Controle preditivo com número reduzido de estados aplicado a conversor multinível modular

• Ano de defesa: 2017
• Autor(a) principal: Crispino, Laís Ferreira
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal do Rio de Janeiro; Brasil; Instituto Alberto Luiz Coimbra de Pós-Graduação e Pesquisa de Engenharia; Programa de Pós-Graduação em Engenharia Elétrica; UFRJ
• Palavras-chave em Português: Conversores elétricos; Distribuição de energia elétrica; CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA::SISTEMAS ELETRICOS DE POTENCIA::MEDICAO, CONTROLE, CORRECAO E PROTECAO DE SISTEMAS ELETRICOS DE POTENCIA
• Link de acesso: http://hdl.handle.net/11422/6095

The modular multilevel converter (MMC) has been a subject of increasing interest for power electronics researchers. Therefore, the use of this topology in the most diverse application areas has been studied, especially in high voltage direct current transmission systems. Thus, this work describes the MMC converter model, considering both half and full-bridge submodules, associated with a model predictive control strategy, which considers a reduced number of states to control the power flow delivered to the grid, through AC current regulation; and for minimizing circulating current components. The algorithm used to implement the balancing of submodules’ capacitor voltages was developed based on sorting methods, but also considering the cost function idea. In order to verify the performance of the adopted strategy, qualitative analyses were made based on simulation results conducted in the environment of circuit simulation PSIM, in which the logic of the control system was programmed in C language. Tests were also carried out on an experimental prototype whose circuit refers to one leg of the MMC, and the results obtained confirm the model. In these tests the circulating current was not analyzed.