Impact of plug-in hybrid electric vehicles on the power grid in a smart environment

This dissertation focuses on the interactions between plug-in hybrid electric vehicles (PHEVs) and the power grid. This study has used the data available through the national household travel survey (NHTS) to obtain realistic information such as the arrival time of vehicles, miles driven, and vehicles types. After data manipulation, the aggregated PHEV charging load profiles (PCLP) have been developed. The results show that the peak of PCLP overlaps with the domestic load peak. In order to alleviate this issue, several charging policies have been developed and their impacts on PCLP have been investigated. As a next contribution, the data obtained from NHTS has been used to extract probability distribution functions (PDF) for the arrival time and the energy required to fully charge PHEVs. These functions have been used to propose an event-based simulation framework which emulates the interactions between the power grid and PHEVs and examines how a power grid with limited capacity responds to PHEV charging demand. For instance, for a particular region in the United States and for a specific penetration level, it is possible to obtain the wait time of vehicles, the number of vehicles receiving service, and the number of vehicles that give up. In a smart environment, it becomes necessary to define and schedule optimum policies for smart charging of PHEVs. Finally, this work proposes an optimization problem to find an optimal charging rate schedule for PHEVs. This rate which is variable by time tries to maximize the energy delivered to vehicles while considering the constraints of the grid as well as the wait time of the vehicles.