Análise de técnicas de otimização por enxame de partículas em sistemas fotovoltaicos conectados à rede elétrica

• Ano de defesa: 2020
• Autor(a) principal: Lopes, Guilherme de Lima
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Tecnológica Federal do Paraná; Cornelio Procopio; Brasil; Programa de Pós-Graduação em Engenharia Elétrica; UTFPR
• Palavras-chave em Português: Sistemas de energia fotovoltaica; Algoritmos computacionais; Controladores elétricos; Photovoltaic power systems; Computer algorithms; Electric controllers; CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA; Engenharia Elétrica
• Link de acesso: http://repositorio.utfpr.edu.br/jspui/handle/1/5430

Considering that the photovoltaic (PV) system may be subject to partial shading conditions, i.e., presenting a global (GMPP) and several local (LMPP) maximum power points, this work presents a study and an analysis of techniques for maximum power point tracking (MPPT). In this work, the MPPT techniques are based on particle swarm optimization (PSO), which can search for GMPP, improving the tracking efficiency of the overall system maximum power point (MPP), as well as and increasing the tracking factor. In this study, the following methods are compared: Perturb and Observe (P&O); Traditional PSO (PSO 1), PSO with non-linear inertial weight (PSO 2); PSO with chaotic improvement (PSO 3); and, finally, the proposed PSO with varying inertial weight from a logistic map (PSO 4). The PSO algorithms are analyzed in a photovoltaic system connected to the single-phase utility grid, which is composed by a step-up DC-DC converter (Boost) and a single-phase full bridge inverter. In addition, the PSO algorithm is also used to optimize the gains of the controllers present both in the multi-loop control of the Boost converter, as well as in the control of the DC bus voltage and the output current loop of the single-phase inverter. Thereby, the presented PV system is able to operate with a better dynamic performance, in addition to presenting better transient and steady-state in the search for the GMPP, since both the search for MPP and the controller gains are optimized. The PV system is analyzed considering the scenario of uniform solar irradiation on the arrangement, as well as, when partial shading conditions occur. By means of simulation results, the MPPT techniques are evaluated to each other comparing the tracking factor, convergence time, power oscillations in steady-state and tracking, efficiency in the search for the GMPP, as well as the total harmonic distortion of the injected grid-current. In addition, the performance analysis of the PV system is performed considering the tuning of the PI compensators by the frequency response analysis method and the PSO associated with the frequency response analysis.