الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 134 |  |  |
| | | from | 134 | | |
Abstract
With an increasing focus on environmental conservation and sustainability, the shift toward clean and renewable energy has become desirable and crucial. Renewable Energy Sources (RESs), such as photovoltaics (PVs) and wind turbines (WTs), have become critical players in this significant transition. A remarkable surge in DC-based energy resources and load demands, including the broad adoption of electric vehicles (EVs) and a consistent rise in solar PV capacity, clearly shows this global trend. This transformation has prompted the emergence of Hybrid AC/DC Microgrids (HMGs), which efficiently integrate various energy sources. These include traditional power sources, RESs, and battery energy storage systems (BESSs). The development of HMGs underscores the need for strategies to manage energy effectively in both grid-connected and islanded scenarios.
This thesis presents a stochastic energy management (EM) approach in multi-area HMGs. The focus is optimizing day-ahead dispatch in grid-connected and islanded modes. The primary objectives in the grid-connected mode is minimizing the daily operational costs and daily energy loss. An essential aspect of this approach is the flexibility it offers to the operator of the HMGs, allowing them to choose the weighting factors that best suit their needs. Furthermore, the approach considers the charging and discharging of EVs while considering the EVs batteries’ degradation cost. In the islanded mode, the objectives are reducing the operational costs and improving the system reliability by minimizing power outages. The strategy also encompasses simultaneous active and reactive power dispatch, incorporating the reactive power costs of Diesel Generators Unit (DUs ) within the operational scheduling process.
A significant challenge in energy management of HMGs is the exposure to multiple sources of uncertainties. Environmental factors such as weather conditions directly influence the output of PV and WTs, while economic uncertainties affect electricity market prices and demand, including the demand at EV charging stations. Accurate modeling of these uncertainties is vital to achieving realistic results. Accordingly, this thesis explores two methodologies in this regard: Unscented Transformations (UT) and Monte Carlo Simulation (MCS) with Fast Forward Scenario reduction (FFS).
To achieve the required objectives of the energy management of the HMG in the two modes of operation, two optimization problems are formulated as a constrained mixed-integer nonlinear programming problem. These optimization problems are solved using several metaheuristic optimization algorithms like Transient Search Optimization (TSO) and Hunger Game Search (HGS). Finally, the effectiveness of this comprehensive approach is investigated through several case studies, simulated on HMGs with multiple areas in both grid and islanded mode.