Estimation of hourly global solar radiation using artificial neural network in Adana province, Turkey

Since global solar radiation (GSR) is an important parameter for the design, installation, and operation of energy-based systems, it to have precise information about it. As indicating devices are expensive their requirements such as maintenance should be carried out, measurement cannot frequently taken. On other hand, measurements different meteorological parameters relative humidity, sunshine duration, ground surface temperature more prevalent in meteorology stations. Therefore, estimation a significant areas where could not performed complete missing databases. Many models, software, simulation programs utilized calculate data, provide economic advantage, obtain high accuracy. The main purpose this study perform Adana, on east Mediterranean Turkey, by using artificial neural network (ANN) model. best performance obtained optimizing neuron numbers used network’s hidden layer with trial error method. With aim, hourly data including wind speed, direction, actual pressure, average taken inputs while target. All these which 2018, has from Turkish State Meteorological Service. A linear correlation coefficient value been 0.87313 mean square (MSE) 1.6184x104 W/m2 testing set. ANN’s testing/validation results show that low MSE, accuracy adequacy Besides, predicted ANN output evaluated remarkably close measured target considering coefficient.