Operation Strategy of EV Aggregators Considering EV Driving Model and Distribution System Operation in Integrated Power and Transportation Systems

Recently, the penetration of electric vehicle (EV) has increased, and several motor companies have declared a paradigm shift to EVs. However, most existing studies focusing on EV charging scheduling in power system operations neglect the integrated operation of transportation and distribution systems. Because EVs affect both systems, it is necessary to consider the grid operation and the individual EV's preferences (charging cost and time) simultaneously. In this study, an integrated power-transportation system structure and operation algorithms are developed to analyze the impact of EVs on the distribution system. In addition, this study proposes an optimal EV driving model that consists of two driving patterns: routine driving and long-distance driving. For routine driving patterns, the Markov chain model is implemented considering the mileage and conditions of individual EVs. Subsequently, long-distance driving patterns are developed to select the optimal route through reinforcement learning. In addition, the EV aggregator structure of EV charging stations can effectively predict local charge demands and perform charge control in real-time. An integrated power-transportation system is developed using the IEEE RBTS system and the integrated system architecture and algorithms are developed using MATLAB. It utilizes the ACOPPF function of MATPOWER to evaluate systematic stability and derive DLMP (Distributed Locational Marginal Price). This study conducts performance verification in various scenarios, confirming the effect of charging control of EV aggregators considering system stability, loss reduction, and cost reduction of the entire system.