Intelligence-Driven Solar Power Prediction: The Impact of Feature Selection and Seasonal Training Approaches on Model Performance
Resumen
This study evaluates machine learning paradigms for photovoltaic power generation forecasting, focusing on the impact of architectural structures and feature selection on predictive stability. Using data from Portugal, we tested multiple combinations of meteorological variables and seasonal training approaches. Results demonstrate that ensemble tree-based methods, particularly Random Forest, offer superior robustness and generalization compared to kernel based and connectionist architectures when handling non-linear environmental transitions. A critical finding reveals that models relying exclusively on historical lags often function as naive predictors, emphasizing the necessity of integrating exogenous meteorological features for effective forecasting. Overall, this research highlights the importance of feature engineering and strategic training in enhancing model performance for intelligent energy management within smart grid contexts.
- Referencias
- Cómo citar
- Del mismo autor
- Métricas
Awad, M., & Khanna, R. (2015). Support vector regression. In Efficient learning machines: Theories, concepts, and applications for engineers and system designers (pp. 67–80). Springer. https://doi.org/10.1007/978-1-4302-5990-9_4
Azevedo, B. F., Rocha, A. M. A., & Pereira, A. I. (2024). Hybrid approaches to optimization and machine learning methods: A systematic literature review. Machine Learning, 113(7), 4055–4097. https://doi.org/10.1007/s10994-023-06467-x
Benti, N. E., Chaka, M. D., & Semie, A. G. (2023). Forecasting renewable energy generation with machine learning and deep learning: Current advances and future prospects. Sustainability, 15(9), Article 7087. https://doi.org/10.3390/su15097087
Bessa, R., Moreira, C., Silva, B., & Matos, M. (2019). Handling renewable energy variability and uncertainty in power system operation. Advances in Energy Systems: The Large-scale Renewable Energy Integration Challenge, 1–26. https://doi.org/10.1002/9781119508311.ch1
Breiman, L. (1999). Random forests - random features (Technical Report No. 567). Department of Statistics, UC Berkeley.
Castro-Gallardo, F., Rabanal-Arabach, J., Rodríguez-Romero, S., Olivares, D., & Fuentealba, E. (2024). Enhancing electric vehicle autonomy with solar energy: A case study of the «Takai Urban» in northern Chile. SiliconPV Conference Proceedings, 2. https://doi.org/10.52825/siliconpv.v2i.1307
Chandwani, C., & Kamalesh, V. N. (2025). Harnessing the Sun: Implementation of statistical, ML and ANN based models for solar forecasting. International Journal of Scientific Research in Science and Technology, 12(2), 631–639. https://doi.org/10.32628/IJSRST251222605
Chen, T., & Guestrin, C. (2016). Xgboost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 785–794). ACM. https://doi.org/10.1145/2939672.2939785
Chodakowska, E., Nazarko, J., & Nazarko, Ł. (2023). Arima models in solar radiation forecasting in different geographic locations. Energies, 16(13), Article 5029. https://doi.org/10.3390/en16135029
Colak, I., Yesilbudak, M., Genc, N., & Bayindir, R. (2015). Multi-period prediction of solar radiation using ARMA and ARIMA models. In 2015 IEEE 14th International Conference on Machine Learning and Applications (ICMLA) (pp. 1045–1049). IEEE. https://doi.org/10.1109/ICMLA.2015.33
Gala, Y., Fernández, Á., Díaz, J., & Dorronsoro, J. R. (2016). Hybrid machine learning forecasting of solar radiation values. Neurocomputing, 176, 48–59. https://doi.org/10.1016/j.neucom.2015.02.078
Jebli, I., Belouadha, F.-Z., Kabbaj, M. I., & Tilioua, A. (2021). Deep learning based models for solar energy prediction. Advances in Science, Technology and Engineering Systems Journal, 6(1), 349–355. https://doi.org/10.25046/aj060140
Khan, S., Ahmad, A., Ahmad, F., Shafaati Shemami, M., Saad Alam, M., & Khateeb, S. (2018). A comprehensive review on solar powered electric vehicle charging system. Smart Science, 6(1), 54–79. https://doi.org/10.1080/23080477.2017.1419054
Kriegeskorte, N. (2015). Deep neural networks: A new framework for modeling biological vision and brain information processing. Annual Review of Vision Science, 1(1), 417–446. https://doi.org/10.1146/annurev-vision-082114-035447
Kumhar, R., Sarkar, A., Mahato, T. K., & Lakra, P. (2024). Grid integrated renewable energy resources with battery storage support for optimized EV charging. In 2024 3rd International Conference on Power Electronics and IoT Applications in Renewable Energy and its Control (PARC) (pp. 453–458). IEEE. https://doi.org/10.1109/PARC59193.2024.10486230
Li, Z., Rahman, S. M. M., Vega, R., & Dong, B. (2016). A hierarchical approach using machine learning methods in solar photovoltaic energy production forecasting. Energies, 9(1), Article 55. https://doi.org/10.3390/en9010055
Nadeem, A., Hanif, M. F., Naveed, M. S., Hassan, M. T., Gul, M., Husnain, N., & Mi, J. (2024). AI-driven precision in solar forecasting: Breakthroughs in machine learning and deep learning. AIMS Geosciences, 10(4), 684–734. https://doi.org/10.3934/geosci.2024035
Odejobi, O. A., Alawode, K. O., & Lawal, M. O. (2024). Efficient method for forecasting solar irradiance - a review. Fudma Journal of Sciences, 8(6), 285–298. https://doi.org/10.33003/fjs-2024-0806-2786
Osório, G. J., Matias, J. C., & Catalão, J. P. (2014). Hybrid evolutionary-adaptive approach to predict electricity prices and wind power in the short-term. In 2014 Power Systems Computation Conference (pp. 1–7). IEEE. https://doi.org/10.1109/PSCC.2014.7038453
Sa'ad, A., Zied, H., & Nyoungue, A. (2020). A day-ahead multi-approach machine learning technique for photovoltaic power forecasting. In 2020 9th International Conference on Renewable Energy Research and Application (ICRERA) (pp. 257–262). IEEE. https://doi.org/10.1109/ICRERA49962.2020.9242897
Teixeira, B., Pinto, T., Catarino, P., Vasco, P., Soares, J., Reis, A., & Barroso, J. (2024). Prediction of electric power generation for green mobility vehicles. In 2024 22nd International Conference on Intelligent Systems Applications to Power Systems (ISAP) (pp. 1–6). IEEE. https://doi.org/10.1109/ISAP63260.2024.10744261
Teixeira, B., Pinto, T., Faria, P., & Vale, Z. (2021). Pv generation forecasting model for energy management in buildings. In EPIA Conference on Artificial Intelligence (pp. 176–182). Springer. https://doi.org/10.1007/978-3-030-86230-5_14
Theocharides, S., Makrides, G., Georghiou, G. E., & Kyprianou, A. (2018). Machine learning algorithms for photovoltaic system power output prediction. In 2018 IEEE International Energy Conference (ENERGYCON) (pp. 1–6). IEEE. https://doi.org/10.1109/ENERGYCON.2018.8398737
Tina, G. M., Ventura, C., Ferlito, S., & De Vito, S. (2021). A state-of-art-review on machine-learning based methods for PV. Applied Sciences, 11(16), Article 7550. https://doi.org/10.3390/app11167550
Umair, M., Hidayat, N. M., Sukri Ahmad, A., Nik Ali, N. H., Mawardi, M. I. M., & Abdullah, E. (2024). A renewable approach to electric vehicle charging through solar energy storage. PLoS ONE, 19(2), Article e0297376. https://doi.org/10.1371/journal.pone.0297376
Vennila, C., Titus, A., Sudha, T. S., Sreenivasulu, U., Reddy, N. P. R., Jamal, K., Lakshmaiah, D., Jagadeesh, P., & Belay, A. (2022). Forecasting solar energy production using machine learning. International Journal of Photoenergy, 2022(1), Article 7797488. https://doi.org/10.1155/2022/7797488
Vidal, R., Zhu, Z., & Haeffele, B. D. (2022). Optimization landscape of neural networks. Mathematical Aspects of Deep Learning, 1, 200. Cambridge University Press. https://doi.org/10.1017/9781009025096.005
Wamburu, J., Raff, C., Irwin, D., & Shenoy, P. (2020). Greening electric bike sharing using solar charging stations. In Proceedings of the 7th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation (pp. 180–189). https://doi.org/10.1145/3408308.3427621
Wan, C., Zhao, J., Song, Y., Xu, Z., Lin, J., & Hu, Z. (2015). Photovoltaic and solar power forecasting for smart grid energy management. CSEE Journal of Power and Energy Systems, 1(4), 38–46. https://doi.org/10.17775/CSEEJPES.2015.00046
Yang, B., Zhu, T., Cao, P., Guo, Z., Zeng, C., Li, D., Chen, Y., Ye, H., Shao, R., & Shu, H. (2021). Classification and summarization of solar irradiance and power forecasting methods: A thorough review. CSEE Journal of Power and Energy Systems, 9(3), 978–995. https://doi.org/10.17775/CSEEJPES.2020.04930
Zhang, T. (2004). Statistical behavior and consistency of classification methods based on convex risk minimization. The Annals of Statistics, 32(1), 56–85. https://doi.org/10.1214/aos/1079120130
Zhou, Y., & Liang, Y. (2017). Critical points of neural networks: Analytical forms and landscape properties (arXiv:1710.11205). arXiv. https://doi.org/10.48550/arXiv.1710.11205