Fault detection and severity classification in HVAC fan coil units using a hybrid autoencoder-based method
Bibliographic record
Abstract
Abstract Efficient HVAC operation is vital to reduce energy waste and discomfort, but systems often face various faults. As automation grows, there is a pressing need for fault detection methods that go beyond basic identification. This study presents a hybrid fault detection framework for fan coil units (FCUs) that leverages simulated fault-labelled data for evaluation and unsupervised learning for anomaly detection. An autoencoder model is trained exclusively on fault-free operational data to learn the normal behaviour of the system. Reconstruction errors from the autoencoder are used to detect anomalies, and a sensitivity analysis is conducted to determine optimal percentile-based thresholds across multiple fault scenarios. These fault specific thresholds are then generalized into three global severity bands based on their statistical distribution. The methodology is validated using a high-fidelity simulated dataset representing an independent fault scenario in a four-pipe FCU system. A test case involving a reverse-acting control fault shows that severity levels correspond to increasing heating and cooling loads, confirming the operational impact of anomaly classification. The proposed approach enables early fault detection, severity quantification, and energy-aware prioritization of maintenance tasks, offering a scalable solution for integration into building automation and facility management systems.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 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".