Optimal centralized and decentralized management strategies for electric vehicles considering customer demand, road and electric grid infrastructure

The electrical grid is a complex network becoming increasingly linked with smart devices and energy sources, such as electric vehicles, appliances, grid support systems, and renewable energy resources. However, the power delivery systems in use today are antiquated and have been in operation for over a hundred years. In this dissertation, several methodologies for optimal management of electric vehicle (EV) fleets connected to the power grid are discussed. First, a hybrid methodology is suggested for determining the quickest way for a vehicle to reach the charging station, taking into account both the distance and the current traffic conditions developed based on graph theory. The strategy is accurate, more efficient, and scalable. Second, a technique that considers the shortest distance to the charging station considering the impact and optimal use of the electric grid is developed. The method takes advantage of distance and simultaneously considers the influence on the grid, such as variations in voltage or power. The procedure is tested and quantitative and qualitative analysis is conducted. Also, with the help of a convex optimization methodology, a speed optimization framework is developed that mitigates range anxiety. Next, an optimization methodology is developed that addresses real-time electric car charging congestion as well as centralized and decentralized charging scheduling of electric vehicles. The charging of plug-in electric vehicles (PEVs) has to be handled through the use of "smart" charging processes to lessen the demand that PEVs have on the electrical grid. These studies examine the impact that the actual implementation of four distinct smart charging architectures has on the electric grid, including a centralized and decentralized design. The capabilities of each method are summarized.
Further, a methodology for demand-side management and distributed load management is developed, considering customer comfort with the help of an electric vehicle fleet. A new mathematical model of household loads such as air conditioners, water heaters, clothes dryers, and dishwashers considering the weather conditions is developed. It was identified that during high temperatures, the system's operational architecture may derive a significant advantage from these massive demand-responsive loads. Further, a robust energy optimization framework is proposed that suggests healthy results to keep the grid stable and sustained after optimizing household loads avoiding customer comfort violation. The proposed methodologies are scalable, field implementable, and have a significant advantage in collectively managing electric vehicle fleets, customer comfort, and energy usage considering road and grid conditions.