ANALYZING THE IMPACT OF PLUG-IN ELECTRIC VEHICLE’S CHARGING LOAD ON THE GRID BASED ON DRIVER’S PERSONAL ATTITUDES TOWARDS PEV USAGE AND CHARGING

Today, the transport sector is responsible for nearly one-quarter of global energy-related direct carbon-dioxide (CO2) emissions and is a significant contributor to air pollution [1]. In the United States, the transportation sector has the highest share (28%) in the mix of green-house gas (GHG) sources [2]. Some of the more developed nations across the globe are now committed to improve the climate and air quality. Countries like China, Europe and the United States are front runners in introducing ambitions policies to incentivize the production and adoption of plug-in electric vehicles (PEV’s). Along with the expected benefits of PEV uptake, large scale deployment poses a challenge for the electric grid, especially at the distribution level, since the charging load of an PEV is substantial. This load is dependent not only on the characteristics of the PEV, but also on its use and charging habits of its user(s). Since a PEV can be directly plugged into the grid at any available point, which may be spatially anywhere in the utility’s service area, it is important to model its accurate use and charging behavior of the users. Having precise knowledge of the load profile, the utilities can have a better economic solution to balancing the supply and demand. In this dissertation, an agent-based model is developed that estimates the impact of charging load of PEVs on the grid. It is based on reasonably realistic diverse human behavior pertaining to day-to-day driving patterns and charging practices and their effect on each other. The model portrays the heterogenous, spatial and temporal nature of this load, which depends on the habits and the interaction among different agents. The model mimics the heterogeneity of choices made by human drivers and its effect on the charging choices of other drivers, which is an important element to consider when depicting human behavior. The model uses travel statistics of conventional personally owned vehicles (POVs) from the National Household Travel Survey (NHTS) conducted by the Federal Highway Administration (FHWA) across different states of the United States from 2016 – 2017. The travel needs are modified to incorporate the effect of EV’s limited range and charging time requirements. A modified GIS map of Collinsville, IL, is used to implement the spatial requirements of travel, with, which highlight exact load points. The agent’s travel and charging choices are modelled with heterogenous rules of engagement with the environment and other agents. Common psychological effects of limited range, long charging times, and range anticipation are applied heterogeneously to all agents to create a macro environment. The resulting charging load is superimposed on existing substation transformer load and voltage profile is analyzed to study the impact of different charging strategies and charging infrastructure availability. Different case studies are analyzed to investigate the effect of the aggregated load of multiple charging points in the respective service areas of the distribution transformers