Análise comparativa de algoritmos de busca de máxima potência em sistemas de geração distribuída

Abstract

Photovoltaic solar energy is one of the most used sources of renewable energy in the world and one of the fastest growing in terms of generation each year, só it is essential that photovoltaic modules generate the maximum power possible, given the constructive characteristics of each module. The main objective of this work is to show, through simulations, the limitation of simple techniques of maximum power point tracking in a photovoltaic module operating under shading conditions. In this context, concepts of local maximum and global maximum in functions are approached, aiming to facilitate the understanding of the reasons that lead simple algorithms to diverge in the search for maximum power generation in a photovoltaic module. As alternatives to the classic tracking methods, results of simulations with deterministic and heuristic methods of maximum power search are presented. The methods are compared by calculating the mean square error and the steady state error. With the results of the simulations, it was concluded that deterministic algorithms, excepting the Levenberg-Mardquardt method, have the same limitations as the classical methods. The heuristic maximum power point tracking methods were able to achieve satisfactory results, better than the other methods, therefore they can be used as an alternative to the classical tracking methods.