PLANNING OF RENEWABLE ENERGY RESOURCES : optimization and stability

Abstract

In the current century, electrical networks have witnessed great developments and continuous increases in the demand for fossil-fuel-based energy, leading to excessive rise in the total production cost and the pollutant gases emitted by thermal plants. Under these circumstances, energy supply from different resources became necessary, such as renewable energy sources (RES) as an alternative solution. These sources, however, are characterized by uncertainty in their operational principle, especially when the system operator needs to define the optimal contribution of each resource to ensure economic efficiency and enhanced grid reliability. However, even with the huge demand met, networks still face other problems such as power loss and voltage instability. Therefore, FACTS devices appear as an effective solution, but they remain expensive. This thesis addresses the growing complexity of modern power systems as they incorporate high shares of variable renewable energy. A unified framework is developed that couples probabilistic modelling of wind, solar, and hydro output using Monte Carlo simulation and specific probability density functions (Weibull distribution for wind speeds, lognormal distribution for solar irradiance, and Gumbel distribution for river flow) with an enhanced metaheuristic optimizer tailored for large-scale optimal power flow. Key developments in the Kepler Optimization Algorithm include a novel exploratory–exploitative search operator for deeper solution-space exploration and a non-dominated sorting scheme to support efficient multi-objective trade-offs. Additionally, the framework incorporates SVC and TCSC devices by determining their optimal sizing and placement to reinforce the transmission lines and buses that demand the most reactive-power support, thereby achieving a cost-effective trade-off between capital investment and operational performance. When validated on a large scale test system, this integrated solution enhances economic efficiency, reduces environmental impact, and bolsters reliability under uncertainty. By combining advanced uncertainty quantification, customized metaheuristics, and targeted network reinforcement, this work provides a versatile, scalable methodology for planning and operating resilient, low-carbon electrical grids.