رسالة جامعية
Analysis, classification and prediction of water and electricity consumption by machine learning techniques ; Analyse, classification et prédiction de consommation d’eau et d’électricité par des techniques de machine learning
| العنوان: | Analysis, classification and prediction of water and electricity consumption by machine learning techniques ; Analyse, classification et prédiction de consommation d’eau et d’électricité par des techniques de machine learning |
|---|---|
| المؤلفون: | Boudhaouia Miled, Aida |
| المساهمون: | Institut de Recherche en Informatique Mathématiques Automatique Signal - IRIMAS - UR 7499 (IRIMAS), Université de Haute-Alsace (UHA) Mulhouse - Colmar (Université de Haute-Alsace (UHA)), Université de Haute Alsace - Mulhouse, Patrice Wira |
| المصدر: | https://theses.hal.science/tel-03562074Test ; Intelligence artificielle [cs.AI]. Université de Haute Alsace - Mulhouse, 2022. Français. ⟨NNT : 2022MULH3724⟩. |
| بيانات النشر: | HAL CCSD |
| سنة النشر: | 2022 |
| مصطلحات موضوعية: | Smart consumption meter, Load curve, Time series, Data analysis, Classification, Clustering, Prediction, Machine learning, Artificial neural networks, SARIMA, LSTM, Réseaux de neurones artificiels, Compteur intelligent de consommation, Courbe de charge, Série temporelle, Analyse de données, Prédiction, Apprentissage automatique, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], [INFO.INFO-SY]Computer Science [cs]/Systems and Control [cs.SY] |
| الوصف: | The objective of this thesis is to study users behavior as water and electricity consumers. This is achieved from data collected by smart meters and available in the form of time series of irregularly spaced events in time that represent the instants of consumption at a single measurement point. These data are also translated into the form of Load Curves (Cdc) generally sampled. After a pre-treatment, the data is used for the supervised classification of daily water quantities with different Machine Learning (ML) algorithms. Then, a new approach was proposed to determine the probability density function of the daily water consumption of users in order to reduce data and optimize computer resources. This has facilitated the development of an algorithm for detecting small as well as large water leaks. Many supervised ML approaches have been applied and individually tested for the classification of electrical devices according to the harmonic current characteristics they generate. Secondly, we have proposed hybrid approaches that combine the previous methods in order to obtain highly efficient classification results. Finally, in the context of consumption forecasting, hybrid models that combine deterministic and learning approaches have been implemented and tested to predict the next moments of consumption or the quantities of consumption expected in the future hours with much better performance. ; L’objectif de cette thèse consiste à étudier le comportement des utilisateurs en tant que consommateurs d’eau et d’électricité. Ceci est réalisé à partir de données collectées par des compteurs intelligents et disponibles sous la forme de séries temporelles d’événements irrégulièrement espacés dans le temps qui représentent les instants de consommation d’un seul point de mesure. Ces données sont également traduites sous la forme de Courbes de charge (Cdc) généralement échantillonnées. Après un pré-traitement, les données servent à la classification supervisée des CdC journalières d’eau avec différents algorithmes de ... |
| نوع الوثيقة: | doctoral or postdoctoral thesis |
| اللغة: | French |
| العلاقة: | NNT: 2022MULH3724; tel-03562074; https://theses.hal.science/tel-03562074Test; https://theses.hal.science/tel-03562074/documentTest; https://theses.hal.science/tel-03562074/file/2022MULH3724_these_BOUDHAOUIA.pdfTest |
| الإتاحة: | https://theses.hal.science/tel-03562074Test https://theses.hal.science/tel-03562074/documentTest https://theses.hal.science/tel-03562074/file/2022MULH3724_these_BOUDHAOUIA.pdfTest |
| حقوق: | info:eu-repo/semantics/OpenAccess |
| رقم الانضمام: | edsbas.23AF0A78 |
| قاعدة البيانات: | BASE |
| الوصف غير متاح. |