Fault detection and identification using a novel process decomposition algorithm for distributed process monitoring
Bibliographic record
Abstract
Recent progress in fault detection and identification increasingly relies on sophisticated techniques for fault detection, applied through either centralized or distributed approaches. Instead of increasing the sophistication of the fault detection method, this work introduces a novel algorithm for determining process blocks of interacting measurements and applies principal component analysis (PCA) at the block level to identify fault occurrences. Additionally, we define a novel contribution map that scales the magnitudes of disparate faults to facilitate the visual identification of abnormal values of measured variables and analysis of fault propagation. Bayesian aggregate fault index and block fault indices vs. time pinpoint origins of the fault. The proposed method yields fault detection rates on par with most sophisticated centralized or distributed methods on the Tennessee Eastman Process (TEP) benchmark. Since the decomposition algorithm relies on the process flowsheet and control loop structures, practicing control engineers can implement the proposed method in a straightforward manner. • Novel process decomposition algorithm based on process flow diagrams and control loop structures. • Monitoring blocks created without using correlation data. • Improved block definitions through community merging based on measurement allocation. • Fault detection rates comparable to complex centralized and distributed monitoring methods. • Simple fault detection method designed for practicing engineers.
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 | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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".