SMAIS: Inteligência Artificial para Serviços Municipalizados de Água e Saneamento
Fuente:
Zenodo
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Recurso digital |
| Sprache: | Portugiesisch |
| Veröffentlicht: |
Zenodo
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866901928079785984 |
|---|---|
| author | Peres, Ricardo Oleiro Araújo, Sara Ferrada, Filipa Nico Casimiro, Paulo Adão, Paula Sapinho, Paulo |
| author_facet | Peres, Ricardo Oleiro Araújo, Sara Ferrada, Filipa Nico Casimiro, Paulo Adão, Paula Sapinho, Paulo |
| contents | <p>Here’s a professional English translation of your text:</p> <p>The water sector in Portugal faces real losses (leaks in networks and service connections) and apparent losses (measurement errors, illicit consumption, billing failures), compromising the financial and environmental sustainability of managing entities. In SMAS Almada, these issues result in resource waste and underbilling, while traditional detection methods, such as physical inspections, are costly, time-consuming, and prone to errors.</p> <p>SMAIS – Artificial Intelligence for Municipal Water and Sanitation Services is an innovative project based on Artificial Intelligence and satellite remote sensing, focused on detecting and monitoring water losses in supply networks, covering both physical leaks and unauthorized consumption. The methodology combines Sentinel-1 SAR and Sentinel-2 multispectral data, meteorological data, consumption histories, and local sensors. These are analyzed by advanced machine learning algorithms (object detection, segmentation, classification, and anomaly detection), identifying inconsistent patterns that indicate leaks or illicit consumption, enabling a non-invasive, scalable, and cost-effective approach.</p> <p>For leak detection, patterns of moisture, spectral variations, and surface displacements that may indicate pipe breaks or degradation are analyzed. For theft detection, computer vision models estimate expected consumption based on visible elements in residential units (pools, lawns, dwellings), cross-referencing these estimates with billing records to identify discrepancies that may indicate illegal connections or meter tampering.</p> <p>Pilot cases in residential areas and public spaces demonstrate the practical feasibility of the solution and its adaptability to different consumption contexts. The results will be integrated into an interactive digital platform, enabling remote inspections, semi-automated analyses, and digital capacity building, aligned with the principles of Trustworthy AI and the European AI Act.</p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_18789326 |
| institution | Zenodo |
| language | por |
| publishDate | 2025 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | SMAIS: Inteligência Artificial para Serviços Municipalizados de Água e Saneamento Peres, Ricardo Oleiro Araújo, Sara Ferrada, Filipa Nico Casimiro, Paulo Adão, Paula Sapinho, Paulo Remote Sensing Technology Artificial Intelligence Deep Learning Computer vision <p>Here’s a professional English translation of your text:</p> <p>The water sector in Portugal faces real losses (leaks in networks and service connections) and apparent losses (measurement errors, illicit consumption, billing failures), compromising the financial and environmental sustainability of managing entities. In SMAS Almada, these issues result in resource waste and underbilling, while traditional detection methods, such as physical inspections, are costly, time-consuming, and prone to errors.</p> <p>SMAIS – Artificial Intelligence for Municipal Water and Sanitation Services is an innovative project based on Artificial Intelligence and satellite remote sensing, focused on detecting and monitoring water losses in supply networks, covering both physical leaks and unauthorized consumption. The methodology combines Sentinel-1 SAR and Sentinel-2 multispectral data, meteorological data, consumption histories, and local sensors. These are analyzed by advanced machine learning algorithms (object detection, segmentation, classification, and anomaly detection), identifying inconsistent patterns that indicate leaks or illicit consumption, enabling a non-invasive, scalable, and cost-effective approach.</p> <p>For leak detection, patterns of moisture, spectral variations, and surface displacements that may indicate pipe breaks or degradation are analyzed. For theft detection, computer vision models estimate expected consumption based on visible elements in residential units (pools, lawns, dwellings), cross-referencing these estimates with billing records to identify discrepancies that may indicate illegal connections or meter tampering.</p> <p>Pilot cases in residential areas and public spaces demonstrate the practical feasibility of the solution and its adaptability to different consumption contexts. The results will be integrated into an interactive digital platform, enabling remote inspections, semi-automated analyses, and digital capacity building, aligned with the principles of Trustworthy AI and the European AI Act.</p> |
| title | SMAIS: Inteligência Artificial para Serviços Municipalizados de Água e Saneamento |
| topic | Remote Sensing Technology Artificial Intelligence Deep Learning Computer vision |
| url | https://doi.org/10.5281/zenodo.18789326 |