Artificial neural networks applied for representation of curves current-voltage of photovoltaic modules

The research of alternative forms of energy production became more important ahead a context which the natural resources are scarce. Among the alternative forms, a resource that distinguishes for using a natural power source is the photovoltaic generator; which is capable to convert solar energy into electric energy. The existing tools and models for prediction and evaluation of photovoltaic modules present the inconvenience to need constant measurements of module temperature, irradiance, current, tension and load, and of the estimate of these parameters for simulation objectives. It is necessary to obtain modules models for different range of these parameters values. Due their generalization capacity, ANN has been proposed in this work as an alternative to represent photovoltaic modules. This representation allows determine the current curve through the voltage, solar irradiance and module temperature. Due the net capacity of reconstruct the curves IV of photovoltaic generation process, it is possible getting a valid photovoltaic systems simulator for different operational conditions. Moreover the test platform and the procedures adopted are presented.