Solar energy analysis and temperature forecast with artificial neural networks for Bilecik province

Nowadays, to be used in all areas of energy and increasing energy demand, increased use and the search for alternative energy sources.Almost all of the climatic conditions of the alternative energy sources to be effective, ( like wind, solar, bioenergy, geothermal etc..) renewable energy sources, has taken its place in history as the most important energy sources. In this study, 2013 year's hourly average wind speed, air pressure, water vapor pressure, the relative humidity and temperature measurements were taken 1794 pieces from the State Meteorology Affairs General Directorate of Bilecik province.For the hourly average air temperature was used feedforward back propagation multi-layered (ANN) Artifical Neural Network Model. And for Artifical Neural Network training, 80% (1435 pieces) of measurement received wind speed, air pressure,the water vapor pressure and relative humidity measurement was used the input layer, and as the output layer the air temperature was taken.The rest of 20% of the input data was used as a test set of Artificial Neural Networks.For 2013 year's 20% of data obtained from the Artificial Neural Networks model of average temperature data for Bilecik province were compared with the actual temperature data of the same year again. As a result, the results of our ANN model's RMSE and MSE values are quite satisfactory, it would be appropriate for the later years of our average hourly temperature of ANN model was seen.