Integration of Hydraulic and Thermal Sensors with Machine Learning For Enhanced Leak Detection and Localization in District Heating Systems
Bibliographic record
Abstract
Accurate leak localization within district heating networks (DHNs) is a significant challenge due to the complex hydraulic dynamics that govern these systems.Traditional methods for leak detection, such as manual inspections and thermal imaging, are often inefficient for real-time applications.Recent advancements in artificial intelligence (AI), particularly artificial neural networks (ANNs), offer promising solutions by analysing pressure and temperature variations in DHN pipelines to identify leak locations.This paper explores the potential of AI-driven techniques for improving leak localization accuracy within DHNs.Using synthetic data generated with a Modelica-based simulation of DHN conditions in Grenoble, France, the study evaluates the impact of pressure and temperature sensors installed in DHN.The model was developed and trained using multi-class classification techniques, with the dataset balanced via Random Under Sampling (RUS) to address class imbalances.Feature selection was performed using Random Forest to identify the most critical input features, which were then used in the ANN for leak localization.The results demonstrate the effectiveness of pressure sensors, particularly in the return lines, for enhancing leak localization precision, while temperature sensors, though less directly indicative of leaks, also contribute valuable insights.The study concludes that AI-based approaches, coupled with strategically placed sensors, can significantly improve the accuracy and efficiency of leak localization in DHNs, contributing to more effective predictive maintenance and reduced system downtime.
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) | 0.000 | 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 | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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".