Forecasting Solar Photovoltaic Power Generation with the Use of Machine Learning Algorithms
J. Environ. Nanotechnol., Volume 15, No 1 (March 2026) pp. 531-542
Abstract
Estimation of solar photovoltaic (PV) power generation is necessary to predict accurately the further functioning of the grid, energy planning, and massive integration of renewable energy sources. Nevertheless, dynamic nature of PV systems, intermittent solar irradiance, and meteorological conditions which change at a very high rate all render it challenging to have a reliable prediction. This paper proposes a Hybrid Machine Learning Approach of a Dynamic Weighting Mechanism (DWM-HMLA) in the short-term solar PV power prediction. The presented framework combines various heterogeneous machine learning models into an ensemble structure, in which a dynamic weighting scheme dynamically modulates the contribution of individual base learners in real-time according to the current environmental factors and recent errors in prediction. The proposed DWM-HMLA was tested with the help of the real-life solar PV data taken in various geographical areas and under various meteorological conditions. The proposed architecture is a combination of Support Vector Regression (SVR), Random Forest (RF), and Artificial Neural Network (ANN) models in an ensemble architecture. It has been experimentally demonstrated that the proposed method attains a Root Mean Square Error (RMSE) of 0.214 kW, a Mean Absolute Error (MAE) of 0.167 kW, and a coefficient of determination (R2 ) of 0.962. The forecasting accuracy of the proposed approach is 12.8% and 15.3% higher in the average of RMSE and MAE respectively. The adaptive weighting mechanism allows the non-linear time patterns to be processed in an effective manner and also ensures robustness when the weather conditions change rapidly. The findings indicate that DWM-HMLA is a stable and scaled forecasting platform, which can be used to integrate renewable energy, balance grids, and energy market. Proper predictions made by the suggested model can help the operators of the systems and policymakers to make the most out of their resource allocation and minimize reliance on standard power generation. The results of forecasting the proposed DWM-HMLA can be implemented directly in the short-term operational decision making in grid management whereby precision intraday solar PV power estimates are needed to adjust unit commitment to the reserve commitment and real-time load balancing.
Full Text
Reference
Ağbulut, Ü., Gürel, A. E. and Biçen, Y., Prediction of daily global solar radiation using different machine learning algorithms: Evaluation and comparison, Renewable Sustainable Energy Rev., 135, 110114(2021).
https://doi.org/10.1016/j.rser.2020.110114
Akhter, M. N., Mekhilef, S., Mokhlis, H. and Mohamed, S. N., Review on forecasting of photovoltaic power generation based on machine learning and metaheuristic techniques, IET Renewable Power Generat., 13(7), 1009–1023(2019).
https://doi.org/10.1049/iet-rpg.2018.5649
De Freitas, V. G. and Alves-Souza, S. N., A Systematic Literature Review on big data for solar photovoltaic electricity generation forecasting, Sustain. Energy Technol. Assess., 31, 54–63(2019).
https://doi.org/10.1016/j.seta.2018.11.008
Feng, Y., Hao, W., Li, H., Cui, N., Gong, D. and Gao, L., Machine learning models to quantify and map daily global solar radiation and photovoltaic power, Renewable Sustain. Energy Rev., 118, 109393(2020).
https://doi.org/10.1016/j.rser.2019.109393
Huang, C. J. and Kuo, P. H., Multiple-input Deep Convolutional Neural Network Model for Short-Term Photovoltaic Power Forecasting, IEEE Access, 7, 74822–74834(2019).
https://doi.org/10.1109/access.2019.2921238
Jebli, I., Belouadha, F. Z., Kabbaj, M. I. and Tilioua, A., Prediction of solar energy guided by pearson correlation using machine learning, Energy, 224, 120109(2021).
https://doi.org/10.1016/j.energy.2021.120109
Khan, W., Walker, S. and Zeiler, W., Improved solar photovoltaic energy generation forecast using deep learning-based ensemble stacking approach, Energy, 240, 122812(2022).
https://doi.org/10.1016/j.energy.2021.122812
Khandakar, A., Chowdhury, E. H. M., Khoda, K. M., Benhmed, K., Touati, F., Al-Hitmi, M. and Gonzales, S. P. J. A., Machine Learning Based Photovoltaics (PV) Power Prediction Using Different Environmental Parameters of Qatar, Energies, 12(14), 2782(2019).
https://doi.org/10.3390/en12142782
Kuzlu, M., Cali, U., Sharma, V. and Guler, O., Gaining Insight into Solar Photovoltaic Power Generation Forecasting Utilizing Explainable Artificial Intelligence Tools, IEEE Access, 8, 187814–187823(2020).
https://doi.org/10.1109/access.2020.3031477
Li, G., Xie, S., Wang, B., Xin, J., Li, Y. and Du, S., Photovoltaic Power Forecasting with a Hybrid Deep Learning Approach, IEEE Access, 8, 175871–175880(2020).
https://doi.org/10.1109/access.2020.3025860
Luo, X., Zhang, D. and Zhu, X., Deep learning-based forecasting of photovoltaic power generation by incorporating domain knowledge, Energy, 225, 120240(2021).
https://doi.org/10.1016/j.energy.2021.120240
Markovics, D. and Mayer, M. J., Comparison of machine learning methods for photovoltaic power forecasting based on numerical weather prediction, Renewable Sustainable Energy Rev., 161, 112364(2022).
https://doi.org/10.1016/j.rser.2022.112364
Mellit, A., Massi, P. A., Ogliari, E., Leva, S. and Lughi, V., Advanced Methods for Photovoltaic Output Power Forecasting: A Review, Appl. Sci., 10(2), 487(2020).
https://doi.org/10.3390/app10020487
Mellit, A., Pavan, A. M. and Lughi, V., Deep learning neural networks for short-term photovoltaic power forecasting, Renewable Energy, 172, 276–288(2021).
https://doi.org/10.1016/j.renene.2021.02.166
Rajagukguk, R. A., Ramadhan, R. A. A. and Lee, H. J., A Review on Deep Learning Models for Forecasting Time Series Data of Solar Irradiance and Photovoltaic Power, Energies, 13(24), 6623(2020).
https://doi.org/10.3390/en13246623
Sharifzadeh, M., Sikinioti-Lock, A. and Shah, N., Machine-learning methods for integrated renewable power generation: A comparative study of artificial neural networks, support vector regression, and Gaussian Process Regression, Renewable Sustainable Energy Rev., 108, 513–538(2019).
https://doi.org/10.1016/j.rser.2019.03.040
Sun, Y., Venugopal, V. and Brandt, A. R., Short-term solar power forecast with deep learning: Exploring optimal input and output configuration, Solar Energy, 188, 730–741(2019).
https://doi.org/10.1016/j.solener.2019.06.041
Theocharides, S., Makrides, G., Livera, A., Theristis, M., Kaimakis, P. and Georghiou, G. E., Day-ahead photovoltaic power production forecasting methodology based on machine learning and statistical post-processing, Appl. Energy, 268, 115023(2020).
https://doi.org/10.1016/j.apenergy.2020.115023
Wang, K., Qi, X. and Liu, H., A comparison of day-ahead photovoltaic power forecasting models based on deep learning neural network, Appl. Energy, 251, 113315(2019).
https://doi.org/10.1016/j.apenergy.2019.113315
Yagli, G. M., Yang, D. and Srinivasan, D., Automatic hourly solar forecasting using machine learning models, Renewable Sustain. Energy Rev., 105, 487–498(2019).
https://doi.org/10.1016/j.rser.2019.02.006
Zhou, Y., Zhou, N., Gong, L. and Jiang, M., Prediction of photovoltaic power output based on similar day analysis, genetic algorithm and extreme learning machine, Energy, 204, 117894(2020).
