Charging infrastructure optimization for plug-in electric vehicles
Conventional light duty vehicle fleets and petroleum are firmly bonded. It is leading to several issues, such as energy security, greenhouse gas (GHG) and criteria pollutant emissions. Plug-in electric vehicles (PEVs), including plug-in hybrid electric vehicles (PHEVs) and battery electric vehicles (BEVs), have the potential to break this bond and improve the energy and environmental landscape of personal transportation. The United States, especially the state of California is very proactive in promoting PEV adoption. Several regulations and laws have been passed to stimulate PEV growth, including the Zero Emission Vehicle (ZEV) Regulation and Assembly Bill 32 (Climate Change). However, society faces three main challenges associated with PEV deployment from the perspective of charging infrastructure. This dissertation addresses each step by step. First, a methodology is established to quantify the energy impact of PEVs. In particular, a travel behavior based PEV operating/charging model is made to characterize fleet-wide energy consumption with different PEV parameters and charging infrastructure scenarios, such as location, power level and charging time strategy. Second, PEVs face the hurdle of access to charging infrastructure. Consequently, question has to be answered as to what types, locations, and quantities of electric vehicle supply equipment (EVSE) will be required. For this purpose, an optimal charging strategy based on 24-hour travel patterns is formulated to minimize operating cost. With that, the approximation of the Level 1 and Level 2 EVSE needed at different types of location categories is proposed. As an alternative infrastructure solution for BEVs, Level 3 DC fast charging stations are also investigated in terms of location allocation. Third, a massive population of PEVs has the potential to change the grid operation. To this end, a durable decentralized charging protocol is proposed and verified for coordinating individual PEVs with grid operation such that grid-level optimality can be achieved.