Integration of electric vehicles into the smart grid and analysis of their effects on the grid with artificial intelligence methods

Energy is among the most important needs of our age. Electric vehicles (EV) have been rapidly increasing in popularity in recent years. The main reasons for this can be said that fossil fuels will run out and electric vehicles do not cause greenhouse gas emissions. Sustainable and clean energy is one of the goals today. The widespread use of electric vehicles is one of the main initiatives that support these advantages. But this increase will bring with it many problems. These vehicles will use electricity as fuel. Therefore, this increase will affect each of the production, transmission and distribution sections of electrical energy. In addition, it will have various effects on all consumers connected to electricity networks. The effects on the electricity grid will not only create infrastructural problems but also affect other consumers. These vehicles will be connected to the electricity grid through charging stations (EACS). Charging stations (CS) need to increase their number faster than electric vehicles. Because nowadays, the main obstacles to the widespread use of electric vehicles are the range problem and charging time. Electric vehicle charging stations are produced in different modes and types. Production standards may vary from country to country. They are produced in different powers. Installation and usage methods also vary. While individual users can install EAC, it can also be installed commercially. For this reason, installations are made by many different users. For this reason, their numbers are increasing rapidly. The increase in the number of EACs and EVs shows that the loads on the electricity networks are also increasing. For this reason, analysis of the effects of ECOMs on electricity networks has gained great importance among academic studies. In this thesis study, the effects of EACs on the electricity network were examined using the Institute of Engineers and Everyone Else (IEEE) electrical network test systems. Matlab Simulink model of IEEE power test systems with different bus numbers was designed and the effects of ECOMs on these networks were examined. Additionally, future impacts were predicted using deep learning (DL) with the collected data. In this way, malfunctions that may occur in connection with infrastructure planning and analysis for networks have become predictable. Another major impact of the widespread use of EACs and EVs on the grid will be sudden load increases. Especially in networks operating at full load during the day, it is possible that loads of greater than one power may be included. For this reason, the losses that will arise in response to this load increase are found by load flow analysis. By finding the existing networks' own losses, the increase in line losses resulting from the addition of EACs to the network as load is calculated and shown using Newton-Raphson load flow analysis. In order to reduce losses, solar power plants (SPPs), which support sustainable energy and the fight against greenhouse gas emissions, have been connected to the system so that the energy is produced where it is consumed, and it has been shown how much the losses have been reduced. For these modeling and analysis, Electrical Transient Analyzer Program (ETAP) was used. Finally, cost analyzes were carried out in order to provide EACS installation and planning advantage to investors. At this stage of the study, the installation types of EACS and the resulting costs are explained in detail. Earnings functions were created and the time spent by investors to recover their costs was estimated using artificial neural networks (ANN).