Hybrid Deep Learning Framework for Short-Term Electricity Generation Forecasting in Türkiye Using Multi-Source Data

Year 2025, Volume: 5 Issue: 3, 152 - 160, 30.10.2025

Hamdullah Karamollaoğlu

https://doi.org/10.5152/tepes.2025.25029

https://izlik.org/JA59UW92YZ

Abstract

The growing complexity of electricity generation, driven by the diversification of energy sources and the integration of renewables, makes accurate short-term forecasting crucial for grid stability and energy security. This study proposes a deep learning-based hybrid forecasting model designed for Türkiye’s dynamic energy landscape. Using hourly electricity production data from December 1, 2019, to March 1, 2025, sourced from the EPİAŞ Transparency Platform, the model analyzes generation patterns across 17 different sources, including both fossil fuels and renewables. The proposed architecture combines Long Short- Term Memory networks and Transformer models to effectively capture complex time-dependent relationships in electricity generation. To improve accuracy, preprocessing techniques such as time-based interpolation, normalization, and principal component analysis were applied. Experimental results demonstrate strong forecasting performance, achieving a mean absolute error of 589.50, a root mean squared error of 762.41, and a coefficient of determination (R2) of 0.98017 for 1-hour ahead predictions, and an R2 of 0.87813 for 1-day ahead predictions. These findings underline the model’s potential to support operational planning, market regulation, and policy-making processes, particularly in emerging economies with dynamic and heterogeneous energy infrastructures.

Keywords

Deep learning , electricity generation , short-term forecasting

References

• 1. J. Zhang, “Energy access challenge and the role of fossil fuels in meeting electricity demand: Promoting renewable energy capacity for sustainable development,” Geoscience Frontiers, vol. 15, no. 5, p. 101873, 2024.
• 2. M. M. Islam et al., “Improving reliability and stability of the power systems: A comprehensive review on the role of energy storage systems to enhance flexibility,” IEEE Access, 2024.
• 3. S. Shahzad, and E. Jasińska, “Renewable revolution: A review of strategic flexibility in future power systems,” Sustainability, vol. 16, no. 13, p. 5454, 2024.
• 4. G.-F. Fan, Y.-Y. Han, J.-W. Li, L.-L. Peng, Y.-H. Yeh, and W.-C. Hong, “A hybrid model for deep learning short-term power load forecasting based on feature extraction statistics techniques,” Expert Syst. Appl., vol. 238, p. 122012, 2024.
• 5. S. Luo, B. Wang, Q. Gao, Y. Wang, and X. Pang, “Stacking integration algorithm based on CNN-BiLSTM-Attention with XGBoost for short-term electricity load forecasting,” Energy Rep., vol. 12, pp. 2676–2689, 2024.
• 6. A. I. Osman et al., “Cost, environmental impact, and resilience of renewable energy under a changing climate: A review,” Environ. Chem. Lett., vol. 21, no. 2, pp. 741–764, 2023.
• 7. S. Mertens, “Design of wind and solar energy supply, to match energy demand,” Cleaner Eng. Technol., vol. 6, p. 100402, 2022.
• 8. M. Bilgili, and E. Pinar, “Gross electricity consumption forecasting using LSTM and SARIMA approaches: A case study of Türkiye,” Energy, vol. 284, p. 128575, 2023.
• 9. D. Saygin, O. B. Tör, M. E. Cebeci, S. Teimourzadeh, and P. Godron, “Increasing Turkey’s power system flexibility for grid integration of 50% renewable energy share,” Energy Strategy Rev., vol. 34, p. 100625, 2021.
• 10. A. Telli, S. Erat, and B. Demir, “Comparison of energy transition of Turkey and Germany: Energy policy, strengths/weaknesses and targets,” Clean Technol. Environ. Policy, vol. 23, no. 2, pp. 413–427, 2021.
• 11. H. Tutar, and M. Atas, “A review on turkey’s renewable energy potential and its usage problems,” Int. J. Energy Econ. Policy, vol. 12, no. 4, pp. 1–9, 2022.
• 12. “Transparency Platform of the Energy Exchange Istanbul (EPİAŞ).” [Online]. Available: seffafli k.epias. com.tr.
• 13. S. Inal, Ö. Yasar, and K. Aydıner, “Importance of domestic coal (lignite) reserves on Turkey’s energy independency,” MT Bilimsel, no. 19, pp. 11–32, 2021.
• 14. K. Kaygusuz, “Renewable and sustainable energy use in Turkey: Areview,” Renew. Sustain. Energy Rev., vol. 6, no. 4, pp. 339–366, 2002.
• 15. C. Kul, L. Zhang, and Y. A. Solangi, “Assessing the renewable energy investment risk factors for sustainable development in Turkey,” J. Cleaner Prod., vol. 276, p. 124164, 2020.
• 16. A. Sherstinsky, “Fundamentals of recurrent neural network (RNN) and long short-term memory (LSTM) network,” Phys. D Nonlinear Phenom., vol. 404, p. 132306, 2020.
• 17. X. Wang, J. Dai, and X. Liu, “A spatial-temporal neural network based on ResNet-Transformer for predicting railroad broken rails,” Adv. Eng. Inform., vol. 65, p. 103126, 2025.
• 18. X. Li, Z. Wang, C. Yang, and A. Bozkurt, “An advanced framework for net electricity consumption prediction: Incorporating novel machine learning models and optimization algorithms,” Energy, vol. 296, p. 131259, 2024.
• 19. S. Aslam, H. Herodotou, S. M. Mohsin, N. Javaid, N. Ashraf, and S. Aslam, “A survey on deep learning methods for power load and renewable energy forecasting in smart microgrids,” Renew. Sustain. Energy Rev., vol. 144, p. 110992, 2021.
• 20. N. Saxena et al., “Hybrid KNN-SVM machine learning approach for solar power forecasting,” Environ. Chall., vol. 14, p. 100838, 2024.
• 21. M. Sawant et al., “A selective review on recent advancements in long, short and ultra-short-term wind power prediction,” Energies, vol. 15, no. 21, p. 8107, 2022.
• 22. Y. Zhang, L. Cui, and W. Yan, “Integrating Kolmogorov–Arnold networks with time series prediction framework in electricity demand forecasting,” Energies, vol. 18, no. 6, p. 1365, 2025.
• 23. S. Wang, J. Shi, W. Yang, and Q. Yin, “High and low frequency wind power prediction based on Transformer and BiGRU-Attention,” Energy, vol. 288, p. 129753, 2024.
• 24. X. Xiang, X. Li, Y. Zhang, and J. Hu, “A short-term forecasting method for photovoltaic power generation based on the TCN-ECANet-GRU hybrid model,” Sci. Rep., vol. 14, no. 1, p. 6744, 2024.
• 25. M. Güldürek, “Short-term wind speed prediction using a hybrid artificial intelligence approach based on dragonfly algorithm: A case study of the Mediterranean climate,” TEPES, vol. 4, No. 2, pp. 84–95, 2024.
• 26. M. S. Ibrahim, S. M. Gharghory, and H. A. Kamal, “A hybrid model of CNN and LSTM autoencoder-based short-term PV power generation forecasting,” Electr. Eng., vol. 106, no. 4, pp. 4239–4255, 2024.
• 27. W. G. Buratto, R. N. Muniz, A. Nied, C. F. de O. Barros, R. Cardoso, and G. V. Gonzalez, “Wavelet CNN‐LSTM time series forecasting of electricity power generation considering biomass thermal systems,” IET Generation Trans. & Dist., vol. 18, no. 21, pp. 3437–3451, 2024.
• 28. Q. Wu, F. Guan, C. Lv, and Y. Huang, “Ultra‐short‐term multi‐step wind power forecasting based on CNN‐LSTM,” IET Renew. Power Gener., vol. 15, no. 5, pp. 1019–1029, 2021.
• 29. S. Patro, and K. K. Sahu, “Normalization: A preprocessing stage,” arXiv Preprint ArXiv:1503.06462, 2015.
• 30. T. Kurita, “Principal component analysis (PCA),” in Computer Vision: A Reference Guide, K. Ikeuchi, Ed. Cham: Springer International Publishing, 2021, pp. 1013–1016.
• 31. S. Wang, and F. Xiao, “AHU sensor fault diagnosis using principal component analysis method,” Energy Build., vol. 36, no. 2, pp. 147–160, 2004.
• 32. S. P. Mishra et al., “Multivariate statistical data analysis-principal component analysis (PCA),” Int. J. Livest. Res., vol. 7, no. 5, pp. 60–78, 2017.
• 33. H. Abdi, and L. J. Williams, “Principal component analysis,” WIREs Computational Stats, vol. 2, no. 4, pp. 433–459, 2010.
• 34. G. Van Houdt, C. Mosquera, and G. Nápoles, “A review on the long short-term memory model,” Artif. Intell. Rev., vol. 53, no. 8, pp. 5929–5955, 2020.
• 35. J. Cheng, L. Dong, and M. Lapata, “Long short-term memory-networks for machine reading,” arXiv Preprint ArXiv:1601.06733, 2016.
• 36. T. Fischer, and C. Krauss, “Deep learning with long short-term memory networks for financial market predictions,” Eur. J. Oper. Res., vol. 270, no. 2, pp. 654–669, 2018.
• 37. Q. Zhang, C. Qin, Y. Zhang, F. Bao, C. Zhang, and P. Liu, “Transformerbased attention network for stock movement prediction,” Expert Syst. Appl., vol. 202, p. 117239, 2022.
• 38. T. P. Nguyen, H. Nam, and D. Kim, “Transformer-based attention network for in-vehicle intrusion detection,” IEEE Access, vol. 11, pp. 55389–55403, 2023.
• 39. S. Reza, M. C. Ferreira, J. J. M. Machado, and J. M. R. Tavares, “A multihead attention-based transformer model for traffic flow forecasting with a comparative analysis to recurrent neural networks,” Expert Syst. Appl., vol. 202, p. 117275, 2022.
• 40. H. Chen, D. Jiang, and H. Sahli, “Transformer encoder with multi-modal multi-head attention for continuous affect recognition,” IEEE Trans. Multimedia, vol. 23, pp. 4171–4183, 2021.
• 41. J. Joy, and M. P. Selvan, “A comprehensive study on the performance of different Multi-class Classification Algorithms and Hyperparameter Tuning Techniques using Optuna,” presented at the International Conference on Computing, Communication, Security and Intelligent Systems (IC3SIS), IEEE, 2022, pp. 1–5.
• 42. S. Hanifi, A. Cammarono, and H. Zare-Behtash, “Advanced hyperparameter optimization of deep learning models for wind power prediction,” Renew. Energy, vol. 221, p. 119700, 2024.
• 43. T. Akiba, S. Sano, T. Yanase, T. Ohta, and M. Koyama, Optuna: A Next- Generation Hyperparameter Optimization Framework,” presented at the Proceedings of the 25th ACM SIGKDD international conference on knowledge discovery & data mining, pp. 2623–2631, 2019.
• 44. C. Banerjee, T. Mukherjee, and E. Pasiliao Jr., “An empirical study on generalizations of the ReLU activation function.” ACM Southeast Conference, 2019, pp. 164–167.
• 45. T. O. Hodson, T. M. Over, and S. S. Foks, “Mean squared error, deconstructed,” J. Adv. Model. Earth Syst., vol. 13, no. 12, p, MS002681, e2021, 2021.
• 46. S. Y. Y. Hamad, T. Ma, and C. Antoniou, “Analysis and prediction of bikesharing traffic flow–Citi bike, New York.”,” presented at the 7th International Conference on Models and Technologies for Intelligent Transportation Systems (MT-ITS), IEEE, 2021, pp. 1–8.
• 47. D. Chicco, M. J. WARRENS, and G. Jurman, “The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation,” PeerJ Comput. Sci., vol. 7, p. e623, 2021.