DENSIFICAÇÃO DE DADOS BASEADA EM MEAN SHIFT APLICADA A PREVISÃO DE GERAÇÃO DE ENERGIA EÓLICA

• Ano de defesa: 2022
• Autor(a) principal: SOUZA, Hellen Dianne Pereira de
• Orientador(a): LIMA, Shigeaki Leite de
• Banca de defesa: LIMA, Shigeaki Leite de , OLIVEIRA, Denisson Queiroz , SAAVEDRA MENDEZ, Osvaldo Ronald , SOUZA, André Nunes de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal do Maranhão
• Programa de Pós-Graduação: PROGRAMA DE PÓS-GRADUAÇÃO EM ENGENHARIA DE ELETRICIDADE/CCET
• Departamento: DEPARTAMENTO DE ENGENHARIA DA ELETRICIDADE/CCET
• País: Brasil
• Palavras-chave em Português: Mean Shift; Teoria da informação; Previsão; Geração de energia; Energia eólica
• Palavras-chave em Inglês: Mean Shift; Information theoretic; Forecast; Energy Generation; Wind energy
• Área do conhecimento CNPq: Transmissão da Energia Elétrica, Distribuição da Energia Elétrica
• Link de acesso: https://tedebc.ufma.br/jspui/handle/tede/3951

Brazil in the last 10 years has a high potential for wind energy production, showing significant growth each year with the installation of new parks and an increase in installed power. In view of the power generation capacity, which makes it possible to generate wind important, how to carry out power which can generate the wind capacity important, as the factors that generate power may generate the wind important. In this sense, the projected wind speed forecast models, numerical weather forecast (NWP) models, many of the weather forecast models occur during generation. Others, such as statistical models and hybrids, perform analyzes based on data from a historical series. Wind power generation is used by many energy companies for the purpose of safety estimation and in this way guaranteeing the resource for the revenue company and the consumer to increase an energy source from source during a certain period of time. Many techniques were calculated with resources and working hours and with the same efficiency, working days and hours. This is an information theory based technique for a wind measurement database, generating new data with the same probability density function as the original set and using these virtual data as forecast scenarios.