Prediction of the Solar Radiation by using Machine Learning Algorithms

Yazıcı K., Geçici E., Şeker Ş.

3rd International Symposium of Scientific Research and Innovative Studies, Balıkesir, Turkey, 15 - 18 March 2023, pp.258-268

  • Publication Type: Conference Paper / Full Text
  • City: Balıkesir
  • Country: Turkey
  • Page Numbers: pp.258-268
  • Yıldız Technical University Affiliated: Yes


Solar radiation emitted as electromagnetic radiation emitted by the sun can be used for heating, conversion to electrical energy and different energy use. In this context, the appropriate periods of solar energy and how much is important. In this way, it can be calculated how much energy can be produced with the existing solar energy. In this study, the rate of radiation emitted from solar energy was predicted by using a dataset including temperature, humidity, pressure, wind direction and speed, and sunrise and sunset variables covering a fourmonth period. In this context, ridge regression, linear regression, lasso, random forest, support vector regressor, decision tree regressor and gradient boosting regressor methods, which are machine learning techniques, were used. In addition to these methods, a method consisting of a combination of bagging tree, random forest and decision tree was added and the best prediction method of solar radiation was tried to be selected among nine different methods. According to the results obtained, it was seen that the best result was obtained with the stacking model consisting of a combination of random forest regressor and decision tree regressor.