This paper presents results of research aimed at improving efficiency of coordinated on-line control of system voltage and reactive power on the basis of developing models, methods, and a control system, integrating traditional numerical techniques with fuzzy logic technology. An approach associated with experimental design is considered to construct diverse types of sensitivity indices, which serve as the basis for multi-attribute evaluating efficiency of control actions and, therefore, for forming rules included in knowledge base. Questions of the construction and tuning of linguistic variables as well as rational implementation of fuzzy logic procedures are discussed. 0-7803-5446-X/99 $10.00 0 1999 IEEE

Resumo -Este artigo apresenta a implementação de um sistema fotovoltaico (PV) monofásico conectado à rede elétrica com duplo estágio de conversão de energia, o qual não é afetado por problemas associados a sombreamentos parciais. Técnicas convencionais de rastreamento de máxima potência, tais como perturba e observa, não rastreia o ponto de máxima potência global (GMPP) quando o arranjo PV opera sob condições de sombreamento parcial. Como resultado, o desempenho geral do sistema diminui. Este trabalho usa o método de otimização por enxame de partículas para rastrear o GMPP, aumentando o desempenho do sistema PV. Simultaneamente, o sistema fotovoltaico é controlado para realizar o condicionamento ativo de potência, realizando supressão das correntes harmônicas e compensação da potência reativa da carga. Além disso, como principal contribuição deste trabalho, é proposta uma malha de controle feed-forward para reduzir variações bruscas de tensão do barramento CC na ocorrência de transientes de radiação solar e/ou temperatura. Tais variações podem influenciar no desempenho do sistema PV bem como ocasionar danos às chaves de potência. Os desempenhos estático e dinâmico do sistema PV são avaliados por meio de resultados experimentais. Palavras-Chave PSO-BASED MPPT TECHNIQUE APPLIED TO A GRID-TIED PV SYSTEM WITH ACTIVE POWER LINE CONDITIONING USING A FEED-FORWARD DC-BUS CONTROL LOOP1 Abstract -This paper presents a double-stage conversion single-phase grid-tied photovoltaic (PV) system implementation, which is not affected to problems associated to mismatching phenomena, such as partial shading. Usually, conventional maximum power point tracking techniques, such as perturb & observe, do not track the global maximum power point (GMPP) when the PV array is operating under partial shading. As result, the overall PV system performance decreases. This paper uses Artigo submetido em 18/01/2016. Primeira revisão em 16/03/2016. Aceito para publicação em 11/04/2016 por recomendação do Editor Marcelo Cabral Cavalcanti.particle swarm optimization algorithm to track the GMPP, increasing the PV system performance. Simultaneously, the PV system is controlled to carry out the active conditioning, suppressing load harmonic currents and compensating reactive power. In addition, as main contribution of this paper, a feed-forward control loop is proposed to reduce sudden variations of the DCbus voltage, in occurrence of solar radiation and/or temperature transients. Such variations can influence the PV system performance and can also cause damage in the switching devices. The static and dynamic PV system performances are evaluated through experimental results.

No abstract

This paper presents results of research aimed at improving the efficiency of coordinated on-line control of voltage and reactive power on the basis of developing models, methods, and a control system integrating traditional numerical methods with fuzzy logic technology. An approach based on experimental design is applied to construct diverse types of sensitivity indices, which serve for evaluating efficiency of control actions and, therefore, for constructing fuzzy production rules included in knowledge base. The rules considered in the paper are applicable for voltage and reactive power control in regulated and deregulated environments. Questions of constructing linguistic variables and tuning a fuzzy model are discussed. Some aspects of rational carrying out fuzzy logic procedures are considered.Index Terms--experimental design, fuzzy logic procedures, fuzzy model tuning, fuzzy production rules, regulated and deregulated environments, sensitivity indices, voltage and reactive power control.

No abstract