Previsão de indicadores de continuidade com a utilização de redes neurais

• Ano de defesa: 2018
• Autor(a) principal: Louback, Filiphe Oliveira
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal do Espírito Santo; BR; Mestrado em Engenharia Elétrica; Centro Tecnológico; UFES; Programa de Pós-Graduação em Engenharia Elétrica
• Palavras-chave em Português: Estimação de indicadores de continuidade; Sistemas de energia elétrica - Qualidade; Indicadores - Qualidade; Redes neurais (Computação); Engenharia Elétrica; 621.3
• Link de acesso: http://repositorio.ufes.br/handle/10/10710

In Brazil, one of the optics through which the energy distributors' performance is evaluated is Quality of Service, which regulates the continuity of electric power supply through the DEC and FEC indicators. The power distributor is subject to penalties in the event of no compliance with regulatory limits. The estimation of these indicators provides a knowledge of the future panorama of the company, allowing the identification of areas that tend to deteriorate over time. The present work consists in the development of a methodology to estimate the indicators of continuity of energy supply using the application of Artificial Neural Networks. The work uses real data from EDP Espírito Santo and aims to estimate the daily DEC and FEC indicators of the São Mateus electrical set. The results obtained from the estimation of DEC and FEC show that the proposed model is feasible. The total errors accumulated in the DEC indicator forecast at the end of January and February 2016 were 1.71% and 0.87%, respectively. In the FEC forecast, the total errors accumulated at the end of January and February 2016 were 7.4468% and 10,9589%, respectively. Therefore, the results obtained by this work allow more adequate decisions are made on the execution of maintenance actions contributing to the operational reliability of the electric system.