-11-10T09:10:59.648Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: material_development.
Observation: 15f99d6bae29a249a71123f418aa0b6f7c9b90d18bc92e2403e0c0fecd793ea0. Source text hash: 7b0b5d577020a6f31f8ce80bd32b7abc36b2afb068f97083fd99cf52a8087f64.
passed · 2026-11-10T09:10:59.648Z · anthropic/claude-opus-5.5
Publication: 70fb1b4c-f8c5-4f29-8cf5-5bb767b3631d
attribution — pass: The reading correctly attributes the work to Jinsong Wang and Kenneth A. Loparo, as listed on the arXiv page. Results are kept as reported results and as the authors' conclusions, not as established fact.
Authors:Jinsong Wang, Kenneth A. Loparo
summary — pass: The title, summary and question faithfully reflect the abstract: the two-stage method, the graph linking features to components and sensors, learned versus manually engineered features, the sequence model, and the authors' stated limits. The question follows from those limits. A general reader can follow the core idea. Suggested tooltips: graph neural network, gated graph neural network, sparse filtering. Optional fidelity additions: the eight accelerometer channels and the actual reported numbers in the prose.
Experiments use eight accelerometer channels from the National Renewable Energy Laboratory gearbox condition-monitoring benchmark.
date — pass: The originalPublished date of 2023-03-06 matches the v1 submission date. The 2 Nov 2026 revision and the feed date are correctly not used as the original publication date.
[Submitted on 6 Mar 2023 (v1), last revised 2 Nov 2026 (this version, v2)]
support — pass: All claims and their excerpts match the source verbatim, including 93.74% vs 89.28% and 90.73% and 90.36%. The closing sentence, that this is a benchmark result and not evidence of performance across turbines or in deployment, stays within the authors' own stated limitation. No benefits or costs are invented.
The evidence is restricted to the evaluated benchmark and operating condition; independent component localization, transfer across turbines, and deployment latency require further validation.
media — pass: The selected media is the verified direct link to the original arXiv abstract page. It is the most direct and relevant 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.
A system of connected simple calculations, loosely inspired by the brain, that learns patterns from data.
Definition referenceA defined set of tests used to compare systems. Results depend on the tasks and measures selected.
Definition referenceThe Sequence publishes machine-learning analysis and interviews whose guest answers retain their own attribution.
Explore The SequenceThe delay between sending a request and receiving the result.
Definition referencearXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.
Explore arXivA defined set of tests used to compare systems. Results depend on the tasks and measures selected.
Definition referenceThe delay between sending a request and receiving the result.
Definition referenceThe delay between sending a request and receiving the result.
Definition referenceA system of connected simple calculations, loosely inspired by the brain, that learns patterns from data.
Definition referenceA defined set of tests used to compare systems. Results depend on the tasks and measures selected.
Definition referenceThe Sequence publishes machine-learning analysis and interviews whose guest answers retain their own attribution.
Explore The SequenceThe delay between sending a request and receiving the result.
Definition referencearXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.
Explore arXivA defined set of tests used to compare systems. Results depend on the tasks and measures selected.
Definition referenceThe delay between sending a request and receiving the result.
Definition referenceThe delay between sending a request and receiving the result.
Definition reference