Análise de modelos para previsões de consumo de energia elétrica e seu uso na redução de custos

Abstract

Machine learning has been widely used in a number of research and application domains, such as search engines, fraud detection, natural language processing, anomaly detection, demand forecasting, and standalone vehicles. In this document, we focus on analyzing the major forecasting models in order to choose the one that best ts the residential energy consumption data. In addition, an analysis was performed on the data, changes that were necessary to keep the data within the chosen requirements, also presented all the results of the chosen models and explained how each model adapted to the data. Moreover, we also show all the technologies used to conduct the present work.