The wind turbine generation system (WTGS) exhibits a nonlinear characteristic and its maximum power point varies with changing atmospheric conditions. In order to operate the WTGS at maximum power output under various wind speeds and to avoid using speed encoder in practical applications, it is necessary to improve the controller system to operate the maximum power points in the WTGS. There are three factors to influence wind generator, the wind speed, power coefficient and the radius of blade. The power coefficient depends on the blade pitch angle and tip speed ratio (TSR). The objective of the dissertation is to develop an intelligent controlled wind energy conversion system (WECS) using AC/DC and DC/AC power converters for grid-connected power application. To achieve a fast and stable response for the real power control, an intelligent controller was proposed, which consists of a fuzzy neural network (FNN), a recurrent fuzzy neural network (RFNN), a wilcoxcon radial basis function network (WRBFN) and a improved Elman neural network (IENN) for MPPT. Furthermore, the parameter of the developed FNN, RFNN, WRBFN and IENN are trained on-line using back-propagation learning algorithm. However, the learning rates in the FNN, RFNN, WRBFN, and IENN are usually selected by trial and error method, which is time-consuming. Therefore, modified particle swarm optimization (MPSO) method is adopted to adjust the learning rates to improve the learning capability of the developed RFNN, WRBFN and IENN. Moreover, presents the estimation of the rotor speed is based on the sliding mode and model reference adaptive system (MRAS) speed observer theory. Furthermore, a sensorless vector-control strategy for an induction generator (IG) operating in a grid-connected variable speed wind energy conversion system can be achieved. On the other hand, a WRBFN based with hill-climb searching (HCS) maximum-power-point-tracking (MPPT) strategy is proposed for permanent magnet synchronous generator (PMSG) with a variable speed wind turbine. Finally, many simulation results are provided to show the effectiveness of the proposed intelligent control wind generation systems.

Orientador: Dr. Alfeu J. Sguarezi Filho / Dissertação (mestrado) - Universidade Federal do ABC, Programa de Pós-Graduação em Engenharia Elétrica, 2015. / Neste trabalho apresenta-se uma configuração do sistema de armazenamento de energia com baterias aplicado a sistemas de geração de energia eólica empregando conversores de fonte Z conectados à rede elétrica. Os geradores de indução gaiola de esquilo, são frequentemente utilizados nos sistemas de geração de energia eólica, por sua robustez, simplicidade, peso menor e custo baixo. Este é conectado diretamente ao conversor de potência bidirecional back to back, pode fornecer potências ativa e reativa à rede elétrica. Além disso, é estudado o conversor de fonte Z aplicado nesta topologia. No entanto, a implantação de sistemas de armazenamento de energia com baterias nos sistemas de geração de energia eólica na atualidade é muito importante, devido à possibilidade de oscilações da tensão e corrente na rede elétrica, portanto, estes podem ajudar à estabilização das tensões, correntes e a frequência na rede elétrica. Este sistema é conectado ao conversor back to back por meio de um conversor elevador-abaixador de corrente contínua. Para controlar a velocidade no eixo do rotor no gerador de indução, a estratégia é baseada no controle direto de torque. Enquanto, para o conversor do lado da rede é empregada a técnica de controle orientado pela tensão. Para o banco de baterias é utilizado o controle da tensão no barramento de corrente contínua e do fluxo na corrente da bateria, utilizando controladores do tipo PI. Com os novos desenvolvimentos tecnológicos nas chaves de potência, são apresentadas topologias de conversores CC-CA como o conversor de fonte Z, este tipo de conversor corrige algumas limitações do conversor back to back, com as características de elevador/abaixador de tensão, sem o uso de dispositivos de comutação, são permitidos os curto-circuitos na chaves, empregando novas técnicas de modulação, e reduz a quantidade harmônica injetada na rede elétrica. Os estudos foram realizados por meio de técnicas de simulação computacional usando modelos matemáticos do sistema estudado para a validação das estratégias de controle empregadas em diferentes condições de operação. Para as simulações empregou-se a ferramenta computacional SimPowerSystems R do Matlab/Simulink R . / This paper presents a battery energy storage system applied to wind power generation based on Z-source inverter connected to the power grid. The squirrel cage induction generators, often used in wind power generation systems, for its robustness, simplicity, lower weight and low cost. This is connected directly to the bidirectional power converter back to back, therefore, and provides active and reactive powers to grid. In addition, it is studied the Z-source inverter applied in this topology. However, the implementation of battery energy storage systems in wind power generation systems, currently is very important, due to possibility of the voltage and current fluctuations in the power grid, so these may to stabilisation of current, voltage and frequency on the grid. This system is connected to back to back converter through a DC-DC converter (buck-boost). For the rotor speed control on induction generator, the strategy is based on direct torque control. While, for the grid side converter is employed the technique of voltage oriented control. For the battery bank voltage control is used on DC-link voltage and battery current flow, through PI type controllers. With the new technological developments in the keys of power, DC converters topologies are presented as the Z-source inverter, this type converter fixes some limitations of the converter back to back, with the characteristics of buck-boost voltage, without the use of switching devices, allowed short-circuits on converter, using new modulation techniques, and reduces the amount injected harmonic to power grid. The studies were performed by means of computer simulation techniques using mathematical models of studied system to validate the control strategies employed in different operating conditions. For the simulations was used the computational tool SimPowerSystems R do Matlab/Simulink R .