Título: Prediction of Electricity Consumption for Optimization of Shared Credit Systems Using Machine Learning
Autores: Rhuan Lucas Alves Soares, Luiz Felipe Pugliese, Elcio Franklin de Arruda, Rodrigo Aparecido da Silva Braga, Giovani Bernardes Vitor
Resumo: Brazil has witnessed remarkable growth in distributed energy generation, driven by government incentives and the credit compensation market. This scenario, particularly within the solar energy sector, has led to the emergence of companies investing in large-scale areas for the installation of solar panels, enabling high-volume energy production and the sale of surplus. However, this exponential growth has introduced significant challenges in the photovoltaic sector, including the need to forecast plant generation, anticipate individual customer consumption, and optimize energy credit allocation. In this context, this work proposes the application of Machine Learning algorithms using Python and cloud resources with SQL queries—specifically through linear regression—to predict each customer’s electricity consumption for the following month. The objective is to achieve the lowest possible error rate and thereby enable optimization of the credit allocation system.
Palavras-chave: Distributed generation; credit clearing market; Machine Learning; cloud resource; consumption forecast.
Páginas: 8
Código DOI: 10.21528/CBIC2025-1166691
Artigo em PDF: CBIC_2025_paper1166691.pdf
Arquivo BibTeX:
CBIC_2025_1166691.bib