Collaborative swarm intelligence to estimate PV parameters

• Autor(a) principal: Nunes, H.G.G.
• Data de Publicação: 2019
• Outros Autores: Pombo, José Álvaro Nunes, Bento, P.M.R., Mariano, S., Calado, M. Do Rosário
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10400.6/7051
• topic: Collaborative swarm intelligence; Hybrid metaheuristic; Parameter estimation; Single-diode model; Double-diode model

To properly evaluate, control and optimize photovoltaic (PV) systems, it is crucial to accurately estimate the equivalent electric circuit parameters from the respective mathematical models that characterize the PV cells or modules behavior. This is currently a hot research topic that has attracted the attention of numerous researchers. In this paper, we propose a new hybrid methodology that combines diversification and intensification mechanisms from different metaheuristics (MHs) to estimate PV parameters precisely. The proposed methodology has the capacity to adapt to the specific optimization problem and maintain diversity when building solutions, thus mitigating premature convergence and population stagnation. This methodology can incorporate several MHs (two or more swarms) with different potentialities, enabling a good balance between diversification and intensification mechanisms. Furthermore, it is able to explore a multidimensional search space in different regions simultaneously. To validate its performance, the proposed methodology was compared with other wellestablished MHs in several benchmark functions, and used to estimate PV parameters in single and double-diode models in two case studies, the first using standard literature data, and the second using measured data from a real application with and without the occurrence of partial shading. The proposed methodology was able to find highly accurate solutions with reduced computational cost and high reliability. Comparisons with the other MHs demonstrate that the proposed methodology presents a very competitive performance when solving the PV parameter estimation problem.