Une étude comparative sur la prédiction de la production de l’énergie photovoltaïque à l’aide des méthodes d’apprentissage automatique

Abstract

This Master’s thesis presents a comparative study of various machine learning algorithms used to predict photovoltaic energy production. Using the Orange3 software, models such as linear regression, decision trees, AdaBoost, and PLS were evaluated based on classification performance metrics. The main objective is to identify the most effective models for forecasting solar energy output using environmental data. The study highlights the growing role of artificial intelligence in the smart management of renewable energy and suggests future directions, including the use of real-time weather data and advanced techniques such as Machine Learning.