Abstract:

The useful energy obtained from solar energy in both thermal and electrical applications depends on the radiation coming to the surface and the amount of this radiation varies according to location, time, climatic conditions and environmental conditions. Determination correctly way of the amount of radiation is important for the efficient design and operation of solar-powered energy systems. In this study, an Artificial Neural Network (ANN) model was developed for solar radiation estimation by using MATLAB application for four cities (Kayseri, Rize, Hakkari and İzmir) located in different climatic zones of Türkiye. While estimating the solar radiation, the effect of the geographical data of the cities consisting of latitude, longitude, and altitude, as well as the effect of meteorological data such as average temperature, relative humidity, cloudiness, and vapor pressure are also discussed. Considering all these factors, the best estimation result in the model established is Hakkari with R2=0.93, followed by İzmir, Kayseri and Rize cities, respectively. For each city, it was seen that there was a correspondence between the values estimated by ANN and the measured values.

Keywords: