Yazar "Elweddad, Mohamed Ali H." seçeneğine göre listele

Listeleniyor 1 - 1 / 1
  • Küçük Resim
    Öğe
    AN OPTIMAL ENERGY MANAGEMENT SYSTEM FOR SUSTAINABLE CITY BASED ON RENEWABLE ENERGY SOURCES
    (2023-07) Elweddad, Mohamed Ali H.
    The present power generation system is confronted with difficulties such as reducing pollution, rising global energy consumption, high reliability requirements, energy cleanliness, and planning constraints. To achieve a sustainable and intelligent energy system, large and central generating stations are converted into small generating systems located close to residential buildings. Therefore, the design and installation of a micro grid lead to energy and cost savings. The micro grid consists of a number of loads, smart meters, communication system, traditional and alternative power generators, and storage systems. Energy management is necessary for the uninterrupted and reliable operation of a micro grid. Thus, energy management should be prioritized when developing multi-source system for economic and sustainable growth. Optimum scheduling of power generators operation leads to proper utilization and optimization of available energy sources while maintaining a balance between supply and demand. Many programs and smart algorithms can be used to manage and control the production and consumption of energy generated in a micro grid in order to calculate operating costs, reduce harmful gas emissions, maximize the use of renewable energy, reduce the cost of energy storage, and respond quickly to high loads., reduce energy cost, and finally simulate the components of the micro grid while carefully considering the constraints. This thesis aims to manage energy within the micro grid to supply residential loads effectively and cheaply. The first objective is to analyze six combinations of different energy sources to determine the best hybrid source in addition to improving the size and number of generation and storage units based on the cheapest total costs of the project. After that, obtaining the best energy source by comparing several economic and environmental factors help the decision-maker determine the best suitable combination for feeding a residential building to ensure optimal control of micro grids by considering reducing energy costs and reducing gas emissions as a main goal. Three research stages were investigated to determine the best hybrid system in terms of cost and sustainability. The first stage is determining the best size and optimization of the proposed system; the goal of this section is to use multi criteria decision-making algorithms to select the optimal design of six energy systems for sustainable energy to supply some buildings located in Tripoli. In this part, the HOMER software results were used to select all of the criteria for decision-making analysis. At first, the study used Homer software to determine optimal energy systems that can meet load demand while minimizing net present cost and the cost of energy. The technical, economic, and environmental results are explored for the most suitable system companion. To select the best HRE system, two decision-making algorithms (Vikor and Topsis) were implemented. Following that, in comparison to the other HRES, the final scores proved that PV /WT/Batt/Diesel generators are the best micro grid component for supplying the building. This proves that significant investment in hybrid PV/WT/Batt/Diesel generator systems will give the Libyan residential sector an excellent chance of achieving sustainable power. The second part of this study is a control strategy including ""ON/OFF"" operation of the available energy sources, including photovoltaic system PV- diesel generator, wind system, and energy storage banks using a Genetic algorithm. Then the output results from the algorithm are used as input data to machine learning models; in this phase, three algorithms were used to predict load and supply dispatch for the next 720 hours. The final part of the study compares the results obtained from the classification algorithms. The tables below show the high performance of the Decision Tree and Random Forest algorithms, where the accuracy reached 100% and 99%, respectively, in addition to the KNN algorithm, which was the worst with an accuracy of 90%.