A Hybrid Physics-Informed Machine Learning Model for Renewable Energy Forecasting and Optimization
DOI:
https://doi.org/10.63876/ijtm.v5i1.178Keywords:
Renewable Energy Forecasting, Physics-Informed Machine Learning, Hybrid Modeling, Energy Optimization, Smart Grid, Sustainable EnergyAbstract
The rapid expansion of renewable energy systems requires accurate forecasting and optimization models to address the inherent uncertainty, variability, and intermittency of energy generation. Conventional data-driven machine learning models have shown strong predictive performance; however, they often require large datasets and may fail to represent the physical behavior of renewable energy systems. This study proposes a hybrid physics-informed machine learning model for renewable energy forecasting and optimization by integrating physical system constraints with data-driven learning algorithms. The proposed model combines historical energy production data, meteorological variables, and domain-specific physical principles, such as solar irradiance behavior, wind speed dynamics, power conversion efficiency, and energy balance constraints, to improve forecasting accuracy and operational reliability. The framework is designed to predict renewable energy output, reduce forecasting error, and support optimal energy dispatch decisions. Machine learning algorithms are used to capture nonlinear relationships among weather conditions, system performance, and energy generation, while physics-informed constraints guide the model to produce physically consistent predictions. This hybrid approach reduces the risk of unrealistic outputs and improves model generalization under changing environmental conditions. The optimization component further supports decision-making by identifying efficient energy allocation, storage utilization, and grid integration strategies. The proposed approach contributes to renewable energy research by bridging the gap between purely empirical forecasting models and physics-based simulation models. It provides a more robust, interpretable, and reliable forecasting framework for smart energy management systems. The findings are expected to support the development of sustainable, intelligent, and resilient renewable energy infrastructures, particularly in contexts where accurate prediction and optimal resource management are essential for reducing dependence on fossil fuels and improving energy system efficiency.
Downloads
References
[1] R. Li, K. Zhang, and Q. Wang, “A data-driven framework for renewable energy sustainability: Reducing subjective weighting with a real-coded accelerated genetic algorithm and projection pursuit,” Renewable and Sustainable Energy Reviews, vol. 242, p. 117292, Dec. 2026, doi: https://doi.org/10.1016/j.rser.2026.117292.
[2] M. E. A. Elsayed, A. A. A. Attia, M. A. Abdelrahman, E.-A. Attia, and M. R. Rashed, “Design for manufacturing a hydraulic gravity energy storage system: a critical review and six-sigma DMADV framework for sustainable energy storage systems,” Sustainable Energy Technologies and Assessments, vol. 92, p. 105169, Aug. 2026, doi: https://doi.org/10.1016/j.seta.2026.105169.
[3] S. Roy and R. Singh, “Strategic evaluation of time-of-use pricing limits for enhanced renewable energy procurement in radial distribution systems,” Utilities Policy, vol. 102, p. 102289, Oct. 2026, doi: https://doi.org/10.1016/j.jup.2026.102289.
[4] L. Yin and K. Wei, “Large language models driven adaptive proximal policy optimization for clean energy system frequency stability control,” Sustainable Energy Technologies and Assessments, vol. 84, p. 104739, Dec. 2025, doi: https://doi.org/10.1016/j.seta.2025.104739.
[5] E. A. Kaplan, Y. Büyükkör, T. Sarıtaş, and A. Aslan, “Artificial intelligence investment and renewable energy generation: Cross-country evidence from 25 European economies,” Sustainable Production and Consumption, p. S2352550926000916, Jul. 2026, doi: https://doi.org/10.1016/j.spc.2026.07.017.
[6] W. Zhang, X. Pan, Q. Mao, and Z. Zhu, “Coordinating demand response and energy storage for flexibility in renewable-integrated smart grids: A three-tier stochastic optimization framework,” Sustainable Energy, Grids and Networks, p. 102456, Jul. 2026, doi: https://doi.org/10.1016/j.segan.2026.102456.
[7] C. Saletti, P. Conti, E. Schito, M. Morini, A. Gambarotta, and D. Testi, “Scenario-based assessment of electrification pathways in district multi-energy systems,” Energy, vol. 359, p. 141479, Sep. 2026, doi: https://doi.org/10.1016/j.energy.2026.141479.
[8] X. Wu et al., “Unsupervised super-resolution for LWIR polarization images via multimodal guidance and physical constraints,” Optics & Laser Technology, vol. 203, p. 115814, Nov. 2026, doi: https://doi.org/10.1016/j.optlastec.2026.115814.
[9] V. Nakama, J. Nájera, D. S. Y. Rosas, and M. Lafoz, “Development of a dynamic–thermal model for an industrial lithium battery integrated into a photovoltaic system,” Journal of Energy Storage, vol. 169, p. 121989, Aug. 2026, doi: https://doi.org/10.1016/j.est.2026.121989.
[10] H. Li et al., “Physics-informed machine learning of hyporheic exchange driven by channel blockage and riverbed morphodynamics,” Journal of Hydrology, vol. 678, p. 136121, Oct. 2026, doi: https://doi.org/10.1016/j.jhydrol.2026.136121.
[11] A. Kerr-German, A. C. Tas, and A. T. Buss, “A multi-method approach to addressing the toddler data desert in attention research,” Cognitive Development, vol. 65, p. 101293, Jan. 2023, doi: https://doi.org/10.1016/j.cogdev.2022.101293.
[12] E. Mansouri, K. E. Melkemi, D. Michelucci, and B. Mokhtari, “A unified optimization framework for parametric modelers with mixed explicit and black-box constraints,” Computer-Aided Design, vol. 197, p. 104084, Aug. 2026, doi: https://doi.org/10.1016/j.cad.2026.104084.
[13] H. Xu et al., “Deep learning models for efficient geotechnical predictions: reducing training effort and data requirements with transfer learning,” Advanced Engineering Informatics, vol. 69, p. 103852, Jan. 2026, doi: https://doi.org/10.1016/j.aei.2025.103852.
[14] T. Prieß, L. Bornemann, and M. Kaltschmitt, “A simulation-based approach to assess the energy consumption and operational feasibility of hydrogen-powered port equipment,” Energy Conversion and Management, vol. 365, p. 121724, Oct. 2026, doi: https://doi.org/10.1016/j.enconman.2026.121724.
[15] P. Han, W. He, Y. Cao, Y. Li, Q. Mu, and Y. Wang, “Lithium-ion battery health assessment method based on belief rule base with interpretability,” Applied Soft Computing, vol. 138, p. 110160, May 2023, doi: https://doi.org/10.1016/j.asoc.2023.110160.
[16] D. Yang, W. Li, G. M. Yagli, and D. Srinivasan, “Operational solar forecasting for grid integration: Standards, challenges, and outlook,” Solar Energy, vol. 224, pp. 930–937, Aug. 2021, doi: https://doi.org/10.1016/j.solener.2021.04.002.
[17] A. J. Garzón-Orduña, O. E. Coronado-Hernández, A. Arrieta-Pastrana, H. M. Ramos, and M. Pérez-Sánchez, “Physics-informed machine learning for hybrid modelling of water leakages induced by rapid valve manoeuvres in water distribution networks,” Engineering Applications of Artificial Intelligence, vol. 167, p. 113887, Mar. 2026, doi: https://doi.org/10.1016/j.engappai.2026.113887.
[18] H. Wu, M. Gui, and D. Wu, “Physics-Informed hybrid machine learning for critical heat flux prediction: A comparative analysis of modeling approaches,” Nuclear Engineering and Design, vol. 445, p. 114434, Dec. 2025, doi: https://doi.org/10.1016/j.nucengdes.2025.114434.
[19] N. Frieß, U. Pferschy, D. Raese, and J. Schauer, “Assessing the potential of forecast-based optimization in renewable energy communities with flexible electricity, heat and mobility resources,” Applied Energy, vol. 401, p. 126664, Dec. 2025, doi: https://doi.org/10.1016/j.apenergy.2025.126664.
[20] J. Liang et al., “Dynamic feature selection improves influenza forecasting accuracy and generalization across countries,” Journal of Advanced Research, p. S2090123226002882, Apr. 2026, doi: https://doi.org/10.1016/j.jare.2026.04.009.
[21] S. Etemadi and M. Khashei, “Accuracy versus reliability-based modelling approaches for medical decision making,” Computers in Biology and Medicine, vol. 141, p. 105138, Feb. 2022, doi: https://doi.org/10.1016/j.compbiomed.2021.105138.
[22] C. Ghenai, M. Albawab, and M. Bettayeb, “Sustainability indicators for renewable energy systems using multi-criteria decision-making model and extended SWARA/ARAS hybrid method,” Renewable Energy, vol. 146, pp. 580–597, Feb. 2020, doi: https://doi.org/10.1016/j.renene.2019.06.157.
