Heuristic-Based Optimization for Handling the Dynamic Response of Multi-Resource Energy Systems

Technological advancements, urbanization, high energy demand, and global requirements to mitigate carbon footprints have led to the adoption of innovative green technologies for energy production. The integration of green technologies with traditional grids offers huge benefits in terms of cost and voltage stability. However, this amalgamation may bring about a power mismatch dilemma due to intermittent renewable energy production and non-linear energy consumption patterns, which can affect the whole system’s reliability and operational efficiency. An efficient Energy Management System (EMS) is essential to deal with uncertainties associated with renewable energy production and load demand while optimizing the operation of distributed energy sources. This research study presents artificial intelligence-based solutions to improve EMS, focusing on optimization of energy sources, forecasting load, and renewable energy production. The findings of the literature review suggest that advanced metaheuristic algorithms can overcome the challenges of trapping in local optima and premature convergence, leading to widespread adoption and effective utilization in theoptimization problem. One study of this thesis proposed an Improved Lévy Flight Grey Wolf Optimization (ILFGWO) algorithm for capacity planning and energy management of a Hybrid Renewable Energy System.To enable a cost-effective transition, optimizing the energy mix at each time step is key to efficiently managing diverse energy sources. The proposed algorithm is applied to minimize three performance metrics: Annualized System Cost (ASC), Levelized Cost of Energy (LCOE), and Net Present Cost (NPC). The proposed approach combines dimension learning-based hunting with multi-neighbor learning and Lévy flight distribution, which improves its exploration and exploitation capabilities. This amalgamation helps the Grey Wolf Optimization (GWO) algorithm in avoiding the stagnation problem and can be verified by comparing its performance with five metaheuristic optimization algorithms. The findings reveal that the proposed algorithm outperforms other algorithms and achieves an ASC of 6574.340 (ke/yr), LCOE of 0.5301 (e/kWh), and NPC of 75407.213 (ke). Furthermore, statistical analysis is performed to demonstrate the effectiveness of the proposed ILFGWO algorithm on 20 independent runs. The proposed technique outperforms other optimization methods by achieving the lowest minimum (6574.340 ×10³ e), maximum (6576.870 ×10³ e), average (6574.848 ×10³ e) cost values and the smallest standard deviation (592.18 e). The paired t-test validates the best performance and reliability of the proposed approach, demonstrating that the comparisons between the proposed method and other optimization techniques are statistically significant. Regarding carbon emissions, the proposed system is expected to avoid carbon emissions by 72.05%. A thesis study implemented a case study on the installation of PV and ESS in each building on the Unipa Campus. The study aims to determine the optimal size of the ESS for each building using PSO with the objective of maximizing the net present value. The findings suggest that those buildings that consume low energy do not need to have a storage system, and excess energy from PV can be directly sent back to the grid. In terms of stability, the Unipa network of energy distribution is modeled using the Naplan tool, and stability analysis is performed to measure the voltage stability. In the case of installing PVand ESS, the network offers a significant improvement in voltage. Therefore, compared to the baseline scenario, when meeting all energy demand from an external grid, installing PV and ESS not only reduces energy cost but also improves the voltage profile.In addition, renewable energies are essential for the clean energy transition needed to achieve the objectives of the European Green Deal and reduce the dependence on fossil fuels and energy imports. The energy transition can strengthen the stability of the energy market, especially in small islands. At the same time, the need for large infrastructure construction could be alleviated in developing areas. In recent years, increasingly stringent laws on polluting emissions and the need to lower production costs have greatly influenced the technologicalprogress in the field of renewable energies. However, in various remote areas and small islands, the production of energy is still dominated by the use of fossil fuels. Some chapters of the thesis discuss the feasibility of implementing a mixed system of renewable sources (wind, photovoltaic, and wave energy) for small islands of the Pacific and Mediterranean sea, considering alternatively the minimization of the LCOE and estimation of the CO2 emission of the proposed energy mix. The study was carried out by estimating energy productionmonthly and annually, considering a mixture of three renewable sources. This method can be easily applied to several small islands, estimating the ability to reduce the production of energy from fossil fuels.