Please use this identifier to cite or link to this item:
|Title:||Electric vehicle charging station placement and management||Authors:||Xiong, Yanhai||Keywords:||DRNTU::Engineering::Electrical and electronic engineering::Electric power||Issue Date:||2018||Source:||Xiong, Y. (2018). Electric vehicle charging station placement and management. Doctoral thesis, Nanyang Technological University, Singapore.||Abstract:||Due to the world’s shortage of fossil fuels and the serious environmental pollution from burning them, seeking alternative energy has become a crucial topic of research. Transportation is one of the main consumers of energy and contributors to air pollution. Electric Vehicles (EVs) move pollution away from urban areas and electricity can be efficiently transformed from both traditional fossil fuels and promising renewable energies like solar energy and tidal energy. EVs, as a replacement of traditional internal combustion engine vehicles, provide an environment-friendly solution to modern cities’ transportation. A rapid growth of EVs has been seen in recent years along with the rising popularity of the notion of smart cities. This calls for an efficient deployment of relevant supporting facilities, among which charging facility is of top priority. Although EVs can be charged at home, it is time-consuming and usually takes 6 to 8 hours, which is at least 12 times the time it takes at charging stations with high voltage. The distribution of charging stations determines EV drivers’ accessibility to energy sources and consequently affects the EV flow and traffic conditions in the road network. Although charging in charging stations is much faster than that with domestic electricity, it can still take several dozens of minutes. Thus in return, the EV drivers’ charging behavior would greatly influence the performance of the charging system, especially the queuing condition in charging stations. This thesis is concerned with optimal placement and efficient management of charging stations. To achieve this goal, we carefully study the interactions between charging stations and EV drivers as well as the bounded rationality of EV drivers in charging activities. In our first step of research, we study the electric vehicle charging station placement problem. We highlight two main factors to consider: traffic congestion and charging station congestion. We also take into consideration the electric vehicle drivers’ strategic charging activities. A congestion game framework is employed in our work to model the electric vehicle drivers’ competitive and self-interested charging activities. We formulate the charging station placement problem as a bi-level optimization problem and propose efficient algorithms for computing optimal solutions. Experimental results show that our approach provides a better result than baseline methods. We then extend to optimal pricing for charging station management. While most existing research works focus on optimizing spatial placement of charging stations, they are inflexible and inefficient against rapidly changing urban structure and traffic pattern. Therefore, this work approaches the management of EV charging stations from the pricing perspective as a more flexible and adaptive complement to established charging station placement. In this work, we build a realistic pricing model in consideration of residential travel pattern and EV drivers’ self-interested charging behavior, traffic congestion, and operating expense of charging stations. We formulate the pricing problem as a mixed integer non-convex optimization problem and propose a scalable algorithm to solve it. Experiments on both mock and real data are conducted, which show scalability of our algorithm as well as our solution’s significant improvement over existing approaches. Last, we study charging behavior of the EV drivers and construct more practical charging behavior models. While previous works assume that EV drivers can reach equilibrium in the charging game, this can rarely happen in real world. Players are limited by partial information and poor computation ability, thus they are bounded rational. Through analyzing EV drivers’ decision-making in the charging process, we propose a 2-Level Nested LQRE charging behavior model that combines LQRE model and level-k thinking model. We design a set of user studies to simulate the charging scenarios, collect data from human players and learn parameters of the 2-Level Nested LQRE charging behavior model. Experimental results show that our charging behavior model well captures the bounded rationality of human players in the charging process. The selection distribution of all players tends to converge after a number of repeated playing. Furthermore, we formulate the charging station placement problem with the 2-Level Nested LQRE model and design a heuristic algorithm to solve it. Our approach obtains placement with a significantly better performance by decreasing more than 8% for the social cost compared with benchmark approaches.||URI:||http://hdl.handle.net/10356/74999||DOI:||10.32657/10356/74999||Fulltext Permission:||open||Fulltext Availability:||With Fulltext|
|Appears in Collections:||IGS Theses|
Updated on May 16, 2021
Updated on May 16, 2021
Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.