Dissertation for the degree of Doctor of Philosophy in speciality 141 “Electric Power Engineering, Electrical Engineering and Electromechanics”. – Institute of Renewable Energy of the National Academy of Sciences of Ukraine, Kyiv, 2026.
The dissertation addresses the improvement of dynamic stability, reliability, and energy efficiency of local combined power supply systems (LCPS) integrating renewable energy sources (RES), inverter-based resources, and energy storage systems (ESS) under conditions of inertia deficit and stochastic operating changes. The object of the research is the conversion, balancing, and control of energy flows in LCPS. The subject of the research comprises methods, models, and algorithms for multilevel adaptive control of their operating modes.
A multilevel adaptive approach has been developed in which the LCPS operating mode is treated as a controllable variable and selected from a set of admissible states with account for RES and ESS conditions, external grid parameters, and optimization criteria. An integral adaptability criterion is proposed to combine reliability, energy efficiency, ESS degradation, and dynamic stability. Mathematical models are developed for Grid-Following, Grid-Forming, islanded operation, zero-power-flow operation, and resynchronization, together with a method for constructing three-dimensional regime maps.
A digital twin of the LCPS has been created, integrating models of RES, ESS, inverter converters, and control algorithms and providing for integration with EMS/SCADA using CIM (IEC 61970/61968) and IEC/TS 62351 cybersecurity mechanisms. The stochastic nature of generation and load is considered using probabilistic models and Monte Carlo simulation. Dynamic stability is assessed using RoCoF, frequency and voltage deviations, settling time, LVRT/HVRT indicators, and ESS resource characteristics.
Numerical simulation results show that adaptive control in GFL/GFM, islanded, and resynchronization modes can reduce RoCoF by 28–35%, shorten frequency settling time by 30–40%, and reduce ESS degradation by 18–25% compared with baseline scenarios. The highest adaptability zone is associated with an ESS state of charge of 50–65%, while a virtual power plant formed on the basis of such LCPS achieves a resynchronization success rate of approximately 95%.
The scientific novelty consists in the development of multilevel adaptive control with dynamic operating-mode selection, further development of models for multicriteria coordination of reliability, energy efficiency, and ESS lifetime, and advancement of digital-twin and regime-map methods. The practical significance lies in the applicability of the proposed approaches to microgrids, critical infrastructure, telecommunication facilities, data centers, and other facilities with RES and ESS. The estimated economic effect of implementing the adaptive algorithm is UAH 2.384 million per year due to extended equipment lifetime and reduced energy losses.