Abstract:

The output power obtained from PV systems depends mainly on solar radiation and other atmospheric conditions. In this study, meteorological data measured in Hakkari province in the Southeast of Turkey has been evaluated as input parameters in the Artificial Neural Networks (ANN) model, which is widely used in the literature in forecasting studies, and the prediction values of solar radiation have been determined at the output of this model. Matlab/Simulink model of PV system with incremental conductivity maximum power point tracking (MPPT) algorithm applied to DC–DC boost power electronics converter has been considered to determine the maximum power under different atmospheric conditions. Output powers of the PV system in Matlab/Simulink environment have been calculated by considering the real solar radiation, ambient temperature and the solar radiation values estimated in the ANN model separately. First, the actual solar radiation and ambient temperature values and then the predicted solar radiation and ambient temperature values have been handled to compare the output powers in the relevant PV system model. Comparison results have been calculated with evaluation metrics commonly used in the literature, and successful results have been obtained with a determination coefficient (R2) value of 0.9705 for solar radiation and 0.9668 for PV system output power.

Keywords: