Development of a generator prognostic tool
Bibliographic record
Abstract
In the past decades, significant improvements in generator diagnostics were made possible by using continuous online measurements and a number of periodic tests. In recent years, the data provided has been converted into more useful information thanks to integrated diagnostic systems. For example, an integrated methodology for generator diagnostics (MIDA) was developed at Hydro-Québec's research institute (IREQ) using a Web-based application. This comprehensive diagnostic system gives the degradation state of generator stator winding insulation by using a portfolio of diagnostic tools. Combining the various results leads to a health index ranging from 1 (good condition) to 5 (worst condition). This system is used by power plant managers as well as technical support and maintenance engineers at Hydro-Québec in the context of condition-based maintenance (CBM). The next step of development is to add new prognostic-related characteristics. This involves automatic identification of active failure mechanisms, root cause analysis and estimation of the stage of advancement of any active mechanism. These characteristics form the basis of predictive maintenance and support the optimization of maintenance strategies. The approach chosen is based on a number of cause- and-effect chains formed by the combination of sequential physical degradation states that ultimately lead to failure. Each combination of physical states is unique and defines a particular failure mechanism. Failure mechanism analysis was followed by identification of all observable symptoms (diagnostics from MIDA) for each physical state. This paper presents a first step toward the development of a prognostic tool, where the modeling of failure mechanisms is combined with automatic analysis of observable symptoms from our diagnostic system. It puts forward probable failure mechanisms for a given generator.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.009 | 0.004 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".