Synthetic Fouling Image Data Generation for Heat Exchanger Predictive Maintenance
Bibliographic record
Abstract
Regular maintenance of heating, ventilation, and air conditioning (HVAC) systems is vital for building sustainability and energy efficiency, as neglect can lead to performance degradation, increased energy consumption, and reduced equipment lifespan. Fouling, the accumulation of unwanted material on surfaces, is a significant issue in HVAC heat exchangers, leading to efficiency losses and increased operational costs. While preventive maintenance is commonly employed, predictive maintenance offers a more effective approach, particularly when detecting fouling that requires visual data. However, the rarity of anomalies in such data complicates the training of deep learning models. This paper introduces an approach for synthetic fouling data generation for predictive maintenance designed to simulate fouling on a heat exchanger in a cooling tower. A synthetic image dataset is generated with two distinct types: continuous growth scenarios for training purposes and scheduled maintenance scenarios for evaluating model performance. Resulting in a collection of 4,260 images over a 60-day period. The effectiveness of the dataset is validated through experiments where a U-Net model is trained for semantic segmentation in fouling detection, achieving an average F1 score of 0.8595 across the test scenarios. These results confirm that the synthetic dataset is effective for training convolutional neural networks for fouling detection, laying the groundwork for integrating such models into a broader predictive maintenance strategy for cooling towers.
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 | 0.000 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| 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".