Performance Evaluation of Photovoltaic Systems by Use of RBF Network Models

Abstract

This paper deals with the problem of modeling and performance evaluation of the solar photovoltaic systems, based on the available operation data measurements. The main goal is to find out the most significant cause-effect relationships between the available measured inputs and the output power of the system. The problem is not easy one, because quite often the available measurements are correlated and in addition they have an indirect influence on the final output of the system - the output power. The use of the Radial Basis Function (RBF) network models is proposed in this paper. Two different models have been created. One of them needs direct available measurements for the solar radiation and the solar panel temperature to predict the output power. The other is indirect model, which first predicts the solar panel temperature, based on other measurements and then is used in another model for predicting the output power. It is shown in the paper that the indirect model is suitable for performance evaluation and simulation of a solar photovoltaic system under different climate conditions, in different parts of the country. This helps to optimize the most suitable location of the solar system, in order to produce a larger amount of renewable energy.