Desenvolvimento de uma ferramenta baseada em sistema multiagente para obtenção da configuração do parque gerador de eletricidade por meio dos Leilões de Energia Nova

• Ano de defesa: 2019
• Autor(a) principal: Rafaella de Souza Henriques
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; Brasil; ENG - DEPARTAMENTO DE ENGENHARIA ELÉTRICA; Programa de Pós-Graduação em Engenharia Elétrica; UFMG
• Palavras-chave em Português: Setor elétrico brasileiro; Leilão de Energia Nova; Sistema multiagente; Engenharia elétrica; Geradores hidrelétricos
• Link de acesso: http://hdl.handle.net/1843/30849

The Brazilian Electricity Sector has tools for its planning, such as DECOMP and NEWAVE, but in terms of execution, there are issues to be addressed. The auction which is the mechanism of expansion of the power generation park in Brazil, often do not reflect the real requirement to meet the demand with the lowest possible cost. Considering the New Energy Auctions, the insufficient advance of their realization is highlighted as deficiency, generating delays in the operation beginning of the generating plants. In addition, the lack of tools to analyze which aspects impact the fall in the price paid for energy in these auctions is also a shortcoming. In order to assist in these issues, this study proposes a tool based on Multiagent System (MAS) to obtain the composition of the New Energy Auctions which ensures the desired configuration of the power generating park, as well as to help in defining the aspects that influence the price paid for energy in the New Energy Auctions. Therefore, it is presented a model with agents who represent the Ministry of Mines and Energy (MME), the National Electric Energy Agency (NEEA), the Electric Energy Trading Chamber (EETC) and the competitors of the New Energy Auctions. This model is composed of a cooperation subsystem, which includes a Multiobjective Mixed Integer Linear (MMIL) model for auction planning, and a competition subsystem, which covers the simulation of the energy product auctions, considering their current systematic. Besides the proposed MAS based model, this study endeavors to use real data from the Brazilian Electricity Sector as input, with the aim of providing a decision support tool for planning and executing of the New Energy Auctions.