Abstract

In the context of global energy transformation and the continuous growth of renewable energy sources (RES) in generation portfolios, managing their stochastic behavior and ensuring their integration into existing power systems has become critically important. This study presents a theoretical foundation and proposes AI-based adaptive algorithms for the predictive control of hybrid energy systems (HES). The objective is to formulate a smart control concept aimed at the comprehensive optimization of both thermodynamic and economic performance of HES. Within this framework, a conceptual SAEO (Smart Adaptive Energy Optimization) model is introduced, integrating RES, energy storage technologies (including hydrogen systems), and a gas-steam combined cycle. The results demonstrate that implementation of the developed adaptive algorithms increases overall system efficiency and reduces the levelized cost of energy compared with traditional control schemes based on fixed logic rules. Based on these findings, it is concluded that the intelligent enhancement of control algorithms is a key prerequisite for achieving a synergistic effect in complex hybrid energy systems. The presented results may be of value to power engineers, AI researchers, and strategic planning specialists in the energy sector.

Keywords

  • hybrid energy system
  • artificial intelligence
  • predictive control
  • thermodynamic efficiency
  • economic efficiency
  • SAEO
  • hydrogen energy
  • renewable energy sources
  • optimization
  • SCADA.

References

  1. 1. International Energy Agency. Renewables 2023. IEA, Paris, 2023. [Electronic resource]. - Access mode: https://www.iea.org/reports/renewables-2023 (date accessed: 10.06.2025).
  2. 2. Jamil I. et al. Predictive evaluation of solar energy variables for a large-scale solar power plant based on triple deep learning forecast models //Alexandria Engineering Journal. – 2023. – Vol. 76. – pp. 51-73. https://doi.org/10.1016/j.aej.2023.06.023.
  3. 3. Upadhyay S., Ahmed I., Mihet-Popa L. Energy management system for an industrial microgrid using optimization algorithms-based reinforcement learning technique //Energies. – 2024. – Vol. 17 (16). – pp. 1-18. https://doi.org/10.3390/en17163898.
  4. 4. Fan Z., Zhang W., Liu W. Multi-agent deep reinforcement learning-based distributed optimal generation control of DC microgrids //IEEE Transactions on Smart Grid. – 2023. – Vol. 14 (5). – pp. 3337-3351. https://doi.org/10.1109/TSG.2023.3237200.
  5. 5. Garip M., Sulukan E., Celiktas M. S. Optimization of a grid-connected hybrid energy system: Techno-economic and environmental assessment //Cleaner Energy Systems. – 2022. – Vol. 3. – pp. 1-14. https://doi.org/10.1016/j.cles.2022.100042.
  6. 6. Ammar M. B., Zdiri M. A., Ammar R. B. Fuzzy logic energy management between stand-alone PV systems //Power. – 2021. – Vol. 9. – pp. 1238-1249.
  7. 7. Akarsu B., GenΓ§ M. S. Optimization of electricity and hydrogen production with hybrid renewable energy systems //Fuel. – 2022. – Vol. 324. https://doi.org/10.1016/j.fuel.2022.124465.
  8. 8. Zhou D., Zhu Z. Urban integrated energy system stochastic robust optimization scheduling under multiple uncertainties //Energy Reports. – 2023. – Vol. 9. – pp. 1357-1366. https://doi.org/10.1016/j.egyr.2023.04.193.
  9. 9. Alqahtani B., Yang J., Paul M. C. A techno-economic-environmental assessment of a hybrid-renewable pumped hydropower energy storage system: A case study of Saudi Arabia //Renewable Energy. – 2024. – Vol. 232. – pp. 1-16. https://doi.org/10.1016/j.renene.2024.121052.
  10. 10. International Renewable Energy Agency. Global hydrogen trade to meet the 1.5Β°C climate goal: Part I – Trade outlook for 2050 and beyond. IRENA, Abu Dhabi, 2022. [Electronic resource]. - Access mode: https://india-re-navigator.com/public/uploads/1651644555-IRENA_Global_Trade_Hydrogen_2022.pdf (date accessed: 16.06.2025).
  11. 11. Global Electricity Review 2024. [Electronic resource]. - Access mode: https://ember-climate.org/insights/research/global-electricity-review-2024/ (date accessed: 20.06.2025).