The purpose of the study was to conduct a comprehensive assessment of the effectiveness of intelligent load management methods within Ukraine’s hybrid power systems, under conditions of an increasing proportion of renewable energy sources and the integration of small modular reactors, and to determine their impact on energy efficiency, operational stability, and the dynamics of daily load fluctuations. The methodology included a step-by-step analysis of energy-efficient control technologies, modelling of the operation of the 330/110 kV Kremenchuk substation based on actual operating profiles, and five-year load forecasting using a Long Short-Term Memory model and optimisation algorithms. The study established that traditional dispatch control mechanisms maintain stability only with low generation variability, while intelligent approaches provide a significant increase in accuracy, reduction of losses and improvement of mode stability. The results showed that optimisation based on a genetic algorithm reduces energy consumption by 12.4% and costs by 9.1%, while the Particle Swarm Optimisation algorithm demonstrated the highest efficiency, reducing energy consumption by up to 18.1%, reducing costs by up to 14.7% and providing the most accurate reproduction of daily profiles. Forecast calculations for the period up to 2030 showed an increase in the daily load amplitude and identified intervals of increased sensitivity in which the use of intelligent control strategies can reduce daily active power deviations by more than 40% and losses to be reduced by almost 17%. The practical significance of the results is determined by the fact that the established patterns can be used to modernise Ukrainian power grids, optimise the integration of renewable generation, increase the reliability of industrial power systems, and justify the introduction of smart technologies in grids with a growing share of unstable sources
Ievgen Alfimov (Fri,) studied this question.