Author : Yanbo Ge
Publisher :
ISBN 13 :
Total Pages : 185 pages
Book Rating : 4.:/5 (11 download)
Book Synopsis Discrete Choice Modeling of Plug-in Electric Vehicle Use and Charging Behavior Using Stated Preference Data by : Yanbo Ge
Download or read book Discrete Choice Modeling of Plug-in Electric Vehicle Use and Charging Behavior Using Stated Preference Data written by Yanbo Ge and published by . This book was released on 2019 with total page 185 pages. Available in PDF, EPUB and Kindle. Book excerpt: Plug-in Electric Vehicles (PEVs) have the potential of reducing gasoline consumption and greenhouse gas emissions in the transportation sector. The net impacts of PEVs – including upstream emissions from electricity generation and the impact these vehicles place on the electricity grid – depend on both the amount of travel conducted by PEV and locations that those PEVs are charged. This dissertation investigates the vehicle use choices and charging decisions of both battery electric vehicles (BEVs) and plug-in hybrid electric vehicles (PHEVs) for both home-based trip tours and long-distance trips using stated preference (SP) data. It presents a novel dynamic discrete choice modeling (DDCM) framework that explicitly accounts for the stochastic nature of the vehicle choice and charging decisions of PEV users: earlier choices on vehicle use and charging influence the utility of the future choices; the expectation of the future options influences those earlier decisions; and choices are made under uncertainty about actual energy consumption and availability of chargers. For home-based trip tours, my results show that BEV users are willing to pay $10-$24 to avoid having to deviate from the originally planned route, which indicates that “range anxiety” of BEV owners – the fear of being stranded in the middle of a trip – is not a crucial issue for home-based trips. Using charging infrastructure development to encourage BEV adoption might be more beneficial than reducing “range anxiety” among the current users, which could entail building charging stations at locations that have more public exposure, such as public parking garages in a city center. When BEVs are on long-distance trips, the cost of deviation is significantly higher: $244, which indicates that BEV owners are likely to be more cautious and view finding a charger off the route much more costly when they are on long-distance trips. Comparing the cost of deviation for home-based tours and long-distance trips, to support the existing users, the most cost-effective places to invest in charging infrastructure are inter-city corridors instead of in-city locations. By comparing the relative size of the coefficient estimates, in this dissertation, I also analyze the monetary value of increasing charging power, moving the charging stations closer to highway exits, and having amenities such as restrooms, restaurants, and Wi-Fi near the charging stations. The comparison between the DDCMs and SDCMs based on simpler decision heuristics shows that for home-based tours, DDCMs only offer a little better prediction rate with a significant cost when it comes to computation time and complexity of model development. For the purpose of demand forecasting of a charging network or site selection for the charging facilities, the SDCMs based on simpler heuristics are recommended for home-based trip tours. For long-distance trips, the charging choices are largely decided by the state of charge (SOC) and deviation, and the characteristics of the charging stations only contribute to a small portion of predictive power. SDCMs outperform the DDCMs for the current sample. However, this could change in the future when the charging network is dense and the characteristics of the charging stations have higher prediction power. For both the home-based tours and long-distance trips, and for both vehicle choices and charging decisions, the decision patterns are likely to be heterogeneous among the PEV owners. The efforts related to the prediction of the future EV charging demand, the policy-making on battery and charging infrastructure development, and the planning/design of the charging network all need to consider these different preferences of the consumers. Due to the heterogeneity of users’ preferences, both increasing battery pack size and reducing station spacing can encourage current BEV owners to use their BEVs for long-distance trips, and one of the two does not substitute the other. Even if a lot of the BEV models offered by the market have 500 miles of range, the density of the public charging network can still play an important role in enabling BEVs for long-distance trips, especially when the battery remains expensive.
