Machine learning-based classification of operating regimes in an urban wastewater treatment plant: a benchmark on reconstructed reference labels
Identification
- Clé publique
- radar:r-922f89f5ebe9f74f73a95053a44ecd58
- Auteurs
- Byron Oviedo
- Année déclarée
- 2026
- Langue déclarée
- en
- Type source
- article
- Première détection
- 2026-11-14T01:59:49.873703+00:00
- Dernière observation des métadonnées
- 2026-11-14T01:59:49.873703+00:00
- Dernière vérification HTTP
- Non effectuée
Institutions déclarées par la source
Universidad Técnica Estatal de Quevedo
Pays d’affiliation déclarés
EC
Résumé documentaire — langue originale
Reliable identification of the operating condition of a wastewater treatment plant (WWTP) is central to sanitary-environmental engineering, since undetected process deviations translate into effluent non-compliance and public-health risk. This study benchmarks supervised machine learning for classifying the daily operating regime of a full-scale, predominantly domestic urban WWTP using the openly available UCI Water Treatment Plant dataset (38 physicochemical determinations at four control points over 527 consecutive days). The target was reconstructed from the date lists documented in the source, which assign every record to one of thirteen classes induced by conceptual clustering; the reconstruction assigns 527 of 527 records and reproduces the declared class counts. Grouped by engineering meaning, the data comprises three substantial regimes of normal operation (328, 116 and 69 days) and only fourteen abnormal days, so the task is posed as regime classification rather than fault detection. Four classifiers were compared under nested cross-validation, repeated stratified cross-validation with confidence intervals, and blocked chronological validation, on three predictor sets. The support vector machine with a radial-basis-function (RBF) kernel and the random forest were strongest on class-balanced criteria and, under a dependence-aware corrected resampled test, statistically indistinguishable (macro-F1 0.782 and 0.758; corrected p ≈ 0.09–0.41). Removing the removal-efficiency variables left performance essentially unchanged (macro-F1 0.786); because every variable entered the clustering, the task is framed as classification against an expert-supervised reference labelling, not the recovery of independent process structure. Under chronological validation macro-F1 fell to 0.536; although an upstream configuration retained 0.749 under random validation, its chronological performance was lower, so an anticipatory role is treated as a hypothesis for prospective study rather than a demonstrated early-warning capability. Permutation importance identified volatile suspended solids at the input to the primary settler as the most consistent predictor in this procedure, though other variables were influential in individual folds. The results support machine learning as a proof-of-concept monitoring layer while delimiting what a single historical plant record can establish.
Identifiants
- OPENALEX
- https://openalex.org/W7213445331
- DOI
- 10.3389/fenvs.2026.1941653
Repères documentaires
- Famille documentaire
- article
- Types déclarés par les sources
- article
- Éditeurs
- Frontiers Media
Les mentions d’entités ne constituent pas une preuve de reconnaissance juridique. La juridiction étudiée est distincte du pays d’affiliation des auteurs.
Sources
OpenAlex · Document 1
Citer cette référence
Byron Oviedo. Machine learning-based classification of operating regimes in an urban wastewater treatment plant: a benchmark on reconstructed reference labels. 2026. https://doi.org/10.3389/fenvs.2026.1941653
JSON · CSL-JSON · BibTeX · RIS · CSV · JSON-LD ·
APA, Chicago, ISO 690 : formats bibliographiques fondés sur les champs disponibles. OSCOLA / Bluebook : données structurées uniquement ; aucune conformité juridique complète n’est revendiquée sans vérification des éléments nécessaires.
- APA
- Byron Oviedo. (2026). Machine learning-based classification of operating regimes in an urban wastewater treatment plant: a benchmark on reconstructed reference labels. https://doi.org/10.3389/fenvs.2026.1941653
- Chicago
- Byron Oviedo. Machine learning-based classification of operating regimes in an urban wastewater treatment plant: a benchmark on reconstructed reference labels. 2026. https://doi.org/10.3389/fenvs.2026.1941653
- ISO 690
- Byron Oviedo. Machine learning-based classification of operating regimes in an urban wastewater treatment plant: a benchmark on reconstructed reference labels. 2026. https://doi.org/10.3389/fenvs.2026.1941653
Warning: Undefined array key "last_link_verified" in /home/clients/e27d81e12f04d2bfd747aa264540ac56/sites/_UDTR_APP/legal_library_reference.php on line 31
UDTR — Bibliothèque juridique mondiale des droits de la Nature. Machine learning-based classification of operating regimes in an urban wastewater treatment plant: a benchmark on reconstructed reference labels. Notice documentaire: https://www.declarationuniverselledesdroitsdelarbre.org/global-rights-of-nature-radar/reference/r-922f89f5ebe9f74f73a95053a44ecd58/ · Dernière vérification HTTP : non effectuée
Données de cette version figée
Dossiers et relations documentaires
Rôles des sources
Source bibliographique / publication d’origine · https://doi.org/10.3389/fenvs.2026.1941653
Un lien vers une autorité ne prouve pas à lui seul une affirmation juridique. Les résumés et les textes intégraux restent soumis aux conditions de leur source.
Statut documentaire
Pertinence à examiner
Bibliothèque
Cette identité n’est pas encore intégrée à la Bibliothèque juridique.