| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 18.72 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
Os algoritmos de rastreamento do Ponto de PotênciaMáxima (MPPT) têm grande relevância
para o rendimento otimizado emsistemas fotovoltaicos conectados à rede. Muitos algoritmos
foram investigados. No entanto, existe uma grande dificuldade em encontrar análises
comparativas com resultados experimentais entre as diferentes técnicas, já que grande parte
das pesquisas se baseiam emrevisões da literatura ou em simulação.
Este trabalho apresenta umanálise experimental entre quatro técnicas MPPT, que estão
entre as mais discutidas na literatura: duas das mais simples (Perturb & Observe e Condutância
Incremental) e duas das mais complexas (Particle Swarm Optimization e Fuzzy
Logic Controller). São utilizados ainda inversores comerciais disponíveis para teste em laboratório,
comobjetivo de entender qual algoritmo MPPT é utilizado, o seu funcionamento e
procedimento de inicialização.
Todos os algoritmos são implementados e validados durante testes experimentais, sob
condições reais, as quais operam os sistemas PV. Os testes buscam ainda garantir as mesmas
condições experimentais, para que a avaliação de desempenho de cada algoritmo reflita o
seu real desempenho em rastrear o ponto de máxima potência.
Os testes experimentais são divididos em 3: testes em condições normais, testes em
condição de sombreamento parcial e testes com inversores PV comerciais.
Para a sua implementação, utilizou-se uma estrutura de potência baseada em um conversor
DC-DC elevador ou conversor Boost, e um inversor monofásico de ponte completa. O
controle e leitura dos sinais é implementado utilizando o Simulink® e dSPACE 1103, placa
controladora em tempo real. Além disso, as técnicas MPPT de três inversores comerciais
também são analisados.
Maximum Power Point tracking algorithms (MPPT) have great relevance for the optimized performance in photovoltaic systems connected to the grid. Many algorithms have been investigated. However, there is a great difficulty in finding comparative analyzes based on experimental results between the different techniques, since most of the research is based on literature reviews or simulation. This work presents an experimental analysis between four MPPT techniques, which are among the most discussed in the literature: two of the simplest (Perturb & Observe and Incremental Conductance) and two of themost complex (Particle SwarmOptimization and Fuzzy Logic Controller). Commercial inverters available for laboratory testing are also used in order to understand whichMPPT algorithmis used, how does it work and how is it initialized. All algorithms are implemented and validated during experimental tests, under real conditions, which operate the PV systems. The tests also seek to guarantee the same experimental conditions, so that the performance evaluation of each algorithm reflects its real performance in tracking the maximumpower point. The experimental tests are divided into 3: tests under normal conditions, tests under partial shading condition and tests with commercial PV inverters. For implementation, a power structure based on a DC-DC Boost converter and a singlephase full-bridge inverter was used. The control and reading of the signals is implemented using Simulink® and dSPACE 1103 real-time controller board. In addition, three MPPT techniques of commercial inverters are also analyzed.
Maximum Power Point tracking algorithms (MPPT) have great relevance for the optimized performance in photovoltaic systems connected to the grid. Many algorithms have been investigated. However, there is a great difficulty in finding comparative analyzes based on experimental results between the different techniques, since most of the research is based on literature reviews or simulation. This work presents an experimental analysis between four MPPT techniques, which are among the most discussed in the literature: two of the simplest (Perturb & Observe and Incremental Conductance) and two of themost complex (Particle SwarmOptimization and Fuzzy Logic Controller). Commercial inverters available for laboratory testing are also used in order to understand whichMPPT algorithmis used, how does it work and how is it initialized. All algorithms are implemented and validated during experimental tests, under real conditions, which operate the PV systems. The tests also seek to guarantee the same experimental conditions, so that the performance evaluation of each algorithm reflects its real performance in tracking the maximumpower point. The experimental tests are divided into 3: tests under normal conditions, tests under partial shading condition and tests with commercial PV inverters. For implementation, a power structure based on a DC-DC Boost converter and a singlephase full-bridge inverter was used. The control and reading of the signals is implemented using Simulink® and dSPACE 1103 real-time controller board. In addition, three MPPT techniques of commercial inverters are also analyzed.
Description
Mestrado de dupla diplomação com a UTFPR - Universidade Tecnológica Federal do Paraná
Keywords
MPPT algorithms