Machine learning for early detection of distillation column flooding
Bibliographic record
Abstract
• Demonstrates effectiveness in realistic navigation scenarios with obstacles via simulation. • The issue with data scarcity surrounding the application of supervised ML for flooding detection is addressed. • The time-series generative adversarial network is used to generate synthetic data while preserving the temporal order. • Sets of industrial data are used to demonstrate the efficacy of the algorithm. Flooding in a distillation column is an abnormal event that limits the operations of the column and eventually leads to plant shutdown if not prevented. Recently, machine learning (ML) has been widely employed in process engineering to uncover critical patterns in data. Supervised ML methods can predict flooding by monitoring pressure changes in the column. However, one of the challenges of applying supervised ML methods for predicting flooding in distillation columns is the need for large volumes of flooding data. Flooding events are rare compared to normal operations, resulting in an imbalanced dataset. To address this, we used a time-series generative adversarial network to generate synthetic flooding data by preserving the temporal patterns of the original dataset. With this extra data, we trained supervised ML models to predict flooding by forecasting the pressure drops. Our results show that flooding can be detected 19 min in advance, and supervised ML methods outperformed unsupervised ML methods like PCA and Autoencoders in early detection.
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) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".