Neural Network Models for Predicting the electrical behavior of the PV systems in a grid connected application

Solar energy is considered as the most important source of renewable energy. Considering that the average amount of energy received in the year about KWH/M2 2000 and the number of sunny hours is more than 2800 hours per year, This paper define a circuit-based simulation model for a PV cell in order to allow estimate the electrical behavior of the cell with respect changes on environmental parameter of temperature and irradiance. The general model was implemented and accepts irradiance and temperature as variable parameters and outputs the I-V characteristic. The final objective is develops a general model to simulate the electrical behavior of the PV systems in a grid connected application. The obtained results confirm the superiority of the RBF technique over the MLP technique in most of the cases, namely, some models with deterministic coefficients near 90 % and low MBE, MAPE and RMSE values.

Keywords:

photovoltaic module , I-V characteristic , ANN models , grid connected , solar panels