A New Generic Approach to Convert FMEA in Causal Trees for the Purpose of Hydro-Generator Rotor Failure Mechanisms Identification
Bibliographic record
Abstract
At Hydro-Québec (HQ), an integrated diagnostic system (MIDA) is currently used to assess hydro-generators health index. This system gives the global health index but does not propose any understanding of active failure mechanisms. At this point, this work needs to be done by experts after analysis of the diagnostic data in MIDA. To relieve the expert from part of this work, a prognostic tool, that uses a Failure Mechanisms and Symptoms Analysis (FMSA), is under development. The approach is based on the understanding of the evolution of degradation processes for each failure mechanism. Failure mechanisms are structured as causal trees and defined as a sequence of physical states starting from a root cause and ending with a failure mode. A physical state corresponds to characteristic degradation condition of a component of the generator. Each physical state being defined by a unique combination of symptoms as measured with diagnostic tools. After consigning all possible mechanisms occurring in both the rotor and the stator, the symptoms logged into a database can be read to automatically identify all active physical state and active failure mechanisms. This approach has been under development in HQ for the stator for a number of years and is now extended to the rotors of hydro-generators.The purpose of this paper is to present the structured method used to build the failure mechanisms from bits and pieces of information (sub-mechanisms) found in the literature and from discussions with experts. This new methodology is based on a two steps process. First, sub-mechanisms were extracted from FMEA in the literature. Then, an algorithm was used to generate a set of causal trees from these sub- mechanisms. The generated results then had to be validated by experts to make sure that automatically generated mechanisms were logical and plausible. The resulting extended failure mechanisms trees built can then be used for the purpose of Root Cause Analysis (RCA), model-based diagnostics and prognosis. This method was developed to be as generic as possible so it could be applied to any complex system.
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 | 0.002 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.003 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.021 | 0.006 |
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".