№2, 2023


Javad V. Najafli

This research paper explores the prediction of solar energy radiation using various machine learning methods and neural networks. The results are presented based on the analysis of four different datasets obtained from solar stations. The study begins with an overview of solar energy in the context of contemporary challenges in the fields of energy and environmental sustainability, and reviews previous research related to the application of artificial intelligence in solar energy. The main contribution of the work lies in the analysis and comparison of diverse machine learning models and neural networks for predicting solar energy radiation. The results are compared considering accuracy metrics (RMSE - Root Mean Squared Error, MAE - Mean Absolute Error, MRE - Mean Relative Error) and execution times for each model. Each model is evaluated on four datasets with different characteristics (pp.32-36).

Keywords: Solar energy, Artifical Intelligence, Forecasting, Solar power generation, Energy sustainability
  • Alzahrani, A., Shamsi, P., Dagli, C., & Ferdowsi, M. (2017). Solar irradiance forecasting using deep neural networks. Procedia Computer Science, 114, 304-313.
  • Aydin, U. (2019). Energy insecurity and renewable energy sources: Prospects and challenges for Azerbaijan.
  • Biau, G., & Scornet, E. (2016). A random forest guided tour. Test, 25, 197-227.
  • Chen, T., He, T., Benesty, M., Khotilovich, V., Tang, Y., Cho, H., ... & Zhou, T. (2015). Xgboost: extreme gradient boosting. R package version 0.4-2, 1(4), 1-4.
  • Culp, M., Johnson, K., & Michailides, G. (2007). ada: An r package for stochastic boosting. Journal of statistical software, 17, 1-27.
  • Gulaliyev, M. G., Mustafayev, E. R., & Mehdiyeva, G. Y. (2020). Assessment of solar energy potential and its ecological-economic efficiency: Azerbaijan case. Sustainability, 12(3), 1116.
  • Hancock, J. T., & Khoshgoftaar, T. M. (2020). CatBoost for big data: an interdisciplinary review. Journal of big data, 7(1), 1-45.
  • Hemmatabady, H., Welsch, B., Formhals, J., & Sass, I. (2022). AI-based enviro-economic optimization of solar-coupled and standalone geothermal systems for heating
  • https://doi.org/10.3390/su12031116
  • Huseynli, B. (2023). Renewable Solar Energy Resources Potential and Strategy in Azerbaijan. International Journal of Energy Economics and Policy, 13(1), 31.
  • Ibrahimov, R. (2010). Azerbaijan energy strategy and the importance of the divercification of exported transport routes. Journal of Qafqaz University, (29).
  • Khodayar, M., Mohammadi, S., Khodayar, M. E., Wang, J., & Liu, G. (2019). Convolutional graph autoencoder: A generative deep neural network for probabilistic spatio-temporal solar irradiance forecasting. IEEE Transactions on Sustainable Energy, 11(2), 571-583.
  • Liu, D., & Sun, K. (2019). Random forest solar power forecast based on classification optimization. Energy, 187, 115940.
  • Mustafayev, F. (2021). The potential role of renewable energy in providing energy security of Azerbaijan. Unpublished. Ph. D. Thesis, University of Gdansk, Sopot, Poland. https://doi.org/10.37474/0365-8554/2020-6-7-57-62 
  • Natekin, A., & Knoll, A. (2013). Gradient boosting machines, a tutorial. Frontiers in neurorobotics, 7, 21.
  • Vidadili, N., Suleymanov, E., Bulut, C., & Mahmudlu, C. (2017). Transition to renewable energy and sustainable energy development in Azerbaijan. Renewable and Sustainable Energy Reviews, 80, 1153-1161.
  • Weisberg, S. (2005). Applied linear regression (Vol. 528). John Wiley & Sons.
  • Zargar, S. (2021). Introduction to sequence learning models: RNN, LSTM, GRU. Department of Mechanical and Aerospace Engineering, North Carolina State University, Raleigh, North Carolina, 27606.
  • https://www.kaggle.com/datasets/ibrahimkiziloklu/solar-radiation-dataset
  • https://www.kaggle.com/datasets/ishaanthareja007/solar-radiation
  • https://www.kaggle.com/datasets/pythonafroz/solar-powe-generation-data
  • https://www.kaggle.com/datasets/vipulgote4/solar-power-generation