Modelagem, estimação de parâmetros e método MPPT para módulo fotovoltaicos

Abstract

The use of photovoltaic systems has been increasing in recent years, demanding the need of new researches to reduce costs and to increase efficiency and reliability of new technologies. In this thesis, a new model for the photovoltaic module, a new MPPT (maximum power point trackers) method and an estimator of irradiance and temperature are developed. Also, it is presented a new application of evolutionary algorithm for parameter estimation of the photovoltaic module. The new model is based on the transfer function of the step response for a first order linear system. Therefore, a simple model with few parameters and low computational effort is shown. Based on this new model, a new MPPT method is proposed. This method calculates the voltage for the maximum power, considering the environmental conditions. It presents the advantages of not oscillating around the maximum power point of the photovoltaic module and of not depending on the sampling interval. Along with the new model, an estimator has been developed for the measurement of irradiance and temperature values from the current and voltage output of the photovoltaic module, eliminating the need of the acquisition sensor for these two measures, thus reducing costs. This estimator can work along with multiple MPPT techniques without affecting its operation. For the parameter estimation, it is proposed to apply an evolutionary algorithm inspired by the natural selection and genetic reproduction. The used data to estimate the parameters comes from the manufacturers or from testing. This methodology has the advantage of not requiring data adjusts for different environmental conditions, besides, the use of complex equations to describe the photovoltaic module is not needed in its application. For analysis and validation of all models and proposed methods, simulation programs have been implemented. The results have been compared with the consolidated methodologies from the literature.